Stratix IV Device Handbook
Volume 1
101 Innovation Drive
San Jose, CA 95134
www.altera.com
SIV5V1-4.6
© 2012 Altera Corporation. All rights reserved. ALTERA, ARRIA, CYCLONE, HARDCOPY, MAX, MEGACORE, NIOS, QUARTUS and STRATIX words and logos
are trademarks of Altera Corporation and registered in the U.S. Patent and Trademark Office and in other countries. All other words and logos identified as
ISO
trademarks or service marks are the property of their respective holders as described at www.altera.com/common/legal.html. Altera warrants performance of its
semiconductor products to current specifications in accordance with Altera's standard warranty, but reserves the right to make changes to any products and 9001:2008
services at any time without notice. Altera assumes no responsibility or liability arising out of the application or use of any information, product, or service Registered
described herein except as expressly agreed to in writing by Altera. Altera customers are advised to obtain the latest version of device specifications before relying
on any published information and before placing orders for products or services.
September 2012
Altera Corporation
Stratix IV Device Handbook
Volume 1
Contents
Chapter Revision Dates . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xi
Section I. Device Core
Chapter 1. Overview for the Stratix IV Device Family
Feature Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1–2
Stratix IV GX Devices . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1–3
Stratix IV E Device . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1–4
Stratix IV GT Devices . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1–5
Architecture Features . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1–6
High-Speed Transceiver Features . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1–6
Highest Aggregate Data Bandwidth . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1–6
Wide Range of Protocol Support . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1–6
Diagnostic Features . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1–7
Signal Integrity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1–7
FPGA Fabric and I/O Features . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1–8
Device Core Features . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1–8
Embedded Memory . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1–8
Digital Signal Processing (DSP) Blocks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1–8
Clock Networks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1–8
PLLs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1–9
I/O Features . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1–9
High-Speed Differential I/O with DPA and Soft-CDR . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1–9
External Memory Interfaces . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1–9
System Integration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1–10
Integrated Software Platform . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1–19
Ordering Information . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1–19
Chapter 2. Logic Array Blocks and Adaptive Logic Modules in Stratix IV Devices
Logic Array Blocks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2–1
LAB Interconnects . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2–4
LAB Control Signals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2–4
Adaptive Logic Modules . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2–5
ALM Operating Modes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2–8
Normal Mode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2–9
Extended LUT Mode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2–11
Arithmetic Mode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2–12
Shared Arithmetic Mode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2–14
LUT-Register Mode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2–15
Register Chain . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2–17
ALM Interconnects . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2–18
Clear and Preset Logic Control . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2–18
LAB Power Management Techniques . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2–19
Chapter 3. TriMatrix Embedded Memory Blocks in Stratix IV Devices
Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3–1
TriMatrix Memory Block Types . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3–3
Parity Bit Support . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3–3
September 2012
Altera Corporation
Stratix IV Device Handbook
Volume 1
iv
Contents
Byte Enable Support . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3–3
Packed Mode Support . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3–5
Address Clock Enable Support . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3–6
Mixed Width Support . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3–7
Asynchronous Clear . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3–7
Error Correction Code (ECC) Support . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3–8
Memory Modes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3–9
Single-Port RAM Mode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3–10
Simple Dual-Port Mode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3–11
True Dual-Port Mode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3–14
Shift-Register Mode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3–16
ROM Mode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3–17
FIFO Mode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3–17
Clocking Modes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3–17
Independent Clock Mode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3–18
Input/Output Clock Mode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3–18
Read/Write Clock Mode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3–18
Single Clock Mode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3–18
Design Considerations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3–18
Selecting TriMatrix Memory Blocks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3–18
Conflict Resolution . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3–19
Read-During-Write Behavior . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3–19
Same-Port Read-During-Write Mode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3–19
Mixed-Port Read-During-Write Mode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3–21
Power-Up Conditions and Memory Initialization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3–23
Power Management . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3–24
Chapter 4. DSP Blocks in Stratix IV Devices
Stratix IV DSP Block Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4–2
Stratix IV Simplified DSP Operation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4–4
Stratix IV Operational Modes Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4–8
Stratix IV DSP Block Resource Descriptions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4–9
Input Registers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4–10
Multiplier and First-Stage Adder . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4–12
Pipeline Register Stage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4–13
Second-Stage Adder . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4–13
Rounding and Saturation Stage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4–14
Second Adder and Output Registers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4–14
Stratix IV Operational Mode Descriptions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4–15
Independent Multiplier Modes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4–15
9-, 12-, and 18-Bit Multiplier . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4–15
36-Bit Multiplier . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4–19
Double Multiplier . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4–20
Two-Multiplier Adder Sum Mode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4–22
18 x 18 Complex Multiply . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4–24
Four-Multiplier Adder . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4–26
High-Precision Multiplier Adder Mode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4–27
Multiply Accumulate Mode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4–29
Shift Modes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4–30
Rounding and Saturation Mode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4–32
DSP Block Control Signals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4–34
Software Support . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4–35
Stratix IV Device Handbook
Volume 1
September 2012 Altera Corporation
Contents
v
Chapter 5. Clock Networks and PLLs in Stratix IV Devices
Clock Networks in Stratix IV Devices . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5–1
Global Clock Networks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5–3
Regional Clock Networks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5–4
Periphery Clock Networks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5–6
Clock Sources Per Quadrant . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5–9
Clock Regions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5–9
Clock Network Sources . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5–10
Dedicated Clock Input Pins . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5–10
LABs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5–10
PLL Clock Outputs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5–11
Clock Input Connections to the PLLs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5–12
Clock Output Connections . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5–13
Clock Control Block . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5–14
Clock Enable Signals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5–17
Clock Source Control for PLLs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5–18
Cascading PLLs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5–19
PLLs in Stratix IV Devices . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5–19
Stratix IV PLL Hardware Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5–23
PLL Clock I/O Pins . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5–24
PLL Control Signals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5–27
pfdena . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5–27
areset . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5–27
locked . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5–27
Clock Feedback Modes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5–28
Source Synchronous Mode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5–29
Source-Synchronous Mode for LVDS Compensation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5–30
No-Compensation Mode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5–30
Normal Mode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5–31
Zero-Delay Buffer (ZDB) Mode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5–31
External Feedback Mode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5–32
Clock Multiplication and Division . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5–33
Post-Scale Counter Cascading . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5–34
Programmable Duty Cycle . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5–35
Programmable Phase Shift . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5–35
Programmable Bandwidth . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5–37
Background . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5–37
Implementation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5–38
Spread-Spectrum Tracking . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5–39
Clock Switchover . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5–39
Automatic Clock Switchover . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5–40
Manual Clock Switchover . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5–43
Guidelines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5–43
PLL Reconfiguration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5–44
PLL Reconfiguration Hardware Implementation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5–45
Post-Scale Counters (C0 to C9) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5–47
Scan Chain Description . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5–48
Charge Pump and Loop Filter . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5–50
Bypassing a PLL . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5–51
Dynamic Phase-Shifting . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5–51
PLL Specifications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5–54
September 2012
Altera Corporation
Stratix IV Device Handbook
Volume 1
vi
Contents
Section II. I/O Interfaces
Chapter 6. I/O Features in Stratix IV Devices
I/O Standards Support . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6–2
I/O Standards and Voltage Levels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6–3
I/O Banks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6–5
Modular I/O Banks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6–8
I/O Structure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6–17
3.3-V I/O Interface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6–19
External Memory Interfaces . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6–19
High-Speed Differential I/O with DPA Support . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6–20
Programmable Current Strength . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6–20
Programmable Slew Rate Control . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6–21
Programmable I/O Delay . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6–22
Programmable IOE Delay . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6–22
Programmable Output Buffer Delay . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6–22
Open-Drain Output . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6–22
Bus Hold . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6–22
Programmable Pull-Up Resistor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6–23
Programmable Pre-Emphasis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6–23
Programmable Differential Output Voltage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6–23
MultiVolt I/O Interface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6–23
On-Chip Termination Support and I/O Termination Schemes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6–24
On-Chip Series (RS) Termination Without Calibration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6–25
On-Chip Series Termination with Calibration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6–26
Left-Shift Series Termination Control . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6–27
On-Chip Parallel Termination with Calibration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6–28
Expanded On-Chip Series Termination with Calibration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6–29
Dynamic On-Chip Termination . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6–29
LVDS Input OCT (RD) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6–31
Summary of OCT Assignments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6–31
OCT Calibration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6–32
OCT Calibration Block Location . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6–32
Sharing an OCT Calibration Block on Multiple I/O Banks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6–34
OCT Calibration Block Modes of Operation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6–35
Power-Up Mode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6–35
User Mode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6–36
OCT Calibration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6–37
Serial Data Transfer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6–37
Example of Using Multiple OCT Calibration Blocks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6–38
RS Calibration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6–38
Termination Schemes for I/O Standards . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6–38
Single-Ended I/O Standards Termination . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6–38
Differential I/O Standards Termination . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6–41
LVDS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6–43
Differential LVPECL . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6–44
RSDS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6–45
Mini-LVDS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6–46
Design Considerations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6–46
I/O Bank Restrictions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6–46
Non-Voltage-Referenced Standards . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6–47
Voltage-Referenced Standards . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6–47
Mixing Voltage-Referenced and Non-Voltage-Referenced Standards . . . . . . . . . . . . . . . . . . . . . 6–47
Stratix IV Device Handbook
Volume 1
September 2012 Altera Corporation
Contents
vii
Chapter 7. External Memory Interfaces in Stratix IV Devices
Memory Interfaces Pin Support . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7–3
Using the RUP and RDN Pins in a DQS/DQ Group Used for Memory Interfaces . . . . . . . . . . . . . . 7–26
Combining ×16/×18 DQS/DQ Groups for a ×36 QDR II+/QDR II SRAM Interface . . . . . . . . . . . 7–26
Rules to Combine Groups . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7–27
Stratix IV External Memory Interface Features . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7–29
DQS Phase-Shift Circuitry . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7–29
DLL . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7–31
Phase Offset Control . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7–41
DQS Logic Block . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7–43
DQS Delay Chain . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7–44
Update Enable Circuitry . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7–44
DQS Postamble Circuitry . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7–45
Leveling Circuitry . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7–46
Dynamic On-Chip Termination Control . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7–48
I/O Element Registers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7–49
Delay Chain . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7–52
I/O Configuration Block and DQS Configuration Block . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7–54
Chapter 8. High-Speed Differential I/O Interfaces and DPA in Stratix IV Devices
Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8–1
Locations of the I/O Banks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8–3
LVDS Channels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8–4
LVDS SERDES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8–8
ALTLVDS Port List . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8–9
Differential Transmitter . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8–11
Programmable VOD and Programmable Pre-Emphasis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8–14
Programmable VOD . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8–15
Programmable Pre-Emphasis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8–16
Differential Receiver . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8–17
Differential I/O Termination . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8–18
Receiver Hardware Blocks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8–19
DPA Block . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8–19
Synchronizer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8–20
Data Realignment Block (Bit Slip) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8–20
Deserializer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8–22
Receiver Data Path Modes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8–22
Non-DPA Mode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8–22
DPA Mode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8–24
Soft-CDR Mode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8–25
LVDS Interface with the Use External PLL Option Enabled . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8–26
Left and Right PLLs (PLL_Lx and PLL_Rx) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8–29
Stratix IV Clocking . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8–30
Source-Synchronous Timing Budget . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8–31
Differential Data Orientation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8–31
Differential I/O Bit Position . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8–31
Transmitter Channel-to-Channel Skew . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8–33
Receiver Skew Margin for Non-DPA Mode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8–33
Differential Pin Placement Guidelines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8–38
Guidelines for DPA-Enabled Differential Channels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8–38
DPA-Enabled Channels and Single-Ended I/Os . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8–38
DPA-Enabled Channel Driving Distance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8–38
Using Corner and Center Left and Right PLLs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8–38
September 2012
Altera Corporation
Stratix IV Device Handbook
Volume 1
viii
Contents
Using Both Center Left and Right PLLs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8–40
Guidelines for DPA-Disabled Differential Channels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8–42
DPA-Disabled Channels and Single-Ended I/Os . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8–42
DPA-Disabled Channel Driving Distance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8–42
Using Corner and Center Left and Right PLLs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8–42
Using Both Center Left and Right PLLs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8–45
Section III. System Integration
Chapter 9. Hot Socketing and Power-On Reset in Stratix IV Devices
Stratix IV Hot-Socketing Specifications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9–1
Stratix IV Devices can be Driven Before Power Up . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9–2
I/O Pins Remain Tri-Stated During Power Up . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9–2
Insertion or Removal of a Stratix IV Device from a Powered-Up System . . . . . . . . . . . . . . . . . . . . . . 9–2
Hot-Socketing Feature Implementation in Stratix IV Devices . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9–3
Power-On Reset Circuitry . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9–4
Power-On Reset Specifications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9–5
Chapter 10. Configuration, Design Security, and Remote System Upgrades in Stratix IV Devices
Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10–1
Configuration Devices . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10–2
Configuration Schemes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10–2
Configuration Features . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10–4
Power-On Reset Circuit . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10–5
VCCPGM Pins . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10–5
VCCPD Pins . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10–5
Fast Passive Parallel Configuration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10–6
FPP Configuration Using a MAX II Device as an External Host . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10–6
FPP Configuration Timing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10–12
FPP Configuration Using a Microprocessor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10–16
Fast Active Serial Configuration (Serial Configuration Devices) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10–16
Estimating Active Serial Configuration Time . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10–22
Programming Serial Configuration Devices . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10–23
Guidelines for Connecting Serial Configuration Devices on an AS Interface . . . . . . . . . . . . . . . . . 10–25
Passive Serial Configuration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10–25
PS Configuration Using a MAX II Device as an External Host . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10–26
PS Configuration Timing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10–31
PS Configuration Using a Microprocessor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10–32
PS Configuration Using a Download Cable . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10–32
JTAG Configuration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10–35
Jam STAPL . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10–40
Device Configuration Pins . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10–40
Configuration Data Decompression . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10–48
Remote System Upgrades . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10–50
Functional Description . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10–51
Enabling Remote Update . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10–53
Configuration Image Types . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10–54
Remote System Upgrade Mode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10–54
Remote Update Mode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10–54
Dedicated Remote System Upgrade Circuitry . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10–57
Remote System Upgrade Registers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10–58
Remote System Upgrade Control Register . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10–58
Remote System Upgrade Status Register . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10–59
Stratix IV Device Handbook
Volume 1
September 2012 Altera Corporation
Contents
ix
Remote System Upgrade State Machine . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10–60
User Watchdog Timer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10–62
Quartus II Software Support . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10–63
ALTREMOTE_UPDATE Megafunction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10–63
Design Security . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10–64
Stratix IV Security Protection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10–65
Security Against Copying . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10–65
Security Against Reverse Engineering . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10–65
Security Against Tampering . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10–65
AES Decryption Block . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10–65
Flexible Security Key Storage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10–65
Stratix IV Design Security Solution . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10–66
Security Modes Available . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10–67
Volatile Key . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10–67
Non-Volatile Key . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10–67
Non-Volatile Key with Tamper Protection Bit Set . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10–67
No Key Operation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10–68
Supported Configuration Schemes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10–68
Chapter 11. SEU Mitigation in Stratix IV Devices
Error Detection Fundamentals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11–2
Configuration Error Detection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11–2
User Mode Error Detection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11–2
Automated Single-Event Upset Detection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11–5
Error Detection Pin Description . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11–5
CRC_ERROR Pin . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11–5
Error Detection Block . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11–6
Error Detection Registers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11–7
Error Detection Timing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11–8
Software Support . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11–10
Recovering From CRC Errors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11–11
Chapter 12. JTAG Boundary-Scan Testing in Stratix IV Devices
BST Architecture . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12–1
BST Operation Control . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12–2
I/O Voltage Support in a JTAG Chain . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12–4
BST Circuitry . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12–4
BSDL Support . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12–4
Chapter 13. Power Management in Stratix IV Devices
Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13–1
Stratix IV Power Technology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13–2
Programmable Power Technology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13–2
Stratix IV External Power Supply Requirements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13–3
Temperature Sensing Diode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13–4
External Pin Connections . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13–4
Additional Information
About this Handbook . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Info–1
How to Contact Altera . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Info–1
Typographic Conventions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Info–1
September 2012
Altera Corporation
Stratix IV Device Handbook
Volume 1
x
Stratix IV Device Handbook
Volume 1
Contents
September 2012 Altera Corporation
Chapter Revision Dates
The chapters in this document, Stratix IV Device Handbook, were revised on the
following dates. Where chapters or groups of chapters are available separately, part
numbers are listed.
Chapter 1.
Overview for the Stratix IV Device Family
Revised:
September 2012
Part Number: SIV51001-3.4
Chapter 2.
Logic Array Blocks and Adaptive Logic Modules in Stratix IV Devices
Revised:
February 2011
Part Number: SIV51002-3.1
Chapter 3.
TriMatrix Embedded Memory Blocks in Stratix IV Devices
Revised:
December 2011
Part Number: SIV51003-3.3
Chapter 4.
DSP Blocks in Stratix IV Devices
Revised:
February 2011
Part Number: SIV51004-3.1
Chapter 5.
Clock Networks and PLLs in Stratix IV Devices
Revised:
September 2012
Part Number: SIV51005-3.4
Chapter 6.
I/O Features in Stratix IV Devices
Revised:
September 2012
Part Number: SIV51006-3.4
Chapter 7.
External Memory Interfaces in Stratix IV Devices
Revised:
February 2011
Part Number: SIV51007-3.2
Chapter 8.
High-Speed Differential I/O Interfaces and DPA in Stratix IV Devices
Revised:
September 2012
Part Number: SIV51008-3.4
Chapter 9.
Hot Socketing and Power-On Reset in Stratix IV Devices
Revised:
February 2011
Part Number: SIV51009-3.2
Chapter 10. Configuration, Design Security, and Remote System Upgrades in Stratix IV Devices
Revised:
September 2012
Part Number: SIV51010-3.5
Chapter 11. SEU Mitigation in Stratix IV Devices
Revised:
February 2011
Part Number: SIV51011-3.2
September 2012
Altera Corporation
Stratix IV Device Handbook
Volume 1
xii
Chapter Revision Dates
Chapter 12. JTAG Boundary-Scan Testing in Stratix IV Devices
Revised:
February 2011
Part Number: SIV51012-3.2
Chapter 13. Power Management in Stratix IV Devices
Revised:
February 2011
Part Number: SIV51013-3.2
Stratix IV Device Handbook
Volume 1
September 2012 Altera Corporation
Section I. Device Core
This section provides a complete overview of all features relating to the Stratix® IV
device family, which is the most architecturally advanced, high-performance,
low-power FPGA in the market place. This section includes the following chapters:
■
Chapter 1, Overview for the Stratix IV Device Family
■
Chapter 2, Logic Array Blocks and Adaptive Logic Modules in Stratix IV Devices
■
Chapter 3, TriMatrix Embedded Memory Blocks in Stratix IV Devices
■
Chapter 4, DSP Blocks in Stratix IV Devices
■
Chapter 5, Clock Networks and PLLs in Stratix IV Devices
Revision History
Refer to each chapter for its own specific revision history. For information on when
each chapter was updated, refer to the Chapter Revision Dates section, which appears
in the full handbook.
September 2012
Altera Corporation
Stratix IV Device Handbook
Volume 1
I–2
Stratix IV Device Handbook
Volume 1
Section I: Device Core
September 2012 Altera Corporation
1. Overview for the Stratix IV Device
Family
September 2012
SIV51001-3.4
SIV51001-3.4
Altera® Stratix® IV FPGAs deliver a breakthrough level of system bandwidth and
power efficiency for high-end applications, allowing you to innovate without
compromise. Stratix IV FPGAs are based on the Taiwan Semiconductor
Manufacturing Company (TSMC) 40-nm process technology and surpass all other
high-end FPGAs, with the highest logic density, most transceivers, and lowest power
requirements.
The Stratix IV device family contains three optimized variants to meet different
application requirements:
■
Stratix IV E (Enhanced) FPGAs—up to 813,050 logic elements (LEs), 33,294 kilobits
(Kb) RAM, and 1,288 18 x 18 bit multipliers
■
Stratix IV GX transceiver FPGAs—up to 531,200 LEs, 27,376 Kb RAM, 1,288
18 x 18-bit multipliers, and 48 full-duplex clock data recovery (CDR)-based
transceivers at up to 8.5 Gbps
■
Stratix IV GT—up to 531,200 LEs, 27,376 Kb RAM, 1,288 18 x 18-bit multipliers,
and 48 full-duplex CDR-based transceivers at up to 11.3 Gbps
The complete Altera high-end solution includes the lowest risk, lowest total cost path
to volume using HardCopy® IV ASICs for all the family variants, a comprehensive
portfolio of application solutions customized for end-markets, and the industry
leading Quartus® II software to increase productivity and performance.
f For information about upcoming Stratix IV device features, refer to the Upcoming
Stratix IV Device Features document.
f For information about changes to the currently published Stratix IV Device Handbook,
refer to the Addendum to the Stratix IV Device Handbook chapter.
This chapter contains the following sections:
■
“Feature Summary” on page 1–2
■
“Architecture Features” on page 1–6
■
“Integrated Software Platform” on page 1–19
■
“Ordering Information” on page 1–19
© 2012 Altera Corporation. All rights reserved. ALTERA, ARRIA, CYCLONE, HARDCOPY, MAX, MEGACORE, NIOS, QUARTUS and STRATIX words and logos
are trademarks of Altera Corporation and registered in the U.S. Patent and Trademark Office and in other countries. All other words and logos identified as
trademarks or service marks are the property of their respective holders as described at www.altera.com/common/legal.html. Altera warrants performance of its
semiconductor products to current specifications in accordance with Altera's standard warranty, but reserves the right to make changes to any products and
services at any time without notice. Altera assumes no responsibility or liability arising out of the application or use of any information, product, or service
described herein except as expressly agreed to in writing by Altera. Altera customers are advised to obtain the latest version of device specifications before relying
on any published information and before placing orders for products or services.
ISO
9001:2008
Registered
Stratix IV Device Handbook
Volume 1
September 2012
Feedback Subscribe
1–2
Chapter 1: Overview for the Stratix IV Device Family
Feature Summary
Feature Summary
The following list summarizes the Stratix IV device family features:
■
Up to 48 full-duplex CDR-based transceivers in Stratix IV GX and GT devices
supporting data rates up to 8.5 Gbps and 11.3 Gbps, respectively
■
Dedicated circuitry to support physical layer functionality for popular serial
protocols, such as PCI Express (PCIe) (PIPE) Gen1 and Gen2, Gbps Ethernet (GbE),
Serial RapidIO, SONET/SDH, XAUI/HiGig, (OIF) CEI-6G, SD/HD/3G-SDI, Fibre
Channel, SFI-5, and Interlaken
■
Complete PCIe protocol solution with embedded PCIe hard IP blocks that
implement PHY-MAC layer, Data Link layer, and Transaction layer functionality
f For more information, refer to the IP Compiler for PCI Express User Guide.
Stratix IV Device Handbook
Volume 1
■
Programmable transmitter pre-emphasis and receiver equalization circuitry to
compensate for frequency-dependent losses in the physical medium
■
Typical physical medium attachment (PMA) power consumption of 100 mW at
3.125 Gbps and 135 mW at 6.375 Gbps per channel
■
72,600 to 813,050 equivalent LEs per device
■
7,370 to 33,294 Kb of enhanced TriMatrix memory consisting of three RAM block
sizes to implement true dual-port memory and FIFO buffers
■
High-speed digital signal processing (DSP) blocks configurable as 9 x 9-bit,
12 x 12-bit, 18 x 18-bit, and 36 x 36-bit full-precision multipliers at up to 600 MHz
■
Up to 16 global clocks (GCLK), 88 regional clocks (RCLK), and 132 periphery
clocks (PCLK) per device
■
Programmable power technology that minimizes power while maximizing device
performance
■
Up to 1,120 user I/O pins arranged in 24 modular I/O banks that support a wide
range of single-ended and differential I/O standards
■
Support for high-speed external memory interfaces including DDR, DDR2,
DDR3 SDRAM, RLDRAM II, QDR II, and QDR II+ SRAM on up to 24 modular
I/O banks
■
High-speed LVDS I/O support with serializer/deserializer (SERDES), dynamic
phase alignment (DPA), and soft-CDR circuitry at data rates up to 1.6 Gbps
■
Support for source-synchronous bus standards, including SGMII, GbE, SPI-4
Phase 2 (POS-PHY Level 4), SFI-4.1, XSBI, UTOPIA IV, NPSI, and CSIX-L1
■
Pinouts for Stratix IV E devices designed to allow migration of designs from
Stratix III to Stratix IV E with minimal PCB impact
September 2012 Altera Corporation
Chapter 1: Overview for the Stratix IV Device Family
Feature Summary
1–3
Stratix IV GX Devices
Stratix IV GX devices provide up to 48 full-duplex CDR-based transceiver channels
per device:
■
Thirty-two out of the 48 transceiver channels have dedicated physical coding
sublayer (PCS) and physical medium attachment (PMA) circuitry and support
data rates between 600 Mbps and 8.5 Gbps
■
The remaining 16 transceiver channels have dedicated PMA-only circuitry and
support data rates between 600 Mbps and 6.5 Gbps
1
The actual number of transceiver channels per device varies with device selection. For
more information about the exact transceiver count in each device, refer to Table 1–1
on page 1–11.
1
For more information about transceiver architecture, refer to the Transceiver
Architecture in Stratix IV Devices chapter.
Figure 1–1 shows a high-level Stratix IV GX chip view.
Figure 1–1. Stratix IV GX Chip View (1)
PLL
General Purpose
I/O and Memory
Interface
PLL
PLL
General Purpose
I/O and Memory
Interface
Transceiver Block
General Purpose I/O and
High-Speed LVDS I/O
with DPA and Soft CDR
PLL
PLL
PLL
PCI Express
Hard IP Block
PLL
PCI Express
Hard IP Block
General Purpose
I/O and
High-Speed
LVDS I/O with
DPA and Soft CDR
PLL
FPGA Fabric
(Logic Elements, DSP,
Embedded Memory,
Clock Networks)
General Purpose
I/O and
High-Speed
LVDS I/O with
DPA and Soft CDR
PLL
General Purpose
I/O and
High-Speed
LVDS I/O with
DPA and Soft CDR
General Purpose
I/O and
High-Speed
LVDS I/O with
DPA and Soft CDR
PCI Express
Hard IP Block
PLL
Transceiver Transceiver Transceiver Transceiver
Block
Block
Block
Block
PLL
PLL
PCI Express
Hard IP Block
Transceiver Transceiver Transceiver Transceiver
Block
Block
Block
Block
General Purpose
I/O and Memory
Interface
General Purpose
I/O and Memory
Interface
600 Mbps-8.5 Gbps CDR-based Transceiver
General Purpose I/O and 150 Mbps-1.6 Gbps
LVDS interface with DPA and Soft-CDR
Note to Figure 1–1:
(1) Resource counts vary with device selection, package selection, or both.
September 2012
Altera Corporation
Stratix IV Device Handbook
Volume 1
1–4
Chapter 1: Overview for the Stratix IV Device Family
Feature Summary
Stratix IV E Device
Stratix IV E devices provide an excellent solution for applications that do not require
high-speed CDR-based transceivers, but are logic, user I/O, or memory intensive.
Figure 1–2 shows a high-level Stratix IV E chip view.
Figure 1–2. Stratix IV E Chip View (1)
General Purpose
I/O and Memory PLL
Interface
PLL
General Purpose
I/O and Memory
Interface
PLL
PLL
General
Purpose
I/O and
High-Speed
LVDS I/O
with DPA
and Soft-CDR
General
Purpose
I/O and
High-Speed
LVDS I/O
with DPA
and Soft-CDR
PLL
PLL
FPGA Fabric
PLL
(Logic Elements, DSP,
Embedded Memory,
Clock Networks)
PLL
General
Purpose
I/O and
High-Speed
LVDS I/O
with DPA
and Soft-CDR
General
Purpose
I/O and
High-Speed
LVDS I/O
with DPA
and Soft-CDR
PLL
PLL
General Purpose
I/O and Memory PLL
Interface
General Purpose I/O and
High-Speed LVDS I/O with DPA
and Soft-CDR
PLL
General Purpose
I/O and Memory
Interface
General Purpose I/O and
150 Mbps-1.6 Gbps
LVDS interface with DPA and Soft-CDR
Note to Figure 1–2:
(1) Resource counts vary with device selection, package selection, or both.
Stratix IV Device Handbook
Volume 1
September 2012 Altera Corporation
Chapter 1: Overview for the Stratix IV Device Family
Feature Summary
1–5
Stratix IV GT Devices
Stratix IV GT devices provide up to 48 CDR-based transceiver channels per device:
■
Thirty-two out of the 48 transceiver channels have dedicated PCS and PMA
circuitry and support data rates between 600 Mbps and 11.3 Gbps
■
The remaining 16 transceiver channels have dedicated PMA-only circuitry and
support data rates between 600 Mbps and 6.5 Gbps
1
The actual number of transceiver channels per device varies with device selection. For
more information about the exact transceiver count in each device, refer to Table 1–7
on page 1–16.
1
For more information about Stratix IV GT devices and transceiver architecture, refer
to the Transceiver Architecture in Stratix IV Devices chapter.
Figure 1–3 shows a high-level Stratix IV GT chip view.
(1)
FPGA Fabric
PLL
(Logic Elements, DSP,
Embedded Memory,
Clock Networks)
PLL
General Purpose
I/O and
High-Speed
LVDS I/O with
DPA and Soft CDR
General Purpose
I/O and
High-Speed
LVDS I/O with
DPA and Soft CDR
PLL
General Purpose
I/O and
High-Speed
LVDS I/O with
DPA and Soft CDR
PLL
General Purpose
I/O and
High-Speed
LVDS I/O with
DPA and Soft CDR
PCI Express
Hard IP Block
PLL
Transceiver Transceiver
Block
Block
General Purpose
I/O and Memory
Interface
Transceiver Transceiver
Block
Block
PLL
PCI Express
Hard IP Block
PLL
PLL
PCI Express
Hard IP Block
Transceiver Transceiver
Block
Block
Transceiver Transceiver
Block
Block
General Purpose
I/O and Memory
Interface
PCI Express
Hard IP Block
Figure 1–3. Stratix IV GT Chip View
PLL
General Purpose
I/O and Memory
Interface
Transceiver Block
General Purpose I/O and
High-Speed LVDS I/O
with DPA and Soft CDR
PLL
PLL
PLL
General Purpose
I/O and Memory
Interface
600 Mbps-11.3 Gbps CDR-based Transceiver
General Purpose I/O and up to 1.6 Gbps
LVDS interface with DPA and Soft-CDR
Note to Figure 1–3:
(1) Resource counts vary with device selection, package selection, or both.
September 2012
Altera Corporation
Stratix IV Device Handbook
Volume 1
1–6
Chapter 1: Overview for the Stratix IV Device Family
Architecture Features
Architecture Features
The Stratix IV device family features are divided into high-speed transceiver features
and FPGA fabric and I/O features.
1
The high-speed transceiver features apply only to Stratix IV GX and Stratix IV GT
devices.
High-Speed Transceiver Features
The following sections describe high-speed transceiver features for Stratix IV GX and
GT devices.
Highest Aggregate Data Bandwidth
Up to 48 full-duplex transceiver channels supporting data rates up to 8.5 Gbps in
Stratix IV GX devices and up to 11.3 Gbps in Stratix IV GT devices.
Wide Range of Protocol Support
Physical layer support for the following serial protocols:
■
Stratix IV GX—PCIe Gen1 and Gen2, GbE, Serial RapidIO, SONET/SDH,
XAUI/HiGig, (OIF) CEI-6G, SD/HD/3G-SDI, Fibre Channel, SFI-5, GPON,
SAS/SATA, HyperTransport 1.0 and 3.0, and Interlaken
■
Stratix IV GT—40G/100G Ethernet, SFI-S, Interlaken, SFI-5.1, Serial RapidIO,
SONET/SDH, XAUI/HiGig, (OIF) CEI-6G, 3G-SDI, and Fibre Channel
■
Extremely flexible and easy-to-configure transceiver data path to implement
proprietary protocols
■
PCIe Support
■
Complete PCIe Gen1 and Gen2 protocol stack solution compliant to PCI
Express base specification 2.0 that includes PHY-MAC, Data Link, and
transaction layer circuitry embedded in PCI Express hard IP blocks
f For more information, refer to the PCI Express Compiler User Guide.
Stratix IV Device Handbook
Volume 1
■
Root complex and end-point applications
■
x1, x4, and x8 lane configurations
■
PIPE 2.0-compliant interface
■
Embedded circuitry to switch between Gen1 and Gen2 data rates
■
Built-in circuitry for electrical idle generation and detection, receiver detect,
power state transitions, lane reversal, and polarity inversion
■
8B/10B encoder and decoder, receiver synchronization state machine, and
± 300 parts per million (ppm) clock compensation circuitry
■
Transaction layer support for up to two virtual channels (VCs)
September 2012 Altera Corporation
Chapter 1: Overview for the Stratix IV Device Family
Architecture Features
■
■
■
1–7
XAUI/HiGig Support
■
Compliant to IEEE802.3ae specification
■
Embedded state machine circuitry to convert XGMII idle code groups (||I||)
to and from idle ordered sets (||A||, ||K||, ||R||) at the transmitter and
receiver, respectively
■
8B/10B encoder and decoder, receiver synchronization state machine, lane
deskew, and ± 100 ppm clock compensation circuitry
GbE Support
■
Compliant to IEEE802.3-2005 specification
■
Automatic idle ordered set (/I1/, /I2/) generation at the transmitter,
depending on the current running disparity
■
8B/10B encoder and decoder, receiver synchronization state machine, and
± 100 ppm clock compensation circuitry
Support for other protocol features such as MSB-to-LSB transmission in
SONET/SDH configuration and spread-spectrum clocking in PCIe configurations
Diagnostic Features
■
Serial loopback from the transmitter serializer to the receiver CDR for transceiver
PCS and PMA diagnostics
■
Reverse serial loopback pre- and post-CDR to transmitter buffer for physical link
diagnostics
■
Loopback master and slave capability in PCI Express hard IP blocks
f For more information, refer to the PCI Express Compiler User Guide.
Signal Integrity
Stratix IV devices simplify the challenge of signal integrity through a number of chip,
package, and board-level enhancements to enable efficient high-speed data transfer
into and out of the device. These enhancements include:
September 2012
■
Programmable 3-tap transmitter pre-emphasis with up to 8,192 pre-emphasis
levels to compensate for pre-cursor and post-cursor inter-symbol interference (ISI)
■
Up to 900% boost capability on the first pre-emphasis post-tap
■
User-controlled and adaptive 4-stage receiver equalization with up to 16 dB of
high-frequency gain
■
On-die power supply regulators for transmitter and receiver phase-locked loop
(PLL) charge pump and voltage controlled oscillator (VCO) for superior noise
immunity
■
On-package and on-chip power supply decoupling to satisfy transient current
requirements at higher frequencies, thereby reducing the need for on-board
decoupling capacitors
■
Calibration circuitry for transmitter and receiver on-chip termination (OCT)
resistors
Altera Corporation
Stratix IV Device Handbook
Volume 1
1–8
Chapter 1: Overview for the Stratix IV Device Family
Architecture Features
FPGA Fabric and I/O Features
The following sections describe the Stratix IV FPGA fabric and I/O features.
Device Core Features
■
Up to 531,200 LEs in Stratix IV GX and GT devices and up to 813,050 LEs in
Stratix IV E devices, efficiently packed in unique and innovative adaptive logic
modules (ALMs)
■
Ten ALMs per logic array block (LAB) deliver faster performance, improved logic
utilization, and optimized routing
■
Programmable power technology, including a variety of process, circuit, and
architecture optimizations and innovations
■
Programmable power technology available to select power-driven compilation
options for reduced static power consumption
Embedded Memory
■
TriMatrix embedded memory architecture provides three different memory block
sizes to efficiently address the needs of diversified FPGA designs:
■
640-bit MLAB
■
9-Kb M9K
■
144-Kb M144K
■
Up to 33,294 Kb of embedded memory operating at up to 600 MHz
■
Each memory block is independently configurable to be a single- or dual-port
RAM, FIFO, ROM, or shift register
Digital Signal Processing (DSP) Blocks
■
Flexible DSP blocks configurable as 9 x 9-bit, 12 x 12-bit, 18 x 18-bit, and 36 x 36-bit
full-precision multipliers at up to 600 MHz with rounding and saturation
capabilities
■
Faster operation due to fully pipelined architecture and built-in addition,
subtraction, and accumulation units to combine multiplication results
■
Optimally designed to support advanced features such as adaptive filtering, barrel
shifters, and finite and infinite impulse response (FIR and IIR) filters
Clock Networks
Stratix IV Device Handbook
Volume 1
■
Up to 16 global clocks and 88 regional clocks optimally routed to meet the
maximum performance of 800 MHz
■
Up to 112 and 132 periphery clocks in Stratix IV GX and Stratix IV E devices,
respectively
■
Up to 66 (16 GCLK + 22 RCLK + 28 PCLK) clock networks per device quadrant in
Stratix IV GX and Stratix IV GT devices
■
Up to 71 (16 GCLK + 22 RCLK + 33 PCLK) clock networks per device quadrant in
Stratix IV E devices
September 2012 Altera Corporation
Chapter 1: Overview for the Stratix IV Device Family
Architecture Features
1–9
PLLs
■
Three to 12 PLLs per device supporting spread-spectrum input tracking,
programmable bandwidth, clock switchover, dynamic reconfiguration, and delay
compensation
■
On-chip PLL power supply regulators to minimize noise coupling
I/O Features
■
Sixteen to 24 modular I/O banks per device with 24 to 48 I/Os per bank designed
and packaged for optimal simultaneous switching noise (SSN) performance and
migration capability
■
Support for a wide range of industry I/O standards, including single-ended
(LVTTL/CMOS/PCI/PCIX), differential (LVDS/mini-LVDS/RSDS),
voltage-referenced single-ended and differential (SSTL/HSTL Class I/II) I/O
standards
■
On-chip series (RS) and on-chip parallel (RT) termination with auto-calibration for
single-ended I/Os and on-chip differential (RD) termination for differential I/Os
■
Programmable output drive strength, slew rate control, bus hold, and weak
pull-up capability for single-ended I/Os
■
User I/O:GND:VCC ratio of 8:1:1 to reduce loop inductance in the package—PCB
interface
■
Programmable transmitter differential output voltage (VOD) and pre-emphasis for
high-speed LVDS I/O
High-Speed Differential I/O with DPA and Soft-CDR
■
Dedicated circuitry on the left and right sides of the device to support differential
links at data rates from 150 Mbps to 1.6 Gbps
■
Up to 98 differential SERDES in Stratix IV GX devices, up to 132 differential
SERDES in Stratix IV E devices, and up to 47 differential SERDES in Stratix IV GT
devices
■
DPA circuitry at the receiver automatically compensates for channel-to-channel
and channel-to-clock skew in source synchronous interfaces
■
Soft-CDR circuitry at the receiver allows implementation of asynchronous serial
interfaces with embedded clocks at up to 1.6 Gbps data rate (SGMII and GbE)
External Memory Interfaces
September 2012
■
Support for existing and emerging memory interface standards such as DDR
SDRAM, DDR2 SDRAM, DDR3 SDRAM, QDRII SRAM, QDRII+ SRAM, and
RLDRAM II
■
DDR3 up to 1,067 Mbps/533 MHz
■
Programmable DQ group widths of 4 to 36 bits (includes parity bits)
■
Dynamic OCT, trace mismatch compensation, read-write leveling, and half-rate
register capabilities provide a robust external memory interface solution
Altera Corporation
Stratix IV Device Handbook
Volume 1
1–10
Chapter 1: Overview for the Stratix IV Device Family
Architecture Features
System Integration
■
All Stratix IV devices support hot socketing
■
Four configuration modes:
■
Passive Serial (PS)
■
Fast Passive Parallel (FPP)
■
Fast Active Serial (FAS)
■
JTAG configuration
■
Ability to perform remote system upgrades
■
256-bit advanced encryption standard (AES) encryption of configuration bits
protects your design against copying, reverse engineering, and tampering
■
Built-in soft error detection for configuration RAM cells
f For more information about how to connect the PLL, external memory interfaces, I/O,
high-speed differential I/O, power, and the JTAG pins to PCB, refer to the
Stratix IV GX and Stratix IV E Device Family Pin Connection Guidelines and the
Stratix IV GT Device Family Pin Connection Guidelines.
Stratix IV Device Handbook
Volume 1
September 2012 Altera Corporation
Table 1–1. Stratix IV GX Device Features (Part 1 of 2)
F1932
F1760
F1932
EP4SGX530
F1760
F1517
F1152
F780
F1932
EP4SGX360
F1760
F1517
F780
F1152
EP4SGX290
F1517
F780
F1152
EP4SGX230
F1517
F780
F1152
EP4SGX180
F1152
EP4SGX110
F780
F1152
Altera Corporation
Package
Option
EP4SGX70
F780
Feature
ALMs
29,040
42,240
70,300
91,200
116,480
141,440
212,480
LEs
72,600
105,600
175,750
228,000
291,200
353,600
531,200
0.6 Gbps8.5 Gbps
Transceivers
(PMA + PCS)
—
16
—
—
16
—
—
16
24
—
—
16
24
—
—
16
24
24
32
—
—
16
24
24
32
24
32
8
—
8
16
—
8
16
—
—
8
16
—
—
16
16
—
—
—
—
16
16
—
—
—
—
—
—
PMA-only
CMU
Channels
(0.6 Gbps6.5 Gbps)
—
8
—
—
8
—
—
8
12
—
—
8
12
—
—
8
12
12
16
—
—
8
12
12
16
12
16
PCI Express
hard IP
Blocks
1
2
1
28
56
28
(1)
0.6 Gbps6.5 Gbps
Transceivers
(PMA + PCS)
Chapter 1: Overview for the Stratix IV Device Family
Architecture Features
September 2012
Table 1–1 lists the Stratix IV GX device features.
(1)
High-Speed
LVDS
SERDES (up
to 1.6 Gbps)
2
28
1
56
2
1
2
2
28
44
88
28
44
88
—
44
1
2
4
1
2
4
—
2
4
88
88
98
2
—
44
—
2
4
4
88
88
98
88
98
(4)
1
1
4
4
4
1–11
Stratix IV Device Handbook
Volume 1
SPI-4.2 Links
F1932
F1760
F1932
EP4SGX530
F1760
F1517
F1152
F780
F1932
EP4SGX360
F1760
F1517
F780
F1152
EP4SGX290
F1517
F780
F1152
EP4SGX230
F1517
F1152
EP4SGX180
F780
F780
F1152
EP4SGX110
F1152
Package
Option
EP4SGX70
F780
Feature
1–12
Stratix IV Device Handbook
Volume 1
Table 1–1. Stratix IV GX Device Features (Part 2 of 2)
M9K Blocks
(256 x
36 bits)
462
660
950
1,235
936
1,248
1,280
M144K
Blocks
(2048 x
72 bits)
16
16
20
22
36
48
64
7,370
9,564
13,627
17,133
17,248
22,564
27,376
384
512
920
1,288
832
Total Memory
(MLAB+M9K
+M144K) Kb
Embedded
Multipliers
18 x 18 (2)
PLLs
4
3
(3)
372
488
372
372
48
8
372
56
4
Speed Grade
(fastest to
slowest) (5)
–2,
–3,
–4
–2,
–3,
–4
–2
,
–3,
–4
–2
,
–3,
–4
–2,
–3,
–4
–2
,
–3,
–4
–2
,
–3,
–4
User I/Os
4
3
6
8
3
6
8
4
56
4
74
4
372
564
74
4
289
564
56
4
–2
,
–3
,
–4
–2,
–3,
–4
–2
,
–3,
–4
–2
–2
–2, –2,
,
,
–3, –3,
–3,
–3,
–4 –4
–4
–4
–2
,
–3,
–4
56
4
6
8
12
12
4
6
74
4
88
0
92
0
289
564
–2, –2, –2, –2,
–3, –3, –3, –3,
–4 –4 –4 –4
–2
,
–3,
–4
1,024
8
12
12
12
12
74
4
88
0
920
880
920
–2
–2, –2, –2,
,
–3, –3, –3,
–3,
–4 –4 –4
–4
–2,
–3,
–4
–2, –3,
–4
–2, –3,
–4
56
4
Notes to Table 1–1:
(1) The total number of transceivers is divided equally between the left and right side of each device, except for the devices in the F780 package. These devices have eight transceiver channels located only
on the right side of the device.
September 2012 Altera Corporation
(2) Four multiplier adder mode.
(3) The user I/Os count from pin-out files includes all general purpose I/O, dedicated clock pins, and dual purpose configuration pins. Transceiver pins and dedicated configuration pins are not included in
the pin count.
(4) Total pairs of high-speed LVDS SERDES take the lowest channel count of RX/TX.
(5) The difference between the Stratix IV GX devices in the –2 and –2x speed grades is the number of available transceiver channels. The –2 device allows you to use the transceiver CMU blocks as
transceiver channels. The –2x device does NOT allow you to use the CMU blocks as transceiver channels. In addition to the reduction of available transceiver channels in the Stratix IV GX –2x device,
the data rates in the –2x device are limited to 6.5 Gbps.
Chapter 1: Overview for the Stratix IV Device Family
Architecture Features
3
1,02
4
1,040
Table 1–2. Stratix IV GX Device Package Options
Altera Corporation
F780
(29 mm x 29 mm)
Device
(1), (2)
F1152
(35 mm x 35 mm)
(6)
(6)
F1152
(35 mm x 35 mm)
(5), (7)
F1517
(40 mm x 40 mm)
(5), (7)
F1760
(42.5 mm x 42.5 mm)
F1932
(45 mm x 45 mm)
(7)
(7)
EP4SGX70
DF29
—
—
HF35
—
—
—
—
EP4SGX110
DF29
—
FF35
HF35
—
—
—
—
EP4SGX180
DF29
—
FF35
—
HF35
KF40
—
—
EP4SGX230
DF29
—
EP4SGX290
—
FF35
—
HF35
KF40
—
—
FH29
(3)
FF35
—
HF35
KF40
KF43
NF45
(3)
FF35
—
—
—
EP4SGX360
—
FH29
EP4SGX530
—
—
HF35
HH35
KF40
(4)
KH40
(4)
KF43
NF45
KF43
NF45
Chapter 1: Overview for the Stratix IV Device Family
Architecture Features
September 2012
Table 1–2 lists the Stratix IV GX device package options.
Notes to Table 1–2:
(1) Device packages in the same column and marked under the same arrow sign have vertical migration capability.
(2) Use the Pin Migration Viewer in the Pin Planner to verify the pin migration compatibility when migrating devices. For more information, refer to I/O Management in the Quartus II Handbook, Volume 2.
(3) The 780-pin EP4SGX290 and EP4SGX360 devices are available only in 33 mm x 33 mm Hybrid flip chip package.
(4) The 1152-pin and 1517-pin EP4SGX530 devices are available only in 42.5 mm x 42.5 mm Hybrid flip chip packages.
(5) When migrating between hybrid and flip chip packages, there is an additional keep-out area. For more information, refer to the Package Information Datasheet for Altera Devices.
(6) Devices listed in this column are available in –2x, –3, and –4 speed grades. These devices do not have on-package decoupling capacitors.
(7) Devices listed in this column are available in –2, –3, and –4 speed grades. These devices have on-package decoupling capacitors. For more information about on-package decoupling capacitor value
in each device, refer to Table 1–3.
1
On-package decoupling reduces the need for on-board or PCB decoupling capacitors by satisfying the transient current
requirements at higher frequencies. The Power Delivery Network design tool for Stratix IV devices accounts for the on-package
decoupling and reflects the reduced requirements for PCB decoupling capacitors.
1–13
Stratix IV Device Handbook
Volume 1
1–14
Stratix IV Device Handbook
Volume 1
Table 1–3 lists the Stratix IV GX device on-package decoupling information.
Table 1–3. Stratix IV GX Device On-Package Decoupling Information (1)
Ordering Information
EP4SGX70
EP4SGX110
EP4SGX180
EP4SGX230
HF35
HF35
HF35
KF40
HF35
KF40
VCC
VCCL_GXB
VCCA_L/R
VCCT and VCCR (Shared)
10nF per bank
(2)
100nF per transceiver block
100nF
1470nF + 147nF per side
21uF + 2470nF
10nF per bank
(2)
100nF per transceiver block
100nF
1470nF + 147nF per side
21uF + 2470nF
10nF per bank
(2)
100nF per transceiver block
100nF
1470nF + 147nF per side
21 uF + 2470 nF
10 nF per bank
(2)
100 nF per transceiver block
100 nF
1470 nF + 147 nF
per side
41 uF + 4470 nF
10 nF per bank
(2)
100 nF per transceiver block
100nF
1470 nF + 147 nF
per side
41 uF + 4470 nF
10 nF per bank
(2)
100 nF per transceiver block
100 nF
1470 nF + 147 nF
per side
41 uF + 4470 nF
10 nF per bank
(2)
100 nF per transceiver block
100 nF
1470 nF + 147 nF
per side
21uF + 2470nF
VCCIO
HF35
EP4SGX290
KF40
KF43
NF45
HF35
EP4SGX360
KF40
KF43
NF45
HH35
KH40
KF43
NF45
Notes to Table 1–3:
September 2012 Altera Corporation
(1) Table 1–3 refers to production devices on-package decoupling. For more information about decoupling design of engineering sample (ES) devices, contact Altera Technical Support.
(2) For I/O banks 3(*), 4(*), 7(*), and 8(*) only. There is no OPD for I/O bank 1(*), 2(*), 5(*), and 6(*).
Chapter 1: Overview for the Stratix IV Device Family
Architecture Features
EP4SGX530
Table 1–4. Stratix IV E Device Features
Feature
EP4SE230
Altera Corporation
Package Pin Count
780
EP4SE360
780
EP4SE530
1152
1152
EP4SE820
1517
1760
1152
1517
ALMs
91,200
141,440
212,480
325,220
LEs
228,000
353,600
531,200
813,050
High-Speed LVDS
SERDES (up to
1.6 Gbps) (1)
56
56
88
88
SPI-4.2 Links
3
3
4
4
M9K Blocks
(256 x 36 bits)
1,235
1,248
1,280
1610
M144K Blocks
(2048 x 72 bits)
22
48
64
60
Total Memory
(MLAB+M9K+
M144K) Kb
17,133
22,564
27,376
33,294
Embedded Multipliers
(18 x 18) (2)
1,288
1,040
1,024
960
PLLs
User I/Os
112
6
88
112
132
4
6
6
4
4
8
8
12
12
8
488
488
744
744
976
976
744
–2, –3, –4
–2, –3, –4
–2, –3, –4
–2, –3, –4
–2, –3, –4
–2, –3, –4
(3)
Speed Grade
(fastest to slowest)
112
1760
12
(4)
–3, –4
976
Chapter 1: Overview for the Stratix IV Device Family
Architecture Features
September 2012
Table 1–4 lists the Stratix IV E device features.
12
(4)
–3, –4
1120
(4)
–3, –4
Notes to Table 1–4:
(1) The user I/O count from the pin-out files include all general purpose I/Os, dedicated clock pins, and dual purpose configuration pins. Transceiver pins
and dedicated configuration pins are not included in the pin count.
(2) Four multiplier adder mode.
(4) This data is preliminary.
1–15
Stratix IV Device Handbook
Volume 1
(3) Total pairs of high-speed LVDS SERDES take the lowest channel count of RX/TX.
1–16
Chapter 1: Overview for the Stratix IV Device Family
Architecture Features
Table 1–5 summarizes the Stratix IV E device package options.
Table 1–5. Stratix IV E Device Package Options (1),
F780
(29 mm x 29 mm)
Device
EP4SE230
(5), (6)
(2)
F1152
(35 mm x 35 mm)
F29
EP4SE360
H29
(3)
F1517
(40 mm x 40 mm)
(5), (7)
(7)
F1760
(42.5 mm x 42.5 mm)
—
—
—
F35
—
—
EP4SE530
—
H35
(4)
H40
(4)
F43
EP4SE820
—
H35
(4)
H40
(4)
F43
(7)
Notes to Table 1–5:
(1) Device packages in the same column and marked under the same arrow sign have vertical migration capability.
(2) Use the Pin Migration Viewer in the Pin Planner to verify the pin migration compatibility when migrating devices. For more information, refer to
I/O Management in the Quartus II Handbook, Volume 2.
(3) The 780-pin EP4SE360 device is available only in the 33 mm x 33 mm Hybrid flip chip package.
(4) The 1152-pin and 1517-pin for EP4SE530 and EP4SE820 devices are available only in the 42.5 mm x 42.5 mm Hybrid flip chip package.
(5) When migrating between hybrid and flip chip packages, there is an additional keep-out area. For more information, refer to the Package
Information Datasheet for Altera Devices.
(6) Devices listed in this column do not have on-package decoupling capacitors.
(7) Devices listed in this column have on-package decoupling capacitors. For more information about on-package decoupling capacitor value for
each device, refer to Table 1–6.
Table 1–6 lists the Stratix IV E on-package decoupling information.
Table 1–6. Stratix IV E Device On-Package Decoupling Information (1)
Ordering Information
EP4SE360
F35
VCC
VCCIO
41 uF + 4470 nF
10 nF per bank
41 uF + 4470 nF
10 nF per bank
41 uF + 4470 nF
10 nF per bank
H35
EP4SE530
H40
F43
H35
EP4SE820
H40
F43
Note to Table 1–6:
(1) Table 1–6 refers to production devices on-package decoupling. For more information about decoupling design of engineering sample (ES)
devices, contact Altera Technical Support.
Table 1–7 lists the Stratix IV GT device features.
Table 1–7. Stratix IV GT Device Features (Part 1 of 2)
Feature
EP4S40G5
EP4S100G2
EP4S100G3
EP4S100G4
1517
1517
1517
1932
1932
ALMs
91,200
212,480
91,200
116,480
141,440
212,480
LEs
228,000
531,200
228,000
291,200
353,600
531,200
36
36
36
48
48
Package Pin Count
Total Transceiver
Channels
Stratix IV Device Handbook
Volume 1
EP4S40G2
EP4S100G5
1517
36
1932
48
September 2012 Altera Corporation
Chapter 1: Overview for the Stratix IV Device Family
Architecture Features
1–17
Table 1–7. Stratix IV GT Device Features (Part 2 of 2)
Feature
EP4S40G2
EP4S40G5
EP4S100G2
EP4S100G3
EP4S100G4
EP4S100G5
10G Transceiver
Channels
(600 Mbps - 11.3 Gbps
with PMA + PCS)
12
12
24
24
24
24
32
8G Transceiver
Channels
(600 Mbps - 8.5 Gbps
with PMA + PCS) (1)
12
12
0
8
8
0
0
PMA-only CMU
Channels
(600 Mbps- 6.5 Gbps)
12
12
12
16
16
12
16
PCIe hard IP Blocks
2
2
2
4
4
2
4
High-Speed LVDS
SERDES
(up to 1.6 Gbps) (2)
46
46
46
47
47
46
47
SP1-4.2 Links
2
2
2
2
2
2
2
M9K Blocks
(256 x 36 bits)
1,235
1,280
1,235
936
1,248
1,280
M144K Blocks
(2048 x 72 bits)
22
64
22
36
48
64
Total Memory (MLAB +
M9K + M144K) Kb
17,133
27,376
17,133
17,248
22,564
27,376
Embedded Multipliers
18 x 18 (3)
1,288
1,024
1,288
832
1,024
1,024
8
8
8
12
12
8
12
654
654
654
781
781
654
781
–1, –2, –3
–1, –2, –3
–1, –2, –3
–1, –2, –3
–1, –2, –3
–1, –2, –3
–1, –2, –3
PLLs
User I/Os
(4), (5)
Speed Grade
(fastest to slowest)
Notes to Table 1–7:
(1) You can configure all 10G transceiver channels as 8G transceiver channels. For example, the EP4S40G2F40 device has twenty-four 8G
transceiver channels and the EP4S100G5F45 device has thirty-two 8G transceiver channels.
(2) Total pairs of high-speed LVDS SERDES take the lowest channel count of RX/TX.
(3) Four multiplier adder mode.
(4) The user I/O count from the pin-out files include all general purpose I/Os, dedicated clock pins, and dual purpose configuration pins. Transceiver
pins and dedicated configuration pins are not included in the pin count.
(5) This data is preliminary.
September 2012
Altera Corporation
Stratix IV Device Handbook
Volume 1
1–18
Chapter 1: Overview for the Stratix IV Device Family
Architecture Features
Table 1–8 lists the resource counts for the Stratix IV GT devices.
Table 1–8. Stratix IV GT Device Package Options
(1), (2)
1517 Pin
(40 mm x 40 mm)
Device
1932 Pin
(45 mm x 45 mm)
(3)
Stratix IV GT 40 G Devices
EP4S40G2
F40
EP4S40G5
—
(4), (5)
H40
—
Stratix IV GT 100 G Devices
EP4S100G2
F40
—
EP4S100G3
—
F45
EP4S100G4
—
F45
EP4S100G5
H40
(4), (5)
F45
Notes to Table 1–8:
(1) This table represents pin compatability; however, it does not include hard IP block placement compatability.
(2) Devices under the same arrow sign have vertical migration capability.
(3) When migrating between hybrid and flip chip packages, there is an additional keep-out area. For more information,
refer to the Altera Device Package Information Data Sheet.
(4) EP4S40G5 and EP4S100G5 devices with 1517 pin-count are only available in 42.5-mm x 42.5-mm Hybrid flip chip
packages.
(5) If you are using the hard IP block, migration is not possible.
Table 1–9 lists the Stratix IV GT on-package decoupling information.
Table 1–9. Stratix IV GT Device On-Package Decoupling Information (1)
Ordering
Information
EP4S40G2F40
EP4S100G2F40
VCC
VCCIO
VCCL_GXB
VCCA_L/R
VCCT_L/R
VCCR_L/R
21 uF + 2470 nF
10 nF per bank
(2)
100 nF per
transceiver block
100 nF
100 nF
100 nF
41 uF + 4470 nF
10 nF per bank
(2)
100 nF per
transceiver block
100 nF
100 nF
100 nF
EP4S100G3F45
EP4S100G4F45
EP4S40G5H40
EP4S100G5H40
EP4S100G5F45
Notes to Table 1–9:
(1) Table 1–9 refers to production devices on-package decoupling. For more information about decoupling design of engineering sample (ES)
devices, contact Altera Technical Support.
(2) For I/O banks 3(*), 4(*), 7(*), and 8(*) only. There is no OPD for I/O bank 1(*), 2(*), 5(*), and 6(*).
Stratix IV Device Handbook
Volume 1
September 2012 Altera Corporation
Chapter 1: Overview for the Stratix IV Device Family
Integrated Software Platform
1–19
Integrated Software Platform
The Quartus II software provides an integrated environment for HDL and schematic
design entry, compilation and logic synthesis, full simulation and advanced timing
analysis, SignalTap II Logic Analyzer, and device configuration of Stratix IV designs.
The Quartus II software provides the MegaWizard Plug-In Manager user interface to
generate different functional blocks, such as memory, PLL, and digital signal
processing logic. For transceivers, the Quartus II software provides the ALTGX
MegaWizard Plug-In Manager interface that guides you through configuration of the
transceiver based on your application requirements.
The Stratix IV GX and GT transceivers allow you to implement low-power and
reliable high-speed serial interface applications with its fully reconfigurable
hardware, optimal signal integrity, and integrated Quartus II software platform.
f For more information about the Quartus II software features, refer to the Quartus II
Handbook.
Ordering Information
This section describes the Stratix IV E, GT, and GX devices ordering information.
Figure 1–4 shows the ordering codes for Stratix IV GX and E devices.
Figure 1–4. Stratix IV GX and E Device Packaging Ordering Information
EP4SGX
230
K
F
40
C
Family Signature
2
ES
Optional Suffix
Indicates specific device options
ES: Engineering sample
N: Lead-free devices
EP4SGX: Stratix IV Transceiver
EP4SE: Stratix IV Logic/Memory
Device Density
Speed Grade
70
110
180
230
290
360
530
820
D: 8
F: 16
H: 24
K: 36
N: 48
Package Type
Operating Temperature
C: Commercial Temperature (tJ=0° C to 85° C)
I: Industrial Temperature (tJ=–40° C to 100° C)
M: Military Temperature (tJ=–55° C to 125° C)
Ball Array Dimension
F: FineLine BGA (FBGA)
H: Hybrid FineLine BGA
September 2012
2, 2x, 3, or 4, with 2 being the fastest
Transceiver Count
Altera Corporation
Corresponds to pin count
29 = 780 pins
35 = 1152 pins
40 = 1517 pins
43 = 1760 pins
45 = 1932 pins
Stratix IV Device Handbook
Volume 1
1–20
Chapter 1: Overview for the Stratix IV Device Family
Ordering Information
Figure 1–5 shows the ordering codes for Stratix IV GT devices.
Figure 1–5. Stratix IV GT Device Packaging Ordering Information
EP4SEP4S
40G
2
230
40
F
C
2
ES
Family Signature
Optional Suffix
Aggregate Bandwidth
Indicates specific device options
ES: Engineering sample
N: Lead-free devices
40G
100G
Device Density
Speed Grade
2 = 230k LEs
3 = 290k LEs
4 = 360k LEs
5 = 530k LEs
1, 2, 3 with 1 being the fastest
Operating Temperature
Package Type
C: Commercial temperature (t J = 0 C to 85 C)
I : Industrial temperature (t J = 0°C to 100°C)
F: FineLine BGA (FBGA)
H: Hybrid FineLine BGA
Ball Array Dimension
Corresponds to pin count
40 = 1517 pins
45 = 1932 pins
o
Document Revision History
Table 1–10 lists the revision history for this chapter.
Table 1–10. Document Revision History (Part 1 of 2)
Date
Version
September 2012
3.4
June 2011
3.3
February 2011
March 2010
Stratix IV Device Handbook
Volume 1
3.2
3.1
Changes
■
Updated Table 1–1 to close FB #30986.
■
Updated Table 1–2 and Table 1–5 to close FB #31127.
■
Added military temperature to Figure 1–4.
■
Updated Table 1–7 and Table 1–8.
■
Applied new template.
■
Minor text edits.
■
Updated Table 1–1, Table 1–2, and Table 1–7.
■
Updated Figure 1–3.
■
Updated the “Stratix IV GT Devices” section.
■
Added two new references to the Introduction section.
■
Minor text edits.
September 2012 Altera Corporation
Chapter 1: Overview for the Stratix IV Device Family
Ordering Information
1–21
Table 1–10. Document Revision History (Part 2 of 2)
Date
Version
November 2009
June 2009
3.0
2.4
April 2009
2.3
March 2009
2.2
March 2009
2.1
November 2008
2.0
Changes
■
Updated the “Stratix IV Device Family Overview”, “Feature Summary”, “Stratix IV GT
Devices”, “High-Speed Transceiver Features”, “FPGA Fabric and I/O Features”, “Highest
Aggregate Data Bandwidth”, “System Integration”, and “Integrated Software Platform”
sections.
■
Added Table 1–3, Table 1–6, and Table 1–9.
■
Updated Table 1–1, Table 1–2, Table 1–4, Table 1–5, Table 1–7, and Table 1–8.
■
Updated Figure 1–3, Figure 1–4, and Figure 1–5.
■
Minor text edits.
■
Updated Table 1–1.
■
Minor text edits.
■
Added Table 1–5, Table 1–6, and Figure 1–3.
■
Updated Figure 1–5.
■
Updated Table 1–1, Table 1–2, Table 1–3, and Table 1–4.
■
Updated “Introduction”, “Feature Summary”, “Stratix IV GX Devices”, “Stratix IV GT
Devices”, “Architecture Features”, and “FPGA Fabric and I/O Features”
■
Updated “Feature Summary”, “Stratix IV GX Devices”, “Stratix IV E Device”, “Stratix IV
GT Devices”, “Signal Integrity”
■
Removed Tables 1-5 and 1-6
■
Updated Figure 1–4
■
Updated “Introduction”, “Feature Summary”, “Stratix IV Device Diagnostic Features”,
“Signal Integrity”, “Clock Networks”,“High-Speed Differential I/O with DPA and SoftCDR”, “System Integration”, and “Ordering Information” sections.
■
Added “Stratix IV GT 100G Devices” and “Stratix IV GT 100G Transceiver Bandwidth”
sections.
■
Updated Table 1–1, Table 1–2, Table 1–3, and Table 1–4.
■
Added Table 1–5 and Table 1–6.
■
Updated Figure 1–3 and Figure 1–4.
■
Added Figure 1–5.
■
Removed “Referenced Documents” section.
■
Updated “Feature Summary” on page 1–1.
■
Updated “Stratix IV Device Diagnostic Features” on page 1–7.
■
Updated “FPGA Fabric and I/O Features” on page 1–8.
■
Updated Table 1–1.
■
Updated Table 1–2.
■
Updated “Table 1–5 shows the total number of transceivers available in the Stratix IV GT
Device.” on page 1–15.
July 2008
1.1
Revised “Introduction”.
May 2008
1.0
Initial release.
September 2012
Altera Corporation
Stratix IV Device Handbook
Volume 1
1–22
Stratix IV Device Handbook
Volume 1
Chapter 1: Overview for the Stratix IV Device Family
Ordering Information
September 2012 Altera Corporation
2. Logic Array Blocks and Adaptive Logic
Modules in Stratix IV Devices
February 2011
SIV51002-3.1
SIV51002-3.1
This chapter describes the features of the logic array blocks (LABs) in the Stratix® IV
core fabric. LABs are made up of adaptive logic modules (ALMs) that you can
configure to implement logic functions, arithmetic functions, and register functions.
LABs and ALMs are the basic building blocks of the Stratix IV device. Use these to
configure logic, arithmetic, and register functions. The ALM provides advanced
features with efficient logic usage and is completely backward-compatible.
This chapter contains the following sections:
■
“Logic Array Blocks”
■
“Adaptive Logic Modules” on page 2–5
Logic Array Blocks
Each LAB consists of ten ALMs, various carry chains, shared arithmetic chains, LAB
control signals, local interconnect, and register chain connection lines. The local
interconnect transfers signals between ALMs in the same LAB. The direct link
interconnect allows the LAB to drive into the local interconnect of its left and right
neighbors. Register chain connections transfer the output of the ALM register to the
adjacent ALM register in the LAB. The Quartus® II Compiler places associated logic in
the LAB or adjacent LABs, allowing the use of local, shared arithmetic chain, and
register chain connections for performance and area efficiency.
© 2011 Altera Corporation. All rights reserved. ALTERA, ARRIA, CYCLONE, HARDCOPY, MAX, MEGACORE, NIOS, QUARTUS and STRATIX words and logos
are trademarks of Altera Corporation and registered in the U.S. Patent and Trademark Office and in other countries. All other words and logos identified as
trademarks or service marks are the property of their respective holders as described at www.altera.com/common/legal.html. Altera warrants performance of its
semiconductor products to current specifications in accordance with Altera's standard warranty, but reserves the right to make changes to any products and
services at any time without notice. Altera assumes no responsibility or liability arising out of the application or use of any information, product, or service
described herein except as expressly agreed to in writing by Altera. Altera customers are advised to obtain the latest version of device specifications before relying
on any published information and before placing orders for products or services.
ISO
9001:2008
Registered
Stratix IV Device Handbook
Volume 1
February 2011
Feedback Subscribe
2–2
Chapter 2: Logic Array Blocks and Adaptive Logic Modules in Stratix IV Devices
Logic Array Blocks
Figure 2–1 shows the Stratix IV LAB structure and interconnects.
Figure 2–1. Stratix IV LAB Structure and Interconnects
C4
C12
Row Interconnects of
Variable Speed & Length
R20
R4
ALMs
Direct link
interconnect from
adjacent block
Direct link
interconnect from
adjacent block
Direct link
interconnect to
adjacent block
Direct link
interconnect to
adjacent block
Local Interconnect
LAB
MLAB
Local Interconnect is Driven
from Either Side by Columns & LABs,
& from Above by Rows
Column Interconnects of
Variable Speed & Length
The LAB of the Stratix IV device has a derivative called memory LAB (MLAB), which
adds look-up table (LUT)-based SRAM capability to the LAB, as shown in Figure 2–2.
The MLAB supports a maximum of 640 bits of simple dual-port static random access
memory (SRAM). You can configure each ALM in an MLAB as either a 64 × 1 or a
32 × 2 block, resulting in a configuration of either a 64 × 10 or a 32 × 20 simple
dual-port SRAM block. MLAB and LAB blocks always coexist as pairs in all Stratix IV
families. MLAB is a superset of the LAB and includes all LAB features.
Stratix IV Device Handbook
Volume 1
February 2011 Altera Corporation
Chapter 2: Logic Array Blocks and Adaptive Logic Modules in Stratix IV Devices
Logic Array Blocks
2–3
f For more information about the MLAB, refer to the chapter.
Figure 2–2. Stratix IV LAB and MLAB Structure
LUT-based-64 x 1
Simple dual-port SRAM
(1)
ALM
LUT-based-64 x 1
Simple dual-port SRAM
(1)
LUT-based-64 x 1
Simple dual-port SRAM
(1)
LUT-based-64 x 1
Simple dual-port SRAM
(1)
LUT-based-64 x 1
Simple dual-port SRAM
(1)
ALM
LAB Control Block
(1)
ALM
LUT-based-64 x 1
Simple dual-port SRAM
(1)
LUT-based-64 x 1
Simple dual-port SRAM
(1)
LUT-based-64 x 1
Simple dual-port SRAM
(1)
LUT-based-64 x 1
Simple dual-port SRAM
(1)
MLAB
ALM
ALM
LAB Control Block
LUT-based-64 x 1
Simple dual-port SRAM
ALM
ALM
ALM
ALM
ALM
LAB
Note to Figure 2–2:
(1) You can use the MLAB ALM as a regular LAB ALM or configure it as a dual-port SRAM, as shown.
February 2011
Altera Corporation
Stratix IV Device Handbook
Volume 1
2–4
Chapter 2: Logic Array Blocks and Adaptive Logic Modules in Stratix IV Devices
Logic Array Blocks
LAB Interconnects
The LAB local interconnect can drive ALMs in the same LAB. It is driven by column
and row interconnects and ALM outputs in the same LAB. Neighboring
LABs/MLABs, M9K RAM blocks, M144K blocks, or digital signal processing (DSP)
blocks from the left or right can also drive the LAB’s local interconnect through the
direct link connection. The direct link connection feature minimizes the use of row
and column interconnects, providing higher performance and flexibility. Each LAB
can drive 30 ALMs through fast-local and direct-link interconnects.
Figure 2–3 shows the direct-link connection.
Figure 2–3. Direct-Link Connection
Direct-link interconnect from the
left LAB, TriMatrix memory
block, DSP block, or IOE output
Direct-link interconnect from the
right LAB, TriMatrix memory
block, DSP block, or IOE output
ALMs
ALMs
Direct-link
interconnect
to right
Direct-link
interconnect
to left
Local
Interconnect
MLAB
LAB
LAB Control Signals
Each LAB contains dedicated logic for driving control signals to its ALMs. Control
signals include three clocks, three clock enables, two asynchronous clears, a
synchronous clear, and synchronous load control signals. This gives a maximum of 10
control signals at a time. Although you generally use synchronous-load and clear
signals when implementing counters, you can also use them with other functions.
Each LAB has two unique clock sources and three clock enable signals, as shown in
Figure 2–4. The LAB control block can generate up to three clocks using two clock
sources and three clock enable signals. Each LAB’s clock and clock enable signals are
linked. For example, any ALM in a particular LAB using the labclk1 signal also uses
the labclkena1 signal. If the LAB uses both the rising and falling edges of a clock, it
also uses two LAB-wide clock signals. De-asserting the clock enable signal turns off
the corresponding LAB-wide clock.
Stratix IV Device Handbook
Volume 1
February 2011 Altera Corporation
Chapter 2: Logic Array Blocks and Adaptive Logic Modules in Stratix IV Devices
Adaptive Logic Modules
2–5
The LAB row clocks [5..0] and LAB local interconnects generate the LAB-wide control
signals. The MultiTrack interconnect’s inherent low skew allows clock and control
signal distribution in addition to data.
Figure 2–4. LAB-Wide Control Signals
There are two unique
clock signals per LAB.
6
Dedicated Row LAB Clocks
6
6
Local Interconnect
Local Interconnect
Local Interconnect
Local Interconnect
Local Interconnect
Local Interconnect
labclk0
labclk2
labclk1
labclkena0
or asyncload
or labpreset
labclkena1
labclkena2
labclr1
syncload
labclr0
synclr
Adaptive Logic Modules
The ALM is the basic building block of logic in the Stratix IV architecture. It provides
advanced features with efficient logic usage. Each ALM contains a variety of
LUT-based resources that can be divided between two combinational adaptive LUTs
(ALUTs) and two registers. With up to eight inputs for the two combinational ALUTs,
one ALM can implement various combinations of two functions. This adaptability
allows an ALM to be completely backward-compatible with four-input LUT
architectures. One ALM can also implement any function with up to six inputs and
certain seven-input functions.
February 2011
Altera Corporation
Stratix IV Device Handbook
Volume 1
2–6
Chapter 2: Logic Array Blocks and Adaptive Logic Modules in Stratix IV Devices
Adaptive Logic Modules
In addition to the adaptive LUT-based resources, each ALM contains two
programmable registers, two dedicated full adders, a carry chain, a shared arithmetic
chain, and a register chain. Through these dedicated resources, an ALM can efficiently
implement various arithmetic functions and shift registers. Each ALM drives all types
of interconnects: local, row, column, carry chain, shared arithmetic chain, register
chain, and direct link. Figure 2–5 shows a high-level block diagram of the Stratix IV
ALM.
Figure 2–5. High-Level Block Diagram of the Stratix IV ALM
shared_arith_in
carry_in
Combinational/Memory ALUT0
reg_chain_in
labclk
To general or
local routing
dataf0
datae0
6-Input LUT
adder0
D
Q
dataa
To general or
local routing
reg0
datab
datac
datad
datae1
adder1
D
Q
6-Input LUT
To general or
local routing
reg1
dataf1
To general or
local routing
Combinational/Memory ALUT1
reg_chain_out
shared_arith_out
Stratix IV Device Handbook
Volume 1
carry_out
February 2011 Altera Corporation
Chapter 2: Logic Array Blocks and Adaptive Logic Modules in Stratix IV Devices
Adaptive Logic Modules
2–7
Figure 2–6 shows a detailed view of all the connections in an ALM.
Figure 2–6. Stratix IV ALM Connection Details
syncload
aclr[1:0]
shared_arith_in
carry_in
clk[2:0]
sclr
reg_chain_in
dataf0
datae0
dataa
datab
GND
4-INPUT
LUT
datac0
+
CLR
D
Q
3-INPUT
LUT
local
interconnect
row, column
direct link routing
row, column
direct link routing
3-INPUT
LUT
4-INPUT
LUT
datac1
+
CLR
D
Q
3-INPUT
LUT
local
interconnect
row, column
direct link routing
row, column
direct link routing
3-INPUT
LUT
VCC
datae1
dataf1
shared_arith_out
carry_out
reg_chain_out
One ALM contains two programmable registers. Each register has data, clock, clock
enable, synchronous and asynchronous clear, and synchronous load and clear inputs.
Global signals, general-purpose I/O pins, or any internal logic can drive the register’s
clock and clear-control signals. Either general-purpose I/O pins or internal logic can
drive the clock enable. For combinational functions, the register is bypassed and the
output of the LUT drives directly to the outputs of an ALM.
Each ALM has two sets of outputs that drive the local, row, and column routing
resources. The LUT, adder, or register outputs can drive these output drivers (refer to
Figure 2–6). For each set of output drivers, two ALM outputs can drive column, row,
or direct-link routing connections. One of these ALM outputs can also drive local
interconnect resources. This allows the LUT or adder to drive one output while the
register drives another output.
February 2011
Altera Corporation
Stratix IV Device Handbook
Volume 1
2–8
Chapter 2: Logic Array Blocks and Adaptive Logic Modules in Stratix IV Devices
Adaptive Logic Modules
This feature, called register packing, improves device utilization because the device
can use the register and the combinational logic for unrelated functions. Another
special packing mode allows the register output to feed back into the LUT of the same
ALM so that the register is packed with its own fan-out LUT. This provides another
mechanism for improved fitting. The ALM can also drive out registered and
unregistered versions of the LUT or adder output.
ALM Operating Modes
The Stratix IV ALM operates in one of the following modes:
■
Normal
■
Extended LUT
■
Arithmetic
■
Shared Arithmetic
■
LUT-Register
Each mode uses ALM resources differently. In each mode, eleven available inputs to
an ALM—the eight data inputs from the LAB local interconnect, carry-in from the
previous ALM or LAB, the shared arithmetic chain connection from the previous
ALM or LAB, and the register chain connection—are directed to different destinations
to implement the desired logic function. LAB-wide signals provide clock,
asynchronous clear, synchronous clear, synchronous load, and clock enable control for
the register. These LAB-wide signals are available in all ALM modes.
For more information about the LAB-wide control signals, refer to “LAB Control
Signals” on page 2–4.
The Quartus II software and supported third-party synthesis tools, in conjunction
with parameterized functions such as the library of parameterized modules (LPM)
functions, automatically choose the appropriate mode for common functions such as
counters, adders, subtractors, and arithmetic functions.
Stratix IV Device Handbook
Volume 1
February 2011 Altera Corporation
Chapter 2: Logic Array Blocks and Adaptive Logic Modules in Stratix IV Devices
Adaptive Logic Modules
2–9
Normal Mode
Normal mode is suitable for general logic applications and combinational functions.
In this mode, up to eight data inputs from the LAB local interconnect are inputs to the
combinational logic. Normal mode allows two functions to be implemented in one
Stratix IV ALM, or a single function of up to six inputs. The ALM can support certain
combinations of completely independent functions and various combinations of
functions that have common inputs.
Figure 2–7 shows the supported LUT combinations in normal mode.
Figure 2–7. ALM in Normal Mode
dataf0
datae0
datac
dataa
(1)
4-Input
LUT
combout0
datab
datad
datae1
dataf1
4-Input
LUT
combout1
dataf0
datae0
datac
dataa
datab
5-Input
LUT
combout0
datad
datae1
dataf1
3-Input
LUT
dataf0
datae0
datac
dataa
datab
5-Input
LUT
4-Input
LUT
datad
datae1
dataf1
dataf0
datae0
datac
dataa
datab
5-Input
LUT
combout0
5-Input
LUT
combout1
dataf0
datae0
dataa
datab
datac
datad
6-Input
LUT
combout0
dataf0
datae0
dataa
datab
datac
datad
6-Input
LUT
combout0
6-Input
LUT
combout1
datad
datae1
dataf1
combout1
combout0
combout1
datae1
dataf1
Note to Figure 2–7:
(1) Combinations of functions with fewer inputs than those shown are also supported. For example, combinations of functions with the following
number of inputs are supported: 4 and 3, 3 and 3, 3 and 2, and 5 and 2.
Normal mode provides complete backward-compatibility with four-input LUT
architectures.
February 2011
Altera Corporation
Stratix IV Device Handbook
Volume 1
2–10
Chapter 2: Logic Array Blocks and Adaptive Logic Modules in Stratix IV Devices
Adaptive Logic Modules
For the packing of 2 five-input functions into one ALM, the functions must have at
least two common inputs. The common inputs are dataa and datab. The combination
of a four-input function with a five-input function requires one common input (either
dataa or datab).
In the case of implementing 2 six-input functions in one ALM, four inputs must be
shared and the combinational function must be the same. In a sparsely used device,
functions that could be placed in one ALM may be implemented in separate ALMs by
the Quartus II software to achieve the best possible performance. As a device begins
to fill up, the Quartus II software automatically uses the full potential of the Stratix IV
ALM. The Quartus II Compiler automatically searches for functions using common
inputs or completely independent functions to be placed in one ALM to make efficient
use of device resources. In addition, you can manually control resource usage by
setting location assignments.
You can implement any six-input function using inputs dataa, datab, datac, datad,
and either datae0 and dataf0 or datae1 and dataf1. If you use datae0 and dataf0, the
output is driven to register0, and/or register0 is bypassed and the data drives out
to the interconnect using the top set of output drivers (refer to Figure 2–8). If you use
datae1 and dataf1, the output either drives to register1 or bypasses register1 and
drives to the interconnect using the bottom set of output drivers. The Quartus II
Compiler automatically selects the inputs to the LUT. ALMs in normal mode support
register packing.
Figure 2–8. Input Function in Normal Mode
dataf0
datae0
dataa
datab
datac
datad
(1)
To general or
local routing
6-Input
LUT
D
Q
To general or
local routing
reg0
datae1
dataf1
(2)
D
labclk
Q
To general or
local routing
reg1
These inputs are available for register packing.
Notes to Figure 2–8:
(1) If you use datae1 and dataf1 as inputs to a six-input function, datae0 and dataf0 are available for register packing.
(2) The dataf1 input is available for register packing only if the six-input function is unregistered.
Stratix IV Device Handbook
Volume 1
February 2011 Altera Corporation
Chapter 2: Logic Array Blocks and Adaptive Logic Modules in Stratix IV Devices
Adaptive Logic Modules
2–11
Extended LUT Mode
Use extended LUT mode to implement a specific set of seven-input functions. The set
must be a 2-to-1 multiplexer fed by two arbitrary five-input functions sharing four
inputs. Figure 2–9 shows the template of supported seven-input functions using
extended LUT mode. In this mode, if the seven-input function is unregistered, the
unused eighth input is available for register packing.
Functions that fit into the template shown in Figure 2–9 occur naturally in designs.
These functions often appear in designs as “if-else” statements in Verilog HDL or
VHDL code.
Figure 2–9. Template for Supported Seven-Input Functions in Extended LUT Mode
datae0
datac
dataa
datab
datad
dataf0
5-Input
LUT
To general or
local routing
combout0
D
5-Input
LUT
Q
To general or
local routing
reg0
datae1
dataf1
(1)
This input is available
for register packing.
Note to Figure 2–9:
(1) If the seven-input function is unregistered, the unused eighth input is available for register packing. The second register, reg1, is
not available.
February 2011
Altera Corporation
Stratix IV Device Handbook
Volume 1
2–12
Chapter 2: Logic Array Blocks and Adaptive Logic Modules in Stratix IV Devices
Adaptive Logic Modules
Arithmetic Mode
Arithmetic mode is ideal for implementing adders, counters, accumulators, wide
parity functions, and comparators. The ALM in arithmetic mode uses two sets of
2 four-input LUTs along with two dedicated full adders. The dedicated adders allow
the LUTs to be available to perform pre-adder logic; therefore, each adder can add the
output of 2 four-input functions.
The four LUTs share dataa and datab inputs. As shown in Figure 2–10, the carry-in
signal feeds to adder0 and the carry-out from adder0 feeds to the carry-in of adder1.
The carry-out from adder1 drives to adder0 of the next ALM in the LAB. ALMs in
arithmetic mode can drive out registered and/or unregistered versions of the adder
outputs.
Figure 2–10. ALM in Arithmetic Mode
carry_in
datae0
adder0
4-Input
LUT
To general or
local routing
D
dataf0
datac
datab
dataa
datad
datae1
Q
To general or
local routing
reg0
4-Input
LUT
adder1
4-Input
LUT
To general or
local routing
D
4-Input
LUT
Q
To general or
local routing
reg1
dataf1
carry_out
While operating in arithmetic mode, the ALM can support simultaneous use of the
adder’s carry output along with combinational logic outputs. In this operation, adder
output is ignored. Using the adder with combinational logic output provides resource
savings of up to 50% for functions that can use this ability.
Arithmetic mode also offers clock enable, counter enable, synchronous up/down
control, add/subtract control, synchronous clear, and synchronous load. The LAB
local interconnect data inputs generate the clock enable, counter enable, synchronous
up/down, and add/subtract control signals. These control signals are good
candidates for the inputs that are shared between the four LUTs in the ALM. The
synchronous clear and synchronous load options are LAB-wide signals that affect all
registers in the LAB. These signals can also be individually disabled or enabled per
register. The Quartus II software automatically places any registers that are not used
by the counter into other LABs.
Stratix IV Device Handbook
Volume 1
February 2011 Altera Corporation
Chapter 2: Logic Array Blocks and Adaptive Logic Modules in Stratix IV Devices
Adaptive Logic Modules
2–13
Carry Chain
The carry chain provides a fast carry function between the dedicated adders in
arithmetic or shared-arithmetic mode. The two-bit carry select feature in Stratix IV
devices halves the propagation delay of carry chains within the ALM. Carry chains
can begin in either the first ALM or the fifth ALM in the LAB. The final carry-out
signal is routed to the ALM, where it is fed to local, row, or column interconnects.
The Quartus II Compiler automatically creates carry-chain logic during design
processing, or you can create it manually during design entry. Parameterized
functions such as LPM functions automatically take advantage of carry chains for the
appropriate functions.
The Quartus II Compiler creates carry chains longer than 20 (10 ALMs in arithmetic or
shared arithmetic mode) by linking LABs together automatically. For enhanced
fitting, a long carry chain runs vertically, allowing fast horizontal connections to
TriMatrix memory and DSP blocks. A carry chain can continue as far as a full column.
To avoid routing congestion in one small area of the device when a high fan-in
arithmetic function is implemented, the LAB can support carry chains that only use
either the top half or bottom half of the LAB before connecting to the next LAB. This
leaves the other half of the ALMs in the LAB available for implementing narrower
fan-in functions in normal mode. Carry chains that use the top five ALMs in the first
LAB carry into the top half of the ALMs in the next LAB within the column. Carry
chains that use the bottom five ALMs in the first LAB carry into the bottom half of the
ALMs in the next LAB within the column. In every alternate LAB column, the top half
can be bypassed; in the other MLAB columns, the bottom half can be bypassed.
For more information about carry-chain interconnects, refer to “ALM Interconnects”
on page 2–18.
February 2011
Altera Corporation
Stratix IV Device Handbook
Volume 1
2–14
Chapter 2: Logic Array Blocks and Adaptive Logic Modules in Stratix IV Devices
Adaptive Logic Modules
Shared Arithmetic Mode
In shared arithmetic mode, the ALM can implement a three-input add within the
ALM. In this mode, the ALM is configured with 4 four-input LUTs. Each LUT either
computes the sum of three inputs or the carry of three inputs. The output of the carry
computation is fed to the next adder (either to adder1 in the same ALM or to adder0 of
the next ALM in the LAB) using a dedicated connection called the shared arithmetic
chain. This shared arithmetic chain can significantly improve the performance of an
adder tree by reducing the number of summation stages required to implement an
adder tree. Figure 2–11 shows the ALM using this feature.
Figure 2–11. ALM in Shared Arithmetic Mode
shared_arith_in
carry_in
labclk
4-Input
LUT
To general or
local routing
D
datae0
datac
datab
dataa
datad
datae1
Q
To general or
local routing
reg0
4-Input
LUT
4-Input
LUT
To general or
local routing
D
4-Input
LUT
Q
To general or
local routing
reg1
carry_out
shared_arith_out
You can find adder trees in many different applications. For example, the summation
of the partial products in a logic-based multiplier can be implemented in a tree
structure. Another example is a correlator function that can use a large adder tree to
sum filtered data samples in a given time frame to recover or de-spread data that was
transmitted using spread-spectrum technology.
Shared Arithmetic Chain
The shared arithmetic chain available in enhanced arithmetic mode allows the ALM
to implement a three-input add. This significantly reduces the resources necessary to
implement large adder trees or correlator functions.
Shared arithmetic chains can begin in either the first or sixth ALM in the LAB. The
Quartus II Compiler creates shared arithmetic chains longer than 20 (10 ALMs in
arithmetic or shared arithmetic mode) by linking LABs together automatically. For
enhanced fitting, a long shared arithmetic chain runs vertically, allowing fast
horizontal connections to the TriMatrix memory and DSP blocks. A shared arithmetic
chain can continue as far as a full column.
Stratix IV Device Handbook
Volume 1
February 2011 Altera Corporation
Chapter 2: Logic Array Blocks and Adaptive Logic Modules in Stratix IV Devices
Adaptive Logic Modules
2–15
Similar to the carry chains, the top and bottom halves of shared arithmetic chains in
alternate LAB columns can be bypassed. This capability allows the shared arithmetic
chain to cascade through half of the ALMs in an LAB while leaving the other half
available for narrower fan-in functionality. Every other LAB column is top-half
by-passable, while the other LAB columns are bottom-half by-passable.
For more information about the shared arithmetic chain interconnect, refer to “ALM
Interconnects” on page 2–18.
LUT-Register Mode
LUT-register mode allows third-register capability within an ALM. Two internal
feedback loops allow combinational ALUT1 to implement the master latch and
combinational ALUT0 to implement the slave latch needed for the third register. The
LUT register shares its clock, clock enable, and asynchronous clear sources with the
top dedicated register. Figure 2–12 shows the register constructed using two
combinational blocks within the ALM.
Figure 2–12. LUT Register from Two Combinational Blocks
sumout
clk
aclr
LUT regout
4-input
LUT
combout
5-input
LUT
combout
sumout
datain(datac)
sclr
February 2011
Altera Corporation
Stratix IV Device Handbook
Volume 1
2–16
Chapter 2: Logic Array Blocks and Adaptive Logic Modules in Stratix IV Devices
Adaptive Logic Modules
Figure 2–13 shows the ALM in LUT-register mode.
Figure 2–13. ALM in LUT-Register Mode with Three-Register Capability
clk [2:0]
DC1
aclr [1:0]
reg_chain_in
datain
lelocal 0
aclr
aclr
sclr
regout
datain
latchout
sdata
leout 0 a
regout
leout 0 b
E0
F1
lelocal 1
aclr
datain
E1
sdata
F0
leout 1 a
regout
leout 1 b
reg_chain_out
Stratix IV Device Handbook
Volume 1
February 2011 Altera Corporation
Chapter 2: Logic Array Blocks and Adaptive Logic Modules in Stratix IV Devices
Adaptive Logic Modules
2–17
Register Chain
In addition to general routing outputs, ALMs in the LAB have register-chain outputs.
Register-chain routing allows registers in the same LAB to be cascaded together. The
register-chain interconnect allows the LAB to use LUTs for a single combinational
function and the registers to be used for an unrelated shift-register implementation.
These resources speed up connections between ALMs while saving local interconnect
resources (refer to Figure 2–14). The Quartus II Compiler automatically takes
advantage of these resources to improve utilization and performance.
Figure 2–14. Register Chain within the LAB
(1)
From previous ALM
within the LAB
reg_chain_in
labclk
To general or
local routing
adder0
D
Q
To general or
local routing
reg0
Combinational
Logic
adder1
D
Q
To general or
local routing
reg1
To general or
local routing
To general or
local routing
adder0
D
Q
To general or
local routing
reg0
Combinational
Logic
adder1
D
Q
To general or
local routing
reg1
To general or
local routing
reg_chain_out
To next ALM
within the LAB
Note to Figure 2–14:
(1) You can use the combinational or adder logic to implement an unrelated, un-registered function.
For more information about the register chain interconnect, refer to “ALM
Interconnects” on page 2–18.
February 2011
Altera Corporation
Stratix IV Device Handbook
Volume 1
2–18
Chapter 2: Logic Array Blocks and Adaptive Logic Modules in Stratix IV Devices
Adaptive Logic Modules
ALM Interconnects
There are three dedicated paths between the ALMs—register cascade, carry chain,
and shared arithmetic chain. Stratix IV devices include an enhanced interconnect
structure in LABs for routing shared arithmetic chains and carry chains for efficient
arithmetic functions. The register chain connection allows the register output of one
ALM to connect directly to the register input of the next ALM in the LAB for fast shift
registers. These ALM-to-ALM connections bypass the local interconnect. The
Quartus II Compiler automatically takes advantage of these resources to improve
utilization and performance. Figure 2–15 shows the shared arithmetic chain, carry
chain, and register chain interconnects.
Figure 2–15. Shared Arithmetic Chain, Carry Chain, and Register Chain Interconnects
Local interconnect
routing among ALMs
in the LAB
Carry chain & shared
arithmetic chain
routing to adjacent ALM
ALM 1
Register chain
routing to adjacent
ALM's register input
ALM 2
Local
interconnect
ALM 3
ALM 4
ALM 5
ALM 6
ALM 7
ALM 8
ALM 9
ALM 10
Clear and Preset Logic Control
LAB-wide signals control the logic for the register’s clear signal. The ALM directly
supports an asynchronous clear function. You can achieve the register preset through
the Quartus II software’s NOT-gate push-back logic option. Each LAB supports up to
two clears.
Stratix IV devices provide a device-wide reset pin (DEV_CLRn) that resets all the
registers in the device. An option set before compilation in the Quartus II software
controls this pin. This device-wide reset overrides all other control signals.
Stratix IV Device Handbook
Volume 1
February 2011 Altera Corporation
Chapter 2: Logic Array Blocks and Adaptive Logic Modules in Stratix IV Devices
Adaptive Logic Modules
2–19
LAB Power Management Techniques
The following techniques are used to manage static and dynamic power consumption
within the LAB:
■
To save AC power, the Quartus II software forces all adder inputs low when ALM
adders are not in use.
■
Stratix IV LABs operate in high-performance mode or low-power mode. The
Quartus II software automatically chooses the appropriate mode for the LAB,
based on the design, to optimize speed versus leakage trade-offs.
■
Clocks represent a significant portion of dynamic power consumption due to their
high switching activity and long paths. The LAB clock that distributes a clock
signal to registers within an LAB is a significant contributor to overall clock power
consumption. Each LAB’s clock and clock enable signal are linked. For example, a
combinational ALUT or register in a particular LAB using the labclk1 signal also
uses the labclkena1 signal. To disable LAB-wide clock power consumption
without disabling the entire clock tree, use LAB-wide clock enable to gate the
LAB-wide clock. The Quartus II software automatically promotes register-level
clock enable signals to the LAB-level. All registers within the LAB that share a
common clock and clock enable are controlled by a shared, gated clock. To take
advantage of these clock enables, use a clock-enable construct in your HDL code
for the registered logic.
f For more information about implementing static and dynamic power consumption
within the LAB, refer to the Power Optimization chapter in volume 2 of the Quartus II
Handbook.
Document Revision History
Table 2–1 lists the revision history for this chapter.
Table 2–1. Document Revision History
Date
Version
February 2011
3.1
November 2009
3.0
June 2009
2.2
March 2009
2.1
November 2008
2.0
May 2008
1.0
February 2011
Altera Corporation
Changes
■
Updated Figure 2–6.
■
Applied new template.
■
Minor text edits.
■
Updated graphics.
■
Minor text edits.
■
Removed the Conclusion section.
■
Added introductory sentences to improve search ability.
■
Minor text edits.
Removed “Referenced Documents” section.
■
Updated Figure 2–6.
■
Made minor editorial changes.
Initial release.
Stratix IV Device Handbook
Volume 1
2–20
Stratix IV Device Handbook
Volume 1
Chapter 2: Logic Array Blocks and Adaptive Logic Modules in Stratix IV Devices
Adaptive Logic Modules
February 2011 Altera Corporation
3. TriMatrix Embedded Memory Blocks in
Stratix IV Devices
December 2011
SIV51003-3.3
SIV51003-3.3
This chapter describes the TriMatrix embedded memory blocks in Stratix® IV devices.
TriMatrix embedded memory blocks provide three different sizes of embedded
SRAM to efficiently address the needs of Stratix IV FPGA designs. TriMatrix memory
includes 640-bit memory logic array blocks (MLABs), 9-Kbit M9K blocks, and
144-Kbit M144K blocks. MLABs have been optimized to implement filter delay lines,
small FIFO buffers, and shift registers. You can use the M9K blocks for general
purpose memory applications and the M144K blocks for processor code storage,
packet buffering, and video frame buffering.
You can independently configure each embedded memory block to be a single- or
dual-port RAM, FIFO buffer, ROM, or shift register using the Quartus® II
MegaWizard™ Plug-In Manager. You can stitch together multiple blocks of the same
type to produce larger memories with minimal timing penalty. TriMatrix memory
provides up to 31,491 Kbits of embedded SRAM at up to 600 MHz operation.
This chapter contains the following sections:
■
“Overview”
■
“Memory Modes” on page 3–9
■
“Clocking Modes” on page 3–17
■
“Design Considerations” on page 3–18
Overview
Table 3–1 lists the features supported by the three sizes of TriMatrix memory.
Table 3–1. Summary of TriMatrix Memory Features (Part 1 of 2)
Feature
Maximum performance
Total RAM bits
(including parity bits)
MLABs
M9K Blocks
M144K Blocks
600 MHz
600 MHz
540 MHz
640
9216
147,456
© 2011 Altera Corporation. All rights reserved. ALTERA, ARRIA, CYCLONE, HARDCOPY, MAX, MEGACORE, NIOS, QUARTUS and STRATIX words and logos
are trademarks of Altera Corporation and registered in the U.S. Patent and Trademark Office and in other countries. All other words and logos identified as
trademarks or service marks are the property of their respective holders as described at www.altera.com/common/legal.html. Altera warrants performance of its
semiconductor products to current specifications in accordance with Altera's standard warranty, but reserves the right to make changes to any products and
services at any time without notice. Altera assumes no responsibility or liability arising out of the application or use of any information, product, or service
described herein except as expressly agreed to in writing by Altera. Altera customers are advised to obtain the latest version of device specifications before relying
on any published information and before placing orders for products or services.
ISO
9001:2008
Registered
Stratix IV Device Handbook
Volume 1
December 2011
Feedback Subscribe
3–2
Chapter 3: TriMatrix Embedded Memory Blocks in Stratix IV Devices
Overview
Table 3–1. Summary of TriMatrix Memory Features (Part 2 of 2)
Feature
MLABs
M9K Blocks
8K × 1
4K × 2
64 × 8
2K × 4
64 × 9
Configurations
(depth × width)
1K × 8
64 × 10
1K × 9
32 × 16
512 × 16
32 × 18
512 × 18
32 × 20
256 × 32
256 × 36
M144K Blocks
16K × 8
16K × 9
8K × 16
8K × 18
4K × 32
4K × 36
2K × 64
2K × 72
Parity bits
Supported
Supported
Supported
Byte enable
Supported
Supported
Supported
—
Supported
Supported
Address clock enable
Supported
Supported
Supported
Single-port memory
Supported
Supported
Supported
Simple dual-port memory
Supported
Supported
Supported
True dual-port memory
—
Supported
Supported
Embedded shift register
Supported
Supported
Supported
ROM
Supported
Supported
Supported
FIFO buffer
Packed mode
Supported
Supported
Supported
Simple dual-port mixed
width support
—
Supported
Supported
True dual-port mixed width
support
—
Supported
Supported
Memory Initialization File
(.mif)
Supported
Supported
Supported
Mixed clock mode
Supported
Supported
Supported
Power-up condition
Outputs cleared if
registered, otherwise reads
memory contents
Outputs cleared
Outputs cleared
Register clears
Output registers
Output registers
Output registers
Write/Read operation
triggering
Write: Falling clock edges
Write and Read: Rising clock
edges
Write and Read: Rising clock
edges
Same-port read-during-write
Outputs set to don’t care
Outputs set to old data or
new data
Outputs set to old data or
new data
Mixed-port read-during-write
Outputs set to old data,
new data, or don’t care
Outputs set to old data or
don’t care
Outputs set to old data or
don’t care
ECC Support
Soft IP support using the
Quartus II software
Soft IP support using the
Quartus II software
Built-in support in ×64-wide
SDP mode or soft IP support
using the Quartus II software
Read: Rising clock edges
(1)
Note to Table 3–1:
(1) The mixed-port read-during-write options of new data or old data are only supported for MLABs when you use both the read address registers
and the output registers.
Stratix IV Device Handbook
Volume 1
December 2011 Altera Corporation
Chapter 3: TriMatrix Embedded Memory Blocks in Stratix IV Devices
Overview
3–3
Table 3–2 lists the capacity and distribution of the TriMatrix memory blocks in each
Stratix IV family member.
Table 3–2. TriMatrix Memory Capacity and Distribution in Stratix IV Devices
MLABs
M9K Blocks
M144K
Blocks
Total Dedicated RAM Bits
(Dedicated Memory Blocks Only)
(Kb)
Total RAM Bits
(Including MLABs)
(Kb)
EP4SE230
4,560
1,235
22
14,283
17,133
EP4SE360
7,072
1,248
48
18,144
22,564
EP4SE530
10,624
1,280
64
20,736
27,376
EP4SE820
16,261
1,610
60
23,130
33,294
EP4SGX70
1,452
462
16
6,462
7,370
EP4SGX110
2,112
660
16
8,244
9,564
Device
EP4SGX180
3,515
950
20
11,430
13,627
EP4SGX230
4,560
1,235
22
14,283
17,133
EP4SGX290
5,824
936
36
13,608
17,248
EP4SGX360
7,072
1,248
48
18,144
22,564
EP4SGX530
10,624
1,280
64
20,736
27,376
EP4S40G2
4,560
1,235
22
14,283
17,133
EP4S40G5
10,624
1280
64
20,736
27,376
EP4S100G2
4,560
1,235
22
14,283
17,133
EP4S100G3
5,824
936
36
13,608
17,248
EP4S100G4
7,072
1,248
48
18,144
22,564
EP4S100G5
10,624
1,280
64
20,736
27,376
TriMatrix Memory Block Types
While the M9K and M144K memory blocks are dedicated resources, the MLABs are
dual-purpose blocks. They can be configured as regular logic array blocks (LABs) or
as MLABs. Ten adaptive logic modules (ALMs) make up one MLAB. You can
configure each ALM in an MLAB as either a 64 × 1 or a 32 × 2 block, resulting in a
64 × 10 or 32 × 20 simple dual-port SRAM block in a single MLAB.
Parity Bit Support
All TriMatrix memory blocks have built-in parity-bit support. The ninth bit associated
with each byte can store a parity bit or serve as an additional data bit. No parity
function is actually performed on the ninth bit.
Byte Enable Support
All TriMatrix memory blocks support byte enables that mask the input data so that
only specific bytes of data are written. The unwritten bytes retain the previously
written values. The write enable (wren) signals, along with the byte enable (byteena)
signals, control the RAM blocks’ write operations.
December 2011
Altera Corporation
Stratix IV Device Handbook
Volume 1
3–4
Chapter 3: TriMatrix Embedded Memory Blocks in Stratix IV Devices
Overview
The default value for the byte enable signals is high (enabled), in which case writing is
controlled only by the write enable signals. The byte enable registers have no clear
port. When using parity bits on the M9K and M144K blocks, the byte enable controls
all nine bits (eight bits of data plus one parity bit). When using parity bits on the
MLAB, the byte-enable controls all 10 bits in the widest mode.
The MSB for the byteena signal corresponds to the MSB of the data bus and the LSB of
the byteena signal corresponds to the LSB of the data bus. For example, if you use a
RAM block in ×18 mode, with byteena = 01, data[8..0] is enabled, and data[17..9]
id disabled. Similarly, if byteena = 11, both data[8..0] and data[17..9] are enabled.
Byte enables are active high.
1
You cannot use the byte enable feature when using the error correction coding (ECC)
feature on M144K blocks.
1
Byte enables are only supported for true dual-port memory configurations when both
the PortA and PortB data widtByths of the individual M9K memory blocks are
multiples of 8 or 9 bits. For example, if you implement a mixed data width memory
configured with portA = 32 and portB = 8 as two separate 16 x 4 bit memories, you
cannot use the byte enable feature.
Figure 3–1 shows how the write enable (wren) and byte enable (byteena) signals
control the operations of the RAM blocks.
When a byte-enable bit is de-asserted during a write cycle, the corresponding data
byte output can appear as either a “don’t care” value or the current data at that
location. The output value for the masked byte is controllable using the Quartus II
software. When a byte-enable bit is asserted during a write cycle, the corresponding
data byte output also depends on the setting chosen in the Quartus II software.
Figure 3–1. Byte Enable Functional Waveform
inclock
wren
address
data
byteena
contents at a0
contents at a1
a0
an
a1
a2
a0
a1
ABCD
XXXX
10
XX
a2
XXXX
01
11
FFFF
XX
ABFF
FFFF
FFCD
FFFF
contents at a2
ABCD
don't care: q (asynch)
doutn
ABXX
XXCD
ABCD
ABFF
FFCD
ABCD
current data: q (asynch)
doutn
ABFF
FFCD
ABCD
ABFF
FFCD
ABCD
Stratix IV Device Handbook
Volume 1
December 2011 Altera Corporation
Chapter 3: TriMatrix Embedded Memory Blocks in Stratix IV Devices
Overview
3–5
Packed Mode Support
Stratix IV M9K and M144K blocks support packed mode. The packed mode feature
packs two independent single-port RAMs into one memory block. The Quartus II
software automatically implements packed mode where appropriate by placing the
physical RAM block into true dual-port mode and using the MSB of the address to
distinguish between the two logical RAMs. The size of each independent single-port
RAM must not exceed half of the target block size.
December 2011
Altera Corporation
Stratix IV Device Handbook
Volume 1
3–6
Chapter 3: TriMatrix Embedded Memory Blocks in Stratix IV Devices
Overview
Address Clock Enable Support
All Stratix IV memory blocks support address clock enable, which holds the previous
address value for as long as the signal is enabled (addressstall = 1). When the
memory blocks are configured in dual-port mode, each port has its own independent
address clock enable. The default value for the address clock enable signals is low
(disabled).
Figure 3–2 shows an address clock enable block diagram. The address clock enable is
referred to by the port name addressstall.
Figure 3–2. Address Clock Enable
address[0]
1
0
address[N]
1
0
address[0]
register
address[0]
address[N]
register
address[N]
addressstall
clock
Figure 3–3 shows the address clock enable waveform during the read cycle.
Figure 3–3. Address Clock Enable During Read Cycle Waveform
inclock
rdaddress
a0
a1
a2
a3
a4
a5
a6
rden
addressstall
latched address
(inside memory)
an
q (synch) doutn-1
q (asynch)
Stratix IV Device Handbook
Volume 1
doutn
a1
a0
doutn
dout0
dout0
a4
dout4
dout1
dout1
a5
dout4
dout5
December 2011 Altera Corporation
Chapter 3: TriMatrix Embedded Memory Blocks in Stratix IV Devices
Overview
3–7
Figure 3–4 shows the address clock enable waveform during the write cycle.
Figure 3–4. Address Clock Enable During the Write Cycle Waveform
inclock
wraddress
a0
a1
a2
a3
a4
a5
a6
00
01
02
03
04
05
06
data
wren
addressstall
latched address
(inside memory)
contents at a0
contents at a1
an
a1
a0
a4
a5
00
XX
XX
01
02
contents at a2
XX
contents at a3
XX
contents at a4
contents at a5
03
04
XX
XX
05
Mixed Width Support
M9K and M144K memory blocks support mixed data widths inherently. MLABs can
support mixed data widths through emulation using the Quartus II software. When
using simple dual-port, true dual-port, or FIFO modes, mixed width support allows
you to read and write different data widths to a memory block. For more information
about the different widths supported per memory mode, refer to “Memory Modes”
on page 3–9.
1
MLABs do not support mixed-width FIFO mode.
Asynchronous Clear
Stratix IV TriMatrix memory blocks support asynchronous clears on output latches
and output registers. Therefore, if your RAM is not using output registers, you can
still clear the RAM outputs using the output latch asynchronous clear. Figure 3–5
shows a waveform of the output latch asynchronous clear function.
Figure 3–5. Output Latch Asynchronous Clear Waveform
outclk
aclr
aclr at latch
q
You can selectively enable asynchronous clears per logical memory using the
Quartus II RAM MegaWizard Plug-In Manager.
f For more information, refer to the Internal Memory (RAM and ROM) User Guide.
December 2011
Altera Corporation
Stratix IV Device Handbook
Volume 1
3–8
Chapter 3: TriMatrix Embedded Memory Blocks in Stratix IV Devices
Overview
Error Correction Code (ECC) Support
Stratix IV M144K blocks have built-in support for error correction code (ECC) when in
×64-wide simple dual-port mode. ECC allows you to detect and correct data errors in
the memory array. The M144K blocks have a single-error-correction
double-error-detection (SECDED) implementation. SECDED can detect and fix a
single bit error in a 64-bit word, or detect two bit errors in a 64-bit word. It cannot
detect three or more errors.
The M144K ECC status is communicated using a three-bit status flag
eccstatus[2..0]. The status flag can be either registered or unregistered. When
registered, it uses the same clock and asynchronous clear signals as the output
registers. When unregistered, it cannot be asynchronously cleared.
Table 3–3 lists the truth table for the ECC status flags.
Table 3–3. Truth Table for ECC Status Flags
Status
eccstatus[2]
eccstatus[1]
eccstatus[0]
No error
0
0
0
Single error and fixed
0
1
1
Double error and no fix
1
0
1
Illegal
0
0
1
Illegal
0
1
0
Illegal
1
0
0
Illegal
1
1
X
1
You cannot use the byte enable feature when ECC is engaged.
1
Read-during-write “old data mode” is not supported when ECC is engaged.
Stratix IV Device Handbook
Volume 1
December 2011 Altera Corporation
Chapter 3: TriMatrix Embedded Memory Blocks in Stratix IV Devices
Memory Modes
3–9
Figure 3–6 shows a diagram of the ECC block of the M144K block.
Figure 3–6. ECC Block Diagram of the M144K Block
8
64
64
SECDED
Encoder
Data Input
8
72
RAM
Array
72
64
SECDED
Encoder
Comparator
8
64
8
8
64
Error
Locator
64
Error
Correction
Block
Flag
Generator
3
Status Flags
64
Data Output
Memory Modes
Stratix IV TriMatrix memory blocks allow you to implement fully synchronous SRAM
memory in multiple modes of operation. M9K and M144K blocks do not support
asynchronous memory (unregistered inputs). MLABs support asynchronous
(flow-through) read operations.
Depending on which TriMatrix memory block you target, you can use the following:
■
“Single-Port RAM Mode” on page 3–10
■
“Simple Dual-Port Mode” on page 3–11
■
“True Dual-Port Mode” on page 3–14
■
“Shift-Register Mode” on page 3–16
■
“ROM Mode” on page 3–17
■
“FIFO Mode” on page 3–17
c When using the memory blocks in ROM, single-port, simple dual-port, or true
dual-port mode, you can corrupt the memory contents if you violate the setup or
hold-time on any of the memory block input registers. This applies to both read and
write operations.
December 2011
Altera Corporation
Stratix IV Device Handbook
Volume 1
3–10
Chapter 3: TriMatrix Embedded Memory Blocks in Stratix IV Devices
Memory Modes
Single-Port RAM Mode
All TriMatrix memory blocks support single-port mode. Single-port mode allows you
to do either one-read or one-write operation at a time. Simultaneous reads and writes
are not supported in single-port mode. Figure 3–7 shows the single-port RAM
configuration.
Figure 3–7. Single-Port RAM
(1)
data[ ]
address[ ]
wren
byteena[]
addressstall
inclock
clockena
rden
aclr
q[]
outclock
Note to Figure 3–7:
(1) You can implement two single-port memory blocks in a single M9K or M144K block. For more information, refer to
“Packed Mode Support” on page 3–5.
During a write operation, RAM output behavior is configurable. If you use the
read-enable signal and perform a write operation with read enable de-activated, the
RAM outputs retain the values they held during the most recent active read enable. If
you activate read enable during a write operation, or if you are not using the
read-enable signal at all, the RAM outputs either show the “new data” being written,
the “old data” at that address, or a “don’t care” value. To choose the desired behavior,
set the read-during-write behavior to either new data, old data, or don’t care in the
RAM MegaWizard Plug-In Manager in the Quartus II software. For more information,
refer to “Read-During-Write Behavior” on page 3–19.
Table 3–4 lists the possible port width configurations for TriMatrix memory blocks in
single-port mode.
Table 3–4. Port Width Configurations for MLABs, M9K, and M144K Blocks (Single-Port Mode)
MLABs
M9K Blocks
8K × 1
64 × 8
64 × 9
Port Width
Configurations
64 × 10
32 × 16
32 × 18
32 × 20
4K × 2
2K × 4
1K × 8
1K × 9
512 × 16
512 × 18
256 × 32
256 × 36
Stratix IV Device Handbook
Volume 1
M144K Blocks
16K × 8
16K × 9
8K × 16
8K × 18
4K × 32
4K × 36
2K × 64
2K × 72
December 2011 Altera Corporation
Chapter 3: TriMatrix Embedded Memory Blocks in Stratix IV Devices
Memory Modes
3–11
Figure 3–8 shows timing waveforms for read and write operations in single-port
mode with unregistered outputs. Registering the RAM’s outputs simply delays the
q output by one clock cycle.
Figure 3–8. Timing Waveform for Read-Write Operations (Single-Port Mode)
clk_a
A0
address
A1
rdena
wrena
bytenna
data_a
01
10
00
A123
B456
C789
A0 (old data) DoldDold23
q_a (asyn)
11
DDDD
B423
EEEE
A1(old data)
FFFF
DDDD
EEEE
Simple Dual-Port Mode
All TriMatrix memory blocks support simple dual-port mode. Simple dual-port mode
allows you to perform one read and one write operation to different locations at the
same time. Write operation happens on port A; read operation happens on port B.
Figure 3–9 shows a simple dual-port configuration.
Figure 3–9. Stratix IV Simple Dual-Port Memory
data[ ]
wraddress[ ]
wren
byteena[]
wr_addressstall
wrclock
wrclocken
aclr
(1)
rdaddress[ ]
rden
q[ ]
rd_addressstall
rdclock
rdclocken
ecc_status
Note to Figure 3–9:
(1) Simple dual-port RAM supports input/output clock mode in addition to read/write clock mode.
Simple dual-port mode supports different read and write data widths (mixed-width
support). Table 3–5 lists the mixed width configurations for M9K blocks in simple
dual-port mode. MLABs do not have native support for mixed-width operation. The
Quartus II software implements mixed-width memories in MLABs by using more
than one MLAB.
Table 3–5. M9K Block Mixed-Width Configurations (Simple Dual-Port Mode) (Part 1 of 2)
Write Port
Read Port
8K × 1
4K × 2
2K × 4
1K × 8
512 × 16
256 × 32
1K × 9
512 × 18
256 × 36
8K × 1
Y
Y
Y
Y
Y
Y
—
—
—
4K × 2
Y
Y
Y
Y
Y
Y
—
—
—
2K × 4
Y
Y
Y
Y
Y
Y
—
—
—
December 2011
Altera Corporation
Stratix IV Device Handbook
Volume 1
3–12
Chapter 3: TriMatrix Embedded Memory Blocks in Stratix IV Devices
Memory Modes
Table 3–5. M9K Block Mixed-Width Configurations (Simple Dual-Port Mode) (Part 2 of 2)
Write Port
Read Port
8K × 1
4K × 2
2K × 4
1K × 8
512 × 16
256 × 32
1K × 9
512 × 18
256 × 36
1K × 8
Y
Y
Y
Y
Y
Y
—
—
—
512 × 16
Y
Y
Y
Y
Y
Y
—
—
—
256 × 32
Y
Y
Y
Y
Y
Y
—
—
—
1K × 9
—
—
—
—
—
—
Y
Y
Y
512 × 18
—
—
—
—
—
—
Y
Y
Y
256 × 36
—
—
—
—
—
—
Y
Y
Y
Table 3–6 lists the mixed-width configurations for M144K blocks in simple dual-port
mode.
Table 3–6. M144K Block Mixed-Width Configurations (Simple Dual-Port Mode)
Write Port
Read Port
16K × 8
8K × 16
4K × 32
2K × 64
16K × 9
8K × 18
4K × 36
2K × 72
16K × 8
Y
Y
Y
Y
—
—
—
—
8K × 16
Y
Y
Y
Y
—
—
—
—
4K × 32
Y
Y
Y
Y
—
—
—
—
2K × 64
Y
Y
Y
Y
—
—
—
—
16K × 9
—
—
—
—
Y
Y
Y
Y
8K × 18
—
—
—
—
Y
Y
Y
Y
4K × 36
—
—
—
—
Y
Y
Y
Y
2K × 72
—
—
—
—
Y
Y
Y
Y
In simple dual-port mode, M9K and M144K blocks support separate write-enable and
read-enable signals. You can save power by keeping the read-enable signal low
(inactive) when not reading. Read-during-write operations to the same address can
either output a “don’t care” value or “old data” value. To choose the desired behavior,
set the read-during-write behavior to either don’t care or old data in the RAM
MegaWizard Plug-In Manager in the Quartus II software. For more information, refer
to “Read-During-Write Behavior” on page 3–19.
MLABs only support a write-enable signal. For MLABs, you can set the same-port
read-during-write behavior to don’t care and the mixed-port read-during-write
behavior to either don’t care or old data. The available choices depend on the
configuration of the MLAB. There is no “new data” option for MLABs.
Stratix IV Device Handbook
Volume 1
December 2011 Altera Corporation
Chapter 3: TriMatrix Embedded Memory Blocks in Stratix IV Devices
Memory Modes
3–13
Figure 3–10 shows timing waveforms for read and write operations in simple
dual-port mode with unregistered outputs. Registering the RAM outputs simply
delays the q output by one clock cycle.
Figure 3–10. Simple Dual-Port Timing Waveforms
wrclock
wren
wraddress
an-1
data
din-1
a0
an
a1
a2
a3
din
a4
a5
din4
din5
a6
din6
rdclock
rden
rdaddress
q (asynch)
bn
b1
b0
doutn-1
b2
b3
dout0
doutn
Figure 3–11 shows timing waveforms for read and write operations in mixed-port
mode with unregistered outputs.
Figure 3–11. Mixed-Port Read-During-Write Timing Waveforms
wrclock
wren
wraddress
an-1
data
din-1
a0
an
a1
a2
din
a3
a4
a5
din4
din5
a6
din6
rdclock
rden
rdaddress
q (asynch)
December 2011
bn
doutn-1
Altera Corporation
b0
doutn
b1
b2
b3
dout0
Stratix IV Device Handbook
Volume 1
3–14
Chapter 3: TriMatrix Embedded Memory Blocks in Stratix IV Devices
Memory Modes
True Dual-Port Mode
Stratix IV M9K and M144K blocks support true dual-port mode. Sometimes called
bi-directional dual-port, this mode allows you to perform any combination of two
port operations: two reads, two writes, or one read and one write at two different
clock frequencies.
Figure 3–12 shows the true dual-port RAM configuration.
Figure 3–12. Stratix IV True Dual-Port Memory
(1)
data_a[ ]
address_a[ ]
wren_a
byteena_a[]
addressstall_a
clock_a
rden_a
aclr_a
q_a[]
data_b[ ]
address_b[]
wren_b
byteena_b[]
addressstall_b
clock_b
rden_b
aclr_b
q_b[]
Note to Figure 3–12:
(1) True dual-port memory supports input/output clock mode in addition to independent clock mode.
The widest bit configuration of the M9K and M144K blocks in true dual-port mode is
as follows:
■
M9K: 512 × 16-bit (or 512 ×18-bit with parity)
■
M144K: 4K × 32-bit (or 4K ×36-bit with parity)
Wider configurations are unavailable because the number of output drivers is
equivalent to the maximum bit width of the respective memory block. Because true
dual-port RAM has outputs on two ports, its maximum width equals half of the total
number of output drivers. Table 3–7 lists the possible M9K block mixed-port width
configurations in true dual-port mode.
Table 3–7. M9K Block Mixed-Width Configuration (True Dual-Port Mode)
Write Port
Read Port
Stratix IV Device Handbook
Volume 1
8K × 1
4K × 2
2K × 4
1K × 8
512 × 16
1K × 9 512 × 18
8K × 1
Y
Y
Y
Y
Y
—
—
4K × 2
Y
Y
Y
Y
Y
—
—
2K × 4
Y
Y
Y
Y
Y
—
—
1K × 8
Y
Y
Y
Y
Y
—
—
512 × 16
Y
Y
Y
Y
Y
—
—
1K × 9
—
—
—
—
—
Y
Y
512 × 18
—
—
—
—
—
Y
Y
December 2011 Altera Corporation
Chapter 3: TriMatrix Embedded Memory Blocks in Stratix IV Devices
Memory Modes
3–15
Table 3–8 lists the possible M144K block mixed-port width configurations in true
dual-port mode.
Table 3–8. M144K Block Mixed-Width Configurations (True Dual-Port Mode)
Write Port
Read Port
16K × 8
8K × 16
4K × 32
16K × 9
8K × 18
4K × 36
16K × 8
Y
Y
Y
—
—
—
8K × 16
Y
Y
Y
—
—
—
4K × 32
Y
Y
Y
—
—
—
16K × 9
—
—
—
Y
Y
Y
8K × 18
—
—
—
Y
Y
Y
4K × 36
—
—
—
Y
Y
Y
In true dual-port mode, M9K and M144K blocks support separate write-enable and
read-enable signals. You can save power by keeping the read-enable signal low
(inactive) when not reading. Read-during-write operations to the same address can
either output “new data” at that location or “old data”. To choose the desired
behavior, set the read-during-write behavior to either new data or old data in the
RAM MegaWizard Plug-In Manager in the Quartus II software. For more information,
refer to “Read-During-Write Behavior” on page 3–19.
In true dual-port mode, you can access any memory location at any time from either
port. When accessing the same memory location from both ports, you must avoid
possible write conflicts. A write conflict happens when you attempt to write to the
same address location from both ports at the same time. This results in unknown data
being stored to that address location. No conflict resolution circuitry is built into the
Stratix IV TriMatrix memory blocks. You must handle address conflicts external to the
RAM block.
Figure 3–13 shows true dual-port timing waveforms for the write operation at port A
and the read operation at port B, with the read-during-write behavior set to new data.
Registering the RAM’s outputs simply delays the q outputs by one clock cycle.
Figure 3–13. True Dual-Port Timing Waveform
clk_a
wren_a
address_a
an-1
an
data_a
din-1
din
q_a (asynch)
din-1
a0
din
a1
dout0
a2
dout1
a3
dout2
a4
a5
a6
din4
din5
din6
dout3
din4
din5
clk_b
wren_b
address_b
q_b (asynch)
December 2011
bn
doutn-1
Altera Corporation
b0
b1
b2
b3
doutn
dout0
dout1
dout2
Stratix IV Device Handbook
Volume 1
3–16
Chapter 3: TriMatrix Embedded Memory Blocks in Stratix IV Devices
Memory Modes
Shift-Register Mode
All Stratix IV memory blocks support shift register mode. Embedded memory block
configurations can implement shift registers for digital signal processing (DSP)
applications, such as finite impulse response (FIR) filters, pseudo-random number
generators, multi-channel filtering, and auto- and cross-correlation functions. These
and other DSP applications require local data storage, traditionally implemented with
standard flipflops that quickly exhaust many logic cells for large shift registers. A
more efficient alternative is to use embedded memory as a shift-register block, which
saves logic cell and routing resources.
The size of a shift register (w × m × n) is determined by the input data width (w), the
length of the taps (m), and the number of taps (n). You can cascade memory blocks to
implement larger shift registers.
Figure 3–14 shows the TriMatrix memory block in shift-register mode.
Figure 3–14. Shift-Register Memory Configuration
w x m x n Shift Register
m-Bit Shift Register
W
W
m-Bit Shift Register
W
W
n Number of Taps
m-Bit Shift Register
W
W
m-Bit Shift Register
W
Stratix IV Device Handbook
Volume 1
W
December 2011 Altera Corporation
Chapter 3: TriMatrix Embedded Memory Blocks in Stratix IV Devices
Clocking Modes
3–17
ROM Mode
All Stratix IV TriMatrix memory blocks support ROM mode. A .mif file initializes the
ROM contents of these blocks. The address lines of the ROM are registered on M9K
and M144K blocks, but can be unregistered on MLABs. The outputs can be registered
or unregistered. Output registers can be asynchronously cleared. The ROM read
operation is identical to the read operation in the single-port RAM configuration.
FIFO Mode
All TriMatrix memory blocks support FIFO mode. MLABs are ideal for designs with
many small, shallow FIFO buffers. To implement FIFO buffers in your design, use the
Quartus II software FIFO MegaWizard Plug-In Manager. Both single- and dual-clock
(asynchronous) FIFO buffers are supported.
f For more information about implementing FIFO buffers, refer to the SCFIFO and
DCFIFO Megafunctions User Guide.
1
MLABs do not support mixed-width FIFO mode.
Clocking Modes
Stratix IV TriMatrix memory blocks support the following clocking modes:
■
“Independent Clock Mode” on page 3–18
■
“Input/Output Clock Mode” on page 3–18
■
“Read/Write Clock Mode” on page 3–18
■
“Single Clock Mode” on page 3–18
c Violating the setup or hold time on the memory block address registers could corrupt
memory contents. This applies to both read and write operations.
Table 3–9 lists which clocking mode/memory mode combinations are supported.
Table 3–9. TriMatrix Memory Clock Modes
True
Dual-Port Mode
Simple
Dual-Port Mode
Single-Port Mode
ROM Mode
FIFO Mode
Independent
Y
—
—
Y
—
Input/output
Y
Y
Y
Y
—
Read/write
—
Y
—
—
Y
Single clock
Y
Y
Y
Y
Y
Clocking Mode
December 2011
Altera Corporation
Stratix IV Device Handbook
Volume 1
3–18
Chapter 3: TriMatrix Embedded Memory Blocks in Stratix IV Devices
Design Considerations
Independent Clock Mode
Stratix IV TriMatrix memory blocks can implement independent clock mode for true
dual-port memories. In this mode, a separate clock is available for each port (clock A
and clock B). Clock A controls all registers on the port A side; clock B controls all
registers on the port B side. Each port also supports independent clock enables for
both port A and port B registers, respectively. Asynchronous clears are supported
only for output latches and output registers on both ports.
Input/Output Clock Mode
Stratix IV TriMatrix memory blocks can implement input/output clock mode for true
dual-port and simple dual-port memories. In this mode, an input clock controls all
registers related to the data input to the memory block including data, address, byte
enables, read enables, and write enables. An output clock controls the data output
registers. Asynchronous clears are available on output latches and output registers
only.
Read/Write Clock Mode
Stratix IV TriMatrix memory blocks can implement read/write clock mode for simple
dual-port memories. In this mode, a write clock controls the data-input,
write-address, and write-enable registers. Similarly, a read clock controls the
data-output, read-address, and read-enable registers. The memory blocks support
independent clock enables for both read and write clocks. Asynchronous clears are
available on data output latches and registers only.
When using read/write clock mode, if you perform a simultaneous read/write to the
same address location, the output read data is unknown. If you require the output
data to be a known value, use either single-clock mode or input/output clock mode
and choose the appropriate read-during-write behavior in the MegaWizard Plug-In
Manager.
Single Clock Mode
Stratix IV TriMatrix memory blocks can implement single-clock mode for true
dual-port, simple dual-port, and single-port memories. In this mode, a single clock,
together with a clock enable, is used to control all registers of the memory block.
Asynchronous clears are available on output latches and output registers only.
Design Considerations
This section describes guidelines for designing with TriMatrix memory blocks.
Selecting TriMatrix Memory Blocks
The Quartus II software automatically partitions user-defined memory into
embedded memory blocks by taking into account both speed and size constraints
placed on your design. For example, the Quartus II software may spread memory out
across multiple memory blocks when resources are available to increase the
performance of the design. You can manually assign memory to a specific block size
using the RAM MegaWizard Plug-In Manager.
Stratix IV Device Handbook
Volume 1
December 2011 Altera Corporation
Chapter 3: TriMatrix Embedded Memory Blocks in Stratix IV Devices
Design Considerations
3–19
MLABs can implement single-port SRAM through emulation using the Quartus II
software. Emulation results in minimal additional logic resources being used. Because
of the dual-purpose architecture of the MLAB, it only has data input registers and
output registers in the block. MLABs gain input address registers and additional data
output registers from ALMs.
f For more information about register packing, refer to the Logic Array Blocks and
Adaptive Logic Modules in Stratix IV Devices chapter.
Conflict Resolution
When using memory blocks in true dual-port mode, it is possible to attempt two write
operations to the same memory location (address). Because no conflict resolution
circuitry is built into the memory blocks, this results in unknown data being written to
that location. Therefore, you must implement conflict resolution logic external to the
memory block to avoid address conflicts.
Read-During-Write Behavior
You can customize the read-during-write behavior of the Stratix IV TriMatrix memory
blocks to suit your design needs. Two types of read-during-write operations are
available: same port and mixed port. Figure 3–15 shows the difference between the
two types.
Figure 3–15. Stratix IV Read-During-Write Data Flow
Port A
data in
Port B
data in
Mixed-port
data flow
Same-port
data flow
Port A
data out
Port B
data out
Same-Port Read-During-Write Mode
This mode applies to either a single-port RAM or the same port of a true dual-port
RAM. For MLABs, the output of the MLABs can only be set to don’t care in same-port
read-during-write mode. In this mode, the output of the MLABs is unknown during a
write cycle. There is a window near the falling edge of the clock during which the
output is unknown. Prior to that window, “old data” is read out; after that window,
“new data” is seen at the output.
December 2011
Altera Corporation
Stratix IV Device Handbook
Volume 1
3–20
Chapter 3: TriMatrix Embedded Memory Blocks in Stratix IV Devices
Design Considerations
Figure 3–16 shows sample functional waveforms of same-port read-during-write
behavior in don’t care mode for MLABs.
Figure 3–16. MLABs Same-Port Read-During Write: Don’t Care Mode
clk_a
address
XX
A0
data_in
XX
FFFF
A1
A2
AAAA
XXXX
wrena
q(unregistered)
q(registered)
XX
FFFF
A0(old data)
AAAA
A1(old data)
XX
FFFF
A2(old data)
AAAA
For M9K and M144K memory blocks, three output choices are available in same-port
read-during-write mode: “new data” (or flow-through) or “old data”. In new data
mode, the “new data” is available on the rising edge of the same clock cycle on which
it was written. In old data mode, the RAM outputs reflect the “old data” at that
address before the write operation proceeds. In don’t care mode, the RAM outputs
“unknown values” for a read-during-write operation.
Figure 3–17 shows sample functional waveforms of same-port read-during-write
behavior in new data mode for M9K and M144K blocks.
Figure 3–17. M9K and M144K Blocks Same-Port Read-During-Write: New Data Mode
clk_a
0A
address
0B
rdena
wrena
bytenna
data_a
q_a (asyn)
Stratix IV Device Handbook
Volume 1
01
10
00
A123
B456
C789
XX23
B423
11
B423
DDDD
EEEE
DDDD
EEEE
FFFF
FFFF
December 2011 Altera Corporation
Chapter 3: TriMatrix Embedded Memory Blocks in Stratix IV Devices
Design Considerations
3–21
Figure 3–18 shows sample functional waveforms of same-port read-during-write
behavior in old data mode for M9K and M144K blocks.
Figure 3–18. M9K and M144K Blocks Same-Port Read-During-Write: Old Data Mode
clk_a
A0
address
A1
rdena
wrena
01
10
00
A123
B456
C789
bytenna
data_a
A0 (old data) DoldDold23
q_a (asyn)
11
DDDD
B423
EEEE
A1(old data)
FFFF
DDDD
EEEE
Mixed-Port Read-During-Write Mode
This mode applies to a RAM in simple or true dual-port mode that has one port
reading from and the other port writing to the same address location with the same
clock.
In this mode, you have two output choices if you use the output register: “old data,”
or “don’t care”. With MLABs, you also have the output register “new data.” In old
data mode, a read-during-write operation to different ports causes the RAM outputs
to reflect the “old data” at that address location. In don’t care mode, the same
operation results in a “don’t care” or “unknown” value on the RAM outputs.
f Read-during-write behavior is controlled with the RAM MegaWizard Plug-In
Manager. For more information, refer to the Internal Memory (RAM and ROM) User
Guide.
Figure 3–19 shows a sample functional waveform of mixed-port read-during-write
behavior for old data mode in MLABs.
Figure 3–19. MLABs Mixed-Port Read-During-Write: Old Data Mode
clk_a
wraddress
A0
A1
rdaddress
A0
A1
data_in
AAAA
BBBB
CCCC
DDDD
EEEE
FFFF
01
10
11
01
10
AAAA
AABB
A1(old data)
DDDD
wrena
byteena_a
q_b(registered)
December 2011
Altera Corporation
11
A0 (old data)
DDEE
Stratix IV Device Handbook
Volume 1
3–22
Chapter 3: TriMatrix Embedded Memory Blocks in Stratix IV Devices
Design Considerations
Figure 3–20 shows a sample functional waveform of mixed-port read-during-write
behavior for don’t care mode in MLABs.
Figure 3–20. MLABs Mixed-Port Read-During-Write: Don’t Care Mode
clk_a
wraddress
A0
A1
rdaddress
A0
A1
data_in
AAAA
BBBB
CCCC
DDDD
EEEE
FFFF
01
10
11
01
10
AAAA
AABB
CCBB
DDDD
DDEE
wrena
byteena_a
11
q_b(registered)
FFEE
Figure 3–21 shows a sample functional waveform of mixed-port read-during-write
behavior for old data mode in M9K and M144K blocks.
Figure 3–21. M9K and M144K Blocks Mixed-Port Read-During Write: Old Data Mode
clk_a&b
wrena
address_a
A1
A0
data_a
AAAA
BBBB
CCCC
bytenna
11
01
10
DDDD
EEEE
FFFF
11
rdenb
address_b
q_b_(asyn)
Stratix IV Device Handbook
Volume 1
A1
A0
A0 (old data)
AAAA
AABB
A1(old data)
DDDD
EEEE
December 2011 Altera Corporation
Chapter 3: TriMatrix Embedded Memory Blocks in Stratix IV Devices
Design Considerations
3–23
Figure 3–22 shows a sample functional waveform of mixed-port read-during-write
behavior for don’t care mode in M9K and M144K blocks.
Figure 3–22. M9K and M144K Blocks Mixed-Port Read-During Write: Don’t Care Mode
clk_a&b
wrena
A0
address_a
A1
data_a
AAAA
BBBB
CCCC
bytenna
11
01
10
DDDD
EEEE
FFFF
11
rdenb
address_b
q_b_(asyn)
A0
A1
XXXX (unknown data)
Mixed-port read-during-write is not supported when two different clocks are used in
a dual-port RAM. The output value is unknown during a dual-clock mixed-port
read-during-write operation.
Power-Up Conditions and Memory Initialization
M9K memory cells are initialized to all zeros through a default .mif file in the
Quartus II software. However, you may specify your own initialization of the
memory cells through a defined .mif file. M144K memory cells are not initialized and;
therefore, come up in an undefined state. This is to prevent the programming file from
being too large. Again, you may specify your own initialization of the memory cells
through a defined .mif file.
MLABs power up to zero if output registers are used and power up reading the
memory contents if output registers are not used. You must take this into
consideration when designing logic that might evaluate the initial power-up values of
the MLAB memory block. For Stratix IV devices, the Quartus II software initializes
the RAM cells to zero unless there is a .mif file specified.
As mentioned, all memory blocks support initialization using a .mif file. You can
create .mif files in the Quartus II software and specify their use with the RAM
MegaWizard Plug-In Manager when instantiating a memory in your design. Even if a
memory is pre-initialized (for example, using a .mif file), it still powers up with its
outputs cleared.
f For more information about .mif files, refer to the Internal Memory (RAM and ROM)
User Guide and the Quartus II Handbook.
December 2011
Altera Corporation
Stratix IV Device Handbook
Volume 1
3–24
Chapter 3: TriMatrix Embedded Memory Blocks in Stratix IV Devices
Design Considerations
Power Management
Stratix IV memory block clock-enables allow you to control clocking of each memory
block to reduce AC power consumption. Use the read-enable signal to ensure that
read operations only occur when you need them to. If your design does not need
read-during-write, you can reduce your power consumption by de-asserting the
read-enable signal during write operations, or any period when no memory
operations occur.
The Quartus II software automatically places any unused memory blocks in
low-power mode to reduce static power.
Document Revision History
Table 3–10 lists the revision history for this chapter.
Table 3–10. Document Revision History
Date
Version
December 2011
February 2011
March 2010
November 2009
June 2009
April 2009
3.3
3.2
3.1
3.0
Changes
■
Updated the “Byte Enable Support” and “Mixed-Port Read-During-Write Mode” sections.
■
Updated Table 3–1.
■
Updated the “Byte Enable Support” and “Power-Up Conditions and Memory Initialization”
sections.
■
Applied new template.
■
Minor text edits.
■
Updated the “Simple Dual-Port Mode”, “Same-Port Read-During-Write Mode”, and
“Mixed-Port Read-During-Write Mode” sections.
■
Updated Figure 3–14.
■
Minor text edits.
■
Updated Table 3–2.
■
Updated the “Simple Dual-Port Mode” section.
■
Minor text edits.
■
Updated graphics.
■
Updated Table 3–1 and Figure 3–2.
■
Updated the “Introduction”, “Byte Enable Support”, “Mixed Width Support”,
“Asynchronous Clear”, “Single-Port RAM”, “Simple Dual-Port Mode”, “True Dual-Port
Mode”, “FIFO Mode”, and “Read/Write Clock Mode” sections.
■
Added introductory sentences to improve search ability.
■
Removed the Conclusion section.
■
Minor text edits.
■
Updated Table 3–2.
■
Updated Table 3–2.
■
Removed “Referenced Documents” section.
2.3
2.2
March 2009
2.1
November 2008
2.0
Updated “Power-Up Conditions and Memory Initialization” on page 3–20
May 2008
1.0
Initial release.
Stratix IV Device Handbook
Volume 1
December 2011 Altera Corporation
4. DSP Blocks in Stratix IV Devices
February 2011
SIV51004-3.1
SIV51004-3.1
This chapter describes how the Stratix® IV device digital signal processing (DSP)
blocks are optimized to support DSP applications requiring high data throughput,
such as finite impulse response (FIR) filters, infinite impulse response (IIR) filters, fast
Fourier transform (FFT) functions, and encoders. You can configure the DSP blocks to
implement one of several operational modes to suit your application. The built-in
shift register chain, multipliers, and adders/subtractors minimize the amount of
external logic to implement these functions, resulting in efficient resource usage and
improved performance and data throughput for DSP applications.
Many complex systems, such as WiMAX, 3GPP WCDMA, high-performance
computing (HPC), voice over Internet protocol (VoIP), H.264 video compression,
medical imaging, and HDTV use sophisticated digital signal processing techniques,
which typically require a large number of mathematical computations. Stratix IV
devices are ideally suited for these tasks because the DSP blocks consist of a
combination of dedicated elements that perform multiplication, addition, subtraction,
accumulation, summation, and dynamic shift operations.
Along with the high-performance Stratix IV soft logic fabric and TriMatrix memory
structures, you can configure DSP blocks to build sophisticated fixed-point and
floating-point arithmetic functions. These can be manipulated easily to implement
common, larger computationally intensive subsystems such as FIR filters, complex
FIR filters, IIR filters, FFT functions, and discrete cosine transform (DCT) functions.
This chapter contains the following sections:
■
“Stratix IV DSP Block Overview” on page 4–2
■
“Stratix IV Simplified DSP Operation” on page 4–4
■
“Stratix IV Operational Modes Overview” on page 4–8
■
“Stratix IV DSP Block Resource Descriptions” on page 4–9
■
“Stratix IV Operational Mode Descriptions” on page 4–15
■
“Software Support” on page 4–35
© 2011 Altera Corporation. All rights reserved. ALTERA, ARRIA, CYCLONE, HARDCOPY, MAX, MEGACORE, NIOS, QUARTUS and STRATIX words and logos
are trademarks of Altera Corporation and registered in the U.S. Patent and Trademark Office and in other countries. All other words and logos identified as
trademarks or service marks are the property of their respective holders as described at www.altera.com/common/legal.html. Altera warrants performance of its
semiconductor products to current specifications in accordance with Altera's standard warranty, but reserves the right to make changes to any products and
services at any time without notice. Altera assumes no responsibility or liability arising out of the application or use of any information, product, or service
described herein except as expressly agreed to in writing by Altera. Altera customers are advised to obtain the latest version of device specifications before relying
on any published information and before placing orders for products or services.
ISO
9001:2008
Registered
Stratix IV Device Handbook
Volume 1
February 2011
Feedback Subscribe
4–2
Chapter 4: DSP Blocks in Stratix IV Devices
Stratix IV DSP Block Overview
Stratix IV DSP Block Overview
Each Stratix IV device has two to seven columns of DSP blocks that implement
multiplication, multiply-add, multiply-accumulate (MAC), and dynamic shift
functions efficiently. Architectural highlights of the Stratix IV DSP block include:
■
High-performance, power optimized, fully registered, and pipelined
multiplication operations
■
Natively supported 9-, 12-, 18-, and 36-bit wordlengths
■
Natively supported 18-bit complex multiplications
■
Efficiently supported floating-point arithmetic formats (24-bit for single precision
and 53-bit for double precision)
■
Signed and unsigned input support
■
Built-in addition, subtraction, and accumulation units to combine multiplication
results efficiently
■
Cascading 18-bit input bus to form the tap-delay line for filtering applications
■
Cascading 44-bit output bus to propagate output results from one block to the next
block without external logic support
■
Rich and flexible arithmetic rounding and saturation units
■
Efficient barrel shifter support
■
Loopback capability to support adaptive filtering
Table 4–1 lists the number of DSP blocks for the Stratix IV device family.
Family
Stratix IV E
Stratix IV GX
DSP Blocks
Table 4–1. Number of DSP Blocks in Stratix IV Devices (Part 1 of 2)
Device
Independent Input and Output Multiplication Operators
High-Precision
Multiplier
Adder Mode
Four
Multiplier
Adder
Mode
9×9
Multipliers
12 × 12
Multipliers
18 × 18
Multipliers
18 × 18
Complex
36 × 36
Multipliers
18 × 36
Multipliers
18 × 18
Multipliers
EP4SE230
161
1,288
966
644
322
322
644
1288
EP4SE360
130
1,040
780
520
260
260
520
1040
EP4SE530
128
1,024
768
512
256
256
512
1024
EP4SE820
120
960
720
480
240
240
480
960
EP4SGX70
48
384
288
192
96
96
192
384
EP4SGX110
64
512
384
256
128
128
256
512
EP4SGX180
115
920
690
460
230
230
460
920
EP4SGX230
161
1,288
966
644
322
322
644
1288
EP4SGX290
104
832
624
416
208
208
416
832
EP4SGX360
(1)
130
1,040
780
520
260
260
520
1,040
EP4SGX360
(2)
128
1,024
768
512
256
256
512
1,024
128
1,024
768
512
256
256
512
1,024
EP4SGX530
Stratix IV Device Handbook
Volume 1
February 2011 Altera Corporation
Chapter 4: DSP Blocks in Stratix IV Devices
Stratix IV DSP Block Overview
4–3
Family
DSP Blocks
Table 4–1. Number of DSP Blocks in Stratix IV Devices (Part 2 of 2)
Device
Stratix IV GT
Independent Input and Output Multiplication Operators
High-Precision
Multiplier
Adder Mode
Four
Multiplier
Adder
Mode
9×9
Multipliers
12 × 12
Multipliers
18 × 18
Multipliers
18 × 18
Complex
36 × 36
Multipliers
18 × 36
Multipliers
18 × 18
Multipliers
EP4S40G2
161
1,288
966
644
322
322
644
1,288
EP4S40G5
128
1,024
768
512
256
256
512
1,024
EP4S100G2
161
1,288
966
644
322
322
644
1,288
EP4S100G3
104
832
624
416
208
208
416
832
EP4S100G4
128
1,024
768
512
256
256
512
1,024
EP4S100G5
128
1,024
768
512
256
256
512
1,024
Notes to Table 4–1:
(1) This is applicable for all packages in EP4SGX360 except F1932.
(2) This is applicable for EP4SGX360F1932 only.
Table 4–1 shows that the largest Stratix IV DSP-centric device provides up to 1288
18 × 18 multiplier functionality in the 36 × 36, complex 18 × 18, and summation
modes.
Each DSP block occupies four LABs in height and can be divided further into two half
blocks that share some common clock signals, but are for all common purposes
identical in functionality. Figure 4–1 shows the layout of each DSP block.
Figure 4–1. Overview of DSP Block Signals
34
Control
144
Input
Data
Half-DSP Block
72
Output
Data
72
Output
Data
288
144
Half-DSP Block
Full DSP Block
February 2011
Altera Corporation
Stratix IV Device Handbook
Volume 1
4–4
Chapter 4: DSP Blocks in Stratix IV Devices
Stratix IV Simplified DSP Operation
Stratix IV Simplified DSP Operation
In Stratix IV devices, the fundamental building block is a pair of 18 × 18-bit
multipliers followed by a first-stage 37-bit addition/subtraction unit, as shown in
Equation 4–1 and Figure 4–2.
1
All signed numbers, input, and output data are represented in 2’s-complement format
only.
Equation 4–1. Multiplier Equation
P[36..0] = A0[17..0] × B0[17..0] ± A1[17..0] × B1[17..0]
Figure 4–2. Basic Two-Multiplier Adder Building Block
A0[17..0]
B0[17..0]
+/A1[17..0]
D
Q
B1[17..0]
D
Q
P[36..0]
The structure shown in Figure 4–2 is useful for building more complex structures,
such as complex multipliers and 36 × 36 multipliers, as described in later sections.
Each Stratix IV DSP block contains four two-multiplier adder units (2 two-multiplier
adder units per half block). Therefore, there are eight 18 × 18 multiplier functionalities
per DSP block.
Stratix IV Device Handbook
Volume 1
February 2011 Altera Corporation
Chapter 4: DSP Blocks in Stratix IV Devices
Stratix IV Simplified DSP Operation
4–5
Following the two-multiplier adder units are the pipeline registers, the second-stage
adders, and an output register stage. You can configure the second-stage adders to
provide the alternative functions per half block, as shown in Equation 4–2 and
Equation 4–3.
Equation 4–2. Four-Multiplier Adder Equation
Z[37..0] = P0[36..0] + P1[36..0]
Equation 4–3. Four-Multiplier Adder Equation (44-Bit Accumulation)
Wn[43..0] = Wn-1[43..0] ± Zn[37..0]
In these equations, n denotes sample time and P[36..0] denotes the result from the
two-multiplier adder units.
Equation 4–2 provides a sum of four 18 × 18-bit multiplication operations
(four-multiplier adder). Equation 4–3 provides a four 18 × 18-bit multiplication
operation but with a maximum 44-bit accumulation capability by feeding the output
of the unit back to itself, as shown in Figure 4–3.
Depending on the mode you select, you can bypass all register stages except
accumulation and loopback mode. In these two modes, one set of registers must be
enabled. If the register set is not enabled, an infinite loop occurs.
Output Register Bank
Adder/
Accumulator
144
Pipeline Register Bank
Input
Data
Input Register Bank
Figure 4–3. Four-Multiplier Adder and Accumulation Capability
44
Result[]
Half-DSP Block
February 2011
Altera Corporation
Stratix IV Device Handbook
Volume 1
4–6
Chapter 4: DSP Blocks in Stratix IV Devices
Stratix IV Simplified DSP Operation
To support commonly found FIR-like structures efficiently, a major addition to the
DSP block in Stratix IV devices is the ability to propagate the result of one half block
to the next half block completely within the DSP block without additional soft logic
overhead. This is achieved by the inclusion of a dedicated addition unit and routing
that adds the 44-bit result of a previous half block with the 44-bit result of the current
block. The 44-bit result is either fed to the next half block or out of the DSP block using
the output register stage, as shown in Figure 4–4. Detailed examples are described in
later sections.
The combination of a fast, low-latency four-multiplier adder unit and the “chained
cascade” capability of the output chaining adder provides the optimal FIR and vector
multiplication capability.
To support single-channel type FIR filters efficiently, you can configure one of the
multiplier input’s registers to form a tap delay line input, saving resources and
providing higher system performance.
Figure 4–4. Output Cascading Feature for FIR Structures
From Previous Half DSP Block
Half DSP Block
Output Register Bank
Round/Saturate
Adder/
Accumulator
144
Pipeline Register Bank
Input
Data
Input Register Bank
44
44
Result[]
44
To Next
Half DSP Block
Also shown in Figure 4–4 is the optional rounding and saturation unit (RSU). This
unit provides a rich set of commonly found arithmetic rounding and saturation
functions used in signal processing.
In addition to the independent multipliers and sum modes, you can use DSP blocks to
perform shift operations. DSP blocks can dynamically switch between logical shift
left/right, arithmetic shift left/right, and rotation operation in one clock cycle.
Stratix IV Device Handbook
Volume 1
February 2011 Altera Corporation
Chapter 4: DSP Blocks in Stratix IV Devices
Stratix IV Simplified DSP Operation
4–7
Figure 4–5 shows a top-level view of the Stratix IV DSP block.
Figure 4–6 on page 4–9 shows a more detailed top-level view of the DSP block.
Figure 4–5. Stratix IV Full DSP Block
From Previous
Half DSP Block
Output Multiplexer
Round/Saturate
Output Register Bank
Output Multiplexter
Round/Saturate
Output Register Bank
Adder/Accumulator
144
Pipeline Register Bank
Input
Data
Input Register Bank
44
Result[]
Top Half DSP Block
Adder/Accumulator
144
Pipeline Register Bank
Input
Data
Input Register Bank
44
Result[]
Bottom Half DSP Block
To Next Half DSP Block
February 2011
Altera Corporation
Stratix IV Device Handbook
Volume 1
4–8
Chapter 4: DSP Blocks in Stratix IV Devices
Stratix IV Operational Modes Overview
Stratix IV Operational Modes Overview
You can use each Stratix IV DSP block in one of five basic operational modes.
Table 4–2 lists the five basic operational modes and the number of multipliers that you
can implement within a single DSP block, depending on the mode.
Table 4–2. Stratix IV DSP Block Operation Modes
Multiplier
in Width
# of
Mults
# per
Block
Signed or
Unsigned
RND,
SAT
In Shift
Register
Chainout
Adder
1st Stage
Add/Sub
2nd
Stage
Add/Acc
9 bits
1
8
Both
No
No
No
—
—
12 bits
1
6
Both
No
No
No
—
—
18 bits
1
4
Both
Yes
Yes
No
—
—
36 bits
1
2
Both
No
No
No
—
—
Double
1
2
Both
No
No
No
—
—
Two-Multiplier
Adder (1)
18 bits
2
4
Signed
Yes
No
No
Both
—
Four-Multiplier
Adder
18 bits
4
2
Both
Yes
Yes
Yes
Both
Add Only
Multiply
Accumulate
18 bits
4
2
Both
Yes
Yes
Yes
Both
Both
1
2
Both
No
No
—
—
—
2
2
Both
No
No
No
—
Add Only
Mode
Independent
Multiplier
Shift
(2)
High Precision
Multiplier Adder
36 bits
(3)
1836
(4)
Notes to Table 4–2:
(1) This mode also supports loopback mode. In loopback mode, the number of loopback multipliers per DSP block is two. You can use the
remaining multipliers in regular two-multiplier adder mode.
(2) Dynamic shift mode supports arithmetic shift left, arithmetic shift right, logical shift left, logical shift right, and rotation operation.
(3) Dynamic shift mode operates on a 32-bit input vector but the multiplier width is configured as 36 bits.
(4) Unsigned value is also supported but you must ensure that the result can be contained within 36 bits.
The DSP block consists of two identical halves (the top half and bottom half). Each
half has four 18 × 18 multipliers.
The Quartus® II software includes megafunctions used to control the mode of
operation of the multipliers. After making the appropriate parameter settings using
the megafunction’s MegaWizard Plug-In Manager, the Quartus II software
automatically configures the DSP block.
Stratix IV DSP blocks can operate in different modes simultaneously. Each half block
is fully independent except for the sharing of the three clock, ena, and aclr signals.
For example, you can break down a single DSP block to operate a 9 × 9 multiplier in
one half block and an 18 × 18 two-multiplier adder in the other half block. This
increases DSP block resource efficiency and allows you to implement more
multipliers within a Stratix IV device. The Quartus II software automatically places
multipliers that can share the same DSP block resources within the same block.
Stratix IV Device Handbook
Volume 1
February 2011 Altera Corporation
Chapter 4: DSP Blocks in Stratix IV Devices
Stratix IV DSP Block Resource Descriptions
4–9
Stratix IV DSP Block Resource Descriptions
The DSP block consists of the following elements:
■
Input register bank
■
Four two-multiplier adders
■
Pipeline register bank
■
Two second-stage adders
■
Four rounding and saturation logic units
■
Second adder register and output register bank
Figure 4–6 shows a detailed overall architecture of the top half of the DSP block.
Table 4–9 on page 4–34 shows a list of DSP block dynamic signals.
Figure 4–6. Half DSP Block Architecture
clock[3..0]
ena[3..0]
alcr[3..0]
chainin[ ] (3)
signa
signb
output_round
output_saturate
rotate
shift_right
zero_loopback
accum_sload
zero_chainout
chainout_round
chainout_saturate
overflow (1)
chainout_sat_overflow (2)
scanina[ ]
datab_3[ ]
Multiplexer
Shift/Rotate
Output Register Bank
Second Round/Saturate
Chainout Adder
Second Adder Register Bank
dataa_3[ ]
First Round/Saturate
datab_2[ ]
Second Stage Adder/Accumulator
dataa_2[ ]
Pipeline Register Bank
datab_1[ ]
Input Register Bank
datab_0[ ]
dataa_1[ ]
First Stage Adder
loopback
First Stage Adder
dataa_0[ ]
result[ ]
Half-DSP Block
scanouta
chainout
Notes to Figure 4–6:
(1) Block output for accumulator overflow and saturate overflow.
(2) Block output for saturation overflow of chainout.
(3) The chainin port must only be connected to chainout of the previous DSP blocks and must not be connected to general routings.
February 2011
Altera Corporation
Stratix IV Device Handbook
Volume 1
4–10
Chapter 4: DSP Blocks in Stratix IV Devices
Stratix IV DSP Block Resource Descriptions
Input Registers
All of the DSP block registers are triggered by the positive edge of the clock signal and
are cleared after power up. Each multiplier operand can feed an input register or go
directly to the multiplier, bypassing the input registers. The following DSP block
signals control the input registers within the DSP block:
■
clock[3..0]
■
ena[3..0]
■
aclr[3..0]
Every DSP block has nine 18-bit data input register banks per half DSP block. Every
half DSP block has the option to use the eight data register banks as inputs to the four
multipliers. The special ninth register bank is a delay register required by modes that
use both the cascade and chainout features of the DSP block. Use the ninth register
bank to balance the latency requirements when using the chained cascade feature.
Stratix IV Device Handbook
Volume 1
February 2011 Altera Corporation
Chapter 4: DSP Blocks in Stratix IV Devices
Stratix IV DSP Block Resource Descriptions
4–11
A feature of the input register bank is to support a tap delay line. Therefore, the top
leg of the multiplier input (A) can be driven from general routing or from the cascade
chain, as shown in Figure 4–7. Table 4–9 on page 4–34 lists the DSP block dynamic
signals.
Figure 4–7. Input Register of a Half DSP Block
clock[3..0]
ena[3..0]
aclr[3..0]
signa
signb
scanina[17..0]
dataa_0[17..0]
loopback
datab_0[17..0]
+/-
dataa_1[17..0]
datab_1[17..0]
dataa_2[17..0]
datab_2[17..0]
+/-
dataa_3[17..0]
datab_3[17..0]
Delay
Register
scanouta
At compile time, you must select whether the A-input comes from general routing or
from the cascade chain. In cascade mode, the dedicated shift outputs from one
multiplier block and directly feeds the input registers of the adjacent multiplier below
it (within the same half DSP block) or the first multiplier in the next half DSP block, to
form an 8-tap shift register chain per DSP Block. The DSP block can increase the
length of the shift register chain by cascading to the lower DSP blocks. The dedicated
shift register chain spans a single column, but you can implement longer shift register
chains requiring multiple columns using the regular FPGA routing resources.
February 2011
Altera Corporation
Stratix IV Device Handbook
Volume 1
4–12
Chapter 4: DSP Blocks in Stratix IV Devices
Stratix IV DSP Block Resource Descriptions
Shift registers are useful in DSP functions such as FIR filters. When implementing
18 × 18 or smaller width multipliers, you do not need external logic to create the shift
register chain because the input shift registers are internal to the DSP block. This
implementation significantly reduces the logical element (LE) resources required,
avoids routing congestion, and results in predictable timing.
The first multiplier in every half DSP block (top- and bottom-half) in Stratix IV
devices has a multiplexer for the first multiplier B-input (lower-leg input) register to
select between general routing and loopback, as shown in Figure 4–6 on page 4–9. In
loopback mode, the most significant 18-bit registered outputs are connected as
feedback to the multiplier input of the first top multiplier in each half DSP block.
Loopback modes are used by recursive filters where the previous output is needed to
compute the current output.
Loopback mode is described in “Two-Multiplier Adder Sum Mode” on page 4–22.
Table 4–3 lists input register modes for the DSP block.
Table 4–3. Input Register Modes
Register Input Mode
(1)
Parallel input
Shift register input
Loopback input
(2)
(3)
9×9
12 × 12
18 × 18
36 × 36
Double
Y
Y
Y
Y
Y
—
—
Y
—
—
—
—
Y
—
—
Notes to Table 4–3:
(1) Multiplier operand input wordlengths are statically configured at compile time.
(2) Available only on the A-operand.
(3) Only one loopback input is allowed per half block. For more information, refer to Figure 4–15 on page 4–24.
Multiplier and First-Stage Adder
The multiplier stage natively supports 9 × 9, 12 × 12, 18 × 18, or 36 × 36 multipliers.
Other wordlengths are padded up to the nearest appropriate native wordlength; for
example, 16 × 16 would be padded up to use 18 × 18. For more information, refer to
“Independent Multiplier Modes” on page 4–15. Depending on the data width of the
multiplier, a single DSP block can perform many multiplications in parallel.
Each multiplier operand can be a unique signed or unsigned number. Two dynamic
signals, signa and signb, control the representation of each operand, respectively. A
logic 1 value on the signa/signb signal indicates that data A/data B is a signed
number; a logic 0 value indicates an unsigned number. Table 4–4 lists the sign of the
multiplication result for the various operand sign representations. The result of the
multiplication is signed if any one of the operands is a signed value.
Table 4–4. Multiplier Sign Representation
Stratix IV Device Handbook
Volume 1
Data A (signa Value)
Data B (signb Value)
Result
Unsigned (logic 0)
Unsigned (logic 0)
Unsigned
Unsigned (logic 0)
Signed (logic 1)
Signed
Signed (logic 1)
Unsigned (logic 0)
Signed
Signed (logic 1)
Signed (logic 1)
Signed
February 2011 Altera Corporation
Chapter 4: DSP Blocks in Stratix IV Devices
Stratix IV DSP Block Resource Descriptions
4–13
Each half block has its own signa and signb signal. Therefore, all of the data A inputs
feeding the same half DSP block must have the same sign representation. Similarly, all
of the data B inputs feeding the same half DSP block must have the same sign
representation. The multiplier offers full precision regardless of the sign
representation in all operational modes except for full precision 18 × 18 loopback and
two-multiplier adder modes. For more information, refer to “Two-Multiplier Adder
Sum Mode” on page 4–22.
1
By default, when the signa and signb signals are unused, the Quartus II software sets
the multiplier to perform unsigned multiplication.
Figure 4–6 on page 4–9 shows that the outputs of the multipliers are the only outputs
that can feed into the first-stage adder. There are four first-stage adders in a DSP block
(two adders per half DSP block). The first-stage adder block has the ability to perform
addition and subtraction. The control signal for addition or subtraction is static and
has to be configured after compile time. The first-stage adders are used by the sum
modes to compute the sum of two multipliers, 18 × 18-complex multipliers, and to
perform the first stage of a 36 × 36 multiply and shift operations.
Depending on your specifications, the output of the first-stage adder has the option to
feed into the pipeline registers, second-stage adder, rounding and saturation unit, or
output registers.
Pipeline Register Stage
Figure 4–6 on page 4–9 shows that the output from the first-stage adder can either
feed or bypass the pipeline registers. Pipeline registers increase the DSP block’s
maximum performance (at the expense of extra cycles of latency), especially when
using the subsequent DSP block stages. Pipeline registers split up the long signal path
between the input registers/multiplier/first-stage adder and the second-stage adder/
round-and-saturation/output registers, creating two shorter paths.
Second-Stage Adder
There are four individual 44-bit second-stage adders per DSP block (two adders
per half DSP block). You can configure the second-stage adders as follows:
1
■
The final stage of a 36-bit multiplier
■
A sum of four (18 × 18)
■
An accumulator (44-bits maximum)
■
A chained output summation (44-bits maximum)
You can use the chained-output adder at the same time as a second-level adder in
chained output summation mode.
The output of the second-stage adder has the option to go into the rounding and
saturation logic unit or the output register.
1
February 2011
You cannot use the second-stage adder independently from the multiplier and
first-stage adder.
Altera Corporation
Stratix IV Device Handbook
Volume 1
4–14
Chapter 4: DSP Blocks in Stratix IV Devices
Stratix IV DSP Block Resource Descriptions
Rounding and Saturation Stage
The rounding and saturation logic units are located at the output of the 44-bit
second-stage adder (the rounding logic unit followed by the saturation logic unit).
There are two rounding and saturation logic units per half DSP block. The input to the
rounding and saturation logic unit can come from one of the following stages:
■
Output of the multiplier (independent multiply mode in 18 × 18)
■
Output of the first-stage adder (two-multiplier adder)
■
Output of the pipeline registers
■
Output of the second-stage adder (four-multiplier adder and multiply-accumulate
mode in 18 × 18)
These stages are described in “Stratix IV Operational Mode Descriptions” on
page 4–15.
The rounding and saturation logic unit is controlled by the dynamic rounding and
saturate signals, respectively. A logic 1 value on the rounding and/or saturate
signals enables the rounding and/or saturate logic unit, respectively.
1
You can use the rounding and saturation logic units together or independently.
Second Adder and Output Registers
The second adder register and output register banks are two banks of 44-bit registers
that you can combine to form larger 72-bit banks to support 36 × 36 output results.
The outputs of the different stages in the Stratix IV devices are routed to the output
registers through an output selection unit. Depending on the operational mode of the
DSP block, the output selection unit selects whether the outputs of the DSP blocks
comes from the outputs of the multiplier block, first-stage adder, pipeline registers,
second-stage adder, or the rounding and saturation logic unit. The output selection
unit is set automatically by the software, based on the DSP block operational mode
you specified, and has the option to either drive or bypass the output registers. The
exception is when you use the block in shift mode, in which case you dynamically
control the output-select multiplexer directly.
When the DSP block is configured in chained cascaded output mode, both of the
second-stage adders are used. Use the first one for performing a four-multiplier
adder; use the second for the chainout adder.
The outputs of the four-multiplier adder are routed to the second-stage adder
registers before they enter the chainout adder. The output of the chainout adder goes
to the regular output register bank. Depending on the configuration, you can route
the chainout results to the input of the next half block’s chainout adder input or to the
general fabric (functioning as regular output registers). For more information, refer to
“Stratix IV Operational Mode Descriptions” on page 4–15.
Stratix IV Device Handbook
Volume 1
February 2011 Altera Corporation
Chapter 4: DSP Blocks in Stratix IV Devices
Stratix IV Operational Mode Descriptions
4–15
The second-stage and output registers are triggered by the positive edge of the clock
signal and are cleared after power up. The following DSP block signals control the
output registers within the DSP block:
■
clock[3..0]
■
ena[3..0]
■
aclr[3..0]
Stratix IV Operational Mode Descriptions
This section contains an explanation of different operational modes in Stratix IV
devices.
Independent Multiplier Modes
In independent input and output multiplier mode, the DSP block performs individual
multiplication operations for general-purpose multipliers.
9-, 12-, and 18-Bit Multiplier
You can configure each DSP block multiplier for 9-, 12-, or 18-bit multiplication. A
single DSP block can support up to eight individual 9 × 9 multipliers, six individual
12 × 12 multipliers, or four individual 18 × 18 multipliers. For operand widths up to
9 bits, a 9 × 9 multiplier is implemented. For operand widths from 10 to 12 bits, a
12 × 12 multiplier is implemented, and for operand widths from 13 to 18 bits, an
18 × 18 multiplier is implemented. This is done by the Quartus II software by
zero-padding the LSBs. Figure 4–8, Figure 4–9, and Figure 4–10 show the DSP block in
the independent multiplier operation. Table 4–9 on page 4–34 lists the dynamic
signals for the DSP block.
February 2011
Altera Corporation
Stratix IV Device Handbook
Volume 1
4–16
Chapter 4: DSP Blocks in Stratix IV Devices
Stratix IV Operational Mode Descriptions
Figure 4–8. 18-Bit Independent Multiplier Mode Shown for a Half DSP Block
signa
clock[3..0]
signb
overflow (1)
Pipeline Register Bank
18
dataa_1[17..0]
Input Register Bank
18
datab_0[17..0]
18
datab_1[17..0]
36
result_0[ ]
Output Register Bank
18
dataa_0[17..0]
Round/Saturate
output_round
output_saturate
Round/Saturate
ena[3..0]
aclr[3..0]
36
result_1[ ]
Half-DSP Block
Note to Figure 4–8:
(1) Block output for accumulator overflow and saturate overflow.
Stratix IV Device Handbook
Volume 1
February 2011 Altera Corporation
Chapter 4: DSP Blocks in Stratix IV Devices
Stratix IV Operational Mode Descriptions
4–17
Figure 4–9. 12-Bit Independent Multiplier Mode Shown for a Half DSP Block
clock[3..0]
ena[3..0]
aclr[3..0]
signa
signb
12
dataa_0[11..0]
24
result_0[ ]
12
Output Register Bank
12
datab_1[11..0]
Pipeline Register Bank
12
dataa_1[11..0]
Input Register Bank
datab_0[11..0]
24
result_1[ ]
12
dataa_2[11..0]
24
result_2[ ]
12
datab_2[11..0]
Half-DSP Block
February 2011
Altera Corporation
Stratix IV Device Handbook
Volume 1
4–18
Chapter 4: DSP Blocks in Stratix IV Devices
Stratix IV Operational Mode Descriptions
Figure 4–10. 9-Bit Independent Multiplier Mode Shown for a Half Block
clock[3..0]
ena[3..0]
aclr[3..0]
signa
signb
9
dataa_0[8..0]
18
result_0[ ]
9
datab_0[8..0]
9
dataa_1[8..0]
Output Register Bank
9
dataa_2[8..0]
Pipeline Register Bank
9
datab_1[8..0]
Input Register Bank
18
result_1[ ]
18
result_2[ ]
9
datab_2[8..0]
9
dataa_3[8..0]
18
result_3[ ]
9
datab_3[8..0]
Half-DSP Block
The multiplier operands can accept signed integers, unsigned integers, or a
combination of both. You can change the signa and signb signals dynamically and
can register the signals in the DSP block. Additionally, the multiplier inputs and
results can be registered independently. You can use the pipeline registers within the
DSP block to pipeline the multiplier result, increasing the performance of the DSP
block.
1
Stratix IV Device Handbook
Volume 1
The rounding and saturation logic unit is supported for 18-bit independent multiplier
mode only.
February 2011 Altera Corporation
Chapter 4: DSP Blocks in Stratix IV Devices
Stratix IV Operational Mode Descriptions
4–19
36-Bit Multiplier
You can efficiently construct a 36 × 36 multiplier using four 18 × 18 multipliers. This
simplification fits conveniently into one half DSP block and is implemented in the
DSP block automatically by selecting 36 × 36 mode. Stratix IV devices can have up to
two 36-bit multipliers per DSP block (one 36-bit multiplier per half DSP block). The
36-bit multiplier is also under the independent multiplier mode but uses the entire
half DSP block, including the dedicated hardware logic after the pipeline registers to
implement the 36 × 36 bit multiplication operation, as shown in Figure 4–11.
The 36-bit multiplier is useful for applications requiring more than 18-bit precision;
for example, for the mantissa multiplication portion of single precision and extended
single precision floating-point arithmetic applications.
Figure 4–11. 36-Bit Independent Multiplier Mode Shown for a Half DSP Block
clock[3..0]
ena[3..0]
aclr[3..0]
signa
signb
dataa_0[35..18]
datab_0[35..18]
datab_0[17..0]
+
Output Register Bank
dataa_0[35..18]
Input Register Bank
datab_0[35..18]
Pipeline Register Bank
+
dataa_0[17..0]
72
result[ ]
+
dataa_0[17..0]
datab_0[17..0]
Half-DSP Block
February 2011
Altera Corporation
Stratix IV Device Handbook
Volume 1
4–20
Chapter 4: DSP Blocks in Stratix IV Devices
Stratix IV Operational Mode Descriptions
Double Multiplier
You can configure the Stratix IV DSP block to efficiently support a signed or unsigned
54 × 54-bit multiplier that is required to compute the mantissa portion of an IEEE
double-precision floating point multiplication. You can build a 54 × 54-bit multiplier
using basic 18 × 18 multipliers, shifters, and adders. In order to efficiently use the
Stratix IV DSP block’s built-in shifters and adders, a special double mode (partial
54 × 54 multiplier) is available that is a slight modification to the basic 36 × 36
multiplier mode, as shown in Figure 4–12 and Figure 4–13.
Figure 4–12. Double Mode Shown for a Half DSP Block
clock[3..0]
ena[3..0]
aclr[3..0]
signa
signb
dataa_0[35..18]
datab_0[35..18]
datab_0[17..0]
+
Output Register Bank
dataa_0[35..18]
Input Register Bank
datab_0[35..18]
Pipeline Register Bank
+
dataa_0[17..0]
72
result[ ]
+
dataa_0[17..0]
datab_0[17..0]
Half-DSP Block
Stratix IV Device Handbook
Volume 1
February 2011 Altera Corporation
Chapter 4: DSP Blocks in Stratix IV Devices
Stratix IV Operational Mode Descriptions
4–21
Figure 4–13. Unsigned 54 × 54 Multiplier for a Half-DSP Block
clock[3..0]
ena[3..0]
aclr[3..0]
"0"
"0"
dataa[53..36]
signa
signb
Two Multiplier
Adder Mode
+
36
datab[53..36]
dataa[35..18]
Double Mode
55
datab[35..18]
dataa[53..36]
datab[17..0]
dataa[35..18]
Final Adder (implemented with ALUT logic)
datab[53..36]
dataa[53..36]
Shifters and Adders
datab[53..36]
dataa[17..0]
108
result[ ]
36 x 36 Mode
datab[35..18]
dataa[35..18]
Shifters and Adders
datab[35..18]
dataa[17..0]
72
datab[17..0]
dataa[17..0]
datab[17..0]
Unsigned 54 X 54 Multiplier
February 2011
Altera Corporation
Stratix IV Device Handbook
Volume 1
4–22
Chapter 4: DSP Blocks in Stratix IV Devices
Stratix IV Operational Mode Descriptions
Two-Multiplier Adder Sum Mode
In a two-multiplier adder configuration, the DSP block can implement four 18-bit
two-multiplier adders (2 two-multiplier adders per half DSP block). You can
configure the adders to take the sum or difference of two multiplier outputs. You must
select summation or subtraction at compile time. The two-multiplier adder function
is useful for applications such as FFTs, complex FIR, and IIR filters. Figure 4–14 on
page 4–23 shows the DSP block configured in two-multiplier adder mode.
Loopback mode is the other sub-feature of the two-multiplier adder mode.
Figure 4–15 on page 4–24 shows the DSP block configured in the loopback mode. This
mode takes the 36-bit summation result of the two multipliers and feeds back the
most significant 18-bits to the input. The lower 18-bits are discarded. You have the
option to disable or zero-out the loopback data by using the dynamic zero_loopback
signal. A logic 1 value on the zero_loopback signal selects the zeroed data or
disables the looped back data, while a logic 0 selects the looped back data.
1
You must select the option to use loopback mode or the general two-multiplier adder
mode at compile time.
For two-multiplier adder mode, if all the inputs are full 18-bit and unsigned, the result
requires 37 bits. As the output data width in two-multiplier adder mode is limited to
36 bits, this 37-bit output requirement is not allowed. Any other combination that
does not violate the 36-bit maximum result is permitted; for example, two 16 × 16
signed two-multiplier adders is valid.
Two-multiplier adder mode supports the rounding and saturation logic unit. You can
use the pipeline registers and output registers within the DSP block to pipeline the
multiplier-adder result, increasing the performance of the DSP block.
Stratix IV Device Handbook
Volume 1
February 2011 Altera Corporation
Chapter 4: DSP Blocks in Stratix IV Devices
Stratix IV Operational Mode Descriptions
4–23
Figure 4–14. Two-Multiplier Adder Mode Shown for a Half DSP Block
signa
clock[3..0]
ena[3..0]
aclr[3..0]
signb
output_round
output_saturate
overflow (1)
Output Register Bank
Round/Saturate
dataa_1[17..0]
+
Pipeline Register Bank
datab_0[17..0]
Input Register Bank
dataa_0[17..0]
result[ ]
datab_1[17..0]
Half-DSP Block
Note to Figure 4–14:
(1) Block output for accumulator overflow and saturate overflow.
February 2011
Altera Corporation
Stratix IV Device Handbook
Volume 1
4–24
Chapter 4: DSP Blocks in Stratix IV Devices
Stratix IV Operational Mode Descriptions
Figure 4–15. Loopback Mode for a Half DSP Block
clock[3..0]
ena[3..0]
aclr[3..0]
signa
signb
output_round
output_saturate
zero_loopback
overflow (1)
Output Register Bank
dataa_1[17..0]
+
Round/Saturate
datab_0[17..0]
Pipeline Register Bank
loopback
Input Register Bank
dataa_0[17..0]
result[ ]
datab_1[17..0]
Half-DSP Block
Note to Figure 4–15:
(1) Block output for accumulator overflow and saturate overflow.
18 x 18 Complex Multiply
You can configure the DSP block to implement complex multipliers using
two-multiplier adder mode. A single half DSP block can implement one 18-bit
complex multiplier.
Equation 4–4 shows a complex multiplication.
Equation 4–4. Complex Multiplication Equation
(a + jb) × (c + jd) = ((a × c) – (b × d)) + j((a × d) + (b × c))
Stratix IV Device Handbook
Volume 1
February 2011 Altera Corporation
Chapter 4: DSP Blocks in Stratix IV Devices
Stratix IV Operational Mode Descriptions
4–25
To implement this complex multiplication within the DSP block, the real part
((a × c) – (b × d)) is implemented using two multipliers feeding one subtractor block
while the imaginary part ((a × d) + (b × c)) is implemented using another two
multipliers feeding an adder block. Figure 4–16 shows an 18-bit complex
multiplication. This mode automatically assumes all inputs are using signed
numbers.
Figure 4–16. Complex Multiplier Using Two-Multiplier Adder Mode
clock[3..0]
ena[3..0]
aclr[3..0]
signa
signb
A
C
36
AxC BxD
Real Part
Output Register Bank
Pipeline Register Bank
D
Input Register Bank
B
36
AxD
BxC
Imaginary Part
Half-DSP Block
February 2011
Altera Corporation
Stratix IV Device Handbook
Volume 1
4–26
Chapter 4: DSP Blocks in Stratix IV Devices
Stratix IV Operational Mode Descriptions
Four-Multiplier Adder
In the four-multiplier adder configuration shown in Figure 4–17, the DSP block can
implement two four-multiplier adders (one four-multiplier adder per half DSP block).
These modes are useful for implementing one-dimensional and two-dimensional
filtering applications. The four-multiplier adder is performed in two addition stages.
The outputs of two of the four multipliers are initially summed in the two first-stage
adder blocks. The results of these two adder blocks are then summed in the
second-stage adder block to produce the final four-multiplier adder result, as shown
by Equation 4–2 on page 4–5 and Equation 4–3 on page 4–5.
Figure 4–17. Four-Multiplier Adder Mode Shown for a Half DSP Block
signa
signb
clock[3..0]
ena[3..0]
aclr[3..0]
output_round
output_saturate
overflow (1)
dataa_0[ ]
datab_0[ ]
+
Output Register Bank
+
Round/Saturate
dataa_2[ ]
Input Register Bank
datab_1[ ]
Pipeline Register Bank
dataa_1[ ]
result[ ]
datab_2[ ]
+
dataa_3[ ]
datab_3[ ]
Half-DSP Block
Note to Figure 4–17:
(1) Block output for accumulator overflow and saturate overflow.
Stratix IV Device Handbook
Volume 1
February 2011 Altera Corporation
Chapter 4: DSP Blocks in Stratix IV Devices
Stratix IV Operational Mode Descriptions
4–27
Four-multiplier adder mode supports the rounding and saturation logic unit. You can
use the pipeline registers and output registers within the DSP block to pipeline the
multiplier-adder result, increasing the performance of the DSP block.
High-Precision Multiplier Adder Mode
In a high-precision multiplier adder configuration, shown in Figure 4–18 on
page 4–28, the DSP block can implement 2 two-multiplier adders, with multiplier
precision of 18 x 36 (one two-multiplier adder per half DSP block). This mode is useful
in filtering or FFT applications where a data path greater than 18 bits is required, yet
18 bits is sufficient for the coefficient precision. This can occur where the data has a
high dynamic range. If the coefficients are fixed, as in FFT and most filter applications,
the precision of 18 bits provide a dynamic range over 100 dB, if the largest coefficient
is normalized to the maximum 18-bit representation.
In these situations, the data path can be up to 36 bits, allowing sufficient capacity for
bit growth or gain changes in the signal source without loss of precision. This mode is
also extremely useful in single precision block floating point applications.
The high-precision multiplier adder is performed in two stages. The 18 × 36 multiply
is divided into two 18 × 18 multipliers. The multiplier with the LSB of the data source
is performed unsigned, while the multiplier with the MSB of the data source can be
signed or unsigned. The latter multiplier has its result left shifted by 18 bits prior to
the first adder stage, creating an effective 18 x 36 multiplier. The results of these two
adder blocks are then summed in the second stage adder block to produce the final
result:
Z[54..0] = P0[53..0] + P1[53..0]
where:
P0 = A[17..0] × B[35..0]
P1 = C[17..0] × D[35..0]
February 2011
Altera Corporation
Stratix IV Device Handbook
Volume 1
4–28
Chapter 4: DSP Blocks in Stratix IV Devices
Stratix IV Operational Mode Descriptions
Figure 4–18. High-Precision Multiplier Adder Configuration
signa
signb
clock[3..0]
ena[3..0]
aclr[3..0]
overflow (1)
dataA[0:17]
dataB[0:17]
Pipeline Register Bank
dataC[0:17]
Input Register Bank
dataB[18:35]
P0
<<18
+
Output Register Bank
+
dataA[0:17]
result[ ]
dataD[0:17]
+
P1
dataC[0:17]
<<18
dataD[18:35]
Half-DSP Block
Note to Figure 4–18:
(1) Block output for accumulator overflow and saturate overflow.
Stratix IV Device Handbook
Volume 1
February 2011 Altera Corporation
Chapter 4: DSP Blocks in Stratix IV Devices
Stratix IV Operational Mode Descriptions
4–29
Multiply Accumulate Mode
In multiply accumulate mode, the second-stage adder is configured as a 44-bit
accumulator or subtractor. The output of the DSP block is looped back to the
second-stage adder and added or subtracted with the two outputs of the first-stage
adder block according to Equation 4–3 on page 4–5. Figure 4–19 shows the DSP block
configured to operate in multiply accumulate mode.
Figure 4–19. Multiply Accumulate Mode Shown for a Half DSP Block
clock[3..0]
ena[3..0]
aclr[3..0]
signa
signb
output_round
output_saturate
chainout_sat_overflow (1)
accum_sload
dataa_0[ ]
datab_0[ ]
+
Output Register Bank
Round/Saturate
+
Second Register Bank
dataa_2[ ]
Input Register Bank
datab_1[ ]
Pipeline Register Bank
dataa_1[ ]
44
result[ ]
datab_2[ ]
+
dataa_3[ ]
datab_3[ ]
Half-DSP Block
Note to Figure 4–19:
(1) Block output for saturation overflow of chainout.
A single DSP block can implement up to two independent 44-bit accumulators.
February 2011
Altera Corporation
Stratix IV Device Handbook
Volume 1
4–30
Chapter 4: DSP Blocks in Stratix IV Devices
Stratix IV Operational Mode Descriptions
Use the dynamic accum_sload control signal to clear the accumulation. A logic 1
value on the accum_sload signal synchronously loads the accumulator with the
multiplier result only, while a logic 0 enables accumulation by adding or subtracting
the output of the DSP block (accumulator feedback) to the output of the multiplier
and first-stage adder.
1
You must configure the control signal for the accumulator and subtractor if static at
compile time.
This mode supports the rounding and saturation logic unit because it is configured as
an 18-bit multiplier accumulator. You can use the pipeline registers and output
registers within the DSP block to increase the performance of the DSP block.
Shift Modes
Stratix IV devices support the following shift modes for 32-bit input only:
1
■
Arithmetic shift left, ASL[N]
■
Arithmetic shift right, ASR[32-N]
■
Logical shift left, LSL[N]
■
Logical shift right, LSR[32-N]
■
32-bit rotator or barrel shifter, ROT[N]
You can switch between these modes using the dynamic rotate and shift control
signals.
You can use shift mode in a Stratix IV device by using a soft embedded processor such
as Nios® II to perform the dynamic shift and rotate operation. Figure 4–20 on
page 4–31 shows the shift mode configuration.
Shift mode makes use of the available multipliers to logically or arithmetically shift
left, right, or rotate the desired 32-bit data. You can configure the DSP block similar to
the independent 36-bit multiplier mode to perform shift mode operations.
Arithmetic shift right requires a signed input vector. During an arithmetic shift right,
the sign is extended to fill the MSB of the 32-bit vector. The logical shift right uses an
unsigned input vector. During a logical shift right, zeros are padded in the MSBs,
shifting the 32-bit vector to the right. The barrel shifter uses unsigned input vector
and implements a rotation function on a 32-bit word length.
Two control signals, rotate and shift_right, together with the signa and signb
signals, determine the shifting operation. Table 4–5 on page 4–31 lists examples of
shift operations.
Stratix IV Device Handbook
Volume 1
February 2011 Altera Corporation
Chapter 4: DSP Blocks in Stratix IV Devices
Stratix IV Operational Mode Descriptions
4–31
Figure 4–20. Shift Operation Mode Shown for a Half DSP Block
signa
signb
rotate
shift_right
clock[3..0]
ena[3..0]
aclr[3..0]
dataa_0[35..18]
datab_0[35..18]
+
dataa_0[35..18]
+
Shift/Rotate
datab_0[35..18]
Output Register Bank
Input Register Bank
Pipeline Register Bank
dataa_0[17..0]
32
result[ ]
datab_0[17..0]
+
dataa_0[17..0]
datab_0[17..0]
Half-DSP Block
Table 4–5. Examples of Shift Operations
Example
Signa
Signb
Shift
Rotate
A-input
B-input
Result
Logical Shift Left
LSL[N]
Unsigned
Unsigned
0
0
0xAABBCCDD
0x0000100
0xBBCCDD00
Logical Shift Right
LSR[32-N]
Unsigned
Unsigned
1
0
0xAABBCCDD
0x0000100
0x000000AA
Arithmetic Shift Left
ASL[N]
Signed
Unsigned
0
0
0xAABBCCDD
0x0000100
0xBBCCDD00
Arithmetic Shift Right
ASR[32-N]
Signed
Unsigned
1
0
0xAABBCCDD
0x0000100
0xFFFFFFAA
Unsigned
Unsigned
0
1
0xAABBCCDD
0x0000100
0xBBCCDDAA
Rotation ROT[N]
February 2011
Altera Corporation
Stratix IV Device Handbook
Volume 1
4–32
Chapter 4: DSP Blocks in Stratix IV Devices
Stratix IV Operational Mode Descriptions
Rounding and Saturation Mode
Rounding and saturation functions are often required in DSP arithmetic. Use
rounding to limit bit growth and its side effects; use saturation to reduce overflow and
underflow side effects.
Two rounding modes are supported in Stratix IV devices:
1
■
Round-to-nearest-integer mode
■
Round-to-nearest-even mode
You must select one of these two options at compile time.
Round-to-nearest-integer provides the biased rounding support and is the simplest
form of rounding commonly used in DSP arithmetic. The round-to-nearest-even
method provides unbiased rounding support and is used where DC offsets are a
concern. Table 4–6 lists how round-to-nearest-even works.
Table 4–6. Example of Round-To-Nearest-Even Mode
6- to 4-bits
Rounding
Odd/Even
(Integer)
Fractional
Add to Integer
Result
010111
x
> 0.5 (11)
1
0110
001101
x
< 0.5 (01)
0
0011
001010
Even (0010)
= 0.5 (10)
0
0010
001110
Odd (0011)
= 0.5 (10)
1
0100
110111
x
> 0.5 (11)
1
1110
101101
x
< 0.5 (01)
0
1011
110110
Odd (1101)
= 0.5 (10)
1
1110
110010
Even (1100)
= 0.5 (10)
0
1100
Table 4–7 lists examples of the difference between the two modes. In this example, a
6-bit input is rounded to 4 bits. The main difference between the two rounding
options is when the residue bits are exactly halfway between its nearest two integers
and the LSB is zero (even).
Table 4–7. Comparison of Round-to-Nearest-Integer and Round-to-Nearest-Even
Stratix IV Device Handbook
Volume 1
Round-To-Nearest-Integer
Round-To-Nearest-Even
010111 ➱ 0110
010111 ➱ 0110
001101 ➱ 0011
001101 ➱ 0011
001010 ➱ 0011
001010 ➱ 0010
001110 ➱ 0100
001110 ➱ 0100
110111 ➱ 1110
110111 ➱ 1110
101101 ➱ 1011
101101 ➱ 1011
110110 ➱ 1110
110110 ➱ 1110
110010 ➱ 1101
110010 ➱ 1100
February 2011 Altera Corporation
Chapter 4: DSP Blocks in Stratix IV Devices
Stratix IV Operational Mode Descriptions
4–33
Two saturation modes are supported in Stratix IV:
1
■
Asymmetric saturation mode
■
Symmetric saturation mode
You must select one of the two options at compile time.
In 2’s-complement format, the maximum negative number that can be represented is
–2(n–1), while the maximum positive number is 2(n–1) – 1. Symmetrical saturation limits
the maximum negative number to –2(n–1) + 1. For example, for 32 bits:
■
Asymmetric 32-bit saturation: Max = 0x7FFFFFFF, Min = 0x80000000
■
Symmetric 32-bit saturation: Max = 0x7FFFFFFF, Min = 0x80000001
Table 4–8 lists how saturation works. In this example, a 44-bit input is saturated to
36-bits.
Table 4–8. Examples of Saturation
44- to 36-Bits Saturation
Symmetric SAT Result
Asymmetric SAT Result
5926AC01342h
7FFFFFFFFh
7FFFFFFFFh
ADA38D2210h
800000001h
800000000h
Stratix IV devices have up to 16 configurable bit positions out of the 44-bit bus
([43:0]) for the rounding and saturate logic unit, providing higher flexibility. These
16-bit positions are located at bits [21:6] for rounding and [43:28] for saturation, as
shown in Figure 4–21.
1
You must select the 16 configurable bit positions at compile time.
Figure 4–21. Rounding and Saturation Locations
16 User defined SAT Positions (bit 43-28)
43
42
29
28
1
0
16 User defined RND Positions (bit 21-6)
43
42
1
February 2011
21
20
7
6
0
For symmetric saturation, the RND bit position is also used to determine where the
LSP for the saturated data is located.
Altera Corporation
Stratix IV Device Handbook
Volume 1
4–34
Chapter 4: DSP Blocks in Stratix IV Devices
Stratix IV Operational Mode Descriptions
Use the rounding and saturation function just described in regular supported
multiplication operations, as specified in Table 4–2 on page 4–8. However, for
accumulation-type operations, use the following convention:
The functionality of the round logic unit is in the format of:
Result = RND[S(A × B)], when used for an accumulation type of operation.
Likewise, the functionality of the saturation logic unit is in the format of:
Result = SAT[S(A × B)], when used for an accumulation type of operation.
If you use both the rounding and saturation logic units for an accumulation type of
operation, the format is:
Result = SAT[RND[S(A × B)]]
DSP Block Control Signals
The Stratix IV DSP block is configured using a set of static and dynamic signals. You
can configure the DSP block dynamic signals. You can set the signals to toggle or not
toggle at run time. Table 4–9 lists the dynamic signals for the DSP block.
Table 4–9. DSP Block Dynamic Signals (Part 1 of 2)
Signal Name
Function
Count
■
signa
Signed/unsigned control for all multipliers and adders.
■
signb
■
signa for “multiplicand” input bus to dataa[17:0] to each
multiplier
■
signb for “multiplier” input bus datab[17:0] to each multiplier
■
signa = 1, signb = 1 for signed-signed multiplication
■
signa = 1, signb = 0 for signed-unsigned multiplication
■
signa = 0, signb = 1 for unsigned-signed multiplication
■
signa = 0, signb = 0 for unsigned-unsigned multiplication
2
Round control for the first stage round and saturation block.
output_round
■
output_round = 1 for rounding on multiply output
■
output_round = 0 for normal multiply output
1
Round control for the second stage round and saturation block.
chainout_round
output_saturate
chainout_saturate
Stratix IV Device Handbook
Volume 1
■
chainout_round = 1 for rounding multiply output
■
chainout_round = 0 for normal multiply output
1
Saturation control for the first stage round and saturation block for
Q-format multiply. If you enable both rounding and saturation,
saturation is done on the rounded result.
■
output_saturate = 1 for saturation support
■
output_saturate = 0 for no saturation support
Saturation control for the second stage round and saturation block for
Q-format multiply. If you enable both rounding and saturation,
saturation is done on the rounded result.
■
chainout_saturate = 1 for saturation support
■
chainout_saturate = 0 for no saturation support
1
1
February 2011 Altera Corporation
Chapter 4: DSP Blocks in Stratix IV Devices
Software Support
4–35
Table 4–9. DSP Block Dynamic Signals (Part 2 of 2)
Signal Name
Function
Count
Dynamically specifies whether the accumulator value is zero.
accum_sload
■
accum_sload = 0, accumulation input is from the output registers
■
accum_sload = 1, accumulation input is set to zero
1
zero_chainout
Dynamically specifies whether the chainout value is zero.
1
zero_loopback
Dynamically specifies whether the loopback value is zero.
1
rotate
rotate = 1, the rotation feature is enabled
1
shift_right
shift_right = 1, the shift right feature is enabled
1
Total Signals per Half Block
11
clock0
clock1
clock2
DSP-block-wide clock signals.
4
Input and Pipeline Register enable signals.
4
DSP block-wide asynchronous clear signals (active low).
4
clock3
ena0
ena1
ena2
ena3
aclr0
aclr1
aclr2
aclr3
Total Count per Full Block
34
Software Support
Altera provides two distinct methods for implementing various modes of the DSP
block in a design—instantiation and inference. Both methods use the following
Quartus II megafunctions:
■
lpm_mult
■
altmult_add
■
altmult_accum
■
altfp_mult
To use the DSP block, instantiate the megafunctions in the Quartus II software.
Alternatively, with inference, create an HDL design and synthesize it using a
third-party synthesis tool (such as LeonardoSpectrum, Synplify, or Quartus II
Native Synthesis) that infers the appropriate megafunction by recognizing
multipliers, multiplier adders, multiplier accumulators, and shift functions. Using
either method, the Quartus II software maps the functionality to the DSP blocks
during compilation.
f For instructions about using these megafunctions and the MegaWizard Plug-In
Manager, refer to Quartus II software Help.
February 2011
Altera Corporation
Stratix IV Device Handbook
Volume 1
4–36
Chapter 4: DSP Blocks in Stratix IV Devices
Software Support
f For more information, refer to the “Synthesis” section in volume 1 of the Quartus II
Handbook.
Document Revision History
Table 4–10 lists the revision history for this chapter.
Table 4–10. Document Revision History
Date
Version
February 2011
November 2009
3.1
3.0
June 2009
2.3
April 2009
2.2
March 2009
2.1
November 2008
May 2008
Stratix IV Device Handbook
Volume 1
2.0
1.0
Changes
■
Applied new template.
■
Minor text edits.
■
Updated Table 4–1.
■
Updated “Stratix IV Simplified DSP Operation” section.
■
Updated graphics.
■
Minor text edits.
■
Added an introductory paragraph to increase search ability.
■
Removed the Conclusion section.
■
Updated Table 4–1.
■
Updated Table 4–1.
■
Removed “Referenced Documents” section.
■
Updated Table 4–2.
■
Updated Figure 4–16.
■
Updated Figure 4–18.
Initial release.
February 2011 Altera Corporation
5. Clock Networks and PLLs in Stratix IV
Devices
September 2012
SIV51005-3.4
SIV51005-3.4
This chapter describes the hierarchical clock networks and phase-locked loops (PLLs)
which have advanced features in Stratix® IV devices. It includes details about the
ability to reconfigure the PLL counter clock frequency and phase shift in real time,
allowing you to sweep PLL output frequencies and dynamically adjust the output
clock phase shift.
The Quartus® II software enables the PLLs and their features without external
devices. The following sections describe the Stratix IV clock networks and PLLs in
detail:
■
“Clock Networks in Stratix IV Devices” on page 5–1
■
“PLLs in Stratix IV Devices” on page 5–19
Clock Networks in Stratix IV Devices
The global clock networks (GCLKs), regional clock networks (RCLKs), and periphery
clock networks (PCLKs) available in Stratix IV devices are organized into hierarchical
clock structures that provide up to 236 unique clock domains (16 GCLKs + 88 RCLKs
+ 132 PCLKs) within the Stratix IV device and allow up to 71 unique GCLK, RCLK,
and PCLK clock sources (16 GCLKs + 22 RCLKs + 33 PCLKs) per device quadrant.
Table 5–1 lists the clock resources available in Stratix IV devices.
Table 5–1. Clock Resources in Stratix IV Devices (Part 1 of 2)
Clock Resource
Number of Resources Available
Source of Clock Resource
Clock input pins
32 Single-ended
(16 Differential)
CLK[0..15]p and CLK[0..15]n pins
GCLK networks
16
CLK[0..15]p and CLK[0..15]n pins, PLL clock outputs, and
logic array
RCLK networks
64/88
PCLK networks
GCLKs/RCLKs per
quadrant
(1)
56/88/112/132 (33 per device
quadrant) (2)
32/38
(3)
CLK[0..15]p and CLK[0..15]n pins, PLL clock outputs, and
logic array
DPA clock outputs, PLD-transceiver interface clocks, horizontal
I/O pins, and logic array
16 GCLKs + 16 RCLKs
16 GCLKs + 22 RCLKs
© 2012 Altera Corporation. All rights reserved. ALTERA, ARRIA, CYCLONE, HARDCOPY, MAX, MEGACORE, NIOS, QUARTUS and STRATIX words and logos
are trademarks of Altera Corporation and registered in the U.S. Patent and Trademark Office and in other countries. All other words and logos identified as
trademarks or service marks are the property of their respective holders as described at www.altera.com/common/legal.html. Altera warrants performance of its
semiconductor products to current specifications in accordance with Altera's standard warranty, but reserves the right to make changes to any products and
services at any time without notice. Altera assumes no responsibility or liability arising out of the application or use of any information, product, or service
described herein except as expressly agreed to in writing by Altera. Altera customers are advised to obtain the latest version of device specifications before relying
on any published information and before placing orders for products or services.
ISO
9001:2008
Registered
Stratix IV Device Handbook
Volume 1
September 2012
Feedback Subscribe
5–2
Chapter 5: Clock Networks and PLLs in Stratix IV Devices
Clock Networks in Stratix IV Devices
Table 5–1. Clock Resources in Stratix IV Devices (Part 2 of 2)
Clock Resource
GCLKs/RCLKs per
device
Number of Resources Available
80/104
(4)
Source of Clock Resource
16 GCLKs + 64 RCLKs
16 GCLKs + 88 RCLKs
Notes to Table 5–1:
(1) There are 64 RCLKs in the EP4S40G2, EP4S100G2, EP4SE230, EP4SGX70, EP4SGX110, EP4SGX180, and EP4SGX230 devices. There are 88
RCLKs in the EP4S40G5, EP4S100G3, EP4S100G4, EP4S100G5, EP4SE360, EP4SE530, EP4SE820, EP4SGX290, EP4SGX360, and
EP4SGX530 devices.
(2) There are 56 PCLKs in the EP4SGX70, and EP4SGX110 devices. There are 88 PCLKs in the EP4S40G2, EP4S100G2, EP4SE230, EP4SE360,
EP4SGX180, EP4SGX230, EP4SGX290, and EP4SGX360 devices. There are 112 PCLKs in the EP4S40G5, EP4S100G3, EP4S100G4,
EP4S100G5, EP4SE530 and EP4SGX530 devices. There are 132 PCLKs in the EP4SE820 device.
(3) There are 32 GCLKs/RCLKs per quadrant in the EP4S40G2, EP4S100G2, EP4SE230, EP4SGX70, EP4SGX110, EP4SGX180, and EP4SGX230
devices. There are 38 GCLKs/RCLKs per quadrant in the EP4S40G5, EP4S100G3, EP4S100G4, EP4S100G5, EP4SE360, EP4SE530, EP4SE820,
EP4SGX290, EP4SGX360, and EP4SGX530 devices.
(4) There are 80 GCLKs/RCLKs per entire device in the EP4S40G2, EP4S100G2, EP4SE230, EP4SGX70, EP4SGX110, EP4SGX180, and EP4SGX230
devices. There are 104 GCLKs/RCLKS per entire device in the EP4S40G5, EP4S100G3, EP4S100G4, EP4S100G5, EP4SE360, EP4SE530,
EP4SE820, EP4SGX290, EP4SGX360, and EP4SGX530 devices.
Stratix IV devices have up to 32 dedicated single-ended clock pins or 16 dedicated
differential clock pins (CLK[0..15]p and CLK[0..15]n) that can drive either the GCLK
or RCLK networks. These clock pins are arranged on the four sides of the Stratix IV
device, as shown in Figure 5–1 through Figure 5–4 on page 5–5.
f For more information about how to connect the clock input pins, refer to the
Stratix IV GX and Stratix IV E Device Family Pin Connection Guidelines.
Stratix IV Device Handbook
Volume 1
September 2012 Altera Corporation
Chapter 5: Clock Networks and PLLs in Stratix IV Devices
Clock Networks in Stratix IV Devices
5–3
Global Clock Networks
Stratix IV devices provide up to 16 GCLKs that can drive throughout the device,
serving as low-skew clock sources for functional blocks such as adaptive logic
modules (ALMs), digital signal processing (DSP) blocks, TriMatrix memory blocks,
and PLLs. Stratix IV device I/O elements (IOEs) and internal logic can also drive
GCLKs to create internally generated global clocks and other high fan-out control
signals; for example, synchronous or asynchronous clears and clock enables.
Figure 5–1 shows the CLK pins and PLLs that can drive the GCLK networks in
Stratix IV devices.
Figure 5–1. GCLK Networks
CLK[12..15]
T1 T2
L1
R1
GCLK[12..15]
CLK[0..3]
L2 GCLK[0..3]
L3
GCLK[8..11] R2
CLK[8..11]
R3
GCLK[4..7]
L4
R4
B1 B2
CLK[4..7]
September 2012
Altera Corporation
Stratix IV Device Handbook
Volume 1
5–4
Chapter 5: Clock Networks and PLLs in Stratix IV Devices
Clock Networks in Stratix IV Devices
Regional Clock Networks
RCLK networks only pertain to the quadrant they drive into. RCLK networks provide
the lowest clock delay and skew for logic contained within a single device quadrant.
The Stratix IV device IOEs and internal logic within a given quadrant can also drive
RCLKs to create internally generated regional clocks and other high fan-out control
signals; for example, synchronous or asynchronous clears and clock enables.
Figure 5–2 through Figure 5–4 on page 5–5 show the CLK pins and PLLs that can
drive the RCLK networks in Stratix IV devices.
Figure 5–2. RCLK Networks (EP4SE230, EP4SGX70, and EP4SGX110 Devices)
(1)
CLK[12..15]
T1
RCLK[54..63] RCLK[44..53]
RCLK[38..43]
RCLK[0..5]
CLK[0..3] L2
Q1
Q2
Q4
Q3
RCLK[6..11]
R2 CLK[8..11]
RCLK[32..37]
RCLK[12..21] RCLK[22..31]
B1
CLK[4..7]
Note to Figure 5–2:
(1) A maximum of four signals from the core can drive into each group of RCLKs. For example, only four core signals can drive into RCLK[0..5] and
another four core signals can drive into RCLK[54..63] at any one time.
Stratix IV Device Handbook
Volume 1
September 2012 Altera Corporation
Chapter 5: Clock Networks and PLLs in Stratix IV Devices
Clock Networks in Stratix IV Devices
5–5
Figure 5–3. RCLK Networks (EP4S40G2, EP4S100G2, EP4SGX180, and EP4SGX230 Devices)
(1)
CLK[12..15]
T1 T2
RCLK[54..63] RCLK[44..53]
RCLK[0..5]
RCLK[38..43]
Q1 Q2
L2
CLK[0..3]
L3
R2
CLK[8..11]
R3
Q4 Q3
RCLK[6..11]
RCLK[32..37]
RCLK[12..21] RCLK[22..31]
B1 B2
CLK[4..7]
Note to Figure 5–3:
(1) A maximum of four signals from the core can drive into each group of RCLKs. For example, only four core signals can drive into RCLK[0..5] and
another four core signals can drive into RCLK[54..63] at any one time.
Figure 5–4. RCLK Networks (EP4S40G5, EP4S100G3, EP4S100G4, EP4S100G5, EP4SE360, EP4SE530, EP4SE820,
EP4SGX290, EP4SGX360, and EP4SGX530 Devices) (1), (2), (3)
CLK[12..15]
T1 T2
L1
R1
RCLK[82..87] RCLK[54..63] RCLK[44..53] RCLK[76..81]
RCLK[0..5]
CLK[0..3] L2
L3
RCLK[38..43]
Q1
Q2
Q4
Q3
RCLK[6..11]
R2 CLK[8..11]
R3
RCLK[32..37]
RCLK[64..69] RCLK[12..21] RCLK[22..31] RCLK[70..75]
L4
R4
B1 B2
CLK[4..7]
Notes to Figure 5–4:
(1) The corner RCLK[64..87] can only be fed by their respective corner PLL outputs. For more information about connectivity, refer to Table 5–6 on
page 5–13.
(2) The EP4S40G5 and EP4SE360 devices have up to eight PLLs. For more information about PLL availability, refer to Table 5–7 on page 5–19.
(3) A maximum of four signals from the core can drive into each group of RCLKs. For example, only four core signals can drive into RCLK[0..5] and
another four core signals can drive into RCLK[54..63] at any one time.
September 2012
Altera Corporation
Stratix IV Device Handbook
Volume 1
5–6
Chapter 5: Clock Networks and PLLs in Stratix IV Devices
Clock Networks in Stratix IV Devices
Periphery Clock Networks
PCLK networks shown in Figure 5–5 through Figure 5–8 on page 5–8 are collections of
individual clock networks driven from the periphery of the Stratix IV device. Clock
outputs from the dynamic phase aligner (DPA) block, programmable logic device
(PLD)-transceiver interface clocks, I/O pins, and internal logic can drive the PCLK
networks.
PCLKs have higher skew when compared with GCLK and RCLK networks. You can
use PCLKs for general purpose routing to drive signals into and out of the Stratix IV
device.
Figure 5–5. PCLK Networks (EP4SGX70 and EP4SGX110 Devices)
CLK[12..15]
T1
PCLK[42..56]
PCLK[0..13]
CLK[0..3] L2
Q1
Q2
Q4
Q3
PCLK[14..27]
R2 CLK[8..11]
PCLK[28..41]
B1
CLK[4..7]
Stratix IV Device Handbook
Volume 1
September 2012 Altera Corporation
Chapter 5: Clock Networks and PLLs in Stratix IV Devices
Clock Networks in Stratix IV Devices
5–7
Figure 5–6. PCLK Networks (EP4S40G2, EP4S100G2, EP4SE230, EP4SE360, EP4SGX180, EP4SGX230, EP4SGX290, and
EP4SGX360 Devices) (1)
CLK[12..15]
T1 T2
CLK[0..3]
PCLK[0..10]
PCLK[77..87]
PCLK[11..21]
PCLK[66..76]
L2
Q1 Q2
R2
L3
Q4 Q3
R3
PCLK[22..32]
PCLK[55..65]
PCLK[33..43]
PCLK[44..54]
CLK[8..11]
B1 B2
CLK[4..7]
Note to Figure 5–6:
(1) The EP4SE230 device has four PLLs. The EP4SGX290 and EP4SGX360 devices have up to 12 PLLs. For more information about PLL availability,
refer to Table 5–7 on page 5–19.
Figure 5–7. PCLK Networks (EP4S40G5, EP4S100G3, EP4S100G4, EP4S100G5, EP4SE530, and EP4SGX530 Devices) (1)
CLK[12..15]
T1 T2
L1
R1
PCLK[98..111]
PCLK[0..13]
PCLK[14..27]
CLK[0..3]
PCLK[84..97]
L2
Q1
Q2
R2
L3
Q4
Q3
R3
PCLK[28..41]
PCLK[70..83]
PCLK[42..55]
PCLK[56..69]
L4
CLK[8..11]
R4
B1 B2
CLK[4..7]
Note to Figure 5–7:
(1) The EP4S40G5 device has eight PLLs. For more information about PLL availability, refer to Table 5–7 on page 5–19.
September 2012
Altera Corporation
Stratix IV Device Handbook
Volume 1
5–8
Chapter 5: Clock Networks and PLLs in Stratix IV Devices
Clock Networks in Stratix IV Devices
Figure 5–8. PCLK Networks (EP4SE820 Device)
CLK[12..15]
T1 T2
L1
R1
PCLK[0..15]
PCLK[116..131]
PCLK[16..32]
CLK[0..3]
PCLK[99..115]
L2
Q1
Q2
R2
L3
Q4
Q3
R3
PCLK[33..49]
PCLK[82..98]
PCLK[50..65]
PCLK[66..81]
L4
CLK[8..11]
R4
B1 B2
CLK[4..7]
Stratix IV Device Handbook
Volume 1
September 2012 Altera Corporation
Chapter 5: Clock Networks and PLLs in Stratix IV Devices
Clock Networks in Stratix IV Devices
5–9
Clock Sources Per Quadrant
There are 26 section clock (SCLK) networks available in each spine clock that can
drive six row clocks in each logic array block (LAB) row, nine column I/O clocks, and
three core reference clocks. The SCLKs are the clock resources to the core functional
blocks, PLLs, and I/O interfaces of the device. Figure 5–9 shows that the SCLKs can
be driven by the GCLK, RCLK, PCLK, or the PLL feedback clock networks in each
spine clock.
1
A spine clock is another layer of routing below the GCLKs, RCLKs, and PCLKs before
each clock is connected to clock routing for each LAB row. The settings for spine
clocks are transparent to all users. The Quartus II software automatically routes the
spine clock based on the GCLK, RCLK, and PCLKs.
Figure 5–9. Hierarchical Clock Networks per Spine Clock (1)
9
GCLK
PLL feedback clock (4)
16
3
16 (2)
PCLK
Column I/O clock (5)
SCLK 26
3
Core reference clock (6)
22 (3)
RCLK
6
Row clock (7)
Notes to Figure 5–9:
(1) The GCLK, RCLK, PCLK, and PLL feedback clocks share the same routing to the SCLKs. The total number of clock
resources must not exceed the SCLK limits in each region to ensure successful design fitting in the Quartus II
software.
(2) There are up to 16 PCLKs that can drive the SCLKs in each spine clock in the largest device.
(3) There are up to 22 RCLKs that can drive the SCLKs in each spine clock in the largest device.
(4) The PLL feedback clock is the clock from the PLL that drives into the SCLKs.
(5) The column I/O clock is the clock that drives the column I/O core registers and I/O interfaces.
(6) The core reference clock is the clock that feeds into the PLL as the PLL reference clock.
(7) The row clock is the clock source to the LAB, memory blocks, and row I/O interfaces in the core row.
Clock Regions
Stratix IV devices provide up to 104 distinct clock domains (16 GCLKs + 88 RCLKs) in
the entire device. You can use these clock resources to form the following types of
clock regions:
■
Entire device
■
Regional
■
Dual-regional
To form the entire device clock region, a source (not necessarily a clock signal) drives a
GCLK network that can be routed through the entire device. This clock region has the
maximum delay when compared with other clock regions, but allows the signal to
reach every destination within the device. This is a good option for routing global
reset and clear signals or routing clocks throughout the device.
September 2012
Altera Corporation
Stratix IV Device Handbook
Volume 1
5–10
Chapter 5: Clock Networks and PLLs in Stratix IV Devices
Clock Networks in Stratix IV Devices
To form a RCLK region, a source drives a single quadrant of the device. This clock
region provides the lowest skew within a quadrant and is a good option if all the
destinations are within a single device quadrant.
To form a dual-regional clock region, a single source (a clock pin or PLL output)
generates a dual-regional clock by driving two RCLK networks (one from each
quadrant). This technique allows destinations across two device quadrants to use the
same low-skew clock. The routing of this signal on an entire side has approximately
the same delay as a RCLK region. Internal logic can also drive a dual-regional clock
network. Corner PLL outputs only span one quadrant, they cannot generate a
dual-regional clock network. Figure 5–10 shows the dual-regional clock region.
Figure 5–10. Stratix IV Dual-Regional Clock Region
Clock pins or PLL outputs
can drive half of the device to
create side-wide clocking
regions for improved
interface timing.
Clock Network Sources
In Stratix IV devices, clock input pins, PLL outputs, and internal logic can drive the
GCLK and RCLK networks. For connectivity between dedicated pins CLK[0..15] and
the GCLK and RCLK networks, refer to Table 5–2 and Table 5–3 on page 5–11.
Dedicated Clock Input Pins
Clock pins can be either differential clocks or single-ended clocks. Stratix IV devices
support 16 differential clock inputs or 32 single-ended clock inputs. You can also use
dedicated clock input pins CLK[15..0] for high fan-out control signals such as
asynchronous clears, presets, and clock enables for protocol signals such as TRDY and
IRDY for PCIe through the GCLK or RCLK networks.
LABs
You can drive each GCLK and RCLK network using LAB-routing to enable internal
logic to drive a high fan-out, low-skew signal.
1
Stratix IV Device Handbook
Volume 1
Stratix IV PLLs cannot be driven by internally generated GCLKs or RCLKs. The input
clock to the PLL has to come from dedicated clock input pins or pin/PLL-fed GCLKs
or RCLKs.
September 2012 Altera Corporation
Chapter 5: Clock Networks and PLLs in Stratix IV Devices
Clock Networks in Stratix IV Devices
5–11
PLL Clock Outputs
Stratix IV PLLs can drive both GCLK and RCLK networks, as described in Table 5–5
on page 5–13 and Table 5–6 on page 5–13.
Table 5–2 lists the connection between the dedicated clock input pins and GCLKs.
Table 5–2. Clock Input Pin Connectivity to the GCLK Networks
CLK (p/n Pins)
Clock Resources
0
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
GCLK0
Y
Y
Y
Y
—
—
—
—
—
—
—
—
—
—
—
—
GCLK1
Y
Y
Y
Y
—
—
—
—
—
—
—
—
—
—
—
—
GCLK2
Y
Y
Y
Y
—
—
—
—
—
—
—
—
—
—
—
—
GCLK3
Y
Y
Y
Y
—
—
—
—
—
—
—
—
—
—
—
—
GCLK4
—
—
—
—
Y
Y
Y
Y
—
—
—
—
—
—
—
—
GCLK5
—
—
—
—
Y
Y
Y
Y
—
—
—
—
—
—
—
—
GCLK6
—
—
—
—
Y
Y
Y
Y
—
—
—
—
—
—
—
—
GCLK7
—
—
—
—
Y
Y
Y
Y
—
—
—
—
—
—
—
—
GCLK8
—
—
—
—
—
—
—
—
Y
Y
Y
Y
—
—
—
—
GCLK9
—
—
—
—
—
—
—
—
Y
Y
Y
Y
—
—
—
—
GCLK10
—
—
—
—
—
—
—
—
Y
Y
Y
Y
—
—
—
—
GCLK11
—
—
—
—
—
—
—
—
Y
Y
Y
Y
—
—
—
—
GCLK12
—
—
—
—
—
—
—
—
—
—
—
—
Y
Y
Y
Y
GCLK13
—
—
—
—
—
—
—
—
—
—
—
—
Y
Y
Y
Y
GCLK14
—
—
—
—
—
—
—
—
—
—
—
—
Y
Y
Y
Y
GCLK15
—
—
—
—
—
—
—
—
—
—
—
—
Y
Y
Y
Y
Table 5–3 lists the connectivity between the dedicated clock input pins and RCLKs in
Stratix IV devices. A given clock input pin can drive two adjacent RCLK networks to
create a dual-regional clock network.
Table 5–3. Clock Input Pin Connectivity to the RCLK Networks (Part 1 of 2)
CLK (p/n Pins)
Clock Resource
0
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
RCLK [0, 4, 6, 10]
Y
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
RCLK [1, 5, 7, 11]
—
Y
—
—
—
—
—
—
—
—
—
—
—
—
—
—
RCLK [2, 8]
—
—
Y
—
—
—
—
—
—
—
—
—
—
—
—
—
RCLK [3, 9]
—
—
—
Y
—
—
—
—
—
—
—
—
—
—
—
—
RCLK [13, 17, 21, 23,
27, 31]
—
—
—
—
Y
—
—
—
—
—
—
—
—
—
—
—
RCLK [12, 16, 20, 22,
26, 30]
—
—
—
—
—
Y
—
—
—
—
—
—
—
—
—
—
RCLK [15, 19, 25, 29]
—
—
—
—
—
—
Y
—
—
—
—
—
—
—
—
—
RCLK [14, 18, 24, 28]
—
—
—
—
—
—
—
Y
—
—
—
—
—
—
—
—
RCLK [35, 41]
—
—
—
—
—
—
—
—
Y
—
—
—
—
—
—
—
September 2012
Altera Corporation
Stratix IV Device Handbook
Volume 1
5–12
Chapter 5: Clock Networks and PLLs in Stratix IV Devices
Clock Networks in Stratix IV Devices
Table 5–3. Clock Input Pin Connectivity to the RCLK Networks (Part 2 of 2)
CLK (p/n Pins)
Clock Resource
0
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
RCLK [34, 40]
—
—
—
—
—
—
—
—
—
Y
—
—
—
—
—
—
RCLK [33, 37, 39, 43]
—
—
—
—
—
—
—
—
—
—
Y
—
—
—
—
—
RCLK [32, 36, 38, 42]
—
—
—
—
—
—
—
—
—
—
—
Y
—
—
—
—
RCLK [47, 51, 57, 61]
—
—
—
—
—
—
—
—
—
—
—
—
Y
—
—
—
RCLK [46, 50, 56, 60]
—
—
—
—
—
—
—
—
—
—
—
—
—
Y
—
—
RCLK [45, 49, 53, 55,
59, 63]
—
—
—
—
—
—
—
—
—
—
—
—
—
—
Y
—
RCLK [44, 48, 52, 54,
58, 62]
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
Y
Clock Input Connections to the PLLs
Table 5–4 lists the dedicated clock input pin connectivity to Stratix IV PLLs.
Table 5–4. Device PLLs and PLL Clock Pin Drivers
Dedicated Clock
Input Pin
CLK (p/n Pins)
(1), (2)
PLL Number
L1 (3)
L2
L3
L4 (3)
B1
B2
R1 (3)
R2
R3
R4 (3)
T1
T2
CLK0
Y
Y
Y
Y
—
—
—
—
—
—
—
—
CLK1
Y
Y
Y
Y
—
—
—
—
—
—
—
—
CLK2
Y
Y
Y
Y
—
—
—
—
—
—
—
—
CLK3
Y
Y
Y
Y
—
—
—
—
—
—
—
—
CLK4
—
—
—
—
Y
Y
—
—
—
—
—
—
CLK5
—
—
—
—
Y
Y
—
—
—
—
—
—
CLK6
—
—
—
—
Y
Y
—
—
—
—
—
—
CLK7
—
—
—
—
Y
Y
—
—
—
—
—
—
CLK8
—
—
—
—
—
—
Y
Y
Y
Y
—
—
CLK9
—
—
—
—
—
—
Y
Y
Y
Y
—
—
CLK10
—
—
—
—
—
—
Y
Y
Y
Y
—
—
CLK11
—
—
—
—
—
—
Y
Y
Y
Y
—
—
CLK12
—
—
—
—
—
—
—
—
—
—
Y
Y
CLK13
—
—
—
—
—
—
—
—
—
—
Y
Y
CLK14
—
—
—
—
—
—
—
—
—
—
Y
Y
CLK15
—
—
—
—
—
—
—
—
—
—
Y
Y
Notes to Table 5–4:
(1) For single-ended clock inputs, only the CLK<#>p pin has a dedicated connection to the PLL. If you use the CLK<#>n pin, a global clock is used.
(2) For the availability of the clock input pins in each device density, refer to the “Stratix IV Device Pin-Out Files” section of the Pin-Out Files for
Altera Devices site.
(3) These are non-compensated clock input paths. For the compensated input for these PLLs, use the corresponding PLL_[L, R][1,4]_CLK input
pin.
Stratix IV Device Handbook
Volume 1
September 2012 Altera Corporation
Chapter 5: Clock Networks and PLLs in Stratix IV Devices
Clock Networks in Stratix IV Devices
1
5–13
Dedicated clock pins can drive PLLs over dedicated routing; they do not require the
global or regional network. Compensated inputs, which are a subset of dedicated
clock pins, drive PLLs that can only compensate the input delay when a dedicated
clock pin is in the same I/O bank as the PLL used.
Clock Output Connections
PLLs in Stratix IV devices can drive up to 20 RCLK networks and four GCLK
networks. For Stratix IV PLL connectivity to GCLK networks, refer to Table 5–5. The
Quartus II software automatically assigns PLL clock outputs to RCLK and GCLK
networks.
Table 5–5 lists how the PLL clock outputs connect to the GCLK networks.
Table 5–5. Stratix IV PLL Connectivity to the GCLK Networks
(1)
PLL Number
Clock Network
L1
L2
L3
L4
B1
B2
R1
R2
R3
R4
T1
T2
GCLK0
Y
Y
Y
Y
—
—
—
—
—
—
—
—
GCLK1
Y
Y
Y
Y
—
—
—
—
—
—
—
—
GCLK2
Y
Y
Y
Y
—
—
—
—
—
—
—
—
GCLK3
Y
Y
Y
Y
—
—
—
—
—
—
—
—
GCLK4
—
—
—
—
Y
Y
—
—
—
—
—
—
GCLK5
—
—
—
—
Y
Y
—
—
—
—
—
—
GCLK6
—
—
—
—
Y
Y
—
—
—
—
—
—
GCLK7
—
—
—
—
Y
Y
—
—
—
—
—
—
GCLK8
—
—
—
—
—
—
Y
Y
Y
Y
—
—
GCLK9
—
—
—
—
—
—
Y
Y
Y
Y
—
—
GCLK10
—
—
—
—
—
—
Y
Y
Y
Y
—
—
GCLK11
—
—
—
—
—
—
Y
Y
Y
Y
—
—
GCLK12
—
—
—
—
—
—
—
—
—
—
Y
Y
GCLK13
—
—
—
—
—
—
—
—
—
—
Y
Y
GCLK14
—
—
—
—
—
—
—
—
—
—
Y
Y
GCLK15
—
—
—
—
—
—
—
—
—
—
Y
Y
Note to Table 5–5:
(1) Only PLL counter outputs C0 - C3 can drive the GCLK networks.
Table 5–6 lists how the PLL clock outputs connect to the RCLK networks.
Table 5–6. Stratix IV RCLK Outputs From the PLL Clock Outputs
(1)
(Part 1 of 2)
PLL Number
Clock Resource
September 2012
L1
L2
L3
L4
B1
B2
R1
R2
R3
R4
T1
T2
RCLK[0..11]
—
Y
Y
—
—
—
—
—
—
—
—
—
RCLK[12..31]
—
—
—
—
Y
Y
—
—
—
—
—
—
RCLK[32..43]
—
—
—
—
—
—
—
Y
Y
—
—
—
RCLK[44..63]
—
—
—
—
—
—
—
—
—
—
Y
Y
Altera Corporation
Stratix IV Device Handbook
Volume 1
5–14
Chapter 5: Clock Networks and PLLs in Stratix IV Devices
Clock Networks in Stratix IV Devices
Table 5–6. Stratix IV RCLK Outputs From the PLL Clock Outputs
(1)
(Part 2 of 2)
PLL Number
Clock Resource
L1
L2
L3
L4
B1
B2
R1
R2
R3
R4
T1
T2
RCLK[64..69]
—
—
—
Y
—
—
—
—
—
—
—
—
RCLK[70..75]
—
—
—
—
—
—
—
—
—
Y
—
—
RCLK[76..81]
—
—
—
—
—
—
Y
—
—
—
—
—
RCLK[82..87]
Y
—
—
—
—
—
—
—
—
—
—
—
Note to Table 5–6:
(1) All PLL counter outputs can drive the RCLK networks.
Clock Control Block
Every GCLK and RCLK network has its own clock control block. The control block
provides the following features:
■
Clock source selection (dynamic selection for GCLKs)
■
Global clock multiplexing
■
Clock power down (static or dynamic clock enable or disable)
Figure 5–11 and Figure 5–12 show the GCLK and RCLK select blocks, respectively.
You can select the clock source for the GCLK select block either statically or
dynamically. You can statically select the clock source using a setting in the Quartus II
software or you can dynamically select the clock source using internal logic to drive
the multiplexer-select inputs. When selecting the clock source dynamically, you can
select either PLL outputs (such as C0 or C1) or a combination of clock pins or PLL
outputs.
Figure 5–11. Stratix IV GCLK Control Block
CLKp
Pins
PLL Counter
Outputs
CLKSELECT[1..0]
(1)
2
2
CLKn
Pin
Internal
Logic
2
Static Clock
Select (2)
This multiplexer
supports user-controllable
dynamic switching
Enable/
Disable
Internal
Logic
GCLK
Notes to Figure 5–11:
(1) When the device is operating in user mode, you can dynamically control the clock select signals through internal
logic.
(2) When the device is operation in user mode, you can only set the clock select signals through a configuration file
(SRAM object file [.sof] or programmer object file [.pof]) and cannot be dynamically controlled.
Stratix IV Device Handbook
Volume 1
September 2012 Altera Corporation
Chapter 5: Clock Networks and PLLs in Stratix IV Devices
Clock Networks in Stratix IV Devices
5–15
The mapping between the input clock pins, PLL counter outputs, and clock control
block inputs is as follows:
■
inclk[0] and inclk[1]—can be fed by any of the four dedicated clock pins on the
same side of the Stratix IV device
■
inclk[2]—can be fed by PLL counters C0 and C2 from the two center PLLs on the
same side of the Stratix IV device
■
inclk[3]—can be fed by PLL counters C1 and C3 from the two center PLLs on the
same side of the Stratix IV device
The corner PLLs (L1, L4, R1, and R4) and the corresponding clock input pins
(PLL_L1_CLK and so forth) do not support dynamic selection for the GCLK network.
The clock source selection for the GCLK and RCLK networks from the corner PLLs
(L1, L4, R1, and R4) and the corresponding clock input pins (PLL_L1_CLK and so forth)
are controlled statically using configuration bit settings in the configuration file (.sof
or .pof) generated by the Quartus II software.
Figure 5–12. RCLK Control Block
CLKp
Pin
PLL Counter
Outputs
CLKn
Pin (2)
2
Internal
Logic
Static Clock Select (1)
Enable/
Disable
Internal
Logic
RCLK
Notes to Figure 5–12:
(1) When the device is operation in user mode, you can only set the clock select signals through a configuration file (.sof
or .pof) and cannot be dynamically controlled.
(2) The CLKn pin is not a dedicated clock input when used as a single-ended PLL clock input.
You can only control the clock source selection for the RCLK select block statically
using configuration bit settings in the configuration file (.sof or .pof) generated by the
Quartus II software.
You can power down the Stratix IV clock networks using both static and dynamic
approaches. When a clock network is powered down, all the logic fed by the clock
network is in off-state, thereby reducing the overall power consumption of the device.
The unused GCLK and RCLK networks are automatically powered down through
configuration bit settings in the configuration file (.sof or .pof) generated by the
Quartus II software. The dynamic clock enable or disable feature allows the internal
logic to control power-up or power-down synchronously on the GCLK and RCLK
networks, including dual-regional clock regions. This function is independent of the
PLL and is applied directly on the clock network, as shown in Figure 5–11 and
Figure 5–12.
September 2012
Altera Corporation
Stratix IV Device Handbook
Volume 1
5–16
Chapter 5: Clock Networks and PLLs in Stratix IV Devices
Clock Networks in Stratix IV Devices
You can set the input clock sources and the clkena signals for the GCLK and RCLK
network multiplexers through the Quartus II software using the ALTCLKCTRL
megafunction. You can also enable or disable the dedicated external clock output pins
using the ALTCLKCTRL megafunction. Figure 5–13 shows the external PLL output
clock control block.
1
When using the ALTCLKCTRL megafunction to implement dynamic clock source
selection, the inputs from the clock pins feed the inclk[0..1] ports of the multiplexer,
while the PLL outputs feed the inclk[2..3] ports. You can choose from among these
inputs using the CLKSELECT[1..0] signal.
f For more information, refer to the Clock Control Block (ALTCLKCTRL) Megafunction
User Guide.
Figure 5–13. Stratix IV External PLL Output Clock Control Block
PLL Counter
Outputs
7 or 10
Static Clock Select (1)
Enable/
Disable
Internal
Logic
IOE (2)
Internal
Logic
Static Clock
Select (1)
PLL_<#>_CLKOUT pin
Notes to Figure 5–13:
(1) When the device is operation in user mode, you can only set the clock select signals through a configuration file (.sof
or .pof) and cannot be dynamically controlled.
(2) The clock control block feeds to a multiplexer within the PLL_<#>_CLKOUT pin’s IOE. The PLL_<#>_CLKOUT
pin is a dual-purpose pin. Therefore, this multiplexer selects either an internal signal or the output of the clock control
block.
Stratix IV Device Handbook
Volume 1
September 2012 Altera Corporation
Chapter 5: Clock Networks and PLLs in Stratix IV Devices
Clock Networks in Stratix IV Devices
5–17
Clock Enable Signals
Figure 5–14 shows how the clock enable and disable circuit of the clock control block
is implemented in Stratix IV devices.
Figure 5–14. clkena Implementation
(1)
(1)
clkena
output of clock
select mux
Q
D
R1
(2)
Q
D
R2
GCLK/
RCLK/
PLL_<#>_CLKOUT (1)
Notes to Figure 5–14:
(1) The R1 and R2 bypass paths are not available for the PLL external clock outputs.
(2) The select line is statically controlled by a bit setting in the configuration file (.sof or .pof).
In Stratix IV devices, the clkena signals are supported at the clock network level
instead of at the PLL output counter level. This allows you to gate off the clock even
when you are not using a PLL. You can also use the clkena signals to control the
dedicated external clocks from the PLLs. Figure 5–15 shows a waveform example for
a clock output enable. clkena is synchronous to the falling edge of the clock output.
Stratix IV devices also have an additional metastability register that aids in
asynchronous enable and disable of the GCLK and RCLK networks. You can
optionally bypass this register in the Quartus II software.
Figure 5–15. clkena Signals
(1)
output of clock
select mux
clkena
output of AND gate
with R2 bypassed
output of AND gate
with R2 not bypassed
Note to Figure 5–15:
(1) You can use the clkena signals to enable or disable the GCLK and RCLK networks or the PLL_<#>_CLKOUT pins.
September 2012
Altera Corporation
Stratix IV Device Handbook
Volume 1
5–18
Chapter 5: Clock Networks and PLLs in Stratix IV Devices
Clock Networks in Stratix IV Devices
The PLL can remain locked independent of the clkena signals because the
loop-related counters are not affected. This feature is useful for applications that
require a low-power or sleep mode. The clkena signal can also disable clock outputs if
the system is not tolerant of frequency over-shoot during resynchronization.
Clock Source Control for PLLs
The clock input to Stratix IV PLLs comes from clock input multiplexers. The clock
multiplexer inputs come from dedicated clock input pins, PLLs through the GCLK
and RCLK networks, or from dedicated connections between adjacent top/bottom
and left/right PLLs. The clock input sources to top/bottom and left/right PLLs (L2,
L3, T1, T2, B1, B2, R2, and R3) are shown in Figure 5–16; the corresponding clock
input sources to left and right PLLs (L1, L4, R1, and R4) are shown in Figure 5–17.
The multiplexer select lines are only set in the configuration file (.sof or .pof). After
programmed, this block cannot be changed without loading a new configuration file
(.sof or .pof). The Quartus II software automatically sets the multiplexer select signals
depending on the clock sources selected in the design.
Figure 5–16. Clock Input Multiplexer Logic for L2, L3, T1, T2, B1, B2, R2, and R3 PLLs
(1)
clk[n+3..n] (2)
GCLK / RCLK input (3)
4
inclk0
To the clock
switchover block
Adjacent PLL output
(1)
inclk1
4
Notes to Figure 5–16:
(1) When the device is operating in user mode, input clock multiplexing is controlled through a configuration file (.sof
or .pof) only and cannot be dynamically controlled.
(2) n=0 for L2 and L3 PLLs; n=4 for B1 and B2 PLLs; n=8 for R2 and R3 PLLs, and n=12 for T1 and T2 PLLs.
(3) You can drive the GCLK or RCLK input using an output from another PLL, a pin-driven GCLK or RCLK, or through a
clock control block provided the clock control block is fed by an output from another PLL or a pin-driven dedicated
GCLK or RCLK. An internally generated global signal or general purpose I/O pin cannot drive the PLL.
Figure 5–17. Clock Input Multiplexer Logic for L1, L4, R1, and R4 PLLs
PLL_<L1/L4/R1/R4>_CLK (1)
inclk0
GCLK/RCLK (2)
4
CLK[0..3] or CLK[8..11] (3)
inclk1
4
Notes to Figure 5–17:
(1) Dedicated clock input pins to the PLLs are L1, L4, R1, and R4, respectively. For example, PLL_L1_CLK is the
dedicated clock input for PLL_L1.
(2) You can drive the GCLK or RCLK input using an output from another PLL, a pin-driven GCLK or RCLK, or through a
clock control block provided the clock control block is fed by an output from another PLL or a pin-driven dedicated
GCLK or RCLK. An internally generated global signal or general purpose I/O pin cannot drive the PLL.
(3) The center clock pins can feed the corner PLLs on the same side directly through a dedicated path. However, these
paths may not be fully compensated.
Stratix IV Device Handbook
Volume 1
September 2012 Altera Corporation
Chapter 5: Clock Networks and PLLs in Stratix IV Devices
PLLs in Stratix IV Devices
5–19
Cascading PLLs
You can cascade the left/right and top/bottom PLLs through the GCLK and RCLK
networks. In addition, where two left/right or top/bottom PLLs exist next to each
other, there is a direct connection between them that does not require the GCLK or
RCLK network. Using this path reduces clock jitter when cascading PLLs.
1
Stratix IV GX devices allow cascading the left and right PLLs to transceiver PLLs
(CMU PLLs and receiver CDRs).
f For more information, refer to the “FPGA Fabric PLLs -Transceiver PLLs Cascading”
section in the Transceiver Clocking in Stratix IV Devices chapter.
When cascading PLLs in Stratix IV devices, the source (upstream) PLL must have a
low-bandwidth setting while the destination (downstream) PLL must have a
high-bandwidth setting. Ensure that there is no overlap of the bandwidth ranges of
the two PLLs.
f For more information about PLL cascading in external memory interfaces designs,
refer to the External Memory PHY Interface (ALTMEMPHY) (nonAFI) Megafunction User
Guide.
PLLs in Stratix IV Devices
Stratix IV devices offer up to 12 PLLs that provide robust clock management and
synthesis for device clock management, external system clock management, and
high-speed I/O interfaces. The nomenclature for the PLLs follows their geographical
location in the device floor plan. The PLLs that reside on the top and bottom sides of
the device are named PLL_T1, PLL_T2, PLL_B1 and PLL_B2; the PLLs that reside on the
left and right sides of the device are named PLL_L1, PLL_L2, PLL_L3, PLL_L4, PLL_R1,
PLL_R2, PLL_R3, and PLL_R4.
Table 5–7 lists the number of PLLs available in the Stratix IV device family.
Table 5–7. PLL Availability for Stratix IV Devices (Part 1 of 2)
Device
Package
L1
L2
L3
L4
T1
T2
B1
B2
R1
R2
R3
R4
EP4S40G2
F1517
—
Y
Y
—
Y
Y
Y
Y
—
Y
Y
—
EP4S40G5
H1517
—
Y
Y
—
Y
Y
Y
Y
—
Y
Y
—
EP4S100G2
F1517
—
Y
Y
—
Y
Y
Y
Y
—
Y
Y
—
EP4S100G3
F1932
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
EP4S100G4
F1932
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
H1517
—
Y
Y
—
Y
Y
Y
Y
—
Y
Y
—
F1932
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
F780
—
Y
—
—
Y
—
Y
—
—
Y
—
—
H780
—
Y
—
—
Y
—
Y
—
—
Y
—
—
F1152
—
Y
Y
—
Y
Y
Y
Y
—
Y
Y
—
EP4S100G5
EP4SE230
EP4SE360
September 2012
Altera Corporation
Stratix IV Device Handbook
Volume 1
5–20
Chapter 5: Clock Networks and PLLs in Stratix IV Devices
PLLs in Stratix IV Devices
Table 5–7. PLL Availability for Stratix IV Devices (Part 2 of 2)
Device
EP4SE530
EP4SE820
EP4SGX70
EP4SGX110
EP4SGX180
EP4SGX230
EP4SGX290
EP4SGX360
EP4SGX530
Package
L1
L2
L3
L4
T1
T2
B1
B2
R1
R2
R3
R4
H1152
—
Y
Y
—
Y
Y
Y
Y
—
Y
Y
—
H1517
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
F1760
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
H1152
—
Y
Y
—
Y
Y
Y
Y
—
Y
Y
—
H1517
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
F1760
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
F780
—
Y
—
—
Y
—
Y
—
—
—
—
—
F1152
—
Y
—
—
Y
—
Y
—
—
Y
—
—
F780
—
Y
—
—
Y
—
Y
—
—
—
—
—
F1152
—
Y
—
—
Y
—
Y
—
—
Y
—
—
F780
—
Y
—
—
Y
—
Y
—
—
—
—
—
F1152
—
Y
—
—
Y
Y
Y
Y
—
Y
—
—
F1517
—
Y
Y
—
Y
Y
Y
Y
—
Y
Y
—
F780
—
Y
—
—
Y
—
Y
—
—
—
—
—
F1152
—
Y
—
—
Y
Y
Y
Y
—
Y
—
—
F1517
—
Y
Y
—
Y
Y
Y
Y
—
Y
Y
—
H780
—
—
—
—
Y
Y
Y
Y
—
—
—
—
F1152
—
Y
—
—
Y
Y
Y
Y
—
Y
—
—
F1517
—
Y
Y
—
Y
Y
Y
Y
—
Y
Y
—
F1760
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
F1932
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
H780
—
—
—
—
Y
Y
Y
Y
—
—
—
—
F1152
—
Y
—
—
Y
Y
Y
Y
—
Y
—
—
F1517
—
Y
Y
—
Y
Y
Y
Y
—
Y
Y
—
F1760
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
F1932
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
H1152
—
Y
—
—
Y
Y
Y
Y
—
Y
—
—
H1517
—
Y
Y
—
Y
Y
Y
Y
—
Y
Y
—
F1760
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
F1932
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
All Stratix IV PLLs have the same core analog structure with only minor differences in
the features that are supported. Table 5–8 lists the features of top/bottom and
left/right PLLs in Stratix IV devices.
Table 5–8. PLL Features in Stratix IV Devices (Part 1 of 2)
Feature
(1)
Stratix IV Top/Bottom PLLs
Stratix IV Left/Right PLLs
C (output) counters
10
7
M, N, C counter sizes
1 to 512
1 to 512
6 single-ended or 4 single-ended and 1
differential pair
2 single-ended or 1 differential pair
Dedicated clock outputs
Stratix IV Device Handbook
Volume 1
September 2012 Altera Corporation
Chapter 5: Clock Networks and PLLs in Stratix IV Devices
PLLs in Stratix IV Devices
5–21
Table 5–8. PLL Features in Stratix IV Devices (Part 2 of 2)
Feature
Clock input pins
(2)
External feedback input pin
(1)
Stratix IV Top/Bottom PLLs
Stratix IV Left/Right PLLs
4 single-ended or 4 differential pin pairs
4 single-ended or 4 differential pin
pairs
Single-ended or differential
Single-ended only
Spread-spectrum input clock tracking
Yes
(3)
Yes
Through GCLK and RCLK and a dedicated
path between adjacent PLLs
PLL cascading
Compensation modes
Through GCLK and RCLK and
dedicated path between adjacent PLLs
(4)
All except LVDS clock network
compensation
All except external feedback mode
when using differential I/Os
No
Yes
PLL drives LVDSCLK and LOADEN
VCO output drives the DPA clock
Phase shift resolution
(3)
No
Down to 96.125 ps
Yes
(5)
Down to 96.125 ps
Programmable duty cycle
Yes
Yes
Output counter cascading
Yes
Yes
Input clock switchover
Yes
Yes
(5)
Notes to Table 5–8:
(1) While there is pin compatibility, there is no hard IP block placement compatibility.
(2) General purpose I/O pins cannot drive the PLL clock input pins.
(3) Provided input clock jitter is within input jitter tolerance specifications.
(4) The dedicated path between adjacent PLLs is not available on L1, L4, R1, and R4 PLLs.
(5) The smallest phase shift is determined by the voltage-controlled oscillator (VCO) period divided by eight. For degree increments, the Stratix IV
device can shift all output frequencies in increments of at least 45°. Smaller degree increments are possible depending on the frequency and
divide parameters.
September 2012
Altera Corporation
Stratix IV Device Handbook
Volume 1
5–22
Chapter 5: Clock Networks and PLLs in Stratix IV Devices
PLLs in Stratix IV Devices
Figure 5–18 shows the location of PLLs in Stratix IV devices.
Figure 5–18. PLL Locations in Stratix IV Devices
Top/Bottom PLLs
Top/Bottom PLLs
CLK[12..15]
T1 T2
PLL_L1_CLK
Left/Right PLLs
CLK[0..3]
Left/Right PLLs
PLL_L4_CLK
L1
Q1 Q2
L2
L3
Q4 Q3
L4
R1
PLL_R1_CLK
R2
R3
CLK[8..11]
R4
PLL-R4_CLK
Left/Right PLLs
Left/Right PLLs
B1 B2
CLK[4..7]
Top/Bottom PLLs
Stratix IV Device Handbook
Volume 1
Top/Bottom PLLs
September 2012 Altera Corporation
Chapter 5: Clock Networks and PLLs in Stratix IV Devices
PLLs in Stratix IV Devices
5–23
Stratix IV PLL Hardware Overview
Stratix IV devices contain up to 12 PLLs with advanced clock management features.
The goal of a PLL is to synchronize the phase and frequency of an internal or external
clock to an input reference clock. There are a number of components that comprise a
PLL to achieve this phase alignment.
Stratix IV PLLs align the rising edge of the input reference clock to a feedback clock
using the phase-frequency detector (PFD). The falling edges are determined by the
duty-cycle specifications. The PFD produces an up or down signal that determines
whether the VCO must operate at a higher or lower frequency. The output of the PFD
feeds the charge pump and loop filter, which produces a control voltage for setting the
VCO frequency. If the PFD produces an up signal, the VCO frequency increases. A
down signal decreases the VCO frequency. The PFD outputs these up and down
signals to a charge pump. If the charge pump receives an up signal, current is driven
into the loop filter. Conversely, if the charge pump receives a down signal, current is
drawn from the loop filter.
The loop filter converts these up and down signals to a voltage that is used to bias the
VCO. The loop filter also removes glitches from the charge pump and prevents
voltage over-shoot, which filters the jitter on the VCO. The voltage from the loop filter
determines how fast the VCO operates. A divide counter (m) is inserted in the
feedback loop to increase the VCO frequency above the input reference frequency.
VCO frequency (fVCO) is equal to (m) times the input reference clock (fREF). The input
reference clock (fREF) to the PFD is equal to the input clock (fIN) divided by the
pre-scale counter (N). Therefore, the feedback clock (fFB) applied to one input of the
PFD is locked to the fREF that is applied to the other input of the PFD.
The VCO output from the left and right PLLs can feed seven post-scale counters
(C[0..6]), while the corresponding VCO output from the top and bottom PLLs can
feed ten post-scale counters (C[0..9]). These post-scale counters allow a number of
harmonically related frequencies to be produced by the PLL.
September 2012
Altera Corporation
Stratix IV Device Handbook
Volume 1
5–24
Chapter 5: Clock Networks and PLLs in Stratix IV Devices
PLLs in Stratix IV Devices
Figure 5–19 shows a simplified block diagram of the major components of the
Stratix IV PLL.
Figure 5–19. Stratix IV PLL Block Diagram
To DPA block on
Left/Right PLLs
Lock
Circuit
pfdena
Casade output
to adjacent PLL
locked
/2, /4
÷C0
4
÷n
inclk0
inclk1
GCLK/RCLK
Clock
Switchover
Block
PFD
CP
LF
VCO
8
÷2
(2)
8
÷C1
8
÷C2
clkswitch
clkbad0
clkbad1
activeclock
÷C3
Cascade input
from adjacent PLL
÷Cn (1)
÷m
no compensation mode
ZDB, External feedback modes
LVDS Compensation mode
Source Synchronous, normal modes
PLL Output Mux
GCLKs
Dedicated
clock inputs
RCLKs
External clock
outputs
DIFFIOCLK from
Left/Right PLLs
LOAD_EN from
Left/Right PLLs
FBOUT (3)
External
memory
interface DLL
FBIN
DIFFIOCLK network
GCLK/RCLK network
Notes to Figure 5–19:
(1) The number of post-scale counters is seven for left and right PLLs and ten for top and bottom PLLs.
(2) This is the VCO post-scale counter K.
(3) The FBOUT port is fed by the M counter in Stratix IV PLLs.
1
You can drive the GCLK or RCLK inputs using an output from another PLL, a
pin-driven GCLK or RCLK, or through a clock control block provided the clock
control block is fed by an output from another PLL or a pin-driven dedicated GCLK
or RCLK. An internally generated global signal or general purpose I/O pin cannot
drive the PLL.
PLL Clock I/O Pins
Each top and bottom PLL supports six clock I/O pins, organized as three pairs of
pins:
Stratix IV Device Handbook
Volume 1
■
1st pair—two single-ended I/O or one differential I/O
■
2nd pair—two single-ended I/O or one differential external feedback input
(FBp/FBn)
■
3rd pair—two single-ended I/O or one differential input
September 2012 Altera Corporation
Chapter 5: Clock Networks and PLLs in Stratix IV Devices
PLLs in Stratix IV Devices
5–25
Figure 5–20 shows the clock I/O pins associated with the top and bottom PLLs.
Figure 5–20. External Clock Outputs for Top and Bottom PLLs
Internal Logic
C0
C1
C2
C3
Top/Bottom
PLLs
C4
C5
C6
C7
C8
C9
m(fbout)
clkena0 (3)
clkena2 (3)
clkena4 (3)
clkena1 (3)
clkena3 (3)
clkena5 (3)
PLL_<#>_CLKOUT0p (1), (2)
PLL_<#>_FBp/CLKOUT1 (1), (2)
PLL_<#>_CLKOUT0n (1), (2)
PLL_<#>_CLKOUT3
(1), (2)
PLL_<#>_FBn/CLKOUT2 (1), (2)
PLL_<#>_CLKOUT4
(1), (2)
Notes to Figure 5–20:
(1) You can feed these clock output pins using any one of the C[9..0], m counters.
(2) The CLKOUT0p and CLKOUT0n pins can be either single-ended or differential clock outputs. The CLKOUT1 and CLKOUT2 pins are
dual-purpose I/O pins that you can use as two single-ended outputs or one differential external feedback input pin. The CLKOUT3 and CLKOUT4
pins are two single-ended output pins.
(3) These external clock enable signals are available only when using the ALTCLKCTRL megafunction.
September 2012
Altera Corporation
Stratix IV Device Handbook
Volume 1
5–26
Chapter 5: Clock Networks and PLLs in Stratix IV Devices
PLLs in Stratix IV Devices
Any of the output counters (C[9..0] on the top and bottom PLLs and C[6..0] on the
left and right PLLs) or the M counter can feed the dedicated external clock outputs, as
shown in Figure 5–20 and Figure 5–21. Therefore, one counter or frequency can drive
all output pins available from a given PLL.
Each left and right PLL supports two clock I/O pins, configured as either two
single-ended I/Os or one differential I/O pair. When using both pins as single-ended
I/Os, one of them can be the clock output while the other pin is the external feedback
input (FB) pin. Therefore, for single-ended I/O standards, the left and right PLLs only
support external feedback mode.
Figure 5–21. External Clock Outputs for Left and Right PLLs
Internal Logic
C0
C1
C2
LEFT/RIGHT
PLLs
C3
C4
C5
C6
m(fbout)
clkena0 (3)
clkena1 (3)
PLL_<L2, L3, R2, R3>_CLKOUT0n/FB_CLKOUT0p (1), (2)
PLL_<L2, L3, R2, R3>_FB_CLKOUT0p/CLKOUT0n (1), (2)
Notes to Figure 5–21:
(1) You can feed these clock output pins using any one of the C[6..0], m counters.
(2) The CLKOUT0p and CLKOUT0n pins are dual-purpose I/O pins that you can use as two single-ended outputs or one single-ended output and
one external feedback input pin.
(3) These external clock enable signals are available only when using the ALTCLKCTRL megafunction.
Each pin of a single-ended output pair can either be in-phase or 180° out-of-phase.
The Quartus II software places the NOT gate in the design into the IOE to implement
the 180° phase with respect to the other pin in the pair. The clock output pin pairs
support the same I/O standards as standard output pins (in the top and bottom
banks) as well as LVDS, LVPECL, differential High-Speed Transceiver Logic (HSTL),
and differential SSTL.
f To determine which I/O standards are supported by the PLL clock input and output
pins, refer to the I/O Features in Stratix IV Devices chapter.
Stratix IV PLLs can also drive out to any regular I/O pin through the GCLK or RCLK
network. You can also use the external clock output pins as user I/O pins if you do
not need external PLL clocking.
Stratix IV Device Handbook
Volume 1
September 2012 Altera Corporation
Chapter 5: Clock Networks and PLLs in Stratix IV Devices
PLLs in Stratix IV Devices
5–27
PLL Control Signals
You can use the pfdena, areset, and locked signals to observe and control PLL
operation and resynchronization.
pfdena
Use the pfdena signal to maintain the most recent locked frequency so your system
has time to store its current settings before shutting down. The pfdena signal controls
the PFD output with a programmable gate. If you disable PFD, the VCO operates at
its most recent set value of control voltage and frequency, with some long-term drift to
a lower frequency. The PLL continues running even if it goes out-of-lock or the input
clock is disabled. You can use either your own control signal or the control signals
available from the clock switchover circuit (activeclock, clkbad[0], or clkbad[1]) to
control pfdena.
areset
The areset signal is the reset or resynchronization input for each PLL. The device
input pins or internal logic can drive these input signals. When areset is driven high,
the PLL counters reset, clearing the PLL output and placing the PLL out-of-lock. The
VCO is then set back to its nominal setting. When areset is driven low again, the PLL
resynchronizes to its input as it re-locks.
You must assert the areset signal every time the PLL loses lock to guarantee the
correct phase relationship between the PLL input and output clocks. You can set up
the PLL to automatically reset (self reset) after a loss-of-lock condition using the
Quartus II MegaWizard Plug-In Manager. You must include the areset signal in
designs if either of the following conditions is true:
1
■
PLL reconfiguration or clock switchover is enabled in the design
■
Phase relationships between the PLL input and output clocks must be maintained
after a loss-of-lock condition
If the input clock to the PLL is not toggling or is unstable after power up, assert the
areset signal after the input clock is stable and within specifications.
locked
The locked signal output of the PLL indicates that the PLL has locked onto the
reference clock and the PLL clock outputs are operating at the desired phase and
frequency set in the Quartus II MegaWizard Plug-In Manager. The lock detection
circuit provides a signal to the core logic that gives an indication when the feedback
clock has locked onto the reference clock both in phase and frequency.
1
September 2012
Altera recommends using the areset and locked signals in your designs to control
and observe the status of your PLL.
Altera Corporation
Stratix IV Device Handbook
Volume 1
5–28
Chapter 5: Clock Networks and PLLs in Stratix IV Devices
PLLs in Stratix IV Devices
Clock Feedback Modes
Stratix IV PLLs support up to six different clock feedback modes. Each mode allows
clock multiplication and division, phase shifting, and programmable duty cycle.
Table 5–9 lists the clock feedback modes supported by the Stratix IV device PLLs.
Table 5–9. Clock Feedback Mode Availability
Availability
Clock Feedback Mode
Top and Bottom PLLs
Left and Right PLLs
Source-synchronous
Yes
Yes
No-compensation
Yes
Yes
Normal
Yes
Yes
Yes
Yes
Zero-delay buffer (ZDB)
(1)
Yes
LVDS compensation
No
External feedback
Yes
(2)
Yes
Notes to Table 5–9:
(1) The high-bandwidth PLL setting is not supported in external feedback mode.
(2) External feedback mode is supported for single-ended inputs and outputs only on the left and right PLLs.
1
Stratix IV Device Handbook
Volume 1
The input and output delays are fully compensated by a PLL only when using the
dedicated clock input pins associated with a given PLL as the clock source. For
example, when using PLL_T1 in normal mode, the clock delays from the input pin to
the PLL clock output-to-destination register are fully compensated, provided the
clock input pin is one of the following two pins: CLK14 and CLK15. Compensated pins
are only in the same I/O bank as the PLL. When an RCLK or GCLK network drives
the PLL, the input and output delays may not be fully compensated in the Quartus II
software. Another example is when you configure PLL_T2 in zero-delay buffer mode
and the PLL input is driven by a dedicated clock input pin, a fully compensated clock
path results in zero-delay between the clock input and one of the output clocks from
the PLL. If the PLL input is instead fed by a non-dedicated input (using the GCLK
network), the output clock may not be perfectly aligned with the input clock.
September 2012 Altera Corporation
Chapter 5: Clock Networks and PLLs in Stratix IV Devices
PLLs in Stratix IV Devices
5–29
Source Synchronous Mode
If data and clock arrive at the same time on the input pins, the same phase
relationship is maintained at the clock and data ports of any IOE input register.
Figure 5–22 shows an example waveform of the clock and data in this mode. Altera
recommends source synchronous mode for source-synchronous data transfers. Data
and clock signals at the IOE experience similar buffer delays as long as you use the
same I/O standard.
Figure 5–22. Phase Relationship Between Clock and Data in Source-Synchronous Mode
Data pin
PLL
reference clock
at input pin
Data at register
Clock at register
Source-synchronous mode compensates for the delay of the clock network used plus
any difference in the delay between these two paths:
■
Data pin to the IOE register input
■
Clock input pin to the PLL PFD input
The Stratix IV PLL can compensate multiple pad-to-input-register paths, such as a
data bus when it is set to use source-synchronous compensation mode. You can use
the “PLL Compensation” assignment in the Quartus II software Assignment Editor to
select which input pins are used as the PLL compensation targets. You can include
your entire data bus, provided the input registers are clocked by the same output of a
source-synchronous-compensated PLL. In order for the clock delay to be properly
compensated, all of the input pins must be on the same side of the device. The PLL
compensates for the input pin with the longest pad-to-register delay among all input
pins in the compensated bus.
If you do not make the “PLL Compensation” assignment, the Quartus II software
automatically selects all of the pins driven by the compensated output of the PLL as
the compensation target.
September 2012
Altera Corporation
Stratix IV Device Handbook
Volume 1
5–30
Chapter 5: Clock Networks and PLLs in Stratix IV Devices
PLLs in Stratix IV Devices
Source-Synchronous Mode for LVDS Compensation
The goal of source-synchronous mode is to maintain the same data and clock timing
relationship seen at the pins of the internal serializer/deserializer (SERDES) capture
register, except that the clock is inverted (180° phase shift). Thus, source-synchronous
mode ideally compensates for the delay of the LVDS clock network plus any
difference in delay between these two paths:
■
Data pin-to-SERDES capture register
■
Clock input pin-to-SERDES capture register. In addition, the output counter must
provide the 180° phase shift
Figure 5–23 shows an example waveform of the clock and data in LVDS mode.
Figure 5–23. Phase Relationship Between the Clock and Data in LVDS Mode
Data pin
PLL
reference clock
at input pin
Data at register
Clock at register
No-Compensation Mode
In no-compensation mode, the PLL does not compensate for any clock networks. This
mode provides better jitter performance because the clock feedback into the PFD
passes through less circuitry. Both the PLL internal- and external-clock outputs are
phase-shifted with respect to the PLL clock input. Figure 5–24 shows an example
waveform of the PLL clocks’ phase relationship in no-compensation mode.
Figure 5–24. Phase Relationship Between the PLL Clocks in No Compensation Mode
Phase Aligned
PLL Reference
Clock at the
Input Pin
PLL Clock at the
Register Clock Port (1)
External PLL Clock Outputs (1)
Note to Figure 5–24:
(1) The PLL clock outputs lag the PLL input clocks depending on routine delays.
Stratix IV Device Handbook
Volume 1
September 2012 Altera Corporation
Chapter 5: Clock Networks and PLLs in Stratix IV Devices
PLLs in Stratix IV Devices
5–31
Normal Mode
An internal clock in normal mode is phase-aligned to the input clock pin. The external
clock-output pin has a phase delay relative to the clock input pin if connected in this
mode. The Quartus II software timing analyzer reports any phase difference between
the two. In normal mode, the delay introduced by the GCLK or RCLK network is fully
compensated. Figure 5–25 shows an example waveform of the PLL clocks’ phase
relationship in normal mode.
Figure 5–25. Phase Relationship Between the PLL Clocks in Normal Mode
Phase Aligned
PLL Reference
Clock at the
Input Pin
PLL Clock at the
Register Clock Port
Dedicated PLL Clock Outputs (1)
Note to Figure 5–25:
(1) The external clock output can lead or lag the PLL internal clock signals.
Zero-Delay Buffer (ZDB) Mode
In ZDB mode, the external clock output pin is phase-aligned with the clock input pin
for zero-delay through the device. When using this mode, you must use the same I/O
standard on the input clocks and output clocks to guarantee clock alignment at the
input and output pins. ZDB mode is supported on all Stratix IV PLLs.
When using Stratix IV PLLs in ZDB mode, along with single-ended I/O standards, to
ensure phase alignment between the CLK pin and the external clock output (CLKOUT)
pin, you must instantiate a bi-directional I/O pin in the design to serve as the
feedback path connecting the FBOUT and FBIN ports of the PLL. The PLL uses this
bi-directional I/O pin to mimic, and compensate for, the output delay from the clock
output port of the PLL to the external clock output pin. Figure 5–26 shows ZDB mode
in Stratix IV PLLs. When using ZDB mode, you cannot use differential I/O standards
on the PLL clock input or output pins.
1
September 2012
The bi-directional I/O pin that you instantiate in your design must always be
assigned a single-ended I/O standard.
Altera Corporation
Stratix IV Device Handbook
Volume 1
5–32
Chapter 5: Clock Networks and PLLs in Stratix IV Devices
PLLs in Stratix IV Devices
1
When using ZDB mode, to avoid signal reflection, do not place board traces on the
bi-directional I/O pin.
Figure 5–26. ZDB Mode in Stratix IV PLLs
inclk
÷n
PFD
CP/LF
VCO
÷C0
PLL_<#>_CLKOUT#
÷C1
PLL_<#>_CLKOUT#
÷m
fbout
bidirectional
I/O pin (1)
fbin
Note to Figure 5–26:
(1) The bidirectional I/O pin must be assigned to the PLL_<#>_FB_CLKOUT0p pin for left and right PLLs and to the PLL_<#>_FBp_/CLKOUT1 pin for
top and bottom PLLs.
Figure 5–27 shows an example waveform of the PLL clocks’ phase relationship in
ZDB mode.
Figure 5–27. Phase Relationship Between the PLL Clocks in ZDB Mode
Phase Aligned
PLL Reference
Clock at the
Input Pin
PLL Clock at the
Register Clock Port (1)
Dedicated PLL
Clock Outputs
Note to Figure 5–27:
(1) The internal PLL clock output can lead or lag the external PLL clock outputs.
External Feedback Mode
In external feedback mode, the external feedback input pin (fbin) is phase-aligned
with the clock input pin, as shown in Figure 5–28. Aligning these clocks allows you to
remove clock delay and skew between devices. This mode is supported on all
Stratix IV PLLs.
In external feedback mode, the output of the M counter (FBOUT) feeds back to the PLL
fbin input (using a trace on the board) becoming part of the feedback loop. Also, use
one of the dual-purpose external clock outputs as the fbin input pin in this mode.
When using external feedback mode, you must use the same I/O standard on the
input clock, feedback input, and output clocks. Left and right PLLs support this mode
when using single-ended I/O standards only.
Stratix IV Device Handbook
Volume 1
September 2012 Altera Corporation
Chapter 5: Clock Networks and PLLs in Stratix IV Devices
PLLs in Stratix IV Devices
5–33
Figure 5–28 shows an example waveform of the phase relationship between the PLL
clocks in external feedback mode.
Figure 5–28. Phase Relationship Between the PLL Clocks in External Feedback Mode
Phase Aligned
PLL Reference
Clock at the
Input Pin
PLL Clock at
the Register
Clock Port (1)
Dedicated PLL
Clock Outputs (1)
fbin Clock Input Pin
Note to Figure 5–28:
(1) The PLL clock outputs can lead or lag the fbin clock input.
Figure 5–29 shows external feedback mode implementation in Stratix IV devices.
Figure 5–29. External Feedback Mode in Stratix IV Devices
inclk
÷n
PFD
CP/LF
VCO
PLL_<#>_CLKOUT#
÷C0
PLL_<#>_CLKOUT#
÷C1
÷m
fbout
fbin
external
board
trace
Clock Multiplication and Division
Each Stratix IV PLL provides clock synthesis for PLL output ports using
M/(N* post-scale counter) scaling factors. The input clock is divided by a pre-scale
factor, n, and is then multiplied by the m feedback factor. The control loop drives the
VCO to match fin (M/N). Each output port has a unique post-scale counter that
divides down the high-frequency VCO. For multiple PLL outputs with different
frequencies, the VCO is set to the least common multiple of the output frequencies
that meets its frequency specifications. For example, if the output frequencies required
from one PLL are 33 and 66 MHz, the Quartus II software sets the VCO to 660 MHz
(the least common multiple of 33 and 66 MHz within the VCO range). Then the
post-scale counters scale down the VCO frequency for each output port.
September 2012
Altera Corporation
Stratix IV Device Handbook
Volume 1
5–34
Chapter 5: Clock Networks and PLLs in Stratix IV Devices
PLLs in Stratix IV Devices
Each PLL has one pre-scale counter, n, and one multiply counter, m, with a range of
1 to 512 for both m and n. The n counter does not use duty-cycle control because the
only purpose of this counter is to calculate frequency division. There are seven generic
post-scale counters per left or right PLL and ten post-scale counters per top or bottom
PLL that can feed the GCLKs, RCLKs, or external clock outputs. These post-scale
counters range from 1 to 512 with a 50% duty cycle setting. The high- and low-count
values for each counter range from 1 to 256. The sum of the high- and low-count
values chosen for a design selects the divide value for a given counter.
The Quartus II software automatically chooses the appropriate scaling factors
according to the input frequency, multiplication, and division values entered into the
ALTPLL megafunction.
Post-Scale Counter Cascading
Stratix IV PLLs support post-scale counter cascading to create counters larger than
512. This is automatically implemented in the Quartus II software by feeding the
output of one C counter into the input of the next C counter, as shown in Figure 5–30.
Figure 5–30. Counter Cascading
VCO Output
C0
VCO Output
C1
VCO Output
C2
VCO Output
C3
VCO Output
C4
from preceding
post-scale counter
VCO Output
Cn
(1)
Note to Figure 5–30:
(1) N = 6 or N = 9
When cascading post-scale counters to implement a larger division of the
high-frequency VCO clock, the cascaded counters behave as one counter with the
product of the individual counter settings. For example, if C0 = 40 and C1 = 20, the
cascaded value is C0 × C1 = 800.
1
Stratix IV Device Handbook
Volume 1
Post-scale counter cascading is set in the configuration file. You cannot set this using
PLL reconfiguration.
September 2012 Altera Corporation
Chapter 5: Clock Networks and PLLs in Stratix IV Devices
PLLs in Stratix IV Devices
5–35
Programmable Duty Cycle
The programmable duty cycle allows PLLs to generate clock outputs with a variable
duty cycle. This feature is supported on the PLL post-scale counters. The duty-cycle
setting is achieved by a low and high time-count setting for the post-scale counters. To
determine duty cycle choices, the Quartus II software uses the frequency input and
the required multiply or divide rate. The post-scale counter value determines the
precision of the duty cycle. Precision is defined as 50% divided by the post-scale
counter value. For example, if the C0 counter is 10, steps of 5% are possible for
duty-cycle choices from 5% to 90%.
If the PLL is in external feedback mode, set the duty cycle for the counter driving the
fbin pin to 50%. Combining the programmable duty cycle with programmable phase
shift allows the generation of precise non-overlapping clocks.
Programmable Phase Shift
Use phase shift to implement a robust solution for clock delays in Stratix IV devices.
Implement phase shift by using a combination of the VCO phase output and the
counter starting time. A combination of VCO phase output and counter starting time
is the most accurate method of inserting delays because it is only based on counter
settings, which are independent of process, voltage, and temperature (PVT).
You can phase-shift the output clocks from the Stratix IV PLLs in either of these two
resolutions:
■
Fine resolution using VCO phase taps
■
Coarse resolution using counter starting time
Implement fine-resolution phase shifts by allowing any of the output counters
(C[n..0]) or the m counter to use any of the eight phases of the VCO as the reference
clock. This allows you to adjust the delay time with a fine resolution. Equation 5–1
shows the minimum delay time that you can insert using this method.
Equation 5–1. Fine-Resolution Phase Shift
Φfine =
1
T
=
8 VCO
N
1
=
8fVCO 8MfREF
where fREF is the input reference clock frequency.
For example, if fREF is 100 MHz, N is 1, and M is 8, then fVCO is 800 MHz and fine
equals 156.25 ps. This phase shift is defined by the PLL operating frequency, which is
governed by the reference clock frequency and the counter settings.
September 2012
Altera Corporation
Stratix IV Device Handbook
Volume 1
5–36
Chapter 5: Clock Networks and PLLs in Stratix IV Devices
PLLs in Stratix IV Devices
Equation 5–2 shows the coarse-resolution phase shifts are implemented by delaying
the start of the counters for a predetermined number of counter clocks.
Equation 5–2. Coarse-Resolution Phase Shift
Φcoarse =
C − 1 (C − 1)N
=
fVco
MfREF
where C is the count value set for the counter delay time (this is the initial setting in
the “PLL usage” section of the compilation report in the Quartus II software). If the
initial value is 1, C – 1 = 0° phase shift.
Figure 5–31 shows an example of phase-shift insertion with fine resolution using the
VCO phase-taps method. The eight phases from the VCO are shown and labeled for
reference. For this example, CLK0 is based on the 0phase from the VCO and has the C
value for the counter set to one. The CLK1 signal is divided by four, two VCO clocks
for high time and two VCO clocks for low time. CLK1 is based on the 135° phase tap
from the VCO and also has the C value for the counter set to one. In this case, the two
clocks are offset by 3 FINE. CLK2 is based on the 0phase from the VCO but has the
C value for the counter set to three. This arrangement creates a delay of 2 COARSE
(two complete VCO periods).
Figure 5–31. Delay Insertion Using VCO Phase Output and Counter Delay Time
1/8 tVCO
tVCO
0
45
90
135
180
225
270
315
CLK0
td0-1
CLK1
td0-2
CLK2
You can use coarse- and fine-phase shifts to implement clock delays in Stratix IV
devices.
Stratix IV devices support dynamic phase-shifting of VCO phase taps only. You can
reconfigure the phase shift any number of times. Each phase shift takes about one
SCANCLK cycle, allowing you to implement large phase shifts quickly.
Stratix IV Device Handbook
Volume 1
September 2012 Altera Corporation
Chapter 5: Clock Networks and PLLs in Stratix IV Devices
PLLs in Stratix IV Devices
5–37
Programmable Bandwidth
Stratix IV PLLs provide advanced control of the PLL bandwidth using the PLL loop’s
programmable characteristics, including loop filter and charge pump.
Background
PLL bandwidth is the measure of the PLL’s ability to track the input clock and its
associated jitter. The closed-loop gain 3 dB frequency in the PLL determines PLL
bandwidth. Bandwidth is approximately the unity gain point for open loop PLL
response. As Figure 5–32 shows, these points correspond to approximately the same
frequency. Stratix IV PLLs provide three bandwidth settings—low, medium (default),
and high.
Figure 5–32. Open- and Closed-Loop Response Bode Plots
Open-Loop Reponse Bode Plot
Increasing the PLL's
bandwidth in effect pushes
the open loop response out.
0 dB
Gain
Frequency
Closed-Loop Reponse Bode Plot
Gain
Frequency
September 2012
Altera Corporation
Stratix IV Device Handbook
Volume 1
5–38
Chapter 5: Clock Networks and PLLs in Stratix IV Devices
PLLs in Stratix IV Devices
A high-bandwidth PLL provides a fast lock time and tracks jitter on the reference
clock source, passing it through to the PLL output. A low-bandwidth PLL filters out
reference clock jitter but increases lock time. Stratix IV PLLs allow you to control the
bandwidth over a finite range to customize the PLL characteristics for a particular
application. The programmable bandwidth feature in Stratix IV PLLs benefits
applications requiring clock switchover.
A high-bandwidth PLL can benefit a system that must accept a spread-spectrum clock
signal. Stratix IV PLLs can track a spread-spectrum clock by using a high-bandwidth
setting. Using a low-bandwidth setting in this case could cause the PLL to filter out
the jitter on the input clock.
A low-bandwidth PLL can benefit a system using clock switchover. When clock
switchover occurs, the PLL input temporarily stops. A low-bandwidth PLL reacts
more slowly to changes on its input clock and takes longer to drift to a lower
frequency (caused by input stopping) than a high-bandwidth PLL.
Implementation
Traditionally, external components such as the VCO or loop filter control a PLL’s
bandwidth. Most loop filters consist of passive components such as resistors and
capacitors that take up unnecessary board space and increase cost. With Stratix IV
PLLs, all the components are contained within the device to increase performance and
decrease cost.
When you specify the bandwidth setting (low, medium, or high) in the ALTPLL
MegaWizardPlug-in Manager, the Quartus II software automatically sets the
corresponding charge pump and loop filter (Icp, R, C) values to achieve the desired
bandwidth range.
Figure 5–33 shows the loop filter and components that you can set using the
Quartus II software. The components are the loop filter resistor, R, the high frequency
capacitor, Ch, and the charge pump current, IUP or IDN.
Figure 5–33. Loop Filter Programmable Components
IUP
PFD
R
Ch
IDN
C
Stratix IV Device Handbook
Volume 1
September 2012 Altera Corporation
Chapter 5: Clock Networks and PLLs in Stratix IV Devices
PLLs in Stratix IV Devices
5–39
Spread-Spectrum Tracking
Stratix IV devices can accept a spread-spectrum input with typical modulation
frequencies. However, the device cannot automatically detect that the input is a
spread-spectrum signal. Instead, the input signal looks like deterministic jitter at the
input of the PLL. Stratix IV PLLs can track a spread-spectrum input clock as long as it
is within input-jitter tolerance specifications. Stratix IV devices cannot internally
generate spread-spectrum clocks.
Clock Switchover
The clock switchover feature allows the PLL to switch between two reference input
clocks. Use this feature for clock redundancy or for a dual-clock domain application
such as in a system that turns on the redundant clock if the previous clock stops
running. The design can perform clock switchover automatically when the clock is no
longer toggling or based on a user control signal, clkswitch.
The following clock switchover modes are supported in Stratix IV PLLs:
September 2012
■
Automatic switchover—The clock sense circuit monitors the current reference
clock and if it stops toggling, automatically switches to the other inclk0 or inclk1
clock.
■
Manual clock switchover—Clock switchover is controlled using the clkswitch
signal. When the clkswitch signal goes from logic low to logic high, and stays
high for at least three clock cycles, the reference clock to the PLL is switched from
inclk0 to inclk1, or vice-versa.
■
Automatic switchover with manual override—This mode combines automatic
switchover and manual clock switchover. When the clkswitch signal goes high, it
overrides the automatic clock switchover function. As long as the clkswitch signal
is high, further switchover action is blocked.
Altera Corporation
Stratix IV Device Handbook
Volume 1
5–40
Chapter 5: Clock Networks and PLLs in Stratix IV Devices
PLLs in Stratix IV Devices
Stratix IV PLLs support a fully configurable clock switchover capability. Figure 5–34
shows a block diagram of the automatic switchover circuit built into the PLL. When
the current reference clock is not present, the clock sense block automatically switches
to the backup clock for PLL reference. The clock switchover circuit also sends out
three status signals—clkbad[0], clkbad[1], and activeclock—from the PLL to
implement a custom switchover circuit in the logic array. You can select a clock source
as the backup clock by connecting it to the inclk1 port of the PLL in your design.
Figure 5–34. Automatic Clock Switchover Circuit Block Diagram
clkbad[0]
clkbad[1]
activeclock
Switchover
State
Machine
Clock
Sense
clksw
Clock Switch
Control Logic
clkswitch
inclk0
n Counter
inclk1
muxout
PFD
refclk
fbclk
Automatic Clock Switchover
Use the switchover circuitry to automatically switch between inclk0 and inclk1
when the current reference clock to the PLL stops toggling. For example, in
applications that require a redundant clock with the same frequency as the reference
clock, the switchover state machine generates a signal (clksw) that controls the
multiplexer select input, as shown in Figure 5–34. In this case, inclk1 becomes the
reference clock for the PLL. When using automatic switchover mode, you can switch
back and forth between inclk0 and inclk1 any number of times when one of the two
clocks fails and the other clock is available.
When using automatic clock switchover mode, the following requirements must be
satisfied:
■
Both clock inputs must be running
■
The period of the two clock inputs can differ by no more than 100% (2×)
If the current clock input stops toggling while the other clock is also not toggling,
switchover is not initiated and the clkbad[0..1] signals are not valid. Also, if both
clock inputs are not the same frequency, but their period difference is within 100%, the
clock sense block detects when a clock stops toggling, but the PLL may lose lock after
the switchover is completed and needs time to re-lock.
1
Stratix IV Device Handbook
Volume 1
Altera recommends resetting the PLL using the areset signal to maintain the phase
relationships between the PLL input and output clocks when using clock switchover.
September 2012 Altera Corporation
Chapter 5: Clock Networks and PLLs in Stratix IV Devices
PLLs in Stratix IV Devices
5–41
In automatic switchover mode, the clkbad[0] and clkbad[1] signals indicate the
status of the two clock inputs. When they are asserted, the clock sense block has
detected that the corresponding clock input has stopped toggling. These two signals
are not valid if the frequency difference between inclk0 and inclk1 is greater than
20%.
The activeclock signal indicates which of the two clock inputs (inclk0 or inclk1) is
being selected as the reference clock to the PLL. When the frequency difference
between the two clock inputs is more than 20%, the activeclock signal is the only
valid status signal.
Figure 5–35 shows an example waveform of the switchover feature when using
automatic switchover mode. In this example, the inclk0 signal is stuck low. After the
inclk0 signal is stuck at low for approximately two clock cycles, the clock sense
circuitry drives the clkbad[0] signal high. Also, because the reference clock signal is
not toggling, the switchover state machine controls the multiplexer through the
clkswitch signal to switch to the backup clock, inclk1.
Figure 5–35. Automatic Switchover After Loss of Clock Detection
inclk0
inclk1
(1)
muxout
clkbad0
clkbad1
activeclock
Note to Figure 5–35:
(1) Switchover is enabled on the falling edge of inclk0 or inclk1, depending on which clock is available. In this figure,
switchover is enabled on the falling edge of inclk1.
Manual Override
In automatic switchover with manual override mode, you can use the clkswitch
input for user- or system-controlled switch conditions. You can use this mode for
same-frequency switchover, or to switch between inputs of different frequencies. For
example, if inclk0 is 66 MHz and inclk1 is 200 MHz, you must control switchover
using clkswitch because the automatic clock-sense circuitry cannot monitor clock
input (inclk0 and inclk1) frequencies with a frequency difference of more than 100%
(2×). This feature is useful when the clock sources originate from multiple cards on the
backplane, requiring a system-controlled switchover between the frequencies of
operation. You must choose the backup clock frequency and set the m, n, c, and k
counters accordingly so the VCO operates within the recommended operating
frequency range of 600 to 1,600 MHz. The ALTPLL MegaWizard Plug-in Manager
notifies you if a given combination of inclk0 and inclk1 frequencies cannot meet this
requirement.
September 2012
Altera Corporation
Stratix IV Device Handbook
Volume 1
5–42
Chapter 5: Clock Networks and PLLs in Stratix IV Devices
PLLs in Stratix IV Devices
Figure 5–36 shows a clock switchover waveform controlled by clkswitch. In this case,
both clock sources are functional and inclk0 is selected as the reference clock;
clkswitch goes high, which starts the switchover sequence. On the falling edge of
inclk0, the counter’s reference clock, muxout, is gated off to prevent clock glitching.
On the falling edge of inclk1, the reference clock multiplexer switches from inclk0 to
inclk1 as the PLL reference and the activeclock signal changes to indicate which
clock is currently feeding the PLL.
Figure 5–36. Clock Switchover Using the clkswitch (Manual) Control
(1)
inclk0
inclk1
muxout
clkswitch
activeclock
clkbad0
clkbad1
Note to Figure 5–36:
(1) To initiate a manual clock switchover event, both inclk0 and inclk1 must be running when the clkswitch signal
goes high.
In automatic override with manual switchover mode, the activeclock signal mirrors
the clkswitch signal. As both clocks are still functional during the manual switch,
neither clkbad signal goes high. Because the switchover circuit is positive-edge
sensitive, the falling edge of the clkswitch signal does not cause the circuit to switch
back from inclk1 to inclk0. When the clkswitch signal goes high again, the process
repeats. clkswitch and automatic switch only work if the clock being switched to is
available. If the clock is not available, the state machine waits until the clock is
available.
Stratix IV Device Handbook
Volume 1
September 2012 Altera Corporation
Chapter 5: Clock Networks and PLLs in Stratix IV Devices
PLLs in Stratix IV Devices
5–43
Manual Clock Switchover
In manual clock switchover mode, the clkswitch signal controls whether inclk0 or
inclk1 is selected as the input clock to the PLL. By default, inclk0 is selected. A
low-to-high transition on clkswitch and clkswitch being held high for at least three
inclk cycles initiates a clock switchover event. You must bring clkswitch back low
again in order to perform another switchover event in the future. If you do not require
another switchover event in the future, you can leave clkswitch in a logic high state
after the initial switch. Pulsing clkswitch high for at least three inclk cycles performs
another switchover event. If inclk0 and inclk1 are different frequencies and are
always running, the clkswitch minimum high time must be greater than or equal to
three of the slower frequency inclk0 or inclk1 cycles. Figure 5–37 shows a block
diagram of the manual switchover circuit.
Figure 5–37. Manual Clock Switchover Circuitry in Stratix IV PLLs
clkswitch
Clock Switch
Control Logic
inclk0
n Counter
PFD
inclk1
muxout
refclk
fbclk
f For more information about PLL software support in the Quartus II software, refer to
the Phase-Locked Loop (ALTPLL) Megafunction User Guide.
Guidelines
When implementing clock switchover in Stratix IV PLLs, use the following
guidelines:
■
Automatic clock switchover requires that the inclk0 and inclk1 frequencies be
within 100% (2×) of each other. Failing to meet this requirement causes the
clkbad[0] and clkbad[1] signals to not function properly.
■
When using manual clock switchover, the difference between inclk0 and inclk1
can be more than 100% (2×). However, differences in frequency, phase, or both, of
the two clock sources will likely cause the PLL to lose lock. Resetting the PLL
ensures that the correct phase relationships are maintained between the input and
output clocks.
1
■
September 2012
Both inclk0 and inclk1 must be running when the clkswitch signal goes
high to initiate the manual clock switchover event. Failing to meet this
requirement causes the clock switchover to not function properly.
Applications that require a clock switchover feature and a small frequency drift
must use a low-bandwidth PLL. The low-bandwidth PLL reacts more slowly than
a high-bandwidth PLL to reference input clock changes. When switchover
happens, a low-bandwidth PLL propagates the stopping of the clock to the output
more slowly than a high-bandwidth PLL. However, be aware that the
low-bandwidth PLL also increases lock time.
Altera Corporation
Stratix IV Device Handbook
Volume 1
5–44
Chapter 5: Clock Networks and PLLs in Stratix IV Devices
PLLs in Stratix IV Devices
■
After a switchover occurs, there may be a finite resynchronization period for the
PLL to lock onto a new clock. The exact amount of time it takes for the PLL to
re-lock depends on the PLL configuration.
■
The phase relationship between the input clock to the PLL and the output clock
from the PLL is important in your design. Assert areset for at least 10 ns after
performing a clock switchover. Wait for the locked signal to go high and be stable
before re-enabling the output clocks from the PLL.
■
Figure 5–38 shows how the VCO frequency gradually decreases when the current
clock is lost and then increases as the VCO locks on to the backup clock.
Figure 5–38. VCO Switchover Operating Frequency
Primary Clock Stops Running
Switchover Occurs
VCO Tracks Secondary Clock
ΔFvco
■
Disable the system during clock switchover if it is not tolerant of frequency
variations during the PLL resynchronization period. You can use the clkbad[0]
and clkbad[1] status signals to turn off the PFD (PFDENA = 0) so the VCO
maintains its most recent frequency. You can also use the state machine to switch
over to the secondary clock. When the PFD is re-enabled, output clock-enable
signals (clkena) can disable clock outputs during the switchover and
resynchronization period. When the lock indication is stable, the system can
re-enable the output clocks.
PLL Reconfiguration
PLLs use several divide counters and different VCO phase taps to perform frequency
synthesis and phase shifts. In Stratix IV PLLs, you can reconfigure both the counter
settings and phase-shift the PLL output clock in real time. You can also change the
charge pump and loop-filter components, which dynamically affects PLL bandwidth.
You can use these PLL components to update the output-clock frequency and PLL
bandwidth and to phase-shift in real time, without reconfiguring the entire Stratix IV
device.
The ability to reconfigure the PLL in real time is useful in applications that operate at
multiple frequencies. It is also useful in prototyping environments, allowing you to
sweep PLL output frequencies and adjust the output-clock phase dynamically. For
instance, a system generating test patterns is required to generate and transmit
patterns at 75 or 150 MHz, depending on the requirements of the device under test.
Reconfiguring the PLL components in real time allows you to switch between two
such output frequencies within a few microseconds. You can also use this feature to
adjust clock-to-out (tCO) delays in real time by changing the PLL output clock phase
shift. This approach eliminates the need to regenerate a configuration file with the
new PLL settings.
Stratix IV Device Handbook
Volume 1
September 2012 Altera Corporation
Chapter 5: Clock Networks and PLLs in Stratix IV Devices
PLLs in Stratix IV Devices
5–45
PLL Reconfiguration Hardware Implementation
The following PLL components are reconfigurable in real time:
■
Pre-scale counter (n)
■
Feedback counter (m)
■
Post-scale output counters (C0 - C9)
■
Post VCO Divider (K)
■
Dynamically adjust the charge-pump current (Icp) and loop-filter components
(R, C) to facilitate reconfiguration of the PLL bandwidth
Figure 5–39 shows how you can dynamically adjust the PLL counter settings by
shifting their new settings into a serial shift-register chain or scan chain. Serial data is
input to the scan chain using the scandata port. Shift registers are clocked by scanclk.
The maximum scanclk frequency is 100 MHz. Serial data is shifted through the scan
chain as long as the scanclkena signal stays asserted. After the last bit of data is
clocked, asserting the configupdate signal for at least one scanclk clock cycle causes
the PLL configuration bits to be synchronously updated with the data in the scan
registers.
Figure 5–39. PLL Reconfiguration Scan Chain
(1)
from m counter
from n counter
LF/K/CP (3)
PFD
VCO
scandata
scanclkena
configupdate
/Ci (2)
inclk
scandataout
/Ci-1
/C2
/C1
/C0
/m
/n
scandone
scanclk
Notes to Figure 5–39:
(1) Stratix IV left and right PLLs support C0 - C6 counters.
(2) i = 6 or i = 9.
(3) This figure shows the corresponding scan register for the K counter in between the scan registers for the charge pump and loop filter. The K
counter is physically located after the VCO.
1
September 2012
The counter settings are updated synchronously to the clock frequency of the
individual counters. Therefore, all counters are not updated simultaneously.
Altera Corporation
Stratix IV Device Handbook
Volume 1
5–46
Chapter 5: Clock Networks and PLLs in Stratix IV Devices
PLLs in Stratix IV Devices
Table 5–10 lists how these signals can be driven by the PLD logic array or I/O pins.
Table 5–10. Real-Time PLL Reconfiguration Ports
PLL Port Name
Description
Source
Destination
scandata
Serial input data stream to scan
chain.
Logic array or I/O pin
PLL reconfiguration circuit
scanclk
Serial clock input signal. This clock
can be free running.
GCLK, RCLK or I/O pins
PLL reconfiguration circuit
scanclkena
Enables scanclk and allows the
scandata to be loaded in the scan
chain. Active high.
Logic array or I/O pin
PLL reconfiguration circuit
configupdate
Writes the data in the scan chain to
the PLL. Active high.
Logic array or I/O pin
PLL reconfiguration circuit
scandone
Indicates when the PLL has finished
reprogramming. A rising edge
indicates the PLL has begun
reprogramming. A falling edge
indicates the PLL has finished
reprogramming.
PLL reconfiguration circuit
Logic array or I/O pins
scandataout
Used to output the contents of the
scan chain.
PLL reconfiguration circuit
Logic array or I/O pins
To reconfigure the PLL counters, follow these steps:
1. The scanclkena signal is asserted at least one scanclk cycle prior to shifting in the
first bit of scandata (D0).
2. Serial data (scandata) is shifted into the scan chain on the second rising edge of
scanclk.
3. After all 234 bits (top and bottom PLLs) or 180 bits (left and right PLLs) have been
scanned into the scan chain, the scanclkena signal is de-asserted to prevent
inadvertent shifting of bits in the scan chain.
4. The configupdate signal is asserted for one scanclk cycle to update the PLL
counters with the contents of the scan chain.
5. The scandone signal goes high, indicating the PLL is being reconfigured. A falling
edge indicates the PLL counters have been updated with new settings.
6. Reset the PLL using the areset signal if you make any changes to the M, N, or
post-scale output C counters or to the Icp, R, or C settings.
7. You can repeat steps 1-5 to reconfigure the PLL any number of times.
Stratix IV Device Handbook
Volume 1
September 2012 Altera Corporation
Chapter 5: Clock Networks and PLLs in Stratix IV Devices
PLLs in Stratix IV Devices
5–47
Figure 5–40 shows a functional simulation of the PLL reconfiguration feature.
Figure 5–40. PLL Reconfiguration Waveform
(LSB)
D0
SCANDATA
(MSB)
Dn
SCANCLK
SCANCLKENA
D0_old
SCANDATAOUT
Dn_old
Dn
CONFIGUPDATE
SCANDONE
ARESET
1
When you reconfigure the counter clock frequency, you cannot reconfigure the
corresponding counter phase shift settings using the same interface. Instead,
reconfigure the phase shifts in real time using the dynamic phase shift reconfiguration
interface. If you reconfigure the counter frequency, but wish to keep the same
non-zero phase shift setting (for example, 90°) on the clock output, you must
reconfigure the phase shift immediately after reconfiguring the counter clock
frequency.
Post-Scale Counters (C0 to C9)
You can reconfigure the multiply or divide values and duty cycle of post-scale
counters in real time. Each counter has an 8-bit high-time setting and an 8-bit
low-time setting. The duty cycle is the ratio of output high- or low-time to the total
cycle time, which is the sum of the two. Additionally, these counters have two control
bits, rbypass, for bypassing the counter, and rselodd, to select the output clock duty
cycle.
When the rbypass bit is set to 1, it bypasses the counter, resulting in a divide by 1.
When the rbypass bit is set to 0, the high- and low-time counters are added to
compute the effective division of the VCO output frequency. For example, if the
post-scale divide factor is 10, the high- and low-count values can be set to 5 and 5,
respectively, to achieve a 50% - 50% duty cycle. The PLL implements this duty cycle
by transitioning the output clock from high to low on the rising edge of the VCO
output clock. However, a 4 and 6 setting for the high- and low-count values,
respectively, produces an output clock with a 40% - 60% duty cycle.
September 2012
Altera Corporation
Stratix IV Device Handbook
Volume 1
5–48
Chapter 5: Clock Networks and PLLs in Stratix IV Devices
PLLs in Stratix IV Devices
The rselodd bit indicates an odd divide factor for the VCO output frequency along
with a 50% duty cycle. For example, if the post-scale divide factor is 3, the high- and
low-time count values could be set to 2 and 1, respectively, to achieve this division.
This implies a 67% - 33% duty cycle. If you need a 50% - 50% duty cycle, you can set
the rselodd control bit to 1 to achieve this duty cycle despite an odd division factor.
The PLL implements this duty cycle by transitioning the output clock from high to
low on a falling edge of the VCO output clock. When you set rselodd = 1, you
subtract 0.5 cycles from the high time and you add 0.5 cycles to the low time. For
example:
■
High-time count = 2 cycles
■
Low-time count = 1 cycle
■
rselodd = 1 effectively equals:
■
High-time count = 1.5 cycles
■
Low-time count = 1.5 cycles
■
Duty cycle = (1.5/3) % high-time count and (1.5/3) % low-time count
Scan Chain Description
The length of the scan chain varies for different Stratix IV PLLs. The top and bottom
PLLs have ten post-scale counters and a 234-bit scan chain, while the left and right
PLLs have seven post-scale counters and a 180-bit scan chain. Table 5–11 lists the
number of bits for each component of a Stratix IV PLL.
Table 5–11. Top and Bottom PLL Reprogramming Bits (Part 1 of 2)
Number of Bits
Block Name
Total
Counter
C9
(1)
16
2
18
C8
16
2
18
C7
16
2
18
16
2
18
C5
16
2
18
C4
16
2
18
C3
16
2
18
C2
16
2
18
C1
16
2
18
C0
16
2
18
M
16
2
18
N
16
2
18
Charge Pump Current
0
3
3
VCO Post-Scale divider (K)
1
0
1
C6
Stratix IV Device Handbook
Volume 1
(2)
Other
(3)
September 2012 Altera Corporation
Chapter 5: Clock Networks and PLLs in Stratix IV Devices
PLLs in Stratix IV Devices
5–49
Table 5–11. Top and Bottom PLL Reprogramming Bits (Part 2 of 2)
Number of Bits
Block Name
Total
Counter
Loop Filter Capacitor
(4)
Other
(1)
0
2
2
Loop Filter Resistor
0
5
5
Unused CP/LF
0
7
7
Total number of bits
—
—
234
Notes to Table 5–11:
(1) Includes two control bits, rbypass, for bypassing the counter, and rselodd, to select the output clock duty
cycle.
(2) The LSB for the C9 low-count value is the first bit shifted into the scan chain for the top and bottom PLLs.
(3) The LSB for the C6 low-count value is the first bit shifted into the scan chain for the left and right PLLs.
(4) The MSB for the loop filter is the last bit shifted into the scan chain.
Table 5–11 lists the scan chain order of PLL components for the top and bottom PLLs,
which have 10 post-scale counters. The order of bits is the same for the left and right
PLLs, but the reconfiguration bits start with the C6 post-scale counter.
Figure 5–41 shows the scan-chain order of PLL components for the top and bottom
PLLs.
(1)
Figure 5–41. Scan-Chain Order of PLL Components for Top and Bottom PLLs
DATAIN
LF
K
CP
LSB
MSB
C6
C4
C5
C7
C8
N
M
C0
C3
C2
C1
DATAOUT
C9
Note to Figure 5–41:
(1) Left and right PLLs have the same scan-chain order. The post-scale counters end at C6.
Figure 5–42 shows the scan-chain bit-order sequence for post-scale counters in all
Stratix IV PLLs.
Figure 5–42. Scan-Chain Bit-Order Sequence for Post-Scale Counters in Stratix IV PLLs
DATAOUT
September 2012
HB
HB
HB
HB
HB
HB
HB
HB
0
1
2
3
4
5
6
7
LB
LB
LB
LB
LB
LB
LB
LB
0
1
2
3
4
5
6
7
Altera Corporation
rbypass
DATAIN
rselodd
Stratix IV Device Handbook
Volume 1
5–50
Chapter 5: Clock Networks and PLLs in Stratix IV Devices
PLLs in Stratix IV Devices
Charge Pump and Loop Filter
You can reconfigure the charge-pump and loop-filter settings to update the PLL
bandwidth in real time.
Table 5–12 lists the possible settings for charge pump current (Icp) values for
Stratix IV PLLs.
Table 5–12. Charge Pump Current Bit Settings
CP[2]
CP[1]
CP[0]
Decimal Value for Setting
0
0
0
0
0
0
1
1
0
1
1
3
1
1
1
7
Table 5–13 lists the possible settings for loop-filter resistor (R) values for Stratix IV
PLLs.
Table 5–13. Loop-Filter Resistor Bit Settings
LFR[4]
LFR[3]
LFR[2]
LFR[1]
LFR[0]
Decimal Value for Setting
0
0
0
0
0
0
0
0
0
1
1
3
0
0
1
0
0
4
0
1
0
0
0
8
1
0
0
0
0
16
1
0
0
1
1
19
1
0
1
0
0
20
1
1
0
0
0
24
1
1
0
1
1
27
1
1
1
0
0
28
1
1
1
1
0
30
Table 5–14 lists the possible settings for loop-filter capacitor (C) values for Stratix IV
PLLs.
Table 5–14. Loop-Filter Capacitor Bit Settings
Stratix IV Device Handbook
Volume 1
LFC[1]
LFC[0]
Decimal Value for Setting
0
0
0
0
1
1
1
1
3
September 2012 Altera Corporation
Chapter 5: Clock Networks and PLLs in Stratix IV Devices
PLLs in Stratix IV Devices
5–51
Bypassing a PLL
Bypassing a PLL counter results in a multiply (m counter) or a divide (n and C0 to C9
counters) factor of one.
Table 5–15 lists the settings for bypassing the counters in Stratix IV PLLs.
Table 5–15. PLL Counter Settings
PLL Scan Chain Bits [0..8] Settings
LSB
MSB
X
X
X
X
X
X
X
X
1
(1)
X
X
X
X
X
X
X
X
0
(1)
Description
PLL counter bypassed
PLL counter not bypassed because
bit 8 (MSB) is set to 0
Note to Table 5–15:
(1) Counter-bypass bit.
1
To bypass any of the PLL counters, set the bypass bit to 1. The values on the other bits
are ignored. To bypass the VCO post-scale counter (K), set the corresponding bit to 0.
Dynamic Phase-Shifting
The dynamic phase-shifting feature allows the output phases of individual PLL
outputs to be dynamically adjusted relative to each other and to the reference clock,
without having to send serial data through the scan chain of the corresponding PLL.
This feature simplifies the interface and allows you to quickly adjust the clock-to-out
(tCO) delays by changing the output clock phase-shift in real time. This adjustment is
achieved by incrementing or decrementing the VCO phase-tap selection to a given C
counter or to the M counter. The phase is shifted by 1/8 of the VCO frequency at a
time. The output clocks are active during this phase-reconfiguration process.
Table 5–16 lists the control signals that are used for dynamic phase-shifting.
Table 5–16. Dynamic Phase-Shifting Control Signals (Part 1 of 2)
Signal Name
Description
Source
PHASECOUNTERSELECT
[3..0]
Counter select. Four bits decoded to
select either the M or one of the C
counters for phase adjustment. One
address maps to select all C counters.
This signal is registered in the PLL on
the rising edge of SCANCLK.
Logic array or I/O pins
PLL reconfiguration circuit
PHASEUPDOWN
Selects dynamic phase shift direction;
1 = UP; 0 = DOWN. Signal is registered
in the PLL on the rising edge of
SCANCLK.
Logic array or I/O pin
PLL reconfiguration circuit
PHASESTEP
Logic high enables dynamic phase
shifting.
Logic array or I/O pin
PLL reconfiguration circuit
September 2012
Altera Corporation
Destination
Stratix IV Device Handbook
Volume 1
5–52
Chapter 5: Clock Networks and PLLs in Stratix IV Devices
PLLs in Stratix IV Devices
Table 5–16. Dynamic Phase-Shifting Control Signals (Part 2 of 2)
Signal Name
Description
Source
Destination
SCANCLK
Free running clock from the core used
in combination with PHASESTEP to
enable and disable dynamic phase
shifting. Shared with SCANCLK for
dynamic reconfiguration.
GCLK, RCLK or I/O pin
PLL reconfiguration circuit
PHASEDONE
When asserted, this indicates to
core-logic that the phase adjustment is
complete and the PLL is ready to act
on a possible second adjustment
pulse. Asserts based on internal PLL
timing. De-asserts on the rising edge
of SCANCLK.
PLL reconfiguration
circuit
Logic array or I/O pins
Table 5–17 lists the PLL counter selection based on the corresponding
PHASECOUNTERSELECT setting.
Table 5–17. Phase Counter Select Mapping
PHASECOUNTERSELECT[3]
[2]
[1]
[0]
Selects
0
0
0
0
All Output Counters
0
0
0
1
M Counter
0
0
1
0
C0 Counter
0
0
1
1
C1 Counter
0
1
0
0
C2 Counter
0
1
0
1
C3 Counter
0
1
1
0
C4 Counter
0
1
1
1
C5 Counter
1
0
0
0
C6 Counter
1
0
0
1
C7 Counter
1
0
1
0
C8 Counter
1
0
1
1
C9 Counter
To perform one dynamic phase-shift, follow these steps:
1. Set PHASEUPDOWN and PHASECOUNTERSELECT as required.
2. Assert PHASESTEP for at least two SCANCLK cycles. Each PHASESTEP pulse enables
one phase shift.
3. Deassert PHASESTEP after PHASEDONE goes low.
4. Wait for PHASEDONE to go high.
5. Repeat steps 1-4 as many times as required to perform multiple phase-shifts.
The PHASEUPDOWN and PHASECOUNTERSELECT signals are synchronous to SCANCLK and
must meet tsu/th requirements with respect to SCANCLK edges.
Stratix IV Device Handbook
Volume 1
September 2012 Altera Corporation
Chapter 5: Clock Networks and PLLs in Stratix IV Devices
PLLs in Stratix IV Devices
1
5–53
You can repeat dynamic phase-shifting indefinitely. For example, in a design where
the VCO frequency is set to 1000 MHz and the output clock frequency is 100 MHz,
performing 40 dynamic phase shifts (each one yields 125 ps phase shift) results in
shifting the output clock by 180°, which is a phase shift of 5 ns.
The PHASESTEP signal is latched on the negative edge of SCANCLK (a,c) and must remain
asserted for at least two SCANCLK cycles. De-assert PHASESTEP after PHASEDONE goes low.
On the second SCANCLK rising edge (b,d) after PHASESTEP is latched, the values of
PHASEUPDOWN and PHASECOUNTERSELECT are latched and the PLL starts dynamic
phase-shifting for the specified counters and in the indicated direction. PHASEDONE is
de-asserted synchronous to SCANCLK at the second rising edge (b,d) and remains low
until the PLL finishes dynamic phase-shifting. Depending on the VCO and SCANCLK
frequencies, PHASEDONE low time may be greater than or less than one SCANCLK cycle.
You can perform another dynamic phase-shift after the PHASEDONE signal goes from
low to high. Each PHASESTEP pulse enables one phase shift. PHASESTEP pulses must be
at least one SCANCLK cycle apart.
Figure 5–43. Dynamic Phase Shifting Waveform
SCANCLK
PHASESTEP
PHASEUPDOWN
PHASECOUNTERSELECT
PHASEDONE
a
b
c
d
PHASEDONE goes low synchronous with SCANCLK
t CONFIGPHASE
Depending on the VCO and SCANCLK frequencies, PHASEDONE low time may be greater
than or less than one SCANCLK cycle.
After PHASEDONE goes from low to high, you can perform another dynamic phase shift.
PHASESTEP pulses must be at least one SCANCLK cycle apart.
f For information about the ALTPLL_RECONFIG MegaWizard Plug-In Manager, refer
to the Phase-Locked Loops Reconfiguration (ALTPLL_RECONFIG) Megafunction User
Guide.
September 2012
Altera Corporation
Stratix IV Device Handbook
Volume 1
5–54
Chapter 5: Clock Networks and PLLs in Stratix IV Devices
PLLs in Stratix IV Devices
PLL Specifications
f For information about PLL timing specifications, refer to the DC and Switching
Characteristics for Stratix IV Devices chapter.
Document Revision History
Table 5–18 lists the revision history for this chapter.
Table 5–18. Document Revision History (Part 1 of 2)
Date
Version
September 2012
3.4
December 2011
3.3
February 2011
March 2010
November 2009
June 2009
Stratix IV Device Handbook
Volume 1
3.2
3.1
3.0
2.3
Changes
Updated the “Periphery Clock Networks” section.
■
Updated the “Dynamic Phase-Shifting” section.
■
Updated Figure 5–43.
■
Updated the “Clock Input Connections to the PLLs,” “PLL Clock I/O Pins,” “Clock
Feedback Modes,” and “Clock Switchover” sections.
■
Updated Table 5–4 and Table 5–8.
■
Updated Figure 5–26, Figure 5–40, and Figure 5–43.
■
Applied new template.
■
Minor text edits.
■
Updated Table 5–3.
■
Updated notes to Figure 5–2, Figure 5–3, Figure 5–4, and Figure 5–9.
■
Added a note to Table 5–5 and Table 5–6.
■
Added two notes to Table 5–4.
■
Updated Figure 5–43.
■
Updated the “Dynamic Phase-Shifting” section.
■
Minor text edits.
■
Updated Table 5–1 and Table 5–7.
■
Updated “Clock Networks in Stratix IV Devices”, “Periphery Clock Networks”, and
“Cascading PLLs” sections.
■
Added Figure 5–5, Figure 5–6, Figure 5–7, Figure 5–8, and Figure 5–9.
■
Added “Clock Sources Per Region” section.
■
Updated Figure 5–40.
■
Removed EP4SE110, EP4SE290, and EP4SE680 devices.
■
Added EP4S40G2, EP4S100G2, EP4S40G5, EP4S100G3, EP4S100G4, EP4S100G5, and
EP4SE820 devices.
■
Updated Table 5–7.
■
Updated the “PLL Reconfiguration Hardware Implementation” and “Zero-Delay Buffer
Mode” sections.
■
Added introductory sentences to improve search ability.
■
Removed the Conclusion section.
■
Minor text edits.
September 2012 Altera Corporation
Chapter 5: Clock Networks and PLLs in Stratix IV Devices
PLLs in Stratix IV Devices
5–55
Table 5–18. Document Revision History (Part 2 of 2)
Date
Version
April 2009
2.2
March 2009
2.1
November 2008
May 2008
September 2012
2.0
1.0
Altera Corporation
Changes
■
Updated Table 5–1 and Table 5–7.
■
Updated Figure 5–3 and Figure 5–4.
■
Updated the “Periphery Clock Networks” section.
■
Updated Table 5–7.
■
Updated Figure 5–34.
■
Updated “Guidelines” section.
■
Removed “Referenced Documents” section.
■
Updated Table 5–7.
■
Updated Note 1 of Figure 5–10.
■
Updated Figure 5–15.
■
Updated Figure 5–20.
■
Added Figure 5–21.
■
Made minor editorial changes.
Initial release.
Stratix IV Device Handbook
Volume 1
5–56
Stratix IV Device Handbook
Volume 1
Chapter 5: Clock Networks and PLLs in Stratix IV Devices
PLLs in Stratix IV Devices
September 2012 Altera Corporation
Section II. I/O Interfaces
This section provides information on Stratix® IV device I/O features, external
memory interfaces, and high-speed differential interfaces with DPA. This section
includes the following chapters:
■
Chapter 6, I/O Features in Stratix IV Devices
■
Chapter 7, External Memory Interfaces in Stratix IV Devices
■
Chapter 8, High-Speed Differential I/O Interfaces and DPA in Stratix IV Devices
Revision History
Refer to each chapter for its own specific revision history. For information on when
each chapter was updated, refer to the Chapter Revision Dates section, which appears
in the full handbook.
September 2012
Altera Corporation
Stratix IV Device Handbook
Volume 1
II–2
Stratix IV Device Handbook
Volume 1
Section II: I/O Interfaces
September 2012 Altera Corporation
6. I/O Features in Stratix IV Devices
September 2012
SIV51006-3.4
SIV51006-3.4
This chapter describes how Stratix IV devices provide I/O capabilities that allow
you to work in compliance with current and emerging I/O standards and
requirements. With these device features, you can reduce board design interface costs
and increase development flexibility.
Altera Stratix IV FPGAs deliver a breakthrough level of system bandwidth and
power efficiency for high-end applications, allowing you to innovate without
compromise. Stratix IV I/Os are specifically designed for ease-of-use and rapid
system integration while simultaneously providing the high bandwidth required to
maximize internal logic capabilities and produce system-level performance.
Stratix IV device I/O capability far exceeds the I/O bandwidth available from
previous generation FPGAs. Independent modular I/O banks with a common bank
structure for vertical migration lend efficiency and flexibility to the high-speed I/O.
Package and die enhancements with dynamic termination and output control provide
best-in-class signal integrity. Numerous I/O features assist high-speed data transfer
into and out of the device, including:
■
Up to 32 full-duplex clock data recovery (CDR)-based transceivers supporting
data rates between 600 Mbps and 8.5 Gbps
■
Dedicated circuitry to support physical layer functionality for popular serial
protocols, such as PCI Express® (PIPE) (PCIe) Gen1 and Gen2, Gigabit Ethernet
(GbE), Serial RapidIO®, SONET/SDH, XAUI/HiGig, (OIF) CEI-6G,
SD/HD/3G-SDI, Fibre Channel, SFI-5, and Interlaken
■
Complete PCIe protocol solution with embedded PCIe hard IP blocks that
implement PHY-MAC layer, data link layer, and transaction layer functionality
■
Single-ended, non-voltage-referenced, and voltage-referenced I/O standards
■
Low-voltage differential signaling (LVDS), reduced swing differential signaling
(RSDS), mini-LVDS, high-speed transceiver logic (HSTL), and SSTL
■
Single data rate (SDR) and half data rate (HDR—half frequency and twice data
width of SDR) input and output options
■
Up to 132 full duplex 1.6 Gbps true LVDS channels (132 Tx + 132 Rx) on the row
I/O banks
■
Hard dynamic phase alignment (DPA) block with serializer/deserializer
(SERDES)
■
Deskew, read and write leveling, and clock-domain crossing functionality
■
Programmable output current strength
■
Programmable slew rate
© 2012 Altera Corporation. All rights reserved. ALTERA, ARRIA, CYCLONE, HARDCOPY, MAX, MEGACORE, NIOS, QUARTUS and STRATIX words and logos
are trademarks of Altera Corporation and registered in the U.S. Patent and Trademark Office and in other countries. All other words and logos identified as
trademarks or service marks are the property of their respective holders as described at www.altera.com/common/legal.html. Altera warrants performance of its
semiconductor products to current specifications in accordance with Altera's standard warranty, but reserves the right to make changes to any products and
services at any time without notice. Altera assumes no responsibility or liability arising out of the application or use of any information, product, or service
described herein except as expressly agreed to in writing by Altera. Altera customers are advised to obtain the latest version of device specifications before relying
on any published information and before placing orders for products or services.
ISO
9001:2008
Registered
Stratix IV Device Handbook
Volume 1
September 2012
Feedback Subscribe
6–2
Chapter 6: I/O Features in Stratix IV Devices
I/O Standards Support
■
Programmable delay
■
Programmable bus-hold circuit
■
Programmable pull-up resistor
■
Open-drain output
■
Serial, parallel, and dynamic on-chip termination (OCT)
■
Differential OCT
■
Programmable pre-emphasis
■
Programmable equalization
■
Programmable differential output voltage (VOD)
This chapter contains the following sections:
■
“I/O Standards Support”
■
“I/O Banks” on page 6–5
■
“I/O Structure” on page 6–17
■
“On-Chip Termination Support and I/O Termination Schemes” on page 6–24
■
“OCT Calibration” on page 6–32
■
“Termination Schemes for I/O Standards” on page 6–38
■
“Design Considerations” on page 6–46
I/O Standards Support
Stratix IV devices support a wide range of industry I/O standards. Table 6–1 lists the
I/O standards Stratix IV devices support, as well as the typical applications. These
devices support VCCIO voltage levels of 3.0, 2.5, 1.8, 1.5, and 1.2 V.
Table 6–1. I/O Standards and Applications for Stratix IV Devices (Part 1 of 2)
I/O Standard
3.3-V LVTTL/LVCMOS
Stratix IV Device Handbook
Volume 1
(1), (2)
Application
General purpose
2.5-V LVCMOS
General purpose
1.8-V LVCMOS
General purpose
1.5-V LVCMOS
General purpose
1.2-V LVCMOS
General purpose
3.0-V PCI/PCI-X
PC and embedded system
SSTL-2 Class I and II
DDR SDRAM
SSTL-18 Class I and II
DDR2 SDRAM
SSTL-15 Class I and II
DDR3 SDRAM
HSTL-18 Class I and II
QDRII/RLDRAM II
HSTL-15 Class I and II
QDRII/QDRII+/RLDRAM II
HSTL-12 Class I and II
General purpose
Differential SSTL-2 Class I and II
DDR SDRAM
Differential SSTL-18 Class I and II
DDR2 SDRAM
September 2012 Altera Corporation
Chapter 6: I/O Features in Stratix IV Devices
I/O Standards Support
6–3
Table 6–1. I/O Standards and Applications for Stratix IV Devices (Part 2 of 2)
I/O Standard
Application
Differential SSTL-15 Class I and II
DDR3 SDRAM
Differential HSTL-18 Class I and II
Clock interfaces
Differential HSTL-15 Class I and II
Clock interfaces
Differential HSTL-12 Class I and II
Clock interfaces
LVDS
High-speed communications
RSDS
Flat panel display
mini-LVDS
Flat panel display
LVPECL
Video graphics and clock distribution
Notes to Table 6–1:
(1) The 3.3-V LVTTL/LVCMOS standard is supported using VCCIO at 3.0 V.
(2) For more information about the 3.3-V LVTTL/LVCMOS standard supported in Stratix IV devices, refer to “3.3-V I/O
Interface” on page 6–19.
f For more information about transceiver supported I/O standards, refer to the
Transceiver Architecture in Stratix IV Devices chapter.
I/O Standards and Voltage Levels
Stratix IV devices support a wide range of industry I/O standards, including
single-ended, voltage-referenced single-ended, and differential I/O standards.
Table 6–2 lists the supported I/O standards and typical values for input and output
VCCIO, VCCPD, VREF, and board VTT.
Table 6–2. I/O Standards and Voltage Levels for Stratix IV Devices (1) (Part 1 of 3)
VCCIO (V)
I/O Standard
Standard
Support
Input Operation
Column
Row
I/O Banks I/O Banks
3.3-V LVTTL
(3)
3.3-V LVCMOS
2.5-V LVCMOS
Output Operation
Column
I/O Banks
Row
I/O Banks
VTT (V)
VCCPD (V)
VREF (V)
(Board
(Pre-Driver (Input Ref
Termination
Voltage)
Voltage)
Voltage)
JESD8-B
3.0/2.5
3.0/2.5
3.0
3.0
3.0
—
—
JESD8-B
3.0/2.5
3.0/2.5
3.0
3.0
3.0
—
—
JESD8-5
3.0/2.5
3.0/2.5
2.5
2.5
2.5
—
—
1.8-V LVCMOS
JESD8-7
1.8/1.5
1.8/1.5
1.8
1.8
2.5
—
—
1.5-V LVCMOS
JESD8-11
1.8/1.5
1.8/1.5
1.5
1.5
2.5
—
—
1.2-V LVCMOS
JESD8-12
1.2
1.2
1.2
1.2
2.5
—
—
3.0-V PCI
PCI
Rev 2.1
3.0
3.0
3.0
3.0
3.0
—
—
3.0-V PCI-X
PCI-X
Rev 1.0
3.0
3.0
3.0
3.0
3.0
—
—
JESD8-9B
(2)
(2)
2.5
2.5
2.5
1.25
1.25
SSTL-2 Class II
JESD8-9B
(2)
(2)
2.5
2.5
2.5
1.25
1.25
SSTL-18 Class I
JESD8-15
(2)
(2)
1.8
1.8
2.5
0.90
0.90
SSTL-18 Class II
JESD8-15
(2)
(2)
1.8
1.8
2.5
0.90
0.90
SSTL-2 Class I
September 2012
Altera Corporation
Stratix IV Device Handbook
Volume 1
6–4
Chapter 6: I/O Features in Stratix IV Devices
I/O Standards Support
Table 6–2. I/O Standards and Voltage Levels for Stratix IV Devices (1) (Part 2 of 3)
VCCIO (V)
I/O Standard
Standard
Support
Input Operation
Column
Row
I/O Banks I/O Banks
Output Operation
Column
I/O Banks
Row
I/O Banks
VTT (V)
VCCPD (V)
VREF (V)
(Board
(Pre-Driver (Input Ref
Termination
Voltage)
Voltage)
Voltage)
—
(2)
(2)
1.5
1.5
2.5
0.75
0.75
—
(2)
(2)
1.5
—
2.5
0.75
0.75
JESD8-6
(2)
(2)
1.8
1.8
2.5
0.90
0.90
JESD8-6
(2)
(2)
1.8
1.8
2.5
0.90
0.90
JESD8-6
(2)
(2)
1.5
1.5
2.5
0.75
0.75
HSTL-15 Class II
JESD8-6
(2)
(2)
1.5
—
2.5
0.75
0.75
HSTL-12 Class I
JESD8-16A
(2)
(2)
1.2
1.2
2.5
0.6
0.6
HSTL-12 Class II
JESD8-16A
(2)
(2)
1.2
—
2.5
0.6
0.6
Differential SSTL-2
Class I
JESD8-9B
(2)
(2)
2.5
2.5
2.5
—
1.25
Differential SSTL-2
Class II
JESD8-9B
(2)
(2)
2.5
2.5
2.5
—
1.25
Differential
SSTL-18 Class I
JESD8-15
(2)
(2)
1.8
1.8
2.5
—
0.90
Differential
SSTL-18 Class II
JESD8-15
(2)
(2)
1.8
1.8
2.5
—
0.90
Differential
SSTL-15 Class I
—
(2)
(2)
1.5
1.5
2.5
—
0.75
Differential
SSTL-15 Class II
—
(2)
(2)
1.5
—
2.5
—
0.75
Differential
HSTL-18 Class I
JESD8-6
(2)
(2)
1.8
1.8
2.5
—
0.90
Differential
HSTL-18 Class II
JESD8-6
(2)
(2)
1.8
1.8
2.5
—
0.90
Differential
HSTL-15 Class I
JESD8-6
(2)
(2)
1.5
1.5
2.5
—
0.75
Differential
HSTL-15 Class II
JESD8-6
(2)
(2)
1.5
—
2.5
—
0.75
Differential
HSTL-12 Class I
JESD8-16A
(2)
(2)
1.2
1.2
2.5
—
0.60
Differential
HSTL-12 Class II
JESD8-16A
(2)
(2)
1.2
—
2.5
—
0.60
SSTL-15 Class I
SSTL-15 Class II
HSTL-18 Class I
HSTL-18 Class II
HSTL-15 Class I
LVDS
(4), (5), (8)
ANSI/TIA/
EIA-644
(2)
(2)
2.5
2.5
2.5
—
—
RSDS
(6), (7), (8)
—
(2)
(2)
2.5
2.5
2.5
—
—
—
(2)
(2)
2.5
2.5
2.5
—
—
mini-LVDS
(6), (7),
(8)
Stratix IV Device Handbook
Volume 1
September 2012 Altera Corporation
Chapter 6: I/O Features in Stratix IV Devices
I/O Banks
6–5
Table 6–2. I/O Standards and Voltage Levels for Stratix IV Devices (1) (Part 3 of 3)
VCCIO (V)
I/O Standard
Standard
Support
Input Operation
Column
Row
I/O Banks I/O Banks
LVPECL
—
(4)
2.5
Output Operation
Column
I/O Banks
Row
I/O Banks
—
—
VTT (V)
VCCPD (V)
VREF (V)
(Board
(Pre-Driver (Input Ref
Termination
Voltage)
Voltage)
Voltage)
2.5
—
—
Notes to Table 6–2:
(1) VCCPD is either 2.5 or 3.0 V. For VCCIO = 3.0 V, VCCPD = 3.0 V. For VCCIO = 2.5 V or less, VCCPD = 2.5 V.
(2) Single-ended HSTL/SSTL, differential SSTL/HSTL, and LVDS input buffers are powered by VCCPD. Row I/O banks support both true differential
input buffers and true differential output buffers. Column I/O banks support true differential input buffers, but not true differential output buffers.
I/O pins are organized in pairs to support differential standards. Column I/O differential HSTL and SSTL inputs use LVDS differential input buffers
without on-chip RD support.
(3) For more information about the 3.3-V LVTTL/LVCMOS standard supported in Stratix IV devices, refer to “3.3-V I/O Interface” on page 6–19.
(4) Column I/O banks support LVPECL I/O standards for input clock operation. Clock inputs on column I/Os are powered by VCCCLKIN when configured
as differential clock inputs. They are powered by VCCIO when configured as single-ended clock inputs. Differential clock inputs in row I/Os are
powered by VCCPD.
(5) Column and row I/O banks support LVDS outputs using two single-ended output buffers, an external one-resistor (LVDS_E_1R), and a
three-resistor (LVDS_E_3R) network.
(6) Row I/O banks support RSDS and mini-LVDS I/O standards using a true LVDS output buffer without a resistor network.
(7) Column and row I/O banks support RSDS and mini-LVDS I/O standards using two single-ended output buffers with one-resistor (RSDS_E_1R
and mini-LVDS_E_1R) and three-resistor (RSDS_E_3R and mini-LVDS_E_3R) networks.
(8) The emulated differential output standard that supports the tri-state feature includes: LVDS_E_1R, LVDS_E_3R, RSDS_E_1R, RSDS_E_3R,
Mini_LVDS_E_1R, and Mini_LVDS_E_3R. For more information, refer to the I/O Buffer (ALTIOBUF) Megafunction User Guide.
f For more information about the electrical characteristics of each I/O standard, refer to
the DC and Switching Characteristics for Stratix IV Devices chapter.
I/O Banks
Stratix IV devices contain up to 24 I/O banks, as shown in Figure 6–1 and Figure 6–2.
The row I/O banks contain true differential input and output buffers and dedicated
circuitry to support differential standards at speeds up to 1.6 Gbps.
Each I/O bank in Stratix IV devices can support high-performance external memory
interfaces with dedicated circuitry. The I/O pins are organized in pairs to support
differential standards. Each I/O pin pair can support both differential input and
output buffers. The only exceptions are the clk[1,3,8,10], PLL_L[1,4]_clk, and
PLL_R[1,4]_clk pins, which support differential input operations only.
f For information about the number of channels available for the LVDS I/O standard,
refer to the High-Speed Differential I/O Interface and DPA in Stratix IV Devices chapter.
For more information about transceiver-bank-related features, refer to the Transceiver
Architecture in Stratix IV Devices chapter.
September 2012
Altera Corporation
Stratix IV Device Handbook
Volume 1
6–6
Chapter 6: I/O Features in Stratix IV Devices
I/O Banks
Bank 1A
Bank 8A
Bank 8B
(1), (2), (3), (4), (5), (6), (7), (8)
Bank 8C
Bank 7B
Bank 7C
I/O banks 8A, 8B, and 8C support all
single-ended and differential input
and output operations except LVPECL,
which is supported on clk input pins only.
Bank 7A
Bank 6C
Bank 5C
Bank 2C
LVPECL I/O standard for input operation on dedicated
clock input pins.
Bank 2B
SSTL-15 Class II, HSTL-15 Class II, HSTL-12 Class II,
differential SSTL-15 Class II, differential HSTL-15
Class II, differential HSTL-12 Class II standards are
only supported for input operations.
I/O banks 4A, 4B, and 4C support all
single-ended and differential input
and output operations except LVPECL,
which is supported on clk input pins only.
Bank 2A
I/O banks 3A, 3B, and 3C support all
single-ended and differential input
and output operations except LVPECL,
which is supported on clk input pins only.
Bank 3A
Bank 3B
Bank 3C
Bank 4C
Bank 4B
Bank 5B
Bank 1C
Row I/O banks support LVTTL, LVCMOS, 2.5-V, 1.8-V,
1.5-V, 1.2-V, SSTL-2 Class I & II, SSTL-18 Class I & II,
SSTL-15 Class I, HSTL-18 Class I & II, HSTL-15 Class I,
HSTL-12 Class I, LVDS, RSDS, mini-LVDS, differential
SSTL-2 Class I & II, differential SSTL-18 Class I & II,
differential SSTL-15 Class I, differential HSTL-18 Class I &
II, differential HSTL-15 Class I, and differential HSTL-12
Class I standards for input and output operations.
Bank 5A
Bank 1B
Bank 6B
I/O banks 7A, 7B, and 7C support all
single-ended and differential input
and output operations except LVPECL,
which is supported on clk input pins only.
Bank 6A
Figure 6–1. Stratix IV E Devices I/0 Banks
Bank 4A
Notes to Figure 6–1:
(1) Differential HSTL and SSTL outputs are not true differential outputs. They use two single-ended outputs with the second output programmed as
inverted.
(2) Column I/O differential HSTL and SSTL inputs use LVDS differential input buffers without differential OCT support.
(3) Column I/O supports LVDS outputs using single-ended buffers and external resistor networks.
(4) Column I/O supports PCI/PCI-X with on-chip clamp diode. Row I/O supports PCI/PCI-X with external clamp diode.
(5) Clock inputs on column I/Os are powered by VCCCLKIN when configured as differential clock inputs. They are powered by VCCIO when configured as
single-ended clock inputs. All outputs use the corresponding bank VCCIO.
(6) Row I/O supports the true LVDS output buffer.
(7) Column and row I/O banks support LVPECL standards for input clock operation.
(8) Figure 6–1 is a top view of the silicon die that corresponds to a reverse view for flip chip packages. It is a graphical representation only.
Stratix IV Device Handbook
Volume 1
September 2012 Altera Corporation
Chapter 6: I/O Features in Stratix IV Devices
I/O Banks
6–7
Bank 3A
Bank 3B
Bank 3C
Bank 4C
Bank 4B
Transceiver Bank
GXBR2
Bank 6C
Bank5C
I/O banks 4A, 4B & 4C support all
single-ended and differential input
and output operation.
I/O banks 3A, 3B & 3C support all
single-ended and differential input
and output operation.
Bank 5A
Bank 2B
Bank 5B
Bank 2C
Bank 1C
Row I/O banks support LVTTL, LVCMOS, 2.5-V, 1.8V, 1.5-V, 1.2-V, SSTL-2 Class I & II, SSTL-18 Class I
& II, SSTL-15 Class I, HSTL-18 Class I & II, HSTL-15
Class I, HSTL-12 Class I, LVDS, RSDS, mini-LVDS,
differential SSTL-2 Class I & II, differential SSTL-18
Class I & II, differential SSTL-15 Class I, differential
HSTL-18 Class I & II, differential HSTL-15 Class I and
differential HSTL-12 Class I standards for input and
output operation.
SSTL-15 class II, HSTL-15 Class II, HSTL-12 Class II,
differential SSTL-15 Class II, differential HSTL-15
Class II, differential HSTL-12 Class II standards are
only supported for input operations
Transceiver Bank
GXBR1
Bank 6B
Bank 1B
I/O banks 7A, 7B & 7C support all
single-ended and differential input
and output operation.
Transceiver Bank
GXBR3
Bank 7A
Transceiver Bank
GXBR0
Bank 1A
Bank 7B
Bank 7C
I/O banks 8A, 8B & 8C support all
single-ended and differential input
and output operation.
Bank 2A
Transceiver Bank
GXBL3
Transceiver Bank
GXBL2
Transceiver Bank
GXBL1
Transceiver Bank
GXBL0
Bank 8C
Bank 8B
Bank 8A
(1), (2), (3), (4), (5), (6), (7), (8), (9)
Bank 6A
Figure 6–2. Stratix IV GX Devices I/O Banks
Bank 4A
Notes to Figure 6–2:
(1) Differential HSTL and SSTL outputs are not true differential outputs. They use two single-ended outputs with the second output programmed as
inverted.
(2) Column I/O differential HSTL and SSTL inputs use LVDS differential input buffers without differential OCT support.
(3) Column I/O supports LVDS outputs using single-ended buffers and external resistor networks.
(4) Column I/O supports PCI/PCI-X with an on-chip clamp diode. Row I/O supports PCI/PCI-X with an external clamp diode.
(5) Clock inputs on column I/Os are powered by VCCCLKIN when configured as differential clock inputs. They are powered by VCCIO when configured as
single-ended clock inputs. All outputs use the corresponding bank VCCIO.
(6) Row I/O supports the true LVDS output buffer.
(7) Column and row I/O banks support LVPECL standards for input clock operation.
(8) Figure 6–2 is a top view of the silicon die that corresponds to a reverse view for flip chip packages. It is a graphical representation only.
(9) Stratix IV devices do not support the PCI clamp diode when VCCIO is 1.2 V, 1.5 V, or 1.8 V.
September 2012
Altera Corporation
Stratix IV Device Handbook
Volume 1
6–8
Chapter 6: I/O Features in Stratix IV Devices
I/O Banks
Modular I/O Banks
The I/O pins in Stratix IV devices are arranged in groups called modular I/O banks.
Depending on device densities, the number of Stratix IV device I/O banks range from
16 to 24. The number of I/O pins on each bank is 24, 32, 36, 40, or 48. Figure 6–4
through Figure 6–16 show the number of I/O pins available in each I/O bank.
In Stratix IV devices, the maximum number of I/O banks per side is either four or six,
depending on the device density. When migrating between devices with a different
number of I/O banks per side, it is the middle or “B” bank that is removed or
inserted. For example, when moving from a 24-bank device to a 16-bank device, the
banks that are dropped are “B” banks, namely: 1B, 2B, 3B, 4B, 5B, 6B, 7B, and 8B.
Similarly, when moving from a 16-bank device to a 24-bank device, the banks that are
added are the same “B” banks.
After migration from a smaller device to a larger device, the bank size increases or
remains the same, but never decreases. For example, the number of I/O pins to a bank
may increase from 24 to 26, 32, 36, 40, 42, or 48, but will never decrease. This is shown
in Figure 6–3.
Figure 6–3. Bank Migration Path with Increasing Device Size
24
Stratix IV Device Handbook
Volume 1
26
32
36
40
42
48
September 2012 Altera Corporation
Chapter 6: I/O Features in Stratix IV Devices
I/O Banks
6–9
Figure 6–4 through Figure 6–16 show the number of I/O pins and packaging
information for different sets of available devices. They show the top view of the
silicon die that corresponds to a reverse view for flip chip packages. They are
graphical representations only.
1
For Figure 6–4 through Figure 6–16, the pin count includes all general purpose I/Os,
dedicated clock pins, and dual purpose configuration pins. Transceiver pins and
dedicated configuration pins are not included in the pin count.
40
Bank 2C
Bank 5C
26
Bank 2A
Bank 5A
32
40
Bank 4A
Bank
Name
Number
of I/Os
40
32
Bank 7A
26
Bank 4C
26
24
Bank 6C
EP4SE230
EP4SE360
24
Bank 1C
Bank 7C
32
Bank 3C
26
24
Bank 6A
24
Bank 1A
Bank 3A
32
Bank 8C
Bank
Name
40
Number
of I/Os
Bank 8A
Figure 6–4. Number of I/Os in Each Bank in EP4SE230 and EP4SE360 Devices in the 780-Pin FineLine BGA Package
24
32
32
24
40
Bank 8B
Bank 8C
Bank 7C
Bank 7B
Altera Corporation
Bank 7A
Bank 4A
40
40
September 2012
Bank 4B
Bank 2A
24
48
Bank 4C
Bank 2C
32
42
EP4SE360
EP4SE530
EP4SE820
Bank 3C
Bank 1C
32
42
Bank 3B
Bank 1A
Bank 3A
48
24
Bank
Name
40
Number
of I/Os
Bank 8A
Figure 6–5. Number of I/Os in Each Bank in EP4SE360, EP4SE530, and EP4SE820 Devices in the 1152-Pin FineLine BGA
Package
Bank 6A
48
Bank 6C
42
Bank 5C
42
Bank 5A
48
Bank
Name
Number
of I/Os
Stratix IV Device Handbook
Volume 1
6–10
Chapter 6: I/O Features in Stratix IV Devices
I/O Banks
48
32
32
48
48
Bank 8B
Bank 8C
Bank 7C
Bank 7B
Bank 7A
Bank
Name
48
Number
of I/Os
Bank 8A
Figure 6–6. Number of I/Os in Each Bank in EP4SE530 and EP4SE820 Devices in the 1517-Pin FineLine BGA Package
50
Bank 1A
Bank 6A
50
24
Bank 1B
Bank 6B
24
42
Bank 1C
Bank 6C
42
42
Bank 2C
Bank 5C
42
24
Bank 2B
Bank 5B
24
50
Bank 2A
Bank 5A
50
48 Bank 4A
48 Bank 4B
32 Bank 4C
32 Bank 3C
48 Bank 3B
48 Bank 3A
EP4SE530
EP4SE820
Bank
Name
Number
of I/Os
48
48
48
48
48
Bank 8B
Bank 8C
Bank 7C
Bank 7B
Bank 7A
Bank
Name
48
Number
of I/Os
Bank 8A
Figure 6–7. Number of I/Os in Each Bank in EP4SE530 and EP4SE820 Devices in the 1760-Pin Fineline BGA Package
50
Bank 1A
Bank 6A
50
36
Bank 1B
Bank 6B
36
50
Bank 1C
Bank 6C
50
50
Bank 2C
Bank 5C
50
36
Bank 2B
Bank 5B
36
50
Bank 2A
Bank 5A
50
Stratix IV Device Handbook
Volume 1
48 Bank 4A
48 Bank 4B
48 Bank 4C
48 Bank 3C
48 Bank 3B
48 Bank 3A
EP4SE530
EP4SE820
Bank
Name
Number
of I/Os
September 2012 Altera Corporation
Chapter 6: I/O Features in Stratix IV Devices
I/O Banks
6–11
40
Bank
GXBR1
Bank 2A
24
24
40
Bank 7C
Bank 7A
Number of
Transceiver
Channels
4
Bank
GXBR0
32
Bank 4A
Bank 2C
40
26
EP4SGX70
EP4SGX110
EP4SGX180
EP4SGX230
Bank 4C
Bank 1C
24
26
Bank 3C
Bank 1A
Bank 3A
32
24
Bank
Name
Bank 8A
40
Number
of I/Os
Bank 8C
Figure 6–8. Number of I/Os in Each Bank in EP4SGX70, EP4SGX110, EP4SGX180, and EP4SGX230 Devices in the 780-Pin
FineLine BGA Package
4
Bank
Name
Number
of I/Os
Altera Corporation
32
40
Bank 7C
Bank 7A
Bank 4C
Bank 4A
32
40
Bank
GXBL0
Bank 3C
4
Number of
Transceiver
Channels
Bank
GXBR1
4
Bank
GXBR0
4
EP4SGX290
EP4SGX360
32
Bank
GXBL1
Bank 3A
4
Number of
Transceiver
Channels
September 2012
32
Bank
1C
40
1
Bank 8C
Bank
Name
40
Number
of I/Os
Bank 8A
Figure 6–9. Number of I/Os in Each Bank in EP4SGX290 and EP4SGX360 Devices in the 780-Pin FineLine BGA Package
Bank
Name
Number
of I/Os
Stratix IV Device Handbook
Volume 1
6–12
Chapter 6: I/O Features in Stratix IV Devices
I/O Banks
Bank
GXBL0
24
24
40
Bank 7C
Bank 7A
40
*Number of
Transceiver
Channels
Bank 4A
4*
Bank 6A
32
Bank 6C
26
Bank
GXBR1
4*
Bank
GXBR0
4*
Bank
Name
Number
of I/Os
40
Bank
GXBL1
Bank 4C
4*
EP4SGX70
EP4SGX110
24
Bank 1C
Bank 3C
26
24
Bank 1A
Bank 3A
32
Bank 8C
Bank
Name
40
Number
of I/Os
Bank 8A
Figure 6–10. Number of I/Os in Each Bank in EP4SGX70 and EP4SGX110 Devices in the 1152-Pin FineLine BGA Package
32
32
24
40
Bank 8C
Bank 7C
Bank 7B
Bank 7A
24
Bank 8B
Bank 4B
Bank 4A
40
40
24
Bank
GXBL0
Bank 4C
4 (2)
32
4 (2)
Bank
GXBL1
Bank 3C
Bank 1C
32
42
EP4SGX180
EP4SGX230
EP4SGX290
EP4SGX360
EP4SGX530
Bank 3B
Bank 1A
Bank 3A
48
24
Bank
Name
40
Number
of I/Os
Bank 8A
Figure 6–11. Number of I/Os in Each Bank in EP4SGX180, EP4SGX230, EP4SGX290, EP4SGX360, and EP4SGX530 Devices
in the 1152-Pin FineLine BGA Package (1), (2)
Bank 6A
48
Bank 6C
42
Bank
GXBR1
4 (2)
Bank
GXBR0
4 (2)
Bank
Name
Number
of I/Os
Notes to Figure 6–11:
(1) Except for the EP4SGX530 device, all listed devices have two variants in the F1152 package option—one with no PMA-only transceiver channels
and the other with two PMA-only transceiver channels for each transceiver bank. The EP4SGX530 device is only offered with two PMA-only
transceiver channels for each transceiver bank in the F1152 package option.
(2) There are two additional PMA-only transceiver channels in each transceiver bank for devices with the PMA-only transceiver package option.
Stratix IV Device Handbook
Volume 1
September 2012 Altera Corporation
Chapter 6: I/O Features in Stratix IV Devices
I/O Banks
6–13
24
32
32
24
40
Bank 8B
Bank 8C
Bank 7C
Bank 7B
Bank 7A
Bank
Name
40
Number
of I/Os
Bank 8A
Figure 6–12. Number of I/Os in Each Bank in EP4SGX180, EP4SGX230, EP4SGX290, EP4SGX360, and EP4SGX530 Devices
in the 1517-Pin FineLine BGA Package (1)
48
Bank 1A
Bank 6A
48
42
Bank 1C
Bank 6C
42
42
Bank 2C
Bank 5C
42
48
Bank 2A
Bank 5A
48
EP4SGX180
EP4SGX230
EP4SGX290
EP4SGX360
EP4SGX530
Bank
GXBR1
4 (1)
4 (1)
Bank
GXBL0
Bank
GXBR0
4 (1)
40
24
32
32
24
40
Bank 4A
Bank
GXBL1
Bank 4B
4 (1)
Bank 4C
4 (1)
Bank 3C
Bank
GXBR2
Bank 3B
Bank
GXBL2
Bank 3A
4 (1)
Bank
Name
Number
of I/Os
Note to Figure 6–12:
(1) There are two additional PMA-only transceiver channels in each transceiver bank.
September 2012
Altera Corporation
Stratix IV Device Handbook
Volume 1
6–14
Chapter 6: I/O Features in Stratix IV Devices
I/O Banks
48
32
32
48
48
Bank 8B
Bank 8C
Bank 7C
Bank 7B
Bank 7A
Bank
Name
48
Number
of I/Os
Bank 8A
Figure 6–13. Number of I/Os in Each Bank in EP4SGX290, EP4SGX360, and EP4SGX530 Devices in the 1932-Pin FineLine
BGA Package (1)
50
Bank 1A
Bank 6A
50
42
Bank 1C
Bank 6C
42
42
Bank 2C
Bank 5C
42
20
Bank 2B
Bank 5B
20
50
Bank 2A
Bank 5A
50
4 (1)
Bank
GXBL3
Bank
GXBR3
4 (1)
4 (1)
Bank
GXBL2
Bank
GXBR2
4 (1)
4 (1)
Bank
GXBL1
4 (1)
4 (1)
Bank
GXBL0
Bank
GXBR1
Bank
GXBR0
48 Bank 4A
48 Bank 4B
32 Bank 4C
32 Bank 3C
48 Bank 3B
48 Bank 3A
EP4SGX530
EP4SGX290
EP4SGX360
4 (1)
Bank
Name
Number
of I/Os
Note to Figure 6–13:
(1) There are two additional PMA-only transceiver channels in each transceiver bank.
Stratix IV Device Handbook
Volume 1
September 2012 Altera Corporation
Chapter 6: I/O Features in Stratix IV Devices
I/O Banks
6–15
48
32
32
48
48
Bank 8B
Bank 8C
Bank 7C
Bank 7B
Bank 7A
Bank
Name
48
Number
of I/Os
Bank 8A
Figure 6–14. Number of I/Os in Each Bank in EP4SGX290, EP4SGX360, and EP4SGX530 Devices in the 1760-Pin FineLine
BGA Package (1)
50
Bank 1A
Bank 6A
50
42
Bank 1C
Bank 6C
42
42
Bank 2C
Bank 5C
42
Bank 5A
50
50
EP4SGX290
EP4SGX360
EP4SGX530
Bank 2A
Bank
GXBR1
4 (1)
4 (1)
Bank
GXBL0
Bank
GXBR0
4 (1)
48
48
32
32
48
48
Bank 4A
Bank
GXBL1
Bank 4B
4 (1)
Bank 4C
4 (1)
Bank 3C
Bank
GXBR2
Bank 3B
Bank
GXBL2
Bank 3A
4 (1)
Bank
Name
Number
of I/Os
Note to Figure 6–14:
(1) There are two additional PMA-only transceiver channels in each transceiver bank.
September 2012
Altera Corporation
Stratix IV Device Handbook
Volume 1
6–16
Chapter 6: I/O Features in Stratix IV Devices
I/O Banks
1
The information in Figure 6–15 and Figure 6–16 applies to Stratix IV GX and GT
devices.
48
32
32
48
48
Bank 8B
Bank 8C
Bank 7C
Bank 7B
Bank 7A
Bank
Name
48
Number
of I/Os
Bank 8A
Figure 6–15. Number of I/Os in Each Bank in EP4S100G3, EP4S100G4, and EP4S100G5 Devices in the 1932-Pin FineLine
BGA Package (1)
40
Bank 1A
Bank 6A
38
21
Bank 1C
Bank 6C
22
21
Bank 2C
Bank 5C
19
13
Bank 2B
Bank 5B
12
41
Bank 2A
Bank 5A
42
4 (1)
Bank
GXBL2
Bank
GXBR2
4 (1)
4 (1)
Bank
GXBL1
Bank
GXBR1
4 (1)
4 (1)
Bank
GXBL0
Bank
GXBR0
4 (1)
Bank 4A
48
Bank 4B
48
Bank 4C
32
Bank 3C
32
Bank 3B
48
48
Bank 3A
EP4S100G3
EP4S100G4
EP4S100G5
Bank
Name
Number
of I/Os
Note to Figure 6–15:
(1) There are two additional PMA-only transceiver channels in each transceiver bank.
Stratix IV Device Handbook
Volume 1
September 2012 Altera Corporation
Chapter 6: I/O Features in Stratix IV Devices
I/O Structure
6–17
24
32
32
24
40
Bank 8B
Bank 8C
Bank 7C
Bank 7B
Bank 7A
Bank
Name
40
Number
of I/Os
Bank 8A
Figure 6–16. Number of I/Os in Each Bank in EP4S40G2, EP4S40G5, EP4S100G2, and EP4S100G5 Devices in the 1517-Pin
FineLine BGA Package (1)
43
Bank 1A
Bank 6A
44
22
Bank 1C
Bank 6C
23
23
Bank 2C
Bank 5C
23
46
Bank 2A
Bank 5A
46
EP4S40G2
EP4S40G5
EP4S100G2
EP4S100G5
Bank
GXBR1
4 (1)
4 (1)
Bank
GXBL0
Bank
GXBR0
4 (1)
40
24
32
32
24
40
Bank 4A
Bank
GXBL1
Bank 4B
4 (1)
Bank 4C
4 (1)
Bank 3C
Bank
GXBR2
Bank 3B
Bank
GXBL2
Bank 3A
4 (1)
Bank
Name
Number
of I/Os
Note to Figure 6–16:
(1) There are two additional PMA-only transceiver channels in each transceiver bank.
I/O Structure
The I/O element (IOE) in Stratix IV devices contain a bidirectional I/O buffer and I/O
registers to support a complete embedded bidirectional single data rate or DDR
transfer. The IOEs are located in I/O blocks around the periphery of the Stratix IV
device. There are up to four IOEs per row I/O block and four IOEs per column I/O
block. The row IOEs drive row, column, or direct link interconnects. The column IOEs
drive column interconnects.
The Stratix IV bidirectional IOE also supports the following features:
September 2012
■
Programmable input delay
■
Programmable output-current strength
■
Programmable slew rate
■
Programmable output delay
■
Programmable bus-hold
■
Programmable pull-up resistor
■
Open-drain output
■
On-chip series termination with calibration
Altera Corporation
Stratix IV Device Handbook
Volume 1
6–18
Chapter 6: I/O Features in Stratix IV Devices
I/O Structure
■
On-chip series termination without calibration
■
On-chip parallel termination with calibration
■
On-chip differential termination
■
PCI clamping diode
I/O registers are composed of the input path for handling data from the pin to the
core, the output path for handling data from the core to the pin, and the output-enable
(OE) path for handling the OE signal to the output buffer. These registers allow faster
source-synchronous register-to-register transfers and resynchronization. The input
path consists of the DDR input registers, alignment and synchronization registers,
and HDR. You can bypass each block of the input path.
The output and OE paths are divided into output or OE registers, alignment registers,
and HDR blocks. You can bypass each block of the output and OE paths.
Figure 6–17 shows the Stratix IV IOE structure.
Figure 6–17. IOE Structure in Stratix IV Devices (1),
(2), (3), (4)
Firm Core
DQS Logic Block
OE Register
D
OE
from
Core
2
Half Data
Rate Block
D6_OCT
D5_OCT
PRN
Q
Dynamic OCT Control (2)
Alignment
Registers
OE Register
D
VCCIO
D5, D6
Delay
PRN
Q
VCCIO
PCI Clamp
Programmable
Pull-Up Resistor
Programmable
Current
Strength and
Slew Rate
Control
Output Register
Write
Data
from
Core
Half Data
Rate Block
4
Alignment
Registers
PRN
D
Q
From OCT
Calibration
Block
Output Buffer
D5, D6
Delay
Output Register
D
Open Drain
PRN
Q
D2 Delay
Input Buffer
D3_0
Delay
clkout
To
Core
D3_1
Delay
To
Core
Read
Data
to
Core
4
Half Data
Rate Block
Alignment and
Synchronization
Registers
D1
Delay
Bus-Hold
Circuit
Input Register
PRN
D
Q
Input Register
Input Register
PRN
D
DQS
CQn
On-Chip
Termination
PRN
Q
D
Q
D4 Delay
clkin
Notes to Figure 6–17:
(1) The following features are not supported by true differential standards: open drain or tri-state output,; programmable current strength and slew
rate control; PCI Clamp; programmable pull-up resistor; bus-hold circuit.
(2) The D3_0 and D3_1 delays have the same available settings in the Quartus® II software
(3) One dynamic OCT control is available per DQ/DQS group.
(4) Column I/O supports PCI/PCI-X with an on-chip clamp diode. Row I/O supports PCI/PCI-X with an external clamp diode.
Stratix IV Device Handbook
Volume 1
September 2012 Altera Corporation
Chapter 6: I/O Features in Stratix IV Devices
I/O Structure
6–19
f For more information about I/O registers and how they are used for memory
applications, refer to the External Memory Interfaces in Stratix IV Devices chapter.
3.3-V I/O Interface
Stratix IV I/O buffers support 3.3-V I/O standards. You can use them as transmitters
or receivers in your system. The output high voltage (VOH), output low voltage (VOL),
input high voltage (VIH), and input low voltage (VIL) levels meet the 3.3-V I/O
standards specifications defined by EIA/JEDEC Standard JESD8-B with margin when
the Stratix IV VCCIO voltage is powered by 3.0 V.
To ensure device reliability and proper operation, when interfacing with a 3.3-V I/O
system using Stratix IV devices, ensure that you do not violate the absolute maximum
ratings of the devices. Altera recommends performing IBIS simulation to determine
that the overshoot and undershoot voltages are within the guidelines.
When using the Stratix IV device as a transmitter, you can use slow slew rate and
series termination to limit overshoot and undershoot at the I/O pins, but they are not
required. Transmission line effects that cause large voltage deviations at the receiver
are associated with an impedance mismatch between the driver and the transmission
lines. By matching the impedance of the driver to the characteristic impedance of the
transmission line, you can significantly reduce overshoot voltage. You can use a series
termination resistor placed physically close to the driver to match the total driver
impedance to the transmission line impedance. Stratix IV devices support series OCT
for all LVTTL and LVCMOS I/O standards in all I/O banks.
When using the Stratix IV device as a receiver, you can use a clamping diode (on-chip
or off-chip) to limit overshoot, though this is not required. Stratix IV devices provide
an optional on-chip PCI-clamping diode for column I/O pins. You can use this diode
to protect the I/O pins against overshoot voltage.
The 3.3-V I/O standard is supported using bank supply voltage (VCCIO) at 3.0 V. In
this method, the clamping diode (on-chip or off-chip), when enabled, can sufficiently
clamp overshoot voltage to within the DC and AC input voltage specifications. The
clamped voltage can be expressed as the sum of the supply voltage (VCCIO) and the
diode forward voltage.
f For more information about the absolute maximum rating and maximum allowed
overshoot during transitions, refer to the DC and Switching Characteristics for Stratix IV
Devices chapter.
External Memory Interfaces
In addition to the I/O registers in each IOE, Stratix IV devices also have dedicated
registers and phase-shift circuitry on all I/O banks for interfacing with external
memory interfaces.
f For more information about external memory interfaces, refer to the External Memory
Interfaces in Stratix IV Devices chapter.
September 2012
Altera Corporation
Stratix IV Device Handbook
Volume 1
6–20
Chapter 6: I/O Features in Stratix IV Devices
I/O Structure
High-Speed Differential I/O with DPA Support
Stratix IV devices have the following dedicated circuitry for high-speed differential
I/O support:
■
Differential I/O buffer
■
Transmitter serializer
■
Receiver deserializer
■
Data realignment
■
Dynamic phase aligner (DPA)
■
Synchronizer (FIFO buffer)
■
Phase-locked loops (PLLs)
f For more information about DPA support, refer to the High-Speed Differential I/O
Interfaces and DPA in Stratix IV Devices chapter.
Programmable Current Strength
The output buffer for each Stratix IV device I/O pin has a programmable current
strength control for certain I/O standards. Use programmable current strength to
mitigate the effects of high signal attenuation due to a long transmission line or a
legacy backplane. The LVTTL, LVCMOS, SSTL, and HSTL standards have several
levels of current strength that you can control. Table 6–3 lists the programmable
current strength for Stratix IV devices.
Table 6–3. Programmable Current Strength (Part 1 of 2) (1),
I/O Standard
3.3-V LVTTL
Stratix IV Device Handbook
Volume 1
(2)
IOH / IOL Current Strength
Setting (mA) for
Column I/O Pins
IOH / IOL Current Strength
Setting (mA) for
Row I/O Pins
16, 12, 8, 4
12, 8, 4
3.3-V LVCMOS
16, 12, 8, 4
8, 4
2.5-V LVCMOS
16, 12, 8, 4
12, 8, 4
1.8-V LVCMOS
12, 10, 8, 6, 4, 2
8, 6, 4, 2
1.5-V LVCMOS
12, 10, 8, 6, 4, 2
8, 6, 4, 2
1.2-V LVCMOS
8, 6, 4, 2
4, 2
SSTL-2 Class I
12, 10, 8
12, 8
SSTL-2 Class II
16
16
SSTL-18 Class I
12, 10, 8, 6, 4
12, 10, 8, 6, 4
SSTL-18 Class II
16, 8
16, 8
SSTL-15 Class I
12, 10, 8, 6, 4
8, 6, 4
SSTL-15 Class II
16, 8
—
HSTL-18 Class I
12, 10, 8, 6, 4
12, 10, 8, 6, 4
HSTL-18 Class II
16
16
HSTL-15 Class I
12, 10, 8, 6, 4
8, 6, 4
HSTL-15 Class II
16
—
September 2012 Altera Corporation
Chapter 6: I/O Features in Stratix IV Devices
I/O Structure
6–21
Table 6–3. Programmable Current Strength (Part 2 of 2) (1),
(2)
IOH / IOL Current Strength
Setting (mA) for
Column I/O Pins
IOH / IOL Current Strength
Setting (mA) for
Row I/O Pins
HSTL-12 Class I
12, 10, 8, 6, 4
8, 6, 4
HSTL-12 Class II
16
—
I/O Standard
Notes to Table 6–3:
(1) The default setting in the Quartus II software is 50-OCT RS without calibration for all non-voltage reference and
HSTL and SSTL Class I I/O standards. The default setting is 25-OCT RS without calibration for HSTL and SSTL
Class II I/O standards.
(2) The 3.3-V LVTTL and 3.3-V LVCMOS are supported using VCCIO and VCCPD at 3.0 V.
1
Altera recommends performing IBIS or SPICE simulations to determine the best
current strength setting for your specific application.
Programmable Slew Rate Control
The output buffer for each Stratix IV device regular- and dual-function I/O pin has a
programmable output slew-rate control that you can configure for low-noise or
high-speed performance. A faster slew rate provides high-speed transitions for
high-performance systems. A slower slew rate can help reduce system noise, but adds
a nominal delay to the rising and falling edges. Each I/O pin has an individual
slew-rate control, allowing you to specify the slew rate on a pin-by-pin basis.
1
You cannot use the programmable slew rate feature when using OCT.
The Quartus II software allows four settings for programmable slew rate control—0,
1, 2, and 3—where 0 is slow slew rate and 3 is fast slew rate. Figure 6–4 lists the
default slew rate settings from the Quartus II software.
Table 6–4. Default Slew Rate Settings
I/O Standard
Slew Rate Option
Default Slew Rate
1.2-V, 1.5-V, 1.8-V, 2.5-V LVCMOS, and 3.3-V LVTTL/LVCMOS
0, 1, 2, 3
3
SSTL-2, SSTL-18, SSTL-15, HSTL-18, HSTL-15, and HSTL-12
0, 1, 2, 3
3
3.0-V PCI/PCI-X
0, 1, 2, 3
3
LVDS_E_1R, mini-LVDS_E_1R, and RSDS_E_1R
0, 1, 2, 3
3
LVDS_E_3R, mini-LVDS_E_3R, and RSDS_E_3R
0, 1, 2, 3
3
You can use faster slew rates to improve the available timing margin in
memory-interface applications or when the output pin has high-capacitive loading.
1
September 2012
Altera recommends performing IBIS or SPICE simulations to determine the best slew
rate setting for your specific application.
Altera Corporation
Stratix IV Device Handbook
Volume 1
6–22
Chapter 6: I/O Features in Stratix IV Devices
I/O Structure
Programmable I/O Delay
The following sections describe programmable IOE delay and programmable output
buffer delay.
Programmable IOE Delay
The Stratix IV device IOE includes programmable delays, shown in Figure 6–17 on
page 6–18, that you can activate to ensure zero hold times, minimize setup times, or
increase clock-to-output times. Each pin can have a different input delay from
pin-to-input register or a delay from output register-to-output pin values to ensure
that the bus has the same delay going into or out of the device. This feature helps read
and time margins because it minimizes the uncertainties between signals in the bus.
f For more information about programmable IOE delay specifications, refer to the
High-Speed Differential I/O Interfaces and DPA in Stratix IV Devices chapter.
Programmable Output Buffer Delay
Stratix IV devices support delay chains built inside the single-ended output buffer, as
shown in Figure 6–17 on page 6–18. The delay chains can independently control the
rising and falling edge delays of the output buffer, providing the ability to adjust the
output-buffer duty cycle, compensate channel-to-channel skew, reduce simultaneous
switching output (SSO) noise by deliberately introducing channel-to-channel skew,
and improve high-speed memory-interface timing margins. Stratix IV devices
support four levels of output buffer delay settings. The default setting is No Delay.
f For more information about programmable output buffer delay specifications, refer to
the High-Speed Differential I/O Interfaces and DPA in Stratix IV Devices chapter.
Open-Drain Output
Stratix IV devices provide an optional open-drain output (equivalent to an open
collector output) for each I/O pin. When configured as open drain, the logic value of
the output is either high-Z or 0. Typically, an external pull-up resistor is required to
provide logic high.
Bus Hold
Each Stratix IV device I/O pin provides an optional bus-hold feature. Bus-hold
circuitry can weakly hold the signal on an I/O pin at its last-driven state. Because the
bus-hold feature holds the last-driven state of the pin until the next input signal is
present, you do not need an external pull-up or pull-down resistor to hold a signal
level when the bus is tri-stated.
Bus-hold circuitry also pulls non-driven pins away from the input threshold voltage
where noise can cause unintended high-frequency switching. You can select this
feature individually for each I/O pin. The bus-hold output drives no higher than
VCCIO to prevent over-driving signals. If you enable the bus-hold feature, you cannot
use the programmable pull-up option. Disable the bus-hold feature if the I/O pin is
configured for differential signals.
Bus-hold circuitry uses a resistor with a nominal resistance (RBH) of approximately
7 k to weakly pull the signal level to the last-driven state.
Stratix IV Device Handbook
Volume 1
September 2012 Altera Corporation
Chapter 6: I/O Features in Stratix IV Devices
I/O Structure
6–23
f For more information about the specific sustaining current driven through this
resistor and the overdrive current used to identify the next-driven input level, refer to
the High-Speed Differential I/O Interfaces and DPA in Stratix IV Devices chapter.
Bus-hold circuitry is active only after configuration. When going into user mode, the
bus-hold circuit captures the value on the pin present at the end of configuration.
Programmable Pull-Up Resistor
Each Stratix IV device I/O pin provides an optional programmable pull-up resistor
during user mode. If you enable this feature for an I/O pin, the pull-up resistor
(typically 25 K ) weakly holds the I/O to the VCCIO level.
Programmable pull-up resistors are only supported on user I/O pins and are not
supported on dedicated configuration pins, JTAG pins, or dedicated clock pins. If you
enable the programmable pull-up option, you cannot use the bus-hold feature.
1
When the optional DEV_OE signal drives low, all the I/O pins remain tri-stated even
with the programmable pull-up option enabled.
Programmable Pre-Emphasis
Stratix IV LVDS transmitters support programmable pre-emphasis to compensate for
the frequency dependent attenuation of the transmission line. The Quartus II software
allows four settings for programmable pre-emphasis.
f For more information about programmable pre-emphasis, refer to the High-Speed
Differential I/O Interfaces and DPA in Stratix IV Devices chapter.
Programmable Differential Output Voltage
Stratix IV LVDS transmitters support programmable VOD. The programmable VOD
settings allow you to adjust output eye height to optimize trace length and power
consumption. A higher VOD swing improves voltage margins at the receiver end; a
smaller VOD swing reduces power consumption. The Quartus II software allows four
settings for programmable VOD.
f For more information about programmable VOD, refer to the High-Speed Differential I/O
Interfaces and DPA in Stratix IV Devices chapter.
MultiVolt I/O Interface
The Stratix IV architecture supports the MultiVolt I/O interface feature that allows the
Stratix IV devices in all packages to interface with systems of different supply
voltages.
You can connect the VCCIO pins to a 1.2-, 1.5-, 1.8-, 2.5-, or 3.0-V power supply,
depending on the output requirements. The output levels are compatible with
systems of the same voltage as the power supply. (For example, when VCCIO pins are
connected to a 1.5-V power supply, the output levels are compatible with 1.5-V
systems.)
September 2012
Altera Corporation
Stratix IV Device Handbook
Volume 1
6–24
Chapter 6: I/O Features in Stratix IV Devices
On-Chip Termination Support and I/O Termination Schemes
f For more information about pin connection guidelines, refer to the Stratix IV GX and
Stratix IV E Device Family Pin Connection Guidelines.
The Stratix IV VCCPD power pins must be connected to a 2.5- or 3.0-V power supply.
Using these power pins to supply the pre-driver power to the output buffers increases
the performance of the output pins. Table 6–5 lists Stratix IV MultiVolt I/O support.
Table 6–5. Stratix IV MultiVolt I/O Support
(1)
Input Signal (V)
VCCIO (V)
Output Signal (V)
(3)
1.2
1.5
1.8
2.5
3.0
3.3
1.2
1.5
1.8
2.5
3.0
3.3
1.2
Y
—
—
—
—
1.5
—
Y
Y
—
—
—
Y
—
—
—
—
—
—
—
Y
—
—
—
—
1.8
—
Y
Y
—
—
—
—
—
Y
—
—
—
2.5
—
—
—
Y
Y (2)
Y (2)
—
—
—
Y
—
—
3.0
—
—
—
Y
Y
Y
—
—
—
—
Y
—
Notes to Table 6–5:
(1) The pin current may be slightly higher than the default value. You must verify that the driving device’s VOL maximum and VOH minimum voltages
do not violate the applicable Stratix IV VIL maximum and VIH minimum voltage specifications.
(2) Altera recommends that you use an external clamping diode on the I/O pins when the input signal is 3.0 V or 3.3 V. You have the option to use
an internal clamping diode for column I/O pins.
(3) Each I/O bank of a Stratix IV device has its own VCCIO pins and supports only one VCCIO, either 1.2, 1.5, 1.8, or 3.0 V. The LVDS I/O standard
is not supported when VCCIO is 3.0 V. The LVDS input operations are supported when VCCIO is 1.2 V, 1.5 V, 1.8 V, or 2.5 V. The LVDS output
operations are only supported when VCCIO is 2.5 V.
On-Chip Termination Support and I/O Termination Schemes
Stratix IV devices feature dynamic series and parallel OCT to provide I/O impedance
matching and termination capabilities. OCT maintains signal quality, saves board
space, and reduces external component costs.
Stratix IV devices support:
■
On-chip series termination (RS) with calibration
■
On-chip series termination (RS) without calibration
■
On-chip Parallel termination (RT) with calibration
■
Dynamic series termination for single-ended I/O standards
■
Dynamic Parallel termination for single-ended I/O standards
■
On-chip differential termination (RD) for differential LVDS I/O standards
Stratix IV devices support OCT in all I/O banks by selecting one of the OCT I/O
standards.
These devices also support OCT RS and RT in the same I/O bank for different I/O
standards if they use the same VCCIO supply voltage. You can independently configure
each I/O in an I/O bank to support OCT RS, programmable current strength, or OCT
RT.
1
Stratix IV Device Handbook
Volume 1
You cannot configure both OCT RS and programmable current strength for the same
I/O buffer.
September 2012 Altera Corporation
Chapter 6: I/O Features in Stratix IV Devices
On-Chip Termination Support and I/O Termination Schemes
6–25
A pair of RUP and RDN pins are available in a given I/O bank and are shared for
series- and parallel-calibrated termination. The RUP and RDN pins share the same VCCIO
and GND, respectively, with the I/O bank where they are located. The RUP and RDN
pins are dual-purpose I/Os and function as regular I/Os if you do not use the
calibration circuit.
For calibration, the connections are as follows:
■
The RUP pin is connected to VCCIO through an external 25- ±1% or 50- ±1%
resistor for an on-chip series termination value of 25-or 50-, respectively.
■
The RDN pin is connected to GND through an external 25- ±1% or 50- ±1%
resistor for an on-chip series termination value of 25-or 50-, respectively.
For on-chip parallel termination, the connections are as follows:
■
The RUP pin is connected to VCCIO through an external 50- ±1% resistor.
■
The RDN pin is connected to GND through an external 50- ±1% resistor.
On-Chip Series (RS) Termination Without Calibration
Stratix IV devices support driver-impedance matching to provide the I/O driver with
controlled output impedance that closely matches the impedance of the transmission
line. As a result, you can significantly reduce reflections. Stratix IV devices support
on-chip series termination for single-ended I/O standards (Figure 6–18).
The RS shown in Figure 6–18 is the intrinsic impedance of the output transistors.
Typical RS values are 25 and 50 . When you select matching impedance, current
strength is no longer selectable.
Figure 6–18. On-Chip Series Termination Without Calibration
Stratix IV Driver
Series Termination
Receiving
Device
VCCIO
RS
ZO = 50 Ω
RS
GND
To use on-chip termination for the SSTL Class I standard, you must select the 50-
on-chip series termination setting, thus eliminating the external 25- RS (to match
the 50- transmission line). For the SSTL Class II standard, you must select the 25-
on-chip series termination setting (to match the 50- transmission line and the
near-end external 50- pull-up to VTT).
September 2012
Altera Corporation
Stratix IV Device Handbook
Volume 1
6–26
Chapter 6: I/O Features in Stratix IV Devices
On-Chip Termination Support and I/O Termination Schemes
On-Chip Series Termination with Calibration
Stratix IV devices support on-chip series termination with calibration in all banks. The
on-chip series termination calibration circuit compares the total impedance of the I/O
buffer to the external 25- ±1% or 50- ±1% resistors connected to the RUP and RDN
pins and dynamically enables or disables the transistors until they match.
The RS shown in Figure 6–19 is the intrinsic impedance of the transistors. Calibration
occurs at the end of device configuration. When the calibration circuit finds the
correct impedance, it powers down and stops changing the characteristics of the
drivers.
Figure 6–19. On-Chip Series Termination with Calibration
Stratix IV Driver
Series Termination
Receiving
Device
VCCIO
RS
ZO = 50 Ω
RS
GND
Table 6–6 lists the I/O standards that support on-chip series termination with and
without calibration.
Table 6–6. Selectable I/O Standards for On-Chip Series Termination with and Without Calibration
(Part 1 of 2)
On-Chip Series Termination Setting
I/O Standard
3.3-V LVTTL/LVCMOS
2.5-V LVCMOS
1.8-V LVCMOS
Stratix IV Device Handbook
Volume 1
Row I/O ()
Column I/O ()
50
50
25
25
50
50
25
25
50
50
25
25
50
1.5-V LVCMOS
50
1.2-V LVCMOS
50
SSTL-2 Class I
50
50
SSTL-2 Class II
25
25
SSTL-18 Class I
50
50
SSTL-18 Class II
25
25
25
50
25
September 2012 Altera Corporation
Chapter 6: I/O Features in Stratix IV Devices
On-Chip Termination Support and I/O Termination Schemes
6–27
Table 6–6. Selectable I/O Standards for On-Chip Series Termination with and Without Calibration
(Part 2 of 2)
On-Chip Series Termination Setting
I/O Standard
Row I/O ()
Column I/O ()
SSTL-15 Class I
50
50
SSTL-15 Class II
—
25
HSTL-18 Class I
50
50
HSTL-18 Class II
25
25
HSTL-15 Class I
50
50
HSTL-15 Class II
—
25
HSTL-12 Class I
50
50
HSTL-12 Class II
—
25
Left-Shift Series Termination Control
Stratix IV devices support left-shift series termination control. You can use left-shift
series termination control to get the calibrated OCT RS with half of the impedance
value of the external reference resistors connected to the RUP and RDN pins. This feature
is useful in applications that require both 25- and 50- calibrated OCT RS at the
same VCCIO. For example, if your application requires 25- and 50- calibrated OCT
RS for SSTL-2 Class I and Class II I/O standards, you only need one OCT calibration
block with 50- external reference resistors.
You can enable the left-shift series termination control feature in the ALTIOBUF
megafunction in the Quartus II software. The Quartus II software only allows
left-shift series termination control for 25- calibrated OCT RS with 50- external
reference resistors connected to the RUP and RDN pins. You can only use left-shift series
termination control for the I/O standards that support 25- calibrated OCT RS .
1
This feature is automatically enabled if you are using a bidirectional I/O with 25-
calibrated OCT RS and 50- parallel OCT.
f For more information about how to enable the left-shift series termination feature in
the ALTIOBUF megafunction, refer to the I/O Buffer (ALTIOBUF) Megafunction User
Guide.
September 2012
Altera Corporation
Stratix IV Device Handbook
Volume 1
6–28
Chapter 6: I/O Features in Stratix IV Devices
On-Chip Termination Support and I/O Termination Schemes
On-Chip Parallel Termination with Calibration
Stratix IV devices support on-chip parallel termination with calibration in all banks.
On-chip parallel termination with calibration is only supported for input
configuration of input and bidirectional pins. Output pin configurations do not
support on-chip parallel termination with calibration. Figure 6–20 shows on-chip
parallel termination with calibration. When you use parallel OCT, the VCCIO of the
bank must match the I/O standard of the pin where the parallel OCT is enabled.
Figure 6–20. On-Chip Parallel Termination with Calibration
Stratix IV OCT
VCCIO
100 Ω
ZO = 50 Ω
V
REF
100 Ω
GND
Transmitter
Receiver
The on-chip parallel termination calibration circuit compares the total impedance of
the I/O buffer to the external 50- ±1% resistors connected to the RUP and RDN pins
and dynamically enables or disables the transistors until they match. Calibration
occurs at the end of device configuration. When the calibration circuit finds the
correct impedance, it powers down and stops changing the characteristics of the
drivers. Table 6–7 lists the I/O standards that support on-chip parallel termination
with calibration.
Table 6–7. Selectable I/O Standards with On-Chip Parallel Termination with Calibration
I/O Standard
Stratix IV Device Handbook
Volume 1
On-Chip Parallel
Termination Setting
(Column I/O) ()
On-Chip Parallel
Termination Setting
(Row I/O) ()
SSTL-2 Class I, II
50
50
SSTL-18 Class I, II
50
50
SSTL-15 Class I, II
50
50
HSTL-18 Class I, II
50
50
HSTL-15 Class I, II
50
50
HSTL-12 Class I, II
50
50
Differential SSTL-2 Class I, II
50
50
Differential SSTL-18 Class I, II
50
50
Differential SSTL-15 Class I, II
50
50
Differential HSTL-18 Class I, II
50
50
Differential HSTL-15 Class I, II
50
50
Differential HSTL-12 Class I, II
50
50
September 2012 Altera Corporation
Chapter 6: I/O Features in Stratix IV Devices
On-Chip Termination Support and I/O Termination Schemes
6–29
Expanded On-Chip Series Termination with Calibration
OCT calibration circuits always adjust OCT RS to match the external resistors
connected to the RUP and RDN pin; however, it is possible to achieve OCT RS values
other than the 25- and 50- resistors. Theoretically, if you need a different OCT RS
value, you can change the resistance connected to the RUP and RDN pins accordingly.
Practically, the OCT RS range that Stratix IV devices support is limited because of
output buffer size and granularity limitations.
The Quartus II software only allows discrete OCT RS calibration settings of 25, 40, 50,
and 60 . You can select the closest discrete value of OCT RS with calibration settings
in the Quartus II software to your system to achieve the closest timing. For example, if
you are using 20- OCT RS with calibration in your system, you can select the 25-
OCT RS with calibration setting in the Quartus II software to achieve the closest
timing.
Table 6–8 lists expanded OCT RS with calibration supported in Stratix IV devices. Use
expanded on-chip series termination with calibration of SSTL and HSTL for
impedance matching to improve signal integrity but do not use it to meet the JEDEC
standard.
Table 6–8. Selectable I/O Standards with Expanded On-Chip Series Termination with Calibration
Range
Expanded OCT RS Range
I/O Standard
Row I/O ()
Column I/O ()
3.3-V LVTTL/LVCMOS
20–60
20–60
2.5-V LVTTL/LVCMOS
20–60
20–60
1.8-V LVTTL/LVCMOS
20–60
20–60
1.5-V LVTTL/LVCMOS
40–60
20–60
1.2-V LVTTL/LVCMOS
40–60
20–60
SSTL-2
20–60
20–60
SSTL-18
20–60
20–60
SSTL-15
40–60
20–60
HSTL-18
20–60
20–60
HSTL-15
40–60
20–60
HSTL-12
40–60
20–60
Dynamic On-Chip Termination
Stratix IV devices support on and off dynamic termination, both series and parallel,
for a bidirectional I/O in all I/O banks. Figure 6–21 shows the termination schemes
supported in Stratix IV devices. Dynamic parallel termination is enabled only when
the bidirectional I/O acts as a receiver and is disabled when it acts as a driver.
Similarly, dynamic series termination is enabled only when the bidirectional I/O acts
as a driver and is disabled when it acts as a receiver. This feature is useful for
terminating any high-performance bidirectional path because signal integrity is
optimized depending on the direction of the data.
Using dynamic OCT helps save power because device termination is internal instead
of external. Termination only switches on during input operation, thus drawing less
static power.
September 2012
Altera Corporation
Stratix IV Device Handbook
Volume 1
6–30
Chapter 6: I/O Features in Stratix IV Devices
On-Chip Termination Support and I/O Termination Schemes
1
When using calibrated input parallel and calibrated output series termination on
bidirectional pins, they must use the same termination value because each I/O pin
can only reference one OCT calibration block. The only exception is when using 50
parallel OCT and 25 series OCT using the left shift series termination control. For
example, you cannot use calibrated 50 parallel OCT on the input buffer of a
bidirectional pin and calibrated 40 series OCT on the output buffer because these
would require two separate calibration blocks with different RUP and RDN resistor
values.
Figure 6–21. Dynamic Parallel OCT in Stratix IV Devices
VCCIO
VCCIO
Transmitter
Receiver
100 Ω
100 Ω
50 Ω
ZO = 50 Ω
100 Ω
100 Ω
50 Ω
GND
GND
Stratix IV OCT
Stratix IV OCT
VCCIO
VCCIO
100 Ω
100 Ω
50 Ω
ZO = 50 Ω
100 Ω
100 Ω
50 Ω
GND
GND
Transmitter
Receiver
Stratix IV OCT
Stratix IV OCT
f For more information about tolerance specifications for OCT with calibration, refer to
the DC and Switching Characteristics for Stratix IV Devices chapter.
Stratix IV Device Handbook
Volume 1
September 2012 Altera Corporation
Chapter 6: I/O Features in Stratix IV Devices
On-Chip Termination Support and I/O Termination Schemes
6–31
LVDS Input OCT (RD)
Stratix IV devices support OCT for differential LVDS input buffers with a nominal
resistance value of 100 , as shown in Figure 6–22. Differential OCT RD can be
enabled in row I/O banks when both the VCCIO and VCCPD is set to 2.5 V. Column I/O
banks do not support OCT RD. Dedicated clock input pairs CLK[1,3,8,10][p,n],
PLL_L[1,4]_CLK[p,n], and PLL_R[1,4]_CLK[p,n] on the row I/O banks of Stratix IV
devices do not support RD termination.
Figure 6–22. Differential Input OCT
Transmitter
Receiver
ZO = 50 Ω
100 Ω
ZO = 50 Ω
f For more information about differential on-chip termination, refer to the High-Speed
Differential I/O Interfaces and DPA in Stratix IV Devices chapter.
Summary of OCT Assignments
Table 6–9 lists the OCT assignments for the Quartus II software version 9.1 and later.
Table 6–9. Summary of OCT Assignments in the Quartus II Software
Assignment Name
Value
Applies To
Parallel 50 with calibration
Input buffers for single-ended and
differential HSTL/SSTL standards
Differential
Input buffers for LVDS receivers on
row I/O banks (1)
Input Termination
Series 25 without
calibration
Series 50 without
calibration
Output Termination
Series 25 with calibration
Series 40 with calibration
Output buffers for single-ended
LVTTL/LVCMOS and HSTL/SSTL
standards as well as differential
HSTL/SSTL standards
Series 50 with calibration
Series 60 with calibration
Note to Table 6–9:
(1) You can enable differential OCT RD in row I/O banks when both VCCIO and VCCPD are set to 2.5 V.
September 2012
Altera Corporation
Stratix IV Device Handbook
Volume 1
6–32
Chapter 6: I/O Features in Stratix IV Devices
OCT Calibration
OCT Calibration
Stratix IV devices support calibrated on-chip series termination (RS) and calibrated
on-chip parallel termination (RT) on all I/O pins. You can calibrate the device’s I/O
bank with any of the OCT calibration blocks available in the device provided the
VCCIO of the I/O bank with the pins using calibrated OCT matches the VCCIO of the
I/O bank with the calibration block and its associated RUP and RDN pins.
OCT Calibration Block Location
Table 6–10 and Table 6–11 list the location of OCT calibration blocks in Stratix IV
devices. For both tables, the following legend applies:
1
■
“Y” indicates I/O banks with OCT calibration block
■
”N” indicates I/O banks without OCT calibration block
■
“—” indicates I/O banks that are not available in the device
Table 6–10 and Table 6–11 do not show transceiver banks and transceiver calibration
blocks.
Table 6–10 lists the OCT calibration blocks in Banks 1A through 4C.
Table 6–10. OCT Calibration Block Counts and Placement in Stratix IV Devices (1A through 4C) (Part 1 of 2)
Device
EP4SE230
EP4SE360
EP4SE530
EP4SE820
EP4SGX70
EP4SGX110
EP4SGX180
EP4SGX230
Stratix IV Device Handbook
Volume 1
Bank
Number of
OCT Blocks
1A
1B
1C
2A
2B
2C
3A
3B
3C
4A
4B
4C
780
8
Y
—
N
Y
—
N
Y
—
N
Y
—
N
780
8
Y
—
N
Y
—
N
Y
—
N
Y
—
N
1152
8
Y
—
N
Y
—
N
Y
N
N
Y
N
N
1152
8
Y
—
N
Y
—
N
Y
N
N
Y
N
N
1517
10
Y
N
N
Y
N
N
Y
N
Y
Y
N
N
1760
10
Y
N
N
Y
N
N
Y
N
Y
Y
N
N
1152
8
Y
—
N
Y
—
N
Y
N
N
Y
N
N
1517
10
Y
N
N
Y
N
N
Y
N
Y
Y
N
N
1760
10
Y
N
N
Y
N
N
Y
N
Y
Y
N
N
780
8
Y
—
N
Y
—
N
Y
—
N
Y
—
N
780
8
Y
—
N
Y
—
N
Y
—
N
Y
—
N
1152
8
Y
—
N
—
—
—
Y
—
N
Y
—
N
780
8
Y
—
N
Y
—
N
Y
—
N
Y
—
N
1152
8
Y
—
N
—
—
—
Y
N
N
Y
N
N
1517
8
Y
—
N
Y
—
N
Y
N
N
Y
N
N
780
8
Y
—
N
Y
—
N
Y
—
N
Y
—
N
1152
8
Y
—
N
—
—
—
Y
N
N
Y
N
N
1517
8
Y
—
N
Y
—
N
Y
N
N
Y
N
N
Pin
September 2012 Altera Corporation
Chapter 6: I/O Features in Stratix IV Devices
OCT Calibration
6–33
Table 6–10. OCT Calibration Block Counts and Placement in Stratix IV Devices (1A through 4C) (Part 2 of 2)
Bank
Number of
OCT Blocks
1A
1B
1C
2A
2B
2C
3A
3B
3C
4A
4B
4C
780
8
—
—
—
—
—
—
Y
—
N
Y
—
N
1152
8
Y
—
N
—
—
—
Y
N
N
Y
N
N
1517
8
Y
—
N
Y
—
N
Y
N
N
Y
N
N
1760
8
Y
—
N
Y
—
N
Y
N
N
Y
N
N
1932
10
Y
N
N
Y
—
N
Y
N
Y
Y
N
N
780
8
—
—
—
—
—
—
Y
—
N
Y
—
N
1152
8
Y
—
N
—
—
—
Y
N
N
Y
N
N
1517
8
Y
—
N
Y
—
N
Y
N
N
Y
N
N
1760
8
Y
—
N
Y
—
N
Y
N
N
Y
N
N
1932
10
Y
N
N
Y
—
N
Y
N
Y
Y
N
N
1152
8
Y
—
N
—
—
—
Y
N
Y
Y
N
N
1517
10
Y
—
N
Y
—
N
Y
N
Y
Y
N
N
1760
10
Y
—
N
Y
—
N
Y
N
Y
Y
N
N
1932
10
Y
—
N
Y
N
N
Y
N
Y
Y
N
N
EP4S40G2
1517
8
Y
—
N
Y
—
N
Y
N
N
Y
N
N
EP4S40G5
1517
10
Y
—
N
Y
—
N
Y
N
Y
Y
N
N
EP4S100G2
1517
8
Y
—
N
Y
—
N
Y
N
N
Y
N
N
EP4S100G3
1932
10
Y
—
N
Y
N
N
Y
N
Y
Y
N
N
EP4S100G4
1932
10
Y
—
N
Y
N
N
Y
N
Y
Y
N
N
1517
10
Y
—
N
Y
—
N
Y
N
Y
Y
N
N
1932
10
Y
—
N
Y
N
N
Y
N
Y
Y
N
N
Device
EP4SGX290
EP4SGX360
EP4SGX530
EP4S100G5
Pin
Table 6–11 lists the OCT calibration blocks in Banks 5A through 8C.
Table 6–11. OCT Calibration Block Counts and Placement in Stratix IV Devices (5A through 8C) (Part 1 of 2)
Device
EP4SE230
EP4SE360
EP4SE530
EP4SE820
EP4SGX70
September 2012
Bank
Number of
OCT Blocks
5A
5B
5C
6A
6B
6C
7A
7B
7C
8A
8B
8C
780
8
Y
—
N
Y
—
N
Y
—
N
Y
—
N
780
8
Y
—
N
Y
—
N
Y
—
N
Y
—
N
1152
8
Y
—
N
Y
—
N
Y
N
N
Y
N
N
1152
8
Y
—
N
Y
—
N
Y
N
N
Y
N
N
1517
10
Y
N
N
Y
N
N
Y
N
N
Y
N
Y
1760
10
Y
N
N
Y
N
N
Y
N
N
Y
N
Y
1152
8
Y
—
N
Y
—
N
Y
N
N
Y
N
N
1517
10
Y
N
N
Y
N
N
Y
N
N
Y
N
Y
1760
10
Y
N
N
Y
N
N
Y
N
N
Y
N
Y
780
8
—
—
—
—
—
—
Y
—
N
Y
—
N
Pin
Altera Corporation
Stratix IV Device Handbook
Volume 1
6–34
Chapter 6: I/O Features in Stratix IV Devices
OCT Calibration
Table 6–11. OCT Calibration Block Counts and Placement in Stratix IV Devices (5A through 8C) (Part 2 of 2)
Device
EP4SGX110
Pin
780
Bank
Number of
OCT Blocks
5A
5B
5C
6A
6B
6C
7A
7B
7C
8A
8B
8C
8
—
—
—
—
—
—
Y
—
N
Y
—
N
1152
8
—
—
—
Y
—
N
Y
—
N
Y
—
N
780
8
—
—
—
—
—
—
Y
—
N
Y
—
N
1152
8
—
—
—
Y
—
N
Y
N
N
Y
Y
N
1517
8
Y
—
N
Y
—
N
Y
N
N
Y
N
N
780
8
—
—
—
—
—
—
Y
—
N
Y
—
N
1152
8
—
—
—
Y
—
N
Y
N
N
Y
Y
N
1517
8
Y
—
N
Y
—
N
Y
N
N
Y
N
N
780
8
—
—
—
—
—
—
Y
—
N
Y
—
N
1152
8
—
—
—
Y
—
N
Y
N
N
Y
N
N
1517
8
Y
—
N
Y
—
N
Y
N
N
Y
N
N
1760
8
Y
—
N
Y
—
N
Y
N
N
Y
N
N
1932
10
Y
—
N
Y
N
N
Y
N
N
Y
N
Y
780
8
—
—
—
—
—
—
Y
—
N
Y
—
N
1152
8
—
—
—
Y
—
N
Y
N
N
Y
N
N
1517
8
Y
—
N
Y
—
N
Y
N
N
Y
N
N
1760
8
Y
—
N
Y
—
N
Y
N
N
Y
N
N
1932
10
Y
—
N
Y
N
N
Y
N
N
Y
N
Y
1152
8
—
—
—
Y
—
N
Y
N
N
Y
N
Y
1517
10
Y
—
N
Y
—
N
Y
N
N
Y
N
Y
1760
10
Y
—
N
Y
—
N
Y
N
N
Y
N
Y
1932
10
Y
N
N
Y
—
N
Y
N
N
Y
N
Y
EP4S40G2
1517
8
Y
—
N
Y
—
N
Y
N
N
Y
N
N
EP4S40G5
1517
10
Y
—
N
Y
—
N
Y
N
N
Y
N
Y
EP4S100G2
1517
8
Y
—
N
Y
—
N
Y
N
N
Y
N
N
EP4S100G3
1932
10
Y
N
N
Y
—
N
Y
N
N
Y
N
Y
EP4S100G4
1932
10
Y
N
N
Y
—
N
Y
N
N
Y
N
Y
1517
10
Y
—
N
Y
—
N
Y
N
N
Y
N
Y
1932
10
Y
N
N
Y
—
N
Y
N
N
Y
N
Y
EP4SGX180
EP4SGX230
EP4SGX290
EP4SGX360
EP4SGX530
EP4S100G5
Sharing an OCT Calibration Block on Multiple I/O Banks
An OCT calibration block has the same VCCIO as the I/O bank that contains the block.
OCT RS calibration is supported on all I/O banks with different VCCIO voltage
standards, up to the number of available OCT calibration blocks. You can configure
the I/O banks to receive calibration codes from any OCT calibration block with the
same VCCIO. All I/O banks with the same VCCIO can share one OCT calibration block,
even if that particular I/O bank has an OCT calibration block.
Stratix IV Device Handbook
Volume 1
September 2012 Altera Corporation
Chapter 6: I/O Features in Stratix IV Devices
OCT Calibration
6–35
For example, Figure 6–23 shows a group of I/O banks that has the same VCCIO
voltage. If a group of I/O banks has the same VCCIO voltage, you can use one OCT
calibration block to calibrate the group of I/O banks placed around the periphery.
Because 3B, 4C, 6C, and 7B have the same VCCIO as bank 7A, you can calibrate all four
I/O banks (3B, 4C, 6C, and 7B) with the OCT calibration block (CB7) located in bank
7A. You can enable this by serially shifting out OCT RS calibration codes from the
OCT calibration block located in bank 7A to the I/O banks located around the
periphery.
1
I/O banks that do not contain calibration blocks share calibration blocks with I/O
banks that do contain calibration blocks.
Figure 6–23 is a top view of the silicon die that corresponds to a reverse view for flip
chip packages. It is a graphical representation only. This figure does not show
transceiver banks and transceiver calibration blocks.
Bank 7A
Bank 7B
Bank 7C
Bank 8C
Bank 8B
Bank 8A
CB 7
Figure 6–23. Example of Calibrating Multiple I/O Banks with One Shared OCT Calibration Block
Bank 1A
Bank 6A
Bank 1B
Bank 6B
Bank 1C
Bank 6C
I/O bank with the same VCCIO
Bank 2C
Bank 5C
I/O bank with different VCCIO
Bank 2B
Bank 5B
Bank 2A
Bank 5A
Bank 4A
Bank 4B
Bank 4C
Bank 3C
Bank 3B
Bank 3A
Stratix IV
OCT Calibration Block Modes of Operation
Stratix IV devices support OCT RS and OCT RT on all I/O banks. The calibration can
occur in either power-up or user mode.
Power-Up Mode
In power-up mode, OCT calibration is automatically performed at power up.
Calibration codes are shifted to selected I/O buffers before transitioning to user
mode.
September 2012
Altera Corporation
Stratix IV Device Handbook
Volume 1
6–36
Chapter 6: I/O Features in Stratix IV Devices
OCT Calibration
User Mode
In user mode, the OCTUSRCLK, ENAOCT, nCLRUSR, and ENASER[9..0] signals are used to
calibrate and serially transfer calibration codes from each OCT calibration block to
any I/O. Table 6–12 lists the user-controlled calibration block signal names and their
descriptions.
Table 6–12. OCT Calibration Block Ports for User Control
Signal Name
Description
OCTUSRCLK
Clock for OCT block.
ENAOCT
Enable OCT Termination (Generated by user IP).
When ENOCT = 0, each signal enables the OCT serializer for the
corresponding OCT calibration block.
ENASER[9..0]
When ENAOCT = 1, each signal enables OCT calibration for the
corresponding OCT calibration block.
S2PENA_<bank#>
Serial-to-parallel load enable per I/O bank.
nCLRUSR
Clear user.
Figure 6–24 shows the flow of the user signal. When ENAOCT is 1, all OCT calibration
blocks are in calibration mode; when ENAOCT is 0, all OCT calibration blocks are in
serial data transfer mode. The OCTUSRCLK clock frequency must be 20 MHz or less.
1
You must generate all user signals on the rising edge of OCTUSRCLK.
Figure 6–24 does not show transceiver banks and transceiver calibration blocks.
CB9
Bank 1A
CB7
CB8
CB0
CB6
ENAOCT, nCLRUSR,
Bank 1B
Bank 1C
S2PENA_1C
Stratix IV
Core
Bank 2C
Bank 6C
S2PENA_6C
Bank 5C
OCTUSRCLK,
ENASER[N]
Bank 5B
CB1
CB5
CB3
Bank 4A
Bank 4B
Bank 4C
Bank 3C
Bank 3B
Bank 3A
Bank 5A
CB4
CB2
Stratix IV Device Handbook
Volume 1
Bank 6A
Bank 6B
S2PENA_4C
Bank 2B
Bank 2A
Bank 7A
Bank 7B
Bank 7C
Bank 8C
Bank 8B
Bank 8A
Figure 6–24. Signals Used for User Mode Calibration
September 2012 Altera Corporation
Chapter 6: I/O Features in Stratix IV Devices
OCT Calibration
6–37
OCT Calibration
Figure 6–25 shows user mode signal-timing waveforms. To calibrate OCT block[N]
(where N is a calibration block number), you must assert ENAOCT one cycle before
asserting ENASER[N]. Also, nCLRUSR must be set to low for one OCTUSRCLK cycle before
the ENASER[N] signal is asserted. Assert the ENASER[N] signals for 1000 OCTUSRCLK
cycles to perform OCTRS and OCTRT calibration. You can de-assert ENAOCT one clock
cycle after the last ENASER is de-asserted.
Serial Data Transfer
After you complete calibration, you must serially shift out the 28-bit OCT calibration
codes (14-bit OCT RS and 14-bit OCT RT) from each OCT calibration block to the
corresponding I/O buffers. Only one OCT calibration block can send out the codes at
any time by asserting only one ENASER[N] signal at a time. After you de-assert ENAOCT,
wait at least one OCTUSRCLK cycle to enable any ENASER[N] signal to begin serial
transfer. To shift the 28-bit code from the OCT calibration block[N], you must assert
ENASER[N] for exactly 28 OCTUSRCLK cycles. Between two consecutive asserted ENASER
signals, there must be at least one OCTUSRCLK cycle gap. (Figure 6–25).
Figure 6–25. OCT User Mode Signal—Timing Waveform for One OCT Block
OCTUSRCLK
ENAOCT
Calibration Phase
nCLRUSR
ENASER0
1000 OCTUSRCLK Cycles
28
OCTUSRCLK
Cycles
ts2p (1)
S2PENA_1A
Note to Figure 6–25:
(1) ts2p 25 ns.
After calibrated codes are shifted in serially to each I/O bank, the calibrated codes
must be converted from serial to parallel format before being used in the I/O buffers.
Figure 6–25 shows the S2PENA signals that can be asserted at any time to update the
calibration codes in each I/O bank. All I/O banks that received the codes from the
same OCT calibration block can have S2PENA asserted at the same time, or at a
different time, even while another OCT calibration block is calibrating and serially
shifting codes. The S2PENA signal is asserted one OCTUSRCLK cycle after ENASER is
de-asserted for at least 25 ns. You cannot use I/Os for transmitting or receiving data
when their S2PENA is asserted for parallel codes transfer.
September 2012
Altera Corporation
Stratix IV Device Handbook
Volume 1
6–38
Chapter 6: I/O Features in Stratix IV Devices
Termination Schemes for I/O Standards
Example of Using Multiple OCT Calibration Blocks
Figure 6–26 shows a signal timing waveform for two OCT calibration blocks doing RS
and RT calibration. Calibration blocks can start calibrating at different times by
asserting the ENASER signals at different times. ENAOCT must remain asserted while any
calibration is ongoing. You must set nCLRUSR low for one OCTUSRCLK cycle before each
ENASER[N] signal is asserted. In Figure 6–26, when you set nCLRUSR to 0 for the second
time to initialize OCT calibration block 0, this does not affect OCT calibration block 1,
whose calibration is already in progress.
Figure 6–26. OCT User-Mode Signal Timing Waveform for Two OCT Blocks
OCTUSRCLK
Calibration Phase
ENAOCT
nCLRUSR
1000 OCTUSRCLK
28 OCTUSRCLK
CY CLE S
CY CLE S
ENASER0
ENASER1
1000 OCTUSRCLK
28 OCTUSRCLK
CY CLE S
CY CLE S
ts2p (1)
S2PENA_1A (2)
ts2p (1)
S2PENA_2A (3)
Notes to Figure 6–26:
(1) ts2p 25 ns.
(2) S2PENA_1A is asserted in Bank 1A for calibration block 0.
(3) S2PENA_2A is asserted in Bank 2A for calibration block 1.
RS Calibration
If only RS calibration is used for an OCT calibration block, its corresponding ENASER
signal only requires to be asserted for 240 OCTUSRCLK cycles.
1
You must assert the ENASER signal for 28 OCTUSRCLK cycles for serial transfer.
Termination Schemes for I/O Standards
The following sections describe the different termination schemes for the I/O
standards used in Stratix IV devices.
Single-Ended I/O Standards Termination
Voltage-referenced I/O standards require both an input reference voltage, VREF, and a
termination voltage, VTT. The reference voltage of the receiving device tracks the
termination voltage of the transmitting device.
Figure 6–27 and Figure 6–28 show the details of SSTL and HSTL I/O termination on
Stratix IV devices.
Stratix IV Device Handbook
Volume 1
September 2012 Altera Corporation
Chapter 6: I/O Features in Stratix IV Devices
Termination Schemes for I/O Standards
1
6–39
In Stratix IV devices, you cannot use series and parallel OCT simultaneously. For
more information, refer to “Dynamic On-Chip Termination” on page 6–29.
Figure 6–27. SSTL I/O Standard Termination
Termination
SSTL Class I
SSTL Class II
External
On-Board
Termination
25 Ω
50 Ω
25 Ω
50 Ω
50 Ω
50 Ω
VREF
Receiver
Transmitter
OCT
Transmit
VTT
50 Ω
Receiver
Transmitter
Stratix IV
Series OCT 25 Ω
VTT
VTT
50 Ω 50 Ω
50 Ω
50 Ω
8
VREF
VREF
Transmitter
Receiver
VCCIO
25 Ω
OCT
Receive
Receiver
Transmitter
Stratix IV
Parallel OCT
VTT
100 Ω
25 Ω
50 Ω
VREF
Transmitter
Receiver
VCCIO
Series OCT
50 Ω
100 Ω
Series OCT
25 Ω
100 Ω
100 Ω
Receiver
VCCIO
100 Ω
50 Ω
100 Ω
100 Ω
50 8
Transmitter
VCCIO
VCCIO
Stratix IV
Parallel OCT
VCCIO
50 Ω
VREF
100 Ω
OCT
in BiDirectional
Pins
50 Ω
VREF
Stratix IV
Series OCT 50 Ω
VTT
VTT
VTT
100 Ω
50 Ω
100 Ω
100 Ω
100 Ω
Series
OCT 50 Ω
Stratix IV
September 2012
Altera Corporation
Stratix IV
Series
OCT 25 Ω
Stratix IV
Stratix IV
Stratix IV Device Handbook
Volume 1
6–40
Chapter 6: I/O Features in Stratix IV Devices
Termination Schemes for I/O Standards
Figure 6–28. HSTL I/O Standard Termination
Termination
HSTL Class II
HSTL Class I
VTT
VTT
VTT
50 Ω 50 Ω
50 Ω
External
On-Board
Termination
50 Ω
50 Ω
VREF
VREF
Transmitter
Receiver
VTT
Stratix IV
Series OCT 50 Ω
Receiver
VTT
Stratix IV
Series OCT 25 Ω
50 Ω
50 Ω
VREF
Receiver
Transmitter
VCCIO
100 Ω
50 Ω
VREF
OCT
Receive
VTT
100 Ω
Stratix IV
Stratix IV Device Handbook
Volume 1
100 Ω
Series OCT
25 Ω
Stratix IV
Parallel OCT
100 Ω
Transmitter
Receiver
VCCIO
VCCIO
100 Ω
50 Ω
100 Ω
VCCIO
50 Ω
100 Ω
VCCIO
100 Ω
Receiver
Stratix IV
Parallel OCT
Receiver
VCCIO
Transmitter
50 Ω
VREF
Transmitter
Series OCT
50 Ω
VTT
50 Ω 50 Ω
50 Ω
VREF
OCT
Transmit
OCT
in BiDirectional
Pins
Transmitter
100 Ω
50 8
100 Ω
Series
OCT 50 Ω
Stratix IV
100 Ω
Stratix IV
100 Ω
Series
OCT 25 Ω
Stratix IV
September 2012 Altera Corporation
Chapter 6: I/O Features in Stratix IV Devices
Termination Schemes for I/O Standards
6–41
Differential I/O Standards Termination
Stratix IV devices support differential SSTL-18 and SSTL-2, differential HSTL-18,
HSTL-15, HSTL-12, LVDS, LVPECL, RSDS, and mini-LVDS. Figure 6–29 through
Figure 6–35 show the details of various differential I/O terminations on these devices.
1
Differential HSTL and SSTL outputs are not true differential outputs. They use two
single-ended outputs with the second output programmed as inverted.
Figure 6–29. Differential SSTL I/O Standard Termination
Termination
Differential SSTL Class II
Differential SSTL Class I
VTT VTT
50 Ω
External
On-Board
Termination
25 Ω
25 Ω
VTT VTT
25 Ω
50 Ω
Receiver
Differential SSTL Class I
Z0= 50 Ω
VTT
VCCIO
50 Ω
100 Ω
Z0= 50 Ω
100 Ω
VTT
VCCIO
GND
100 Ω
50 Ω
Z0= 50 Ω
100 Ω
GND
Altera Corporation
Receiver
Transmitter
Series OCT 25 Ω
VCCIO
Z0= 50 Ω
September 2012
50 Ω
Differential SSTL Class II
Series OCT 50 Ω
Transmitter
50 Ω
50 Ω
50 Ω
25 Ω
50 Ω
Transmitter
OCT
50 Ω
50 Ω
50 Ω
VTT VTT
Receiver
100 Ω
100 Ω
VCCIO
GND
100 Ω
100 Ω
GND
Transmitter
Receiver
Stratix IV Device Handbook
Volume 1
6–42
Chapter 6: I/O Features in Stratix IV Devices
Termination Schemes for I/O Standards
Figure 6–30. Differential HSTL I/O Standard Termination
Termination
Differential HSTL Class II
Differential HSTL Class I
VTT VTT
50 Ω
External
On-Board
Termination
50 Ω
50 Ω
50 Ω
50 Ω
50 Ω
Receiver
50 Ω
Receiver
Transmitter
Differential HSTL Class II
Differential HSTL Class I
Series OCT 50 Ω
Series OCT 25 Ω
VCCIO
Z0= 50 Ω
OCT
Z0= 50 Ω
VTT
VCCIO
50 Ω
100 Ω
Z0= 50 Ω
100 Ω
VCCIO
GND
100 Ω
VTT
50 Ω
Z0= 50 Ω
100 Ω
Receiver
100 Ω
100 Ω
VCCIO
GND
100 Ω
100 Ω
GND
GND
Stratix IV Device Handbook
Volume 1
50 Ω 50 Ω
50 Ω
Transmitter
Transmitter
VTT VTT
VTT VTT
Transmitter
Receiver
September 2012 Altera Corporation
Chapter 6: I/O Features in Stratix IV Devices
Termination Schemes for I/O Standards
6–43
LVDS
The LVDS I/O standard is a differential high-speed, low-voltage swing, low-power,
general-purpose I/O interface standard. In Stratix IV devices, the LVDS I/O standard
requires a 2.5-V VCCIO level. The LVDS input buffer requires 2.5-V VCCPD. Use this
standard in applications requiring high-bandwidth data transfer, such as backplane
drivers and clock distribution. LVDS requires a 100- termination resistor between
the two signals at the input buffer. Stratix IV devices provide an optional 100-
differential termination resistor in the device using on-chip differential termination.
Figure 6–31 shows LVDS termination. The on-chip differential resistor is only
available in the row I/O banks.
Figure 6–31. LVDS I/O Standard Termination
(1)
Termination
LVDS
Differential Outputs
Differential Inputs
External On-Board
Termination
50 Ω
100 Ω
50 Ω
Differential Inputs
Differential Outputs
50 Ω
OCT Receive
(True LVDS
Output)
(2)
100 Ω
50 Ω
Stratix IV OCT
OCT Receive
(Single-Ended
LVDS Output
with One-Resistor
Network,
LVDS_E_1R)
(3)
Differential Inputs
Single-Ended Outputs
≤ 1 inch
50 Ω
100 Ω
Rp
50 Ω
External Resistor
Stratix IV OCT
Single-Ended Outputs
OCT Receive
(Single-Ended
LVDS Output
with Three-Resistor
Network,
LVDS_E_3R)
(3)
Differential Inputs
≤ 1 inch
50 Ω
Rs
100 Ω
Rp
Rs
50 Ω
External Resistor
Stratix IV OCT
Notes to Figure 6–31:
(1) For LVDS output with a three-resistor network, the RS and RP values are 120 and 170 , respectively. For LVDS output with a one-resistor network, the
RP value is 120 .
(2) Side I/O banks support true LVDS output buffers.
(3) Column and side I/O banks support LVDS_E_1R and LVDS_E_3R I/O standards using two single-ended output buffers.
September 2012
Altera Corporation
Stratix IV Device Handbook
Volume 1
6–44
Chapter 6: I/O Features in Stratix IV Devices
Termination Schemes for I/O Standards
Differential LVPECL
In Stratix IV devices, the LVPECL I/O standard is supported on input clock pins on
column and row I/O banks. LVPECL output operation is not supported in Stratix IV
devices. LVDS input buffers are used to support LVPECL input operation. AC
coupling is required when the LVPECL common-mode voltage of the output buffer is
higher than the LVPECL input common-mode voltage. Figure 6–32 shows the
AC-coupled termination scheme. The 50- resistors used at the receiver end are
external to the device.
Figure 6–32. LVPECL AC-Coupled Termination
(1)
Altera FPGA
LVPECL Output Buffer
0.1 μF
0.1 μF
Stratix IV LVPECL
Input Buffer
ZO = 50 Ω
50 Ω
VICM
50 Ω
ZO = 50 Ω
Note to Figure 6–32:
(1) The LVPECL AC-coupled termination is applicable only when you use an Altera FPGA LVPECL transmitter.
DC-coupled LVPECL is supported if the LVPECL output common mode voltage is
within the Stratix IV LVPECL input buffer specification (Figure 6–33).
Figure 6–33. LVPECL DC-Coupled Termination
(1)
Altera FPGA
LVPECL Output Buffer
Stratix IV LVPECL
Input Buffer
ZO = 50 Ω
ZO = 50 Ω
100 Ω
Note to Figure 6–33:
(1) The LVPECL DC-coupled termination is applicable only when you use an Altera FPGA LVPECL transmitter.
Stratix IV Device Handbook
Volume 1
September 2012 Altera Corporation
Chapter 6: I/O Features in Stratix IV Devices
Termination Schemes for I/O Standards
6–45
RSDS
Stratix IV devices support the RSDS output standard with data rates up to 230 Mbps
using LVDS output buffer types. For transmitters, use two single-ended output
buffers with the external one- or three-resistor networks in the column I/O bank, as
shown in Figure 6–34. The one-resistor topology is for data rates up to 200 Mbps. The
three-resistor topology is for data rates above 200 Mbps. The row I/O banks support
RSDS output using true LVDS output buffers without an external resistor network.
Figure 6–34. RSDS I/O Standard Termination
(1)
One-Resistor Network (RSDS_E_1R)
Termination
Three-Resistor Network (RSDS_E_3R)
≤1 inch
External
On-Board
Termination
RP
≤1 inch
50 Ω
50 Ω
RS
100 Ω
RP
50 Ω
50 Ω
100 Ω
RS
Receiver
Transmitter
Stratix IV OCT
≤1 inch
RP
OCT
Transmitter
50 Ω
50 Ω
Transmitter
Receiver
≤ 1 inch
RS
RP
100 Ω
RS
Receiver
Transmitter
Stratix IV OCT
50 Ω
50 Ω
100 Ω
Receiver
Note to Figure 6–34:
(1) The RS and RP values are pending characterization.
A resistor network is required to attenuate the LVDS output-voltage swing to meet
RSDS specifications. You can modify the three-resistor network values to reduce
power or improve noise margin. The resistor values chosen must satisfy Equation 6–1.
Equation 6–1.
R
R s ------p2
-------------------- = 50
R
R s + ------p2
1
Altera recommends performing additional simulations using IBIS models to validate
that custom resistor values meet the RSDS requirements.
f For more information about the RSDS I/O standard, refer to the RSDS Specification
from the National Semiconductor website at www.national.com.
September 2012
Altera Corporation
Stratix IV Device Handbook
Volume 1
6–46
Chapter 6: I/O Features in Stratix IV Devices
Design Considerations
Mini-LVDS
Stratix IV devices support the mini-LVDS output standard with data rates up to
340 Mbps using LVDS output buffer types. For transmitters, use two single-ended
output buffers with external one- or three-resistor networks, as shown in Figure 6–35.
The one-resistor topology is for data rates up to 200 Mbps. The three-resistor topology
is for data rates above 200 Mbps. The row I/O banks support mini-LVDS output using
true LVDS output buffers without an external resistor network.
Figure 6–35. Mini-LVDS I/O Standard Termination
(1)
One-Resistor Network (mini-LVDS_E_1R)
Termination
Three-Resistor Network (mini-LVDS_E_3R)
≤1 inch
External
On-Board
Termination
R
P
50 Ω
50 Ω
≤1 inch
RS
100 Ω
Receiver
R
Receiver
Stratix IV OCT
≤ 1 inch
RS
50 Ω
R
P
100 Ω
RS
OCT
Transmitter
Receiver
100 Ω
Transmitter
Stratix IV OCT
50 Ω
P
50 Ω
RS
Transmitter
≤1 inch
50 Ω
R
P
Transmitter
50 Ω
50 Ω
100 Ω
Receiver
Note to Figure 6–35:
(1) The RS and RP values are pending characterization.
A resistor network is required to attenuate the LVDS output voltage swing to meet the
mini-LVDS specifications. You can modify the three-resistor network values to reduce
power or improve noise margin. The resistor values chosen must satisfy Equation 6–1
on page 6–45.
1
Altera recommends that you perform additional simulations using IBIS models to
validate that custom resistor values meet the RSDS requirements.
f For more information about the mini-LVDS I/O standard, see the mini-LVDS
Specification from the Texas Instruments website at www.ti.com.
Design Considerations
Although Stratix IV devices feature various I/O capabilities for high-performance
and high-speed system designs, there are several other design considerations that
require your attention to ensure the success of your designs.
I/O Bank Restrictions
Each I/O bank can simultaneously support multiple I/O standards. The following
sections provide guidelines for mixing non-voltage-referenced and voltage-referenced
I/O standards in Stratix IV devices.
Stratix IV Device Handbook
Volume 1
September 2012 Altera Corporation
Chapter 6: I/O Features in Stratix IV Devices
Design Considerations
6–47
Non-Voltage-Referenced Standards
Each I/O bank of a Stratix IV device has its own VCCIO pins and supports only one
VCCIO, either 1.2, 1.5, 1.8, 2.5, or 3.0 V. An I/O bank can simultaneously support any
number of input signals with different I/O standard assignments if it meets the VCCIO
and VCCPD requirement, as shown in Table 6–2 on page 6–3.
For output signals, a single I/O bank supports non-voltage-referenced output signals
that are driving at the same voltage as VCCIO. Because an I/O bank can only have one
VCCIO value, it can only drive out that one value for non-voltage-referenced signals.
For example, an I/O bank with a 2.5-V VCCIO setting can support 2.5-V standard
inputs and outputs as well as 3.0-V LVCMOS inputs (but not output or bidirectional
pins).
Voltage-Referenced Standards
To accommodate voltage-referenced I/O standards, each Stratix IV device’s I/O bank
supports multiple VREF pins feeding a common VREF bus. The number of available
VREF pins increases as device density increases. If these pins are not used as VREF pins,
they cannot be used as generic I/O pins and must be tied to VCCIO or GND. Each bank
can only have a single VCCIO voltage level and a single VREF voltage level at a given
time.
An I/O bank featuring single-ended or differential standards can support
voltage-referenced standards if all voltage-referenced standards use the same VREF
setting.
For performance reasons, voltage-referenced input standards use their own VCCPD
level as the power source. This feature allows you to place voltage-referenced input
signals in an I/O bank with a VCCIO of 2.5 V or below. For example, you can place
HSTL-15 input pins in an I/O bank with 2.5-V VCCIO. However, the voltage-referenced
input with parallel OCT enabled requires the VCCIO of the I/O bank to match the
voltage of the input standard.
Voltage-referenced bidirectional and output signals must be the same as the I/O
bank’s VCCIO voltage. For example, you can only place SSTL-2 output pins in an I/O
bank with a 2.5-V VCCIO.
Mixing Voltage-Referenced and Non-Voltage-Referenced Standards
An I/O bank can support both voltage-referenced and non-voltage-referenced pins by
applying each of the rule sets individually. For example, an I/O bank can support
SSTL-18 inputs and 1.8-V inputs and outputs with a 1.8-V VCCIO and a 0.9-V VREF.
Similarly, an I/O bank can support 1.5-V standards, 1.8-V inputs (but not outputs),
and HSTL and HSTL-15 I/O standards with a 1.5-V VCCIO and 0.75-V VREF.
September 2012
Altera Corporation
Stratix IV Device Handbook
Volume 1
6–48
Chapter 6: I/O Features in Stratix IV Devices
Design Considerations
Document Revision History
Table 6–13 lists the revision history for this chapter.
Table 6–13. Document Revision History (Part 1 of 2)
Date
Version
September 2012
3.4
December 2011
3.3
February 2011
March 2010
November 2009
June 2009
April 2009
March 2009
Stratix IV Device Handbook
Volume 1
3.2
3.1
3.0
2.3
2.2
Changes
■
Updated the “Programmable Slew Rate Control” section to close FB #68385.
■
Updated Figure 6–17 to close FB #57979.
Updated Figure 6–2 and Figure 6–17.
■
Updated the “Modular I/O Banks”, “On-Chip Termination Support and I/O Termination
Schemes”, “Dynamic On-Chip Termination”, and “Programmable Pull-Up Resistor”
sections.
■
Updated Figure 6–17, Figure 6–32, and Figure 6–33.
■
Applied new template.
■
Minor text edits.
■
Updated Table 6–2 and Table 6–5.
■
Updated Figure 6–18, Figure 6–19, Figure 6–27, Figure 6–28, and Figure 6–31.
■
Added the “Summary of OCT Assignments” section.
■
Added a note to the “Sharing an OCT Calibration Block on Multiple I/O Banks” section.
■
Updated the “OCT Calibration” section.
■
Minor text edits.
■
Updated Table 6–2, Table 6–4, Table 6–6, Table 6–9, and Table 6–10.
■
Updated Figure 6–1, Figure 6–2, Figure 6–4, Figure 6–5, Figure 6–6, Figure 6–8,
Figure 6–9, Figure 6–10, Figure 6–11, Figure 6–12, Figure 6–13, and Figure 6–31.
■
Added Table 6–8.
■
Added Figure 6–7, Figure 6–14, Figure 6–15, and Figure 6–16.
■
Added “Left-Shift Series Termination Control” and “Expanded On-Chip Series Termination
with Calibration” sections.
■
Updated “MultiVolt I/O Interface”, “RSDS”, “Mini-LVDS”, and “Non-Voltage-Referenced
Standards” sections.
■
Deleted Figure 6-5: Number of I/Os in Each Bank in EP4SE290 and EP4SE360 in the
1517-Pin FineLine BGA Package.
■
Minor text edits.
■
Added introductory sentences to improve search ability.
■
Removed the Conclusion section.
■
Updated Figure 6–2.
■
Updated Table 6–8 and Table 6–9.
■
Deleted Figure 6-14.
■
Updated Table 6–1, Table 6–2,Table 6–3, Table 6–4, Table 6–6, Table 6–8, and Table 6–9.
■
Updated Figure 6–2, Figure 6–7, Figure 6–8, Figure 6–9, Figure 6–10, Figure 6–11, and
Figure 6–12.
■
Added Figure 6–14.
■
Removed Equation 6–2 and “Referenced Documents” section.
2.1
September 2012 Altera Corporation
Chapter 6: I/O Features in Stratix IV Devices
Design Considerations
6–49
Table 6–13. Document Revision History (Part 2 of 2)
Date
Version
Changes
■
Updated “Modular I/O Banks” on page 6–7.
November 2008
2.0
■
Updated Figure 6–3 and Figure 6–21.
■
Made minor editorial changes.
May 2008
1.0
Initial release.
September 2012
Altera Corporation
Stratix IV Device Handbook
Volume 1
6–50
Stratix IV Device Handbook
Volume 1
Chapter 6: I/O Features in Stratix IV Devices
Design Considerations
September 2012 Altera Corporation
7. External Memory Interfaces in
Stratix IV Devices
February 2011
SIV51007-3.2
SIV51007-3.2
This chapter describes external memory interfaces available with the Stratix IV
device family and that family’s silicon capability to support external memory
interfaces. To support the level of system bandwidth achievable with Altera
Stratix IV FPGAs, the devices provide an efficient architecture to quickly and easily fit
wide external memory interfaces within their small modular I/O bank structure. The
I/Os are designed to provide high-performance support for existing and emerging
external double data rate (DDR) memory standards, such as DDR3, DDR2, DDR
SDRAM, QDR II+, QDR II SRAM, and RLDRAM II.
Stratix IV I/O elements provide easy-to-use built-in functionality required for a rapid
and robust implementation with features such as dynamic calibrated on-chip
termination (OCT), trace mismatch compensation, read- and write-leveling circuit for
DDR3 SDRAM interfaces, half data rate (HDR) blocks, and 4- to 36-bit programmable
DQ group widths.
The high-performance memory interface solution is backed-up by a self-calibrating
megafunction (ALTMEMPHY), optimized to take advantage of the Stratix IV I/O
structure and the TimeQuest Timing Analyzer, which completes the picture by
providing the total solution for the highest reliable frequency of operation across
process, voltage, and temperature (PVT) variations.
This chapter contains the following sections:
■
“Memory Interfaces Pin Support” on page 7–3
■
“Stratix IV External Memory Interface Features” on page 7–29
f For more information about external memory system performance specifications,
board design guidelines, timing analysis, simulation, and debugging information,
refer to the External Memory Interface Handbook.
© 2011 Altera Corporation. All rights reserved. ALTERA, ARRIA, CYCLONE, HARDCOPY, MAX, MEGACORE, NIOS, QUARTUS and STRATIX words and logos
are trademarks of Altera Corporation and registered in the U.S. Patent and Trademark Office and in other countries. All other words and logos identified as
trademarks or service marks are the property of their respective holders as described at www.altera.com/common/legal.html. Altera warrants performance of its
semiconductor products to current specifications in accordance with Altera's standard warranty, but reserves the right to make changes to any products and
services at any time without notice. Altera assumes no responsibility or liability arising out of the application or use of any information, product, or service
described herein except as expressly agreed to in writing by Altera. Altera customers are advised to obtain the latest version of device specifications before relying
on any published information and before placing orders for products or services.
ISO
9001:2008
Registered
Stratix IV Device Handbook
Volume 1
February 2011
Feedback Subscribe
7–2
Chapter 7: External Memory Interfaces in Stratix IV Devices
Figure 7–1 shows an overview of the memory interface data path that uses all the
Stratix IV I/O element (IOE) features.
(1), (2)
Figure 7–1. External Memory Interface Data Path Overview
Memory
Stratix IV FPGA
Postamble Enable
Postamble Clock
4n
DPRAM
(2)
DLL
DQS Logic
Block
Postamble
Control
Circuit
DQS Enable
Circuit
2n
2n
Alignment &
Synchronization
Registers
Half Data Rate
Input Registers
DQS (Read) (3)
DDR Input
Registers
n
DQ (Read) (3)
Resynchronization Clock
n
2n
4n
Half-Rate
Resynchronization
Clock
Clock Management & Reset
DQ Write Clock
Half-Rate Clock
2n
Alignment
Registers
Half Data Rate
Output Registers
2
4
Half Data Rate
Output Registers
2
Alignment
Registers
DQ (Write) (3)
DDR Output
and Output
Enable
Registers
DQS (Write) (3)
DDR Output
and Output
Enable
Registers
Alignment Clock
DQS Write Clock
Notes to Figure 7–1:
(1) You can bypass each register block.
(2) The blocks used for each memory interface may differ slightly. The shaded blocks are part of the Stratix IV IOE.
(3) These signals may be bidirectional or unidirectional, depending on the memory standard. When bidirectional, the signal is active during both read
and write operations.
Memory interfaces use Stratix IV device features such as delay-locked loops (DLLs),
dynamic OCT control, read- and write-leveling circuitry, and I/O features such as
OCT, programmable input delay chains, programmable output delay, slew rate
adjustment, and programmable drive strength.
f For more information about I/O features, refer to the I/O Features in Stratix IV Devices
chapter.
The ALTMEMPHY megafunction instantiates a phase-locked loop (PLL) and PLL
reconfiguration logic to adjust the phase shift based on VT variation. vs
f For more information about the Stratix IV PLL, refer to the Clock Networks and PLLs in
Stratix IV Devices chapter. For more information about the ALTMEMPHY
megafunction, refer to the External Memory PHY Interface (ALTMEMPHY) (nonAFI)
Megafunction User Guide.
Stratix IV Device Handbook
Volume 1
February 2011 Altera Corporation
Chapter 7: External Memory Interfaces in Stratix IV Devices
Memory Interfaces Pin Support
7–3
Memory Interfaces Pin Support
A typical memory interface requires data (D, Q, or DQ), data strobe (DQS/CQ and
DQSn/CQn), address, command, and clock pins. Some memory interfaces use data
mask (DM, BWSn, or NWSn) pins to enable write masking and QVLD pins to indicate
that the read data is ready to be captured. This section describes how Stratix IV
devices support all these different pins.
1
If you have more than one clock pair, you must place them in the same DQ group. For
example, if you have two clock pairs, you must place both of them in the same ×4
DQS group.
f For more information about pin connections, refer to the Stratix IV GX and Stratix IV E
Device Family Pin Connection Guidelines.
f For more information about pin planning and pin connections between a Stratix IV
device and an external memory device, refer to the External Memory Interface
Handbook.
DDR3, DDR2, DDR SDRAM, and RLDRAM II devices use the CK and CK# signals to
capture the address and command signals. Generate these signals to mimic the
write-data strobe using Stratix IV DDR I/O registers (DDIOs) to ensure that the
timing relationships between the CK/CK# and DQS signals (tDQSS, tDSS, and tDSH in
DDR3, DDR2, and DDR SDRAM devices or tCKDK in RLDRAM II devices) are met.
QDR II+ and QDR II SRAM devices use the same clock (K/K#) to capture write data,
address, and command signals.
Memory clock pins in Stratix IV devices are generated using a DDIO register going to
differential output pins (refer to Figure 7–2), marked in the pin table with DIFFOUT,
DIFFIO_TX, or DIFFIO_RX prefixes.
February 2011
Altera Corporation
Stratix IV Device Handbook
Volume 1
7–4
Chapter 7: External Memory Interfaces in Stratix IV Devices
Memory Interfaces Pin Support
f For more information about which pins to use for memory clock pins, refer to the
External Memory Interface Handbook.
Figure 7–2. Memory Clock Generation
FPGA LEs
I/O Elements
VCC
D
Q
1
D
Q
mem_clk (2)
0
mem_clk_n (2)
System Clock (3)
Notes to Figure 7–2:
(1) For pin location requirements,refer to the External Memory Interface Handbook.
(2) The mem_clk[0] and mem_clk_n[0] pins for DDR3, DDR2, and DDR SDRAM interfaces use the I/O input buffer for feedback required by
the ALTMEMPHY megafunction for tracking; therefore, use bidirectional I/O buffers for these pins. For memory interfaces using a differential DQS
input, the input feedback buffer is configured as differential input. For memory interfaces using a single-ended DQS input, the input buffer is
configured as a single-ended input. Using a single-ended input feedback buffer requires that I/O standard’s VREF voltage is provided to that I/O
bank’s VREF pins.
(3) To minimize jitter, regional clock networks are required for memory output clock generation.
Stratix IV devices offer differential input buffers for differential read-data strobe and
clock operations. In addition, Stratix IV devices also provide an independent DQS
logic block for each CQn pin for complementary read-data strobe and clock
operations. In the Stratix IV pin tables, the differential DQS pin pairs are denoted as
DQS and DQSn pins, while the complementary CQ signals are denoted as CQ and
CQn pins. DQSn and CQn pins are marked separately in the pin table. Each CQn pin
connects to a DQS logic block and the shifted CQn signals go to the negative-edge
input registers in the DQ IOE registers.
1
Use differential DQS signaling for DDR2 SDRAM interfaces running at or above
333 MHz.
DQ pins can be bidirectional signals, as in DDR3, DDR2, and DDR SDRAM, and
RLDRAM II common I/O (CIO) interfaces, or unidirectional signals, as in QDR II+,
QDR II SRAM, and RLDRAM II separate I/O (SIO) devices. Connect the
unidirectional read-data signals to Stratix IV DQ pins and the unidirectional
write-data signals to a different DQS/DQ group than the read DQS/DQ group.
Furthermore, the write clocks must be assigned to the DQS/DQSn pins associated to
this write DQS/DQ group. Do not use the CQ/CQn pin-pair for write clocks.
1
Stratix IV Device Handbook
Volume 1
Using a DQS/DQ group for the write-data signals minimizes output skew, allows
access to the write-leveling circuitry (for DDR3 SDRAM interfaces), and allows
vertical migration. These pins also have access to deskewing circuitry (using
programmable delay chains) that can compensate for delay mismatch between signals
on the bus.
February 2011 Altera Corporation
Chapter 7: External Memory Interfaces in Stratix IV Devices
Memory Interfaces Pin Support
7–5
The DQS and DQ pin locations are fixed in the pin table. Memory interface circuitry is
available in every Stratix IV I/O bank that does not support transceivers. All the
memory interface pins support the I/O standards required to support DDR3, DDR2,
DDR SDRAM, QDR II+, QDR II SRAM, and RLDRAM II devices.
The Stratix IV device family supports DQS and DQ signals with DQ bus modes of ×4,
×8/×9, ×16/×18, or ×32/×36, although not all devices support DQS bus mode
×32/×36. When any of these pins are not used for memory interfacing, you can use
them as user I/Os. In addition, you can use any DQSn or CQn pins not used for
clocking as DQ (data) pins. Table 7–1 lists pin support per DQS/DQ bus mode,
including the DQS/CQ and DQSn/CQn pin pair.
Table 7–1. Stratix IV DQS/DQ Bus Mode Pins
Mode
×4
×8/×9
(3)
DQSn Support
CQn Support
Yes
No
Parity or DM
(Optional)
No
Typical
Number of
Data Pins
per Group
Maximum
Number of
Data Pins
per Group (2)
No
4
5
QVLD
(Optional)
(6)
(1)
Yes
Yes
Yes
Yes
8 or 9
11
×16/×18
(4)
Yes
Yes
Yes
Yes
16 or 18
23
×32/×36
(5)
Yes
Yes
Yes
Yes
32 or 36
47
×32/×36
(7)
Yes
32 or 36
39
Yes
Yes
No
(8)
Notes to Table 7–1:
(1) The QVLD pin is not used in the ALTMEMPHY megafunction.
(2) This represents the maximum number of DQ pins (including parity, data mask, and QVLD pins) connected to the DQS bus network with
single-ended DQS signaling. When you use differential or complementary DQS signaling, the maximum number of data per group decreases
by one. This number may vary per DQS/DQ group in a particular device. Check the pin table for the exact number per group. For DDR3, DDR2,
and DDR interfaces, the number of pins is further reduced for an interface larger than ×8 due to the need of one DQS pin for each ×8/×9 group
that is used to form the x16/×18 and ×32/×36 groups.
(3) Two ×4 DQS/DQ groups are stitched to make a ×8/×9 group so there are a total of 12 pins in this group.
(4) Four ×4 DQS/DQ groups are stitched to make a ×16/×18 group.
(5) Eight ×4 DQS/DQ groups are stitched to make a ×32/×36 group.
(6) The DM pin can be supported if differential DQS is not used and the group does not have additional signals.
(7) These ×32/×36 DQS/DQ groups are available in EP4SGX290, EP4SGX360, and EP4SGX530 devices in 1152- and 1517-pin FineLine BGA
packages. There are 40 pins in each of these DQS/DQ groups.
(8) There are 40 pins in each of these DQS/DQ groups. The BWSn pins cannot be placed within the same DQS/DQ group as the write data pins
because of insufficient pins available.
Table 7–2 lists the number of DQS/DQ groups available per side in each Stratix IV
device. For a more detailed listing of the number of DQS/DQ groups available per
bank in each Stratix IV device, see Figure 7–3 through Figure 7–19. These figures
represent the die-top view of the Stratix IV device.
Table 7–2. Number of DQS/DQ Groups in Stratix IV Devices per Side (Part 1 of 3)
Device
Package
EP4SGX70
EP4SGX110
EP4SGX180
EP4SGX230
780-pin
FineLine BGA
EP4SGX290
EP4SGX360
780-pin
FineLine BGA
February 2011
Altera Corporation
Side
×4
(2)
(1)
×8/×9
×16/×18
×32/×36
Left
14
6
2
0
Top/Bottom
17
8
2
0
Right
0
0
0
0
Left/Right
0
0
0
0
Top/Bottom
18
8
2
0
(3)
Refer to:
Figure 7–3
Figure 7–5
Stratix IV Device Handbook
Volume 1
7–6
Chapter 7: External Memory Interfaces in Stratix IV Devices
Memory Interfaces Pin Support
Table 7–2. Number of DQS/DQ Groups in Stratix IV Devices per Side (Part 2 of 3)
Device
Package
EP4SE230
EP4SE360
780-pin
FineLine BGA
EP4SGX110
1152-pin
FineLine BGA
(with 16
transceivers)
EP4SGX70
EP4SGX110
1152-pin
FineLine BGA
(with 24
transceivers)
EP4SGX180
EP4SGX230
1152-pin
FineLine BGA
EP4SGX290
EP4SGX360
EP4SGX530
1152-pin
FineLine BGA
EP4SE360
EP4SE530
EP4SE820
Side
×4
(2)
(1)
×8/×9
×16/×18
×32/×36
Left/Right
14
6
2
0
Top/Bottom
17
8
2
0
Right/Left
7
3
1
0
Top/Bottom
17
8
2
0
Right/Left
14
6
2
0
Top/Bottom
17
8
2
0
(3)
Refer to:
Figure 7–4
Figure 7–6
Figure 7–7
Right/Left
13
6
2
0
Top/Bottom
26
12
4
0
Right/Left
13
6
2
0
Top/Bottom
26
12
4
1152-pin
FineLine BGA
All sides
26
12
4
0
Figure 7–10
EP4SGX180
EP4SGX230
1517-pin
FineLine BGA
All sides
26
12
4
0
Figure 7–11
EP4SGX290
EP4SGX360
EP4SGX530
1517-pin
FineLine BGA
Right/Left
26
12
4
0
Top/Bottom
26
12
4
EP4SE530
EP4SE820
1517-pin
FineLine BGA
Right/Left
34
16
6
0
Top/Bottom
38
18
8
4
EP4S40G2
EP4S40G5
EP4S100G2
EP4S100G5
Left
12
3
1
0
1517-pin
FineLine BGA
Top/Bottom
26
12
4
0
Right
11
4
1
0
EP4SGX290
EP4SGX360
EP4SGX530
1760-pin
FineLine BGA
Right/Left
26
12
4
0
Top/Bottom
38
18
8
4
EP4SE530
1760-pin
FineLine BGA
Right/Left
34
16
6
0
Top/Bottom
38
18
8
4
EP4SE820
1760-pin
FineLine BGA
Right/Left
40
18
6
0
Top/Bottom
44
22
10
4
EP4SGX290
EP4SGX360
EP4SGX530
1932-pin
FineLine BGA
Right/Left
29
13
4
0
Top/Bottom
38
18
8
4
Stratix IV Device Handbook
Volume 1
2
2
(4)
(4)
Figure 7–8
Figure 7–9
Figure 7–12
Figure 7–13
Figure 7–14
Figure 7–15
Figure 7–16
Figure 7–17
Figure 7–18
February 2011 Altera Corporation
Chapter 7: External Memory Interfaces in Stratix IV Devices
Memory Interfaces Pin Support
7–7
Table 7–2. Number of DQS/DQ Groups in Stratix IV Devices per Side (Part 3 of 3)
Device
Package
EP4S100G3
EP4S100G4
EP4S100G5
1932-pin
FineLine BGA
Side
×4
(2)
(1)
×8/×9
×16/×18
×32/×36
Left
8
2
0
0
Top/Bottom
38
18
8
4
Right
7
1
0
0
(3)
Refer to:
Figure 7–19
Notes to Table 7–2:
(1) These numbers are preliminary until the devices are available.
(2) Some of the ×4 groups may use RUP and RDN pins. You cannot use these groups if you use the Stratix IV calibrated OCT feature.
(3) To interface with a ×36 QDR II+/QDR II SRAM device in a Stratix IV FPGA that does not support the ×32/×36 DQS/DQ group, refer to “Combining
×16/×18 DQS/DQ Groups for a ×36 QDR II+/QDR II SRAM Interface” on page 7–26.
(4) These ×32/×36 DQS/DQ groups have 40 pins instead of 48 pins per group. BWSn pins cannot be placed within the same DQS/DQ group as the
write data pins because of insufficient pins available.
February 2011
Altera Corporation
Stratix IV Device Handbook
Volume 1
7–8
Chapter 7: External Memory Interfaces in Stratix IV Devices
Memory Interfaces Pin Support
Figure 7–3. Number of DQS/DQ Groups per Bank in EP4SGX70, EP4SGX110, EP4SGX180, and EP4SGX230 Devices in the
780-Pin FineLine BGA Package (1), (2), (3), (4). (5)
DLL0
I/O Bank 8A
I/O Bank 8C
I/O Bank 7C
I/O Bank 7A
40 User I/Os
x4=6
x8/x9=3
x16/x18=1
24 User I/Os
x4=2
x8/x9=1
x16/x18=0
24 User I/Os
x4=3
x8/x9=1
x16/x18=0
40 User I/Os
x4=6
x8/x9=3
x16/x18=1
DLL3
I/O Bank 1A
32 User I/Os
x4=4
x8/x9=2
x16/x18=1
I/O Bank 1C
26 User I/Os
x4=3
x8/x9=1
x16/x18=0
EP4SGX70, EP4SGX110, EP4SGX180, and
EP4SGX230 Devices in the
780-Pin FineLine BGA
I/O Bank 2C
26 User I/Os
x4=3
x8/x9=1
x16/x18=0
I/O Bank 2A
32 User I/Os
x4=4
x8/x9=2
x16/x18=1
DLL1
I/O Bank 3A
I/O Bank 3C
I/O Bank 4C
I/O Bank 4A
40 User I/Os
x4=6
x8/x9=3
x16/x18=1
24 User I/Os
x4=2
x8/x9=1
x16/x18=0
24 User I/Os
x4=3
x8/x9=1
x16/x18=0
40 User I/Os
x4=6
x8/x9=3
x16/x18=1
DLL2
Notes to Figure 7–3:
(1) These numbers are preliminary until the devices are available.
(2) EP4SGX70, EP4SGX110, EP4SGX180, and EP4SGX230 devices do not support ×32/×36 mode. To interface with a ×36 QDR II+/QDR II SRAM
device, refer to “Combining ×16/×18 DQS/DQ Groups for a ×36 QDR II+/QDR II SRAM Interface” on page 7–26.
(3) You can also use DQS/DQSn pins in some of the ×4 groups as RUP and RDN pins, but you cannot use a ×4 group for memory interfaces if two pins
of the ×4 group are used as RUP and RDN pins for OCT calibration. If two pins of a ×4 group are used as RUP and RDN pins for OCT calibration, you
can use the ×16/×18 or ×32/×36 groups that include that ×4 group; however, there are restrictions on using ×8/×9 groups that include that ×4
group.
(4) You can also use some of the DQS/DQ pins in I/O Bank 1C as configuration pins. You cannot use a ×4 DQS/DQ group with any of its pin members
used for configuration purposes. Ensure that the DQS/DQ groups that you have chosen are not also used for configuration because you may lose
up to four ×4 DQS/DQ groups, depending on your configuration scheme.
(5) All I/O pin counts include dedicated clock inputs that you can use for data inputs.
Stratix IV Device Handbook
Volume 1
February 2011 Altera Corporation
Chapter 7: External Memory Interfaces in Stratix IV Devices
Memory Interfaces Pin Support
7–9
Figure 7–4. Number of DQS/DQ Groups per Bank in EP4SE230 and EP4SE360 Devices in the 780-Pin FineLine BGA
Package (1), (2), (3), (4), (5)
DLL0
I/O Bank 8A
I/O Bank 8C
I/O Bank 7C
I/O Bank 7A
40 User I/Os
x4=6
x8/x9=3
x16/x18=1
24 User I/Os
x4=2
x8/x9=1
x16/x18=0
24 User I/Os
x4=3
x8/x9=1
x16/x18=0
40 User I/Os
x4=6
x8/x9=3
x16/x18=1
DLL3
I/O Bank 1A
I/O Bank 6A
32 User I/Os
x4=4
x8/x9=2
x16/x18=1
32 User I/Os
x4=4
x8/x9=2
x16/x18=1
I/O Bank 1C
I/O Bank 6C
26 User I/Os
x4=3
x8/x9=1
x16/x18=0
26 User I/Os
x4=3
x8/x9=1
x16/x18=0
EP4SE230 and EP4SE360 Devices in the
780-Pin FineLine BGA
I/O Bank 2C
I/O Bank 5C
26 User I/Os
x4=3
x8/x9=1
x16/x18=0
26 User I/Os
x4=3
x8/x9=1
x16/x18=0
I/O Bank 2A
I/O Bank 5A
32 User I/Os
x4=4
x8/x9=2
x16/x18=1
32 User I/Os
x4=4
x8/x9=2
x16/x18=1
DLL1
I/O Bank 3A
I/O Bank 3C
I/O Bank 4C
I/O Bank 4A
40 User I/Os
x4=6
x8/x9=3
x16/x18=1
24 User I/Os
x4=2
x8/x9=1
x16/x18=0
24 User I/Os
x4=3
x8/x9=1
x16/x18=0
40 User I/Os
x4=6
x8/x9=3
x16/x18=1
DLL2
Notes to Figure 7–4:
(1) These numbers are preliminary until the devices are available.
(2) EP4SE230 and EP4SE360 devices do not support ×32/×36 mode. To interface with a ×36 QDR II+/QDR II SRAM device, refer to “Combining
×16/×18 DQS/DQ Groups for a ×36 QDR II+/QDR II SRAM Interface” on page 7–26.
(3) You can also use DQS/DQSn pins in some of the ×4 groups as RUP and RDN pins, but you cannot use a ×4 group for memory interfaces if two pins
of the ×4 group are used as RUP and RDN pins for OCT calibration. If two pins of a ×4 group are used as RUP and RDN pins for OCT calibration, you
can use the ×16/×18 or ×32/×36 groups that include that ×4 group; however, there are restrictions on using ×8/×9 groups that include that ×4
group.
(4) You can also use some of the DQS/DQ pins in I/O Bank 1C as configuration pins. You cannot use a ×4 DQS/DQ group with any of its pin members
used for configuration purposes. Ensure that the DQS/DQ groups that you have chosen are not also used for configuration because you may lose
up to four ×4 DQS/DQ groups, depending on your configuration scheme.
(5) All I/O pin counts include dedicated clock inputs that you can use for data inputs.
February 2011
Altera Corporation
Stratix IV Device Handbook
Volume 1
7–10
Chapter 7: External Memory Interfaces in Stratix IV Devices
Memory Interfaces Pin Support
Figure 7–5. Number of DQS/DQ Groups per Bank in EP4SGX290 and EP4SGX360 Devices in the 780-Pin FineLine BGA
Package (1), (2)
DLL0
I/O Bank 8A
I/O Bank 8C
I/O Bank 7C
I/O Bank 7A
40 User I/Os
x4=6
x8/x9=3
x16/x18=1
32 User I/Os
x4=3
x8/x9=1
x16/x18=0
32 User I/Os
x4=3
x8/x9=1
x16//x18=0
40 User I/Os
x4=6
x8/x9=3
x16/x18=1
DLL3
EP4SGX290 and EP4SGX360 Devices
in the 780-Pin FineLine BGA
DLL1
I/O Bank 3A
I/O Bank 3C
I/O Bank 4C
I/O Bank 4A
40 User I/Os
x4=6
x8/x9=3
x16/x18=1
32 User I/Os
x4=3
x8/x9=1
x16/x18=0
32 User I/Os
x4=3
x8/x9=1
x16/x18=0
40 User I/Os
x4=6
x8/x9=3
x16/x18=1
DLL2
Notes to Figure 7–5:
(1) These numbers are preliminary until the devices are available.
(2) EP4SGX290 and EP4SGX360 devices do not support ×32/×36 mode. To interface with a ×36 QDR II+/QDR II SRAM device, refer to “Combining
×16/×18 DQS/DQ Groups for a ×36 QDR II+/QDR II SRAM Interface” on page 7–26.
Stratix IV Device Handbook
Volume 1
February 2011 Altera Corporation
Chapter 7: External Memory Interfaces in Stratix IV Devices
Memory Interfaces Pin Support
7–11
Figure 7–6. Number of DQS/DQ Groups per Bank in EP4SGX110 Devices with 16 Transceivers in the 1152-Pin FineLine
BGA Package (1), (2), (3), (4), (5)
DLL0
I/O Bank 8A
I/O Bank 8C
I/O Bank 7C
I/O Bank 7A
40 User I/Os
x4=6
x8/x9=3
x16/x18=1
24 User I/Os
x4=2
x8/x9=1
x16/x18=0
24 User I/Os
x4=3
x8/x9=1
x16/x18=0
40 User I/Os
x4=6
x8/x9=3
x16/x18=1
DLL3
I/O Bank 1A
I/O Bank 6A
32 User I/Os
x4=4
x8/x9=2
x16/x18=1
32 User I/Os
x4=4
x8/x9=2
x16/x18=1
EP4SGX110 Devices
in the 1152-Pin FineLine BGA
(with 16 Transceivers)
I/O Bank 1C
26 User I/Os
x4=3
x8/x9=1
x16/x18=0
DLL1
I/O Bank 6C
26 User I/Os
x4=3
x8/x9=1
x16/x18=0
I/O Bank 3A
I/O Bank 3C
I/O Bank 4C
I/O Bank 4A
40 User I/Os
x4=6
x8/x9=3
x16/x18=1
24 User I/Os
x4=2
x8/x9=1
x16/x18=0
24 User I/Os
x4=3
x8/x9=1
x16/x18=0
40 User I/Os
x4=6
x8/x9=3
x16/x18=1
DLL2
Notes to Figure 7–6:
(1) These numbers are preliminary until the devices are available.
(2) EP4SGX110 devices do not support ×32/×36 mode. To interface with a ×36 QDR II+/QDR II SRAM device, refer to “Combining ×16/×18 DQS/DQ
Groups for a ×36 QDR II+/QDR II SRAM Interface” on page 7–26.
(3) You can also use DQS/DQSn pins in some of the ×4 groups as RUP and RDN pins, but you cannot use a ×4 group for memory interfaces if two pins
of the ×4 group are used as RUP and RDN pins for OCT calibration. If two pins of a ×4 group are used as RUP and RDN pins for OCT calibration, you
can use the ×16/×18 or ×32/×36 groups that include that ×4 group; however, there are restrictions on using ×8/×9 groups that include that ×4
group.
(4) You can also use some of the DQS/DQ pins in I/O Bank 1C as configuration pins. You cannot use a ×4 DQS/DQ group with any of its pin members
used for configuration purposes. Ensure that the DQS/DQ groups that you have chosen are not also used for configuration because you may lose
up to four ×4 DQS/DQ groups, depending on your configuration scheme.
(5) All I/O pin counts include dedicated clock inputs that you can use for data inputs.
February 2011
Altera Corporation
Stratix IV Device Handbook
Volume 1
7–12
Chapter 7: External Memory Interfaces in Stratix IV Devices
Memory Interfaces Pin Support
Figure 7–7. Number of DQS/DQ Groups per Bank in EP4SGX70 and EP4SGX110 Devices with 24 Transceivers in the
1152-Pin FineLine BGA Package (1), (2), (3), (4), (5)
DLL0
I/O Bank 8A (3)
I/O Bank 8C
I/O Bank 7C
I/O Bank 7A (3)
40 User I/Os
x4=6
x8/x9=3
x16/x18=1
24 User I/Os
x4=2
x8/x9=1
x16/x18=0
24 User I/Os
x4=3
x8/x9=1
x16/x18=0
40 User I/Os
x4=6
x8/x9=3
x16/x18=1
I/O Bank 1A (3)
DLL3
I/O Bank 6A (3)
32 User I/Os
x4=4
x8/x9=2
x16/x18=1
32 User I/Os
x4=4
x8/x9=2
x16/x18=1
I/O Bank 1C (4)
26 User I/Os (5)
x4=3
x8/x9=1
x16/x18=0
I/O Bank 6C
26 User I/Os (5)
x4=3
x8/x9=1
x16/x18=0
EP4SGX70 and EP4SGX110 Devices
in the 1152-Pin FineLine BGA
(with 24 Transceivers)
I/O Bank 6A (3)
32 User I/Os
x4=4
x8/x9=2
x16/x18=1
32 User I/Os
x4=4
x8/x9=2
x16/x18=1
I/O Bank 1C (4)
26 User I/Os (5)
x4=3
x8/x9=1
x16/x18=0
I/O Bank 6C
26 User I/Os (5)
x4=3
x8/x9=1
x16/x18=0
I/O Bank 3A (3)
DLL1
40 User I/Os
x4=6
x8/x9=3
x16/x18=1
I/O Bank 3C
24 User I/Os
x4=2
x8/x9=1
x16/x18=0
I/O Bank 4C
I/O Bank 4A (3)
24 User I/Os
x4=3
x8/x9=1
x16/x18=0
40 User I/Os
x4=6
x8/x9=3
x16/x18=1
DLL2
Notes to Figure 7–7:
(1) These numbers are preliminary until the devices are available.
(2) EP4SGX70 and EP4SGX110 devices do not support ×32/×36 mode. To interface with a ×36 QDR II+/QDR II SRAM device, refer to “Combining
×16/×18 DQS/DQ Groups for a ×36 QDR II+/QDR II SRAM Interface” on page 7–26.
(3) You can also use DQS/DQSn pins in some of the ×4 groups as RUP and RDN pins, but you cannot use a ×4 group for memory interfaces if two pins
of the ×4 group are used as RUP and RDN pins for OCT calibration. If two pins of a ×4 group are used as RUP and RDN pins for OCT calibration, you
can use the ×16/×18 or ×32/×36 groups that include that ×4 group; however, there are restrictions on using ×8/×9 groups that include that ×4
group.
(4) All I/O pin counts include dedicated clock inputs that you can use for data inputs.
(5) You can also use some of the DQS/DQ pins in I/O Bank 1C as configuration pins. You cannot use a ×4 DQS/DQ group with any of its pin members
used for configuration purposes. Ensure that the DQS/DQ groups that you have chosen are not also used for configuration because you may lose
up to four ×4 DQS/DQ groups, depending on your configuration scheme.
Stratix IV Device Handbook
Volume 1
February 2011 Altera Corporation
Chapter 7: External Memory Interfaces in Stratix IV Devices
Memory Interfaces Pin Support
7–13
Figure 7–8. Number of DQS/DQ Groups per Bank in EP4SGX180 and EP4SGX230 Devices in the 1152-Pin FineLine BGA
Package (1), (2), (3), (4), (5)
DLL0
I/O Bank 8A
I/O Bank 8B
I/O Bank 8C
I/O Bank 7C
I/O Bank 7B
I/O Bank 7A
40 User I/Os
x4=6
x8/x9=3
x16/x18=1
24 User I/Os
x4=4
x8/x9=2
x16/x18=1
32 User I/Os
x4=3
x8/x9=1
x16/x18=0
32 User I/Os
x4=3
x8/x9=1
x16//x18=0
24 User I/Os
x4=4
x8/x9=2
x16/x18=1
40 User I/Os
x4=6
x8/x9=3
x16/x18=1
DLL3
I/O Bank 1A
I/O Bank 6A
48 User I/Os
x4=7
x8/x9=3
x16/x18=1
48 User I/Os
x4=7
x8/x9=3
x6/x18=1
EP4SGX180 and EP4SGX230 Devices
in the 1152-Pin FineLine BGA
I/O Bank 1C
I/O Bank 6C
42 User I/Os
x4=6
x8/x9=3
x16/x18=1
42 User I/Os
x4=6
x8/x9=3
x16/x18=1
DLL1
I/O Bank 3A
I/O Bank 3B
I/O Bank 3C
I/O Bank 4C
I/O Bank 4B
I/O Bank 4A
40 User I/Os
x4=6
x8/x9=3
x16/x18=1
24 User I/Os
x4=4
x8/x9=2
x16/x18=1
32 User I/Os
x4=3
x8/x9=1
x16/x18=0
32 User I/Os
x4=3
x8/x9=1
x16/x18=0
24 User I/Os
x4=4
x8/x9=2
x16/x18=1
40 User I/Os
x4=6
x8/x9=3
x16/x18=1
DLL2
Notes to Figure 7–8:
(1) These numbers are preliminary until the devices are available.
(2) EP4SGX180 and EP4SGX230 devices do not support ×32/×36 mode. To interface with a ×36 QDR II+/QDR II SRAM device, refer to “Combining
×16/×18 DQS/DQ Groups for a ×36 QDR II+/QDR II SRAM Interface” on page 7–26.
(3) You can also use DQS/DQSn pins in some of the ×4 groups as RUP and RDN pins, but you cannot use a ×4 group for memory interfaces if two pins
of the ×4 group are used as RUP and RDN pins for OCT calibration. If two pins of a ×4 group are used as RUP and RDN pins for OCT calibration, you
can use the ×16/×18 or ×32/×36 groups that include that ×4 group; however, there are restrictions on using ×8/×9 groups that include that ×4
group.
(4) All I/O pin counts include dedicated clock inputs that you can use for data inputs.
(5) You can also use some of the DQS/DQ pins in I/O Bank 1C as configuration pins. You cannot use a ×4 DQS/DQ group with any of its pin members
used for configuration purposes. Ensure that the DQS/DQ groups that you have chosen are not also used for configuration because you may lose
up to four ×4 DQS/DQ groups, depending on your configuration scheme.
February 2011
Altera Corporation
Stratix IV Device Handbook
Volume 1
7–14
Chapter 7: External Memory Interfaces in Stratix IV Devices
Memory Interfaces Pin Support
Figure 7–9. Number of DQS/DQ Groups per Bank in EP4SGX290, EP4SGX360, and EP4SGX530 Devices in the 1152-Pin
FineLine BGA Package (1), (3), (4), (5)
DLL0
I/O Bank 8A
40 User I/Os
x4=6
x8/x9=3
x16/x18=1
x32/x36=1 (2)
I/O Bank 8B
I/O Bank 8C
I/O Bank 7C
I/O Bank 7B
24 User I/Os
x4=4
x8/x9=2
x16/x18=1
32 User I/Os
x4=3
x8/x9=1
x16/x18=0
32 User I/Os
x4=3
x8/x9=1
x16//x18=0
24 User I/Os
x4=4
x8/x9=2
x16/x18=1
I/O Bank 7A
40 User I/Os
x4=6
x8/x9=3
x16/x18=1
x32/x36=1 (2)
DLL3
I/O Bank 1A
I/O Bank 6A
48 User I/Os
x4=7
x8/x9=3
x16/x18=1
48 User I/Os
x4=7
x8/x9=3
x6/x18=1
EP4SGX290, EP4SGX360, and EP4SGX530 Devices
in the 1152-Pin FineLine BGA
I/O Bank 1C
I/O Bank 6C
42 User I/Os
x4=6
x8/x9=3
x16/x18=1
42 User I/Os
x4=6
x8/x9=3
x16/x18=1
DLL1
I/O Bank 3A
40 User I/Os
x4=6
x8/x9=3
x16/x18=1
x32/x36=1 (2)
I/O Bank 3B
I/O Bank 3C
I/O Bank 4C
I/O Bank 4B
24 User I/Os
x4=4
x8/x9=2
x16/x18=1
32 User I/Os
x4=3
x8/x9=1
x16/x18=0
32 User I/Os
x4=3
x8/x9=1
x16/x18=0
24 User I/Os
x4=4
x8/x9=2
x16/x18=1
I/O Bank 4A
40 User I/Os
x4=6
x8/x9=3
x16/x18=1
x32/x36=1 (2)
DLL2
Notes to Figure 7–9:
(1) These numbers are preliminary until the devices are available.
(2) These ×32/×36 DQS/DQ groups have 40 pins instead of 48 pins per group.
(3) You can also use DQS/DQSn pins in some of the ×4 groups as RUP and RDN pins, but you cannot use a ×4 group for memory interfaces if two pins
of the ×4 group are used as RUP and RDN pins for OCT calibration. If two pins of a ×4 group are used as RUP and RDN pins for OCT calibration, you
can use the ×16/×18 or ×32/×36 groups that include that ×4 group; however, there are restrictions on using ×8/×9 groups that include that ×4
group.
(4) All I/O pin counts include dedicated clock inputs that you can use for data inputs.
(5) You can also use some of the DQS/DQ pins in I/O Bank 1C as configuration pins. You cannot use a ×4 DQS/DQ group with any of its pin members
used for configuration purposes. Ensure that the DQS/DQ groups that you have chosen are not also used for configuration because you may lose
up to four ×4 DQS/DQ groups, depending on your configuration scheme.
Stratix IV Device Handbook
Volume 1
February 2011 Altera Corporation
Chapter 7: External Memory Interfaces in Stratix IV Devices
Memory Interfaces Pin Support
7–15
Figure 7–10. Number of DQS/DQ Groups per Bank in EP4SE360, EP4SE530, and EP4SE820 Devices in the 1152-Pin
FineLine BGA Package (1), (2), (3), (4), (5)
DLL0
I/O Bank 8A
I/O Bank 8B
I/O Bank 8C
I/O Bank 7C
I/O Bank 7B
I/O Bank 7A
40 User I/Os
x4=6
x8/x9=3
x16/x18=1
24 User I/Os
x4=4
x8/x9=2
x16/x18=1
32 User I/Os
x4=3
x8/x9=1
x16/x18=0
32 User I/Os
x4=3
x8/x9=1
x16//x18=0
24 User I/Os
x4=4
x8/x9=2
x16/x18=1
40 User I/Os
x4=6
x8/x9=3
x16/x18=1
DLL3
I/O Bank 1A
I/O Bank 6A
48 User I/Os
x4=7
x8/x9=3
x16/x18=1
48 User I/Os
x4=7
x8/x9=3
x6/x18=1
I/O Bank 1C
I/O Bank 6C
42 User I/Os
x4=6
x8/x9=3
x16/x18=1
42 User I/Os
x4=6
x8/x9=3
x16/x18=1
EP4SE360, EP4SE530
and EP4SE820 Devices
in the 1152-Pin FineLine BGA
I/O Bank 2C
I/O Bank 5C
42 User I/Os
x4=6
x8/x9=3
x16/x18=1
42 User I/Os
x4=6
x8/x9=3
x16/x18=1
I/O Bank 2A
I/O Bank 5A
48 User I/Os
x4=7
x8/x9=3
x16/x18=1
48 User I/Os
x4=7
x8/x9=3
x6/x18=1
DLL1
I/O Bank 3A
I/O Bank 3B
40 User I/Os
x4=6
x8/x9=3
x16/x18=1
24 User I/Os
x4=4
x8/x9=2
x16/x18=1
I/O Bank 3C
32 User I/Os
x4=3
x8/x9=1
x16/x18=0
I/O Bank 4C
I/O Bank 4B
I/O Bank 4A
32 User I/Os
x4=3
x8/x9=1
x16/x18=0
24 User I/Os
x4=4
x8/x9=2
x16/x18=1
40 User I/Os
x4=6
x8/x9=3
x16/x18=1
DLL2
Notes to Figure 7–10:
(1) These numbers are preliminary until the devices are available.
(2) EP4SE360, EP4SE530, and EP4SE820 devices do not support ×32/×36 mode. To interface with a ×36 QDR II+/QDR II SRAM device, refer to
“Combining ×16/×18 DQS/DQ Groups for a ×36 QDR II+/QDR II SRAM Interface” on page 7–26.
(3) You can also use DQS/DQSn pins in some of the ×4 groups as RUP and RDN pins, but you cannot use a ×4 group for memory interfaces if two pins
of the ×4 group are used as RUP and RDN pins for OCT calibration. If two pins of a ×4 group are used as RUP and RDN pins for OCT calibration, you
can use the ×16/×18 or ×32/×36 groups that include that ×4 group, however there are restrictions on using ×8/×9 groups that include that ×4
group.
(4) You can also use some of the DQS/DQ pins in I/O Bank 1C as configuration pins. You cannot use a ×4 DQS/DQ group with any of its pin members
used for configuration purposes. Ensure that the DQS/DQ groups that you have chosen are not also used for configuration because you may lose
up to four ×4 DQS/DQ groups, depending on your configuration scheme.
(5) All I/O pin counts include dedicated clock inputs that you can use for data inputs.
February 2011
Altera Corporation
Stratix IV Device Handbook
Volume 1
7–16
Chapter 7: External Memory Interfaces in Stratix IV Devices
Memory Interfaces Pin Support
Figure 7–11. Number of DQS/DQ Groups per Bank in EP4SGX180 and EP4SGX230 Devices in the 1517-Pin FineLine BGA
Package (1), (2), (3), (4), (5)
DLL0
I/O Bank 8A
I/O Bank 8B
I/O Bank 8C
I/O Bank 7C
I/O Bank 7B
I/O Bank 7A
40 User I/Os
x4=6
x8/x9=3
x16/x18=1
24 User I/Os
x4=4
x8/x9=2
x16/x18=1
32 User I/Os
x4=3
x8/x9=1
x16/x18=0
32 User I/Os
x4=3
x8/x9=1
x16//x18=0
24 User I/Os
x4=4
x8/x9=2
x16/x18=1
40 User I/Os
x4=6
x8/x9=3
x16/x18=1
DLL3
I/O Bank 1A
I/O Bank 6A
48 User I/Os
x4=7
x8/x9=3
x16/x18=1
48 User I/Os
x4=7
x8/x9=3
x6/x18=1
I/O Bank 1C
I/O Bank 6C
42 User I/Os
x4=6
x8/x9=3
x16/x18=1
42 User I/Os
x4=6
x8/x9=3
x16/x18=1
EP4SGX180 and EP4SGX230 Devices
in the 1517-Pin FineLine BGA
I/O Bank 2C
I/O Bank 5C
42 User I/Os
x4=6
x8/x9=3
x16/x18=1
42 User I/Os
x4=6
x8/x9=3
x16/x18=1
I/O Bank 2A
I/O Bank 5A
48 User I/Os
x4=7
x8/x9=3
x16/x18=1
48 User I/Os
x4=7
x8/x9=3
x6/x18=1
DLL1
I/O Bank 3A
I/O Bank 3B
40 User I/Os
x4=6
x8/x9=3
x16/x18=1
24 User I/Os
x4=4
x8/x9=2
x16/x18=1
I/O Bank 3C
32 User I/Os
x4=3
x8/x9=1
x16/x18=0
I/O Bank 4C
I/O Bank 4B
I/O Bank 4A
32 User I/Os
x4=3
x8/x9=1
x16/x18=0
24 User I/Os
x4=4
x8/x9=2
x16/x18=1
40 User I/Os
x4=6
x8/x9=3
x16/x18=1
DLL2
Notes to Figure 7–11:
(1) These numbers are preliminary until the devices are available.
(2) EP4SGX180 and EP4SGX230 devices do not support ×32/×36 mode. To interface with a ×36 QDR II+/QDR II SRAM device, refer to “Combining
×16/×18 DQS/DQ Groups for a ×36 QDR II+/QDR II SRAM Interface” on page 7–26.
(3) You can also use DQS/DQSn pins in some of the ×4 groups as RUP and RDN pins, but you cannot use a ×4 group for memory interfaces if two pins
of the ×4 group are used as RUP and RDN pins for OCT calibration. If two pins of a ×4 group are used as RUP and RDN pins for OCT calibration, you
can use the ×16/×18 or ×32/×36 groups that include that ×4 group, however there are restrictions on using ×8/×9 groups that include that ×4
group.
(4) All I/O pin counts include dedicated clock inputs that you can use for data inputs.
(5) You can also use some of the DQS/DQ pins in I/O Bank 1C as configuration pins. You cannot use a ×4 DQS/DQ group with any of its pin members
used for configuration purposes. Ensure that the DQS/DQ groups that you have chosen are not also used for configuration because you may lose
up to four ×4 DQS/DQ groups, depending on your configuration scheme.
Stratix IV Device Handbook
Volume 1
February 2011 Altera Corporation
Chapter 7: External Memory Interfaces in Stratix IV Devices
Memory Interfaces Pin Support
7–17
Figure 7–12. Number of DQS/DQ Groups per Bank in EP4SGX290, EP4SGX360, and EP4SGX530 Devices in the 1517-Pin
FineLine BGA Package (1), (3), (4), (5)
DLL0
I/O Bank 8A
40 User I/Os
x4=6
x8/x9=3
x16/x18=1
x32/x36=1 (2)
I/O Bank 8B
I/O Bank 8C
I/O Bank 7C
I/O Bank 7B
24 User I/Os
x4=4
x8/x9=2
x16/x18=1
32 User I/Os
x4=3
x8/x9=1
x16/x18=0
32 User I/Os
x4=3
x8/x9=1
x16//x18=0
24 User I/Os
x4=4
x8/x9=2
x16/x18=1
I/O Bank 7A
40 User I/Os
x4=6
x8/x9=3
x16/x18=1
x32/x36=1 (2)
DLL3
I/O Bank 1A
I/O Bank 6A
48 User I/Os
x4=7
x8/x9=3
x16/x18=1
48 User I/Os
x4=7
x8/x9=3
x6/x18=1
I/O Bank 1C
I/O Bank 6C
42 User I/Os
x4=6
x8/x9=3
x16/x18=1
42 User I/Os
x4=6
x8/x9=3
x16/x18=1
EP4SGX290, EP4SGX360, and EP4SGX530 Devices
in the 1517-Pin FineLine BGA
I/O Bank 2C
I/O Bank 5C
42 User I/Os
x4=6
x8/x9=3
x16/x18=1
42 User I/Os
x4=6
x8/x9=3
x16/x18=1
I/O Bank 2A
I/O Bank 5A
48 User I/Os
x4=7
x8/x9=3
x16/x18=1
48 User I/Os
x4=7
x8/x9=3
x6/x18=1
DLL1
I/O Bank 3A
40 User I/Os
x4=6
x8/x9=3
x16/x18=1
x32/x36=1 (2)
I/O Bank 3B
24 User I/Os
x4=4
x8/x9=2
x16/x18=1
I/O Bank 3C
32 User I/Os
x4=3
x8/x9=1
x16/x18=0
I/O Bank 4C
I/O Bank 4B
32 User I/Os
x4=3
x8/x9=1
x16/x18=0
24 User I/Os
x4=4
x8/x9=2
x16/x18=1
I/O Bank 4A
40 User I/Os
x4=6
x8/x9=3
x16/x18=1
x32/x36=1 (2)
DLL2
Notes to Figure 7–12:
(1) These numbers are preliminary until the devices are available.
(2) These ×32/×36 DQS/DQ groups have 40 pins instead of 48 pins per group.
(3) You can also use DQS/DQSn pins in some of the ×4 groups as RUP and RDN pins, but you cannot use a ×4 group for memory interfaces if two pins
of the ×4 group are used as RUP and RDN pins for OCT calibration. If two pins of a ×4 group are used as RUP and RDN pins for OCT calibration, you
can use the ×16/×18 or ×32/×36 groups that include that ×4 group, however there are restrictions on using ×8/×9 groups that include that ×4
group.
(4) All I/O pin counts include dedicated clock inputs that you can use for data inputs.
(5) You can also use some of the DQS/DQ pins in I/O Bank 1C as configuration pins. You cannot use a ×4 DQS/DQ group with any of its pin members
used for configuration purposes. Ensure that the DQS/DQ groups that you have chosen are not also used for configuration because you may lose
up to four ×4 DQS/DQ groups, depending on your configuration scheme.
February 2011
Altera Corporation
Stratix IV Device Handbook
Volume 1
7–18
Chapter 7: External Memory Interfaces in Stratix IV Devices
Memory Interfaces Pin Support
Figure 7–13. Number of DQS/DQ Groups per Bank in EP4SE530 and EP4SE820 Devices in the 1517-pin FineLine BGA
Package (1), (2), (3), (4)
DLL0
I/O Bank 8A
48 User I/Os
x4=8
x8/x9=4
x16/x18=2
x32/x36=1
I/O Bank 8B
48 User I/Os
x4=8
x8/x9=4
x16/x18=2
x32/x36=1
I/O Bank 8C
32 User I/Os
x4=3
x8/x9=1
x16/x18=0
x32/x36=0
I/O Bank 7C
32 User I/Os
x4=3
x8/x9=1
x16/x18=0
x32/x36=0
I/O Bank 7B
48 User I/Os
x4=8
x8/x9=4
x16/x18=2
x32/x36=1
I/O Bank 7A
48 User I/Os
x4=8
x8/x9=4
x16/x18=2
x32/x36=1
I/O Bank 1A
DLL3
I/O Bank 6A
50 User I/Os
x4=7
x8/x9=3
x16/x18=1
x32/x36=0
50 User I/Os
x4=7
x8/x9=3
x16/x18=1
x32/x36=0
I/O Bank 1B
24 User I/Os
x4=4
x8/x9=2
x16/x18=1
x32/x36=0
I/O Bank 6B
24 User I/Os
x4=4
x8/x9=2
x16/x18=1
x32/x36=0
I/O Bank 1C
42 User I/Os
x4=6
x8/x9=3
x16/x18=1
x32/x36=0
I/O Bank 6C
42 User I/Os
x4=6
x8/x9=3
x16/x18=1
x32/x36=0
EP4SE530 and EP4SE820 Devices
in the 1517-Pin FineLine BGA
I/O Bank 2C
I/O Bank 5C
42 User I/Os
x4=6
x8/x9=3
x16/x18=1
x32/x36=0
42 User I/Os
x4=6
x8/x9=3
x16/x18=1
x32/x36=0
I/O Bank 2B
I/O Bank 5B
24 User I/Os
x4=4
x8/x9=2
x16/x18=1
x32/x36=0
24 User I/Os
x4=4
x8/x9=2
x16/x18=1
x32/x36=0
I/O Bank 2A
I/O Bank 5A
50 User I/Os
x4=7
x8/x9=3
x16/x18=1
x32/x36=0
50 User I/Os
x4=7
x8/x9=3
x16/x18=1
x32/x36=0
DLL1
I/O Bank 3A
I/O Bank 3B
I/O Bank 3C
I/O Bank 4C
I/O Bank 4B
I/O Bank 4A
48 User I/Os
x4=8
x8/x9=4
x16/x18=2
x32/x36=1
48 User I/Os
x4=8
x8/x9=4
x16/x18=2
x32/x36=1
32 User I/Os
x4=3
x8/x9=1
x16/x18=0
x32/x36=0
32 User I/Os
x4=3
x8/x9=1
x16/x18=0
x32/x36=0
48 User I/Os
x4=8
x8/x9=4
x16/x18=2
x32/x36=1
48 User I/Os
x4=8
x8/x9=4
x16/x18=2
x32/x36=1
DLL2
Notes to Figure 7–13:
(1) These numbers are preliminary until the devices are available.
(2) You can also use DQS/DQSn pins in some of the ×4 groups as RUP and RDN pins, but you cannot use a ×4 group for memory interfaces if two pins
of the ×4 group are used as RUP and RDN pins for OCT calibration. If two pins of a ×4 group are used as RUP and RDN pins for OCT calibration, you
can use the ×16/×18 or ×32/×36 groups that include that ×4 group, however there are restrictions on using ×8/×9 groups that include that ×4
group.
(3) All I/O pin counts include dedicated clock inputs and dedicated corner PLL clock inputs that you can use for data inputs.
(4) You can also use some of the DQS/DQ pins in I/O Bank 1C as configuration pins. You cannot use a ×4 DQS/DQ group with any of its pin members
used for configuration purposes. Ensure that the DQS/DQ groups that you have chosen are not also used for configuration because you may lose
up to four ×4 DQS/DQ groups, depending on your configuration scheme.
Stratix IV Device Handbook
Volume 1
February 2011 Altera Corporation
Chapter 7: External Memory Interfaces in Stratix IV Devices
Memory Interfaces Pin Support
7–19
Figure 7–14. Number of DQS/DQ Groups per Bank in EP4S40G2, EP4S40G5, EP4S100G2, and EP4S100G5 Devices in the
1517-Pin FineLine BGA Package (1), (2), (3), (4), (5)
DLL0
I/O Bank 8A
40 User I/Os
x4=6
x8/x9=3
x16/x18=1
I/O Bank 8B
24 User I/Os
x4=4
x8/x9=2
x16/x18=1
I/O Bank 8C
32 User I/Os
x4=3
x8/x9=1
x16/x18=0
I/O Bank 7C
32 User I/Os
x4=3
x8/x9=1
x16/x18=0
I/O Bank 7B
24 User I/Os
x4=4
x8/x9=2
x16/x18=1
I/O Bank 7A
40 User I/Os
x4=6
x8/x9=3
x16/x18=1
DLL3
I/O Bank 1A
I/O Bank 6A
43 User I/Os
x4=5
x8/x9=1
x16/x18=0
44 User I/Os
x4=5
x8/x9=1
x16/x18=0
I/O Bank 1C
20 User I/Os
x4=0
x8/x9=0
x16/x18=0
I/O Bank 6C
21 User I/Os
x4=0
x8/x9=0
x16/x18=0
EP4S40G2, EP4S40G5, EP4S100G2, and EP4S100G5 Devices
in the 1517-Pin FineLine BGA
I/O Bank 2C
21 User I/Os
x4=1
x8/x9=0
x16/x18=0
I/O Bank 5C
21 User I/Os
x4=0
x8/x9=0
x16/x18=0
I/O Bank 2A
I/O Bank 5A
46 User I/Os
x4=6
x8/x9=2
x16/x18=1
46 User I/Os
x4=6
x8/x9=3
x16/x18=1
DLL1
I/O Bank 3A
I/O Bank 3B
I/O Bank 3C
I/O Bank 4C
I/O Bank 4B
I/O Bank 4A
40 User I/Os
x4=6
x8/x9=3
x16/x18=1
24 User I/Os
x4=4
x8/x9=2
x16/x18=1
32 User I/Os
x4=3
x8/x9=1
x16/x18=0
32 User I/Os
x4=3
x8/x9=1
x16/x18=0
24 User I/Os
x4=4
x8/x9=2
x16/x18=1
40 User I/Os
x4=6
x8/x9=3
x16/x18=1
DLL2
Notes to Figure 7–14:
(1) These numbers are preliminary until the devices are available.
(2) EP4S40G2, EP4S40G5, EP4S100G2, and EP4S100G5 devices do not support 32/36 mode. To interface with a 36 QDR II+/QDR II SRAM
device, refer to “Combining ×16/×18 DQS/DQ Groups for a ×36 QDR II+/QDR II SRAM Interface” on page 7–26.
(3) You can also use DQS/DQSn pins in some of the ×4 groups as RUP and RDN pins, but you cannot use a ×4 group for memory interfaces if two pins
of the ×4 group are used as RUP and RDN pins for OCT calibration. If two pins of a ×4 group are used as RUP and RDN pins for OCT calibration, you
can use the ×16/×18 or ×32/×36 groups that include that ×4 group, however there are restrictions on using ×8/×9 groups that include that ×4
group.
(4) All I/O pin counts include dedicated clock inputs that you can use for data inputs.
(5) You can also use some of the DQS/DQ pins in I/O Bank 1C as configuration pins. You cannot use a 4 DQS/DQ group with any of its pin members
used for configuration purposes. Make sure that the DQS/DQ groups that you have chosen are not used for configuration as you may lose up to
four 4 DQS/DQ groups, depending on your configuration scheme.
February 2011
Altera Corporation
Stratix IV Device Handbook
Volume 1
7–20
Chapter 7: External Memory Interfaces in Stratix IV Devices
Memory Interfaces Pin Support
Figure 7–15. Number of DQS/DQ Groups per Bank in EP4SGX290, EP4SGX360, and EP4SGX530 Devices in the 1760-Pin
FineLine BGA Package (1), (2), (3), (4)
DLL0
I/O Bank 8A
48 User I/Os
x4=8
x8/x9=4
x16/x18=2
x32/x36=1
I/O Bank 8B
48 User I/Os
x4=8
x8/x9=4
x16/x18=2
x32/x36=1
I/O Bank 8C
32 User I/Os
x4=3
x8/x9=1
x16/x18=0
x32/x36=0
I/O Bank 7C
32 User I/Os
x4=3
x8/x9=1
x16//x18=0
x32/x36=0
I/O Bank 7B
48 User I/Os
x4=8
x8/x9=4
x16/x18=2
x32/x36=1
I/O Bank 7A
48 User I/Os
x4=8
x8/x9=4
x16/x18=2
x32/x36=1
DLL3
I/O Bank 1A
50 User I/Os
x4=7
x8/x9=3
x16/x18=1
x32/x36=0
I/O Bank 6A
50 User I/Os
x4=7
x8/x9=3
x6/x18=1
x32/x36=0
I/O Bank 1C
42 User I/Os
x4=6
x8/x9=3
x16/x18=1
x32/x36=0
I/O Bank 6C
42 User I/Os
x4=6
x8/x9=3
x16/x18=1
x32/x36=0
EP4SGX290, EP4SGX360, and EP4SGX530 Devices
in the 1760-Pin FineLine BGA
I/O Bank 2C
42 User I/Os
x4=6
x8/x9=3
x16/x18=1
x32/x36=0
I/O Bank 5C
42 User I/Os
x4=6
x8/x9=3
x16/x18=1
x32/x36=0
I/O Bank 5A
50 User I/Os
x4=7
x8/x9=3
x16/x18=1
x32/x36=0
I/O Bank 2A
50 User I/Os
x4=7
x8/x9=3
x16/x18=1
x32/x36=0
DLL1
I/O Bank 3A
48 User I/Os
x4=8
x8/x9=4
x16/x18=2
x32/x36=1
I/O Bank 3B
48 User I/Os
x4=8
x8/x9=4
x16/x18=2
x32/x36=1
I/O Bank 3C
32 User I/Os
x4=3
x8/x9=1
x16/x18=0
x32/x36=0
I/O Bank 4C
32 User I/Os
x4=3
x8/x9=1
x16/x18=0
x32/x36=0
I/O Bank 4B
48 User I/Os
x4=8
x8/x9=4
x16/x18=2
x32/x36=1
I/O Bank 4A
48 User I/Os
x4=8
x8/x9=4
x16/x18=2
x32/x36=1
DLL2
Notes to Figure 7–15:
(1) These numbers are preliminary until the devices are available.
(2) You can also use DQS/DQSn pins in some of the ×4 groups as RUP and RDN pins, but you cannot use a ×4 group for memory interfaces if two pins
of the ×4 group are used as RUP and RDN pins for OCT calibration. If two pins of a ×4 group are used as RUP and RDN pins for OCT calibration, you
can use the ×16/×18 or ×32/×36 groups that include that ×4 group, however there are restrictions on using ×8/×9 groups that include that ×4
group.
(3) All I/O pin counts include dedicated clock inputs and dedicated corner PLL clock inputs that you can use for data inputs.
(4) You can also use some of the DQS/DQ pins in I/O Bank 1C as configuration pins. You cannot use a ×4 DQS/DQ group with any of its pin members
used for configuration purposes. Ensure that the DQS/DQ groups that you have chosen are not also used for configuration because you may lose
up to four ×4 DQS/DQ groups, depending on your configuration scheme.
Stratix IV Device Handbook
Volume 1
February 2011 Altera Corporation
Chapter 7: External Memory Interfaces in Stratix IV Devices
Memory Interfaces Pin Support
7–21
Figure 7–16. Number of DQS/DQ Groups per Bank in EP4SE530 Devices in the 1760-Pin FineLine BGA Package
(1), (2), (3),
(4)
DLL0
I/O Bank 8A
48 User I/Os
x4=8
x8/x9=4
x16/x18=2
x32/x36=1
I/O Bank 8B
48 User I/Os
x4=8
x8/x9=4
x16/x18=2
x32/x36=1
I/O Bank 8C
32 User I/Os
x4=3
x8/x9=1
x16/x18=0
x32/x36=0
I/O Bank 7C
32 User I/Os
x4=3
x8/x9=1
x16/x18=0
x32/x36=0
I/O Bank 7B
48 User I/Os
x4=8
x8/x9=4
x16/x18=2
x32/x36=1
I/O Bank 7A
48 User I/Os
x4=8
x8/x9=4
x16/x18=2
x32/x36=1
DLL3
I/O Bank 1A
I/O Bank 6A
50 User I/Os
x4=7
x8/x9=3
x16/x18=1
x32/x36=0
50 User I/Os
x4=7
x8/x9=3
x16/x18=1
x32/x36=0
I/O Bank 1B
24 User I/Os
x4=4
x8/x9=2
x16/x18=1
x32/x36=0
I/O Bank 6B
24 User I/Os
x4=4
x8/x9=2
x16/x18=1
x32/x36=0
I/O Bank 1C
42 User I/Os
x4=6
x8/x9=3
x16/x18=1
x32/x36=0
I/O Bank 6C
42 User I/Os
x4=6
x8/x9=3
x16/x18=1
x32/x36=0
EP4SE530 Devices
in the 1760-Pin FineLine BGA
I/O Bank 2C
I/O Bank 5C
42 User I/Os
x4=6
x8/x9=3
x16/x18=1
x32/x36=0
42 User I/Os
x4=6
x8/x9=3
x16/x18=1
x32/x36=0
I/O Bank 2B
I/O Bank 5B
24 User I/Os
x4=4
x8/x9=2
x16/x18=1
x32/x36=0
24 User I/Os
x4=4
x8/x9=2
x16/x18=1
x32/x36=0
I/O Bank 2A
I/O Bank 5A
50 User I/Os
x4=7
x8/x9=3
x16/x18=1
x32/x36=0
50 User I/Os
x4=7
x8/x9=3
x16/x18=1
x32/x36=0
DLL1
I/O Bank 3A
I/O Bank 3B
I/O Bank 3C
I/O Bank 4C
I/O Bank 4B
I/O Bank 4A
48 User I/Os
x4=8
x8/x9=4
x16/x18=2
x32/x36=1
48 User I/Os
x4=8
x8/x9=4
x16/x18=2
x32/x36=1
32 User I/Os
x4=3
x8/x9=1
x16/x18=0
x32/x36=0
32 User I/Os
x4=3
x8/x9=1
x16/x18=0
x32/x36=0
48 User I/Os
x4=8
x8/x9=4
x16/x18=2
x32/x36=1
48 User I/Os
x4=8
x8/x9=4
x16/x18=2
x32/x36=1
DLL2
Notes to Figure 7–16:
(1) These numbers are preliminary until the devices are available.
(2) You can also use DQS/DQSn pins in some of the ×4 groups as RUP and RDN pins, but you cannot use a ×4 group for memory interfaces if two pins
of the ×4 group are used as RUP and RDN pins for OCT calibration. If two pins of a ×4 group are used as RUP and RDN pins for OCT calibration, you
can use the ×16/×18 or ×32/×36 groups that include that ×4 group, however there are restrictions on using ×8/×9 groups that include that ×4
group.
(3) All I/O pin counts include dedicated clock inputs and dedicated corner PLL clock inputs that you can use for data inputs.
(4) You can also use some of the DQS/DQ pins in I/O Bank 1C as configuration pins. You cannot use a ×4 DQS/DQ group with any of its pin members
used for configuration purposes. Ensure that the DQS/DQ groups that you have chosen are not also used for configuration because you may lose
up to four ×4 DQS/DQ groups, depending on your configuration scheme.
February 2011
Altera Corporation
Stratix IV Device Handbook
Volume 1
7–22
Chapter 7: External Memory Interfaces in Stratix IV Devices
Memory Interfaces Pin Support
Figure 7–17. Number of DQS/DQ Groups per Bank in EP4SE820 Devices in the 1760-pin FineLine BGA Package
(1), (2), (3),
(4)
DLL0
I/O Bank 8A
48 User I/Os
x4=8
x8/x9=4
x16/x18=2
x32/x36=1
I/O Bank 8B
48 User I/Os
x4=8
x8/x9=4
x16/x18=2
x32/x36=1
I/O Bank 8C
48 User I/Os
x4=6
x8/x9=3
x16/x18=1
x32/x36=0
I/O Bank 7C
48 User I/Os
x4=6
x8/x9=3
x16/x18=1
x32/x36=0
I/O Bank 7B
48 User I/Os
x4=8
x8/x9=4
x16/x18=2
x32/x36=1
I/O Bank 7A
48 User I/Os
x4=8
x8/x9=4
x16/x18=2
x32/x36=1
DLL3
I/O Bank 1A
I/O Bank 6A
50 User I/Os
x4=7
x8/x9=3
x16/x18=1
x32/x36=0
50 User I/Os
x4=7
x8/x9=3
x16/x18=1
x32/x36=0
I/O Bank 6B
36 User I/Os
x4=6
x8/x9=3
x16/x18=1
x32/x36=0
I/O Bank 1B
36 User I/Os
x4=6
x8/x9=3
x16/x18=1
x32/x36=0
I/O Bank 1C
50 User I/Os
x4=7
x8/x9=3
x16/x18=1
x32/x36=0
I/O Bank 6C
50 User I/Os
x4=7
x8/x9=3
x16/x18=1
x32/x36=0
EP4SE820 Devices
in the 1760-Pin FineLine BGA
I/O Bank 2C
I/O Bank 5C
50 User I/Os
x4=7
x8/x9=3
x16/x18=1
x32/x36=0
50 User I/Os
x4=7
x8/x9=3
x16/x18=1
x32/x36=0
I/O Bank 2B
I/O Bank 5B
36 User I/Os
x4=6
x8/x9=3
x16/x18=1
x32/x36=0
36 User I/Os
x4=6
x8/x9=3
x16/x18=1
x32/x36=0
I/O Bank 2A
I/O Bank 5A
50 User I/Os
x4=7
x8/x9=3
x16/x18=1
x32/x36=0
50 User I/Os
x4=7
x8/x9=3
x16/x18=1
x32/x36=0
DLL1
I/O Bank 3A
I/O Bank 3B
I/O Bank 3C
I/O Bank 4C
I/O Bank 4B
I/O Bank 4A
48 User I/Os
x4=8
x8/x9=4
x16/x18=2
x32/x36=1
48 User I/Os
x4=8
x8/x9=4
x16/x18=2
x32/x36=1
48 User I/Os
x4=6
x8/x9=3
x16/x18=1
x32/x36=0
48 User I/Os
x4=6
x8/x9=3
x16/x18=1
x32/x36=0
48 User I/Os
x4=8
x8/x9=4
x16/x18=2
x32/x36=1
48 User I/Os
x4=8
x8/x9=4
x16/x18=2
x32/x36=1
DLL2
Notes to Figure 7–17:
(1) These numbers are preliminary until the devices are available.
(2) You can also use DQS/DQSn pins in some of the ×4 groups as RUP and RDN pins, but you cannot use a ×4 group for memory interfaces if two pins
of the ×4 group are used as RUP and RDN pins for OCT calibration. If two pins of a ×4 group are used as RUP and RDN pins for OCT calibration, you
can use the ×16/×18 or ×32/×36 groups that include that ×4 group, however there are restrictions on using ×8/×9 groups that include that ×4
group.
(3) All I/O pin counts include dedicated clock inputs and dedicated corner PLL clock inputs that you can use for data inputs.
(4) You can also use some of the DQS/DQ pins in I/O Bank 1C as configuration pins. You cannot use a ×4 DQS/DQ group with any of its pin members
used for configuration purposes. Ensure that the DQS/DQ groups that you have chosen are not also used for configuration because you may lose
up to four ×4 DQS/DQ groups, depending on your configuration scheme.
Stratix IV Device Handbook
Volume 1
February 2011 Altera Corporation
Chapter 7: External Memory Interfaces in Stratix IV Devices
Memory Interfaces Pin Support
7–23
Figure 7–18. Number of DQS/DQ Groups per Bank in EP4SGX290, EP4SGX360, and EP4SGX530 Devices in the 1932-Pin
FineLine BGA Package (1), (2), (3), (4)
DLL0
I/O Bank 8A
48 User I/Os
x4=8
x8/x9=4
x16/x18=2
x32/x36=1
I/O Bank 8B
48 User I/Os
x4=8
x8/x9=4
x16/x18=2
x32/x36=1
I/O Bank 8C
32 User I/Os
x4=3
x8/x9=1
x16/x18=0
x32/x36=0
I/O Bank 7C
32 User I/Os
x4=3
x8/x9=1
x16/x18=0
x32/x36=0
I/O Bank 7B
48 User I/Os
x4=8
x8/x9=4
x16/x18=2
x32/x36=1
I/O Bank 7A
48 User I/Os
x4=8
x8/x9=4
x16/x18=2
x32/x36=1
I/O Bank 1A
DLL3
I/O Bank 6A
50 User I/Os
x4=7
x8/x9=3
x16/x18=1
x32/x36=0
50 User I/Os
x4=7
x8/x9=3
x16/x18=1
x32/x36=0
I/O Bank 6C
42 User I/Os
x4=6
x8/x9=3
x16/x18=1
x32/x36=0
I/O Bank 1C
42 User I/Os
x4=6
x8/x9=3
x16/x18=1
x32/x36=0
I/O Bank 2C
I/O Bank 5C
42 User I/Os
x4=6
x8/x9=3
x16/x18=1
x32/x36=0
42 User I/Os
x4=6
x8/x9=3
x16/x18=1
x32/x36=0
EP4SGX290, EP4SGX360, and EP4SGX530 Devices
in the 1932-Pin FineLine BGA
I/O Bank 2B
I/O Bank 5B
20 User I/Os
x4=3
x8/x9=1
x16/x18=0
x32/x36=0
20 User I/Os
x4=3
x8/x9=1
x16/x18=0
x32/x36=0
I/O Bank 2A
I/O Bank 5A
50 User I/Os
x4=7
x8/x9=3
x16/x18=1
x32/x36=0
50 User I/Os
x4=7
x8/x9=3
x16/x18=1
x32/x36=0
DLL1
I/O Bank 3A
I/O Bank 3B
I/O Bank 3C
I/O Bank 4C
I/O Bank 4B
I/O Bank 4A
48 User I/Os
x4=8
x8/x9=4
x16/x18=2
x32/x36=1
48 User I/Os
x4=8
x8/x9=4
x16/x18=2
x32/x36=1
32 User I/Os
x4=3
x8/x9=1
x16/x18=0
x32/x36=0
32 User I/Os
x4=3
x8/x9=1
x16/x18=0
x32/x36=0
48 User I/Os
x4=8
x8/x9=4
x16/x18=2
x32/x36=1
48 User I/Os
x4=8
x8/x9=4
x16/x18=2
x32/x36=1
DLL2
Notes to Figure 7–18:
(1) These numbers are preliminary until the devices are available.
(2) You can also use DQS/DQSn pins in some of the ×4 groups as RUP and RDN pins, but you cannot use a ×4 group for memory interfaces if two pins
of the ×4 group are used as RUP and RDN pins for OCT calibration. If two pins of a ×4 group are used as RUP and RDN pins for OCT calibration, you
can use the ×16/×18 or ×32/×36 groups that include that ×4 group, however there are restrictions on using ×8/×9 groups that include that ×4
group.
(3) All I/O pin counts include dedicated clock inputs and dedicated corner PLL clock inputs that you can use for data inputs.
(4) You can also use some of the DQS/DQ pins in I/O Bank 1C as configuration pins. You cannot use a ×4 DQS/DQ group with any of its pin members
used for configuration purposes. Ensure that the DQS/DQ groups that you have chosen are not also used for configuration because you may lose
up to four ×4 DQS/DQ groups, depending on your configuration scheme.
February 2011
Altera Corporation
Stratix IV Device Handbook
Volume 1
7–24
Chapter 7: External Memory Interfaces in Stratix IV Devices
Memory Interfaces Pin Support
Figure 7–19. Number of DQS/DQ Groups per Bank in EP4S100G3, EP4S100G4, and EP4S100G5 Devices in the 1932-Pin
FineLine BGA Package (1), (2), (3), (4)
DLL0
I/O Bank 8A
48 User I/Os
x4=8
x8/x9=4
x16/x18=2
x32/x36=1
I/O Bank 8B
48 User I/Os
x4=8
x8/x9=4
x16/x18=2
x32/x36=1
I/O Bank 8C
32 User I/Os
x4=3
x8/x9=1
x16/x18=0
x32/x36=0
I/O Bank 7C
32 User I/Os
x4=3
x8/x9=1
x16/x18=0
x32/x36=0
I/O Bank 7B
48 User I/Os
x4=8
x8/x9=4
x16/x18=2
x32/x36=1
I/O Bank 7A
48 User I/Os
x4=8
x8/x9=4
x16/x18=2
x32/x36=1
I/O Bank 1A
DLL3
I/O Bank 6A
38 User I/Os
x4=3
x8/x9=0
x16/x18=0
x32/x36=0
40 User I/Os
x4=3
x8/x9=1
x16/x18=0
x32/x36=0
I/O Bank 6C
20 User I/Os
x4=0
x8/x9=0
x16/x18=0
x32/x36=0
I/O Bank 1C
19 User I/Os
x4=0
x8/x9=0
x16/x18=0
x32/x36=0
I/O Bank 2C
I/O Bank 5C
19 User I/Os
x4=0
x8/x9=0
x16/x18=0
x32/x36=0
17 User I/Os
x4=0
x8/x9=0
x16/x18=0
x32/x36=0
EP4S100G3, EP4S100G4, and EP4S100G5 Devices
in the 1932-Pin FineLine BGA
I/O Bank 2B
I/O Bank 5B
13 User I/Os
x4=1
x8/x9=0
x16/x18=0
x32/x36=0
12 User I/Os
x4=0
x8/x9=0
x16/x18=0
x32/x36=0
I/O Bank 2A
I/O Bank 5A
39 User I/Os
x4=4
x8/x9=1
x16/x18=0
x32/x36=0
40 User I/Os
x4=4
x8/x9=1
x16/x18=0
x32/x36=0
DLL1
I/O Bank 3A
I/O Bank 3B
I/O Bank 3C
I/O Bank 4C
I/O Bank 4B
I/O Bank 4A
48 User I/Os
x4=8
x8/x9=4
x16/x18=2
x32/x36=1
48 User I/Os
x4=8
x8/x9=4
x16/x18=2
x32/x36=1
32 User I/Os
x4=3
x8/x9=1
x16/x18=0
x32/x36=0
32 User I/Os
x4=3
x8/x9=1
x16/x18=0
x32/x36=0
48 User I/Os
x4=8
x8/x9=4
x16/x18=2
x32/x36=1
48 User I/Os
x4=8
x8/x9=4
x16/x18=2
x32/x36=1
DLL2
Notes to Figure 7–19:
(1) These numbers are preliminary until the devices are available.
(2) You can also use DQS/DQSn pins in some of the ×4 groups as RUP and RDN pins, but you cannot use a ×4 group for memory interfaces if two pins
of the ×4 group are used as RUP and RDN pins for OCT calibration. If two pins of a ×4 group are used as RUP and RDN pins for OCT calibration, you
can use the ×16/×18 or ×32/×36 groups that include that ×4 group, however there are restrictions on using ×8/×9 groups that include that ×4
group.
(3) All I/O pin counts include dedicated clock inputs and dedicated corner PLL clock inputs that you can use for data inputs.
(4) You can also use some of the DQS/DQ pins in I/O Bank 1C as configuration pins. You cannot use a ×4 DQS/DQ group with any of its pin members
used for configuration purposes. Ensure that the DQS/DQ groups that you have chosen are not also used for configuration because you may lose
up to four ×4 DQS/DQ groups, depending on your configuration scheme.
The DQS and DQSn pins are listed in the Stratix IV pin tables as DQSXY and DQSnXY,
respectively, where X indicates the DQS/DQ grouping number and Y indicates
whether the group is located on the top (T), bottom (B), left (L), or right (R) side of the
device. The DQS/DQ pin numbering is based on ×4 mode.
The corresponding DQ pins are marked as DQXY, where X indicates which DQS group
the pins belong to and Y indicates whether the group is located on the top (T), bottom
(B), left (L), or right (R) side of the device. For example, DQS1L indicates a DQS pin
located on the left side of the device. The DQ pins belonging to that group are shown
as DQ1L in the pin table. For more information, refer to Figure 7–20.
Stratix IV Device Handbook
Volume 1
February 2011 Altera Corporation
Chapter 7: External Memory Interfaces in Stratix IV Devices
Memory Interfaces Pin Support
1
7–25
The parity, DM, BWSn, NWSn, ECC, and QVLD pins are shown as DQ pins in the pin
table.
The numbering scheme starts from the top-left corner of the device going
counter-clockwise in a die-top view. Figure 7–20 shows how the DQS/DQ groups are
numbered in a die-top view of the device. The top and bottom sides of the device can
contain up to 38 ×4 DQS/DQ groups. The left and right sides of the device can contain
up to 34 ×4 DQS/DQ groups.
Figure 7–20. DQS Pins in Stratix IV I/O Banks
DQS20T
DQS38T
DQS19T
DQS1T
DLL0
DLL3
PLL_T1
PLL_T2
PLL_R1
PLL_L1
8A
8B
8C
7C
7B
7A
DQS1L
DQS34R
1A
6A
1B
6B
1C
6C
DQS17L
DQS18R
PLL_R2
PLL_L2
Stratix IV Device
PLL_R3
PLL_L3
DQS18L
DQS17R
2C
5C
2B
5B
2A
5A
DQS34L
DQS1R
3A
3B
3C
4C
4B
4A
PLL_R4
PLL_L4
PLL_B1
PLL_B2
DLL2
DLL1
DQS1B
February 2011
Altera Corporation
DQS19B
DQS20B
DQS38B
Stratix IV Device Handbook
Volume 1
7–26
Chapter 7: External Memory Interfaces in Stratix IV Devices
Memory Interfaces Pin Support
Using the RUP and RDN Pins in a DQS/DQ Group Used for Memory Interfaces
You can use the DQS/DQSn pins in some of the ×4 groups as RUP and RDN pins (listed
in the pin table). You cannot use a ×4 DQS/DQ group for memory interfaces if any of
its pin members are used as RUP and RDN pins for OCT calibration. You may be able to
use the ×8/×9 group that includes this ×4 DQS/DQ group, if either of the following
applies:
■
You are not using DM pins with your differential DQS pins
■
You are not using complementary or differential DQS pins
You can use the ×8/×9 group because a DQS/DQ ×8/×9 group actually comprises 12
pins, as the groups are formed by stitching two DQS/DQ groups in ×4 mode with six
pins each (refer to Table 7–1 on page 7–5). A typical ×8 memory interface consists of
one DQS, one DM, and eight DQ pins that add up to 10 pins. If you choose your pin
assignment carefully, you can use the two extra pins for RUP and RDN. In a DDR3
SDRAM interface, you must use differential DQS, which means that you only have
one extra pin. In this case, pick different pin locations for the RUP and RDN pins (for
example, in the bank that contains the address and command pins).
You cannot use the RUP and RDN pins shared with DQS/DQ group pins when using
×9 QDR II+/QDR II SRAM devices, as the RUP and RDN pins are dual purpose with
the CQn pins. In this case, pick different pin locations for RUP and RDN pins to avoid
conflict with memory interface pin placement. In this case, you have the choice of
placing the RUP and RDN pins in the data-write group or in the same bank as the
address and command pins.
There is no restriction on using ×16/×18 or ×32/×36 DQS/DQ groups that include the
×4 groups whose pins are being used as RUP and RDN pins, because there are enough
extra pins that can be used as DQS pins.
1
For ×8, ×16/×18, or ×32/×36 DQS/DQ groups whose members are used for RUP and
RDN, you must assign DQS and DQ pins manually. The Quartus® II software might
not be able to place DQS and DQ pins without manual pin assignments, resulting in a
“no-fit”.
Combining ×16/×18 DQS/DQ Groups for a ×36 QDR II+/QDR II SRAM Interface
This implementation combines ×16/×18 DQS/DQ groups to interface with a ×36
QDR II+/QDR II SRAM device. The ×36 read data bus uses two ×16/×18 groups
while the ×36 write data uses another two ×16/×18 or four ×8/×9 groups. The
CQ/CQn signal traces are split on the board trace to connect to two pairs of CQ/CQn
pins in the FPGA. This is the only connection on the board that you need to change for
this implementation. Other QDR II+/QDR II SRAM interface rules for Stratix IV
devices also apply for this implementation.
1
The ALTMEMPHY megafunction and UniPHY-based external memory interface IPs
do not use the QVLD signal, so you can leave the QVLD signal unconnected as in any
QDR II+/QDR II SRAM interface.
f For more information about the ALTMEMPHY megafunction or UniPHY-based IPs,
refer to the External Memory Interface Handbook.
Stratix IV Device Handbook
Volume 1
February 2011 Altera Corporation
Chapter 7: External Memory Interfaces in Stratix IV Devices
Memory Interfaces Pin Support
7–27
Rules to Combine Groups
In 780-, 1152-, and some 1517-pin package devices, there is at most one ×16/×18 group
per I/O sub-bank. You can combine two ×16/×18 groups from a single side of the
device for a ×36 interface.
For devices that do not have four ×16/×18 groups in a single side of the device to
form two ×36 groups for read and write data, you can form one ×36 group on one side
of the device and another ×36 group on the other side of the device.
For vertical migration with the ×36 emulation implementation, check if migration is
possible by enabling device migration in the Quartus II project. The Quartus II
software supports the use of four ×8/×9 DQ groups for write data pins and migration
of these groups across device density. Table 7–3 lists the possible combinations to use
two ×16/×18 DQS/DQ groups to form a ×32/×36 group on Stratix IV devices lacking
a native ×32/×36 DQS/DQ group.
Table 7–3. Possible Group Combinations in Stratix IV Devices (Part 1 of 2)
Package
780-Pin
FineLine BGA
1152-Pin
FineLine BGA
February 2011
Altera Corporation
Device Density
■
EP4SGX70
■
EP4SGX110
■
EP4SGX180
■
EP4SGX230
■
EP4SGX290
■
EP4SGX360
■
EP4SE230
■
EP4SE360
■
EP4SGX70
■
EP4SGX110
■
EP4SGX180
■
EP4SGX230
■
EP4SGX290
(2)
■
EP4SGX360
(2)
■
EP4SGX530
(2)
■
EP4SE360
■
EP4SE530
■
EP4SE820
I/O Sub-Bank Combinations
3A and 4A, 7A and 8A (bottom and top I/O banks)
1A and 2A, 5A and 6A (left and right I/O banks)
3A and 4A, 7A and 8A (bottom and top I/O banks)
3A and 4A, 7A and 8A (bottom and top I/O banks)
(1)
(1)
(1)
1A and 1C, 6A and 6C (left and right I/O banks)
3A and 3B, 4A and 4B (bottom I/O banks)
7A and 7B, 8A and 8B (top I/O banks)
1A and 1C, 2A and 2C (left I/O banks)
3A and 3B, 4A and 4B (bottom I/O banks)
5A and 5C, 6A and 6C (right I/O banks)
7A and 7B, 8A and 8B (top I/O banks)
Stratix IV Device Handbook
Volume 1
7–28
Chapter 7: External Memory Interfaces in Stratix IV Devices
Memory Interfaces Pin Support
Table 7–3. Possible Group Combinations in Stratix IV Devices (Part 2 of 2)
Package
1517-Pin
FineLine BGA
1760-Pin
FineLine BGA
1932-Pin
FineLine BGA
Device Density
■
EP4SGX180
■
EP4SGX230
■
EP4SGX290
(2)
■
EP4SGX360
(2)
■
EP4SGX530
(2)
EP4SE530
(2)
■
EP4SE820
(2)
■
EP4S40G2
■
EP4S40G5
■
EP4S100G2
■
EP4S100G5
■
EP4SGX290
■
EP4SGX360
■
EP4SGX530
■
EP4SE530
(2)
■
EP4SE820
(2)
■
EP4SGX290
(2)
■
EP4SGX360
(2)
■
EP4SGX530
(2)
■
I/O Sub-Bank Combinations
1A and 1C, 2A and 2C (left I/O banks)
3A and 3B, 4A and 4B (bottom I/O banks)
5A and 5C, 6A and 6C (right I/O banks)
7A and 7B, 8A and 8B (top I/O banks)
1A and 1B, 2A and 2B or 1B and 1C, 2B and 2C (left I/O
banks) (3)
5A and 5B, 6A and 6B or 5B and 5C, 6B and 6C (right I/O
banks) (3)
3A and 3B, 4A and 4B (bottom I/O banks)
7A and 7B, 8A and 8B (top I/O banks)
1A and 1C, 2A and 2C (left I/O banks)
3A and 3B, 4A and 4B (bottom I/O banks)
5A and 5C, 6A and 6C (right I/O banks)
7A and 7B, 8A and 8B (top I/O banks)
1A and 1B, 2A and 2B or 1B and 1C, 2B and 2C (left I/O
banks) (3)
5A and 5B, 6A and 6B or 5B and 5C, 6B and 6C (right I/O
banks) (3)
1A and 1C, 2A and 2C (left I/O banks)
5A and 5C, 6A and 6C (right I/O banks)
Notes to Table 7–3:
(1) Each side of the device in these packages has four remaining ×8/×9 groups. You can combine them for the write
side (only) if you want to keep the ×36 QDR II+/QDR II SRAM interface on one side of the device. You must change
the Memory Interface Data Group default assignment from the default 18 to 9 in this case.
(2) This device supports ×36 DQS/DQ groups on the top and bottom I/O banks natively.
(3) Although it is possible to combine the ×16/×18 DQS/DQ groups from I/O banks 1A and 1C, 2A and 2C, 5A and 5C,
and 6A and 6C, Altera does not recommend this due to the size of the package. Similarly, crossing a bank number
(for example, combining groups from I/O banks 6C and 5C) is not supported in this package.
Stratix IV Device Handbook
Volume 1
February 2011 Altera Corporation
Chapter 7: External Memory Interfaces in Stratix IV Devices
Stratix IV External Memory Interface Features
7–29
Stratix IV External Memory Interface Features
Stratix IV devices are rich with features that allow robust high-performance external
memory interfacing. The ALTMEMPHY megafunction allows you to use these
external memory interface features and helps set up the physical interface (PHY) best
suited for your system. This section describes each Stratix IV device feature that is
used in external memory interfaces from the DQS phase-shift circuitry, DQS logic
block, leveling multiplexers, and dynamic OCT control block.
1
The ALTMEMPHY megafunction and the Altera memory controller MegaCore®
functions can run at half the frequency of the I/O interface of the memory devices to
allow better timing management in high-speed memory interfaces. Stratix IV devices
have built-in registers in the IOE to convert data from full-rate (the I/O frequency) to
half-rate (the controller frequency) and vice versa. You can bypass these registers if
your memory controller is not running at half the rate of the I/O frequency. When
using the Altera memory controller MegaCore functions, the ALTMEMPHY
megafunction is instantiated for you.
f For more information about the ALTMEMPHY megafunction, refer to the External
Memory PHY Interface (ALTMEMPHY) (nonAFI) Megafunction User Guide.
DQS Phase-Shift Circuitry
Stratix IV phase-shift circuitry provides phase shift to the DQS/CQ and CQn pins on
read transactions when the DQS/CQ and CQn pins are acting as input clocks or
strobes to the FPGA. The DQS phase-shift circuitry consists of DLLs that are shared
between multiple DQS pins and the phase-offset module to further fine-tune the DQS
phase shift for different sides of the device.
February 2011
Altera Corporation
Stratix IV Device Handbook
Volume 1
7–30
Chapter 7: External Memory Interfaces in Stratix IV Devices
Stratix IV External Memory Interface Features
Figure 7–21 shows how the DQS phase-shift circuitry is connected to the DQS/CQ
and CQn pins in the device where memory interfaces are supported on all sides of the
Stratix IV device.
Figure 7–21. DQS/CQ and CQn Pins and DQS Phase-Shift Circuitry
DQS/CQ
Pin
CQn
Pin
DLL
Reference
Clock
(1), (2)
DQS/CQ
Pin
CQn
Pin
DLL
Reference
Clock
DQS Logic
Blocks
Δt
DQS
Phase-Shift
Circuitry
Δt
to IOE
to IOE
Δt
Δt
to IOE
to IOE
DQS
Phase-Shift
Circuitry
DQS Logic
Blocks
DQS/CQ
Pin
CQn
Pin
DQS/CQ
Pin
CQn
Pin
Δt
to
IOE
Δt
to
IOE
Δt
to
IOE
Δt
to
IOE
DQS
Phase-Shift
Circuitry
to IOE
to IOE
to IOE
to IOE
Δt
Δt
Δt
Δt
to
IOE
Δt
CQn
Pin
to
IOE
Δt
DQS/CQ
Pin
to
IOE
Δt
CQn
Pin
to
IOE
Δt
DQS/CQ
Pin
DQS
Phase-Shift
Circuitry
DLL
Reference
Clock
DLL
Reference
Clock
CQn
Pin
DQS/CQ
Pin
CQn
Pin
DQS/CQ
Pin
Notes to Figure 7–21:
(1) For possible reference input clock pins for each DLL, refer to “DLL” on page 7–31.
(2) You can configure each DQS/CQ and CQn pin with a phase shift based on one of two possible DLL output settings.
DQS phase-shift circuitry is connected to the DQS logic blocks that control each
DQS/CQ or CQn pin. The DQS logic blocks allow the DQS delay settings to be
updated concurrently at every DQS/CQ or CQn pin.
Stratix IV Device Handbook
Volume 1
February 2011 Altera Corporation
Chapter 7: External Memory Interfaces in Stratix IV Devices
Stratix IV External Memory Interface Features
7–31
DLL
DQS phase-shift circuitry uses a DLL to dynamically control the clock delay needed
by the DQS/CQ and CQn pin. The DLL, in turn, uses a frequency reference to
dynamically generate control signals for the delay chains in each of the DQS/CQ and
CQn pins, allowing it to compensate for PVT variations. The DQS delay settings are
Gray-coded to reduce jitter when the DLL updates the settings. The phase-shift
circuitry needs 1,280 clock cycles to lock and calculate the correct input clock period
when the DLL is in low jitter mode. Otherwise, only 256 clock cycles are needed. Do
not send data during these clock cycles because there is no guarantee that it will be
captured properly. As the settings from the DLL may not be stable until this lock
period has elapsed, be aware that anything using these settings (including the
leveling delay system) may be unstable during this period.
1
You can still use the DQS phase-shift circuitry for any memory interfaces that are less
than 100 MHz. However, the DQS signal may not shift over 2.5 ns. Even if the DQS
signal is not shifted exactly to the middle of the DQ valid window, the I/O element
should still be able to capture the data in low-frequency applications in which a large
amount of timing margin is available.
There are a maximum of four DLLs in a Stratix IV device, located in each corner of the
device. These four DLLs support a maximum of four unique frequencies, with each
DLL running at one frequency. Each DLL can have two outputs with different phase
offsets, which allows one Stratix IV device to have eight different DLL phase shift
settings.
February 2011
Altera Corporation
Stratix IV Device Handbook
Volume 1
7–32
Chapter 7: External Memory Interfaces in Stratix IV Devices
Stratix IV External Memory Interface Features
Figure 7–22 shows the DLL and I/O bank locations in Stratix IV devices from a
die-top view if all sides of the device support external memory interfaces.
Figure 7–22. Stratix IV DLL and I/O Bank Locations (Die-Top View)
PLL_L1
8A
8B
8C
PLL_T1
PLL_T2
7C
7B
PLL_R1
7A
6
6
DLL0
DLL3
6
6
1A
6A
1B
6B
1C
6C
PLL_R2
PLL_L2
Stratix IV FPGA
PLL_L3
PLL_R3
2C
5C
2B
5B
5A
2A
6
6
DLL1
6
DLL2
6
PLL_L4
3A
3B
3C
PLL_B1
PLL_B2
4C
4B
4A
PLL_R4
The DLL can access the two adjacent sides from its location within the device. For
example, DLL0 on the top left of the device can access the top side (I/O banks 7A, 7B,
7C, 8A, 8B, and 8C) and the left side of the device (I/O banks 1A, 1B, 1C, 2A, 2B, and
2C). This means that each I/O bank is accessible by two DLLs, giving more flexibility
to create multiple frequencies and multiple-type interfaces. You can have two
different interfaces with the same frequency on the two sides adjacent to a DLL, where
the DLL controls the DQS delay settings for both interfaces.
Each bank can use settings from either or both DLLs the bank is adjacent to. For
example, DQS1L can get its phase-shift settings from DLL0, while DQS2L can get its
phase-shift settings from DLL1. Table 7–4 lists the DLL location and supported I/O
banks for Stratix IV devices.
Stratix IV Device Handbook
Volume 1
February 2011 Altera Corporation
Chapter 7: External Memory Interfaces in Stratix IV Devices
Stratix IV External Memory Interface Features
1
7–33
You can only have one memory interface in each I/O sub-bank (such as I/O
sub-banks 1A, 1B, and 1C) when you use leveling delay chains. This is because there
is only one leveling delay chain per I/O sub-bank.
Table 7–4. DLL Location and Supported I/O Banks
DLL
Location
Accessible I/O Banks
(1)
DLL0
Top-left corner
1A, 1B, 1C, 2A, 2B, 2C, 7A, 7B, 7C, 8A, 8B, 8C
DLL1
Bottom-left corner
1A, 1B, 1C, 2A, 2B, 2C, 3A, 3B, 3C, 4A, 4B, 4C
DLL2
Bottom-right corner
3A, 3B, 3C, 4A, 4B, 4C, 5A, 5B, 5C, 6A, 6B, 6C
DLL3
Top-right corner
5A, 5B, 5C, 6A, 6B, 6C, 7A, 7B, 7C, 8A, 8B, 8C
Note to Table 7–4:
(1) The DLL can access these I/O banks if they are available for memory interfacing.
The reference clock for each DLL may come from PLL output clocks or any of the two
dedicated clock input pins located in either side of the DLL. Table 7–5 through
Table 7–17 lists the available DLL reference clock input resources for the Stratix IV
device family.
1
When you have a dedicated PLL that only generates the DLL input reference clock, set
the PLL mode to No Compensation to achieve better performance or the Quartus II
software changes it automatically. Because the PLL does not use any other outputs, it
does not need to compensate for any clock paths.
Table 7–5. DLL Reference Clock Input for EP4SGX70, EP4SGX110, EP4SGX180, and EP4SGX230 Devices in the 780-Pin
FineLine BGA Package
DLL
DLL0
DLL1
CLKIN
(Top/Bottom)
CLKIN
(Left/Right)
PLL
(Top/Bottom)
PLL
(Left/Right)
PLL
(Corner)
CLK12P
CLK0P
CLK13P
CLK1P
CLK14P
CLK2P
PLL_T1
PLL_L2
—
CLK15P
CLK3P
CLK4P
CLK0P
CLK5P
CLK1P
CLK6P
CLK2P
PLL_B1
—
—
CLK7P
CLK3P
—
PLL_B1
—
—
—
PLL_T1
—
—
CLK4P
DLL2
CLK5P
CLK6P
CLK7P
CLK12P
DLL3
CLK13P
CLK14P
CLK15P
February 2011
Altera Corporation
Stratix IV Device Handbook
Volume 1
7–34
Chapter 7: External Memory Interfaces in Stratix IV Devices
Stratix IV External Memory Interface Features
Table 7–6. DLL Reference Clock Input for EP4SE230 and EP4SE360 Devices in the 780-Pin FineLine BGA Package
DLL
DLL0
DLL1
DLL2
DLL3
CLKIN
(Top/Bottom)
CLKIN
(Left/Right)
CLK12P
CLK0P
CLK13P
CLK1P
CLK14P
CLK2P
CLK15P
CLK3P
CLK4P
CLK0P
CLK5P
CLK1P
CLK6P
CLK2P
CLK7P
CLK3P
CLK4P
CLK8P
CLK5P
CLK9P
CLK6P
CLK10P
CLK7P
CLK11P
CLK12P
CLK8P
CLK13P
CLK9P
CLK14P
CLK10P
CLK15P
CLK11P
PLL
(Top/Bottom)
PLL
(Left/Right)
PLL
(Corner)
PLL_T1
PLL_L2
—
PLL_B1
PLL_L2
—
PLL_B1
PLL_R2
—
PLL_T1
PLL_R2
—
Table 7–7. DLL Reference Clock Input for EP4SGX290 and EP4SGX360 Devices in the 780-Pin FineLine BGA Package
DLL
CLKIN (Top/Bottom)
CLKIN
(Left/Right)
PLL (Top/Bottom)
PLL
(Left/Right)
PLL
(Corner)
—
PLL_T1
—
—
—
PLL_B1
—
—
—
PLL_B2
—
—
—
PLL_T2
—
—
CLK12P
DLL0
CLK13P
CLK14P
CLK15P
CLK4P
DLL1
CLK5P
CLK6P
CLK7P
CLK4P
DLL2
CLK5P
CLK6P
CLK7P
CLK12P
DLL3
CLK13P
CLK14P
CLK15P
Stratix IV Device Handbook
Volume 1
February 2011 Altera Corporation
Chapter 7: External Memory Interfaces in Stratix IV Devices
Stratix IV External Memory Interface Features
7–35
Table 7–8. DLL Reference Clock Input for EP4SGX70 and EP4SGX110 Devices in the 1152-Pin FineLine BGA Package
(with 24 Transceivers)
CLKIN
(Top/Bottom)
CLKIN
(Left/Right)
PLL
(Top/Bottom)
PLL
(Left/Right)
PLL
(Corner)
DLL0
CLK12P
CLK13P
CLK14P
CLK15P
CLK0P
CLK1P
CLK2P
CLK3P
PLL_T1
PLL_L2
—
DLL1
CLK4P
CLK5P
CLK6P
CLK7P
CLK0P
CLK1P
CLK2P
CLK3P
PLL_B1
PLL_L2
—
DLL2
CLK4P
CLK5P
CLK6P
CLK7P
CLK8P
CLK9P
CLK10P
CLK11P
PLL_B1
PLL_R2
—
DLL3
CLK12P
CLK13P
CLK14P
CLK15P
CLK8P
CLK9P
CLK10P
CLK11P
PLL_T1
PLL_R2
—
DLL
Table 7–9. DLL Reference Clock Input for EP4SGX110 Devices in the 1152-Pin FineLine BGA Package (with 16
Transceivers)
DLL
CLKIN
(Top/Bottom)
CLKIN
(Left/Right)
PLL
(Top/Bottom)
PLL
(Left/Right)
PLL
(Corner)
PLL_T1
PLL_L2
—
PLL_B1
—
—
PLL_B1
—
—
PLL_T1
PLL_R2
—
CLK12P
DLL0
CLK13P
CLK0P
CLK14P
CLK1P
CLK15P
CLK4P
DLL1
CLK5P
CLK0P
CLK6P
CLK1P
CLK7P
CLK4P
DLL2
CLK5P
CLK10P
CLK6P
CLK11P
CLK7P
CLK12P
DLL3
CLK13P
CLK10P
CLK14P
CLK11P
CLK15P
February 2011
Altera Corporation
Stratix IV Device Handbook
Volume 1
7–36
Chapter 7: External Memory Interfaces in Stratix IV Devices
Stratix IV External Memory Interface Features
Table 7–10. DLL Reference Clock Input for EP4SGX180, EP4SGX230, EP4SGX290, EP4SGX360, and EP4SGX530 Devices
in the 1152-Pin FineLine BGA Package
DLL
CLKIN
(Top/Bottom)
CLKIN
(Left/Right)
PLL
(Top/Bottom)
PLL
(Left/Right)
PLL
(Corner)
PLL_T1
PLL_L2
—
PLL_B1
—
—
PLL_B2
—
—
PLL_T2
PLL_R2
—
CLK12P
DLL0
CLK13P
CLK0P
CLK14P
CLK1P
CLK15P
CLK4P
DLL1
CLK5P
CLK0P
CLK6P
CLK1P
CLK7P
CLK4P
DLL2
CLK5P
CLK10P
CLK6P
CLK11P
CLK7P
CLK12P
DLL3
CLK13P
CLK10P
CLK14P
CLK11P
CLK15P
Table 7–11. DLL Reference Clock Input for EP4SE360, EP4SE530, and EP4SE820 Devices in the 1152-Pin FineLine BGA
Packages
DLL
DLL0
DLL1
DLL2
DLL3
CLKIN
(Top/Bottom)
CLKIN
(Left/Right)
CLK12P
CLK0P
CLK13P
CLK1P
CLK14P
CLK2P
CLK15P
CLK3P
CLK4P
CLK0P
CLK5P
CLK1P
CLK6P
CLK2P
CLK7P
CLK3P
CLK4P
CLK8P
CLK5P
CLK9P
CLK6P
CLK10P
CLK7P
CLK11P
CLK12P
CLK8P
CLK13P
CLK9P
CLK14P
CLK10P
CLK15P
CLK11P
Stratix IV Device Handbook
Volume 1
PLL
(Top/Bottom)
PLL
(Left/Right)
PLL
(Corner)
PLL_T1
PLL_L2
—
PLL_B1
PLL_L3
—
PLL_B2
PLL_R3
—
PLL_T2
PLL_R2
—
February 2011 Altera Corporation
Chapter 7: External Memory Interfaces in Stratix IV Devices
Stratix IV External Memory Interface Features
7–37
Table 7–12. DLL Reference Clock Input for EP4SE530 and EP4SE820 Devices in the 1517- and 1760-Pin FineLine BGA
Packages
DLL
DLL0
DLL1
DLL2
DLL3
CLKIN
(Top/Bottom)
CLKIN
(Left/Right)
CLK12P
CLK0P
CLK13P
CLK1P
CLK14P
CLK2P
CLK15P
CLK3P
CLK4P
CLK0P
CLK5P
CLK1P
CLK6P
CLK2P
CLK7P
CLK3P
CLK4P
CLK8P
CLK5P
CLK9P
CLK6P
CLK10P
CLK7P
CLK11P
CLK12P
CLK8P
CLK13P
CLK9P
CLK14P
CLK10P
CLK15P
CLK11P
PLL
(Top/Bottom)
PLL
(Left/Right)
PLL
(Corner)
PLL_T1
PLL_L2
PLL_L1
PLL_B1
PLL_L3
PLL_L4
PLL_B2
PLL_R3
PLL_R4
PLL_T2
PLL_R2
PLL_R1
Table 7–13. DLL Reference Clock Input for EP4SGX180, EP4SGX230, EP4SGX290, EP4SGX360, and EP4SGX530 Devices
in the 1517-Pin FineLine BGA Package
DLL
DLL0
DLL1
DLL2
DLL3
February 2011
CLKIN
(Top/Bottom)
CLKIN
(Left/Right)
CLK12P
CLK0P
CLK13P
CLK1P
CLK14P
CLK2P
CLK15P
CLK3P
CLK4P
CLK0P
CLK5P
CLK1P
CLK6P
CLK2P
CLK7P
CLK3P
CLK4P
CLK8P
CLK5P
CLK9P
CLK6P
CLK10P
CLK7P
CLK11P
CLK12P
CLK8P
CLK13P
CLK9P
CLK14P
CLK10P
CLK15P
CLK11P
Altera Corporation
PLL
(Top/Bottom)
PLL
(Left/Right)
PLL
(Corner)
PLL_T1
PLL_L2
—
PLL_B1
PLL_L3
—
PLL_B2
PLL_R3
—
PLL_T2
PLL_R2
—
Stratix IV Device Handbook
Volume 1
7–38
Chapter 7: External Memory Interfaces in Stratix IV Devices
Stratix IV External Memory Interface Features
Table 7–14. DLL Reference Clock Input for EP4S40G2, EP4S40G5, EP4S100G2, and EP4S100G5 Devices in the 1517-Pin
FineLine BGA Package
CLKIN
(Top/Bottom)
CLKIN
(Left/Right)
PLL
(Top/Bottom)
PLL
(Left/Right)
PLL
(Corner)
DLL0
CLK12P
CLK13P
CLK14P
CLK15P
CLK1P
CLK3P
PLL_T1
PLL_L2
—
DLL1
CLK4P
CLK5P
CLK6P
CLK7P
CLK1P
CLK3P
PLL_B1
PLL_L3
—
DLL2
CLK4P
CLK5P
CLK6P
CLK7P
CLK8P
CLK10P
PLL_B2
PLL_R3
—
DLL3
CLK12P
CLK13P
CLK14P
CLK15P
CLK8P
CLK10P
PLL_T2
PLL_R2
—
DLL
Table 7–15. DLL Reference Clock Input for EP4SGX290, EP4SGX360, and EP4SGX530 Devices in the 1760-Pin FineLine
BGA Package
CLKIN
(Top/Bottom)
CLKIN
(Left/Right)
PLL
(Top/Bottom)
PLL
(Left/Right)
PLL
(Corner)
DLL0
CLK12P
CLK13P
CLK14P
CLK15P
CLK0P
CLK1P
CLK2P
CLK3P
PLL_T1
PLL_L2
—
DLL1
CLK4P
CLK5P
CLK6P
CLK7P
CLK0P
CLK1P
CLK2P
CLK3P
PLL_B1
PLL_L3
—
DLL2
CLK4P
CLK5P
CLK6P
CLK7P
CLK8P
CLK9P
CLK10P
CLK11P
PLL_B2
PLL_R3
—
DLL3
CLK12P
CLK13P
CLK14P
CLK15P
CLK8P
CLK9P
CLK10P
CLK11P
PLL_T2
PLL_R2
—
DLL
Stratix IV Device Handbook
Volume 1
February 2011 Altera Corporation
Chapter 7: External Memory Interfaces in Stratix IV Devices
Stratix IV External Memory Interface Features
7–39
Table 7–16. DLL Reference Clock Input for EP4SGX290, EP4SGX360, and EP4SGX530 Devices in the 1932-Pin FineLine
BGA Package
DLL
DLL0
DLL1
DLL2
DLL3
CLKIN
(Top/Bottom)
CLKIN
(Left/Right)
CLK12P
CLK0P
CLK13P
CLK1P
CLK14P
CLK2P
CLK15P
CLK3P
CLK4P
CLK0P
CLK5P
CLK1P
CLK6P
CLK2P
CLK7P
CLK3P
CLK4P
CLK8P
CLK5P
CLK9P
CLK6P
CLK10P
CLK7P
CLK11P
CLK12P
CLK8P
CLK13P
CLK9P
CLK14P
CLK10P
CLK15P
CLK11P
PLL
(Top/Bottom)
PLL
(Left/Right)
PLL
(Corner)
PLL_T1
PLL_L2
PLL_L1
PLL_B1
PLL_L3
PLL_L4
PLL_B2
PLL_R3
PLL_R4
PLL_T2
PLL_R2
PLL_R1
Table 7–17. DLL Reference Clock Input for EP4S100G3, EP4S100G4, and EP4S100G5 Devices in the 1932-Pin FineLine
BGA Package
CLKIN
(Top/Bottom)
CLKIN
(Left/Right)
PLL
(Top/Bottom)
PLL
(Left/Right)
PLL
(Corner)
DLL0
CLK12P
CLK13P
CLK14P
CLK15P
—
PLL_T1
PLL_L2
PLL_L1
DLL1
CLK4P
CLK5P
CLK6P
CLK7P
—
PLL_B1
PLL_L3
PLL_L4
DLL2
CLK4P
CLK5P
CLK6P
CLK7P
CLK9P
CLK11P
PLL_B2
PLL_R3
PLL_R4
DLL3
CLK12P
CLK13P
CLK14P
CLK15P
CLK9P
CLK11P
PLL_T2
PLL_R2
PLL_R1
DLL
February 2011
Altera Corporation
Stratix IV Device Handbook
Volume 1
7–40
Chapter 7: External Memory Interfaces in Stratix IV Devices
Stratix IV External Memory Interface Features
Figure 7–23 shows a simple block diagram of the DLL. The input reference clock goes
into the DLL to a chain of up to 16 delay elements. The phase comparator compares
the signal coming out of the end of the delay chain block to the input reference clock.
The phase comparator then issues the upndn signal to the Gray-code counter. This
signal increments or decrements a six-bit delay setting (DQS delay settings) that
increases or decreases the delay through the delay element chain to bring the input
reference clock and the signals coming out of the delay element chain in phase.
Figure 7–23. Simplified Diagram of the DQS Phase-Shift Circuitry
(1)
addnsub
Phase offset settings
from the logic array
( offset [5:0] )
6
offsetdelayctrlin [5:0]
DLL
aload
Input Reference
Clock (2)
offsetdelayctrlout [5:0]
Phase
Comparator
upndninclkena
6
Phase
Offset
Control
B
offsetdelayctrlout [5:0]
offsetdelayctrlin [5:0]
6
delayctrlout [5:0]
6
6
Phase offset
settings to DQS pins
on top or bottom edge (3)
( offsetctrlout [5:0] )
addnsub
Phase offset settings
from the logic array ( offset [5:0] )
Up/Down
Counter
Delay Chains
6
(dll_offset_ctrl_a)
upndnin
clk
Phase
Offset
Control
A
6
(dll_offset_ctrl_b)
Phase offset
settings to DQS pin
on left or right edge (3)
( offsetctrlout [5:0] )
DQS Delay
Settings (4)
dqsupdate
Notes to Figure 7–23:
(1) All features of the DQS phase-shift circuitry are accessible from the ALTMEMPHY megafunction in the Quartus II software.
(2) The input reference clock for the DQS phase-shift circuitry can come from a PLL output clock or an input clock pin. For more information, refer
to Table 7–5 on page 7–33 through Table 7–17 on page 7–39.
(3) Phase offset settings can only go to the DQS logic blocks.
(4) DQS delay settings can go to the logic array, DQS logic block, and leveling circuitry.
1
In the Quartus II assignment, phase offset control block ‘A’ is designated as
DLLOFFSETCTRL_<coordinate x>_<coordinate y>_N1 and phase offset control block
‘B’ is designated as DLLOFFSETCTRL_<coordinate x>_<coordinate y>_N2.
You can reset the DLL from either the logic array or a user I/O pin. Each time the DLL
is reset, you must wait for 1,280 clock cycles for the DLL to lock before you can
capture the data properly.
Depending on the DLL frequency mode, the DLL can shift the incoming DQS signals
by 0°, 22.5°, 30°, 36°, 45°, 60°, 67.5°, 72°, 90°, 108°, 120°, 135°, 144°, 180°, or 240°. The
shifted DQS signal is then used as the clock for the DQ IOE input registers.
All DQS/CQ and CQn pins, referenced to the same DLL, can have their input signal
phase shifted by a different degree amount but all must be referenced at one
particular frequency. For example, you can have a 90° phase shift on DQS1T and a 60°
phase shift on DQS2T, referenced from a 200-MHz clock. Not all phase-shift
combinations are supported. The phase shifts on the DQS pins referenced by the same
DLL must all be a multiple of 22.5° (up to 90°), 30° (up to 120°), 36° (up to 144°), 45°
(up to 180°), or 60° (up to 240°).
Stratix IV Device Handbook
Volume 1
February 2011 Altera Corporation
Chapter 7: External Memory Interfaces in Stratix IV Devices
Stratix IV External Memory Interface Features
7–41
There are eight different frequency modes for the Stratix IV DLL, as listed in
Table 7–18. Each frequency mode provides different phase shift selections. In
frequency mode 0, 1, 2, and 3, the 6-bit DQS delay settings vary with PVT to
implement the phase-shift delay. In frequency modes 4, 5, 6, and 7, only 5 bits of the
DQS delay settings vary with PVT to implement the phase-shift delay; the most
significant bit of the DQS delay setting is set to 0.
Table 7–18. Stratix IV DLL Frequency Modes
Frequency Mode
Available Phase Shift
Number of Delay Chains
0
22.5, 45, 67.5, 90
16
1
30, 60, 90, 120
12
2
36, 72, 108, 144
10
3
45, 90, 135, 180
8
4
30, 60, 90, 120
12
5
36, 72, 108, 144
10
6
45, 90, 135, 180
8
7
60, 120, 180, 240
6
f For the frequency range of each mode, refer to the DC and Switching Characteristics for
Stratix IV Devices chapter.
For 0° shift, the DQS/CQ signal bypasses both the DLL and DQS logic blocks. The
Quartus II software automatically sets the DQ input delay chains so that the skew
between the DQ and DQS/CQ pin at the DQ IOE registers is negligible when 0° shift
is implemented. You can feed the DQS delay settings to the DQS logic block and logic
array.
The shifted DQS/CQ signal goes to the DQS bus to clock the IOE input registers of the
DQ pins. The signal can also go into the logic array for resynchronization if you are
not using IOE resynchronization registers. The shifted CQn signal can only go to the
negative-edge input register in the DQ IOE and is only used for QDR II+ and QDR II
SRAM interfaces.
Phase Offset Control
Each DLL has two phase-offset modules and can provide two separate DQS delay
settings with independent offsets, one for the top and bottom I/O bank and one for
the left and right I/O bank, so you can fine-tune the DQS phase-shift settings between
two different sides of the device. Even though you have independent phase offset
control, the frequency of the interface using the same DLL must be the same. Use the
phase offset control module for making small shifts to the input signal and use the
DQS phase-shift circuitry for larger signal shifts. For example, if the DLL only offers a
multiple of 30° phase shift, but your interface needs a 67.5° phase shift on the DQS
signal, you can use two delay chains in the DQS logic blocks to give you 60° phase
shift and use the phase offset control feature to implement the extra 7.5° phase shift.
February 2011
Altera Corporation
Stratix IV Device Handbook
Volume 1
7–42
Chapter 7: External Memory Interfaces in Stratix IV Devices
Stratix IV External Memory Interface Features
You can use either a static phase offset or a dynamic phase offset to implement the
additional phase shift. The available additional phase shift is implemented in 2’s:
complement in Gray-code between settings –64 to +63 for frequency mode 0, 1, 2, and
3, and between settings –32 to +31 for frequency modes 4, 5, 6, and 7. An additional bit
indicates whether the setting has a positive or negative value. The settings are linear,
each phase offset setting adds a delay amount specified in the DC and Switching
Characteristics for Stratix IV Devices chapter. The DQS phase shift is the sum of the DLL
delay settings and the user-selected phase offset settings whose top setting is 64 for
frequency modes 0, 1, 2, and 3; and 32 for frequency modes 4, 5, 6, and 7, so the actual
physical offset setting range is 64 or 32 subtracted by the DQS delay settings from the
DLL.
1
When using this feature, you need to monitor the DQS delay settings to know how
many offsets you can add and subtract in the system. Note that the DQS delay settings
output by the DLL are also Gray coded.
For example, if the DLL determines that DQS delay settings of 28 is needed to achieve
a 30° phase shift in DLL frequency mode 1, you can subtract up to 28 phase offset
settings and you can add up to 35 phase offset settings to achieve the optimal delay
that you need. However, if the same DQS delay settings of 28 is needed to achieve 30°
phase shift in DLL frequency mode 4, you can still subtract up to 28 phase offset
settings, but you can only add up to 3 phase offset settings before the DQS delay
settings reach their maximum settings because DLL frequency mode 4 only uses 5-bit
DLL delay settings.
f For more information about the value for each step, refer to the DC and Switching
Characteristics for Stratix IV Devices chapter.
When using static phase offset, you can specify the phase offset amount in the
ALTMEMPHY megafunction as a positive number for addition or a negative number
for subtraction. You can also have a dynamic phase offset that is always added to,
subtracted from, or both added to and subtracted from the DLL phase shift. When
you always add or subtract, you can dynamically input the phase offset amount into
the dll_offset[5..0] port. When you want to both add and subtract dynamically,
you control the addnsub signal in addition to the dll_offset[5..0] signals.
Stratix IV Device Handbook
Volume 1
February 2011 Altera Corporation
7–43
Stratix IV Device Handbook
Volume 1
DQS Logic Block
Each DQS/CQ and CQn pin is connected to a separate DQS logic block, which consists of the DQS delay chains, update enable
circuitry, and DQS postamble circuitry, as shown in Figure 7–24.
Figure 7–24. Stratix IV DQS Logic Block
DQS Delay Chain
DQS Enable
dqsenable (2)
1xx
000 dqsbusout
001
010
011
Bypass
dqsin
DQS bus
6
6
DQS Enable Control
0
1
0
1
6
D
Q
dqsupdateen
Input Reference
Clock (1)
Update
Enable
Circuitry
phasectrlin
6
<dqs_ctrl_latches_enable>
6
delayctrlin
Resynchronization
Clock
clk
4
phaseinvertctrl
0111
0110
0101
0100
0011
0010
0001
0000
Postamble
Enable
0
1
<level_dqs_enable>
postamble control clock
0
0 dqsenableout
0 1
1
1
dqsenablein
enaphasetransferreg
<delay_dqs_enable_by_half_cycle>
February 2011 Altera Corporation
Notes to Figure 7–24:
(1) The input reference clock for the DQS phase-shift circuitry can come from a PLL output clock or an input clock pin. For more information, refer to Table 7–5 on page 7–33 through Table 7–17 on page 7–39.
(2) The dqsenable signal can also come from the Stratix IV FPGA fabric.
Chapter 7: External Memory Interfaces in Stratix IV Devices
Stratix IV External Memory Interface Features
6
offsetctrlin [5:0]
6
Phase offset
1
D Q
settings from the
0
DQS phase-shift
circuitry
<dqs_offsetctrl_enable>
6
DQS delay
settings from the delayctrlin [5:0]
DQS phase-shift
circuitry
dqsbusout
phasectrlin[2:0]
dqsin
DQS/CQ or
CQn Pin
6
PRE
Q D
7–44
Chapter 7: External Memory Interfaces in Stratix IV Devices
Stratix IV External Memory Interface Features
DQS Delay Chain
DQS delay chains consist of a set of variable delay elements to allow the input
DQS/CQ and CQn signals to be shifted by the amount specified by the DQS
phase-shift circuitry or the logic array. There are four delay elements in the DQS delay
chain; the first delay chain closest to the DQS/CQ pin can be shifted either by the
DQS delay settings or by the sum of the DQS delay setting and the phase-offset
setting. The number of delay chains required is transparent because the
ALTMEMPHY megafunction automatically sets it when you choose the operating
frequency. The DQS delay settings can come from the DQS phase-shift circuitry on
either end of the I/O banks or from the logic array.
The delay elements in the DQS logic block have the same characteristics as the delay
elements in the DLL. When the DLL is not used to control the DQS delay chains, you
can input your own Gray-coded 6-bit or 5-bit settings using the
dqs_delayctrlin[5..0] signals available in the ALTMEMPHY megafunction. These
settings control 1, 2, 3, or all 4 delay elements in the DQS delay chains. The
ALTMEMPHY megafunction can also dynamically choose the number of DQS delay
chains needed for the system. The amount of delay is equal to the sum of the delay
element’s intrinsic delay and the product of the number of delay steps and the value
of the delay steps.
You can also bypass the DQS delay chain to achieve a 0° phase shift.
Update Enable Circuitry
Both the DQS delay settings and the phase-offset settings pass through a register
before going into the DQS delay chains. The registers are controlled by the update
enable circuitry to allow enough time for any changes in the DQS delay setting bits to
arrive at all the delay elements. This allows them to be adjusted at the same time. The
update enable circuitry enables the registers to allow enough time for the DQS delay
settings to travel from the DQS phase-shift circuitry or core logic to all the DQS logic
blocks before the next change. It uses the input reference clock or a user clock from the
core to generate the update enable output. The ALTMEMPHY megafunction uses this
circuit by default. Figure 7–25 shows an example waveform of the update enable
circuitry output.
Figure 7–25. DQS Update Enable Waveform
DLL Counter Update
(Every 8 cycles)
DLL Counter Update
(Every 8 cycles)
System Clock
DQS Delay Settings
(Updated every 8 cycles)
6 bit
Update Enable
Circuitry Output
Stratix IV Device Handbook
Volume 1
February 2011 Altera Corporation
Chapter 7: External Memory Interfaces in Stratix IV Devices
Stratix IV External Memory Interface Features
7–45
DQS Postamble Circuitry
For external memory interfaces that use a bidirectional read strobe such as in DDR3,
DDR2, and DDR SDRAM, the DQS signal is low before going to or coming from a
high-impedance state. The state in which DQS is low, just after a high-impedance
state, is called the preamble; the state in which DQS is low, just before it returns to a
high-impedance state, is called the postamble. There are preamble and postamble
specifications for both read and write operations in DDR3, DDR2, and DDR SDRAM.
The DQS postamble circuitry ensures that data is not lost if there is noise on the DQS
line during the end of a read operation that occurs while DQS is in a postamble state.
Stratix IV devices have dedicated postamble registers that you can control to ground
the shifted DQS signal used to clock the DQ input registers at the end of a read
operation. This ensures that any glitches on the DQS input signals during the end of a
read operation that occurs while DQS is in a postamble state do not affect the DQ IOE
registers.
In addition to the dedicated postamble register, Stratix IV devices also have an HDR
block inside the postamble enable circuitry. Use these registers if the controller is
running at half the frequency of the I/Os.
Using the HDR block as the first stage capture register in the postamble enable
circuitry block is optional. The HDR block is clocked by the half-rate
resynchronization clock, which is the output of the I/O clock divider circuit (shown in
Figure 7–31 on page 7–49). There is an AND gate after the postamble register outputs
that is used to avoid postamble glitches from a previous read burst on a
non-consecutive read burst. This scheme allows a half-a-clock cycle latency for
dqsenable assertion and zero latency for dqsenable de-assertion, as shown in
Figure 7–26.
Figure 7–26. Avoiding Glitch on a Non-Consecutive Read Burst Waveform
Postamble glitch
Postamble
Preamble
DQS
Postamble Enable
dqsenable
Delayed by
1/2T logic
February 2011
Altera Corporation
Stratix IV Device Handbook
Volume 1
7–46
Chapter 7: External Memory Interfaces in Stratix IV Devices
Stratix IV External Memory Interface Features
Leveling Circuitry
DDR3 SDRAM unbuffered modules use a fly-by clock distribution topology for better
signal integrity. This means that the CK/CK# signals arrive at each DDR3 SDRAM
device in the module at different times. The difference in arrival time between the first
DDR3 SDRAM device and the last device on the module can be as long as 1.6 ns.
Figure 7–27 shows the clock topology in DDR3 SDRAM unbuffered modules.
Figure 7–27. DDR3 SDRAM Unbuffered Module Clock Topology
DQS/DQ
DQS/DQ
DQS/DQ
DQS/DQ CK/CK# DQS/DQ
DQS/DQ
DQS/DQ
DQS/DQ
Stratix IV Device
Because the data and read strobe signals are still point-to-point, take special care to
ensure that the timing relationship between the CK/CK# and DQS signals (tDQSS,
tDSS, and tDSH) during a write is met at every device on the modules. Furthermore,
read data coming back into the FPGA from the memory is also staggered in a similar
way.
Stratix IV FPGAs have leveling circuitry to address these two situations. There is one
leveling circuitry per I/O sub-bank (for example, I/O sub-bank 1A, 1B, and 1C each
has one leveling circuitry). These delay chains are PVT-compensated by the same DQS
delay settings as the DLL and DQS delay chains.
For frequencies equal to and above 400 MHz, the DLL uses eight delay chains, such
that each delay chain generates a 45° delay. The generated clock phases are
distributed to every DQS logic block that is available in the I/O sub-bank. The delay
chain taps then feeds a multiplexer controlled by the ALTMEMPHY megafunction to
select which clock phases are to be used for that ×4 or × 8 DQS group. Each group can
use a different tap output from the read-leveling and write-leveling delay chains to
compensate for the different CK/CK# delay going into each device on the module.
Stratix IV Device Handbook
Volume 1
February 2011 Altera Corporation
Chapter 7: External Memory Interfaces in Stratix IV Devices
Stratix IV External Memory Interface Features
7–47
Figure 7–28 and Figure 7–29 show the Stratix IV write- and read-leveling circuitry.
Figure 7–28. Stratix IV Write-Leveling Delay Chains and Multiplexers
Write clk
(-900)
(1)
Write-Leveled DQS Clock
Write-Leveled DQ Clock
Note to Figure 7–28:
(1) There is one leveling delay chain per I/O sub-bank (for example, I/O sub-banks 1A, 1B, and 1C). You can only have
one memory interface in each I/O sub-bank when you use the leveling delay chain.
Figure 7–29. Stratix IV Read-Leveling Delay Chains and Multiplexers
(1)
I/O Clock Divider (2)
use_masterin
slaveout
masterin
DQS
delayctrlin
1
0
Half-Rate
Resynchronization Clock
DFF
1
0
clkout
Half-Rate Source
Synchronous Clock
phaseselect
phasectrlin
6
4
phaseinvertctrl
Resynchronization Clock
(resync_clk_2x)
0111
0110
0101
0100
0011
0010
0001
0000
0
1
Read-Leveled Resynchronization Clock
Notes to Figure 7–29:
(1) There is one leveling delay chain per I/O sub-bank (for example, I/O sub-banks 1A, 1B, and 1C). You can only have one memory interface in each
I/O sub-bank when you use the leveling delay chain.
(2) Each divider feeds up to six pins (from a 4 DQS group) in the device. To feed wider DQS groups, you must chain multiple clock dividers together
by feeding the slaveout output of one divider to the masterin input of the neighboring pins’ divider.
The –90° write clock of the ALTMEMPHY megafunction feeds the write-leveling
circuitry to produce the clock to generate the DQS and DQ signals. During
initialization, the ALTMEMPHY megafunction picks the correct write-leveled clock
for the DQS and DQ clocks for each DQS/DQ group after sweeping all the available
clocks in the write calibration process. The DQ clock output is –90° phase-shifted
compared to the DQS clock output.
Similarly, the resynchronization clock feeds the read-leveling circuitry to produce the
optimal resynchronization and postamble clock for each DQS/DQ group in the
calibration process. The resynchronization and postamble clocks can use different
clock outputs from the leveling circuitry. The output from the read-leveling circuitry
can also generate the half-rate resynchronization clock that goes to the FPGA fabric.
February 2011
Altera Corporation
Stratix IV Device Handbook
Volume 1
7–48
Chapter 7: External Memory Interfaces in Stratix IV Devices
Stratix IV External Memory Interface Features
1
The ALTMEMPHY megafunction dynamically calibrates the alignment for read- and
write-leveling during the initialization process.
f For more information about the ALTMEMPHY megafunction, refer to the External
Memory PHY Interface (ALTMEMPHY) (nonAFI) Megafunction User Guide.
Dynamic On-Chip Termination Control
Figure 7–30 shows the dynamic OCT control block. The block includes all the registers
needed to dynamically turn on OCT RT during a read and turn OCT RT off during a
write.
f For more information about dynamic on-chip termination control, refer to the I/O
Features in Stratix IV Devices chapter.
Figure 7–30. Stratix IV Dynamic OCT Control Block
OCT Control
OCT Enable
2
DFF
OCT HalfRate Clock
HDR
Block
DFF
Resynchronization
Registers
Write
Clock (1)
OCT Control Path
Note to Figure 7–30:
(1) The write clock comes from either the PLL or the write-leveling delay chain.
Stratix IV Device Handbook
Volume 1
February 2011 Altera Corporation
Chapter 7: External Memory Interfaces in Stratix IV Devices
Stratix IV External Memory Interface Features
7–49
I/O Element Registers
The IOE registers are expanded to allow source-synchronous systems to have faster
register-to-register transfers and resynchronization. Both top and bottom and left and
right IOEs have the same capability. Left and right IOEs have extra features to support
LVDS data transfer.
Figure 7–31 shows the registers available in the Stratix IV input path. The input path
consists of the DDR input registers, resynchronization registers, and HDR block. You
can bypass each block of the input path.
Figure 7–31. Stratix IV IOE Input Registers
DQ
(1)
Double Data Rate Input Registers
D
Q
DFF
Input Reg AI
D
DQS/CQ (3), (9)
Differential
Input
Buffer
DQSn (9)
CQn (4)
Q
neg_reg_out
DFF
Input Reg BI
0
1
D
Q
Half Data Rate Registers
DFF
Input Reg C I
directin
Alignment & Synchronization Registers
D
Q
D
0
1
Q
datain [0]
D
Q
dataout
D
DFF
DFF
1
Q
To Core
dataout [0]
(7)
DFF
Q
DFF
DFF
D
enaphasetransferreg
enainputcycledelay
<bypass_output_register>(10)
D
Q
1
D
DFF
Q
0
1
DFF
dataout
D
DFF
DFF
1
Q
DFF
D
Q
DFF
I/O Clock
Divider (6)
dataoutbypass
(8)
Q
DFF
Q
(2)
Resynchronization Clock
(resync_clk_2x) (5)
D
0
D
0
Q
Q
DFF
datain [1]
D
To Core
dataout[2]
(7)
0
D
0
1
D
To Core
dataout [1]
(7)
To Core
dataout [3]
(7)
Q
DFF
Half-Rate Resynchronization Clock (resync_clk_1x)
to core (7)
Notes to Figure 7–31:
(1) You can bypass each register block in this path.
(2) This is the 0-phase resynchronization clock (from the read-leveling delay chain).
(3) The input clock can be from the DQS logic block (whether the postamble circuitry is bypassed or not) or from a global clock line.
(4) This input clock comes from the CQn logic block.
(5) This resynchronization clock comes from a PLL through the clock network (resync_ck_2x).
(6) The I/O clock divider resides adjacent to the DQS logic block. In addition to the PLL and read-leveled resync clock, the I/O clock divider can also
be fed by the DQS bus or CQn bus.
(7) The half-rate data and clock signals feed into a dual-port RAM in the FPGA core.
(8) You can dynamically change the dataoutbypass signal after configuration to select either the directin input or the output from the half data
rate register to feed dataout.
(9) The DQS and DQSn signals must be inverted for DDR, DDR2, and DDR3 interfaces. When using Altera’s memory interface IPs, the DQS and DQSn
signals are automatically inverted.
(10) The bypass_output_register option allows you to select either the output from the second mux or the output of the fourth alignment/
synchronization register to feed dataout.
February 2011
Altera Corporation
Stratix IV Device Handbook
Volume 1
7–50
Chapter 7: External Memory Interfaces in Stratix IV Devices
Stratix IV External Memory Interface Features
There are three registers in the DDR input registers block. Two registers capture data
on the positive and negative edges of the clock, while the third register aligns the
captured data. You can choose to use the same clock for the positive edge and
negative edge registers, or two complementary clocks (DQS/CQ for the positive-edge
register and DQSn/CQn for the negative-edge register). The third register that aligns
the captured data uses the same clock as the positive edge registers.
The resynchronization registers consist of up to three levels of registers to
resynchronize the data to the system clock domain. These registers are clocked by the
resynchronization clock that is either generated by the PLL or the read-leveling delay
chain. The outputs of the resynchronization registers can go straight to the core or to
the HDR blocks, which are clocked by the divided-down resynchronization clock.
For more information about the read-leveling delay chain, refer to “Leveling
Circuitry” on page 7–46.
Figure 7–32 shows the registers available in the Stratix IV output and output-enable
paths. The path is divided into the HDR block, resynchronization registers, and
output and output-enable registers. The device can bypass each block of the output
and output-enable path.
Stratix IV Device Handbook
Volume 1
February 2011 Altera Corporation
7–51
Stratix IV Device Handbook
Volume 1
Figure 7–32. Stratix IV IOE Output and Output-Enable Path Registers
(1)
Half Data Rate to Single Data Rate Output-Enable Registers
From Core (2)
Alignment Registers (4)
D
Q
Double Data Rate Output-Enable Registers
DFF
DFF
From Core (2)
0
1
D
Q
D
DFF
D
Q
D
D
D
Q
Q
Q
DFF
Q
OE Reg A OE
DFF
OR2
1
DFF
DFF
0
DFF
D
Half Data Rate to Single Data Rate Output Registers
Q
Alignment Registers (4)
OE Reg B OE
From Core
(wdata2) (2)
D
Q
Double Data Rate Output Registers
DFF
DFF
0
D
Q
D
Q
D
1
From Core
(wdata0) (2)
D
DFF
D
Q
D
Q
Q
Q
TRI
DFF
Output Reg Ao
DFF
DFF
D
Output Reg Bo
0
1
D
Q
DFF
D
Q
Q
DFF
Q
DFF
From Core
(wdata1) (2)
D
Q
D
D
Q
Q
DFF
DFF
DFF
Half-Rate Clock (3)
February 2011 Altera Corporation
Alignment
Clock (3)
Write
Clock (5)
Notes to Figure 7–32:
(1) You can bypass each register block of the output and output-enable paths.
(2) Data coming from the FPGA core are at half the frequency of the memory interface clock frequency in half-rate mode.
(3) The half-rate clock comes from the PLL, while the alignment clock comes from the write-leveling delay chains.
(4) These registers are only used in DDR3 SDRAM interfaces for write-leveling purposes.
(5) The write clock can come from either the PLL or from the write-leveling delay chain. The DQ write clock and DQS write clock have a 90° offset between them.
Chapter 7: External Memory Interfaces in Stratix IV Devices
Stratix IV External Memory Interface Features
D
DQ or DQS
DFF
DFF
From Core
(wdata3) (2)
1
0
7–52
Chapter 7: External Memory Interfaces in Stratix IV Devices
Stratix IV External Memory Interface Features
The output path is designed to route combinatorial or registered SDR outputs and
full-rate or half-rate DDR outputs from the FPGA core. Half-rate data is converted to
full-rate using the HDR block, clocked by the half-rate clock from the PLL. The
resynchronization registers are also clocked by the same 0° system clock, except in the
DDR3 SDRAM interface. In DDR3 SDRAM interfaces, the leveling registers are
clocked by the write-leveling clock.
For more information about the write-leveling delay chain, refer to “Leveling
Circuitry” on page 7–46.
The output-enable path has a structure similar to the output path. You can have a
combinatorial or registered output in SDR applications and you can use half-rate or
full-rate operation in DDR applications. Also, the ouput-enable path’s
resynchronization registers have a structure similar to the output path registers,
ensuring that the output-enable path goes through the same delay and latency as the
output path.
Delay Chain
Stratix IV devices have run-time adjustable delay chains in the I/O blocks and the
DQS logic blocks. You can control the delay chain setting through the I/O or the DQS
configuration block output. Figure 7–33 shows the delay chain ports.
Figure 7–33. Delay Chain
delayctrlin [3..0]
<use finedelayctrlin>
finedelayctrlin
datain
Δt
0
dataout
Δt
1
Every I/O block contains the following:
Stratix IV Device Handbook
Volume 1
■
Two delay chains in a series between the output registers and the output buffer
■
One delay chain between the input buffer and the input register
■
Two delay chains between the output enable and the output buffer
■
Two delay chains between the OCT RT enable control register and the output
buffer
February 2011 Altera Corporation
Chapter 7: External Memory Interfaces in Stratix IV Devices
Stratix IV External Memory Interface Features
7–53
Figure 7–34 shows the delay chains in an I/O block.
Figure 7–34. Delay Chains in an I/O Block
rtena
oe
octdelaysetting1 (only)
D5 OCT
Delay
Chain
D5 OutputEnable Delay
Chain
octdelaysetting2 (only)
D6 OCT
Delay
Chain
D6 OutputEnable Delay
Chain
(outputdelaysetting1 +
outputfinedelaysetting1)
(outputdelaysetting2 +
outputfinedelaysetting2)
D5 Delay
Delay
Chain
D6 Delay
Delay
Chain
0
1
(outputdelaysetting2 + outputfinedelaysetting2) or
(outputonlydelaysetting2 + outputonlyfinedelaysetting2)
D1 Delay
Delay Chain
(padtoinputregisterdelaysetting +
padtoinputregisterfinedelaysetting)
Each DQS logic block contains a delay chain after the dqsbusout output and another
delay chain before the dqsenable input. Figure 7–35 shows the delay chains in the
DQS input path.
Figure 7–35. Delay Chains in the DQS Input Path
(dqsbusoutdelaysetting +
dqsbusoutfinedelaysetting)
DQS
DQS
Delay
Chain
DQS
Enable
D4 Delay
Chain
dqsin
dqsbusout
dqsenable
(dqsenabledelaysetting +
dqsenablefinedelaysetting)
T11 Delay
Chain
DQS
Enable
Control
February 2011
Altera Corporation
Stratix IV Device Handbook
Volume 1
7–54
Chapter 7: External Memory Interfaces in Stratix IV Devices
Stratix IV External Memory Interface Features
I/O Configuration Block and DQS Configuration Block
The I/O configuration block and the DQS configuration block are shift registers that
you can use to dynamically change the settings of various device configuration bits.
The shift registers power-up low. Every I/O pin contains one I/O configuration
register, while every DQS pin contains one DQS configuration block in addition to the
I/O configuration register. Figure 7–36 shows the I/O configuration block and the
DQS configuration block circuitry.
Figure 7–36. I/O Configuration Block and DQS Configuration Block
bit 0
bit 1
MSB
bit 2
datain
update
ena
clk
Table 7–19 lists the I/O configuration block bit sequence.
Table 7–19. I/O Configuration Block Bit Sequence
Bit
Bit Name
0..3
outputdelaysetting1[0..3]
4..6
outputdelaysetting2[0..2]
7..10
padtoinputregisterdelaysetting[0..3]
Table 7–20 lists the DQS configuration block bit sequence.
Table 7–20. DQS Configuration Block Bit Sequence (Part 1 of 2)
Stratix IV Device Handbook
Volume 1
Bit
Bit Name
0..3
dqsbusoutdelaysetting[0..3]
4..6
dqsinputphasesetting[0..2]
7..10
dqsenablectrlphasesetting[0..3]
11..14
dqsoutputphasesetting[0..3]
15..18
dqoutputphasesetting[0..3]
19..22
resyncinputphasesetting[0..3]
23
dividerphasesetting
24
enaoctcycledelaysetting
25
enainputcycledelaysetting
26
enaoutputcycledelaysetting
27..29
dqsenabledelaysetting[0..2]
30..33
octdelaysetting1[0..3]
February 2011 Altera Corporation
Chapter 7: External Memory Interfaces in Stratix IV Devices
Stratix IV External Memory Interface Features
7–55
Table 7–20. DQS Configuration Block Bit Sequence (Part 2 of 2)
Bit
Bit Name
34..36
octdelaysetting2[0..2]
37
enadataoutbypass
38
enadqsenablephasetransferreg
39
enaoctphasetransferreg
40
enaoutputphasetransferreg
41
enainputphasetransferreg
42
resyncinputphaseinvert
43
dqsenablectrlphaseinvert
44
dqoutputphaseinvert
45
dqsoutputphaseinvert
Document Revision History
Table 7–21 lists the revision history for this chapter.
Table 7–21. Document Revision History (Part 1 of 2)
Date
Version
February 2011
March 2010
February 2011
3.2
3.1
Altera Corporation
Changes
■
Updated Table 7–5, Table 7–6, Table 7–11, Table 7–19, and Table 7–20.
■
Added Table 7–12.
■
Updated Figure 7–36.
■
Removed Table 7-1 and Table 7-6.
■
Applied new template.
■
Minor text edits.
■
Updated Figure 7–8, Figure 7–11, Figure 7–23, Figure 7–24, Figure 7–29, Figure 7–31,
and Figure 7–36.
■
Added Figure 7–9 and Figure 7–12.
■
Added Table 7–7.
■
Updated Table 7–1, Table 7–2, Table 7–3, Table 7–4, Table 7–6, Table 7–8 and Table 7–19.
■
Added note to the “Memory Interfaces Pin Support” section.
■
Changed “DLL1 through DLL4” to “DLL0 through DLL3” throughout.
■
Added frequency mode 7 throughout.
■
Minor text edits.
Stratix IV Device Handbook
Volume 1
7–56
Chapter 7: External Memory Interfaces in Stratix IV Devices
Stratix IV External Memory Interface Features
Table 7–21. Document Revision History (Part 2 of 2)
Date
Version
November 2009
June 2009
April 2009
March 2009
November 2008
May 2008
Stratix IV Device Handbook
Volume 1
3.0
2.3
2.2
2.1
2.0
1.0
Changes
■
Updated the “Memory Interfaces Pin Support” and “Combining ×16/×18 DQS/DQ Groups
for a ×36 QDR II+/QDR II SRAM Interface” sections.
■
Updated Table 7–1, Table 7–2, Table 7–7, and Table 7–12.
■
Updated Figure 7–3, Figure 7–4, Figure 7–5, Figure 7–6, Figure 7–7, Figure 7–8,
Figure 7–9, Figure 7–10, Figure 7–11, Figure 7–13, Figure 7–14, Figure 7–15, and
Figure 7–16.
■
Added Figure 7–12 and Figure 7–17.
■
Added Table 7–14, Table 7–17, Table 7–19, and Table 7–20.
■
Added “Delay Chain” and “I/O Configuration Block and DQS Configuration Block”
sections.
■
Removed Figure 7-8 and Figure 7-12.
■
Removed Table 7-1, Table 7-2, and Table 7-24.
■
Minor text edits.
■
Updated “Overview” and “Leveling Circuitry”.
■
Updated Figure 7–26 and Figure 7–27.
■
Updated Table 7–3.
■
Added introductory sentences to improve search ability.
■
Removed the Conclusion section.
■
Updated Table 7–5, Table 7–6, Table 7–15, and Table 7–17
■
Removed Figure 7-12, Figure 7-13, and Figure 7-20
■
Updated Table 7–1, Table 7–5, Table 7–8, Table 7–12, Table 7–13, Table 7–14,
Table 7–15, and Table 7–17.
■
Replaced Table 7–6.
■
Added Table 7–11 and Table 7–16.
■
Updated Figure 7–3, Figure 7–6, Figure 7–8, Figure 7–9, and Figure 7–11.
■
Added Figure 7–7, Figure 7–11, Figure 7–12, Figure 7–13, and Figure 7–20.
■
Updated “Combining ×16/×18 DQS/DQ Groups for ×36 QDR II+/QDR II SRAM Interface”.
■
Updated “Rules to Combine Groups”.
■
Removed “Referenced Documents” section.
■
Updated Table 7–1, Table 7–2, Table 7–3, Table 7–4, Table 7–5, and Table 7–6.
■
Added Table 7–7.
■
Updated Figure 7–1 and Figure 7–19.
■
Updated “Combining ×16/×18 DQS/DQ groups for ×36 QDR II+/QDR II SRAM Interface”
on page 7–26.
■
Updated “Rules to Combine Groups” on page 7–27.
■
Updated “DQS Phase-Shift Circuitry” on page 7–29.
■
Updated Table 7–9, Table 7–10, Table 7–11, Table 7–13, Table 7–13, Table 7–14,
Table 7–15, Table 7–15, Table 7–16, and Table 7–18.
■
Updated Figure 7–30 and Figure 7–31.
■
Made minor editorial changes.
Initial release.
February 2011 Altera Corporation
8. High-Speed Differential I/O Interfaces
and DPA in Stratix IV Devices
September 2012
SIV51008-3.4
SIV51008-3.4
This chapter describes the significant advantages of the high-speed differential I/O
interfaces and the dynamic phase aligner (DPA) over single-ended I/Os and their
contribution to the overall system bandwidth achievable with Stratix® IV FPGAs. All
references to Stratix IV devices in this chapter apply to Stratix IV E, GT, and GX
devices.
The Stratix IV device family consists of the Stratix IV E (Enhanced) devices without
high-speed clock data recovery (CDR) based transceivers, Stratix IV GT devices with
up to 48 CDR-based transceivers running up to 11.3 Gbps, and Stratix IV GX devices
with up to 48 CDR-based transceivers running up to 8.5 Gbps.
The following sections describe the Stratix IV high-speed differential I/O interfaces
and DPA:
■
“Locations of the I/O Banks” on page 8–3
■
“LVDS Channels” on page 8–4
■
“LVDS SERDES” on page 8–8
■
“ALTLVDS Port List” on page 8–9
■
“Differential Transmitter” on page 8–11
■
“Differential Receiver” on page 8–17
■
“LVDS Interface with the Use External PLL Option Enabled” on page 8–26
■
“Left and Right PLLs (PLL_Lx and PLL_Rx)” on page 8–29
■
“Stratix IV Clocking” on page 8–30
■
“Source-Synchronous Timing Budget” on page 8–31
■
“Differential Pin Placement Guidelines” on page 8–38
Overview
All Stratix IV E, GX, and GT devices have built-in serializer/deserializer (SERDES)
circuitry that supports high-speed LVDS interfaces at data rates of up to 1.6 Gbps.
SERDES circuitry is configurable to support source-synchronous communication
protocols such as Utopia, Rapid I/O, XSBI, small form factor interface (SFI), serial
peripheral interface (SPI), and asynchronous protocols such as SGMII and Gigabit
Ethernet.
© 2012 Altera Corporation. All rights reserved. ALTERA, ARRIA, CYCLONE, HARDCOPY, MAX, MEGACORE, NIOS, QUARTUS and STRATIX words and logos
are trademarks of Altera Corporation and registered in the U.S. Patent and Trademark Office and in other countries. All other words and logos identified as
trademarks or service marks are the property of their respective holders as described at www.altera.com/common/legal.html. Altera warrants performance of its
semiconductor products to current specifications in accordance with Altera's standard warranty, but reserves the right to make changes to any products and
services at any time without notice. Altera assumes no responsibility or liability arising out of the application or use of any information, product, or service
described herein except as expressly agreed to in writing by Altera. Altera customers are advised to obtain the latest version of device specifications before relying
on any published information and before placing orders for products or services.
ISO
9001:2008
Registered
Stratix IV Device Handbook
Volume 1
September 2012
Feedback Subscribe
8–2
Chapter 8: High-Speed Differential I/O Interfaces and DPA in Stratix IV Devices
Overview
The Stratix IV device family has the following dedicated circuitry for high-speed
differential I/O support:
■
Differential I/O buffer
■
Transmitter serializer
■
Receiver deserializer
■
Data realignment
■
DPA
■
Synchronizer (FIFO buffer)
■
Phase-locked loops (PLLs) (located on left and right sides of the device)
For high-speed differential interfaces, the Stratix IV device family supports the
following differential I/O standards:
■
LVDS
■
Mini-LVDS
■
Reduced swing differential signaling (RSDS)
In the Stratix IV device family, I/Os are divided into row and column I/Os. Figure 8–1
shows I/O bank support for the Stratix IV device family. The row I/Os provide
dedicated SERDES circuitry.
Figure 8–1. I/O Bank Support in the Stratix IV Device Family (1),
(2), (3), (4)
LVDS I/Os
Row I/Os with
Dedicated
SERDES Circuitry (3), (4)
LVDS Interface
with 'Use External PLL'
Option Enabled
Column I/Os (1), (2)
LVDS Interface
with 'Use External PLL'
Option Disabled
Notes to Figure 8–1:
(1) Column input buffers are true LVDS buffers, but do not support 100-differential on-chip termination.
(2) Column output buffers are single ended and need external termination schemes to support LVDS, mini-LVDS, and RSDS standards. For more
information, refer to the I/O Features in Stratix IV Devices chapter.
(3) Row input buffers are true LVDS buffers and support 100-differential on-chip termination.
(4) Row output buffers are true LVDS buffers.
Stratix IV Device Handbook
Volume 1
September 2012 Altera Corporation
Chapter 8: High-Speed Differential I/O Interfaces and DPA in Stratix IV Devices
Locations of the I/O Banks
8–3
The ALTLVDS transmitter and receiver requires various clock and load enable signals
from a left or right PLL. The Quartus® II software provides the following two choices
when configuring the LVDS SERDES circuitry when using the PLL:
1
■
LVDS interface with the Use External PLL option enabled—You control the PLL
settings, such as dynamically reconfiguring the PLL to support different data
rates, dynamic phase shift, and so on. You must enable the Use External PLL
option in the ALTLVDS_TX and ALTLVDS_RX megafunctions, using the
ALTLVDS MegaWizard Plug-in Manager software. You also must instantiate an
ALTPLL megafunction to generate the various clocks and load enable signals. For
more information, refer to “LVDS Interface with the Use External PLL Option
Enabled” on page 8–26.
■
LVDS interface with the Use External PLL option disabled—The Quartus II
software configures the PLL settings automatically. The software is also
responsible for generating the various clock and load enable signals based on the
input reference clock and data rate selected.
Both choices target the same physical PLL; the only difference is the additional
flexibility provided when an LVDS interface has the Use External PLL option enabled.
Locations of the I/O Banks
Stratix IV I/Os are divided into 16 to 24 I/O banks. The dedicated circuitry that
supports high-speed differential I/Os is located in banks in the right and left side of
the device. Figure 8–2 shows a high-level chip overview of the Stratix IV E device.
Figure 8–2. High-Speed Differential I/Os with DPA Locations in Stratix IV E Devices
General Purpose
I/O and Memory
Interface
PLL
PLL
General Purpose
I/O and
High-Speed
LVDS I/O with
DPA and Soft CDR
PLL
FPGA Fabric
(Logic Elements, DSP,
Embedded Memory,
Clock Networks)
PLL
General Purpose
I/O and Memory
Interface
September 2012
Altera Corporation
PLL
PLL
General Purpose
I/O and
High-Speed
LVDS I/O with
DPA and Soft CDR
PLL
General Purpose
I/O and
High-Speed
LVDS I/O with
DPA and Soft CDR
PLL
General Purpose
I/O and
High-Speed
LVDS I/O with
DPA and Soft CDR
PLL
General Purpose
I/O and Memory
Interface
PLL
PLL
PLL
General Purpose
I/O and Memory
Interface
Stratix IV Device Handbook
Volume 1
8–4
Chapter 8: High-Speed Differential I/O Interfaces and DPA in Stratix IV Devices
LVDS Channels
Figure 8–3 shows a high-level chip overview of the Stratix IV GT and GX devices.
Figure 8–3. High-Speed Differential I/Os with DPA Locations in Stratix IV GT and GX Devices
PLL
PLL
General Purpose
I/O and Memory
Interface
PLL
PCI Express
Hard IP Block
PLL
PLL
PLL
PCI Express
Hard IP Block
PLL
General Purpose
I/O and
High-Speed
LVDS I/O with
DPA and Soft CDR
General Purpose
I/O and
High-Speed
LVDS I/O with
DPA and Soft CDR
PLL
FPGA Fabric
(Logic Elements, DSP,
Embedded Memory,
Clock Networks)
General Purpose
I/O and
High-Speed
LVDS I/O with
DPA and Soft CDR
PLL
General Purpose
I/O and
High-Speed
LVDS I/O with
DPA and Soft CDR
PCI Express
Hard IP Block
PLL
General Purpose
I/O and Memory
Interface
Transceiver Transceiver Transceiver Transceiver
Block
Block
Block
Block
PLL
PLL
PCI Express
Hard IP Block
Transceiver Transceiver Transceiver Transceiver
Block
Block
Block
Block
General Purpose
I/O and Memory
Interface
General Purpose
I/O and Memory
Interface
LVDS Channels
The Stratix IV device family supports LVDS on both row and column I/O banks. Row
I/Os support true LVDS input with 100- differential input termination (OCT RD),
and true LVDS output buffers. Column I/Os supports true LVDS input buffers
without OCT RD. Alternately, you can configure the row and column LVDS pins as
emulated LVDS output buffers that use two single-ended output buffers with an
external resistor network to support LVDS, mini-LVDS, and RSDS standards.
Stratix IV devices offer single-ended I/O refclk support for the LVDS.
Dedicated SERDES and DPA circuitries are implemented on the row I/O banks to
further enhance LVDS interface performance in the device. For column I/O banks,
SERDES is implemented in the core logic because there is no dedicated SERDES
circuitry on column I/O banks.
1
Stratix IV Device Handbook
Volume 1
Emulated differential output buffers support tri-state capability starting with the
Quartus II software version 9.1.
September 2012 Altera Corporation
Chapter 8: High-Speed Differential I/O Interfaces and DPA in Stratix IV Devices
LVDS Channels
8–5
Table 8–1 and Table 8–2 list the maximum number of row and column LVDS I/Os
supported in Stratix IV E devices. You can design the LVDS I/Os as true LVDS buffers
or emulated LVDS buffers, as long as the combination of the two do not exceed the
maximum count.
For example, there are a total of 112 LVDS pairs on row I/Os in the 780-pin EP4SE230
device (refer to Table 8–1). You can design up to a maximum of 56 true LVDS input
buffers and 56 true LVDS output buffers, or up to a maximum of 112 emulated LVDS
output buffers. For the 780-pin EP4SE230 device (refer to Table 8–2), there are a total
of 128 LVDS pairs on column I/Os. You can design up to a maximum of 64 true LVDS
input buffers and 64 emulated LVDS output buffers, or up to a maximum of 128
emulated LVDS output buffers.
Table 8–1. LVDS Channels Supported in Stratix IV E Device Row I/O Banks (1),
(2), (3)
Device
780-Pin FineLine BGA
1152-Pin FineLine BGA
1517-Pin FineLine BGA
1760- Pin FineLine BGA
EP4SE230
56 Rx or eTx + 56 Tx
or eTx
—
—
—
EP4SE360
56 Rx or eTx + 56 Tx
or eTx (4)
88 Rx or eTx + 88 Tx
or eTx
—
—
EP4SE530
—
88 Rx or eTx + 88 Tx
or eTx (5)
112 Rx or eTx + 112 Tx
or eTx (6)
112 Rx or eTx + 112 Tx
or eTx
EP4SE820
—
88 Rx or eTx + 88 Tx
or eTx
112 Rx or eTx + 112 Tx
or eTx
132 Rx or eTx + 132 Tx
or eTx
Notes to Table 8–1:
(1) Receiver (Rx) = true LVDS input buffers with OCT RD, Transmitter (Tx) = true LVDS output buffers, eTx = emulated LVDS output buffers (either
LVDS_E_1R or LVDS_E_3R).
(2) The LVDS Rx and Tx channels are equally divided between the left and right sides of the device.
(3) The LVDS channel count does not include dedicated clock input pins.
(4) EP4SE360 devices are offered in the H780 package instead of the F780 package.
(5) EP4SE530 devices are offered in the H1152 package instead of the F1152 package.
(6) EP4SE530 devices are offered in the H1517 package instead of the F1517 package.
Table 8–2. LVDS Channels Supported in Stratix IV E Device Column I/O Banks (1),
(2), (3)
Device
780-Pin FineLine BGA
1152-Pin FineLine BGA
1517-Pin FineLine BGA
1760-Pin FineLine BGA
EP4SE230
64 Rx or eTx + 64 eTx
—
—
—
96 Rx or eTx + 96 eTx
—
—
96 Rx or eTx + 96 eTx
128 Rx or eTx + 128 eTx
(5)
(6)
64 Rx or eTx + 64 eTx
EP4SE360
(4)
EP4SE530
—
EP4SE820
—
96 Rx or eTx + 96 eTx
128 Rx or eTx + 128 eTx
128 Rx or eTx + 128 eTx 144 Rx or eTx + 144 eTx
Notes to Table 8–2:
(1) Rx = true LVDS input buffers without OCT RD, eTx = emulated LVDS output buffers (either LVDS_E_1R or LVDS_E_3R).
(2) The LVDS Rx and Tx channels are equally divided between the top and bottom sides of the device.
(3) The LVDS channel count does not include dedicated clock input pins.
(4) EP4SE360 devices are offered in the H780 package instead of the F780 package.
(5) EP4SE530 devices are offered in the H1152 package instead of the F1152 package.
(6) EP4SE530 devices are offered in the H1517 package instead of the F1517 package.
September 2012
Altera Corporation
Stratix IV Device Handbook
Volume 1
8–6
Chapter 8: High-Speed Differential I/O Interfaces and DPA in Stratix IV Devices
LVDS Channels
Table 8–3 and Table 8–4 list the maximum number of row and column LVDS I/Os
supported in Stratix IV GT devices.
Table 8–3. LVDS Channels Supported in Stratix IV GT Device Row I/O Banks (1),
(2)
Device
1517-pin FineLine BGA
1932-pin FineLine BGA
EP4S40G2
46 Rx or eTx + 73 Tx or eTx
—
EP4S40G5
46 Rx or eTx + 73 Tx or eTx
—
EP4S100G2
46 Rx or eTx + 73 Tx or eTx
—
EP4S100G3
—
47 Rx or eTx + 56 Tx or eTx
EP4S100G4
—
47 Rx or eTx + 56 Tx or eTx
EP4S100G5
46 Rx or eTx + 73 Tx or eTx
47 Rx or eTx + 56 Tx or eTx
Notes to Table 8–3:
(1) Rx = true LVDS input buffers with OCT RD, eTx = emulated LVDS output buffers (either LVDS_E_1R or
LVDS_E_3R).
(2) The LVDS Rx and Tx channel count does not include dedicated clock input pins.
Table 8–4. LVDS Channels Supported in Stratix IV GT Device Column I/O Banks (1),
(2)
Device
1517-pin FineLine BGA
1932-pin FineLine BGA
EP4S40G2
96 Rx or eTx + 96 eTx
—
EP4S40G5
96 Rx or eTx + 96 eTx
—
EP4S100G2
96 Rx or eTx + 96 eTx
—
EP4S100G3
—
128 Rx or eTx + 128 eTx
EP4S100G4
—
128 Rx or eTx + 128 eTx
EP4S100G5
96 Rx or eTx + 96 eTx
128 Rx or eTx + 128 eTx
Notes to Table 8–4:
(1) Rx = true LVDS input buffers without OCT RD, eTx = emulated LVDS output buffers (either LVDS_E_1R or
LVDS_E_3R).
(2) The LVDS Rx and Tx channel count does not include dedicated clock input pins.
Table 8–5 and Table 8–6 list the maximum number of row and column LVDS I/Os
supported in Stratix IV GX devices.
Table 8–5. LVDS Channels Supported in Stratix IV GX Device Row I/O Banks
780-Pin
FineLine BGA
1152-Pin
FineLine BGA
EP4SGX70
28 Rx or eTx +
28 Tx or eTx
—
EP4SGX110
28 Rx or eTx +
28 Tx or eTx
EP4SGX180
EP4SGX230
Device
EP4SGX290
1152-Pin
FineLine BGA
(1), (2), (3)
(Part 1 of 2)
1517-Pin
FineLine BGA
1760-Pin
FineLine BGA
1932-Pin
FineLine BGA
56 Rx or eTx +
56 Tx or eTx
—
—
—
28 Rx or eTx +
28 Tx or eTx
56 Rx or eTx +
56 Tx or eTx
—
—
—
28 Rx or eTx +
28 Tx or eTx
44 Rx or eTx +
44 Tx or eTx
44 Rx or eTx +
44 Tx or eTx
88 Rx or eTx +
88 Tx or eTx
—
—
28 Rx or eTx +
28 Tx or eTx
44 Rx or eTx +
44 Tx or eTx
44 Rx or eTx +
44 Tx or eTx
88 Rx or eTx +
88 Tx or eTx
—
—
44 Rx or eTx +
44 Tx or eTx
44 Rx or eTx +
44 Tx or eTx
88 Rx or eTx +
88 Tx or eTx
88 Rx or eTx +
88 Tx or eTx
98 Rx or eTx +
98 Tx or eTx
—
Stratix IV Device Handbook
Volume 1
(5)
(4)
September 2012 Altera Corporation
Chapter 8: High-Speed Differential I/O Interfaces and DPA in Stratix IV Devices
LVDS Channels
8–7
Table 8–5. LVDS Channels Supported in Stratix IV GX Device Row I/O Banks
Device
780-Pin
FineLine BGA
EP4SGX360
—
EP4SGX530
(5)
—
1152-Pin
FineLine BGA
1152-Pin
FineLine BGA
(1), (2), (3)
(Part 2 of 2)
(4)
1517-Pin
FineLine BGA
1760-Pin
FineLine BGA
1932-Pin
FineLine BGA
44 Rx or eTx +
44 Tx or eTx
44 Rx or eTx +
44 Tx or eTx
88 Rx or eTx +
88 Tx or eTx
88 Rx or eTx +
88 Tx or eTx
98 Rx or eTx +
98 Tx or eTx
—
44 Rx or eTx +
44 Tx or eTx (6)
88 Rx or eTx +
88 Tx or eTx (7)
88 Rx or eTx +
88 Tx or eTx
98 Rx or eTx +
98 Tx or eTx
Notes to Table 8–5:
(1) Rx = true LVDS input buffers with OCT RD, Tx = true LVDS output buffers, eTx = emulated LVDS output buffers (either LVDS_E_1R or
LVDS_E_3R).
(2) The LVDS Rx and Tx channels are equally divided between the left and right sides of the device, except for the devices in the 780-pin Fineline
BGA. These devices have the LVDS Rx and Tx located on the left side of the device.
(3) The LVDS channel count does not include dedicated clock input pins.
(4) This package supports PMA-only transceiver channels.
(5) EP4SGX290 and EP4SGX360 devices are offered in the H780 package instead of the F780 package.
(6) EP4SGX530 devices are offered in the H1152 package instead of the F1152 package.
(7) EP4SGX530 devices are offered in the H1517 package instead of the F1517 package.
Table 8–6. LVDS Channels Supported in Stratix IV GX Device Column I/O Banks (1),
1152-Pin
FineLine BGA
(2), (3)
1517-Pin
FineLine BGA
1760-Pin
FineLine BGA
1932-Pin
FineLine BGA
64 Rx or eTx +
64 eTx
—
—
—
64 Rx or eTx +
64 eTx
64 Rx or eTx +
64 eTx
—
—
—
64 Rx or eTx +
64 eTx
96 Rx or eTx +
96 eTx
96 Rx or eTx +
96 eTx
96 Rx or eTx +
96 eTx
—
—
EP4SGX230
64 Rx or eTx +
64 eTx
96 Rx or eTx +
96 eTx
96 Rx or eTx +
96 eTx
96 Rx or eTx +
96 eTx
—
—
EP4SGX290
72 Rx or eTx +
72 eTx (5)
96 Rx or eTx +
96 eTx
96 Rx or eTx +
96 eTx
96 Rx or eTx +
96 eTx
128 Rx or eTx +
128 eTx
128 Rx or eTx +
128 eTx (8)
EP4SGX360
72 Rx or eTx +
72 eTx (5)
96 Rx or eTx +
96 eTx
96 Rx or eTx +
96 eTx
96 Rx or eTx +
96 eTx
128 Rx or eTx +
128 eTx
128 Rx or eTx +
128 eTx (8)
EP4SGX530
—
—
96 Rx or eTx +
96 eTx (6)
96 Rx or eTx +
96 eTx (7)
128 Rx or eTx +
128 eTx
128 Rx or eTx +
128 eTx
780-Pin
FineLine BGA
1152-Pin
FineLine BGA
EP4SGX70
64 Rx or eTx +
64 eTx
—
EP4SGX110
64 Rx or eTx +
64 eTx
EP4SGX180
Device
(4)
Notes to Table 8–6:
(1) Rx = true LVDS input buffers without OCT RD, eTx = emulated LVDS output buffers (either LVDS_E_1R or LVDS_E_3R).
(2) The LVDS Rx and Tx channels are equally divided between the left and right sides of the device.
(3) The LVDS channel count does not include dedicated clock input pins.
(4) This package supports PMA-only transceiver channels.
(5) EP4SGX290 and EP4SGX360 devices are offered in the H780 package instead of the F780 package.
(6) EP4SGX530 devices are offered in the H1152 package instead of the F1152 package.
(7) EP4SGX530 devices are offered in the H1517 package instead of the F1517 package.
(8) The Quartus II software version 9.0 does not support EP4SGX290 and EP4SGX360 devices in the 1932-Pin FineLine BGA package. These
devices will be supported in a future release of the Quartus II software.
September 2012
Altera Corporation
Stratix IV Device Handbook
Volume 1
8–8
Chapter 8: High-Speed Differential I/O Interfaces and DPA in Stratix IV Devices
LVDS SERDES
LVDS SERDES
Figure 8–4 shows a transmitter and receiver block diagram for the LVDS SERDES
circuitry in the left and right banks. This diagram shows the interface signals of the
transmitter and receiver data path. For more information, refer to “Differential
Transmitter” on page 8–11 and “Differential Receiver” on page 8–17.
Figure 8–4. LVDS SERDES
(1), (2), (3)
Serializer
tx_in
2
IOE Supports SDR, DDR, or
Non-Registered Datapath
IOE
tx_out
+
-
10
DIN DOUT
LVDS Transmitter
tx_coreclock
3
(LVDS_LOAD_EN, diffioclk,
tx_coreclock)
IOE Supports SDR, DDR, or
Non-Registered Datapath
10
2
LVDS Receiver
+
-
IOE
rx_out
rx_in
Synchronizer
FPGA
Fabric
Deserializer
Bit Slip
DOUT DIN
DOUT DIN
DPA Circuitry
Retimed
Data
DOUT DIN
DIN
diffioclk
2
(LOAD_EN, diffioclk)
Clock MUX
DPA_diffioclk
LVDS_diffioclk
DPA Clock
3
(DPA_LOAD_EN,
DPA_diffioclk,
rx_divfwdclk)
rx_divfwdclk
rx_outclock
3
(LVDS_LOAD_EN,
LVDS_diffioclk,
rx_outclock
LVDS Clock Domain
DPA Clock Domain
8 Serial LVDS
Clock Phases
Left/Right PLL
rx_inclock/tx_inclock
Notes to Figure 8–4:
(1) This diagram shows a shared PLL between the transmitter and receiver. If the transmitter and receiver are not sharing the same PLL, the two left
and right PLLs are required.
(2) In SDR and DDR modes, the data width is 1 and 2 bits, respectively.
(3) The tx_in and rx_out ports have a maximum data width of 10 bits.
Stratix IV Device Handbook
Volume 1
September 2012 Altera Corporation
Chapter 8: High-Speed Differential I/O Interfaces and DPA in Stratix IV Devices
ALTLVDS Port List
8–9
ALTLVDS Port List
Table 8–7 lists the interface signals for an LVDS transmitter and receiver.
Table 8–7. Port List of the LVDS Interface (ALTLVDS)
Port Name
Input /
Output
(1), (2)
(Part 1 of 3)
Description
PLL Signals
pll_areset
Input
Asynchronous reset to the LVDS transmitter and receiver PLL. The
minimum pulse width requirement for this signal is 10 ns.
Input
The data bus width per channel is the same as the serialization factor (SF).
Input data must be synchronous to the tx_coreclock signal.
LVDS Transmitter Interface Signals
tx_in[ ]
Reference clock input for the transmitter PLL.
Input
tx_inclock
The ALTLVDS MegaWizard Plug-In Manager software automatically selects
the appropriate PLL multiplication factor based on the data rate and
reference clock frequency selection.
For more information about the allowed frequency range for this reference
clock, refer to the “High-Speed I/O Specification” section in the DC and
Switching Characteristics for Stratix IV Devices chapter.
Input
This port is instantiated only when you select the Use External PLL option
in the MegaWizard Plug-In Manager software. This input port must be
driven by the PLL instantiated though the ALTPLL MegaWizard Plug-In
Manager software.
tx_out
Output
LVDS transmitter serial data output port. tx_out is clocked by a serial clock
generated by the left and right PLL.
tx_outclock
Output
The frequency of this clock is programmable to be the same as the data
rate, half the data rate, or one-fourth the data rate. The phase offset of this
clock, with respect to the serial data, is programmable in increments of 45°.
tx_enable
(3)
FPGA fabric-transmitter interface clock. The parallel transmitter data
generated in the FPGA fabric must be clocked with this clock.
tx_coreclock
(3)
Output
tx_locked
September 2012
Output
Altera Corporation
This port is not available when you select the Use External PLL option in the
MegaWizard Plug-In Manager software. The FPGA fabric-transmitter
interface clock must be driven by the PLL instantiated through the ALTPLL
MegaWizard Plug-In Manager software.
When high, this signal indicates that the transmitter PLL is locked to the
input reference clock.
Stratix IV Device Handbook
Volume 1
8–10
Chapter 8: High-Speed Differential I/O Interfaces and DPA in Stratix IV Devices
ALTLVDS Port List
Table 8–7. Port List of the LVDS Interface (ALTLVDS)
(1), (2)
Port Name
Input /
Output
(Part 2 of 3)
Description
LVDS Receiver Interface Signals
rx_in
Input
LVDS receiver serial data input port.
Reference clock input for the receiver PLL.
rx_inclock
Input
The ALTLVDS MegaWizard Plug-In Manager software automatically selects
the appropriate PLL multiplication factor based on the data rate and
reference clock frequency selection.
For more information about the allowed frequency range for this reference
clock, refer to the “High-Speed I/O Specification” section in the DC and
Switching Characteristics for Stratix IV Devices chapter.
Input
Edge-sensitive bit-slip control signal. Each rising edge on this signal causes
the data re-alignment circuitry to shift the word boundary by one bit. The
minimum pulse width requirement is one parallel clock cycle. There is no
maximum pulse width requirement.
Input
When low, the DPA tracks any dynamic phase variations between the clock
and data. When high, the DPA holds the last locked phase and does not
track any dynamic phase variations between the clock and data. This port is
not available in non-DPA mode.
Input
This port is instantiated only when you select the Use External PLL option
in the MegaWizard Plug-In Manager software. This input port must be
driven by the PLL instantiated though the ALTPLL MegaWizard Plug-In
Manager software.
Output
Receiver parallel data output. The data bus width per channel is the same as
the deserialization factor (DF). The output data is synchronous to the
rx_outclock signal in non-DPA and DPA modes. It is synchronous to the
rx_divfwdclk signal in soft-CDR mode.
rx_outclock
Output
Parallel output clock from the receiver PLL. The parallel data output from
the receiver is synchronous to this clock in non-DPA and DPA modes. This
port is not available when you select the Use External PLL option in the
MegaWizard Plug-In Manager software. The FPGA fabric-receiver interface
clock must be driven by the PLL instantiated through the ALTPLL
MegaWizard Plug-In Manager software.
rx_locked
Output
When high, this signal indicates that the receiver PLL is locked to
rx_inclock.
rx dpa locked
Output
This signal only indicates an initial DPA lock condition to the optimum
phase after power up or reset. This signal is not de-asserted if the DPA
selects a new phase out of the eight clock phases to sample the received
data. You must not use the rx_dpa_locked signal to determine a DPA
loss-of-lock condition.
rx_cda_max
Output
Data re-alignment (bit slip) roll-over signal. When high for one parallel clock
cycle, this signal indicates that the user-programmed number of bits for the
word boundary to roll-over have been slipped.
rx_divfwdclk
Output
Parallel DPA clock to the FPGA fabric logic array. The parallel receiver
output data to the FPGA fabric logic array is synchronous to this clock in
soft-CDR mode. This signal is not available in non-DPA and DPA modes.
dpa_pll_recal
Input
Enable PLL calibration dynamically without resetting the DPA circuitry or
the PLL.
rx_channel_data_align
rx_dpll_hold
rx_enable (3)
rx_out[ ]
Stratix IV Device Handbook
Volume 1
September 2012 Altera Corporation
Chapter 8: High-Speed Differential I/O Interfaces and DPA in Stratix IV Devices
Differential Transmitter
Table 8–7. Port List of the LVDS Interface (ALTLVDS)
Input /
Output
Port Name
(1), (2)
8–11
(Part 3 of 3)
Description
Output
Busy signal that is asserted high when the PLL calibration occurs.
Input
Asynchronous reset to the DPA circuitry and FIFO. The minimum pulse
width requirement for this reset is one parallel clock cycle. This signal
resets DPA and FIFO blocks.
rx_fifo_reset
Input
Asynchronous reset to the FIFO between the DPA and the data realignment
circuits. The synchronizer block must be reset after a DPA loses lock
condition and the data checker shows corrupted received data. The
minimum pulse width requirement for this reset is one parallel clock cycle.
This signal resets the FIFO block.
rx_cda_reset
Input
Asynchronous reset to the data realignment circuitry. The minimum pulse
width requirement for this reset is one parallel clock cycle. This signal
resets the data realignment block.
dpa_pll_cal_busy
Reset Signals
rx_reset
Notes to Table 8–7:
(1) Unless stated, signals are valid in all three modes (non-DPA, DPA, and soft-CDR) for a single channel.
(2) All reset and control signals are active high.
(3) For more information, refer to “LVDS Interface with the Use External PLL Option Enabled” on page 8–26.
f For more information about the LVDS transmitter and receiver settings using
ALTLVDS_TX and ALTLVDS_RX megafunction, refer to the ALTLVDS Megafunction
User Guide.
Differential Transmitter
The Stratix IV transmitter has a dedicated circuitry to provide support for LVDS
signaling. The dedicated circuitry consists of a differential buffer, a serializer, and left
and right PLLs that can be shared between the transmitter and receiver. The
differential buffer can drive out LVDS, mini-LVDS, and RSDS signaling levels. The
serializer takes up to 10 bits wide parallel data from the FPGA fabric, clocks it into the
load registers, and serializes it using shift registers clocked by the left and right PLL
before sending the data to the differential buffer. The MSB of the parallel data is
transmitted first.
1
September 2012
When using emulated LVDS I/O standards at the differential transmitter, the SERDES
circuitry must be implemented in logic cells but not hard SERDES.
Altera Corporation
Stratix IV Device Handbook
Volume 1
8–12
Chapter 8: High-Speed Differential I/O Interfaces and DPA in Stratix IV Devices
Differential Transmitter
The load enable (LVDS_LOAD_EN) signal and the diffioclk signal (the clock running at
serial data rate) generated from PLL_Lx (left PLL) or PLL_Rx (right PLL) clocks the load
and shift registers. You can statically set the serialization factor to ×3, ×4, ×6, ×7, ×8, or
×10 using the Quartus II software. The load enable signal is derived from the
serialization factor setting. Figure 8–5 shows a block diagram of the Stratix IV
transmitter.
Figure 8–5. Stratix IV Transmitter
(1), (2)
Serializer
tx_in 10
DIN
2
IOE
IOE supports SDR, DDR, or
Non-Registered Datapath
+
-
DOUT
tx_out
FPGA
Fabric
LVDS Transmitter
tx_coreclock
3 (LVDS_LOAD_EN, diffioclk, tx_coreclock)
Left/Right PLL
tx_inclock
LVDS Clock Domain
Notes to Figure 8–5:
(1) In SDR and DDR modes, the data width is 1 and 2 bits, respectively.
(2) The tx_in port has a maximum data width of 10 bits.
You can configure any Stratix IV transmitter data channel to generate a
source-synchronous transmitter clock output. This flexibility allows the placement of
the output clock near the data outputs to simplify board layout and reduce
clock-to-data skew. Different applications often require specific clock-to-data
alignments or specific data-rate-to-clock-rate factors. The transmitter can output a
clock signal at the same rate as the data with a maximum frequency of 800 MHz. The
output clock can also be divided by a factor of 1, 2, 4, 6, 8, or 10, depending on the
serialization factor. You can set the phase of the clock in relation to the data at 0° or
180° (edge or center aligned). The left and right PLLs (PLL_Lx and PLL_Rx) provide
additional support for other phase shifts in 45° increments. These settings are made
statically in the Quartus II MegaWizard Plug-In Manager software.
Stratix IV Device Handbook
Volume 1
September 2012 Altera Corporation
Chapter 8: High-Speed Differential I/O Interfaces and DPA in Stratix IV Devices
Differential Transmitter
8–13
Figure 8–6 shows the Stratix IV transmitter in clock output mode. In clock output
mode, you can use an LVDS channel as a clock output channel.
Figure 8–6. Stratix IV Transmitter in Clock Output Mode
Transmitter Circuit
Parallel
Series
Txclkout+
Txclkout–
FPGA
Fabric
Left/Right
PLL
diffioclk
LVDS_LOAD_EN
You can bypass the Stratix IV serializer to support DDR (×2) and SDR (×1) operations
to achieve a serialization factor of 2 and 1, respectively. The I/O element (IOE)
contains two data output registers that can each operate in either DDR or SDR mode.
Figure 8–7 shows the serializer bypass path.
Figure 8–7. Serializer Bypass in Stratix IV Devices (1),
Serializer
tx_in 2
DIN
2
(2), (3)
IOE supports SDR, DDR, or
Non-Registered Datapath
IOE
+
-
DOUT
tx_out
FPGA
Fabric
LVDS Transmitter
tx_coreclock
(LVDS_LOAD_EN, diffioclk, tx_coreclock)
3
Left/Right PLL
Notes to Figure 8–7:
(1) All disabled blocks and signals are grayed out.
(2) In DDR mode, tx_inclock clocks the IOE register. In SDR mode, data is directly passed through the IOE.
(3) In SDR and DDR modes, the data width to the IOE is 1 and 2 bits, respectively.
September 2012
Altera Corporation
Stratix IV Device Handbook
Volume 1
8–14
Chapter 8: High-Speed Differential I/O Interfaces and DPA in Stratix IV Devices
Differential Transmitter
Programmable VOD and Programmable Pre-Emphasis
Stratix IV LVDS transmitters support programmable pre-emphasis and
programmable VOD. Pre-emphasis increases the amplitude of the high-frequency
component of the output signal, and thus helps to compensate for the
frequency-dependent attenuation along the transmission line. Figure 8–8 shows the
differential LVDS output.
Figure 8–8. Differential VOD
Single-Ended Waveform
Positive Channel (p)
VOD
Negative Channel (n)
VCM
Ground
VOD (diff peak - peak) = 2 x VOD(single-ended)
Differential Waveform
VOD
p - n = 0V
VOD
Figure 8–9 shows the LVDS output with pre-emphasis.
Figure 8–9. Programmable Pre-Emphasis (1)
OUT
VP
VOD
OUT
VP
Note to Figure 8–9:
(1) VP— voltage boost from pre-emphasis. VOD— Differential output voltage (peak-peak).
Stratix IV Device Handbook
Volume 1
September 2012 Altera Corporation
Chapter 8: High-Speed Differential I/O Interfaces and DPA in Stratix IV Devices
Differential Transmitter
8–15
Pre-emphasis is an important feature for high-speed transmission. Without
pre-emphasis, the output current is limited by the VOD setting and the output
impedance of the driver. At high frequency, the slew rate may not be fast enough to
reach full VOD before the next edge, producing pattern-dependent jitter.
With pre-emphasis, the output current is boosted momentarily during switching to
increase the output slew rate. The overshoot introduced by the extra current happens
only during switching and does not ring, unlike the overshoot caused by signal
reflection. The amount of pre-emphasis needed depends on the attenuation of the
high-frequency component along the transmission line. The Quartus II software
allows four settings for programmable pre-emphasis—zero (0), low (1), medium (2),
and high (3). The default setting is low.
The VOD is also programmable with four settings: low (0), medium low (1), medium
high (2), and high (3). The default setting is medium low.
Programmable VOD
You can statically assign the VOD settings from the Assignment Editor. Table 8–8 lists
the assignment name for programmable VOD and its possible values in the Quartus II
software Assignment Editor.
Table 8–8. Quartus II Software Assignment Editor
To
tx_out
Assignment name
Programmable Differential Output Voltage (VOD)
Allowed values
0, 1, 2, 3
Figure 8–10 shows the assignment of programmable VOD for a transmit data output
from the Quartus II software Assignment Editor.
Figure 8–10. Quartus II Software Assignment Editor—Programmable VOD
September 2012
Altera Corporation
Stratix IV Device Handbook
Volume 1
8–16
Chapter 8: High-Speed Differential I/O Interfaces and DPA in Stratix IV Devices
Differential Transmitter
Programmable Pre-Emphasis
Four different settings are allowed for pre-emphasis from the Assignment Editor for
each LVDS output channel. Table 8–9 lists the assignment name and its possible
values for programmable pre-emphasis in the Quartus II software Assignment Editor.
Table 8–9. Quartus II Software Assignment Editor
To
tx_out
Assignment name
Programmable Pre-emphasis
Allowed values
0, 1, 2, 3
Figure 8–11 shows the assignment of programmable pre-emphasis for a transmit data
output port from the Quartus II software Assignment Editor.
Figure 8–11. Quartus II Software Assignment Editor – Programmable Pre-Emphasis
Stratix IV Device Handbook
Volume 1
September 2012 Altera Corporation
Chapter 8: High-Speed Differential I/O Interfaces and DPA in Stratix IV Devices
Differential Receiver
8–17
Differential Receiver
The Stratix IV device family has a dedicated circuitry to receive high-speed
differential signals in row I/Os. Figure 8–12 shows the hardware blocks of the
Stratix IV receiver. The receiver has a differential buffer and left and right PLLs that
can be shared between the transmitter and receiver, a DPA block, a synchronizer, a
data realignment block, and a deserializer. The differential buffer can receive LVDS,
mini-LVDS, and RSDS signal levels, which are statically set in the Quartus II software
Assignment Editor.
The left and right PLL receives the external clock input and generates different phases
of the same clock. The DPA block chooses one of the clocks from the left and right PLL
and aligns the incoming data on each channel. The synchronizer circuit is a 1 bit wide
by 6 bit deep FIFO buffer that compensates for any phase difference between the DPA
clock and the data realignment block. If necessary, the user-controlled data
realignment circuitry inserts a single bit of latency in the serial bit stream to align to
the word boundary. The deserializer includes shift registers and parallel load
registers, and sends a maximum of 10 bits to the internal logic.
The Stratix IV device family supports three different receiver modes:
■
“Non-DPA Mode” on page 8–22
■
“DPA Mode” on page 8–24
■
“Soft-CDR Mode” on page 8–25
The physical medium connecting the transmitter and receiver LVDS channels may
introduce skew between the serial data and the source-synchronous clock. The
instantaneous skew between each LVDS channel and the clock also varies with the
jitter on the data and clock signals as seen by the receiver. The three different modes—
non-DPA, DPA, and soft-CDR—provide different options to overcome skew between
the source synchronous clock (non-DPA, DPA) /reference clock (soft-CDR) and the
serial data.
1
September 2012
Only non-DPA mode requires manual skew adjustment.
Altera Corporation
Stratix IV Device Handbook
Volume 1
8–18
Chapter 8: High-Speed Differential I/O Interfaces and DPA in Stratix IV Devices
Differential Receiver
Non-DPA mode allows you to statically select the optimal phase between the source
synchronous clock and the received serial data to compensate skew. In DPA mode, the
DPA circuitry automatically chooses the best phase to compensate for the skew
between the source synchronous clock and the received serial data. Soft-CDR mode
provides opportunities for synchronous and asynchronous applications for
chip-to-chip and short reach board-to-board applications for SGMII protocols.
Figure 8–12. Receiver Block Diagram
(1), (2)
LVDS Receiver
IOE Supports SDR, DDR, or Non-Registered Datapath
2
rx_out
+
IOE
10
rx_in
Synchronizer
Deserializer
Bit Slip
DOUT DIN
DOUT DIN
DPA Circuitry
Retimed
Data
DOUT DIN
DIN
FPGA
Fabric
2
DPA Clock
LVDS_diffiioclk
Clock Mux
rx_divfwdclk
DPA_diffioclk
diffioclk
(LOAD_EN, diffioclk)
3
(DPA_LOAD_EN,
DPA_diffioclk,
rx_divfwdclk)
rx_outclock
3
(LVDS_LOAD_EN,
LVDS_diffioclk,
rx_outclk)
8 Serial LVDS
Clock Phases
LVDS Clock Domain
DPA Clock Domain
Left/Right PLL
rx_inclock
Notes to Figure 8–12:
(1) In SDR and DDR modes, the data width from the IOE is 1 and 2 bits, respectively.
(2) The rx_out port has a maximum data width of 10 bits.
Differential I/O Termination
The Stratix IV device family provides a 100- on-chip differential termination option
on each differential receiver channel for LVDS standards. On-chip termination saves
board space by eliminating the need to add external resistors on the board. You can
enable on-chip termination in the Quartus II software Assignment Editor.
On-chip differential termination is supported on all row I/O pins and dedicated clock
input pins (CLK[0,2,9,11]). It is not supported for column I/O pins, dedicated clock
input pins (CLK[1,3,8,10]), or the corner PLL clock inputs.
Stratix IV Device Handbook
Volume 1
September 2012 Altera Corporation
Chapter 8: High-Speed Differential I/O Interfaces and DPA in Stratix IV Devices
Differential Receiver
8–19
Figure 8–13 shows device on-chip termination.
Figure 8–13. On-Chip Differential I/O Termination
Stratix IV Differential
Receiver with On-Chip
100 Ω Termination
LVDS
Transmitter
Z0 = 50 Ω
RD
Z0 = 50 Ω
Receiver Hardware Blocks
The differential receiver has the following hardware blocks:
■
“DPA Block” on page 8–19
■
“Synchronizer” on page 8–20
■
“Data Realignment Block (Bit Slip)” on page 8–20
■
“Deserializer” on page 8–22
DPA Block
The DPA block takes in high-speed serial data from the differential input buffer and
selects one of the eight phases generated by the left and right PLL to sample the data.
The DPA chooses a phase closest to the phase of the serial data. The maximum phase
offset between the received data and the selected phase is 1/8 UI, which is the
maximum quantization error of the DPA. The eight phases of the clock are equally
divided, offering a 45° resolution.
Figure 8–14 shows the possible phase relationships between the DPA clocks and the
incoming serial data.
Figure 8–14. DPA Clock Phase to Serial Data Timing Relationship
rx_in
D0
D1
D2
D3
D4
(1)
Dn
0˚
45˚
90˚
135˚
180˚
225˚
270˚
315˚
Tvco
0.125Tvco
Note to Figure 8–14:
(1) TVCO is defined as the PLL serial clock period.
September 2012
Altera Corporation
Stratix IV Device Handbook
Volume 1
8–20
Chapter 8: High-Speed Differential I/O Interfaces and DPA in Stratix IV Devices
Differential Receiver
The DPA block continuously monitors the phase of the incoming serial data and
selects a new clock phase if needed. You can prevent the DPA from selecting a new
clock phase by asserting the optional RX_DPLL_HOLD port, which is available for each
channel.
DPA circuitry does not require a fixed training pattern to lock to the optimum phase
out of the eight phases. After reset or power up, DPA circuitry requires transitions on
the received data to lock to the optimum phase. An optional output port,
RX_DPA_LOCKED, is available to indicate an initial DPA lock condition to the optimum
phase after power up or reset. This signal is not de-asserted if the DPA selects a new
phase out of the eight clock phases to sample the received data. Do not use the
rx_dpa_locked signal to determine a DPA loss-of-lock condition. Use data checkers
such as a cyclic redundancy check (CRC) or diagonal interleaved parity (DIP-4) to
validate the data.
An independent reset port, RX_RESET, is available to reset the DPA circuitry. DPA
circuitry must be retrained after reset.
1
The DPA block is bypassed in non-DPA mode.
Synchronizer
The synchronizer is a 1 bit wide and 6 bit deep FIFO buffer that compensates for the
phase difference between DPA_diffioclk, which is the optimal clock selected by the
DPA block, and LVDS_diffioclk, which is produced by the left and right PLL. The
synchronizer can only compensate for phase differences, not frequency differences
between the data and the receiver’s input reference clock.
An optional port, RX_FIFO_RESET, is available to the internal logic to reset the
synchronizer. The synchronizer is automatically reset when the DPA first locks to the
incoming data. Altera recommends using RX_FIFO_RESET to reset the synchronizer
when the DPA signals a loss-of-lock condition and the data checker indicates
corrupted received data.
1
The synchronizer circuit is bypassed in non-DPA and soft-CDR mode.
Data Realignment Block (Bit Slip)
Skew in the transmitted data along with skew added by the link causes
channel-to-channel skew on the received serial data streams. If the DPA is enabled,
the received data is captured with different clock phases on each channel. This may
cause the received data to be misaligned from channel to channel. To compensate for
this channel-to-channel skew and establish the correct received word boundary at
each channel, each receiver channel has a dedicated data realignment circuit that
realigns the data by inserting bit latencies into the serial stream.
An optional RX_CHANNEL_DATA_ALIGN port controls the bit insertion of each receiver
independently controlled from the internal logic. The data slips one bit on the rising
edge of RX_CHANNEL_DATA_ALIGN. The requirements for the RX_CHANNEL_DATA_ALIGN
signal include:
Stratix IV Device Handbook
Volume 1
■
The minimum pulse width is one period of the parallel clock in the logic array.
■
The minimum low time between pulses is one period of the parallel clock.
■
This is an edge-triggered signal.
September 2012 Altera Corporation
Chapter 8: High-Speed Differential I/O Interfaces and DPA in Stratix IV Devices
Differential Receiver
■
8–21
Valid data is available two parallel clock cycles after the rising edge of
RX_CHANNEL_DATA_ALIGN.
Figure 8–15 shows receiver output (RX_OUT) after one bit slip pulse with the
deserialization factor set to 4.
Figure 8–15. Data Realignment Timing
rx_inclock
rx_in
3
2
1
0
3
2
1
0
3
2
1
0
rx_outclock
rx_channel_data_align
rx_out
3210
321x
xx21
0321
The data realignment circuit can have up to 11 bit-times of insertion before a rollover
occurs. The programmable bit rollover point can be from 1 to 11 bit-times,
independent of the deserialization factor. The programmable bit rollover point must
be set equal to or greater than the deserialization factor, allowing enough depth in the
word alignment circuit to slip through a full word. You can set the value of the bit
rollover point using the MegaWizard Plug-In Manager software. An optional status
port, RX_CDA_MAX, is available to the FPGA fabric from each channel to indicate when
the preset rollover point is reached.
Figure 8–16 shows a preset value of four bit-times before rollover occurs. The
rx_cda_max signal pulses for one rx_outclock cycle to indicate that rollover has
occurred.
Figure 8–16. Receiver Data Re-alignment Rollover
rx_inclock
rx_channel_data_align
rx_outclock
rx_cda_max
September 2012
Altera Corporation
Stratix IV Device Handbook
Volume 1
8–22
Chapter 8: High-Speed Differential I/O Interfaces and DPA in Stratix IV Devices
Differential Receiver
Deserializer
You can statically set the deserialization factor to 3, 4, 6, 7, 8, or 10 by using the
Quartus II software. You can bypass the Stratix IV deserializer in the Quartus II
MegaWizard Plug-In Manager software to support DDR (×2) or SDR (×1) operations,
as shown Figure 8–17. The DPA and data realignment circuit cannot be used when the
deserializer is bypassed. The IOE contains two data input registers that can operate in
DDR or SDR mode.
Figure 8–17. Deserializer Bypass in Stratix IV Devices (1),
(2), (3)
LVDS Receiver
IOE Supports SDR, DDR, or Non-Registered Datapath
2
rx_out
+
IOE
2
rx_in
Synchronizer
Deserializer
Bit Slip
DOUT DIN
DOUT DIN
DPA Circuitry
Retimed
Data
DOUT DIN
DIN
FPGA
Fabric
2
DPA Clock
rx_divfwdclk
DPA_diffioclk
Clock Mux
LVDS_diffiioclk
diffioclk
(LOAD_EN, diffioclk)
3
(DPA_LOAD_EN,
DPA_diffioclk,
rx_divfwdclk)
rx_outclock
3
(LVDS_LOAD_EN,
LVDS_diffioclk,
rx_outclk)
8 Serial LVDS
Clock Phases
Left/Right PLL
Notes to Figure 8–17:
(1) All disabled blocks and signals are grayed out.
(2) In DDR mode, rx_inclock clocks the IOE register. In SDR mode, data is directly passed through the IOE.
(3) In SDR and DDR modes, the data width from the IOE is 1 and 2 bits, respectively.
Receiver Data Path Modes
The Stratix IV device family supports three receiver datapath modes—non-DPA
mode, DPA mode, and soft-CDR mode.
Non-DPA Mode
Figure 8–18 shows the non-DPA datapath block diagram. In non-DPA mode, the DPA
and synchronizer blocks are disabled. Input serial data is registered at the rising or
falling edge of the serial LVDS_diffioclk clock produced by the left and right PLL.
You can select the rising/falling edge option using the ALTLDVS MegaWizard
Plug-In Manager software. Both data realignment and deserializer blocks are clocked
by the LVDS_diffioclk clock, which is generated by the left and right PLL.
Stratix IV Device Handbook
Volume 1
September 2012 Altera Corporation
Chapter 8: High-Speed Differential I/O Interfaces and DPA in Stratix IV Devices
Differential Receiver
1
8–23
When using non-DPA receivers, you must drive the PLL from a dedicated and
compensated clock input pin. Compensated clock inputs are dedicated clock pins in
the same I/O bank as the PLL.
f For more information about dedicated and compensated clock inputs, refer to the
Clock Networks and PLLs in Stratix IV Devices chapter.
Figure 8–18. Receiver Data Path in Non-DPA Mode
(1), (2)
LVDS Receiver
IOE Supports SDR, DDR, or Non-Registered Datapath
2
rx_out
+
IOE
10
rx_in
Synchronizer
Deserializer
Bit Slip
DOUT DIN
DOUT DIN
DPA
P Circuitr y
Retimed
Data
DOUT DIN
N
DIN
FPGA
Fabric
2
DPA
P Clock
L
LVDS_diffiioclk
Clock Mux
rx_divfwdclk
DPA_diffioclk
P
diffioclk
(LOAD_EN, diffioclk)
3
(DPA_LO
P
AD_EN,
DPA_diffioclk,
P
rx_divfwdclk)
rx_outclock
3
(LVDS_LOAD_EN,
LVDS_diffioclk,
rx_outclk)
8 Serial LVDS
L
Clock Phases
Left/Right PLL
rx_inclock
LVDS Clock Domain
Notes to Figure 8–18:
(1) In SDR and DDR modes, the data width from the IOE is 1 and 2 bits, respectively.
(2) The rx_out port has a maximum data width of 10 bits.
September 2012
Altera Corporation
Stratix IV Device Handbook
Volume 1
8–24
Chapter 8: High-Speed Differential I/O Interfaces and DPA in Stratix IV Devices
Differential Receiver
DPA Mode
Figure 8–19 shows the DPA mode datapath, where all the hardware blocks mentioned
in “Receiver Hardware Blocks” on page 8–19 are active. The DPA block chooses the
best possible clock (DPA_diffioclk) from the eight fast clocks sent by the left and right
PLL. This serial DPA_diffioclk clock is used for writing the serial data into the
synchronizer. A serial LVDS_diffioclk clock is used for reading the serial data from
the synchronizer. The same LVDS_diffioclk clock is used in data realignment and
deserializer blocks.
Figure 8–19. Receiver Datapath in DPA Mode
(1), (2), (3)
LVDS Receiver
IOE Supports SDR, DDR, or Non-Registered Datapath
2
rx_out
+
IOE
10
rx_in
Synchronizer
Deserializer
Bit Slip
DOUT DIN
DOUT DIN
DPA Circuitry
Retimed
Data
DOUT DIN
DIN
FPGA
Fabric
2
DPA Clock
LVDS_diffiioclk
Clock Mux
rx_divfwdclk
DPA_diffioclk
diffioclk
(LOAD_EN, diffioclk)
3
(DPA_LOAD_EN,
DPA_diffioclk,
rx_divfwdclk)
rx_outclock
3
(LVDS_LOAD_EN,
LVDS_diffioclk,
rx_outclk)
8 Serial LVDS
Clock Phases
LVDS Clock Domain
DPA Clock Domain
Left/Right PLL
rx_inclock
Notes to Figure 8–19:
(1) All disabled blocks and signals are grayed out.
(2) In SDR and DDR modes, the data width from the IOE is 1 and 2 bits, respectively.
(3) The rx_out port has a maximum data width of 10 bits.
Stratix IV Device Handbook
Volume 1
September 2012 Altera Corporation
Chapter 8: High-Speed Differential I/O Interfaces and DPA in Stratix IV Devices
Differential Receiver
8–25
Soft-CDR Mode
The Stratix IV LVDS channel offers soft-CDR mode to support the Gigabit Ethernet
and SGMII protocols. A receiver PLL uses the local clock source for reference.
Figure 8–20 shows the soft-CDR mode datapath.
Figure 8–20. Receiver Datapath in Soft-CDR Mode
(1), (2), (3)
LVDS Receiver
IOE Supports SDR, DDR, or Non-Registered Datapath
2
rx_out
+
IOE
10
rx_in
Synchronizer
Deserializer
Bit Slip
DOUT DIN
DOUT DIN
DPA Circuitry
Retimed
Data
DOUT DIN
DIN
FPGA
Fabric
2
DPA Clock
LVDS_diffiioclk
Clock Mux
rx_divfwdclk
DPA_diffioclk
diffioclk
(LOAD_EN, diffioclk)
3
(DPA_LOAD_EN,
DPA_diffioclk,
rx_divfwdclk)
rx_outclock
3
(LVDS_LOAD_EN,
LVDS_diffioclk,
rx_outclk)
8 Serial LVDS
Clock Phases
LVDS Clock Domain
DPA Clock Domain
Left/Right PLL
rx_inclock
Notes to Figure 8–20:
(1) All disabled blocks and signals are grayed out.
(2) In SDR and DDR modes, the data width from the IOE is 1 and 2 bits, respectively.
(3) The rx_out port has a maximum data width of 10 bits.
In soft-CDR mode, the synchronizer block is inactive. The DPA circuitry selects an
optimal DPA clock phase to sample the data. Use the selected DPA clock for bit-slip
operation and deserialization. The DPA block also forwards the selected DPA clock,
divided by the deserialization factor called rx_divfwdclk, to the FPGA fabric, along
with the deserialized data. This clock signal is put on the periphery clock (PCLK)
network. When using soft-CDR mode, the rx_reset port must not be asserted after
the DPA training is asserted because the DPA will continuously choose new phase
taps from the PLL to track parts per million (PPM) differences between the reference
clock and incoming data.
f For more information about periphery clock networks, refer to the Clock Networks and
PLLs in Stratix IV Devices chapter.
September 2012
Altera Corporation
Stratix IV Device Handbook
Volume 1
8–26
Chapter 8: High-Speed Differential I/O Interfaces and DPA in Stratix IV Devices
LVDS Interface with the Use External PLL Option Enabled
You can use every LVDS channel in soft-CDR mode and can drive the FPGA fabric
using the periphery clock network in the Stratix IV device family. The rx_dpa_locked
signal is not valid in soft-CDR mode because the DPA continuously changes its phase
to track PPM differences between the upstream transmitter and the local receiver
input reference clocks. The parallel clock rx_outclock, generated by the left and right
PLL, is also forwarded to the FPGA fabric.
LVDS Interface with the Use External PLL Option Enabled
The ALTLVDS MegaWizard Plug-In Manager software provides an option for
implementing the LVDS interface with the Use External PLL option. With this option
enabled you can control the PLL settings, such as dynamically reconfiguring the PLL
to support different data rates, dynamic phase shift, and other settings. You also must
instantiate an ALTPLL megafunction to generate the various clock and load enable
signals.
When you enable the Use External PLL option with the ALTLVDS transmitter and
receiver, the following signals are required from the ALTPLL megafunction:
1
■
Serial clock input to the SERDES of the ALTLVDS transmitter and receiver
■
Load enable to the SERDES of the ALTLVDS transmitter and receiver
■
Parallel clock used to clock the transmitter FPGA fabric logic and parallel clock
used for the receiver rx_syncclock port and receiver FPGA fabric logic
■
Asynchronous PLL reset port of the ALTLVDS receiver
As an example, Table 8–10 describes the serial clock output, load enable output, and
parallel clock output generated on ports c0, c1, and c2, respectively, along with the
locked signal of the ALTPLL instance. You can choose any of the PLL output clock
ports to generate the interface clocks.
f With soft SERDES, a different clocking requirement is needed. For more information,
refer to the LVDS SERDES Transmitter/Receiver (ALTLVDS_RX/TX) Megafunction User
Guide.
1
The high-speed clock generated from the PLL is intended to clock the LVDS SERDES
circuitry only. Do not use the high-speed clock to drive other logic because the
allowed frequency to drive the core logic is restricted by the PLL FOUT specification.
For more information about the FOUT specification, refer to the DC and Switching
Characteristics for Stratix IV Devices chapter.
Table 8–10 lists the signal interface between the output ports of the ALTPLL
megafunction and the input ports of the ALTLVDS transmitter and receiver.
Table 8–10. Signal Interface Between ALTPLL and ALTLVDS_TX and ALTLVDS_RX Megafunctions (Part 1 of 2)
From the ALTPLL
Megafunction
Serial clock output (c0)
Load enable output (c1)
Stratix IV Device Handbook
Volume 1
To the ALTLVDS Transmitter
(1)
To the ALTLVDS Receiver
tx_inclock (serial clock input to the
transmitter)
rx_inclock (serial clock input)
tx_enable (load enable to the transmitter)
rx_enable (load enable for the
deserializer)
September 2012 Altera Corporation
Chapter 8: High-Speed Differential I/O Interfaces and DPA in Stratix IV Devices
LVDS Interface with the Use External PLL Option Enabled
8–27
Table 8–10. Signal Interface Between ALTPLL and ALTLVDS_TX and ALTLVDS_RX Megafunctions (Part 2 of 2)
From the ALTPLL
Megafunction
Parallel clock output (c2)
To the ALTLVDS Transmitter
To the ALTLVDS Receiver
Parallel clock used inside the transmitter core
logic in the FPGA fabric
rx_syncclock (parallel clock input) and
parallel clock used inside the receiver
core logic in the FPGA fabric
~(locked)
pll_areset (asynchronous PLL reset
port) (2)
—
Notes to Table 8–10:
(1) The serial clock output (c0) can only drive tx_inclock on the ALTLVDS transmitter and rx_inclock on the ALTLVDS receiver. This clock
cannot drive the core logic.
(2) The pll_areset signal is automatically enabled for the LVDS receiver in external PLL mode. This signal does not exist for LVDS transmitter
instantiation when the external PLL option is enabled.
1
The rx_syncclock port is automatically enabled in an LVDS receiver in external PLL
mode. The Quartus II compiler errors out if this port is not connected, as shown in
Figure 8–21.
When generating the ALTPLL megafunction, the Left/Right PLL option is configured
to set up the PLL in LVDS mode. Figure 8–21 shows the connection between the
ALTPLL and ALTLVDS_TX and ALTLVDS_RX megafunctions.
Figure 8–21. LVDS Interface with the ALTPLL Megafunction (1)
FPGA Fabric
LVDS Transmitter
(ALTLVDS)
tx_inclock
Transmitter Core Logic
tx_in
tx_enable
tx_coreclk
c0
c1
c2
rx_coreclk
Receiver Core Logic
LVDS Receiver
(ALTLVDS)
rx_inclock
rx_out
ALTPLL
inclk0
pll_areset
locked
rx_enable
rx_syncclock
pll_areset
Note to Figure 8–21:
(1) Instantiation of pll_areset is optional for the ALTPLL instantiation.
September 2012
Altera Corporation
Stratix IV Device Handbook
Volume 1
8–28
Chapter 8: High-Speed Differential I/O Interfaces and DPA in Stratix IV Devices
LVDS Interface with the Use External PLL Option Enabled
Example 8–1 shows how to generate three output clocks using an ALTPLL
megafunction.
Example 8–1. Generating Three Output Clocks Using an ALTPLL Megafunction
LVDS data rate = 1 Gbps; serialization factor = 10; input reference clock = 100 MHz
The following settings are used when generating the three output clocks using an ALTPLL megafunction.
The serial clock must be 1000 MHz and the parallel clock must be 100 MHz (serial clock divided by the
serialization factor):
■
■
■
c0
■
Frequency = 1000 MHz (multiplication factor = 10 and division factor = 1)
■
Phase shift = –180° with respect to the voltage-controlled oscillator (VCO) clock
■
Duty cycle = 50%
c1
■
Frequency = (1000/10) = 100 MHz (multiplication factor = 1 and division factor = 1)
■
Phase shift = (10 - 2) × 360/10 = 288° [(deserialization factor - 2)/deserialization factor] × 360°
■
Duty cycle = (100/10) = 10% (100 divided by the serialization factor)
c2
■
Frequency = (1000/10) = 100 MHz (multiplication factor = 1 and division factor = 1)
■
Phase shift = (–180/10) = –18° (c0 phase shift divided by the serialization factor)
■
Duty cycle = 50%
Stratix IV Device Handbook
Volume 1
September 2012 Altera Corporation
Chapter 8: High-Speed Differential I/O Interfaces and DPA in Stratix IV Devices
Left and Right PLLs (PLL_Lx and PLL_Rx)
8–29
The Equation 8–1 calculations for phase shift assume that the input clock and serial
data are edge aligned. Introducing a phase shift of –180° to sampling clock (c0)
ensures that the input data is center-aligned with respect to the c0, as shown in
Figure 8–22.
Figure 8–22. Phase Relationship for External PLL Interface Signals
inclk0
VCO clk
(internal PLL clk)
c0 (-180
phase shift)
c1 (288
phase shift)
c2 (-18
phase shift)
Serial data
D1
D2
D3
D4
D5
D6
D7
D8
D9
D10
Left and Right PLLs (PLL_Lx and PLL_Rx)
The Stratix IV device family contains up to eight left and right PLLs with up to four
PLLs located on the left side and four on the right side of the device. The left PLLs can
support high-speed differential I/O banks on the left side; the right PLLs can support
high-speed differential I/O banks on the right side of the device. The high-speed
differential I/O receiver and transmitter channels use these left and right PLLs to
generate the parallel clocks (rx_outclock and tx_outclock) and high-speed clocks
(diffioclk).
Figure 8–2 on page 8–3 and Figure 8–3 on page 8–4 show the locations of the left and
right PLLs for Stratix IV E, GT, and GX devices. The PLL VCO operates at the clock
frequency of the data rate. Clock switchover and dynamic reconfiguration are allowed
using the left and right PLL in high-speed differential I/O support mode.
f For more information, refer to the Clock Networks and PLLs in Stratix IV Devices
chapter.
September 2012
Altera Corporation
Stratix IV Device Handbook
Volume 1
8–30
Chapter 8: High-Speed Differential I/O Interfaces and DPA in Stratix IV Devices
Stratix IV Clocking
Stratix IV Clocking
The left and right PLLs feed into the differential transmitter and receive channels
through the LVDS and DPA clock network. The center left and right PLLs can clock
the transmitter and receive channels above and below them. The corner left and right
PLLs can drive I/Os in the banks adjacent to them.
Figure 8–23 shows center PLL clocking in the Stratix IV device family. For more
information about PLL clocking restrictions, refer to “Differential Pin Placement
Guidelines” on page 8–38.
Figure 8–23. LVDS/DPA Clocks in the Stratix IV Device Family with Center PLLs
4
LVDS
Clock
DPA
Clock
Quadrant
Quadrant
DPA
Clock
LVDS
4
Clock
4
4
2
Center
PLL_L2
Center
PLL_R2
Center
PLL_L3
Center
PLL_R3
2
2
2
4
4
4
LVDS
Clock
DPA
Clock
Quadrant
Quadrant
DPA
Clock
LVDS
4
Clock
Figure 8–24 shows center and corner PLL clocking in the Stratix IV device family. For
more information about PLL clocking restrictions, refer to “Differential Pin Placement
Guidelines” on page 8–38.
Figure 8–24. LVDS/DPA Clocks in the Stratix IV Device Family with Center and Corner PLLs
Corner
PLL_R1
Corner
PLL_L1
2
2
4
LVDS
Clock
DPA
Clock
Quadrant
Quadrant
DPA
Clock
LVDS
4
Clock
4
4
2
2
Center
PLL_L2
Center
PLL_R2
Center
PLL_L3
Center
PLL_R3
2
2
4
4
4
LVDS
Clock
DPA
Clock
Quadrant
Quadrant
DPA
Clock
LVDS 4
Clock
2
2
Corner
PLL_L4
Stratix IV Device Handbook
Volume 1
Corner
PLL_R4
September 2012 Altera Corporation
Chapter 8: High-Speed Differential I/O Interfaces and DPA in Stratix IV Devices
Source-Synchronous Timing Budget
8–31
Source-Synchronous Timing Budget
This section describes the timing budget, waveforms, and specifications for
source-synchronous signaling in the Stratix IV device family. LVDS I/O standards
enable high-speed data transmission. This high data transmission rate results in better
overall system performance. To take advantage of fast system performance, it is
important to understand how to analyze timing for these high-speed signals. Timing
analysis for the differential block is different from traditional synchronous timing
analysis techniques.
Instead of focusing on clock-to-output and setup times, source synchronous timing
analysis is based on the skew between the data and the clock signals. High-speed
differential data transmission requires the use of timing parameters provided by IC
vendors and is strongly influenced by board skew, cable skew, and clock jitter. This
section defines the source-synchronous differential data orientation timing
parameters, the timing budget definitions for the Stratix IV device family, and how to
use these timing parameters to determine a design’s maximum performance.
Differential Data Orientation
There is a set relationship between an external clock and the incoming data. For
operations at 1 Gbps and a serialization factor of 10, the external clock is multiplied by
10. You can set phase-alignment in the PLL to coincide with the sampling window of
each data bit. The data is sampled on the falling edge of the multiplied clock.
Figure 8–25 shows the data bit orientation of the ×10 mode.
Figure 8–25. Bit Orientation in the Quartus II Software
inclock/outclock
10 LVDS Bits
MSB
data in
9
8
7
6
5
4
3
LSB
2
1
0
Differential I/O Bit Position
Data synchronization is necessary for successful data transmission at high
frequencies. Figure 8–26 shows the data bit orientation for a channel operation. This
figure is based on the following:
September 2012
■
Serialization factor equals the clock multiplication factor
■
Edge alignment is selected for phase alignment
■
Implemented in hard SERDES
Altera Corporation
Stratix IV Device Handbook
Volume 1
8–32
Chapter 8: High-Speed Differential I/O Interfaces and DPA in Stratix IV Devices
Source-Synchronous Timing Budget
For other serialization factors, use the Quartus II software tools to find the bit position
within the word. Table 8–11 lists the bit positions after deserialization.
Figure 8–26. Bit-Order and Word Boundary for One Differential Channel (1)
Transmitter Channel
Operation (x8 Mode)
tx_outclock
tx_out
X
Current Cycle
Next Cycle
Previous Cycle
X X X X X X X 7 6 5 4 3 2 1 0 X X X X X X X X
MSB
LSB
Receiver Channel
Operation (x8 Mode)
rx_inclock
rx_in
7
6
5
4
3
2
1
0
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
rx_outclock
rx_out [7..0]
XXXXXXXX
XXXXXXXX
XXXX7654
3210XXXX
Note to Figure 8–26:
(1) These are only functional waveforms and are not intended to convey timing information.
Table 8–11 lists the conventions for differential bit naming for 18 differential channels.
The MSB and LSB positions increase with the number of channels used in a system.
Table 8–11. Differential Bit Naming
Internal 8-Bit Parallel Data
Receiver Channel Data Number
Stratix IV Device Handbook
Volume 1
MSB Position
LSB Position
1
7
0
2
15
8
3
23
16
4
31
24
5
39
32
6
47
40
7
55
48
8
63
56
9
71
64
10
79
72
11
87
80
12
95
88
13
103
96
14
111
104
15
119
112
16
127
120
17
135
128
18
143
136
September 2012 Altera Corporation
Chapter 8: High-Speed Differential I/O Interfaces and DPA in Stratix IV Devices
Source-Synchronous Timing Budget
8–33
Transmitter Channel-to-Channel Skew
Transmitter channel-to-channel skew (TCCS) is an important parameter based on the
Stratix IV transmitter in a source synchronous differential interface. This parameter is
used in receiver skew margin calculation. For more information, refer to “Receiver
Skew Margin for Non-DPA Mode” on page 8–33.
TCCS is the difference between the fastest and slowest data output transitions,
including the TCO variation and clock skew. For LVDS transmitters, the TimeQuest
Timing Analyzer provides a TCCS report, which shows TCCS values for serial output
ports.
f You can get the TCCS value from the TCCS report (report_TCCS) in the Quartus II
compilation report under the TimeQuest Timing Analyzer, or from the DC and
Switching Characteristics for Stratix IV Devices chapter.
Receiver Skew Margin for Non-DPA Mode
Changes in system environment, such as temperature, media (cable, connector, or
PCB), and loading effect the receiver’s setup and hold times; internal skew affects the
sampling ability of the receiver.
Different modes of LVDS receivers use different specifications that can help in
deciding the ability to sample the received serial data correctly. In DPA mode, you
must use DPA jitter tolerance instead of receiver input skew margin (RSKM).
In non-DPA mode, use TCCS, RSKM, and sampling window (SW) specifications for
high-speed source-synchronous differential signals in the receiver data path. The
relationship between RSKM, TCCS, and SW is expressed by the RSKM equation
shown in Equation 8–1.
Equation 8–1. RSKM
TUI – SW – TCCS
RSKM = ---------------------------------------------2
Conventions used for the equation:
September 2012
■
Time unit interval (TUI)—Time period of the serial data.
■
RSKM—The timing margin between the receiver’s clock input and the data input
sampling window.
■
SW—The period of time that the input data must be stable to ensure that data is
successfully sampled by the LVDS receiver. The SW is a device property and varies
with device speed grade.
■
TCCS—The timing difference between the fastest and the slowest output edges,
including tCO variation and clock skew, across channels driven by the same PLL.
The clock is included in the TCCS measurement.
Altera Corporation
Stratix IV Device Handbook
Volume 1
8–34
Chapter 8: High-Speed Differential I/O Interfaces and DPA in Stratix IV Devices
Source-Synchronous Timing Budget
Figure 8–27 shows the relationship between the RSKM, TCCS, and the receiver’s SW.
You must calculate the RSKM value to decide whether or not data can be sampled
properly by the LVDS receiver with the given data rate and device. A positive RSKM
value indicates that the LVDS receiver can sample the data properly, whereas a
negative RSKM indicates that it cannot.
Figure 8–27. Differential High-Speed Timing Diagram and Timing Budget for Non-DPA Mode
Timing Diagram
External
Input Clock
Time Unit Interval (TUI)
Internal
Clock
TCCS
TCCS
Receiver
Input Data
RSKM
SW
RSKM
Internal
Clock
Falling Edge
Timing Budget
TUI
External
Clock
Clock Placement
Internal
Clock
Synchronization
Transmitter
Output Data
RSKM
RSKM
TCCS
TCCS
2
Receiver
Input Data
SW
Stratix IV Device Handbook
Volume 1
September 2012 Altera Corporation
Chapter 8: High-Speed Differential I/O Interfaces and DPA in Stratix IV Devices
Source-Synchronous Timing Budget
8–35
For LVDS receivers, the Quartus II software provides an RSKM report showing the
SW, TUI, and RSKM values for non-DPA mode. You can generate the RSKM report by
executing the report_RSKM command in the TimeQuest Timing Analyzer. You can find
the RSKM report in the Quartus II compilation report under the TimeQuest Timing
Analyzer section.
1
In order to obtain the RSKM value, you must assign an appropriate input delay to the
LVDS receiver through the TimeQuest Timing Analyzer constraints menu.
For assigning input delay, follow these steps:
1. The Quartus II TimeQuest Timing Analyzer GUI has many options for setting the
constraints and analyzing the design. Figure 8–28 shows various commands on
the Constraints menu. For setting input delay, you must select the Set Input Delay
option.
Figure 8–28. Selection of Constraint Menu in TimeQuest Timing Analyzer
September 2012
Altera Corporation
Stratix IV Device Handbook
Volume 1
8–36
Chapter 8: High-Speed Differential I/O Interfaces and DPA in Stratix IV Devices
Source-Synchronous Timing Budget
2. Figure 8–29 shows the setting parameters for the Set Input Delay option. The
clock name must reference the source synchronous clock that feeds the LVDS
receiver. Select the desired clock using the pull-down menu.
Figure 8–29. Input Time Delay Assignment Through TimeQuest Timing Analyzer
3. Figure 8–30 shows the Targets option. You can view a list of all available ports
using the List option in the Name Finder window.
Figure 8–30. Name Finder Window in Set Input Delay Option
Stratix IV Device Handbook
Volume 1
September 2012 Altera Corporation
Chapter 8: High-Speed Differential I/O Interfaces and DPA in Stratix IV Devices
Source-Synchronous Timing Budget
8–37
4. Select the LVDS receiver serial input ports (from the list) according to the input
delay you set. Click OK.
5. In the Set Input Delay window, set the appropriate values in the Input Delay
Options section and Delay value.
6. Click Run to incorporate these values in the TimeQuest Timing Analyzer.
7. Assign the appropriate delay for all the LVDS receiver input ports following these
steps. If you have already assigned Input Delay and you need to add more delay
to that input port, use the Add Delay option in the Set Input Delay window.
1
If no input delay is set in the TimeQuest Timing Analyzer, the receiver
channel-to-channel skew (RCCS) defaults to zero. You can also directly set the input
delay in a Synopsys Design Constraint file (.sdc) using the set_input_delay
command.
f For more information about .sdc commands and the TimeQuest Timing Analyzer,
refer to the Quartus II TimeQuest Timing Analyzer chapter in volume 3 of the Quartus II
Development Software Handbook.
Example 8–2 shows the RSKM calculation.
Example 8–2. RSKM
Data Rate: 1 Gbps, Board channel-to-channel skew = 200 ps
For Stratix IV devices:
TCCS = 100 ps (pending characterization)
SW = 300 ps (pending characterization)
TUI = 1000 ps
Total RCCS = TCCS + Board channel-to-channel skew= 100 ps + 200 ps
= 300 ps
RSKM= TUI - SW - RCCS
= 1000 ps - 300 ps - 300 ps
= 400 ps > 0
Because the RSKM > 0 ps, receiver non-DPA mode must work correctly.
1
September 2012
You can also calculate RSKM using the steps described in “Guidelines for DPAEnabled Differential Channels” on page 8–38.
Altera Corporation
Stratix IV Device Handbook
Volume 1
8–38
Chapter 8: High-Speed Differential I/O Interfaces and DPA in Stratix IV Devices
Differential Pin Placement Guidelines
Differential Pin Placement Guidelines
To ensure proper high-speed operation, differential pin placement guidelines have
been established. The Quartus II compiler automatically checks that these guidelines
are followed and issues an error message if they are not met.
This section is divided into pin placement guidelines with and without DPA usage
because DPA usage adds some constraints on the placement of high-speed differential
channels.
1
DPA-enabled differential channels refer to DPA mode or soft-CDR mode; DPA
disabled channels refer to non-DPA mode.
Guidelines for DPA-Enabled Differential Channels
The Stratix IV device family has differential receivers and transmitters in I/O banks
on the left and right sides of the device. Each receiver has a dedicated DPA circuit to
align the phase of the clock to the data phase of its associated channel. When you use
DPA-enabled channels in differential banks, you must adhere to the guidelines listed
in the following sections.
DPA-Enabled Channels and Single-Ended I/Os
When you enable a DPA channel in a bank, both single-ended I/Os and differential
I/O standards are allowed in the bank.
■
Single-ended I/Os are allowed in the same I/O bank, as long as the single-ended
I/O standard uses the same VCCIO as the DPA-enabled differential I/O bank.
■
Single-ended inputs can be in the same logic array block (LAB) row as a
differential channel using the SERDES circuitry.
■
DDIO can be placed within the same LAB row as a SERDES differential channel
but half rate DDIO (single data rate) output pins cannot be placed within the same
LAB row as a receiver SERDES differential channel. The input register must be
implemented within the FPGA fabric logic.
DPA-Enabled Channel Driving Distance
If the number of DPA channels driven by each left and right PLL exceeds 25 LAB
rows, Altera recommends implementing data realignment (bit slip) circuitry for all
the DPA channels.
Using Corner and Center Left and Right PLLs
If a differential bank is being driven by two left and right PLLs, where the corner left
and right PLL is driving one group and the center left and right PLL is driving
another group, there must be at least one row of separation between the two groups of
DPA-enabled channels (refer to Figure 8–31). The two groups can operate at
independent frequencies.
Stratix IV Device Handbook
Volume 1
September 2012 Altera Corporation
Chapter 8: High-Speed Differential I/O Interfaces and DPA in Stratix IV Devices
Differential Pin Placement Guidelines
8–39
You do not need a separation if a single left and right PLL is driving the DPA-enabled
channels as well as DPA-disabled channels.
Figure 8–31. Corner and Center Left and Right PLLs Driving DPA-Enabled Differential I/Os in the
Same Bank
Corner
Left /Right PLL
Reference
CLK
DPA -enabled
Diff I/O
DPA - enabled
Diff I/O
DPA - enabled
Diff I/O
Channels
driven by
Corner
Left/Right
PLL
DPA - enabled
Diff I/O
DPA - enabled
Diff I/O
Diff I/O
One Unused
Channel for Buffer
DPA-enabled
Diff I/O
DPA-enabled
Diff I/O
DPA -enabled
Diff I/O
Channels
driven by
Center
Left/Right
PLL
DPA- enabled
Diff I/O
Reference
CLK
Center
Left /Right PLL
September 2012
Altera Corporation
Stratix IV Device Handbook
Volume 1
8–40
Chapter 8: High-Speed Differential I/O Interfaces and DPA in Stratix IV Devices
Differential Pin Placement Guidelines
Using Both Center Left and Right PLLs
You can use both center left and right PLLs to drive DPA-enabled channels
simultaneously, as long as they drive these channels in their adjacent banks only, as
shown in Figure 8–32.
If one of the center left and right PLLs drives the top and bottom banks, you cannot
use the other center left and right PLL to drive differential channels, as shown in
Figure 8–32.
If the top PLL_L2 and PLL_R2 drives DPA-enabled channels in the lower differential
bank, the PLL_L3 and PLL_R3 cannot drive DPA-enabled channels in the upper
differential banks and vice versa. In other words, the center left and right PLLs cannot
drive cross-banks simultaneously, as shown in Figure 8–33.
Figure 8–32. Center Left and Right PLLs Driving DPA-Enabled Differential I/Os
Stratix IV Device Handbook
Volume 1
DPA-enabled
Diff I/O
DPA-enabled
Diff I/O
DPA-enabled
Diff I/O
DPA-enabled
Diff I/O
DPA-enabled
Diff I/O
Reference
CLK
DPA-enabled
Diff I/O
DPA-enabled
Diff I/O
DPA-enabled
Diff I/O
Reference
CLK
Center
Left/Right PLL
(PLL_L2/PLL_R2)
Center
Left/Right PLL
(PLL_L2/PLL_R2)
Center
Left/Right PLL
(PLL_L3/PLL_R3)
Center
Left/Right PLL
(PLL_L3/PLL_R3)
Reference
CLK
DPA-enabled
Diff I/O
DPA-enabled
Diff I/O
DPA-enabled
Diff I/O
Reference
CLK
DPA-enabled
Diff I/O
DPA-enabled
Diff I/O
DPA-enabled
Diff I/O
DPA-enabled
Diff I/O
DPA-enabled
Diff I/O
Unused
PLL
September 2012 Altera Corporation
Chapter 8: High-Speed Differential I/O Interfaces and DPA in Stratix IV Devices
Differential Pin Placement Guidelines
8–41
Figure 8–33. Invalid Placement of DPA-Enabled Differential I/Os Driven by Both Center Left and
Right PLLs
DPA-enabled
Diff I/O
DPA-enabled
Diff I/O
DPA-enabled
Diff I/O
DPA-enabled
Diff I/O
Reference
CLK
Center Left /Right
PLL
Center Left /Right
PLL
Reference
CLK
DPA-enabled
Diff I/O
DPA-enabled
Diff I/O
DPA-enabled
Diff I/O
DPA-enabled
Diff I/O
September 2012
Altera Corporation
Stratix IV Device Handbook
Volume 1
8–42
Chapter 8: High-Speed Differential I/O Interfaces and DPA in Stratix IV Devices
Differential Pin Placement Guidelines
Guidelines for DPA-Disabled Differential Channels
When you use DPA-disabled channels in the left and right banks of a Stratix IV
device, you must adhere to the guidelines in the following sections.
1
When using non-DPA receivers, you must drive the PLL from a dedicated and
compensated clock input pin. Compensated clock inputs are dedicated clock pins in
the same I/O bank as the PLL.
f For more information about dedicated and compensated clock inputs, refer to the
Clock Networks and PLLs in Stratix IV Devices chapter.
DPA-Disabled Channels and Single-Ended I/Os
The placement rules for DPA-disabled channels and single-ended I/Os are the same
as those for DPA-enabled channels and single-ended I/Os. For more information,
refer to “DPA-Enabled Channels and Single-Ended I/Os” on page 8–38.
DPA-Disabled Channel Driving Distance
Each left and right PLL can drive all the DPA-disabled channels in the entire bank.
Using Corner and Center Left and Right PLLs
You can use a corner left and right PLL to drive all transmitter channels and a center
left and right PLL to drive all DPA-disabled receiver channels within the same
differential bank. In other words, a transmitter channel and a receiver channel in the
same LAB row can be driven by two different PLLs, as shown in Figure 8–34.
A corner left and right PLL and a center left and right PLL can drive duplex channels
in the same differential bank, as long as the channels driven by each PLL are not
interleaved. Separation is not necessary between the group of channels driven by the
corner and center left and right PLLs, as shown in Figure 8–34 and Figure 8–35.
Stratix IV Device Handbook
Volume 1
September 2012 Altera Corporation
Chapter 8: High-Speed Differential I/O Interfaces and DPA in Stratix IV Devices
Differential Pin Placement Guidelines
8–43
Figure 8–34. Corner and Center Left and Right PLLs Driving DPA-Disabled Differential I/Os in the
Same Bank
Corner Left/Right
Corner Left/ Right
PLL
PLL
Reference
CLK
Diff RX
Diff TX
Diff RX
Diff TX
Diff RX
Diff TX
Diff RX
Diff TX
Diff RX
Diff TX
Diff RX
Diff TX
Diff RX
Diff TX
Diff RX
Diff TX
Diff RX
Diff TX
DPA-disabled
Diff I/O
Diff RX
Diff TX
DPA -disabled
Diff I /O
Reference
CLK
Center Left/Right
PLL
September 2012
Altera Corporation
Reference
CLK
DPA-disabled
Diff I/O
DPA-disabled
Diff I/O
DPA-disabled
Diff I/O
DPA-disabled
Diff I/O
DPA-disabled
Diff I/O
DPA-disabled
Diff I/O
DPA-disabled
Diff I/O
DPA-disabled
Diff I/O
Channels
driven by
Corner
Left/Right
PLL
No
separation
buffer
needed
Channels
driven by
Center
Left/Right
PLL
Reference
CLK
Center Left/Right
PLL
Stratix IV Device Handbook
Volume 1
8–44
Chapter 8: High-Speed Differential I/O Interfaces and DPA in Stratix IV Devices
Differential Pin Placement Guidelines
Figure 8–35. Invalid Placement of DPA-Disabled Differential I/Os Due to Interleaving of Channels
Driven by the Corner and Center Left and Right PLLs
Corner Left/Right
PLL
Reference CLK
DPA-disabled
Diff I/O
DPA-disabled
Diff I/O
DPA-disabled
Diff I/O
DPA-disabled
Diff I/O
DPA-disabled
Diff I/O
DPA-disabled
Diff I/O
DPA-disabled
Diff I/O
DPA-disabled
Diff I/O
DPA-disabled
Diff I/O
DPA-disabled
Diff I/O
Reference CLK
Center Left/Right
PLL
Stratix IV Device Handbook
Volume 1
September 2012 Altera Corporation
Chapter 8: High-Speed Differential I/O Interfaces and DPA in Stratix IV Devices
Differential Pin Placement Guidelines
8–45
Using Both Center Left and Right PLLs
You can use both center left and right PLLs simultaneously to drive DPA-disabled
channels on upper and lower differential banks. Unlike DPA-enabled channels, the
center left and right PLLs can drive cross-banks. For example, the upper-center left
and right PLL can drive the lower differential bank at the same time the lower center
left and right PLL is driving the upper differential bank, and vice versa, as shown in
Figure 8–36.
Figure 8–36. Both Center Left and Right PLLs Driving Cross-Bank DPA-Disabled Channels
Simultaneously
DPA-disabled
Diff I/O
DPA-disabled
Diff I/O
DPA-disabled
Diff I/O
DPA-disabled
Diff I/O
Reference
CLK
Center
Left/Right PLL
Center
Left/Right PLL
Reference
CLK
DPA-disabled
Diff I/O
DPA-disabled
Diff I/O
DPA-disabled
Diff I/O
DPA-disabled
Diff I/O
September 2012
Altera Corporation
Stratix IV Device Handbook
Volume 1
8–46
Chapter 8: High-Speed Differential I/O Interfaces and DPA in Stratix IV Devices
Differential Pin Placement Guidelines
Document Revision History
Table 8–12 lists the revision history for this chapter.
Table 8–12. Document Revision History (Part 1 of 2)
Date
Version
September 2012
December 2011
February 2011
March 2010
November 2009
June 2009
April 2009
March 2009
Stratix IV Device Handbook
Volume 1
3.4
3.3
Changes
■
Updated Figure 8–22 to close FB case #.28708.
■
Updated the “Soft-CDR Mode” section to close FB #41886.
■
Updated the “Overview” and “ALTLVDS Port List” sections.
■
Updated Table 8–10.
■
Updated Table 8–10.
■
Updated the “Differential Transmitter”, “Non-DPA Mode”, “LVDS Interface with the Use
External PLL Option Enabled”, “Deserializer”, and “Guidelines for DPA-Disabled
Differential Channels” sections.
■
Applied new template.
■
Minor text edits.
■
Removed note 7 from Table 8–1 and Table 8–2.
■
Updated Figure 8–5.
■
Updated the “LVDS Channels” section.
■
Updated Table 8–7.
■
Added a note to the “LVDS Interface with the Use External PLL Option Enabled” and
“ALTLVDS Port List” sections.
■
Minor text edits.
■
Changed “dedicated LVDS” to “true LVDS”.
■
Removed EP4SE110, EP4SE290, and EP4SE680 devices.
■
Added EP4SE820 and Stratix IV GT devices.
■
Updated “LVDS Channels”, “Differential Transmitter”, “Soft-CDR Mode”, and “DPAEnabled Channels and Single-Ended I/Os” sections.
■
Updated Table 8–1, Table 8–2, Table 8–5, and Table 8–6.
■
Added Table 8–3 and Table 8–4.
■
Updated Example 8–1.
■
Updated Figure 8–22.
■
Minor text edits.
■
Added an introductory paragraph to increase search ability.
■
Minor text edits.
■
Updated “Introduction”.
■
Updated Figure 8–3.
■
Removed Table 8-5 and Table 8-6.
■
Updated “Introduction”, “Stratix IV LVDS Channels”, “Stratix IV Differential Transmitter”,
“Differential I/O Termination”, and “Dynamic Phase Alignment (DPA) Block” sections.
■
Updated Table 8–1, Table 8–2, Table 8–3, Table 8–4, and Table 8–7.
■
Added Table 8–5 and Table 8–6.
■
Updated Figure 8–2.
■
Removed “Referenced Documents” section.
3.2
3.1
3.0
2.3
2.2
2.1
September 2012 Altera Corporation
Chapter 8: High-Speed Differential I/O Interfaces and DPA in Stratix IV Devices
Differential Pin Placement Guidelines
8–47
Table 8–12. Document Revision History (Part 2 of 2)
Date
Version
November 2008
May 2008
September 2012
2.0
1.0
Altera Corporation
Changes
■
Updated Figure 8–2, Figure 8–3, Figure 8–21, Figure 8–34.
■
Removed Figure 8–31.
■
Updated Table 8–1, Table 8–10.
■
Updated “Differential Pin Placement Guidelines” section.
Initial release.
Stratix IV Device Handbook
Volume 1
8–48
Stratix IV Device Handbook
Volume 1
Chapter 8: High-Speed Differential I/O Interfaces and DPA in Stratix IV Devices
Differential Pin Placement Guidelines
September 2012 Altera Corporation
Section III. System Integration
This section includes the following chapters:
■
Chapter 9, Hot Socketing and Power-On Reset in Stratix IV Devices
■
Chapter 10, Configuration, Design Security, and Remote System Upgrades in
Stratix IV Devices
■
Chapter 11, SEU Mitigation in Stratix IV Devices
■
Chapter 12, JTAG Boundary-Scan Testing in Stratix IV Devices
■
Chapter 13, Power Management in Stratix IV Devices
Revision History
Refer to each chapter for its own specific revision history. For information on when
each chapter was updated, refer to the Chapter Revision Dates section, which appears
in the full handbook.
September 2012
Altera Corporation
Stratix IV Device Handbook
Volume 1
III–2
Stratix IV Device Handbook
Volume 1
Section III: System Integration
September 2012 Altera Corporation
9. Hot Socketing and Power-On Reset in
Stratix IV Devices
February 2011
SIV51009-3.2
SIV51009-3.2
This chapter describes hot-socketing specifications, power-on reset (POR)
requirements, and their implementation in Stratix® IV devices.
Stratix IV devices offer hot socketing, also known as hot plug-in or hot swap, and
power sequencing support without the use of external devices. You can insert or
remove a Stratix IV device or a board in a system during system operation without
causing undesirable effects to the running system bus or board that is inserted into the
system.
The hot-socketing feature also removes some of the difficulty when you use Stratix IV
devices on PCBs that contain a mixture of 3.0-, 2.5-, 1.8-, 1.5-, and 1.2-V devices.
The Stratix IV hot-socketing feature provides:
■
Board or device insertion and removal without external components or board
manipulation
■
Support for any power-up sequence with the exception that VCC must power up
fully before VCCAUX for all Stratix IV production devices
■
I/O buffers non-intrusive to system buses during hot insertion
This section also describes POR circuitry in Stratix IV devices. POR circuitry keeps the
devices in the reset state until the power supply outputs are within operating range
(provided VCC powers up fully before VCCAUX).
This chapter contains the following sections:
■
“Stratix IV Hot-Socketing Specifications”
■
“Hot-Socketing Feature Implementation in Stratix IV Devices” on page 9–2
■
“Power-On Reset Circuitry” on page 9–3
■
“Power-On Reset Specifications” on page 9–4
Stratix IV Hot-Socketing Specifications
Stratix IV devices are hot-socketing compliant without the need for external
components or special design requirements. Hot-socketing support in Stratix IV
devices has the following advantages:
■
“Stratix IV Devices can be Driven Before Power Up” on page 9–2
■
“I/O Pins Remain Tri-Stated During Power Up” on page 9–2
■
“Insertion or Removal of a Stratix IV Device from a Powered-Up System” on
page 9–2
© 2011 Altera Corporation. All rights reserved. ALTERA, ARRIA, CYCLONE, HARDCOPY, MAX, MEGACORE, NIOS, QUARTUS and STRATIX words and logos
are trademarks of Altera Corporation and registered in the U.S. Patent and Trademark Office and in other countries. All other words and logos identified as
trademarks or service marks are the property of their respective holders as described at www.altera.com/common/legal.html. Altera warrants performance of its
semiconductor products to current specifications in accordance with Altera's standard warranty, but reserves the right to make changes to any products and
services at any time without notice. Altera assumes no responsibility or liability arising out of the application or use of any information, product, or service
described herein except as expressly agreed to in writing by Altera. Altera customers are advised to obtain the latest version of device specifications before relying
on any published information and before placing orders for products or services.
ISO
9001:2008
Registered
Stratix IV Device Handbook
Volume 1
February 2011
Feedback Subscribe
9–2
Chapter 9: Hot Socketing and Power-On Reset in Stratix IV Devices
Stratix IV Hot-Socketing Specifications
Stratix IV Devices can be Driven Before Power Up
You can drive signals into I/O pins, dedicated input pins, and dedicated clock pins of
Stratix IV devices before or during power up or power down without damaging the
device.
I/O Pins Remain Tri-Stated During Power Up
A device that does not support hot socketing can interrupt system operation or cause
contention by driving out before or during power up. In a hot-socketing situation, the
Stratix IV device’s output buffers are turned off during system power up or power
down. Also, the Stratix IV device does not drive out until the device is configured and
working within the recommended operating conditions.
Insertion or Removal of a Stratix IV Device from a Powered-Up System
Devices that do not support hot socketing can short power supplies when powered
up through the device signal pins. This irregular power up can damage both the
driving and driven devices and can disrupt card power up.
You can insert a Stratix IV device into or remove it from a powered-up system board
without damaging the system board or interfering with its operation.
You can power up or power down the VCCIO, VCC, VCCPGM, and VCCPD supplies in any
sequence (with any time between them) which are monitored by the hot socket circuit.
In addition, all other power supplies for the device can be powered up or down in any
sequence. Individual power supply ramp-up and ramp-down rates range from 50 µs
to 100 ms. During hot socketing, the I/O pin capacitance is less than 15 pF and the
clock pin capacitance is less than 20 pF.
1
To successfully power-up and exit POR on production devices, fully power VCC
before VCCAUX begins to ramp.
A possible concern regarding hot socketing is the potential for “latch-up.” Stratix IV
devices are immune to latch-up when hot socketing. Latch-up can occur when
electrical subsystems are hot socketed into an active system. During hot socketing, the
signal pins can be connected and driven by the active system before the power supply
can provide current to the device’s power and ground planes. This condition can lead
to latch-up and cause a low-impedance path from power to ground within the device.
As a result, the device draws a large amount of current, possibly causing electrical
damage.
Stratix IV Device Handbook
Volume 1
February 2011 Altera Corporation
Chapter 9: Hot Socketing and Power-On Reset in Stratix IV Devices
Hot-Socketing Feature Implementation in Stratix IV Devices
9–3
Hot-Socketing Feature Implementation in Stratix IV Devices
The hot-socketing feature turns off the output buffer during power up and power
down of the VCC, VCCAUX, VCCIO, VCCPGM, or VCCPD power supplies. The hot-socketing
circuitry generates an internal HOTSCKT signal when the VCC, VCCAUX, VCCIO, VCCPGM,
or VCCPD power supplies are below the threshold voltage. Hot-socketing circuitry is
designed to prevent excess I/O leakage during power up. When the voltage ramps up
very slowly, it is still relatively low, even after the POR signal is released and the
configuration is completed. The CONF_DONE, nCEO, and nSTATUS pins fail to respond, as
the output buffer cannot flip from the state set by the hot-socketing circuit at this low
voltage. Therefore, the hot-socketing circuitry has been removed from these
configuration pins to make sure that they are able to operate during configuration.
Thus, it is expected behavior for these pins to drive out during power-up and
power-down sequences.
Figure 9–1 shows the Stratix IV device’s I/O pin circuitry.
Figure 9–1. Hot-Socketing Circuitry for Stratix IV Devices
Power On
Reset
Monitor
VCCIO
Weak
Pull-Up
Resistor
PAD
R
Output Enable
Voltage
Tolerance
Control
Hot Socket
Output
Pre-Driver
Input Buffer
to Logic Array
The POR circuit monitors the voltage level of the power supplies (VCC, VCCAUX, VCCPT,
VCCPGM, and VCCPD) and keeps the I/O pins tri-stated until the device is in user mode.
The weak pull-up resistor (R) in the Stratix IV input/output element (IOE) keeps the
I/O pins from floating. The 3.0-V tolerance control circuit permits the I/O pins to be
driven by 3.0 V before the VCC, VCCAUX, VCCPT, VCCPGM, or VCCPD supplies are
powered. It also prevents the I/O pins from driving out when the device is not in user
mode. To successfully power-up and exit POR on production devices, fully power
VCC before VCCAUX begins to ramp.
1
February 2011
Altera uses GND as a reference for hot-socketing operations and I/O buffer designs.
To ensure proper operation, you must connect the GND between boards before
connecting the power supplies. This prevents the GND on your board from being
pulled up inadvertently by a path to power through other components on your board.
A pulled up GND could otherwise cause an out-of-specification I/O voltage or
current condition with the Altera device.
Altera Corporation
Stratix IV Device Handbook
Volume 1
9–4
Chapter 9: Hot Socketing and Power-On Reset in Stratix IV Devices
Power-On Reset Circuitry
Power-On Reset Circuitry
When power is applied to a Stratix IV device, a POR event occurs if the power supply
reaches the recommended operating range within the maximum power supply ramp
time (tRAMP). If tRAMP is not met, the device I/O pins and programming registers
remain tri-stated, during which device configuration could fail. The maximum tRAMP
for Stratix IV devices is 100 ms; the minimum tRAMP is 50 µs. When the PORSEL pin is
high, the maximum TRAMP for Stratix IV devices is 4 ms.
Stratix IV devices provide a dedicated input pin (PORSEL) to select a POR delay time
during power up. When the PORSEL pin is connected to GND, the POR delay time is
100 to 300 ms. When the PORSEL pin is set to high, the POR delay time is 4 to 12 ms.
The POR block consists of a regulator POR, satellite POR, and main POR to check the
power supply levels for proper device configuration.
The satellite POR monitors the following:
1
■
VCCPD and VCCPGM power supplies that are used in the I/O buffers and for device
programming
■
VCCAUX power supply which is the auxiliary supply for the programmable power
technology
■
VCC and VCCPT power supplies that are used in the device core
Altera requires powering up VCC before VCCAUX.
The main POR waits for satellite POR and the regulator POR to release the POR
signal. Until the release of the POR signal, the device configuration cannot start.
The internal configuration memory supply that is used during device configuration is
checked by the regulator POR block and is gated in the main POR block for the final
POR trip. Figure 9–2 shows a simplified diagram of the POR block.
1
All configuration-related dedicated and dual function I/O pins must be powered by
VCCPGM.
Figure 9–2. Simplified POR Diagram for Stratix IV Devices
Regulator POR
Main POR
V CCPGM
V CCPD
POR Pulse
Setting
V CC
Satellite POR
POR
V CCPT
V CCAUX
Stratix IV Device Handbook
Volume 1
PORSEL
February 2011 Altera Corporation
Chapter 9: Hot Socketing and Power-On Reset in Stratix IV Devices
Power-On Reset Specifications
9–5
Power-On Reset Specifications
Table 9–1 lists the power supplies that the POR circuit monitors.
1
Altera requires powering up VCC before VCCAUX.
Table 9–1. Power Supplies Monitored by the POR Circuitry
Power Supply
Description
Setting (V)
VCC
Core and periphery power supply
0.9
VCCPT
Programmable power technology power supply
1.5
VCCPD
I/O pre-driver power supply
VCCPGM
Configuration pins power supply
VCCAUX
Auxiliary supply for the programmable power technology
2.5, 3.0
1.8, 2.5, 3.0
2.5
Table 9–2 lists the power supplies that the POR circuit does not monitor.
Table 9–2. Power Supplies Not Monitored by the POR Circuitry (Note 1)
Power Supply
Description
Setting (V)
1.2, 1.5, 1.8,
2.5, 3.0
VCCIO
I/O power supply
VCCA_PLL
PLL analog global power supply
2.5
VCCD_PLL
PLL digital power supply
0.9
VCC_CLKIN
PLL differential clock input power supply (top and bottom I/O
banks only)
2.5
VCCBAT
Battery back-up power supply for design security volatile key
storage
1.2-3.3
Note to Table 9–2:
(1) The transceiver supplies are not monitored by POR.
1
VCCIO, VCCA_PLL, VCCD_PLL, VCC_CLKIN, and VCCBAT are not monitored by POR and have
no affect on the device configuration.
The POR specification is designed to ensure that all the circuits in the Stratix IV device
are at certain known states during power up.
The POR signal pulse width is programmable using the PORSEL input pin. When the
PORSEL pin is connected to GND, the POR delay time is 100 to 300 ms. When the
PORSEL pin is set to high, the POR delay time is 4 to 12 ms.
f For more information about the POR specification, refer to the DC and Switching
Characteristics for Stratix IV Devices chapter.
February 2011
Altera Corporation
Stratix IV Device Handbook
Volume 1
9–6
Chapter 9: Hot Socketing and Power-On Reset in Stratix IV Devices
Power-On Reset Specifications
Document Revision History
Table 9–3 lists the revision history for this chapter.
Table 9–3. Document Revision History
Date
Version
February 2011
March 2010
November 2009
June 2009
March 2009
November 2008
Changes
■
Updated Table 9–2.
■
Updated the “Power-On Reset Circuitry”, “Power-On Reset Specifications”, and “Insertion
or Removal of a Stratix IV Device from a Powered-Up System” sections.
■
Applied new template.
■
Minor text edits.
■
Updated the introduction and the “Stratix IV Hot-Socketing Specifications”, “Insertion or
Removal of a Stratix IV Device from a Powered-Up System”, “Hot-Socketing Feature
Implementation in Stratix IV Devices”, “Power-On Reset Circuitry”, and “Power-On Reset
Specifications” sections.
■
Updated Table 9–1 and Table 9–2.
■
Updated Figure 9–2.
■
Minor text edits.
■
Updated graphics.
■
Minor text edits.
■
Updated Table 9–2.
■
Added introductory sentences to improve search ability.
■
Removed the Conclusion section.
■
Minor text edits.
■
Changed all “Stratix IV E” to “Stratix IV”.
■
Updated “Stratix IV Hot-Socketing Specifications” and “Hot-Socketing Feature
Implementation in Stratix IV Devices” sections.
■
Updated Figure 9–2.
■
Removed “Referenced Documents” section.
■
Updated “Hot-Socketing Feature Implementation in Stratix IV Devices” on page 9–2.
■
Updated “Power-On Reset Circuitry” on page 9–4.
■
Updated Table 9–1.
■
Made minor editorial changes.
3.2
3.1
3.0
2.2
2.1
2.0
July 2008
1.1
Revised “Introduction”.
May 2008
1.0
Initial release.
Stratix IV Device Handbook
Volume 1
February 2011 Altera Corporation
10. Configuration, Design Security, and
Remote System Upgrades in Stratix IV
Devices
September 2012
SIV51010-3.5
SIV51010-3.5
This chapter describes the configuration, design security, and remote system
upgrades in Stratix® IV devices. To save configuration memory space and time,
Stratix IV devices provide configuration data decompression. They also provide a
built-in design security feature that protects your designs against IP theft and
tampering of your configuration files.
Stratix IV devices also offer remote system upgrade capability so that you can
upgrade your system in real-time through any network. This helps to deliver feature
enhancements and bug fixes and provides error detection, recovery, and status
information to ensure reliable reconfiguration.
Overview
This chapter describes supported configuration schemes for Stratix IV devices,
instructions about how to execute the required configuration schemes, and the
necessary pin settings.
Stratix IV devices use SRAM cells to store configuration data. As SRAM is volatile,
you must download configuration data to the Stratix IV device each time the device
powers up. You can configure Stratix IV devices using one of four configuration
schemes:
■
Fast passive parallel (FPP)
■
Fast active serial (AS)
■
Passive serial (PS)
■
Joint Test Action Group (JTAG)
All configuration schemes use either an external controller (for example, a MAX® II
device or microprocessor), a configuration device, or a download cable. For more
information, refer to “Configuration Features” on page 10–4.
This chapter includes the following sections:
■
“Configuration Schemes” on page 10–2
■
“Configuration Features” on page 10–4
■
“Fast Passive Parallel Configuration” on page 10–6
■
“Fast Active Serial Configuration (Serial Configuration Devices)” on page 10–16
■
“Passive Serial Configuration” on page 10–24
■
“JTAG Configuration” on page 10–34
■
“Device Configuration Pins” on page 10–39
© 2012 Altera Corporation. All rights reserved. ALTERA, ARRIA, CYCLONE, HARDCOPY, MAX, MEGACORE, NIOS, QUARTUS and STRATIX words and logos
are trademarks of Altera Corporation and registered in the U.S. Patent and Trademark Office and in other countries. All other words and logos identified as
trademarks or service marks are the property of their respective holders as described at www.altera.com/common/legal.html. Altera warrants performance of its
semiconductor products to current specifications in accordance with Altera's standard warranty, but reserves the right to make changes to any products and
services at any time without notice. Altera assumes no responsibility or liability arising out of the application or use of any information, product, or service
described herein except as expressly agreed to in writing by Altera. Altera customers are advised to obtain the latest version of device specifications before relying
on any published information and before placing orders for products or services.
ISO
9001:2008
Registered
Stratix IV Device Handbook
Volume 1
September 2012
Feedback Subscribe
10–2
Chapter 10: Configuration, Design Security, and Remote System Upgrades in Stratix IV Devices
Configuration Schemes
■
“Configuration Data Decompression” on page 10–47
■
“Remote System Upgrades” on page 10–49
■
“Remote System Upgrade Mode” on page 10–53
■
“Dedicated Remote System Upgrade Circuitry” on page 10–56
■
“Quartus II Software Support” on page 10–62
■
“Design Security” on page 10–63
Configuration Devices
Altera® serial configuration devices support a single-device and multi-device
configuration solution for Stratix IV devices and are used in the fast AS configuration
scheme. Serial configuration devices offer a low-cost, low pin-count configuration
solution.
f For information about serial configuration devices, refer to the Serial Configuration
Devices (EPCS1, EPCS4, EPCS16, EPCS64, and EPCS128) Data Sheet in volume 2 of the
Configuration Handbook.
1
All minimum timing information in this chapter covers the entire Stratix IV family.
Some devices may work at less than the minimum timing stated in this handbook due
to process variation.
Configuration Schemes
Select the configuration scheme by driving the Stratix IV device MSEL pins either high
or low, as shown in Table 10–1. The MSEL input buffers are powered by the VCC power
supply. Altera recommends hard wiring the MSEL[] pins to VCCPGM and GND. The
MSEL[2..0] pins have 5-k internal pull-down resistors that are always active.
During power-on reset (POR) and during reconfiguration, the MSEL pins must be at
VIL and VIH levels of VCCPGM voltage to be considered logic low and logic high.
1
To avoid problems with detecting an incorrect configuration scheme, hardwire the
MSEL[] pins to VCCPGM and GND without pull-up or pull-down resistors. Do not drive
the MSEL[] pins by a microprocessor or another device.
Table 10–1. Configuration Schemes for Stratix IV Devices (Part 1 of 2)
Configuration Scheme
MSEL2
MSEL1
MSEL0
Fast passive parallel
0
0
0
Passive serial
0
1
0
Fast AS (40 MHz) (1)
0
1
1
Remote system upgrade fast AS (40 MHz) (1)
0
1
1
FPP with design security feature and/or decompression enabled (2)
0
0
1
Stratix IV Device Handbook
Volume 1
September 2012 Altera Corporation
Chapter 10: Configuration, Design Security, and Remote System Upgrades in Stratix IV Devices
Configuration Schemes
10–3
Table 10–1. Configuration Schemes for Stratix IV Devices (Part 2 of 2)
Configuration Scheme
JTAG-based configuration (4)
MSEL2
MSEL1
MSEL0
(3)
(3)
(3)
Notes to Table 10–1:
(1) Stratix IV devices only support fast AS configuration. You must use either EPCS64 or EPCS128 devices to configure a Stratix IV device in fast
AS mode.
(2) These modes are only supported when using a MAX II device or a microprocessor with flash memory for configuration. In these modes, the
host system must output a DCLK that is ×4 the data rate.
(3) Do not leave the MSEL pins floating, connect them to VCCPGM or GND. These pins support the non-JTAG configuration scheme used in
production. If you only use the JTAG configuration, connect the MSEL pins to GND.
(4) The JTAG-based configuration takes precedence over other configuration schemes, which means the MSEL pin settings are ignored. The
JTAG-based configuration does not support the design security or decompression features.
Table 10–2 lists the uncompressed raw binary file (.rbf) configuration file sizes for
Stratix IV devices.
Table 10–2. Uncompressed Raw Binary File (.rbf) Sizes for Stratix IV Devices
Device
Data Size (Bits)
EP4SE230
94,557,472
EP4SE360
128,395,584
EP4SE530
171,722,064
EP4SE820
241,684,472
EP4SGX70
47,833,352
EP4SGX110
47,833,352
EP4SGX180
94,557,472
EP4SGX230
94,557,472
EP4SGX290
EP4SGX360
128,395,584
171,722,064 (1)
128,395,584
171,722,064 (1)
EP4SGX530
171,722,064
EP4S40G2
94,557,472
EP4S40G5
171,722,064
EP4S100G2
94,557,472
EP4S100G3
171,722,064
EP4S100G4
171,722,064
EP4S100G5
171,722,064
Note to Table 10–2:
(1) This only applies to the F45 package.
Use the data in Table 10–2 to estimate the file size before design compilation. Different
configuration file formats; for example, a hexidecimal (.hex) or tabular text file (.ttf)
format, have different file sizes. Refer to the Quartus® II software for the different
types of configuration file and file sizes. However, for any specific version of the
Quartus II software, any design targeted for the same device has the same
uncompressed configuration file size. If you are using compression, the file size can
vary after each compilation because the compression ratio depends on the design.
September 2012
Altera Corporation
Stratix IV Device Handbook
Volume 1
10–4
Chapter 10: Configuration, Design Security, and Remote System Upgrades in Stratix IV Devices
Configuration Features
f For more information about setting device configuration options or creating
configuration files, refer to the Device Configuration Options and Configuration File
Formats chapters in volume 2 of the Configuration Handbook.
Configuration Features
Stratix IV devices offer design security, decompression, and remote system upgrade
features. Design security using configuration bitstream encryption is available in
Stratix IV devices, which protects your designs. Stratix IV devices can receive a
compressed configuration bitstream and decompress this data in real-time, reducing
storage requirements and configuration time. You can make real-time system
upgrades from remote locations of your Stratix IV designs with the remote system
upgrade feature.
Table 10–3 lists which configuration features you can use in each configuration
scheme.
Table 10–3. Configuration Features for Stratix IV Devices
Configuration
Scheme
Configuration Method
Decompression
Design
Security
Remote
System
Upgrade
FPP
MAX II device or a microprocessor with flash memory
Y (1)
Y (1)
Y
Fast AS
Serial configuration device
Y
Y
Y
MAX II device or a microprocessor with flash memory
Y
Y
—
Download cable
Y
Y
—
MAX II device or a microprocessor with flash memory
—
—
—
Download cable
—
—
—
PS
JTAG
Note to Table 10–3:
(1) In these modes, the host system must send a DCLK that is ×4 the data rate.
You can also refer to the following:
■
For more information about the configuration data decompression feature, refer to
“Configuration Data Decompression” on page 10–47.
■
For more information about the remote system upgrade feature, refer to “Remote
System Upgrades” on page 10–49.
■
For more information about the design security feature, refer to “Design Security”
on page 10–63.
If your system already contains a common flash interface (CFI) flash memory, you can
use it for Stratix IV device configuration storage as well. The MAX II parallel flash
loader (PFL) feature in MAX II devices provides an efficient method to program CFI
flash memory devices through the JTAG interface and provides the logic to control
configuration from the flash memory device to the Stratix IV device. Both PS and FPP
configuration modes are supported using this PFL feature.
f For more information about PFL, refer to Parallel Flash Loader Megafunction User Guide.
For more information about programming Altera serial configuration devices, refer to
“Programming Serial Configuration Devices” on page 10–22.
Stratix IV Device Handbook
Volume 1
September 2012 Altera Corporation
Chapter 10: Configuration, Design Security, and Remote System Upgrades in Stratix IV Devices
Configuration Features
10–5
Power-On Reset Circuit
The POR circuit keeps the entire system in reset until the power supply voltage levels
have stabilized on power-up. After power-up, the device does not release nSTATUS
until VCC, VCCAUX, VCCPT, VCCPGM, and VCCPD are above the device’s POR trip point.
On power down, brown-out occurs if the VCC, VCCAUX, VCCPT, VCCPGM, or VCCPD drops
below the threshold voltage.
In Stratix IV devices, a pin-selectable option (PORSEL) is provided that allows you to
select between the standard POR time or fast POR time. When PORSEL is driven low,
the standard POR time is 100 ms < TPOR < 300 ms, which has a lower power-ramp rate.
When PORSEL is driven high, the fast POR time is 4 ms < TPOR < 12 ms.
VCCPGM Pins
Stratix IV devices have a power supply, VCCPGM, for all the dedicated configuration
pins and dual function pins. The supported configuration voltage is 1.8, 2.5, and 3.0 V.
Stratix IV devices do not support 1.5 V configuration.
Use the VCCPGM pin to power all dedicated configuration inputs, dedicated
configuration outputs, dedicated configuration bidirectional pins, and some of the
dual functional pins that you use for configuration. With VCCPGM, the configuration
input buffers do not have to share power lines with the regular I/O buffer in
Stratix IV devices.
The operating voltage for the configuration input pin is independent of the I/O banks
power supply VCCIO during configuration. Therefore, Stratix IV devices do not need
configuration voltage constraints on VCCIO .
VCCPD Pins
Stratix IV devices have a dedicated programming power supply, VCCPD, which must
be connected to 3.0 V/2.5 V to power the I/O pre-drivers and JTAG I/O pins (TCK,
TMS, TDI, TDO, and TRST).
1
VCCPGM and VCCPD must ramp up from 0 V to the desired voltage level within 100 ms
when PORSEL is low or 4 ms when PORSEL is high. If these supplies are not ramped up
within this specified time, your Stratix IV device will not configure successfully. If
your system cannot ramp up the power supplies within 100 ms or 4 ms, you must
hold nCONFIG low until all the power supplies are stable.
1
VCCPD must be greater than or equal to VCCIO of the same bank. If VCCIO of the bank is
set to 3.0 V, VCCPD must be powered up to 3.0 V. If the VCCIO of the bank is powered to
2.5 V or lower, VCCPD must be powered up to 2.5 V.
For more information about configuration pins power supply, refer to “Device
Configuration Pins” on page 10–39.
September 2012
Altera Corporation
Stratix IV Device Handbook
Volume 1
10–6
Chapter 10: Configuration, Design Security, and Remote System Upgrades in Stratix IV Devices
Fast Passive Parallel Configuration
Fast Passive Parallel Configuration
Fast passive parallel configuration in Stratix IV devices is designed to meet the
continuously increasing demand for faster configuration times. Stratix IV devices are
designed with the capability of receiving byte-wide configuration data per clock
cycle.
You can perform FPP configuration of Stratix IV devices using an intelligent host,
such as a MAX II device or a microprocessor.
FPP Configuration Using a MAX II Device as an External Host
FPP configuration using an external host provides the fastest method to configure
Stratix IV devices. In this configuration scheme, you can use a MAX II device as an
intelligent host that controls the transfer of configuration data from a storage device,
such as flash memory, to the target Stratix IV device. You can store configuration data
in .rbf, .hex, or .ttf format. When using the MAX II device as an intelligent host, a
design that controls the configuration process, such as fetching the data from flash
memory and sending it to the device, must be stored in the MAX II device.
1
If you are using the Stratix IV decompression and/or design security features, the
external host must be able to send a DCLK frequency that is ×4 the data rate.
The ×4 DCLK signal does not require an additional pin and is sent on the DCLK pin. The
maximum DCLK frequency is 125 MHz, which results in a maximum data rate of
250 Mbps. If you are not using the Stratix IV decompression or design security
features, the data rate is ×8 the DCLK frequency.
Figure 10–1 shows the configuration interface connections between the Stratix IV
device and a MAX II device for single device configuration.
Figure 10–1. Single Device FPP Configuration Using an External Host
Memory
ADDR DATA[7..0]
VCCPGM (1) VCCPGM (1) VCCPGM/VCCIO (2)
10 kΩ
10 kΩ
10 kΩ
Stratix IV Device
MSEL[2..0]
CONF_DONE
GND
nSTATUS
External Host
(MAX II Device or
Microprocessor)
nCE
GND
nCEO
N.C.
DATA[7..0]
nCONFIG
DCLK
Note to Figure 10–1:
(1) Connect the resistor to a supply that provides an acceptable input signal for the Stratix IV device. VCCPGM must be
high enough to meet the VIH specification of the I/O on the device and the external host. Altera recommends powering
up all configuration system I/Os with VCCPGM.
(2) A pull-up or pull-down resistor helps keep the nCONFIG line in a known state when the external host is not driving
the line.
Stratix IV Device Handbook
Volume 1
September 2012 Altera Corporation
Chapter 10: Configuration, Design Security, and Remote System Upgrades in Stratix IV Devices
Fast Passive Parallel Configuration
10–7
After power-up, the Stratix IV device goes through a POR. The POR delay depends on
the PORSEL pin setting. When PORSEL is driven low, the standard POR time is
100 ms < TPOR < 300 ms. When PORSEL is driven high, the fast POR time is
4 ms < TPOR < 12 ms. During POR, the device resets, holds nSTATUS low, and tri-states
all user I/O pins. After the device successfully exits POR, all user I/O pins continue to
be tri-stated. If nIO_pullup is driven low during power up and configuration, the user
I/O pins and dual-purpose I/O pins have weak pull-up resistors, which are on (after
POR) before and during configuration. If nIO_pullup is driven high, the weak pull-up
resistors are disabled.
The configuration cycle consists of three stages: reset, configuration, and initialization.
While nCONFIG or nSTATUS are low, the device is in the reset stage. To initiate
configuration, the MAX II device must drive the nCONFIG pin from low to high.
1
To begin the configuration process, you must fully power VCCPT, VCC, VCCPD, and
VCCPGM of the banks where the configuration pins reside to the appropriate voltage
levels.
When nCONFIG goes high, the device comes out of reset and releases the open-drain
nSTATUS pin, which is then pulled high by an external 10-k pull-up resistor. After
nSTATUS is released, the device is ready to receive configuration data and the
configuration stage begins. When nSTATUS is pulled high, the MAX II device places
the configuration data one byte at a time on the DATA[7..0] pins.
1
A pull-up or pull-down resistor helps keep the nCONFIG line in a known state when
the external host (a Max II CPLD or a microcontroller) is not driving the line. For
example, during external host reprogramming or power-up where the I/O driving
nCONFIG may be tri-stated. If a pull-up resistor is added to the nCONFIG line, the FPGA
stays in user mode if the external host is being reprogrammed. If a pull-down resistor
is added to the nCONFIG line, the FPGA goes into reset mode if the external host is
being reprogrammed. Whenever the nCONFIG line is released high, ensure that the first
DCLK and DATA are not driven unintentionally.
1
Stratix IV devices receive configuration data on the DATA[7..0] pins and the clock is
received on the DCLK pin. Data is latched into the device on the rising edge of DCLK.
If you are using the Stratix IV decompression and/or design security features,
configuration data is latched on the rising edge of every first DCLK cycle out of the four
DCLK cycles. Altera recommends that you to keep the data on DATA[7. . 0] stable for
the next 3 clock cycles when the data is being processed. You can only stop DCLK after
three clock cycles after the last data is latched.
Data is continuously clocked into the target device until CONF_DONE goes high. The
CONF_DONE pin goes high one byte early in FPP modes. The last byte is required for
FPP mode. After the device has received the next-to-last byte of the configuration data
successfully, it releases the open-drain CONF_DONE pin, which is pulled high by an
external 10-kpull-up resistor. A low-to-high transition on CONF_DONE indicates
configuration is complete and initialization of the device can begin. The CONF_DONE
pin must have an external 10-k pull-up resistor for the device to initialize.
September 2012
Altera Corporation
Stratix IV Device Handbook
Volume 1
10–8
Chapter 10: Configuration, Design Security, and Remote System Upgrades in Stratix IV Devices
Fast Passive Parallel Configuration
In Stratix IV devices, the initialization clock source is either the internal oscillator or
the optional CLKUSR pin. By default, the internal oscillator is the clock source for
initialization. If you use the internal oscillator, the Stratix IV device provides itself
with enough clock cycles for proper initialization. Therefore, if the internal oscillator
is the initialization clock source, sending the entire configuration file to the device is
sufficient to configure and initialize the device. Driving DCLK to the device after
configuration is complete does not affect device operation.
You can also synchronize initialization of multiple devices or delay initialization with
the CLKUSR option. You can turn on the Enable user-supplied start-up clock
(CLKUSR) option in the Quartus II software from the General tab of the Device and
Pin Options dialog box. Supplying a clock on CLKUSR does not affect the configuration
process. The CONF_DONE pin goes high one byte early in FPP modes. The last byte is
required for FPP mode. After the CONF_DONE pin transitions high, CLKUSR is enabled
after the time specified at tCD2CU. After this time period elapses, Stratix IV devices
require 8,532 clock cycles to initialize properly and enter user mode. Stratix IV devices
support a CLKUSR fMAX of 125 MHz.
An optional INIT_DONE pin is available, which signals the end of initialization and the
start of user-mode with a low-to-high transition. This Enable INIT_DONE Output
option is available in the Quartus II software from the General tab of the Device and
Pin Options dialog box. If you use the INIT_DONE pin, it is high because of an external
10-k pull-up resistor when nCONFIG is low and during the beginning of
configuration. After the option bit to enable INIT_DONE is programmed into the device
(during the first frame of configuration data), the INIT_DONE pin goes low. When
initialization is complete, the INIT_DONE pin is released and pulled high. The MAX II
device must be able to detect this low-to-high transition, which signals the device has
entered user mode. When initialization is complete, the device enters user mode. In
user-mode, the user I/O pins no longer have weak pull-up resistors and function as
assigned in your design.
1
Two DCLK falling edges are required after CONF_DONE goes high to begin the
initialization of the device for both uncompressed and compressed bitstream in FPP.
To ensure DCLK and DATA[7..0] are not left floating at the end of configuration, the
MAX II device must drive them either high or low, whichever is convenient on your
board. The DATA[7..0] pins are available as user I/O pins after configuration. When
you select the FPP scheme as a default in the Quartus II software, these I/O pins are
tri-stated in user mode. To change this default option in the Quartus II software, select
the Dual-Purpose Pins tab of the Device and Pin Options dialog box.
The configuration clock (DCLK) speed must be below the specified frequency to ensure
correct configuration. No maximum DCLK period exists, which means you can pause
configuration by halting DCLK for an indefinite amount of time.
Stratix IV Device Handbook
Volume 1
September 2012 Altera Corporation
Chapter 10: Configuration, Design Security, and Remote System Upgrades in Stratix IV Devices
Fast Passive Parallel Configuration
1
10–9
If you need to stop DCLK, it can only be stopped:
■
three clock cycles after the last data byte was latched into the Stratix IV device
when you use the decompression and/or design security features.
■
two clock cycles after the last data byte was latched into the Stratix IV device when
you do not use the Stratix IV decompression and/or design security features.
By stopping DCLK, the configuration circuit allows enough clock cycles to process the
last byte of latched configuration data. When the clock restarts, the MAX II device
must provide data on the DATA[7..0] pins prior to sending the first DCLK rising edge.
If an error occurs during configuration, the device drives its nSTATUS pin low, resetting
itself internally. The low signal on the nSTATUS pin also alerts the MAX II device that
there is an error. If the Auto-restart configuration after error option (available in the
Quartus II software from the General tab of the Device and Pin Options dialog box)
is turned on, the device releases nSTATUS after a reset time-out period (a maximum of
500 s). After nSTATUS is released and pulled high by a pull-up resistor, the MAX II
device can try to reconfigure the target device without needing to pulse nCONFIG low.
If this option is turned off, the MAX II device must generate a low-to-high transition
(with a low pulse of at least 2 s) on nCONFIG to restart the configuration process.
1
If you have enabled the Auto-restart configuration after error option, the nSTATUS pin
transitions from high to low and back again to high when a configuration error is
detected. This appears as a low pulse at the nSTATUS pin with a minimum pulse width
of 10 s to a maximum pulse width of 500 s, as defined in the tSTATUS specification.
The MAX II device can also monitor the CONF_DONE and INIT_DONE pins to ensure
successful configuration. The MAX II device must monitor the CONF_DONE pin to detect
errors and determine when programming completes. If all the configuration data is
sent, but the CONF_DONE or INIT_DONE signals have not gone high, the MAX II device
reconfigures the target device.
1
If you use the optional CLKUSR pin and nCONFIG is pulled low to restart the
configuration during device initialization, ensure that CLKUSR continues toggling
during the time nSTATUS is low (a maximum of 500 s).
When the device is in user mode, initiating reconfiguration is done by transitioning
the nCONFIG pin low-to-high. The nCONFIG pin must be low for at least 2 s. When
nCONFIG is pulled low, the device also pulls nSTATUS and CONF_DONE low and all I/O
pins are tri-stated. After nCONFIG returns to a logic high level and nSTATUS is released
by the device, reconfiguration begins.
September 2012
Altera Corporation
Stratix IV Device Handbook
Volume 1
10–10
Chapter 10: Configuration, Design Security, and Remote System Upgrades in Stratix IV Devices
Fast Passive Parallel Configuration
Figure 10–2 shows how to configure multiple Stratix IV devices using a MAX II
device. This circuit is similar to the FPP configuration circuit for a single device,
except the devices are cascaded for multi-device configuration.
Figure 10–2. Multi-Device FPP Configuration Using an External Host
Memory
ADDR DATA[7..0]
VCCPGM (1) VCCPGM (1) VCCPGM/VCCIO (2)
10 kΩ
10 kΩ
10 kΩ
Stratix IV Device 2
Stratix IV Device 1
MSEL[2..0]
MSEL[2..0]
CONF_DONE
CONF_DONE
GND
nSTATUS
External Host
(MAX II Device or
Microprocessor)
nCE
nCEO
GND
nSTATUS
nCE
nCEO
N.C.
GND
DATA[7..0]
DATA[7..0]
nCONFIG
nCONFIG
DCLK
DCLK
Note to Figure 10–2:
(1) Connect the pull-up resistor to a supply that provides an acceptable input signal for all Stratix IV devices in the chain. VCCPGM must be high enough
to meet the VIH specification of the I/O standard on the device and the external host. Altera recommends powering up all configuration system
I/Os with VCCPGM.
(2) A pull-up or pull-down resistor helps keep the nCONFIG line in a known state when the external host is not driving the line.
In a multi-device FPP configuration, the first device’s nCE pin is connected to GND
while its nCEO pin is connected to nCE of the next device in the chain. The last device’s
nCE input comes from the previous device, while its nCEO pin is left floating. After the
first device completes configuration in a multi-device configuration chain, its nCEO pin
drives low to activate the second device’s nCE pin, which prompts the second device
to begin configuration. The second device in the chain begins configuration within
one clock cycle; therefore, the transfer of data destinations is transparent to the
MAX II device. All other configuration pins (nCONFIG, nSTATUS, DCLK, DATA[7..0], and
CONF_DONE) are connected to every device in the chain. The configuration signals may
require buffering to ensure signal integrity and prevent clock skew problems. Ensure
that the DCLK and DATA lines are buffered for every fourth device. Because all device
CONF_DONE pins are tied together, all devices initialize and enter user mode at the same
time.
All nSTATUS and CONF_DONE pins are tied together; if any device detects an error,
configuration stops for the entire chain and you must reconfigure the entire chain. For
example, if the first device flags an error on nSTATUS, it resets the chain by pulling its
nSTATUS pin low. This behavior is similar to a single device detecting an error.
If the Auto-restart configuration after error option is turned on, the devices release
their nSTATUS pins after a reset time-out period (a maximum of 500 s). After all
nSTATUS pins are released and pulled high, the MAX II device tries to reconfigure the
chain without pulsing nCONFIG low. If this option is turned off, the MAX II device
must generate a low-to-high transition (with a low pulse of at least 2 s) on nCONFIG to
restart the configuration process.
Stratix IV Device Handbook
Volume 1
September 2012 Altera Corporation
Chapter 10: Configuration, Design Security, and Remote System Upgrades in Stratix IV Devices
Fast Passive Parallel Configuration
1
10–11
If you have enabled the Auto-restart configuration after error option, the nSTATUS pin
transitions from high to low and back again to high when a configuration error is
detected. This appears as a low pulse at the nSTATUS pin with a minimum pulse width
of 10 s to a maximum pulse width of 500 s, as defined in the tSTATUS specification.
In a multi-device FPP configuration chain, all Stratix IV devices in the chain must
either enable or disable the decompression and/or design security features. You
cannot selectively enable the decompression and/or design security features for each
device in the chain because of the DATA and DCLK relationship. If the chain contains
devices that do not support design security, use a serial configuration scheme.
If a system has multiple devices that contain the same configuration data, tie all
device nCE inputs to GND and leave the nCEO pins floating. All other configuration
pins (nCONFIG, nSTATUS, DCLK, DATA[7..0], and CONF_DONE) are connected to every
device in the chain. Configuration signals may require buffering to ensure signal
integrity and prevent clock skew problems. Ensure that the DCLK and DATA lines are
buffered for every fourth device. Devices must be the same density and package. All
devices start and complete configuration at the same time.
Figure 10–3 shows a multi-device FPP configuration when both Stratix IV devices are
receiving the same configuration data.
Figure 10–3. Multiple-Device FPP Configuration Using an External Host When Both Devices Receive the Same Data
Memory
ADDR DATA[7..0]
VCCPGM (1) VCCPGM (1) VCCPGM/VCCIO (2)
10 kΩ
10 kΩ
Stratix IV Device
10 kΩ
Stratix IV Device
MSEL[2..0]
MSEL[2..0]
GND
CONF_DONE
nSTATUS
nCE
External Host
(MAX II Device or
Microprocessor)
GND
nCEO
GND
CONF_DONE
nSTATUS
nCE
N.C. (3)
nCEO
N.C.
GND
DATA[7..0]
DATA[7..0]
nCONFIG
nCONFIG
DCLK
DCLK
Notes to Figure 10–3:
(1) Connect the resistor to a supply that provides an acceptable input signal for all Stratix IV devices in the chain. VCCPGM must be high enough to
meet the VIH specification of the I/O on the device and the external host. Altera recommends powering up all configuration system I/Os with VCCPGM.
(2) A pull-up or pull-down resistor helps keep the nCONFIG line in a known state when the external host is not driving the line.
(3) The nCEO pins of both Stratix IV devices are left unconnected when configuring the same configuration data into multiple devices.
You can use a single configuration chain to configure Stratix IV devices with other
Altera devices that support FPP configuration, such as other types of Stratix devices.
To ensure that all devices in the chain complete configuration at the same time, or that
an error flagged by one device initiates reconfiguration in all devices, tie all of the
device CONF_DONE and nSTATUS pins together.
f For more information about configuring multiple Altera devices in the same
configuration chain, refer to the Configuring Mixed Altera FPGA Chains in volume 2 of
the Configuration Handbook.
September 2012
Altera Corporation
Stratix IV Device Handbook
Volume 1
10–12
Chapter 10: Configuration, Design Security, and Remote System Upgrades in Stratix IV Devices
Fast Passive Parallel Configuration
FPP Configuration Timing
Figure 10–4 shows the timing waveform for an FPP configuration when using a
MAX II device as an external host. This waveform shows the timing when you have
not enabled the decompression and design security features.
Figure 10–4. FPP Configuration Timing Waveform (Note 1), (2)
tCF2ST1
tCFG
tCF2CK
nCONFIG
nSTATUS (3)
tSTATUS
tCF2ST0
t
CLK
CONF_DONE (4)
tCF2CD
tST2CK
tCH tCL
(5)
DCLK
tDH
DATA[7..0]
(6)
Byte 0 Byte 1 Byte 2 Byte 3
Byte n-2 Byte n-1
User Mode
Byte n
tDSU
User I/O
High-Z
User Mode
INIT_DONE
tCD2UM
Notes to Figure 10–4:
(1) Use this timing waveform when you have not enabled the decompression and design security features.
(2) The beginning of this waveform shows the device in user mode. In user mode, nCONFIG, nSTATUS, and CONF_DONE are at logic high levels.
When nCONFIG is pulled low, a reconfiguration cycle begins.
(3) After power-up, the Stratix IV device holds nSTATUS low for the time of the POR delay.
(4) After power-up, before and during configuration, CONF_DONE is low.
(5) Do not leave DCLK floating after configuration. You can drive it high or low, whichever is more convenient.
(6) DATA[7..0] are available as user I/O pins after configuration except for some exceptions on Stratix IV GT devices. The state of these pins
depends on the dual-purpose pin settings.
Table 10–4 lists the timing parameters for Stratix IV devices for an FPP configuration
when you have not enabled the decompression and design security features.
Table 10–4. FPP Timing Parameters for Stratix IV Devices (Part 1 of 2) (Note 1), (2)
Minimum
Symbol
Parameter
Stratix IV
(7)
Stratix IV
(8)
Maximum
Stratix IV
(9)
Stratix IV
(7)
Stratix IV
(8)
Stratix IV
(9)
Units
tCF2CD
nCONFIG low to CONF_DONE
low
—
800
ns
tCF2ST0
nCONFIG low to nSTATUS
low
—
800
ns
tCFG
nCONFIG low pulse width
2
—
s
tSTATUS
nSTATUS low pulse width
10
500 (3)
s
tCF2ST1
nCONFIG high to nSTATUS
high
—
500 (4)
s
tCF2CK
nCONFIG high to first rising
edge on DCLK
500
—
s
Stratix IV Device Handbook
Volume 1
September 2012 Altera Corporation
Chapter 10: Configuration, Design Security, and Remote System Upgrades in Stratix IV Devices
Fast Passive Parallel Configuration
10–13
Table 10–4. FPP Timing Parameters for Stratix IV Devices (Part 2 of 2) (Note 1), (2)
Minimum
Symbol
Parameter
Stratix IV
(7)
Stratix IV
(8)
Maximum
Stratix IV
(9)
Stratix IV
(7)
Stratix IV
(8)
Stratix IV
(9)
Units
tST2CK
nSTATUS high to first rising
edge of DCLK
2
—
s
tDSU
Data setup time before
rising edge on DCLK
4
—
ns
tDH
Data hold time after rising
edge on DCLK
1
—
ns
TR
Input rise time
—
40
ns
t
Input fall time
—
40
ns
tCD2UM
CONF_DONE high to user
mode (5)
55
150
s
tCD2CU
CONF_DONE high to CLKUSR
enabled
—
—
tCD2UMC
CONF_DONE high to user
mode with CLKUSR option
on
—
—
tCH
DCLK high time (6)
3.6
4.5
5.6
—
ns
tCL
DCLK low time (6)
3.6
4.5
5.6
—
ns
tCLK
DCLK period (6)
8
10
12.5
—
ns
fMAX
DCLK frequency
4 × maximum
DCLK period
tCD2CU + (8532 × CLKUSR
period)
—
125
100
80
MHz
Notes to Table 10–4:
(1) This information is preliminary.
(2) Use these timing parameters when you have not enabled the decompression and design security features.
(3) You can obtain this value if you do not delay the configuration by extending the nCONFIG or nSTATUS low pulse width.
(4) This value is applicable if you do not delay the configuration by externally holding nSTATUS low.
(5) The minimum and maximum numbers apply only if you chose the internal oscillator as the clock source for starting the device.
(6) Adding up tCH and tCL equals to tCLK. When EP4SE230 tCH is 3.6 ns (min), tCL must be 4.4 ns and vice versa.
(7) Applicable to EP4SE230, EP4SE360, EP4SGX70, EP4SGX110, EP4SGX180, EP4SGX230, EP4SGX290 (except F45 package), EP4SGX360 (except
F45 package), EP4S40G2, EP4S100G2 devices.
(8) Applicable to EP4SE530, EP4SGX290 (only for F45 package), EP4SGX360 (only for F45 package), EP4SGX530, EP4S40G5, EP4S100G3,
EP4S100G4, EP4S100G5 devices.
(9) Applicable to EP4SE820 only.
September 2012
Altera Corporation
Stratix IV Device Handbook
Volume 1
10–14
Chapter 10: Configuration, Design Security, and Remote System Upgrades in Stratix IV Devices
Fast Passive Parallel Configuration
Figure 10–5 shows the timing waveform for an FPP configuration when using a
MAX II device as an external host. This waveform shows the timing when you have
enabled the decompression and/or design security features.
Figure 10–5. FPP Configuration Timing Waveform with Decompression or Design Security Feature Enabled (Note 1), (2)
tCF2ST1
tCFG
tCF2CK
nCONFIG
nSTATUS (3)
CONF_DONE (4)
tSTATUS
tCF2ST0
tCF2CD
DCLK
tCL
tST2CK
tCH
1
2
3
4
1
2
3
(7)
4
1
3
(5)
4
tCLK
DATA[7..0]
Byte 0
Byte 1
tDH
tDH
tDSU
Byte 2
Byte (n-1)
Byte n
User Mode
User Mode
High-Z
User I/O
(6)
INIT_DONE
tCD2UM
Notes to Figure 10–5:
(1) Use this timing waveform when you have enabled the decompression and/or design security features.
(2) The beginning of this waveform shows the device in user-mode. In user-mode, nCONFIG, nSTATUS, and CONF_DONE are at logic high levels.
When nCONFIG is pulled low, a reconfiguration cycle begins.
(3) After power-up, the Stratix IV device holds nSTATUS low for the time of the POR delay.
(4) After power-up, before and during configuration, CONF_DONE is low.
(5) Do not leave DCLK floating after configuration. You can drive it high or low, whichever is more convenient.
(6) DATA[7..0] are available as user I/O pins after configuration except for some exceptions on Stratix IV GT devices. The state of these pins
depends on the dual-purpose pin settings.
(7) If needed, you can pause DCLK by holding it low. When DCLK restarts, the external host must provide data on the DATA[7..0] pins prior to
sending the first DCLK rising edge.
Table 10–5 lists the timing parameters for Stratix IV devices for an FPP configuration
when you enable the decompression and/or the design security features.
Table 10–5. FPP Timing Parameters for Stratix IV Devices with the Decompression and/or Design Security Features
Enabled (Note 1), (2) (Part 1 of 2)
Minimum
Symbol
Parameter
Stratix IV
(7)
Stratix IV
(8)
Maximum
Stratix IV
(9)
Stratix IV
(7)
Stratix IV
(8)
Stratix IV
(9)
Units
tCF2CD
nCONFIG low to CONF_DONE
low
—
800
ns
tCF2ST0
nCONFIG low to nSTATUS low
—
800
ns
tCFG
nCONFIG low pulse width
2
—
s
tSTATUS
nSTATUS low pulse width
10
500 (3)
s
tCF2ST1
nCONFIG high to nSTATUS
high
—
500 (4)
s
tCF2CK
nCONFIG high to first rising
edge on DCLK
500
—
s
Stratix IV Device Handbook
Volume 1
September 2012 Altera Corporation
Chapter 10: Configuration, Design Security, and Remote System Upgrades in Stratix IV Devices
Fast Passive Parallel Configuration
10–15
Table 10–5. FPP Timing Parameters for Stratix IV Devices with the Decompression and/or Design Security Features
Enabled (Note 1), (2) (Part 2 of 2)
Minimum
Symbol
Parameter
Stratix IV
(7)
Stratix IV
(8)
Maximum
Stratix IV
(9)
Stratix IV
(7)
Stratix IV
(8)
Stratix IV
(9)
Units
tST2CK
nSTATUS high to first rising
edge of DCLK
2
—
s
tDSU
Data setup time before rising
edge on DCLK
4
—
ns
tDH
Data hold time after rising
edge on DCLK
3/(DCLK frequency) + 1
—
s
tDATA
Data rate
—
250
Mbps
tR
Input rise time
—
40
ns
t
Input fall time
—
40
ns
tCD2UM
CONF_DONE high to user mode
(5)
55
150
s
tCD2CU
CONF_DONE high to CLKUSR
enabled
—
—
tCD2UMC
CONF_DONE high to user mode
with CLKUSR option on (5)
—
—
tCH
DCLK high time (6)
3.6
4.5
5.6
—
ns
tCL
DCLK low time (6)
3.6
4.5
5.6
—
ns
tCLK
DCLK period (6)
8
10
12.5
—
ns
fMAX
DCLK frequency
4 × maximum
DCLK period
tCD2CU + (8532 × CLKUSR period)
—
125
100
80
MHz
Notes to Table 10–5:
(1) This information is preliminary.
(2) Use these timing parameters when you use the decompression and/or design security features.
(3) You can obtain this value if you do not delay the configuration by extending the nCONFIG or nSTATUS low pulse width.
(4) This value is applicable if you do not delay the configuration by externally holding nSTATUS low.
(5) The minimum and maximum numbers apply only if you chose the internal oscillator as the clock source for starting the device.
(6) Adding up tCH and tCL equals to tCLK. When EP4SE230 tCH is 3.6 ns (min), tCL must be 4.4 ns and vice versa.
(7) Applicable for EP4SE230, EP4SE360, EP4SGX70, EP4SGX110, EP4SGX180, EP4SGX230, EP4SGX290 (except F45 package), EP4SGX360 (except
F45 package), EP4S40G2, EP4S100G2 devices.
(8) Applicable for EP4SE530, EP4SGX290 (only for F45 package), EP4SGX360 (only for F45 package), EP4SGX530, EP4S40G5, EP4S100G3,
EP4S100G4, EP4S100G5 devices.
(9) Applicable to EP4SE820 only.
f For more information about device configuration options and how to create
configuration files, refer to the Device Configuration Options and Configuration File
Formats chapters in volume 2 of the Configuration Handbook.
September 2012
Altera Corporation
Stratix IV Device Handbook
Volume 1
10–16
Chapter 10: Configuration, Design Security, and Remote System Upgrades in Stratix IV Devices
Fast Active Serial Configuration (Serial Configuration Devices)
FPP Configuration Using a Microprocessor
In this configuration scheme, a microprocessor can control the transfer of
configuration data from a storage device, such as flash memory, to the target
Stratix IV device.
All information in “FPP Configuration Using a MAX II Device as an External Host”
on page 10–6 is also applicable when using a microprocessor as an external host. Refer
to this section for all configuration and timing information.
Fast Active Serial Configuration (Serial Configuration Devices)
In the fast AS configuration scheme, Stratix IV devices are configured using a serial
configuration device. These configuration devices are low-cost devices with
non-volatile memory that feature a simple four-pin interface and a small form factor.
The largest serial configuration device currently supports 128 MBits of configuration
bitstream. Use the Stratix IV decompression features or select an FPP or PS
configuration scheme for EP4SE360, EP4SGX290, EP4S40G5, EP4S100G3 and larger
devices.
f For more information about serial configuration devices, refer to the Serial
Configuration Devices (EPCS1, EPCS4, EPCS16, EPCS64, and EPCS128) Data Sheet
chapter in volume 2 of the Configuration Handbook.
Serial configuration devices provide a serial interface to access configuration data.
During device configuration, Stratix IV devices read configuration data using the
serial interface, decompress data if necessary, and configure their SRAM cells. This
scheme is referred to as the AS configuration scheme because the Stratix IV device
controls the configuration interface. This scheme contrasts with the PS configuration
scheme where the configuration device controls the interface.
1
Stratix IV Device Handbook
Volume 1
The Stratix IV decompression and design security features are fully available when
configuring your Stratix IV device using fast AS mode.
September 2012 Altera Corporation
Chapter 10: Configuration, Design Security, and Remote System Upgrades in Stratix IV Devices
Fast Active Serial Configuration (Serial Configuration Devices)
10–17
Serial configuration devices have a four-pin interface—serial clock input (DCLK), serial
data output (DATA), AS data input (ASDI), and an active-low chip select (nCS). This
four-pin interface connects to Stratix IV device pins, as shown in Figure 10–6.
Figure 10–6. Single Device Fast AS Configuration
VCCPGM (1) VCCPGM (1) VCCPGM (1)
10 kΩ
10 kΩ
10 kΩ
Serial Configuration
Device
Stratix IV Device
nSTATUS
CONF_DONE
nCONFIG
nCE
nCEO
GND
VCCPGM
DATA
DATA0
DCLK
DCLK
MSEL2
nCS
nCSO
MSEL1
ASDO
MSEL0
ASDI
(2)
N.C.
GND
Notes to Figure 10–6:
(1) Connect the pull-up resistors to VCCPGM at a 3.0-V supply.
(2) Stratix IV devices use the ASDO-to-ASDI path to control the configuration device.
You can power the EPCS serial configuration device with 3.0 V when you configure
the Stratix IV FPGA using Active Serial (AS) configuration mode. This is feasible
because the power supply to the EPCS device ranges between 2.7 V and 3.6 V. You do
not need a dedicated 3.3 V power supply to power the EPCS device. The EPCS device
and the VCCPGM pins on the Stratix IV device may share the same 3.0 V power supply.
After power-up, the Stratix IV devices go through a POR. The POR delay depends on
the PORSEL pin setting. When PORSEL is driven low, the standard POR time is
100 ms < TPOR < 300 ms. When PORSEL is driven high, the fast POR time is
4 ms < TPOR < 12 ms. During POR, the device resets, holds nSTATUS and CONF_DONE
low, and tri-states all user I/O pins. After the device successfully exits POR, all the
user I/O pins continue to be tri-stated. If nIO_pullup is driven low during power-up
and configuration, the user I/O pins and dual-purpose I/O pins will have weak
pull-up resistors, which are on (after POR) before and during configuration. If
nIO_pullup is driven high, the weak pull-up resistors are disabled.
The configuration cycle consists of three stages—reset, configuration, and
initialization. While nCONFIG or nSTATUS are low, the device is in reset. After POR, the
Stratix IV device releases nSTATUS, which is pulled high by an external 10-k pull-up
resistor and enters configuration mode.
1
To begin configuration, power the VCC, VCCIO, VCCPGM, and VCCPD voltages (for the
banks where the configuration pins reside) to the appropriate voltage levels.
The serial clock (DCLK) generated by the Stratix IV device controls the entire
configuration cycle and provides timing for the serial interface. Stratix IV devices use
an internal oscillator to generate DCLK. Using the MSEL[] pins, you can select to use a
40 MHz oscillator.
September 2012
Altera Corporation
Stratix IV Device Handbook
Volume 1
10–18
Chapter 10: Configuration, Design Security, and Remote System Upgrades in Stratix IV Devices
Fast Active Serial Configuration (Serial Configuration Devices)
In fast AS configuration schemes, Stratix IV devices drive out control signals on the
falling edge of DCLK. The serial configuration device responds to the instructions by
driving out configuration data on the falling edge of DCLK. Then the data is latched
into the Stratix IV device on the following falling edge of DCLK.
In configuration mode, Stratix IV devices enable the serial configuration device by
driving the nCSO output pin low, which connects to the chip select (nCS) pin of the
configuration device. The Stratix IV device uses the serial clock (DCLK) and serial data
output (ASDO) pins to send operation commands and/or read address signals to the
serial configuration device. The configuration device provides data on its serial data
output (DATA) pin, which connects to the DATA0 input of the Stratix IV devices.
After all the configuration bits are received by the Stratix IV device, it releases the
open-drain CONF_DONE pin, which is pulled high by an external 10-k resistor.
Initialization begins only after the CONF_DONE signal reaches a logic high level. All AS
configuration pins (DATA0, DCLK, nCSO, and ASDO) have weak internal pull-up resistors
that are always active. After configuration, these pins are set as input tri-stated and
are driven high by the weak internal pull-up resistors. The CONF_DONE pin must have
an external 10-k pull-up resistor in order for the device to initialize.
In Stratix IV devices, the initialization clock source is either the internal oscillator or
the optional CLKUSR pin. By default, the internal oscillator is the clock source for
initialization. If you use the internal oscillator, the Stratix IV device provides itself
with enough clock cycles for proper initialization. You also have the flexibility to
synchronize initialization of multiple devices or to delay initialization with the CLKUSR
option. You can turn on the Enable user-supplied start-up clock (CLKUSR) option in
the Quartus II software from the General tab of the Device and Pin Options dialog
box. When you select the Enable user supplied start-up clock option, the CLKUSR pin
is the initialization clock source. Supplying a clock on CLKUSR does not affect the
configuration process. After all configuration data is accepted and CONF_DONE goes
high, CLKUSR is enabled after four clock cycles of DCLK. After this time period elapses,
Stratix IV devices require 8,532 clock cycles to initialize properly and enter user mode.
Stratix IV devices support a CLKUSR fMAX of 125 MHz.
An optional INIT_DONE pin is available, which signals the end of initialization and the
start of user-mode with a low-to-high transition. The Enable INIT_DONE Output
option is available in the Quartus II software from the General tab of the Device and
Pin Options dialog box. If you use the INIT_DONE pin, it is high due to an external
10-k pull-up resistor when nCONFIG is low and during the beginning of
configuration. After the option bit to enable INIT_DONE is programmed into the device
(during the first frame of configuration data), the INIT_DONE pin goes low. When
initialization is complete, the INIT_DONE pin is released and pulled high. This
low-to-high transition signals that the device has entered user mode. When
initialization is complete, the device enters user mode. In user mode, the user I/O
pins no longer have weak pull-up resistors and function as assigned in your design.
If an error occurs during configuration, Stratix IV devices assert the nSTATUS signal
low, indicating a data frame error, and the CONF_DONE signal stays low. If the
Auto-restart configuration after error option (available in the Quartus II software
from the General tab of the Device and Pin Options dialog box) is turned on, the
Stratix IV device resets the configuration device by pulsing nCSO, releases nSTATUS
after a reset time-out period (a maximum of 500 µs), and retries configuration. If this
option is turned off, the system must monitor nSTATUS for errors and then pulse
nCONFIG low for at least 2 s to restart configuration.
Stratix IV Device Handbook
Volume 1
September 2012 Altera Corporation
Chapter 10: Configuration, Design Security, and Remote System Upgrades in Stratix IV Devices
Fast Active Serial Configuration (Serial Configuration Devices)
1
10–19
If you have enabled the Auto-restart configuration after error option, the nSTATUS pin
transitions from high to low and back again to high when a configuration error is
detected. This appears as a low pulse at the nSTATUS pin with a minimum pulse width
of 10 s to a maximum pulse width of 500 s, as defined in the tSTATUS specification.
When the Stratix IV device is in user mode, you can initiate reconfiguration by pulling
the nCONFIG pin low. The nCONFIG pin must be low for at least 2 s. When nCONFIG is
pulled low, the device also pulls nSTATUS and CONF_DONE low and all I/O pins are
tri-stated. After nCONFIG returns to a logic high level and nSTATUS is released by the
Stratix IV device, reconfiguration begins.
1
If you wish to gain control of the EPCS pins, hold the nCONFIG pin low and pull the
nCE pin high. This causes the device to reset and tri-state the AS configuration pins.
The timing parameters for AS mode are not listed here because the tCF2CD, tCF2ST0, tCFG,
tSTATUS, tCF2ST1, and tCD2UM timing parameters are identical to the timing parameters
for PS mode listed in Table 10–7 on page 10–30.
You can configure multiple Stratix IV devices using a single serial configuration
device. You can cascade multiple Stratix IV devices using the chip-enable (nCE) and
chip-enable-out (nCEO) pins. The first device in the chain must have its nCE pin
connected to GND. You must connect its nCEO pin to the nCE pin of the next device in
the chain. When the first device captures all of its configuration data from the
bitstream, it drives the nCEO pin low, enabling the next device in the chain. You must
leave the nCEO pin of the last device unconnected. The nCONFIG, nSTATUS, CONF_DONE,
DCLK, and DATA0 pins of each device in the chain are connected (refer to Figure 10–7).
The first Stratix IV device in the chain is the configuration master and controls
configuration of the entire chain. You must connect its MSEL pins to select the AS
configuration scheme. The remaining Stratix IV devices are configuration slaves. You
must connect their MSEL pins to select the PS configuration scheme. Any other Altera
device that supports PS configuration can also be part of the chain as a configuration
slave.
September 2012
Altera Corporation
Stratix IV Device Handbook
Volume 1
10–20
Chapter 10: Configuration, Design Security, and Remote System Upgrades in Stratix IV Devices
Fast Active Serial Configuration (Serial Configuration Devices)
Figure 10–7 shows the pin connections for the multi-device fast AS configuration.
Figure 10–7. Multi-Device Fast AS Configuration
VCCPGM (1) VCCPGM (1) VCCPGM (1)
10 kΩ
10 kΩ
10 kΩ
Serial Configuration
Device
Stratix IV Device Master
Stratix IV Device Slave
nSTATUS
CONF_DONE
nCONFIG
nCE
nCEO
GND
DATA
nSTATUS
CONF_DONE
nCONFIG
nCE
nCEO
VCCPGM
DATA0
VCCPGM
DATA0
DCLK
DCLK
MSEL2
nCS
nCSO
MSEL1
ASDI
ASDO
MSEL0
DCLK
GND
N.C.
MSEL2
MSEL1
MSEL0
GND
Buffers (2)
Notes to Figure 10–7:
(1) Connect the pull-up resistors to VCCPGM at a 3.0-V supply.
(2) Connect the repeater buffers between the Stratix IV master and slave device(s) for DATA[0] and DCLK. This is to prevent potential signal
integrity and clock skew problems.
As shown in Figure 10–7, the nSTATUS and CONF_DONE pins on all target devices are
connected together with external pull-up resistors. These pins are open-drain
bidirectional pins on the devices. When the first device asserts nCEO (after receiving all
of its configuration data), it releases its CONF_DONE pin. But the subsequent devices in
the chain keep this shared CONF_DONE line low until they have received their
configuration data. When all target devices in the chain have received their
configuration data and have released CONF_DONE, the pull-up resistor drives a high
level on this line and all devices simultaneously enter initialization mode.
If an error occurs at any point during configuration, the nSTATUS line is driven low by
the failing device. If you enable the Auto-restart configuration after error option,
reconfiguration of the entire chain begins after a reset time-out period (a maximum of
500 s). If you did not enable the Auto-restart configuration after error option, the
external system must monitor nSTATUS for errors and then pulse nCONFIG low to
restart configuration. The external system can pulse nCONFIG if it is under system
control rather than tied to VCCGPM.
1
Stratix IV Device Handbook
Volume 1
If you have enabled the Auto-restart configuration after error option, the nSTATUS pin
transitions from high to low and back again to high when a configuration error is
detected. This appears as a low pulse at the nSTATUS pin with a minimum pulse width
of 10 s to a maximum pulse width of 500 s, as defined in the tSTATUS specification.
September 2012 Altera Corporation
Chapter 10: Configuration, Design Security, and Remote System Upgrades in Stratix IV Devices
Fast Active Serial Configuration (Serial Configuration Devices)
1
10–21
While you can cascade Stratix IV devices, you cannot cascade or chain together serial
configuration devices.
If the configuration bitstream size exceeds the capacity of a serial configuration
device, you must select a larger configuration device and/or enable the compression
feature. When configuring multiple devices, the size of the bitstream is the sum of the
individual device configuration bitstreams.
A system may have multiple devices that contain the same configuration data. In
active serial chains, you can implement this by storing one copy of the .sof in the
serial configuration device. The same copy of the .sof configures the master Stratix IV
device and all remaining slave devices concurrently. All Stratix IV devices must be the
same density and package.
To configure four identical Stratix IV devices with the same .sof, set up the chain as
shown in Figure 10–8. The first device is the master device and its MSEL pins must be
set to select AS configuration. The other three slave devices are set up for concurrent
configuration and their MSEL pins must be set to select PS configuration. The nCE input
pins from the master and slave are connected to GND, and the DATA and DCLK pins
connect in parallel to all four devices. During the configuration cycle, the master
device reads its configuration data from the serial configuration device and transmits
the configuration data to all three slave devices, configuring all of them
simultaneously.
September 2012
Altera Corporation
Stratix IV Device Handbook
Volume 1
10–22
Chapter 10: Configuration, Design Security, and Remote System Upgrades in Stratix IV Devices
Fast Active Serial Configuration (Serial Configuration Devices)
Figure 10–8 shows the multi-device fast AS configuration when the devices receive
the same data using a single .sof.
Figure 10–8. Multi-Device Fast AS Configuration When the Devices Receive the Same Data Using a Single .sof
Stratix IV
Device Slave
nSTATUS
CONF_DONE
nCONFIG
nCE
nCEO
N.C.
VCCPGM (1) VCCPGM (1) VCCPGM (1)
DATA0
10 kΩ
10 kΩ
MSEL2
DCLK
10 kΩ
VCCPGM
MSEL1
MSEL0
GND
Stratix IV
Device Master
Serial Configuration
Device
nSTATUS
CONF_DONE
nCONFIG
nCE
nCEO
DATA0
DCLK
DCLK
nCS
nCSO
ASDI
ASDO
N.C.
GND
VCCPGM
GND
DATA
Stratix IV
Device Slave
MSEL2
nSTATUS
CONF_DONE
nCONFIG
nCE
DATA0
MSEL2
DCLK
MSEL1
nCEO
N.C.
VCCPGM
MSEL1
MSEL0
MSEL0
GND
GND
Stratix IV
Device Slave
Buffers (2)
nSTATUS
CONF_DONE
nCONFIG
nCE
nCEO
DATA0
DCLK
MSEL2
N.C.
VCCPGM
MSEL1
MSEL0
GND
Notes to Figure 10–8:
(1) Connect the pull-up resistors to VCCPGM at a 3.0-V supply.
(2) Connect the repeater buffers between the Stratix IV master and slave device(s) for DATA[0] and DCLK. This is to prevent potential signal
integrity and clock skew problems.
Estimating Active Serial Configuration Time
Active serial configuration time is dominated by the time it takes to transfer data from
the serial configuration device to the Stratix IV device. This serial interface is clocked
by the Stratix IV DCLK output (generated from an internal oscillator) and must be set to
40 MHz (25 ns).Therefore, the minimum configuration time estimate for an EP4SE230
device (94, 600, 000 bits of uncompressed data) is:
RBF Size × (minimum DCLK period / 1 bit per DCLK cycle) = estimated minimum
configuration time
94, 600, 000 bits × (25 ns / 1 bit) = 2365 ms
Enabling compression reduces the amount of configuration data that is transmitted to
the Stratix IV device, which also reduces configuration time. On average, compression
reduces configuration time, depending on the design.
Stratix IV Device Handbook
Volume 1
September 2012 Altera Corporation
Chapter 10: Configuration, Design Security, and Remote System Upgrades in Stratix IV Devices
Fast Active Serial Configuration (Serial Configuration Devices)
10–23
Programming Serial Configuration Devices
Serial configuration devices are non-volatile, flash-memory-based devices. You can
program these devices in-system using the USB-Blaster™, EthernetBlaster™, or
ByteBlaster™ II download cable. Alternatively, you can program them using the
Altera programming unit (APU), supported third-party programmers, or a
microprocessor with the SRunner software driver.
You can perform in-system programming of serial configuration devices using the
conventional AS programming interface or the JTAG interface solution.
Because serial configuration devices do not support the JTAG interface, the
conventional method to program them is using the AS programming interface. The
configuration data used to program serial configuration devices is downloaded using
programming hardware.
During in-system programming, the download cable disables device access to the AS
interface by driving the nCE pin high. Stratix IV devices are also held in reset by a low
level on nCONFIG. After programming is complete, the download cable releases nCE
and nCONFIG, allowing the pull-down and pull-up resistors to drive GND and VCCPGM,
respectively. Figure 10–9 shows the download cable connections for the serial
configuration device.
Altera has developed Serial FlashLoader (SFL), an in-system programming solution
for serial configuration devices using the JTAG interface. This solution requires the
Stratix IV device to be a bridge between the JTAG interface and the serial
configuration device.
f For more information about SFL, refer to AN 370: Using the Serial FlashLoader with
Quartus II Software.
September 2012
Altera Corporation
Stratix IV Device Handbook
Volume 1
10–24
Chapter 10: Configuration, Design Security, and Remote System Upgrades in Stratix IV Devices
Fast Active Serial Configuration (Serial Configuration Devices)
f For more information about the USB-Blaster download cable, refer to the USB-Blaster
Download Cable User Guide. For more information about the ByteBlaster II cable, refer
to the ByteBlaster II Download Cable User Guide. For more information about the
EthernetBlaster download cable, refer to the EthernetBlaster Communications Cable User
Guide.
Figure 10–9. In-System Programming of Serial Configuration Devices
VCCPGM (1) VCCPGM (1) VCCPGM (1)
10 kΩ
10 kΩ
10 kΩ
Stratix IV Device
CONF_DONE
nSTATUS
Serial
Configuration
Device
nCEO
N.C.
nCONFIG
nCE
10 kΩ
VCCPGM
DATA
DATA0
DCLK
DCLK
nCS
nCSO
MSEL1
ASDI
ASDO
MSEL0
MSEL2
GND
Pin 1
VCCPGM (2)
USB Blaster or ByteBlaser II
(AS Mode)
10-Pin Male Header
Notes to Figure 10–9:
(1) Connect these pull-up resistors to VCCPGM at a 3.0-V supply.
(2) Power up the USB-ByteBlaster, ByteBlaster II, or EthernetBlaster cable’s VCC(TRGT) with VCCPGM.
You can program serial configuration devices with the Quartus II software using the
Altera programming hardware and the appropriate configuration device
programming adapter.
In production environments, you can program serial configuration devices using
multiple methods. You can use Altera programming hardware or other third-party
programming hardware to program blank serial configuration devices before they are
mounted on PCBs. Alternatively, you can use an on-board microprocessor to program
the serial configuration device in-system using C-based software drivers provided by
Altera.
Stratix IV Device Handbook
Volume 1
September 2012 Altera Corporation
Chapter 10: Configuration, Design Security, and Remote System Upgrades in Stratix IV Devices
Passive Serial Configuration
10–25
You can program a serial configuration device in-system by an external
microprocessor using SRunner. SRunner is a software driver developed for embedded
serial configuration device programming, which can be easily customized to fit in
different embedded systems. SRunner is able to read raw programming data (.rpd)
and write to serial configuration devices. The serial configuration device
programming time using SRunner is comparable to the programming time with the
Quartus II software.
f For more information about SRunner, refer to AN 418: SRunner: An Embedded Solution
for Serial Configuration Device Programming and the source code on the Altera website
at www.altera.com.
f For more information about programming serial configuration devices, refer to the
Serial Configuration Devices (EPCS1, EPCS4, EPCS16, EPCS64, and EPCS128) Data Sheet
chapter in volume 2 of the Configuration Handbook.
Guidelines for Connecting Serial Configuration Devices on an AS Interface
For single- and multi-device AS configurations, the board trace length and loading
between the supported serial configuration device and the Stratix IV device family
must follow the recommendations listed in Table 10–6.
Table 10–6. Maximum Trace Length and Loading for the AS Configuration
Maximum Board Trace Length
from the Stratix IV Device to
the Serial Configuration
Device (Inches)
Maximum Board Load (pF)
DCLK
10
15
DATA[0]
10
30
nCSO
10
30
ASDO
10
30
Stratix IV Device AS Pins
Passive Serial Configuration
You can program a PS configuration for Stratix IV devices using an intelligent host,
such as a MAX II device or microprocessor with flash memory, or a download cable.
In the PS scheme, an external host (a MAX II device, embedded processor, or host PC)
controls configuration. Configuration data is clocked into the target Stratix IV device
using the DATA0 pin at each rising edge of DCLK.
1
September 2012
The Stratix IV decompression and design security features are fully available when
configuring your Stratix IV device using PS mode.
Altera Corporation
Stratix IV Device Handbook
Volume 1
10–26
Chapter 10: Configuration, Design Security, and Remote System Upgrades in Stratix IV Devices
Passive Serial Configuration
PS Configuration Using a MAX II Device as an External Host
In this configuration scheme, you can use a MAX II device as an intelligent host that
controls the transfer of configuration data from a storage device, such as flash
memory, to the target Stratix IV device. You can store configuration data in .rbf, .hex,
or .ttf format.
Figure 10–10 shows the configuration interface connections between a Stratix IV
device and a MAX II device for single device configuration.
Figure 10–10. Single Device PS Configuration Using an External Host
Memory
ADDR
VCCPGM (1)
VCCPGM (1) VCCPGM/VCCIO (2)
DATA0
10 k Ω
10 k Ω
10 kΩ
Stratix IV Device
CONF_DONE
nSTATUS
External Host
(MAX II Device or
Microprocessor)
nCE
nCEO
GND
DATA0
MSEL2
nCONFIG
MSEL1
DCLK
N.C.
VCCPGM
MSEL0
GND
Note to Figure 10–10:
(1) Connect the resistor to a supply that provides an acceptable input signal for the Stratix IV device. VCCPGM must be
high enough to meet the VIH specification of the I/O on the device and the external host. Altera recommends powering
up all configuration system I/Os with VCCPGM.
(2) A pull-up or pull-down resistor helps keep the nCONFIG line in a known state when the external host is not driving
the line.
After power-up, Stratix IV devices go through a POR. The POR delay depends on the
PORSEL pin setting. When PORSEL is driven low, the standard POR time is
100 ms < TPOR < 300 ms. When PORSEL is driven high, the fast POR time is
4 ms < TPOR < 12 ms. During POR, the device resets, holds nSTATUS low, and tri-states
all user I/O pins. After the device successfully exits POR, all user I/O pins continue to
be tri-stated. If nIO_pullup is driven low during power-up and configuration, the user
I/O pins and dual-purpose I/O pins will have weak pull-up resistors that are on
(after POR) before and during configuration. If nIO_pullup is driven high, the weak
pull-up resistors are disabled.
The configuration cycle consists of three stages—reset, configuration, and
initialization. While nCONFIG or nSTATUS are low, the device is in reset. To initiate
configuration, the MAX II device must generate a low-to-high transition on the
nCONFIG pin.
1
Stratix IV Device Handbook
Volume 1
VCC, VCCIO, VCCPGM, and VCCPD of the banks where the configuration pins reside must
be fully powered to the appropriate voltage levels to begin the configuration process.
September 2012 Altera Corporation
Chapter 10: Configuration, Design Security, and Remote System Upgrades in Stratix IV Devices
Passive Serial Configuration
10–27
When nCONFIG goes high, the device comes out of reset and releases the open-drain
nSTATUS pin, which is then pulled high by an external 10-k pull-up resistor. After
nSTATUS is released, the device is ready to receive configuration data and the
configuration stage begins. When nSTATUS is pulled high, the MAX II device places
the configuration data one bit at a time on the DATA0 pin. If you are using
configuration data in .rbf, .hex, or .ttf format, you must send the LSB of each data byte
first. For example, if the .rbf contains the byte sequence 02 1B EE 01 FA, the serial
bitstream you must transmit to the device is
0100-0000 1101-1000 0111-0111 1000-0000 0101-1111.
1
A pull-up or pull-down resistor helps keep the nCONFIG line in a known state when
the external host (a Max II CPLD or a microcontroller) is not driving the line. For
example, during external host reprogramming or power-up where the I/O driving
nCONFIG may be tri-stated. If a pull-up resistor is added to the nCONFIG line, the FPGA
stays in user mode if the external host is being reprogrammed. If a pull-down resistor
is added to the nCONFIG line, the FPGA goes into reset mode if the external host is
being reprogrammed. Whenever the nCONFIG line is released high, ensure that the first
DCLK and DATA are not driven unintentionally.
The Stratix IV device receives configuration data on the DATA0 pin and the clock is
received on the DCLK pin. Data is latched into the device on the rising edge of DCLK.
Data is continuously clocked into the target device until CONF_DONE goes high. After
the device has received all configuration data successfully, it releases the open-drain
CONF_DONE pin, which is pulled high by an external 10-kpull-up resistor. A
low-to-high transition on CONF_DONE indicates configuration is complete and
initialization of the device can begin. The CONF_DONE pin must have an external 10-k
pull-up resistor for the device to initialize.
In Stratix IV devices, the initialization clock source is either the internal oscillator or
the optional CLKUSR pin. By default, the internal oscillator is the clock source for
initialization. If you use the internal oscillator, the Stratix IV device provides itself
with enough clock cycles for proper initialization. Therefore, if the internal oscillator
is the initialization clock source, sending the entire configuration file to the device is
sufficient to configure and initialize the device. Driving DCLK to the device after
configuration is complete does not affect device operation.
You also have the flexibility to synchronize initialization of multiple devices or to
delay initialization with the CLKUSR option. You can turn on the Enable user-supplied
start-up clock (CLKUSR) option in the Quartus II software from the General tab of
the Device and Pin Options dialog box. If you supply a clock on CLKUSR, it will not
affect the configuration process. After all configuration data has been accepted and
CONF_DONE goes high, CLKUSR is enabled after the time specified at tCD2CU. After this
time period elapses, Stratix IV devices require 8,532 clock cycles to initialize properly
and enter user mode. Stratix IV devices support a CLKUSR fMAX of 125 MHz.
An optional INIT_DONE pin is available that signals the end of initialization and the
start of user-mode with a low-to-high transition. The Enable INIT_DONE Output
option is available in the Quartus II software from the General tab of the Device and
Pin Options dialog box. If you use the INIT_DONE pin, it is high due to an external
10-k pull-up resistor when nCONFIG is low and during the beginning of
configuration. After the option bit to enable INIT_DONE is programmed into the device
(during the first frame of configuration data), the INIT_DONE pin goes low. When
September 2012
Altera Corporation
Stratix IV Device Handbook
Volume 1
10–28
Chapter 10: Configuration, Design Security, and Remote System Upgrades in Stratix IV Devices
Passive Serial Configuration
initialization is complete, the INIT_DONE pin is released and pulled high. The MAX II
device must be able to detect this low-to-high transition that signals the device has
entered user mode. When initialization is complete, the device enters user mode. In
user-mode, the user I/O pins no longer have weak pull-up resistors and function as
assigned in your design.
1
Two DCLK falling edges are required after CONF_DONE goes high to begin the
initialization of the device for both uncompressed and compressed bitstream in PS.
To ensure DCLK and DATA0 are not left floating at the end of configuration, the MAX II
device must drive them either high or low, whichever is convenient on your board.
The DATA[0] pin is available as a user I/O pin after configuration. When you chose the
PS scheme as a default in the Quartus II software, this I/O pin is tri-stated in user
mode and must be driven by the MAX II device. To change this default option in the
Quartus II software, select the Dual-Purpose Pins tab of the Device and Pin Options
dialog box.
The configuration clock (DCLK) speed must be below the specified frequency to ensure
correct configuration. No maximum DCLK period exists, which means you can pause
the configuration by halting DCLK for an indefinite amount of time.
If an error occurs during configuration, the device drives its nSTATUS pin low, resetting
itself internally. The low signal on the nSTATUS pin also alerts the MAX II device that
there is an error. If the Auto-restart configuration after error option (available in the
Quartus II software from the General tab of the Device and Pin Options dialog box)
is turned on, the Stratix IV device releases nSTATUS after a reset time-out period (a
maximum of 500 s). After nSTATUS is released and pulled high by a pull-up resistor,
the MAX II device can try to reconfigure the target device without needing to pulse
nCONFIG low. If this option is turned off, the MAX II device must generate a
low-to-high transition (with a low pulse of at least 2 s) on nCONFIG to restart the
configuration process.
1
If you have enabled the Auto-restart configuration after error option, the nSTATUS pin
transitions from high to low and back again to high when a configuration error is
detected. This appears as a low pulse at the nSTATUS pin with a minimum pulse width
of 10 s to a maximum pulse width of 500 s, as defined in the tSTATUS specification.
The MAX II device can also monitor the CONF_DONE and INIT_DONE pins to ensure
successful configuration. The CONF_DONE pin must be monitored by the MAX II device
to detect errors and determine when programming completes. If all configuration
data is sent, but CONF_DONE or INIT_DONE have not gone high, the MAX II device must
reconfigure the target device.
1
If you use the optional CLKUSR pin and nCONFIG is pulled low to restart configuration
during device initialization, you must ensure that CLKUSR continues toggling during
the time nSTATUS is low (a maximum of 500 s).
When the device is in user-mode, you can initiate a reconfiguration by transitioning
the nCONFIG pin low-to-high. The nCONFIG pin must be low for at least 2 s. When
nCONFIG is pulled low, the device also pulls nSTATUS and CONF_DONE low and all I/O
pins are tri-stated. After nCONFIG returns to a logic high level and nSTATUS is released
by the device, reconfiguration begins.
Stratix IV Device Handbook
Volume 1
September 2012 Altera Corporation
Chapter 10: Configuration, Design Security, and Remote System Upgrades in Stratix IV Devices
Passive Serial Configuration
10–29
Figure 10–11 shows how to configure multiple devices using a MAX II device. This
circuit is similar to the PS configuration circuit for a single device, except the Stratix IV
devices are cascaded for multi-device configuration.
Figure 10–11. Multi-Device PS Configuration Using an External Host
Memory
ADDR
VCCPGM (1) VCCPGM (1) VCCPGM/VCCIO (2)
DATA0
10 k Ω
10 k Ω
10 kΩ
Stratix IV Device 1
CONF_DONE
CONF_DONE
nSTATUS
nSTATUS
nCE
nCE
External Host
(MAX II Device or
Microprocessor)
Stratix IV Device 2
nCEO
GND
DATA0
MSEL2
VCCPGM
nCEO
MSEL2
DATA0
nCONFIG
MSEL1
nCONFIG
MSEL1
DCLK
MSEL0
DCLK
MSEL0
N.C.
VCCPGM
GND
GND
Note to Figure 10–11:
(1) Connect the resistor to a supply that provides an acceptable input signal for all Stratix IV devices in the chain. VCCPGM must be high enough to
meet the VIH specification of the I/O on the device and the external host. Altera recommends powering up all configuration system I/Os with VCCPGM.
(2) A pull-up or pull-down resistor helps keep the nCONFIG line in a known state when the external host is not driving the line.
In multi-device PS configuration, the first device’s nCE pin is connected to GND, while
its nCEO pin is connected to nCE of the next device in the chain. The last device’s nCE
input comes from the previous device, while its nCEO pin is left floating. After the first
device completes configuration in a multi-device configuration chain, its nCEO pin
drives low to activate the second device’s nCE pin, which prompts the second device
to begin configuration. The second device in the chain begins configuration within
one clock cycle. Therefore, the transfer of data destinations is transparent to the
MAX II device. All other configuration pins (nCONFIG, nSTATUS, DCLK, DATA0, and
CONF_DONE) are connected to every device in the chain. Configuration signals can
require buffering to ensure signal integrity and prevent clock skew problems. Ensure
that the DCLK and DATA lines are buffered for every fourth device. Because all device
CONF_DONE pins are tied together, all devices initialize and enter user mode at the same
time.
Because all nSTATUS and CONF_DONE pins are tied together, if any device detects an
error, configuration stops for the entire chain and you must reconfigure the entire
chain. For example, if the first device flags an error on nSTATUS, it resets the chain by
pulling its nSTATUS pin low. This behavior is similar to a single device detecting an
error.
If the Auto-restart configuration after error option is turned on, the devices release
their nSTATUS pins after a reset time-out period (a maximum of 500 s). After all
nSTATUS pins are released and pulled high, the MAX II device can try to reconfigure
the chain without needing to pulse nCONFIG low. If this option is turned off, the
MAX II device must generate a low-to-high transition (with a low pulse of at least
2 s) on nCONFIG to restart the configuration process.
September 2012
Altera Corporation
Stratix IV Device Handbook
Volume 1
10–30
Chapter 10: Configuration, Design Security, and Remote System Upgrades in Stratix IV Devices
Passive Serial Configuration
1
If you have enabled the Auto-restart configuration after error option, the nSTATUS pin
transitions from high to low and back again to high when a configuration error is
detected. This appears as a low pulse at the nSTATUS pin with a minimum pulse width
of 10 s to a maximum pulse width of 500 s, as defined in the tSTATUS specification.
In your system, you can have multiple devices that contain the same configuration
data. To support this configuration scheme, all device nCE inputs are tied to GND,
while the nCEO pins are left floating. All other configuration pins (nCONFIG, nSTATUS,
DCLK, DATA0, and CONF_DONE) are connected to every device in the chain.
Configuration signals can require buffering to ensure signal integrity and prevent
clock skew problems. Ensure that the DCLK and DATA lines are buffered for every fourth
device. Devices must be the same density and package. All devices start and complete
configuration at the same time.
Figure 10–12 shows multi-device PS configuration when both Stratix IV devices are
receiving the same configuration data.
Figure 10–12. Multiple-Device PS Configuration When Both Devices Receive the Same Data
Memory
ADDR
VCCPGM (1) VCCPGM (1) V
CCPGM/VCCIO (2)
DATA0
10 k Ω
10 k Ω
10 kΩ
Stratix IV Device
Stratix IV Device
CONF_DONE
CONF_DONE
nSTATUS
External Host
(MAX II Device or
Microprocessor)
nCE
nCEO
GND
DATA0
MSEL2
N.C. (3)
nCEO
nSTATUS
nCE
VCCPGM GND
DATA0
MSEL2
nCONFIG
MSEL1
nCONFIG
MSEL1
DCLK
MSEL0
DCLK
MSEL0
N.C. (3)
VCCPGM
GND
GND
Notes to Figure 10–12:
(1) Connect the resistor to a supply that provides an acceptable input signal for all Stratix IV devices in the chain. VCCPGM must be high enough to
meet the VIH specification of the I/O on the device and the external host. Altera recommends powering up all configuration system I/Os with VCCPGM.
(2) A pull-up or pull-down resistor helps keep the nCONFIG line in a known state when the external host is not driving the line.
(3) The nCEO pins of both devices are left unconnected when configuring the same configuration data into multiple devices.
You can use a single configuration chain to configure Stratix IV devices with other
Altera devices. To ensure that all devices in the chain complete configuration at the
same time, or that an error flagged by one device initiates reconfiguration in all
devices, all of the device CONF_DONE and nSTATUS pins must be tied together.
f For more information about configuring multiple Altera devices in the same
configuration chain, refer to the Configuring Mixed Altera FPGA Chains chapter in
volume 2 of the Configuration Handbook.
Stratix IV Device Handbook
Volume 1
September 2012 Altera Corporation
Chapter 10: Configuration, Design Security, and Remote System Upgrades in Stratix IV Devices
Passive Serial Configuration
10–31
PS Configuration Timing
Figure 10–13 shows the timing waveform for PS configuration when using a MAX II
device as an external host.
Figure 10–13. PS Configuration Timing Waveform (Note 1)
tCF2ST1
tCFG
tCF2CK
nCONFIG
nSTATUS (2)
tSTATUS
tCF2ST0
t
CLK
CONF_DONE (3)
tCF2CD
tST2CK
tCH tCL
(4)
DCLK
tDH
Bit 0 Bit 1 Bit 2 Bit 3
DATA
(5)
Bit n
tDSU
High-Z
User I/O
User Mode
INIT_DONE
tCD2UM
Notes to Figure 10–13:
(1) The beginning of this waveform shows the device in user mode. In user mode, nCONFIG, nSTATUS, and CONF_DONE are at logic high levels.
When nCONFIG is pulled low, a reconfiguration cycle begins.
(2) After power-up, the Stratix IV device holds nSTATUS low for the time of the POR delay.
(3) After power-up, before and during configuration, CONF_DONE is low.
(4) Do not leave DCLK floating after configuration. You can drive it high or low, whichever is more convenient.
(5) DATA[0] is available as a user I/O pin after configuration. The state of this pin depends on the dual-purpose pin settings.
Table 10–7 lists the timing parameters for Stratix IV devices for PS configuration.
Table 10–7. PS Timing Parameters for Stratix IV Devices (Part 1 of 2) (Note 1)
Symbol
Parameter
Minimum
Maximum
Units
ns
tCF2CD
nCONFIG low to CONF_DONE low
—
800
tCF2ST0
nCONFIG low to nSTATUS low
—
800
ns
tCFG
nCONFIG low pulse width
2
—
s
tSTATUS
nSTATUS low pulse width
10
500 (2)
s
tCF2ST1
nCONFIG high to nSTATUS high
—
500 (3)
s
tCF2CK
nCONFIG high to first rising edge on DCLK
500
—
s
tST2CK
nSTATUS high to first rising edge of DCLK
2
—
s
tDSU
Data setup time before rising edge on DCLK
4
—
ns
tDH
Data hold time after rising edge on DCLK
0
—
ns
tCH
DCLK high time (5)
3.2
—
ns
tCL
DCLK low time (5)
3.2
—
ns
tCLK
DCLK period (5)
8
—
ns
fMAX
DCLK frequency
—
125
MHz
tR
Input rise time
—
40
ns
September 2012
Altera Corporation
Stratix IV Device Handbook
Volume 1
10–32
Chapter 10: Configuration, Design Security, and Remote System Upgrades in Stratix IV Devices
Passive Serial Configuration
Table 10–7. PS Timing Parameters for Stratix IV Devices (Part 2 of 2) (Note 1)
Symbol
Parameter
Minimum
Maximum
Units
tF
Input fall time
—
40
ns
tCD2UM
CONF_DONE high to user mode (4)
55
150
s
tCD2CU
CONF_DONE high to CLKUSR enabled
4 × maximum
DCLK period
—
—
tCD2UMC
CONF_DONE high to user mode with CLKUSR option on
tCD2CU + (8532
CLKUSR
period)
—
—
Notes to Table 10–7:
(1) This information is preliminary.
(2) This value is applicable if you do not delay the configuration by extending the nCONFIG or nSTATUS low pulse width.
(3) This value is applicable if you do not delay the configuration by externally holding nSTATUS low.
(4) The minimum and maximum numbers apply only if you choose the internal oscillator as the clock source for starting the device.
(5) Adding up tCH and tCL equals to tCLK. When tCH is 3.2 ns (min), tCL must be 4.8 ns and vice versa.
f Device configuration options and how to create configuration files are described in
the Device Configuration Options and Configuration File Formats chapters in volume 2 of
the Configuration Handbook.
PS Configuration Using a Microprocessor
In this PS configuration scheme, a microprocessor controls the transfer of
configuration data from a storage device, such as flash memory, to the target
Stratix IV device.
For more information about configuration and timing information, refer to “PS
Configuration Using a MAX II Device as an External Host” on page 10–25. This
section is also applicable when using a microprocessor as an external host.
PS Configuration Using a Download Cable
1
In this section, the generic term “download cable” includes the Altera USB-Blaster
universal serial bus (USB) port download cable, MasterBlaster serial/USB
communications cable, ByteBlaster II parallel port download cable, ByteBlasterMV
parallel port download cable, and EthernetBlaster download cable.
In a PS configuration with a download cable, an intelligent host (such as a PC)
transfers data from a storage device to the device using the USB Blaster, MasterBlaster,
ByteBlaster II, EthernetBlaster, or ByteBlasterMV cable.
After power-up, Stratix IV devices go through a POR. The POR delay depends on the
PORSEL pin setting. When PORSEL is driven low, the standard POR time is
100 ms < TPOR < 300 ms. When PORSEL is driven high, the fast POR time is
4 ms < TPOR < 12 ms. During POR, the device resets, holds nSTATUS low, and tri-states
all user I/O pins. After the device successfully exits POR, all user I/O pins continue to
be tri-stated. If nIO_pullup is driven low during power-up and configuration, the user
I/O pins and dual-purpose I/O pins will have weak pull-up resistors, which are on
(after POR) before and during configuration. If nIO_pullup is driven high, the weak
pull-up resistors are disabled.
Stratix IV Device Handbook
Volume 1
September 2012 Altera Corporation
Chapter 10: Configuration, Design Security, and Remote System Upgrades in Stratix IV Devices
Passive Serial Configuration
10–33
The configuration cycle consists of three stages—reset, configuration, and
initialization. While nCONFIG or nSTATUS are low, the device is in reset. To initiate
configuration in this scheme, the download cable generates a low-to-high transition
on the nCONFIG pin.
1
To begin configuration, power the VCC, VCCIO, VCCPGM, and VCCPD voltages (for the
banks where the configuration pins reside) to the appropriate voltage levels.
When nCONFIG goes high, the device comes out of reset and releases the open-drain
nSTATUS pin, which is then pulled high by an external 10-k pull-up resistor. After
nSTATUS is released, the device is ready to receive configuration data and the
configuration stage begins. The programming hardware or download cable then
places the configuration data one bit at a time on the device’s DATA0 pin. The
configuration data is clocked into the target device until CONF_DONE goes high. The
CONF_DONE pin must have an external 10-k pull-up resistor for the device to initialize.
When using a download cable, setting the Auto-restart configuration after error
option does not affect the configuration cycle because you must manually restart
configuration in the Quartus II software when an error occurs. Additionally, the
Enable user-supplied start-up clock (CLKUSR) option has no affect on the device
initialization because this option is disabled in the .sof when programming the device
using the Quartus II programmer and download cable. Therefore, if you turn on the
CLKUSR option, you do not need to provide a clock on CLKUSR when you are
configuring the device with the Quartus II programmer and a download cable.
Figure 10–14 shows PS configuration for Stratix IV devices using a USB Blaster,
EthernetBlaster, MasterBlaster, ByteBlaster II, or ByteBlasterMV cable.
Figure 10–14. PS Configuration Using a USB Blaster, EthernetBlaster, MasterBlaster, ByteBlaster II, or ByteBlasterMV
Cable
VCCPGM (1)
VCCPGM (1)
10 kΩ
(2)
VCCPGM (1)
VCCPGM (1) VCCPGM (1)
10 kΩ
10 kΩ
Stratix IV Device
VCCPGM
10 kΩ
(2)
MSEL2
10 kΩ
CONF_DONE
nSTATUS
MSEL1
MSEL0
GND
nCE
GND
DCLK
DATA0
nCONFIG
nCEO
Download Cable
10-Pin Male Header
(PS Mode)
N.C.
Pin 1
VCCPGM (1)
GND
VIO (3)
Shield
GND
Notes to Figure 10–14:
(1) Connect the pull-up resistor to the same supply voltage (VCCPGM) as the USB Blaster, MasterBlaster (VIO pin), ByteBlaster II, ByteBlasterMV, or
EthernetBlaster cable.
(2) You only need the pull-up resistors on DATA0 and DCLK if the download cable is the only configuration scheme used on your board. This ensures
that DATA0 and DCLK are not left floating after configuration. For example, if you are also using a configuration device, you do not need the
pull-up resistors on DATA0 and DCLK.
(3) Pin 6 of the header is a VIO reference voltage for the MasterBlaster output driver. VIO must match the device’s VCCPGM. For more information about
this value, refer to the MasterBlaster Serial/USB Communications Cable User Guide. In the USB-Blaster, ByteBlaster II, and ByteBlasterMV cable,
this pin is a no connect.
September 2012
Altera Corporation
Stratix IV Device Handbook
Volume 1
10–34
Chapter 10: Configuration, Design Security, and Remote System Upgrades in Stratix IV Devices
Passive Serial Configuration
You can use a download cable to configure multiple Stratix IV devices by connecting
each device’s nCEO pin to the subsequent device’s nCE pin. The first device’s nCE pin is
connected to GND, while its nCEO pin is connected to the nCE of the next device in the
chain. The last device’s nCE input comes from the previous device, while its nCEO pin is
left floating. All other configuration pins (nCONFIG, nSTATUS, DCLK, DATA0, and
CONF_DONE) are connected to every device in the chain. Because all CONF_DONE pins are
tied together, all devices in the chain initialize and enter user mode at the same time.
In addition, because the nSTATUS pins are tied together, the entire chain halts
configuration if any device detects an error. The Auto-restart configuration after
error option does not affect the configuration cycle because you must manually restart
the configuration in the Quartus II software when an error occurs.
Figure 10–15 shows how to configure multiple Stratix IV devices with a download
cable.
Figure 10–15. Multi-Device PS Configuration Using a USB Blaster, EthernetBlaster, MasterBlaster, ByteBlaster II, or
ByteBlasterMV Cable
VCCPGM (1)
10 kΩ
VCCPGM (1) VCCPGM (1)
10 kΩ
VCCPGM (1)
GND
GND
VCCPGM (1)
(2)
Pin 1
VCCPGM (1)
GND
VIO (3)
nCEO
nCE
10 kΩ
VCCPGM (1)
10 kΩ
CONF_DONE
nSTATUS
DCLK
MSEL2
MSEL1
MSEL0
(2)
10 kΩ
Stratix IV Device 1
Download Cable
10-Pin Male Header
(PS Mode)
VCCPGM (1)
DATA0
nCONFIG
GND
Stratix IV Device 2
MSEL2
MSEL1
MSEL0
CONF_DONE
nSTATUS
DCLK
GND
nCEO
N.C.
nCE
DATA0
nCONFIG
Notes to Figure 10–15:
(1) Connect the pull-up resistor to the same supply voltage (VCCPGM) as the USB Blaster, MasterBlaster (VIO pin), ByteBlaster II, ByteBlasterMV, or
EthernetBlaster cable.
(2) You only need the pull-up resistors on DATA0 and DCLK if the download cable is the only configuration scheme used on your board. This is to
ensure that DATA0 and DCLK are not left floating after configuration. For example, if you are also using a configuration device, you do not need
the pull-up resistors on DATA0 and DCLK.
(3) Pin 6 of the header is a VIO reference voltage for the MasterBlaster output driver. VIO must match the device’s VCCPGM. For more information about
this value, refer to the MasterBlaster Serial/USB Communications Cable User Guide. In the USB-Blaster, ByteBlaster II, and ByteBlasterMV cables,
this pin is a no connect.
Stratix IV Device Handbook
Volume 1
September 2012 Altera Corporation
Chapter 10: Configuration, Design Security, and Remote System Upgrades in Stratix IV Devices
JTAG Configuration
10–35
f For more information about how to use the USB Blaster, MasterBlaster, ByteBlaster II,
or ByteBlasterMV cables, refer to the following user guides:
■
USB-Blaster Download Cable User Guide
■
MasterBlaster Serial/USB Communications Cable User Guide
■
ByteBlaster II Download Cable User Guide
■
ByteBlasterMV Download Cable User Guide
■
EthernetBlaster Communications Cable User Guide
JTAG Configuration
JTAG has developed a specification for boundary-scan testing. This boundary-scan
test (BST) architecture offers the capability to efficiently test components on PCBs
with tight lead spacing. The BST architecture can test pin connections without using
physical test probes and capture functional data while a device is operating normally.
You can also use JTAG circuitry to shift configuration data into the device. The
Quartus II software automatically generates .sofs that you can use for JTAG
configuration with a download cable in the Quartus II software programmer.
f For more information about JTAG boundary-scan testing and commands available
using Stratix IV devices, refer to the following documents:
■
JTAG Boundary Scan Testing in Stratix IV Devices chapter
■
Programming Support for Jam STAPL Language
Stratix IV devices are designed such that JTAG instructions have precedence over any
device configuration modes. Therefore, JTAG configuration can take place without
waiting for other configuration modes to complete. For example, if you attempt JTAG
configuration of Stratix IV devices during PS configuration, PS configuration is
terminated and JTAG configuration begins.
1
You cannot use the Stratix IV decompression or design security features if you are
configuring your Stratix IV device when using JTAG-based configuration.
1
A device operating in JTAG mode uses four required pins, TDI, TDO, TMS, and TCK, and
one optional pin, TRST. The TCK pin has an internal weak pull-down resistor, while the
TDI, TMS, and TRST pins have weak internal pull-up resistors (typically 25 k). The
JTAG output pin TDO and all JTAG input pins are powered by 2.5-V/3.0-V VCCPD. All
the JTAG pins only support the LVTTL I/O standard.
All user I/O pins are tri-stated during JTAG configuration.
f All the JTAG pins are powered by the VCCPD power supply of I/O bank 1A. For more
information about how to connect a JTAG chain with multiple voltages across the
devices in the chain, refer to the JTAG Boundary Scan Testing in Stratix IV Devices
chapter.
September 2012
Altera Corporation
Stratix IV Device Handbook
Volume 1
10–36
Chapter 10: Configuration, Design Security, and Remote System Upgrades in Stratix IV Devices
JTAG Configuration
During JTAG configuration, you can download data to the device on the PCB through
the USB Blaster, MasterBlaster, ByteBlaster II, EthernetBlaster, or ByteBlasterMV
download cable. Configuring devices through a cable is similar to programming
devices in-system, except you must connect the TRST pin to VCCPD. This ensures that
the TAP controller is not reset.
Figure 10–16 shows JTAG configuration of a single Stratix IV device when using a
download cable.
Figure 10–16. JTAG Configuration of a Single Device Using a Download Cable
VCCPD (1)
(5)
VCCPGM
VCCPD (1)
VCCPGM
10 kΩ
Stratix IV Device
10 kΩ
nCE (4)
GND N.C.
(2)
(2)
(2)
nCE0
nSTATUS
CONF_DONE
nCONFIG
MSEL[2..0]
DCLK
(5)
TCK
TDO
TMS
TDI
Download Cable
10-Pin Male Header
(JTAG Mode)
(Top View)
VCCPD (1)
TRST
Pin 1
VCCPD (1)
GND
VIO (3)
1 kΩ
GND
GND
Notes to Figure 10–16:
(1) Connect the pull-up resistor to the same supply voltage as the USB Blaster, MasterBlaster (VIO pin), ByteBlaster II, ByteBlasterMV, or
EthernetBlaster cable. The voltage supply can be connected to the VCCPD of the device.
(2) Connect the nCONFIG and MSEL[2..0] pins to support a non-JTAG configuration scheme. If you only use the JTAG configuration, connect
nCONFIG to VCCPGM and MSEL[2..0] to GND. Pull DCLK either high or low, whichever is convenient on your board.
(3) Pin 6 of the header is a VIO reference voltage for the MasterBlaster output driver. VIO must match the device’s VCCPD. For more information about
this value, refer to the MasterBlaster Serial/USB Communications Cable User Guide. In the USB-Blaster, ByteBlaster II, and ByteBlasterMV cable,
this pin is a no connect.
(4) You must connect nCE to GND or driven low for successful JTAG configuration.
(5) The pull-up resistor value can vary from 1 k to 10 k .
Stratix IV Device Handbook
Volume 1
September 2012 Altera Corporation
Chapter 10: Configuration, Design Security, and Remote System Upgrades in Stratix IV Devices
JTAG Configuration
10–37
To configure a single device in a JTAG chain, the programming software places all
other devices in bypass mode. In bypass mode, devices pass programming data from
the TDI pin to the TDO pin through a single bypass register without being affected
internally. This scheme enables the programming software to program or verify the
target device. Configuration data driven into the device appears on the TDO pin one
clock cycle later.
The Quartus II software verifies successful JTAG configuration upon completion. At
the end of configuration, the software checks the state of CONF_DONE through the JTAG
port. When the Quartus II software generates a JAM file (.jam) for a multi-device
chain, it contains instructions so that all the devices in the chain are initialized at the
same time. If CONF_DONE is not high, the Quartus II software indicates that
configuration has failed. If CONF_DONE is high, the software indicates that
configuration was successful. After the configuration bitstream is transmitted serially
using the JTAG TDI port, the TCK port is clocked an additional 1,094 cycles to perform
device initialization.
Stratix IV devices have dedicated JTAG pins that always function as JTAG pins. Not
only can you perform JTAG testing on Stratix IV devices before and after, but also
during configuration. While other device families do not support JTAG testing during
configuration, Stratix IV devices support the bypass, ID code, and sample instructions
during configuration without interrupting configuration. All other JTAG instructions
may only be issued by first interrupting configuration and reprogramming the I/O
pins using the CONFIG_IO instruction.
The CONFIG_IO instruction allows I/O buffers to be configured using the JTAG port
and when issued, interrupts configuration. This instruction allows you to perform
board-level testing prior to configuring the Stratix IV device or waiting for a
configuration device to complete configuration. After configuration has been
interrupted and JTAG testing is complete, you must reconfigure the part using JTAG
(PULSE_CONFIG instruction) or by pulsing nCONFIG low.
The chip-wide reset (DEV_CLRn) and chip-wide output enable (DEV_OE) pins on
Stratix IV devices do not affect JTAG boundary-scan or programming operations.
Toggling these pins does not affect JTAG operations (other than the usual
boundary-scan operation).
When designing a board for JTAG configuration for Stratix IV devices, consider the
dedicated configuration pins. Table 10–8 lists how these pins are connected during
JTAG configuration.
Table 10–8. Dedicated Configuration Pin Connections During JTAG Configuration (Part 1 of 2)
Signal
September 2012
Description
nCE
On all Stratix IV devices in the chain, nCE must be driven low by connecting it to
GND, pulling it low using a resistor, or driving it by some control circuitry. For
devices that are also in multi-device FPP, AS, or PS configuration chains, the nCE
pins must be connected to GND during JTAG configuration or JTAG must be
configured in the same order as the configuration chain.
nCEO
On all Stratix IV devices in the chain, you can leave nCEO floating or connected to
the nCE of the next device.
MSEL
Do not leave these pins floating. These pins support whichever non-JTAG
configuration is used in production. If you only use JTAG configuration, tie these
pins to GND.
Altera Corporation
Stratix IV Device Handbook
Volume 1
10–38
Chapter 10: Configuration, Design Security, and Remote System Upgrades in Stratix IV Devices
JTAG Configuration
Table 10–8. Dedicated Configuration Pin Connections During JTAG Configuration (Part 2 of 2)
Signal
Description
nCONFIG
Driven high by connecting to VCCPGM, pulling up using a resistor, or driven high by
some control circuitry.
nSTATUS
Pull to VCCPGM using a 10-k resistor. When configuring multiple devices in the
same JTAG chain, each nSTATUS pin must be pulled up to VCCPGM individually.
CONF_DONE
Pull to VCCPGM using a 10-k resistor. When configuring multiple devices in the
same JTAG chain, each CONF_DONE pin must be pulled up to VCCPGM individually.
CONF_DONE going high at the end of JTAG configuration indicates successful
configuration.
DCLK
Do not leave DCLK floating. Drive low or high, whichever is more convenient on
your board.
When programming a JTAG device chain, one JTAG-compatible header is connected
to several devices. The number of devices in the JTAG chain is limited only by the
drive capability of the download cable. When four or more devices are connected in a
JTAG chain, Altera recommends buffering the TCK, TDI, and TMS pins with an on-board
buffer.
JTAG-chain device programming is ideal when the system contains multiple devices,
or when testing your system using JTAG BST circuitry.
Figure 10–17 shows a multi-device JTAG configuration when using a download cable.
Figure 10–17. JTAG Configuration of Multiple Devices Using a Download Cable
Stratix IV Device
Download Cable
10-Pin Male Header
(JTAG Mode)
VCCPGM
(2)
Pin 1
VCCPGM
(5)
VCCPD (1)
(1) VCCPD
(2)
(5)
VCCPD (1)
VIO
(3)
VCCPGM
(2)
DCLK
MSEL[2..0]
nCE (4)
10 kΩ
nSTATUS
nCONFIG
TRST
TDI
TMS
TDO
TCK
10 kΩ
(2)
nSTATUS
nCONFIG
(2)
DCLK
(2)
MSEL[2..0]
CONF_DONE
CONF_DONE
(2)
DCLK
(2)
MSEL[2..0]
VCCPD (1)
VCCPGM
VCCPGM
10 kΩ
10 kΩ
nSTATUS
nCONFIG
CONF_DONE
(2)
VCCPGM
10 kΩ
10 kΩ
VCCPD (1)
Stratix II or Stratix II GX
Stratix
IV Device
Device
Stratix IV Device
nCE (4)
TRST
TDI
TMS
VCCPD (1)
TDO
TCK
nCE (4)
TRST
TDI
TMS
TDO
TCK
1 kΩ
Notes to Figure 10–17:
(1) Connect the pull-up resistor to the same supply voltage as the USB Blaster, MasterBlaster (VIO pin), ByteBlaster II, ByteBlasterMV, or
EthernetBlaster cable. Connect the voltage supply to VCCPD of the device.
(2) Connect the nCONFIG and MSEL[2..0] pins to support a non-JTAG configuration scheme. If you only use a JTAG configuration, connect
nCONFIG to VCCPGM and MSEL[2..0] to GND. Pull DCLK either high or low, whichever is convenient on your board.
(3) Pin 6 of the header is a VIO reference voltage for the MasterBlaster output driver. VIO must match the device’s VCCPD. For more information about
this value, refer to the MasterBlaster Serial/USB Communications Cable User Guide. In the USB-Blaster, ByteBlaster II, and ByteBlasterMV cables,
this pin is a no connect.
(4) You must connect nCE to GND or drive it low for successful JTAG configuration.
(5) The pull-up resistor value can vary from 1 k to 10 k .
Stratix IV Device Handbook
Volume 1
September 2012 Altera Corporation
Chapter 10: Configuration, Design Security, and Remote System Upgrades in Stratix IV Devices
JTAG Configuration
10–39
You must connect the nCE pin to GND or drive it low during JTAG configuration. In
multi-device FPP, AS, and PS configuration chains, the first device’s nCE pin is
connected to GND, while its nCEO pin is connected to nCE of the next device in the
chain. The last device’s nCE input comes from the previous device, while its nCEO pin is
left floating. In addition, the CONF_DONE and nSTATUS signals are all shared in
multi-device FPP, AS, or PS configuration chains so the devices can enter user mode at
the same time after configuration is complete. When the CONF_DONE and nSTATUS
signals are shared among all the devices, you must configure every device when JTAG
configuration is performed.
If you only use JTAG configuration, Altera recommends connecting the circuitry as
shown in Figure 10–17, where each of the CONF_DONE and nSTATUS signals are isolated,
so that each device can enter user mode individually.
After the first device completes configuration in a multi-device configuration chain,
its nCEO pin drives low to activate the second device’s nCE pin, which prompts the
second device to begin configuration. Therefore, if these devices are also in a JTAG
chain, ensure the nCE pins are connected to GND during JTAG configuration or that
the devices are JTAG configured in the same order as the configuration chain. As long
as the devices are JTAG configured in the same order as the multi-device
configuration chain, the nCEO of the previous device drives the nCE of the next device
low when it has successfully been JTAG configured.
You can place other Altera devices that have JTAG support in the same JTAG chain for
device programming and configuration.
1
JTAG configuration support is enhanced and allows more than 17 Stratix IV devices to
be cascaded in a JTAG chain.
f For more information about configuring multiple Altera devices in the same
configuration chain, refer to the Configuring Mixed Altera FPGA Chains chapter in
volume 2 of the Configuration Handbook.
You can configure Stratix IV devices using multiple configuration schemes on the
same board. Combining JTAG configuration with AS configuration on your board is
useful in the prototyping environment because it allows multiple methods to
configure your FPGA.
f For more information about combining JTAG configuration with other configuration
schemes, refer to the Combining Different Configuration Schemes chapter in volume 2 of
the Configuration Handbook.
September 2012
Altera Corporation
Stratix IV Device Handbook
Volume 1
10–40
Chapter 10: Configuration, Design Security, and Remote System Upgrades in Stratix IV Devices
Device Configuration Pins
Figure 10–18 shows JTAG configuration of a Stratix IV device using a microprocessor.
Figure 10–18. JTAG Configuration of a Single Device Using a Microprocessor
VCCPGM (1)
VCCPGM (1)
Memory
ADDR
Stratix IV Device
10 kΩ
10 kΩ
DATA
nSTATUS
VCCPD
TRST
TDI (4)
TCK (4)
TMS (4)
TDO (4)
Microprocessor
CONF_DONE
DCLK
nCONFIG
MSEL[2..0]
nCEO
(2)
(2)
(2)
N.C.
(3) nCE
GND
Notes to Figure 10–18:
(1) Connect the pull-up resistor to a supply that provides an acceptable input signal for all Stratix IV devices in the chain.
VCCPGM must be high enough to meet the VIH specification of the I/O on the device.
(2) Connect the nCONFIG and MSEL[2..0] pins to support a non-JTAG configuration scheme. If you use only a JTAG
configuration, connect nCONFIG to VCCGPM and MSEL[2..0] to GND. Pull DCLK either high or low, whichever is
convenient on your board.
(3) Connect nCE to GND or drive it low for successful JTAG configuration.
(4) The microprocessor must use the same I/O standard as VCCPD to drive the JTAG pins.
Jam STAPL
Jam STAPL, JEDEC standard JESD-71, is a standard file format for in-system
programmability (ISP) purposes. Jam STAPL supports programming or configuration
of programmable devices and testing of electronic systems, using the IEEE 1149.1
JTAG interface. Jam STAPL is a freely licensed open standard.
The Jam Player provides an interface for manipulating the IEEE Std. 1149.1 JTAG TAP
state machine.
f For more information about JTAG and Jam STAPL in embedded environments, refer
to Using Jam STAPL for ISP via an Embedded Processor. To download the Jam Player,
visit the Altera website at www.altera.com.
Device Configuration Pins
The following tables list the connections and functionality of all the
configuration-related pins on Stratix IV devices. Table 10–9 lists the Stratix IV
configuration pins and their power supply.
Table 10–9. Stratix IV Configuration Pin Summary (Part 1 of 2) (Note 1)
Description
Input/Output
Dedicated
Powered By
Configuration Mode
TDI
Input
Yes
VCCPD
JTAG
TMS
Input
Yes
VCCPD
JTAG
TCK
Input
Yes
VCCPD
JTAG
TRST
Input
Yes
VCCPD
JTAG
TDO
Output
Yes
VCCPD
JTAG
CRC_ERROR
Output
—
Pull-up
Optional, all modes
Stratix IV Device Handbook
Volume 1
September 2012 Altera Corporation
Chapter 10: Configuration, Design Security, and Remote System Upgrades in Stratix IV Devices
Device Configuration Pins
10–41
Table 10–9. Stratix IV Configuration Pin Summary (Part 2 of 2) (Note 1)
Description
Input/Output
Dedicated
Powered By
Configuration Mode
DATA0
Input
—
VCCPGM/VCCIO (3)
All modes except JTAG
DATA[7..1]
Input
—
VCCPGM/VCCIO (3)
FPP
INIT_DONE
Output
—
Pull-up
Optional, all modes
CLKUSR
Input
—
VCCPGM/VCCIO (3)
Optional
nSTATUS
Bidirectional
Yes
VCCPGM/Pull-up
All modes
Input
Yes
VCCPGM
All modes
Bidirectional
Yes
VCCPGM/Pull-up
All modes
nCONFIG
Input
Yes
VCCPGM
All modes
PORSEL
Input
Yes
VCC (2)
All modes
Yes
VCCPGM
AS
nCE
CONF_DONE
ASDO
(4)
Output
nCSO
(4)
Output
Yes
VCCPGM
AS
Input
Yes
VCCPGM
PS, FPP
Output
Yes
VCCPGM
AS
Input
Yes
VCC (2)
All modes
Output
Yes
VCCPGM
All modes
Input
Yes
VCC (2)
All modes
DCLK (4)
nIO_PULLUP
nCEO
MSEL[2..0]
Notes to Table 10–9:
(1) The total number of pins is 29. The total number of dedicated pins is 18.
(2) Although MSEL[2..0], PORSEL, and nIO_PULLUP are powered up by VCC, Altera recommends connecting these pins to VCCPGM or GND directly
without using a pull-up or pull-down resistor.
(3) These pins are powered up by VCCPGM during configuration. These pins are powered up by VCCIO if they are used as regular I/O in user mode.
(4) To tri-state this pin, in the Quartus II software, on the Assignments menu, select Device. On the Device page, select Device and Pin Options...
On the Device and Pin Options page, select Configuration and select the Enable input tri-state on active configuration pins in user mode
option.
September 2012
Altera Corporation
Stratix IV Device Handbook
Volume 1
10–42
Chapter 10: Configuration, Design Security, and Remote System Upgrades in Stratix IV Devices
Device Configuration Pins
Table 10–10 lists the dedicated configuration pins. You must connect these pins
properly on your board for successful configuration. Some of these pins may not be
required for your configuration schemes.
Table 10–10. Dedicated Configuration Pins on the Stratix IV Device (Part 1 of 4)
Pin Name
User Mode
Configuration
Scheme
Pin Type
Description
Dedicated power pin. Use this pin to power all dedicated
configuration inputs, dedicated configuration outputs,
dedicated configuration bidirectional pins, and some of the
dual functional pins that are used for configuration.
VCCPGM
N/A
All
Power
You must connect this pin to 1.8, 2.5, or 3.0 V. VCCPGM must
ramp-up from 0 V to VCCPGM within 100 ms when PORSEL is
low or 4 ms when PORSEL is high. If VCCPGM is not ramped
up within this specified time, your Stratix IV device will not
configure successfully. If your system does not allow a
VCCPGM ramp-up within 100 ms or 4 ms, you must hold
nCONFIG low until all power supplies are stable.
Dedicated power pin. Use this pin to power the I/O
pre-drivers, JTAG input and output pins, and design
security circuitry.
VCCPD
N/A
All
Power
You must connect this pin to 2.5 V or 3.0 V, depending on
the I/O standards selected. For the 3.0-V I/O standard,
VCCPD = 3.0 V. For the 2.5 V or below I/O standards,
VCCPD = 2.5 V.
VCCPD must ramp-up from 0 V to 2.5 V / 3.0 V within
100 ms when PORSEL is low or 4 ms when PORSEL is high.
If VCCPD is not ramped up within this specified time, your
Stratix IV device will not configure successfully. If your
system does not allow a VCCPD to ramp-up time within
100 ms or 4 ms, you must hold nCONFIG low until all
power supplies are stable.
PORSEL
N/A
All
Input
Dedicated input that selects between a standard POR time
or a fast POR time. A logic low selects a standard POR time
setting of 100 ms < TPOR < 300 ms and a logic high selects
a fast POR time setting of 4 ms < TPOR < 12 ms.
The PORSEL input buffer is powered by VCC and has an
internal 5-kpull-down resistor that is always active. Tie
the PORSEL pin directly to VCCPGM or GND.
nIO_PULLUP
N/A
All
Input
Dedicated input that chooses whether the internal pull-up
resistors on the user I/O pins and dual-purpose I/O pins
(nCSO, nASDO, DATA[7..0], CLKUSR, and INIT_DONE) are
on or off before and during configuration. A logic high turns
off the weak internal pull-up resistors; a logic low turns
them on.
The nIO-PULLUP input buffer is powered by VCC and has an
internal 5-k pull-down resistor that is always active. The
nIO-PULLUP can be tied directly to VCCPGM, using a 1-k
pull-up resistor or tied directly to GND, depending on your
device requirements.
Stratix IV Device Handbook
Volume 1
September 2012 Altera Corporation
Chapter 10: Configuration, Design Security, and Remote System Upgrades in Stratix IV Devices
Device Configuration Pins
10–43
Table 10–10. Dedicated Configuration Pins on the Stratix IV Device (Part 2 of 4)
Pin Name
User Mode
Configuration
Scheme
Pin Type
Description
Three-bit configuration input that sets the Stratix IV device
configuration scheme. For the appropriate connections,
refer to Table 10–1 on page 10–2.
MSEL[2..0]
N/A
All
Input
You must hardwire these pins to VCCPGM or GND.
The MSEL[2..0] pins have internal 5-k pull-down
resistors that are always active.
nCONFIG
N/A
All
Input
Configuration control input. Pulling this pin low during
user-mode causes the device to lose its configuration data,
enter a reset state, and tri-state all I/O pins. Returning this
pin to a logic high level initiates a reconfiguration.
Configuration is possible only if this pin is high, except in
JTAG programming mode, when nCONFIG is ignored.
The device drives nSTATUS low immediately after power-up
and releases it after the POR time.
During user mode and regular configuration, this pin is
pulled high by an external 10-k resistor.
This pin, when driven low by the Stratix IV device, indicates
that the device has encountered an error during
configuration.
nSTATUS
N/A
All
Bidirectional
open-drain
■
Status output—If an error occurs during configuration,
nSTATUS is pulled low by the target device.
■
Status input—If an external source drives the nSTATUS
pin low during configuration or initialization, the target
device enters an error state.
Driving nSTATUS low after configuration and initialization
does not affect the configured device. If you use a
configuration device, driving nSTATUS low causes the
configuration device to attempt to configure the device, but
because the device ignores transitions on nSTATUS in user
mode, the device does not reconfigure. To initiate a
reconfiguration, nCONFIG must be pulled low.
If you have enabled the Auto-restart configuration after
error option, the nSTATUS pin transitions from high to low
and back again to high when a configuration error is
detected. This appears as a low pulse at the pin with a
minimum pulse width of 10 s to a maximum pulse width
of 500 s, as defined in the tSTATUS specification.
September 2012
Altera Corporation
Stratix IV Device Handbook
Volume 1
10–44
Chapter 10: Configuration, Design Security, and Remote System Upgrades in Stratix IV Devices
Device Configuration Pins
Table 10–10. Dedicated Configuration Pins on the Stratix IV Device (Part 3 of 4)
Pin Name
User Mode
Configuration
Scheme
Pin Type
Description
If VCCPGM is not fully powered up, the following could occur:
nSTATUS
(continued)
—
—
—
■
VCCPGM is powered high enough for the nSTATUS buffer
to function properly and nSTATUS is driven low. When
VCCPGM is ramped up, POR trips and nSTATUS is released
after POR expires.
■
VCCPGM is not powered high enough for the nSTATUS
buffer to function properly. In this situation, nSTATUS
might appear logic high, triggering a configuration
attempt that would fail because POR did not yet trip.
When VCCPD is powered up, nSTATUS is pulled low
because POR did not yet trip. When POR trips after
VCCPGM is powered up, nSTATUS is released and pulled
high. At that point, reconfiguration is triggered and the
device is configured.
Status output. The target device drives the CONF_DONE pin
low before and during configuration. After all the
configuration data is received without error and the
initialization cycle starts, the target device releases
CONF_DONE.
CONF_DONE
N/A
All
Bidirectional
open-drain
Status input. After all the data is received and CONF_DONE
goes high, the target device initializes and enters user
mode. The CONF_DONE pin must have an external 10-k
pull-up resistor for the device to initialize.
Driving CONF_DONE low after configuration and initialization
does not affect the configured device.
nCE
N/A
All
Input
Active-low chip enable. The nCE pin activates the device
with a low signal to allow configuration. The nCE pin must
be held low during configuration, initialization, and user
mode. In single device configuration, it must be tied low. In
multi-device configuration, nCE of the first device is tied
low, while its nCEO pin is connected to nCE of the next
device in the chain.
The nCE pin must also be held low for successful JTAG
programming of the device.
nCEO
N/A
All
Output
Output that drives low when device configuration is
complete. In single device configuration, this pin is left
floating. In multi-device configuration, this pin feeds the
next device’s nCE pin. The nCEO of the last device in the
chain is left floating.
The nCEO pin is powered by VCCPGM.
ASDO
N/A
AS
Output
Control signal from the Stratix IV device to the serial
configuration device in AS mode used to read out
configuration data.
In AS mode, ASDO has an internal pull-up resistor that is
always active.
Stratix IV Device Handbook
Volume 1
September 2012 Altera Corporation
Chapter 10: Configuration, Design Security, and Remote System Upgrades in Stratix IV Devices
Device Configuration Pins
10–45
Table 10–10. Dedicated Configuration Pins on the Stratix IV Device (Part 4 of 4)
Pin Name
nCSO
User Mode
Configuration
Scheme
N/A
AS
Pin Type
Description
Output
Output control signal from the Stratix IV device to the serial
configuration device in AS mode that enables the
configuration device.
In AS mode, nCSO has an internal pull-up resistor that is
always active.
In PS and FPP configurations, DCLK is the clock input used
to clock data from an external source into the target device.
Data is latched into the device on the rising edge of DCLK.
DCLK
N/A
Synchronous
configuration
schemes
(PS, FPP, AS)
In AS mode, DCLK is an output from the Stratix IV device
that provides timing for the configuration interface. In AS
mode, DCLK has an internal pull-up resistor (typically
25 k) that is always active.
Input
(PS, FPP)
Output (AS)
In AS configuration schemes, this pin is driven into an
inactive state after configuration completes. You can use
this pin as a user I/O during user mode.
In PS or FPP schemes that use a control host, you must
drive DCLK either high or low, whichever is more
convenient. In passive schemes, you cannot use DCLK as a
user I/O during user mode.
Toggling this pin after configuration does not affect the
configured device.
DATA0
N/A in AS
mode. I/O
in PS or
FPP mode.
Data input. In serial configuration modes, bit-wide
configuration data is presented to the target device on the
DATA0 pin.
PS, FPP, AS
Input
In AS mode, DATA0 has an internal pull-up resistor that is
always active.
After PS or FPP configuration, DATA0 is available as a user
I/O pin. The state of this pin depends on the Dual-Purpose
Pin settings.
Data inputs. Byte-wide configuration data is presented to
the target device on DATA[7..0].
DATA[7..1]
September 2012
I/O
Altera Corporation
Parallel
configuration
schemes
(FPP)
Inputs
In serial configuration schemes, they function as user I/O
pins during configuration, which means they are tri-stated.
After FPP configuration, DATA[7..1] are available as user
I/O pins. The state of these pins depends on the
Dual-Purpose Pin settings.
Stratix IV Device Handbook
Volume 1
10–46
Chapter 10: Configuration, Design Security, and Remote System Upgrades in Stratix IV Devices
Device Configuration Pins
Table 10–11 lists the optional configuration pins. If these optional configuration pins
are not enabled in the Quartus II software, they are available as general-purpose user
I/O pins. Therefore, during configuration, these pins function as user I/O pins and
are tri-stated with weak pull-up resistors.
Table 10–11. Optional Configuration Pins
Pin Name
CLKUSR
INIT_DONE
DEV_OE
DEV_CLRn
User Mode
N/A if option is on.
I/O if option is off.
N/A if option is on.
I/O if option is off.
N/A if option is on.
I/O if option is off.
N/A if option is on.
I/O if option is off.
Stratix IV Device Handbook
Volume 1
Pin Type
Input
Description
Optional user-supplied clock input synchronizes the initialization of
one or more devices. Enable this pin by turning on the Enable
user-supplied start-up clock (CLKUSR) option in the Quartus II
software.
Output
open-drain
Use as a status pin to indicate when the device has initialized and is
in user mode. When nCONFIG is low and during the beginning of
configuration, the INIT_DONE pin is tri-stated and pulled high due to
an external 10-k pull-up resistor. After the option bit to enable
INIT_DONE is programmed into the device (during the first frame of
configuration data), the INIT_DONE pin goes low. When initialization
is complete, the INIT_DONE pin is released and pulled high and the
device enters user mode. Thus, the monitoring circuitry must be able
to detect a low-to-high transition. Enable this pin by turning on the
Enable INIT_DONE output option in the Quartus II software.
Input
Optional pin that allows you to override all tri-states on the device.
When this pin is driven low, all I/O pins are tri-stated. When this pin
is driven high, all I/O pins behave as programmed. Enable this pin by
turning on the Enable device-wide output enable (DEV_OE) option
in the Quartus II software.
Input
Optional pin that allows you to override all clears on all device
registers. When this pin is driven low, all registers are cleared. When
this pin is driven high, all registers behave as programmed. Enable
this pin by turning on the Enable device-wide reset (DEV_CLRn)
option in the Quartus II software.
September 2012 Altera Corporation
Chapter 10: Configuration, Design Security, and Remote System Upgrades in Stratix IV Devices
Device Configuration Pins
10–47
Table 10–12 lists the dedicated JTAG pins. JTAG pins must be kept stable before and
during configuration to prevent accidental loading of JTAG instructions. The TDI,
TMS, and TRST pins have weak internal pull-up resistors, while TCK has a weak
internal pull-down resistor (typically 25 k ). If you plan to use the SignalTap®
embedded logic array analyzer, you must connect the JTAG pins of the Stratix IV
device to a JTAG header on your board.
Table 10–12. Dedicated JTAG Pins
Pin
Name
User
Mode
TDI
TDO
N/A
N/A
Pin Type
Test data
input
Test data
output
Description
Serial input pin for instructions as well as test and programming data. Data is shifted on
the rising edge of TCK. The TDI pin is powered by the 2.5-V/3.0-V VCCPD supply.
If the JTAG interface is not required on your board, you can disable the JTAG circuitry by
connecting this pin to logic high using a 1-k resistor.
Serial data output pin for instructions as well as test and programming data. Data is
shifted out on the falling edge of TCK. The pin is tri-stated if data is not being shifted out of
the device. The TDO pin is powered by VCCPD. For recommendations about connecting a
JTAG chain with multiple voltages across the devices in the chain, refer to the JTAG
Boundary Scan Testing in Stratix IV Devices chapter.
If the JTAG interface is not required on your board, you can disable the JTAG circuitry by
leaving this pin unconnected.
TMS
N/A
Input pin that provides the control signal to determine the transitions of the TAP controller
state machine. TMS is evaluated on the rising edge of TCK. Therefore, you must set up TMS
before the rising edge of TCK. Transitions within the state machine occur on the falling
Test mode edge of TCK after the signal is applied to TMS. The TMS pin is powered by 2.5-V/3.0-V
select
VCCPD.
If the JTAG interface is not required on your board, you can disable the JTAG circuitry by
connecting this pin to logic high using a 1-k resistor.
TCK
N/A
Test clock
input
Clock input to the BST circuitry. Some operations occur at the rising edge, while others
occur at the falling edge. The TCK pin is powered by the 2.5-V/3.0-V VCCPD supply.
It is expected that the clock input waveform have a nominal 50% duty cycle.
If the JTAG interface is not required on your board, you can disable the JTAG circuitry by
connecting TCK to GND.
Active-low input to asynchronously reset the boundary-scan circuit. The TRST pin is
optional according to IEEE Std. 1149.1. The TRST pin is powered by the 2.5-V/3.0-V VCCPD
supply.
TRST
N/A
Test reset
input
Hold TMS at 1 or keep TCK static while TRST is changed from 0 to 1.
(optional)
If the JTAG interface is not required on your board, you can disable the JTAG circuitry by
connecting the TRST pin to GND.
f For more information about the pin connection recommendations, refer to the
Stratix IV GX and Stratix IV E Device Family Pin Connection Guidelines.
September 2012
Altera Corporation
Stratix IV Device Handbook
Volume 1
10–48
Chapter 10: Configuration, Design Security, and Remote System Upgrades in Stratix IV Devices
Configuration Data Decompression
Configuration Data Decompression
Stratix IV devices support configuration data decompression, which saves
configuration memory space and time. This feature allows you to store compressed
configuration data in configuration devices or other memory and transmit this
compressed bitstream to Stratix IV devices. During configuration, the Stratix IV
device decompresses the bitstream in real time and programs its SRAM cells.
1
Preliminary data indicates that compression typically reduces the configuration
bitstream size by 30% to 55% based on the designs used.
Stratix IV devices support decompression in the FPP (when using a MAX II device or
microprocessor + flash), fast AS, and PS configuration schemes. The Stratix IV
decompression feature is not available in the JTAG configuration scheme.
In PS mode, use the Stratix IV decompression feature because sending compressed
configuration data reduces configuration time.
When you enable compression, the Quartus II software generates configuration files
with compressed configuration data. This compressed file reduces the storage
requirements in the configuration device or flash memory, and decreases the time
needed to transmit the bitstream to the Stratix IV device. The time required by a
Stratix IV device to decompress a configuration file is less than the time needed to
transmit the configuration data to the device.
There are two ways to enable compression for Stratix IV bitstreams—before design
compilation (in the Compiler Settings menu) and after design compilation (in the
Convert Programming Files window).
To enable compression in the project’s Compiler Settings menu, follow these steps:
1. On the Assignments menu, click Device to bring up the Settings dialog box.
2. After selecting your Stratix IV device, open the Device and Pin Options window.
Stratix IV Device Handbook
Volume 1
September 2012 Altera Corporation
Chapter 10: Configuration, Design Security, and Remote System Upgrades in Stratix IV Devices
Configuration Data Decompression
10–49
3. In the Configuration settings tab, turn on Generate compressed bitstreams (as
shown in Figure 10–19).
Figure 10–19. Enabling Compression for Stratix IV Bitstreams in Compiler Settings
You can also enable compression when creating programming files from the Convert
Programming Files window. To do this, follow these steps:
1. On the File menu, click Convert Programming Files.
2. Select the programming file type (.pof, .sram, .hex, .rbf, or .ttf).
3. For .pof output files, select a configuration device.
4. In the Input files to convert box, select SOF Data.
5. Select Add File and add a Stratix IV device .sof file.
6. Select the name of the file you added to the SOF Data area and click Properties.
7. Check the Compression check box.
September 2012
Altera Corporation
Stratix IV Device Handbook
Volume 1
10–50
Chapter 10: Configuration, Design Security, and Remote System Upgrades in Stratix IV Devices
Remote System Upgrades
When multiple Stratix IV devices are cascaded, you can selectively enable the
compression feature for each device in the chain if you are using a serial configuration
scheme. Figure 10–20 shows a chain of two Stratix IV devices. The first Stratix IV
device has compression enabled; therefore, receives a compressed bitstream from the
configuration device. The second Stratix IV device has the compression feature
disabled and receives uncompressed data.
In a multi-device FPP configuration chain (with a MAX II device or microprocessor +
flash), all Stratix IV devices in the chain must either enable or disable the
decompression feature. You cannot selectively enable the compression feature for
each device in the chain because of the DATA and DCLK relationship.
Figure 10–20. Compressed and Uncompressed Configuration Data in the Same Configuration File
Serial Configuration Data
Serial Configuration
Device
Uncompressed
Configuration
Data
Compressed
Configuration
Data
Decompression
Controller
Stratix IV
Device
Stratix IV
Device
nCE
nCEO
nCE
nCEO
N.C.
GND
You can generate programming files for this setup by clicking Convert Programming
Files on the File menu in the Quartus II software.
Remote System Upgrades
This section describes the functionality and implementation of the dedicated remote
system upgrade circuitry. It also defines several concepts related to remote system
upgrade, including factory configuration, application configuration, remote update
mode, and user watchdog timer. Additionally, this section provides design guidelines
for implementing remote system upgrades with the supported configuration
schemes.
System designers sometimes face challenges such as shortened design cycles,
evolving standards, and system deployments in remote locations. Stratix IV devices
help overcome these challenges with their inherent reprogrammability and dedicated
circuitry to perform remote system upgrades. Remote system upgrades help deliver
feature enhancements and bug fixes without costly recalls, reduce time-to-market,
extend product life, and avoid system downtime.
Stratix IV devices feature dedicated remote system upgrade circuitry. Soft logic (either
the Nios® II embedded processor or user logic) implemented in a Stratix IV device can
download a new configuration image from a remote location, store it in configuration
memory, and direct the dedicated remote system upgrade circuitry to initiate a
reconfiguration cycle. The dedicated circuitry performs error detection during and
after the configuration process, recovers from any error condition by reverting back to
a safe configuration image, and provides error status information.
Stratix IV Device Handbook
Volume 1
September 2012 Altera Corporation
Chapter 10: Configuration, Design Security, and Remote System Upgrades in Stratix IV Devices
Remote System Upgrades
10–51
Remote system upgrade is supported in fast AS Stratix IV configuration schemes. You
can also implement remote system upgrade in conjunction with advanced Stratix IV
features such as real-time decompression of configuration data and design security
using the advanced encryption standard (AES) for secure and efficient field upgrades.
The largest serial configuration device currently supports 128 Mbits of configuration
bitstream.
1
Stratix IV devices only support remote system upgrade in the single device fast AS
configuration scheme. Because the largest serial configuration device currently
supports 128 Mbits of configuration bitstream, the remote system upgrade feature is
not supported in EP4SGX290, EP4SE360, and larger devices.
1
The remote system upgrade feature is not supported in a multi-device chain.
Functional Description
The dedicated remote system upgrade circuitry in Stratix IV devices manages remote
configuration and provides error detection, recovery, and status information. User
logic or a Nios II processor implemented in the Stratix IV device logic array provides
access to the remote configuration data source and an interface to the system’s
configuration memory.
Stratix IV devices have remote system upgrade processes that involve the following
steps:
1. A Nios II processor (or user logic) implemented in the Stratix IV device logic array
receives new configuration data from a remote location. The connection to the
remote source uses a communication protocol such as the transmission control
protocol/Internet protocol (TCP/IP), peripheral component interconnect (PCI),
user datagram protocol (UDP), universal asynchronous receiver/transmitter
(UART), or a proprietary interface.
2. The Nios II processor (or user logic) stores this new configuration data in
non-volatile configuration memory.
3. The Nios II processor (or user logic) initiates a reconfiguration cycle with the new
or updated configuration data.
4. The dedicated remote system upgrade circuitry detects and recovers from any
error(s) that might occur during or after the reconfiguration cycle and provides
error status information to the user design.
September 2012
Altera Corporation
Stratix IV Device Handbook
Volume 1
10–52
Chapter 10: Configuration, Design Security, and Remote System Upgrades in Stratix IV Devices
Remote System Upgrades
Figure 10–21 shows the steps required for performing remote configuration updates.
(The numbers in Figure 10–21 coincide with the steps just mentioned.)
Figure 10–21. Functional Diagram of Stratix IV Remote System Upgrade
1
2
Development
Location
Data
Stratix IV
Device
Control Module
Data
Configuration
Memory
Data
Stratix IV Configuration
3
Figure 10–22 shows a block diagram for implementing a remote system upgrade with
the Stratix IV fast AS configuration scheme.
Figure 10–22. Remote System Upgrade Block Diagram for Stratix IV Fast AS Configuration
Scheme
Stratix IV
Device
Nios II Processor
or User Logic
Serial
Configuration
Device
You must set the mode select pins (MSEL[2..0]) to fast AS mode to use remote system
upgrade in your system. Table 10–13 lists the MSEL pin settings for Stratix IV devices in
standard configuration mode and remote system upgrade mode. The following
sections describe remote update of the remote system upgrade mode.
For more information about standard configuration schemes supported in Stratix IV
devices, refer to “Configuration Schemes” on page 10–2.
Table 10–13. Remote System Upgrade Modes in Stratix IV Devices
Configuration Scheme
Fast AS (40 MHz)
MSEL[2..0]
Remote System Upgrade Mode
011
Standard
011
Remote update (1)
Note to Table 10–13:
(1) All EPCS densities are able to support DCLK up to 40 MHz, but batches of EPCS1 and EPCS4 manufactured on
0.18-m process geometry can only support DCLK up to 20 MHz. For more information, refer to the Serial
Configuration Devices (EPCS1, EPCS4, EPCS16, EPCS64, and EPCS128) Data Sheet chapter in volume 2 of the
Configuration Handbook.
Stratix IV Device Handbook
Volume 1
September 2012 Altera Corporation
Chapter 10: Configuration, Design Security, and Remote System Upgrades in Stratix IV Devices
Remote System Upgrades
1
10–53
When using fast AS mode, you must select remote update mode in the Quartus II
software and insert the ALTREMOTE_UPDATE megafunction to access the circuitry.
For more information, refer to “ALTREMOTE_UPDATE Megafunction” on
page 10–62.
Enabling Remote Update
You can enable remote update for Stratix IV devices in the Quartus II software before
design compilation (in the Compiler Settings menu). In remote update mode, the
auto-restart configuration after error option is always enabled. To enable remote
update in the project’s compiler settings, in the Quartus II software, follow these
steps:
1. On the Assignment menu, click Device. The Settings dialog box appears.
2. Click Device and Pin Options. The Device and Pin Options dialog box appears.
3. Click the Configuration tab.
4. From the Configuration scheme list, select Active Serial (you can also use
Configuration Device) (Figure 10–23).
5. From the Configuration Mode list, select Remote (Figure 10–23).
6. Click OK.
7. In the Settings dialog box, click OK.
Figure 10–23. Enabling Remote Update for Stratix IV Devices in the Compiler Settings Menu
September 2012
Altera Corporation
Stratix IV Device Handbook
Volume 1
10–54
Chapter 10: Configuration, Design Security, and Remote System Upgrades in Stratix IV Devices
Remote System Upgrade Mode
Configuration Image Types
When performing a remote system upgrade, Stratix IV device configuration
bitstreams are classified as factory configuration images or application configuration
images. An image, also referred to as a configuration, is a design loaded into the
Stratix IV device that performs certain user-defined functions.
Each Stratix IV device in your system requires one factory image or the addition of
one or more application images. The factory image is a user-defined fall-back, or safe
configuration, and is responsible for administering remote updates in conjunction
with the dedicated circuitry. Application images implement user-defined
functionality in the target Stratix IV device. You may include the default application
image functionality in the factory image.
A remote system upgrade involves storing a new application configuration image or
updating an existing one using the remote communication interface. After an
application configuration image is stored or updated remotely, the user design in the
Stratix IV device initiates a reconfiguration cycle with the new image. Any errors
during or after this cycle are detected by the dedicated remote system upgrade
circuitry and cause the device to automatically revert to the factory image. The factory
image then performs error processing and recovery. The factory configuration is
written to the serial configuration device only once by the system manufacturer and
must not be remotely updated. On the other hand, application configurations may be
remotely updated in the system. Both images can initiate system reconfiguration.
Remote System Upgrade Mode
Remote system upgrade has one mode of operation—remote update mode. Remote
update mode allows you to determine the functionality of your system after
power-up and offers several features.
Remote Update Mode
In remote update mode, Stratix IV devices load the factory configuration image after
power up. The user-defined factory configuration determines which application
configuration is to be loaded and triggers a reconfiguration cycle. The factory
configuration may also contain application logic.
When used with serial configuration devices, remote update mode allows an
application configuration to start at any flash sector boundary. For example, this
translates to a maximum of 128 sectors in the EPCS64 device and 32 sectors in the
EPCS16 device, where the minimum size of each page is 512 KBits. Altera
recommends not using the same page in the serial configuration devices for two
images. Additionally, remote update mode features a user watchdog timer that
determines the validity of an application configuration.
When a Stratix IV device is first powered up in remote update mode, it loads the
factory configuration located at page zero (page registers PGM[23:0] = 24'b0). Always
store the factory configuration image for your system at page address zero. This
corresponds to the start address location 0×000000 in the serial configuration device.
Stratix IV Device Handbook
Volume 1
September 2012 Altera Corporation
Chapter 10: Configuration, Design Security, and Remote System Upgrades in Stratix IV Devices
Remote System Upgrade Mode
10–55
The factory image is user-designed and contains soft logic to:
■
Process any errors based on status information from the dedicated remote system
upgrade circuitry
■
Communicate with the remote host and receive new application configurations
and store this new configuration data in the local non-volatile memory device
■
Determine which application configuration is to be loaded into the Stratix IV
device
■
Enable or disable the user watchdog timer and load its time-out value (optional)
■
Instruct the dedicated remote system upgrade circuitry to initiate a
reconfiguration cycle
Figure 10–24 shows the transitions between the factory and application
configurations in remote update mode.
Figure 10–24. Transitions Between Configurations in Remote Update Mode
Configuration Error
Set Control Register
and Reconfigure
Power Up
Configuration
Error
Factory
Configuration
(page 0)
Application 1
Configuration
Reload a
Different Application
Reload a
Different Application
Set Control Register
and Reconfigure
Application n
Configuration
Configuration Error
After power up or a configuration error, the factory configuration logic is loaded
automatically. The factory configuration also must specify whether to enable the user
watchdog timer for the application configuration and if enabled, to include the timer
setting information.
The user watchdog timer ensures that the application configuration is valid and
functional. The timer must be continually reset within a specific amount of time
during user mode operation of an application configuration. Only valid application
configurations contain the logic to reset the timer in user mode. This timer reset logic
must be part of a user-designed hardware and/or software health monitoring signal
that indicates error-free system operation. If the timer is not reset in a specific amount
of time; for example, the user application configuration detects a functional problem
or if the system hangs, the dedicated circuitry updates the remote system upgrade
status register, triggering the loading of the factory configuration.
September 2012
Altera Corporation
Stratix IV Device Handbook
Volume 1
10–56
Chapter 10: Configuration, Design Security, and Remote System Upgrades in Stratix IV Devices
Remote System Upgrade Mode
1
The user watchdog timer is automatically disabled for factory configurations. For
more information about the user watchdog timer, refer to “User Watchdog Timer” on
page 10–61.
If there is an error while loading the application configuration, the cause of the
reconfiguration is written by the dedicated circuitry to the remote system upgrade
status register. Actions that cause the remote system upgrade status register to be
written are:
■
nSTATUS driven low externally
■
Internal CRC error
■
User watchdog timer time-out
■
A configuration reset (logic array nCONFIG signal or external nCONFIG pin assertion
to low)
Stratix IV devices automatically load the factory configuration located at page address
zero. This user-designed factory configuration can read the remote system upgrade
status register to determine the reason for the reconfiguration. The factory
configuration then takes appropriate error recovery steps and writes to the remote
system upgrade control register to determine the next application configuration to be
loaded.
When Stratix IV devices successfully load the application configuration, they enter
into user mode. In user mode, the soft logic (Nios II processor or state machine and
the remote communication interface) assists the Stratix IV device in determining
when a remote system update is arriving. When a remote system update arrives, the
soft logic receives the incoming data, writes it to the configuration memory device,
and triggers the device to load the factory configuration. The factory configuration
reads the remote system upgrade status register and control register, determines the
valid application configuration to load, writes the remote system upgrade control
register accordingly, and initiates system reconfiguration.
Stratix IV Device Handbook
Volume 1
September 2012 Altera Corporation
Chapter 10: Configuration, Design Security, and Remote System Upgrades in Stratix IV Devices
Dedicated Remote System Upgrade Circuitry
10–57
Dedicated Remote System Upgrade Circuitry
This section describes the implementation of the Stratix IV remote system upgrade
dedicated circuitry. The remote system upgrade circuitry is implemented in hard
logic. This dedicated circuitry interfaces to the user-defined factory and application
configurations implemented in the Stratix IV device logic array to provide the
complete remote configuration solution. The remote system upgrade circuitry
contains the remote system upgrade registers, a watchdog timer, and a state machine
that controls those components.
Figure 10–25 shows the data path for the remote system upgrade block.
Figure 10–25. Remote System Upgrade Circuit Data Path (Note 1)
Internal Oscillator
Status Register (SR)
[4..0]
Control Register
[37..0]
Logic Array
Update Register
[37..0]
update
Shift Register
dout
Bit [4..0]
din
dout
capture
RSU
State
Machine
din
Bit [37..0]
capture
time-out
User
Watchdog
Timer
clkout capture update
Logic Array
clkin
RU_DOUT
RU_SHIFTnLD
RU_CAPTnUPDT
RU_CLK
RU_DIN
RU_nCONFIG
RU_nRSTIMER
Logic Array
Note to Figure 10–25:
(1) The RU_DOUT, RU_SHIFTnLD, RU_CAPTnUPDT, RU_CLK, RU_DIN, RU_nCONFIG, and RU_nRSTIMER signals are internally
controlled by the ALTREMOTE_UPDATE megafunction.
September 2012
Altera Corporation
Stratix IV Device Handbook
Volume 1
10–58
Chapter 10: Configuration, Design Security, and Remote System Upgrades in Stratix IV Devices
Dedicated Remote System Upgrade Circuitry
Remote System Upgrade Registers
The remote system upgrade block contains a series of registers that store the page
addresses, watchdog timer settings, and status information. Table 10–14 lists these
registers.
Table 10–14. Remote System Upgrade Registers
Register
Description
Shift register
This register is accessible by the logic array and allows the update, status, and control registers to be
written and sampled by user logic.
Control register
This register contains the current page address, user watchdog timer settings, and one bit specifying
whether the current configuration is a factory configuration or an application configuration. During a read
operation in an application configuration, this register is read into the shift register. When a
reconfiguration cycle is initiated, the contents of the update register are written into the control register.
Update register
This register contains data similar to that in the control register. However, it can only be updated by the
factory configuration by shifting data into the shift register and issuing an update operation. When a
reconfiguration cycle is triggered by the factory configuration, the control register is updated with the
contents of the update register. During a capture in a factory configuration, this register is read into the
shift register.
Status register
This register is written to by the remote system upgrade circuitry on every reconfiguration to record the
cause of the reconfiguration. This information is used by the factory configuration to determine the
appropriate action following a reconfiguration. During a capture cycle, this register is read into the shift
register.
The remote system upgrade control and status registers are clocked by the 10-MHz
internal oscillator (the same oscillator that controls the user watchdog timer).
However, the remote system upgrade shift and update registers are clocked by the
user clock input (RU_CLK).
Remote System Upgrade Control Register
The remote system upgrade control register stores the application configuration page
address and user watchdog timer settings. The control register functionality depends
on the remote system upgrade mode selection. In remote update mode, the control
register page address bits are set to all zeros (24'b0 = 0×000000) at power up to load
the factory configuration. A factory configuration in remote update mode has write
access to this register.
Figure 10–26 and Table 10–15 specify the control register bit positions. In the figure,
the numbers show the bit position of a setting within a register. For example, bit
number 25 is the enable bit for the watchdog timer.
Figure 10–26. Remote System Upgrade Control Register
37 36 35 34 33 32 31 30 29 28 27 26
Wd_timer[11..0]
Stratix IV Device Handbook
Volume 1
25
Wd_en
24 23 22 .. 3
PGM[23..0]
2
1
0
AnF
September 2012 Altera Corporation
Chapter 10: Configuration, Design Security, and Remote System Upgrades in Stratix IV Devices
Dedicated Remote System Upgrade Circuitry
10–59
The application-not-factory (AnF) bit indicates whether the current configuration
loaded in the Stratix IV device is the factory configuration or an application
configuration. This bit is set low by the remote system upgrade circuitry when an
error condition causes a fall-back to the factory configuration. When the AnF bit is
high, the control register access is limited to read operations. When the AnF bit is low,
the register allows write operations and disables the watchdog timer.
In remote update mode, the factory configuration design sets this bit high (1'b1) when
updating the contents of the update register with the application page address and
watchdog timer settings.
Table 10–15 lists the remote system upgrade control register contents.
Table 10–15. Remote System Upgrade Control Register Contents
Remote System
Upgrade Mode
Value (2)
AnF (1)
Remote update
1'b0
PGM[23..0]
Remote update
24'b0×000000
AS configuration start address
(StAdd[23..0])
Wd_en
Remote update
1'b0
User watchdog timer enable bit
Control Register Bit
Definition
Application not factory
User watchdog time-out value
Remote update
Wd_timer[11..0]
12'b000000000000
(most significant 12 bits of 29-bit
count value: {Wd_timer[11..0],
17'b0})
Notes to Table 10–15:
(1) In remote update mode, the remote configuration block does not update the AnF bit automatically (you can update it manually).
(2) This is the default value of the control register bit.
Remote System Upgrade Status Register
The remote system upgrade status register specifies the reconfiguration trigger
condition. The various trigger and error conditions include:
September 2012
■
Cyclic redundancy check (CRC) error during application configuration
■
nSTATUS assertion by an external device due to an error
■
Stratix IV device logic array triggered a reconfiguration cycle, possibly after
downloading a new application configuration image
■
External configuration reset (nCONFIG) assertion
■
User watchdog timer time-out
Altera Corporation
Stratix IV Device Handbook
Volume 1
10–60
Chapter 10: Configuration, Design Security, and Remote System Upgrades in Stratix IV Devices
Dedicated Remote System Upgrade Circuitry
Figure 10–27 and Table 10–16 specify the contents of the status register. The numbers
in the figure show the bit positions within a 5-bit register.
Figure 10–27. Remote System Upgrade Status Register
4
Wd
3
2
1
nCONFIG Core_nCONFIG nSTATUS
0
CRC
Table 10–16. Remote System Upgrade Status Register Contents
Status Register Bit
Definition
POR Reset Value
CRC (from the configuration)
CRC error caused reconfiguration
1 bit '0'
nSTATUS
nSTATUS caused reconfiguration
1 bit '0'
Device logic array caused reconfiguration
1 bit '0'
nCONFIG caused reconfiguration
1 bit '0'
Watchdog timer caused reconfiguration
1 bit '0'
CORE_nCONFIG (1)
nCONFIG
Wd
Note to Table 10–16:
(1) Logic array reconfiguration forces the system to load the application configuration data into the Stratix IV device. This occurs after the factory
configuration specifies the appropriate application configuration page address by updating the update register.
Remote System Upgrade State Machine
The remote system upgrade control and update registers have identical bit
definitions, but serve different roles (refer to Table 10–14 on page 10–57). While both
registers can only be updated when the device is loaded with a factory configuration
image, the update register writes are controlled by the user logic; the control register
writes are controlled by the remote system upgrade state machine.
In factory configurations, the user logic sends the AnF bit (set high), the page address,
and the watchdog timer settings for the next application configuration bit to the
update register. When the logic array configuration reset (RU_nCONFIG) goes low, the
remote system upgrade state machine updates the control register with the contents
of the update register and initiates system reconfiguration from the new application
page.
1
To ensure successful reconfiguration between the pages, assert the RU_nCONFIG signal
for a minimum of 250 ns. This is equivalent to strobing the reconfiguration input of
the ALTREMOTE_UPDATE megafunction high for a minimum of 250 ns.
In the event of an error or reconfiguration trigger condition, the remote system
upgrade state machine directs the system to load a factory or application
configuration (page zero or page one, based on the mode and error condition) by
setting the control register accordingly. Table 10–17 lists the contents of the control
register after such an event occurs for all possible error or trigger conditions.
Stratix IV Device Handbook
Volume 1
September 2012 Altera Corporation
Chapter 10: Configuration, Design Security, and Remote System Upgrades in Stratix IV Devices
Dedicated Remote System Upgrade Circuitry
10–61
The remote system upgrade status register is updated by the dedicated error
monitoring circuitry after an error condition but before the factory configuration is
loaded.
Table 10–17. Control Register Contents after an Error or Reconfiguration Trigger Condition
Reconfiguration Error/Trigger
Control Register Setting
Remote Update
nCONFIG reset
All bits are 0
nSTATUS error
All bits are 0
CORE triggered reconfiguration
Update register
CRC error
All bits are 0
Wd time out
All bits are 0
Capture operations during factory configuration access the contents of the update
register. This feature is used by the user logic to verify that the page address and
watchdog timer settings were written correctly. Read operations in application
configurations access the contents of the control register. This information is used by
the user logic in the application configuration.
September 2012
Altera Corporation
Stratix IV Device Handbook
Volume 1
10–62
Chapter 10: Configuration, Design Security, and Remote System Upgrades in Stratix IV Devices
Dedicated Remote System Upgrade Circuitry
User Watchdog Timer
The user watchdog timer prevents a faulty application configuration from stalling the
device indefinitely. The system uses the timer to detect functional errors after an
application configuration is successfully loaded into the Stratix IV device.
The user watchdog timer is a counter that counts down from the initial value loaded
into the remote system upgrade control register by the factory configuration. The
counter is 29 bits wide and has a maximum count value of 229. When specifying the
user watchdog timer value, specify only the most significant 12 bits. The granularity
of the timer setting is 217 cycles. The cycle time is based on the frequency of the
10-MHz internal oscillator. Table 10–18 lists the operating range of the 10-MHz
internal oscillator.
Table 10–18. 10-MHz Internal Oscillator Specifications (Note 1)
Minimum
Typical
Maximum
Units
4.3
5.3
10
MHz
Note to Table 10–18:
(1) These values are preliminary.
The user watchdog timer begins counting after the application configuration enters
device user mode. This timer must be periodically reloaded or reset by the application
configuration before the timer expires by asserting RU_nRSTIMER. If the application
configuration does not reload the user watchdog timer before the count expires, a
time-out signal is generated by the remote system upgrade dedicated circuitry. The
time-out signal tells the remote system upgrade circuitry to set the user watchdog
timer status bit (Wd) in the remote system upgrade status register and reconfigures the
device by loading the factory configuration.
1
To allow remote system upgrade dedicated circuitry to reset the watchdog timer, you
must assert the RU_nRSTIMER signal active for a minimum of 250 ns. This is equivalent
to strobing the reset_timer input of the ALTREMOTE_UPDATE megafunction high
for a minimum of 250 ns.
The user watchdog timer is not enabled during the configuration cycle of the device.
Errors during configuration are detected by the CRC engine. Also, the timer is
disabled for factory configurations. Functional errors should not exist in the factory
configuration because it is stored and validated during production and is never
updated remotely.
1
Stratix IV Device Handbook
Volume 1
The user watchdog timer is disabled in factory configurations and during the
configuration cycle of the application configuration. It is enabled after the application
configuration enters user mode.
September 2012 Altera Corporation
Chapter 10: Configuration, Design Security, and Remote System Upgrades in Stratix IV Devices
Quartus II Software Support
10–63
Quartus II Software Support
The Quartus II software provides the flexibility to include the remote system upgrade
interface between the Stratix IV device logic array and the dedicated circuitry,
generate configuration files for production, and allows remote programming of the
system configuration memory.
The ALTREMOTE_UPDATE megafunction is the implementation option in the
Quartus II software that you use for the interface between the remote system upgrade
circuitry and the device logic array interface. Using the megafunction block instead of
creating your own logic saves design time and offers more efficient logic synthesis
and device implementation.
ALTREMOTE_UPDATE Megafunction
The ALTREMOTE_UPDATE megafunction provides a memory-like interface to the
remote system upgrade circuitry and handles the shift register read and write
protocol in the Stratix IV device logic. This implementation is suitable for designs that
implement the factory configuration functions using a Nios II processor or user logic
in the device.
Figure 10–28 shows the interface signals between the ALTREMOTE_UPDATE
megafunction and Nios II processor or user logic.
Figure 10–28. Interface Signals between the ALTREMOTE_UPDATE Megafunction and the Nios II Processor
ALTREMOTE_UPDATE
read_param
write_param
param[2..0]
data_in[23..0]
Nios II Processor or
User Logic
reconfig
reset_timer
clock
reset
busy
data_out[23..0]
f For more information about the ALTREMOTE_UPDATE megafunction and the
description of ports shown in Figure 10–28, refer to the Remote Update Circuitry
(ALTREMOTE_UPDATE) Megafunction User Guide.
September 2012
Altera Corporation
Stratix IV Device Handbook
Volume 1
10–64
Chapter 10: Configuration, Design Security, and Remote System Upgrades in Stratix IV Devices
Design Security
Design Security
This section provides an overview of the design security feature and its
implementation on Stratix IV devices using the advanced encryption standard (AES).
It also covers the new security modes available in Stratix IV devices.
As Stratix IV devices continue play a role in larger and more critical designs in
competitive commercial and military environments, it is increasingly important to
protect the designs from copying, reverse engineering, and tampering.
Stratix IV devices address these concerns with both volatile and non-volatile security
feature support. Stratix IV devices have the ability to decrypt configuration bitstreams
using the AES algorithm, an industry-standard encryption algorithm that is FIPS-197
certified. Stratix IV devices have a design security feature that utilizes a 256-bit
security key.
Stratix IV devices store configuration data in SRAM configuration cells during device
operation. Because SRAM is volatile, the SRAM cells must be loaded with
configuration data each time the device powers up. It is possible to intercept
configuration data when it is being transmitted from the memory source (flash
memory or a configuration device) to the device. The intercepted configuration data
could then be used to configure another device.
When using the Stratix IV design security feature, the security key is stored in the
Stratix IV device. Depending on the security mode, you can configure the Stratix IV
device using a configuration file that is encrypted with the same key, or for board
testing, configured with a normal configuration file.
The design security feature is available when configuring Stratix IV devices using FPP
configuration mode with an external host (such as a MAX II device or
microprocessor), or when using fast AS or PS configuration schemes. The design
security feature is also available in remote update with fast AS configuration mode.
The design security feature is not available when you are configuring your Stratix IV
device using JTAG-based configuration. For more information, refer to “Supported
Configuration Schemes” on page 10–67.
1
Stratix IV Device Handbook
Volume 1
When using a serial configuration scheme such as PS or fast AS, configuration time is
the same whether or not you enable the design security feature. If the FPP scheme is
used with the design security or decompression feature, a ×4 DCLK is required. This
results in a slower configuration time when compared with the configuration time of
a Stratix IV device that has neither the design security nor the decompression feature
enabled.
September 2012 Altera Corporation
Chapter 10: Configuration, Design Security, and Remote System Upgrades in Stratix IV Devices
Design Security
10–65
Stratix IV Security Protection
Stratix IV device designs are protected from copying, reverse engineering, and
tampering using configuration bitstream encryption.
Security Against Copying
The security key is securely stored in the Stratix IV device and cannot be read out
through any interfaces. In addition, as configuration file read-back is not supported in
Stratix IV devices, the design information cannot be copied.
Security Against Reverse Engineering
Reverse engineering from an encrypted configuration file is very difficult and time
consuming because the Stratix IV configuration file formats are proprietary and the
file contains millions of bits which require specific decryption. Reverse engineering
the Stratix IV device is just as difficult because the device is manufactured on the most
advanced 40-nm process technology.
Security Against Tampering
The non-volatile keys are one-time programmable. After the Tamper Protection bit is
set in the key programming file generated by the Quartus II software, the Stratix IV
device can only be configured with configuration files encrypted with the same key.
AES Decryption Block
The main purpose of the AES decryption block is to decrypt the configuration
bitstream prior to entering data decompression or configuration.
Prior to receiving encrypted data, you must enter and store the 256-bit security key in
the device. You can choose between a non-volatile security key and a volatile security
key with battery backup.
The security key is scrambled prior to storing it in the key storage to make it more
difficult for anyone to retrieve the stored key using de-capsulation of the device.
Flexible Security Key Storage
Stratix IV devices support two types of security key programming—volatile and
non-volatile keys. Table 10–19 lists the differences between volatile keys and
non-volatile keys.
Table 10–19. Security Key Options (Part 1 of 2)
Options
September 2012
Volatile Key
Non-Volatile Key
Key programmability
Reprogrammable and erasable
One-time programmable
External battery
Required
Not required
Key programming method (1)
On-board
On and off board
Altera Corporation
Stratix IV Device Handbook
Volume 1
10–66
Chapter 10: Configuration, Design Security, and Remote System Upgrades in Stratix IV Devices
Design Security
Table 10–19. Security Key Options (Part 2 of 2)
Options
Design protection
Volatile Key
Secure against copying and
reverse engineering
Non-Volatile Key
Secure against copying and
reverse engineering. Tamper
resistant if tamper protection
bit is set.
Note to Table 10–19:
(1) Key programming is carried out using the JTAG interface.
You can program the non-volatile key to the Stratix IV device without an external
battery. Also, there are no additional requirements to any of the Stratix IV power
supply inputs.
VCCBAT is a dedicated power supply for volatile key storage and not shared with other
on-chip power supplies, such as VCCIO or VCC. VCCBAT continuously supplies power to
the volatile register regardless of the on-chip supply condition.
1
After power-up, you must wait 300 ms (PORSEL = 0) or 12 ms (PORSEL = 1) before
beginning key programming to ensure that VCCBAT is at full rail.
1
For more information about how to calculate the key retention time of the battery
used for volatile key storage, refer to the Stratix III, Stratix IV, Stratix V, HardCopy III
and HardCopy IV PowerPlay Early Power Estimator.
f For more information about battery specifications, refer to the DC and Switching
Characteristics for Stratix IV Devices chapter.
f For more information about the VCCBAT pin connection recommendations, refer to the
Stratix IV GX and Stratix IV E Device Family Pin Connection Guidelines.
Stratix IV Design Security Solution
Stratix IV devices are SRAM-based devices. To provide design security, Stratix IV
devices require a 256-bit security key for configuration bitstream encryption.
You can carry out secure configuration in the following steps, as shown in
Figure 10–29:
1. Program the security key into the Stratix IV device.
2. Program the user-defined 256-bit AES keys to the Stratix IV device through the
JTAG interface.
3. Encrypt the configuration file and store it in the external memory.
4. Encrypt the configuration file with the same 256-bit keys used to program the
Stratix IV device. Encryption of the configuration file is done using the Quartus II
software. The encrypted configuration file is then loaded into the external
memory, such as a configuration or flash device.
5. Configure the Stratix IV device.
Stratix IV Device Handbook
Volume 1
September 2012 Altera Corporation
Chapter 10: Configuration, Design Security, and Remote System Upgrades in Stratix IV Devices
Design Security
10–67
At system power-up, the external memory device sends the encrypted configuration
file to the Stratix IV device.
Figure 10–29. Design Security (Note 1)
Stratix IV Device
User-Defined
Step 1
Key Storage
AES Key
AES
Decryption
Step 3
Encrypted
Step 2
Memory or
Configuration
Configuration
File
Device
Note to Figure 10–29:
(1) Step 1, Step 2, and Step 3 correspond to the procedure described in “Design Security” on page 10–63.
Security Modes Available
The following security modes are available on the Stratix IV device.
Volatile Key
Secure operation with volatile key programmed and required external battery: this
mode accepts both encrypted and unencrypted configuration bitstreams. Use the
unencrypted configuration bitstream support for board-level testing only.
Non-Volatile Key
Secure operation with one time programmable (OTP) security key programmed: this
mode accepts both encrypted and unencrypted configuration bitstreams. Use the
unencrypted configuration bitstream support for board level testing only.
Non-Volatile Key with Tamper Protection Bit Set
Secure operation in tamper resistant mode with OTP security key programmed: only
encrypted configuration bitstreams are allowed to configure the device. Tamper
protection disables JTAG configuration with unencrypted configuration bitstream.
September 2012
Altera Corporation
Stratix IV Device Handbook
Volume 1
10–68
Chapter 10: Configuration, Design Security, and Remote System Upgrades in Stratix IV Devices
Design Security
1
Enabling the tamper protection bit disables test mode in Stratix IV devices. This
process is irreversible and prevents Altera from conducting carry-out failure analysis
if test mode is disabled. Contact Altera Technical Support to enable the tamper
protection bit.
No Key Operation
Only unencrypted configuration bitstreams are allowed to configure the device.
Table 10–20 lists the different security modes and configuration bitstream supported
for each mode.
Table 10–20. Security Modes Supported
Mode (1)
Volatile key
Non-volatile key
Non-volatile key with tamper
protection bit set
Function
Configuration File
Secure
Encrypted
Board-level testing
Unencrypted
Secure
Encrypted
Board-level testing
Unencrypted
Secure (tamper resistant) (2)
Encrypted
Notes to Table 10–20:
(1) In No key operation, only the unencrypted configuration file is supported.
(2) The tamper protection bit setting does not prevent the device from being reconfigured.
Supported Configuration Schemes
The Stratix IV device supports only selected configuration schemes, depending on the
security mode you select when you encrypt the Stratix IV device.
Figure 10–30 shows the restrictions of each security mode when encrypting Stratix IV
devices.
Figure 10–30. Security Modes in Stratix IV Devices—Sequence and Restrictions
No Key
Volatile Key
Unencrypted or
Encrypted
Configuration File
Unencrypted
Configuration File
Non-Volatile Key
Unencrypted or
Encrypted
Configuration File
Non-Volatile Key
with
Tamper-Protection
Bit Set
Encrypted
Configuration File
Stratix IV Device Handbook
Volume 1
September 2012 Altera Corporation
Chapter 10: Configuration, Design Security, and Remote System Upgrades in Stratix IV Devices
Design Security
10–69
Table 10–21 lists the configuration modes allowed in each of the security modes.
Table 10–21. Allowed Configuration Modes for Various Security Modes
Configuration
File
Security Mode
No key
Unencrypted
Secure with volatile key
Board-level testing with
volatile key
Encrypted
Unencrypted
Secure with non-volatile key
Board-level testing with
non-volatile key
Encrypted
Unencrypted
Secure in tamper resistant
mode using non-volatile key
with tamper protection set
Encrypted
(Note 1)
Allowed Configuration Modes
All configuration modes that do not engage the design security feature.
■
Passive serial with AES (and/or with decompression)
■
Fast passive parallel with AES (and/or with decompression)
■
Remote update fast AS with AES (and/or with decompression)
■
Fast AS (and/or with decompression)
All configuration modes that do not engage the design security feature.
■
Passive serial with AES (and/or with decompression)
■
Fast passive parallel with AES (and/or with decompression)
■
Remote update fast AS with AES (and/or with decompression)
■
Fast AS (and/or with decompression)
All configuration modes that do not engage the design security feature.
■
Passive serial with AES (and/or with decompression)
■
Fast passive parallel with AES (and/or with decompression)
■
Remote update fast AS with AES (and/or with decompression)
■
Fast AS (and/or with decompression)
Note to Table 10–21:
(1) There is no impact to the configuration time required when compared with unencrypted configuration modes except FPP with AES (and/or
decompression), which requires a DCLK that is ×4 the data rate.
You can use the design security feature with other configuration features, such as
compression and remote system upgrade features. When you use compression with
the design security feature, the configuration file is first compressed and then
encrypted using the Quartus II software. During configuration, the Stratix IV device
first decrypts and then decompresses the configuration file.
Document Revision History
Table 10–22 lists the revision history for this chapter.
Table 10–22. Document Revision History (Part 1 of 2)
Date
Version
September 2012
December 2011
September 2012
3.5
3.4
Altera Corporation
Changes
■
Updated the “FPP Configuration Using a MAX II Device as an External Host” section to
close FB #36583 and #63157.
■
Updated the “Estimating Active Serial Configuration Time” section to close FB #64163.
■
Updated the “PS Configuration Using a MAX II Device as an External Host” section to
close FB #63157.
■
Updated Figure 10–1, Figure 10–2, Figure 10–3, Figure 10–10,Figure 10–11 and
Figure 10–12 to close FB #63155.
Updated Table 10–2 and Table 10–7.
Stratix IV Device Handbook
Volume 1
10–70
Chapter 10: Configuration, Design Security, and Remote System Upgrades in Stratix IV Devices
Design Security
Table 10–22. Document Revision History (Part 2 of 2)
Date
Version
April 2011
February 2011
March 2010
November 2009
June 2009
3.2
3.1
2.3
March 2009
2.1
Stratix IV Device Handbook
Volume 1
Updated the “FPP Configuration Using a MAX II Device as an External Host”, “Fast Active
Serial Configuration (Serial Configuration Devices)”, and “PS Configuration Using a
MAX II Device as an External Host”.
■
Updated Table 10–10.
■
Updated the “Fast Active Serial Configuration (Serial Configuration Devices)”, “FPP
Configuration Using a MAX II Device as an External Host” “Configuration Data
Decompression”, and “User Watchdog Timer” sections.
■
Updated Table 10–2, Table 10–4, Table 10–5, Table 10–7, and Table 10–9.
■
Applied new template.
■
Minor text edits.
■
Added the “Guidelines for Connecting Serial Configuration Devices on an AS Interface”
section.
■
Updated the “Power-On Reset Circuit” and “Fast Active Serial Configuration (Serial
Configuration Devices)” sections.
■
Updated Table 10–2, Table 10–4, Table 10–5, Table 10–10, and Table 10–13.
■
Updated Figure 10–16 and Figure 10–17 with Note 5.
■
Updated Figure 10–4, Figure 10–5, and Figure 10–13.
■
Updated the reference in the “Configuration Schemes” section.
■
Updated Table 10–1 and Table 10–2.
■
Updated the “FPP Configuration Using a MAX II Device as an External Host”,“Fast Active
Serial Configuration (Serial Configuration Devices)”, “Device Configuration Pins”,
“Remote System Upgrades”, “Remote System Upgrade Mode”, “Estimating Active Serial
Configuration Time”, “Remote System Upgrade State Machine”, and “User Watchdog
Timer” sections.
■
Removed Table 10-4, Table 10-7, Table 10-8, and Table 10-25.
■
Minor text edits.
■
Updated the “VCCPD Pins”, “FPP Configuration Using a MAX II Device as an External
Host”, “Estimating Active Serial Configuration Time”, “Fast Active Serial Configuration
(Serial Configuration Devices)”, “Remote System Upgrades”, “PS Configuration Using a
MAX II Device as an External Host”, and “PS Configuration Using a Download Cable”
sections.
■
Updated Table 10–3, Table 10–13 and Table 10–2.
■
Added introductory sentences to improve search ability.
■
Removed the Conclusion section.
■
Minor text edits.
■
Updated Table 10–2.
■
Updated Table 10–1, Table 10–2, and Table 10–9.
■
Removed “Referenced Documents” section.
■
Updated “Fast Active Serial Configuration (Serial Configuration Devices)” and “JTAG
Configuration” sections.
■
Updated Figure 10–4, Figure 10–5, Figure 10–6, and Figure 10–13.
■
Updated Table 10–2 and Table 10–13.
3.0
2.2
May 2008
■
3.3
April 2009
November 2008
Changes
2.0
1.0
Initial release.
September 2012 Altera Corporation
11. SEU Mitigation in Stratix IV Devices
February 2011
SIV51011-3.2
SIV51011-3.2
This chapter describes how to use the error detection cyclical redundancy check
(CRC) feature when a Stratix® IV device is in user mode and recovers from CRC
errors. The purpose of the error detection CRC feature in the Stratix IV device is to
detect a flip in any of the configuration random access memory (CRAM) bits in
Stratix IV devices due to a soft error. With the error detection circuitry, you can
continuously verify the integrity of the configuration CRAM bits.
In critical applications such as avionics, telecommunications, system control, and
military applications, it is important to be able to do the following:
1
■
Confirm that the configuration data stored in a Stratix IV device is correct
■
Alert the system to the occurrence of a configuration error
The error detection feature is enhanced in the Stratix IV device family. Similar to
Stratix III devices, the error detection and recovery time for single-event upset (SEU)
in Stratix IV devices is reduced when compared with Stratix II devices.
f For more information about test methodology for enhanced error detection in
Stratix IV devices, refer to AN 539: Test Methodology of Error Detection and Recovery
using CRC in Altera FPGA Devices.
Dedicated circuitry is built into Stratix IV devices and consists of a CRC error
detection feature that optionally checks for SEUs continuously and automatically.
1
For Stratix IV devices, the error detection CRC feature is provided in the Quartus® II
software version 8.0 and onwards.
Using error detection CRC for the Stratix IV device family has no impact on fitting or
performance of your device.
This chapter contains the following sections:
■
“Error Detection Fundamentals” on page 11–2
■
“Configuration Error Detection” on page 11–2
■
“User Mode Error Detection” on page 11–2
■
“Error Detection Pin Description” on page 11–5
■
“Error Detection Block” on page 11–6
■
“Error Detection Timing” on page 11–8
■
“Recovering From CRC Errors” on page 11–11
© 2011 Altera Corporation. All rights reserved. ALTERA, ARRIA, CYCLONE, HARDCOPY, MAX, MEGACORE, NIOS, QUARTUS and STRATIX words and logos
are trademarks of Altera Corporation and registered in the U.S. Patent and Trademark Office and in other countries. All other words and logos identified as
trademarks or service marks are the property of their respective holders as described at www.altera.com/common/legal.html. Altera warrants performance of its
semiconductor products to current specifications in accordance with Altera's standard warranty, but reserves the right to make changes to any products and
services at any time without notice. Altera assumes no responsibility or liability arising out of the application or use of any information, product, or service
described herein except as expressly agreed to in writing by Altera. Altera customers are advised to obtain the latest version of device specifications before relying
on any published information and before placing orders for products or services.
ISO
9001:2008
Registered
Stratix IV Device Handbook
Volume 1
February 2011
Feedback Subscribe
11–2
Chapter 11: SEU Mitigation in Stratix IV Devices
Error Detection Fundamentals
Error Detection Fundamentals
Error detection determines whether the data received is corrupted during
transmission. To accomplish this, the transmitter uses a function to calculate a
checksum value for the data and appends the checksum to the original data frame.
The receiver uses the same calculation methodology to generate a checksum for the
received data frame and compares the received checksum to the transmitted
checksum. If the two checksum values are equal, the received data frame is correct
and no data corruption occurred during transmission or storage.
The error detection CRC feature uses the same concept. When Stratix IV devices are
configured successfully and are in user mode, the error detection CRC feature ensures
the integrity of the configuration data.
1
There are two CRC error checks. One CRC error check always runs during
configuration and a second optional CRC error check runs in the background in user
mode. Both CRC error checks use the same CRC polynomial but different error
detection implementations. For more information, refer to the “Configuration Error
Detection” and “User Mode Error Detection” sections.
Configuration Error Detection
In configuration mode, a frame-based CRC is stored within the configuration data
and contains the CRC value for each data frame.
During configuration, the Stratix IV device calculates the CRC value based on the
frame of data that is received and compares it against the frame CRC value in the data
stream. Configuration continues until either the device detects an error or
configuration is completed.
In Stratix IV devices, the CRC value is calculated during the configuration stage. A
parallel CRC engine generates 16 CRC check bits per frame and then stores them in
CRAM. The CRAM chain used for storing the CRC check bits is 16 bits wide and its
length is equal to the number of frames in the device.
User Mode Error Detection
Stratix IV devices have built-in error detection circuitry to detect data corruption by
soft errors in the CRAM cells. This feature allows all CRAM contents to be read and
verified to match a configuration-computed CRC value. Soft errors are changes in a
CRAM bit state due to an ionizing particle.
The error detection capability continuously computes the CRC of the configured
CRAM bits and compares it with the pre-calculated CRC. If the CRCs match, there is
no error in the current configuration CRAM bits. The process of error detection
continues until the device is reset (by setting nCONFIG low).
If you enable the CRC error detection option in the Quartus II software, after the
device transitions into user mode, the error detection process is enabled. The internal
100 MHz configuration oscillator is divided down by a factor of two to 256 (at powers
of two) to be used as the clock source during the error detection process. You must set
the clock divide factor in the Quartus II software.
Stratix IV Device Handbook
Volume 1
February 2011 Altera Corporation
Chapter 11: SEU Mitigation in Stratix IV Devices
User Mode Error Detection
11–3
A single 16-bit CRC calculation is done on a per-frame basis. After it has finished the
CRC calculation for a frame, the resulting 16-bit signature is hex 0000 if there are no
CRAM bit errors detected in a frame by the error detection circuitry and the output
signal CRC_ERROR is 0. If a CRAM bit error is detected by the circuitry within a frame in
the device, the resulting signature is non-zero. This causes the CRC engine to start
searching for the error bit location.
Error detection in Stratix IV devices calculates CRC check bits for each frame and
pulls the CRC_ERROR pin high when it detects bit errors in the chip. Within a frame, it
can detect all single-bit, double-bit, and three-bit errors. The probability of more than
three CRAM bits being flipped by an SEU event is very low. In general, for all error
patterns the probability of detection is 99.998%.
The CRC engine reports the bit location and determines the type of error for all
single-bit errors and over 99.641% of double-adjacent errors. The probability of other
error patterns is very low and report of the location of bit flips is not guaranteed by
the CRC engine.
You can also read-out the error bit location through the JTAG and the core interface.
Shift these bits out through either the SHIFT_EDERROR_REG JTAG instruction or the core
interface before the CRC detects the next error in another frame. If the next frame also
has an error, you must shift these bits out within the amount of time of one frame CRC
verification. You can choose to extend this time interval by slowing down the error
detection clock frequency, but this slows down the error recovery time for the SEU
event. For the minimum update interval for Stratix IV devices, refer to Table 11–6 on
page 11–9. If these bits are not shifted out before the next error location is found, the
previous error location and error message is overwritten by the new information. The
CRC circuit continues to run, and if an error is detected, you must decide whether to
complete a reconfiguration or to ignore the CRC error.
The error detection logic continues to calculate the CRC_ERROR and 16-bit signatures for
the next frame of data regardless if any error has occurred in the current frame or not.
You need to monitor these signals and take the appropriate actions if a soft error
occurs.
The error detection circuitry in Stratix IV devices uses a 16-bit CRC-ANSI standard
(16-bit polynomial) as the CRC generator.
The computed 16-bit CRC signature for each frame is stored in the registers within the
core. The total storage register size is 16 (the number of bits per frame) × the number
of frames.
The Stratix IV device error detection feature does not check memory blocks and I/O
buffers. Thus, the CRC_ERROR signal might stay solid high or low depending on the
error status of the previously checked CRAM frame. The I/O buffers are not verified
during error detection because these bits use flipflops as storage elements that are
more resistant to soft errors when compared with CRAM cells. The support parity bits
of MLAB, M9K, and M144K are used to check the contents of the memory blocks for
any errors. The M144K TriMatrix memory block has a built-in error correction code
block that checks and corrects the errors in the block.
f For more information, refer to the TriMatrix Embedded Memory Blocks in Stratix IV
Devices chapter.
February 2011
Altera Corporation
Stratix IV Device Handbook
Volume 1
11–4
Chapter 11: SEU Mitigation in Stratix IV Devices
User Mode Error Detection
A JTAG instruction, EDERROR_INJECT, is provided to test the capability of the error
detection block. This instruction is able to change the content of the 21-bit JTAG fault
injection register that is used for error injection in Stratix IV devices, enabling the
testing of the error detection block.
1
You can only execute the EDERROR_INJECT JTAG instruction when the device is in user
mode.
Table 11–1 lists the description of the EDERROR_INJECT JTAG instruction.
Table 11–1. EDERROR_INJECT JTAG Instruction
JTAG Instruction
Instruction Code
Description
EDERROR_INJECT
00 0001 0101
This instruction controls the 21-bit JTAG fault
injection register, which is used for error
injection.
You can create a Jam™ file (.jam) to automate the testing and verification process.
This allows you to verify the CRC functionality in-system, on-the-fly, without having
to reconfigure the device. You can then switch to the CRC circuit to check for real
errors induced by an SEU.
You can introduce a single-error or double-errors adjacent to each other to the
configuration memory. This provides an extra way to facilitate design verification and
system fault tolerance characterization. Use the JTAG fault injection register with the
EDERROR_INJECT instruction to flip the readback bits. The Stratix IV device is then
forced into error test mode.
The content of the JTAG fault injection register is not loaded into the fault injection
register during the processing of the last and first frame. It is only loaded at the end of
this period.
1
You can only introduce error injection in the first data frame, but you can monitor the
error information at any time. For more information about the JTAG fault injection
register and fault injection register, refer to “Error Detection Registers” on page 11–7.
Table 11–2 lists how the fault injection register is implemented and describes error
injection.
Table 11–2. Fault Injection Register
Bit
Bit[20..19]
Bit[18..8]
Bit[7..0]
Description
Error Type
Byte Location of
the Injected Error
Error Byte Value
Depicts the location
of the injected error
in the first data
frame.
Depicts the location
of the bit error and
corresponds to the
error injection type
selection.
Error Type
(1)
Error injection type
Bit[20]
Bit[19]
0
1
Single-byte error injection
1
0
Double-adjacent byte error injection
0
0
No error injection
Content
Note to Table 11–2:
(1) Bit[20] and Bit[19] cannot both be set to 1 as this is not a valid selection. The error detection circuitry decodes this as no error injection.
Stratix IV Device Handbook
Volume 1
February 2011 Altera Corporation
Chapter 11: SEU Mitigation in Stratix IV Devices
Error Detection Pin Description
1
11–5
After the test completes, Altera recommends reconfiguring the device.
Automated Single-Event Upset Detection
Stratix IV devices offer on-chip circuitry for automated checking of SEU detection.
Some applications that require the device to operate error-free in high-neutron flux
environments require periodic checks to ensure continued data integrity. The error
detection CRC feature ensures data reliability and is one of the best options for
mitigating SEU.
You can implement the error detection CRC feature with existing circuitry in
Stratix IV devices, eliminating the need for external logic. The CRC_ERROR pin reports a
soft error when the configuration CRAM data is corrupted. You must decide whether
to reconfigure the device or to ignore the error.
Error Detection Pin Description
Depending on the type of error detection feature you choose, you must use different
error detection pins to monitor the data during user mode.
CRC_ERROR Pin
Table 11–3 describes the CRC_ERROR pin.
Table 11–3. CRC_ERROR Pin Description
Pin Name
CRC_ERROR
Pin Type
I/O and
open-drain
1
Description
Active-high signal indicates that the error detection circuit has detected errors in the
configuration CRAM bits. This pin is optional and is used when the error detection CRC
circuit is enabled. When the error detection CRC circuit is disabled, it is a user I/O pin.
To use the CRC_ERROR pin, you can either tie this pin to VCCPGM through a 10k resistor or,
depending on the input voltage specification of the system receiving the signal, you can tie
this pin to a different pull-up voltage.
The WYSIWYG function performs optimization on the Verilog Quartus Mapping
(VQM) netlist within the Quartus II software.
f For more information about the stratixiv_crcblock WYSIWYG function, refer to the
AN 539: Test Methodology of Error Detection and Recovery using CRC in Altera FPGA
Devices.
f For more information about the CRC_ERROR pin for Stratix IV devices, refer to Device
Pin-Outs on the Altera website.
February 2011
Altera Corporation
Stratix IV Device Handbook
Volume 1
11–6
Chapter 11: SEU Mitigation in Stratix IV Devices
Error Detection Block
Error Detection Block
You can enable the Stratix IV device error detection block in the Quartus II software
(refer to “Software Support” on page 11–10). This block contains the logic necessary to
calculate the 16-bit CRC signature for the configuration CRAM bits in the device.
The CRC circuit continues running even if an error occurs. When a soft error occurs,
the device sets the CRC_ERROR pin high. Two types of CRC detection checks the
configuration bits:
■
■
1
Stratix IV Device Handbook
Volume 1
CRAM error checking ability (16-bit CRC), which occurs during user mode to be
used by the CRC_ERROR pin.
■
For each frame of data, the pre-calculated 16-bit CRC enters the CRC circuit at
the end of the frame data and determines whether there is an error or not.
■
If an error occurs, the search engine starts to find the location of the error.
■
The error messages are shifted out through the JTAG instruction or core
interface logics while the error detection block continues running.
■
The JTAG interface reads out the 16-bit CRC result for the first frame and also
shifts the 16-bit CRC bits to the 16-bit CRC storage registers for test purposes.
■
Single error, double errors, or double-errors adjacent to each other are
deliberately introduced to configuration memory for testing and design
verification.
16-bit CRC that is embedded in every configuration data frame.
■
During configuration, after a frame of data is loaded into the Stratix IV device,
the pre-computed CRC is shifted into the CRC circuitry.
■
At the same time, the CRC value for the data frame shifted-in is calculated. If
the pre-computed CRC and calculated CRC values do not match, nSTATUS is set
low. Every data frame has a 16-bit CRC; therefore, there are many 16-bit CRC
values for the whole configuration bitstream. Every device has different
lengths of configuration data frame.
The “Error Detection Block” section describes the 16-bit CRC only when the device is
in user mode.
February 2011 Altera Corporation
Chapter 11: SEU Mitigation in Stratix IV Devices
Error Detection Block
11–7
Error Detection Registers
There is one set of 16-bit registers in the error detection circuitry that stores the
computed CRC signature. A non-zero value on the syndrome register causes the
CRC_ERROR pin to be set high.
Figure 11–1 shows the error detection circuitry, syndrome registers, and error injection
block.
Figure 11–1. Error Detection Block Diagram
16-Bit CRC
Calculation and Error
Readback bit
stream with
expected CRC
included
Syndrome
Search Engine
Error Detection
State Machine
8
Register
Control Signals
30
16
Error Message
CRC_ERROR
Register
46
Error Injection Block
Fault Injection
Register
JTAG Update
User Update
Register
Register
JTAG Shift
User Shift
Register
Register
JTAG Fault
Injection Register
JTAG TDO
General Routing
Table 11–4 lists the registers shown in Figure 11–1.
Table 11–4. Error Detection Registers (Part 1 of 2)
Register
Description
Syndrome Register
This register contains the CRC signature of the current frame through the error detection
verification cycle. The CRC_ERROR signal is derived from the contents of this register.
Error Message
Register
This 46-bit register contains information on the error type, location of the error, and the actual
syndrome. The types of errors and location reported are single- and double-adjacent bit errors.
The location bits for other types of errors are not identified by the error message register. The
content of the register can be shifted out through the SHIFT_EDERROR_REG JTAG instruction or to
the core through the core interface.
February 2011
Altera Corporation
Stratix IV Device Handbook
Volume 1
11–8
Chapter 11: SEU Mitigation in Stratix IV Devices
Error Detection Timing
Table 11–4. Error Detection Registers (Part 2 of 2)
Register
Description
JTAG Update Register
This register is automatically updated with the contents of the error message register one cycle
after the 46-bit register content is validated. It includes a clock enable that must be asserted prior
to being sampled into the JTAG shift register. This requirement ensures that the JTAG update
register is not being written into by the contents of the error message register at the same time
that the JTAG shift register is reading its contents.
User Update Register
This register is automatically updated with the contents of the Error Message Register, one cycle
after the 46-bit register content is validated. It includes a clock enable that must be asserted prior
to being sampled into the User Shift Register. This requirement ensures that the User Update
Register is not being written into by the contents of the Error Message Register at exactly the
same time that the User Shift Register is reading its contents.
JTAG Shift Register
This register is accessible by the JTAG interface and allows the contents of the JTAG Update
Register to be sampled and read by the JTAG instruction SHIFT_EDERROR_REG.
User Shift Register
This register is accessible by the core logic and allows the contents of the User Update Register to
be sampled and read by user logic.
JTAG Fault Injection
Register
This 21-bit register is fully controlled by the JTAG instruction EDERROR_INJECT. This register
holds the information of the error injection that you want in the bitstream.
Fault Injection Register
The content of the JTAG Fault Injection Register is loaded into this 21-bit register when it is being
updated.
Error Detection Timing
When you enable the CRC feature through the Quartus II software, the device
automatically activates the CRC process after entering user mode, after configuration,
and after initialization is complete.
If an error is detected within a frame, CRC_ERROR is driven high at the end of the error
location search, after the error message register is updated. At the end of this cycle,
the CRC_ERROR pin is pulled low for a minimum of 32 clock cycles. If the next frame
contains an error, CRC_ERROR is driven high again after the error message register is
overwritten by the new value. You can start to unload the error message on each
rising edge of the CRC_ERROR pin. Error detection runs until the device is reset.
The error detection circuitry runs off an internal configuration oscillator with a divisor
that sets the maximum frequency. Table 11–5 lists the minimum and maximum error
detection frequencies based on the best performance of the internal configuration
oscillator.
Table 11–5. Minimum and Maximum Error Detection Frequencies
Device Type
Error Detection
Frequency
Maximum Error
Detection Frequency
Minimum Error Detection
Frequency
Valid Divisors (n)
Stratix IV
100 MHz / 2n
50 MHz
390 kHz
1, 2, 3, 4, 5, 6, 7, 8
Stratix IV Device Handbook
Volume 1
February 2011 Altera Corporation
Chapter 11: SEU Mitigation in Stratix IV Devices
Error Detection Timing
11–9
You can set a lower clock frequency by specifying a division factor in the Quartus II
software (refer to “Software Support” on page 11–10). The divisor is a power of two,
in which n is between 1 and 8. The divisor ranges from 2 through 256. Refer to
Equation 11–1.
Equation 11–1.
100 MHz
error detection frequency = ----------------------n
2
1
The error detection frequency reflects the frequency of the error detection process for
a frame because the CRC calculation in the Stratix IV device is done on a per-frame
basis.
You must monitor the error message to avoid missing information in the error
message register. The error message register is updated whenever an error occurs. The
minimum interval time between each update for the error message register depends
on the device and the error detection clock frequency.
Table 11–6 lists the estimated minimum interval time between each update for the
error message register for Stratix IV devices.
Table 11–6. Minimum Update Interval for Error Message Register
(1)
Device
Timing Interval (s)
EP4SGX70
13.8
EP4SGX110
13.8
EP4SGX180
19.8
EP4SGX230
19.8
EP4SGX290
21.8
EP4SGX360
21.8
EP4SGX530
26.8
EP4SE230
19.8
EP4SE360
21.8
EP4SE530
26.8
EP4SE820
33.8
EP4S40G2
19.8
EP4S40G5
26.8
EP4S100G2
19.8
EP4S100G3
26.8
EP4S100G4
26.8
EP4S100G5
26.8
Note to Table 11–6:
(1) These timing numbers are preliminary.
CRC calculation time for the error detection circuitry to check from the first until the
last frame depends on the device and the error detection clock frequency.
February 2011
Altera Corporation
Stratix IV Device Handbook
Volume 1
11–10
Chapter 11: SEU Mitigation in Stratix IV Devices
Error Detection Timing
Table 11–7 lists the estimated time for each CRC calculation with minimum and
maximum clock frequencies for Stratix IV devices. The minimum CRC calculation
time is calculated by using the maximum error detection frequency with a divisor
factor of one, and the maximum CRC calculation time is calculated by using the
minimum error detection frequency with a divisor factor of eight.
Table 11–7. CRC Calculation Time
(1)
Device
Minimum Time (ms)
Maximum Time (s)
EP4SGX70
111
30.90
EP4SGX110
111
30.90
EP4SGX180
225
62.44
EP4SGX230
225
62.44
EP4SGX290
296
82.05
EP4SGX360
296
82.05
EP4SGX530
398
110.38
EP4SE230
225
62.44
EP4SE360
296
82.05
EP4SE530
398
110.38
EP4SE820
577
160.00
EP4S40G2
225
62.44
EP4S40G5
398
110.38
EP4S100G2
225
62.44
EP4S100G3
398
110.38
EP4S100G4
398
110.38
EP4S100G5
398
110.38
Note to Table 11–7:
(1) These timing numbers are preliminary.
Software Support
The Quartus II software version 8.0 and onwards supports the error detection CRC
feature for Stratix IV devices. Enabling this feature generates the CRC_ERROR output to
the optional dual purpose CRC_ERROR pin.
The error detection CRC feature is controlled by the Device and Pin Options dialog
box in the Quartus II software.
To enable the error detection feature using CRC, follow these steps:
1. Open the Quartus II software and load a project using a Stratix IV device.
2. On the Assignments menu, click Settings. The Settings dialog box is shown.
3. In the Category list, select Device. The Device page is shown.
4. Click Device and Pin Options. The Device and Pin Options dialog box is shown
(refer to Figure 11–2).
Stratix IV Device Handbook
Volume 1
February 2011 Altera Corporation
Chapter 11: SEU Mitigation in Stratix IV Devices
Recovering From CRC Errors
11–11
5. In the Device and Pin Options dialog box, click the Error Detection CRC tab.
6. Turn on Enable error detection CRC (Figure 11–2).
Figure 11–2. Enabling the Error Detection CRC Feature in the Quartus II Software
7. In the Divide error check frequency by pull-down list, enter a valid divisor as
listed in Table 11–5 on page 11–8.
1
The divide value divides the frequency of the configuration oscillator output clock
that clocks the CRC circuitry.
8. Click OK.
Recovering From CRC Errors
The system that the Stratix IV device resides in must control device reconfiguration.
After detecting an error on the CRC_ERROR pin, strobing the nCONFIG signal low directs
the system to perform the reconfiguration at a time when it is safe for the system to
reconfigure the device.
When the data bit is rewritten with the correct value by reconfiguring the device, the
device functions correctly.
While soft errors are uncommon in Altera devices, certain high-reliability applications
require a design to account for these errors.
February 2011
Altera Corporation
Stratix IV Device Handbook
Volume 1
11–12
Chapter 11: SEU Mitigation in Stratix IV Devices
Recovering From CRC Errors
Document Revision History
Table 11–8 lists the revision history for this chapter.
Table 11–8. Document Revision History
Date
Version
February 2011
March 2010
November 2009
3.2
3.1
3.0
June 2009
2.3
April 2009
2.2
March 2009
2.1
Changes
■
Applied new template.
■
Minor Text edits.
■
Updated Table 11–3 and Table 11–6.
■
Minor text edits.
■
Updated Table 11–3, Table 11–5, Table 11–6, and Table 11–7.
■
Updated the “CRC_ERROR Pin” section.
■
Minor text edits.
■
Added an introductory paragraph to increase search ability.
■
Removed the Conclusion section.
■
Minor text edits.
■
Updated Table 11–6 and Table 11–7.
■
Updated “Error Detection Timing” section.
■
Updated Table 11–6.
■
Added Table 11–7.
■
Removed “Critical Error Detection”, “Critical Error Pin”, and “Referenced Documents”
sections.
November 2008
2.0
Minor text edits.
May 2008
1.0
Initial release.
Stratix IV Device Handbook
Volume 1
February 2011 Altera Corporation
12. JTAG Boundary-Scan Testing in
Stratix IV Devices
February 2011
SIV51012-3.2
SIV51012-3.2
The IEEE Std. 1149.1 boundary-scan test (BST) circuitry available in Stratix® IV
devices provides a cost-effective and efficient way to test systems that contain devices
with tight lead spacing. Circuit boards with Altera and other IEEE Std.
1149.1-compliant devices can use EXTEST, SAMPLE/PRELOAD, and BYPASS modes to
create serial patterns that internally test the pin connections between devices and
check device operation.
This chapter describes how to use the IEEE Std. 1149.1 BST circuitry in Stratix IV
devices. The features are similar to Stratix III devices, unless stated otherwise in this
chapter.
This chapter contains the following sections:
■
“BST Architecture”
■
“BST Operation Control” on page 12–2
■
“I/O Voltage Support in a JTAG Chain” on page 12–4
■
“BST Circuitry” on page 12–4
■
“BSDL Support” on page 12–4
BST Architecture
A device operating in IEEE Std. 1149.1 BST mode uses four required pins, TDI, TDO,
TMS, TCK, and one optional pin, TRST. The TCK pin has an internal weak pull-down
resistor, while the TDI, TMS, and TRST pins have internal weak pull-up resistors. The
TDO output pin and all the JTAG input pins are powered by the 2.5-V/3.0-V VCCPD
supply of I/O bank 1A. All user I/O pins are tri-stated during JTAG configuration.
f For more information about the description and functionality of all JTAG pins,
registers used by the IEEE Std. 1149.1 BST circuitry, and the test access port (TAP)
controller, refer to the IEEE 1149.1 (JTAG) Boundary-Scan Testing in Stratix III Devices
chapter in volume 1 of the Stratix III Device Handbook.
© 2011 Altera Corporation. All rights reserved. ALTERA, ARRIA, CYCLONE, HARDCOPY, MAX, MEGACORE, NIOS, QUARTUS and STRATIX words and logos
are trademarks of Altera Corporation and registered in the U.S. Patent and Trademark Office and in other countries. All other words and logos identified as
trademarks or service marks are the property of their respective holders as described at www.altera.com/common/legal.html. Altera warrants performance of its
semiconductor products to current specifications in accordance with Altera's standard warranty, but reserves the right to make changes to any products and
services at any time without notice. Altera assumes no responsibility or liability arising out of the application or use of any information, product, or service
described herein except as expressly agreed to in writing by Altera. Altera customers are advised to obtain the latest version of device specifications before relying
on any published information and before placing orders for products or services.
ISO
9001:2008
Registered
Stratix IV Device Handbook
Volume 1
February 2011
Feedback Subscribe
12–2
Chapter 12: JTAG Boundary-Scan Testing in Stratix IV Devices
BST Operation Control
BST Operation Control
Table 12–1 lists the boundary-scan register length for Stratix IV devices.
Table 12–1. Boundary-Scan Register Length in Stratix IV Devices
Device
Boundary-Scan Register Length
EP4SGX70
1506
EP4SGX110
1506
EP4SGX180
2274
EP4SGX230
2274
EP4SGX290
(1)
2682
EP4SGX360
(1)
2682
EP4SGX530
2970
EP4SE230
2274
EP4SE360
2682
EP4SE530
2970
EP4SE820
3402
EP4S40G2
2274
EP4S40G5
2970
EP4S100G2
2274
EP4S100G3
2970
EP4S100G4
2970
EP4S100G5
2970
Note to Table 12–1:
(1) For the F1932 package of EP4SGX290 and EP4SGX360 devices, the boundary-scan register length is 2970.
Table 12–2 lists the IDCODE information for Stratix IV devices.
Table 12–2. IDCODE Information for Stratix IV Devices (Part 1 of 2)
IDCODE (32 Bits)
Device
(1)
Version (4 Bits)
Part Number (16 Bits)
Manufacturer Identity
(11 Bits)
LSB
(1 Bit) (2)
EP4SGX70
0000
0010 0100 0010 0000
000 0110 1110
1
EP4SGX110
0000
0010 0100 0000 0000
000 0110 1110
1
EP4SGX180
0000
0010 0100 0010 0001
000 0110 1110
1
EP4SGX230
0000
0010 0100 0000 1001
000 0110 1110
1
EP4SGX290
(3)
0000
0010 0100 0010 0010
000 0110 1110
1
EP4SGX290
(4)
0000
0010 0100 0100 0011
000 0110 1110
1
EP4SGX360
(3)
0000
0010 0100 0000 0010
000 0110 1110
1
EP4SGX360
(4)
0000
0010 0100 1000 0011
000 0110 1110
1
EP4SGX530
0000
0010 0100 0000 0011
000 0110 1110
1
EP4SE230
0000
0010 0100 0001 0001
000 0110 1110
1
EP4SE360
0000
0010 0100 0001 0010
000 0110 1110
1
Stratix IV Device Handbook
Volume 1
February 2011 Altera Corporation
Chapter 12: JTAG Boundary-Scan Testing in Stratix IV Devices
BST Operation Control
12–3
Table 12–2. IDCODE Information for Stratix IV Devices (Part 2 of 2)
IDCODE (32 Bits)
(1)
Device
Version (4 Bits)
Part Number (16 Bits)
Manufacturer Identity
(11 Bits)
LSB
(1 Bit) (2)
EP4SE530
0000
0010 0100 0001 0011
000 0110 1110
1
EP4SE820
0000
0010 0100 0000 0100
000 0110 1110
1
EP4S40G2
(5)
0000
0010 0100 0100 0001
000 0110 1110
1
EP4S40G5
(6)
0000
0010 0100 0010 0011
000 0110 1110
1
0000
0010 0100 0100 0001
000 0110 1110
1
0000
0010 0100 1010 0011
000 0110 1110
1
EP4S100G2
(5)
EP4S100G3
EP4S100G4
EP4S100G5
(6)
0000
0010 0100 0110 0011
000 0110 1110
1
0000
0010 0100 0010 0011
000 0110 1110
1
Notes to Table 12–2:
(1) The MSB is on the left.
(2) The LSB of the IDCODE is always 1.
(3) The IDCODE is applicable for all packages except F1932.
(4) The IDCODE is applicable for package F1932 only.
(5) For the ES1 device, the IDCODE is the same as the IDCODE of EP4SGX230.
(6) For the ES1 device, the IDCODE is the same as the IDCODE of EP4SGX530.
1
If the device is in reset state, when the nCONFIG or nSTATUS signal is low, the device
IDCODE might not be read correctly. To read the device IDCODE correctly, you must
issue the IDCODE JTAG instruction only when the nSTATUS signal is high.
f For more information about the following topics, refer to the IEEE 1149.1 (JTAG)
Boundary-Scan Testing in Stratix III Devices chapter in volume 1 of the Stratix III Device
Handbook:
February 2011
■
JTAG instruction codes with descriptions
■
TAP controller state-machine
■
Timing requirements for IEEE Std. 1149.1 signals
■
Instruction mode
■
Mandatory JTAG instructions (SAMPLE/PRELOAD, EXTEST, and BYPASS)
■
Optional JTAG instructions (IDCODE, USERCODE, CLAMP, and HIGHZ)
Altera Corporation
Stratix IV Device Handbook
Volume 1
12–4
Chapter 12: JTAG Boundary-Scan Testing in Stratix IV Devices
I/O Voltage Support in a JTAG Chain
I/O Voltage Support in a JTAG Chain
The JTAG chain supports several devices. However, you must use caution if the chain
contains devices that have different VCCIO levels.
f For more information, refer to the IEEE 1149.1 (JTAG) Boundary-Scan Testing in
Stratix III Devices chapter in volume 1 of the Stratix III Device Handbook.
BST Circuitry
The IEEE Std. 1149.1 BST circuitry is enabled after device power-up. You can perform
BST on Stratix IV devices before, during, and after configuration. Stratix IV devices
support BYPASS, IDCODE, and SAMPLE JTAG instructions during configuration without
interrupting configuration. To send all other JTAG instructions, you must interrupt
configuration using the CONFIG_IO JTAG instruction.
f For more information, refer to AN 39: IEEE Std. 1149.1 (JTAG) Boundary-Scan Testing in
Altera Devices.
f For more information about using the CONFIG_IO JTAG instruction for dynamic I/O
buffer configuration, considerations when performing BST for configured devices,
and JTAG pin connections to mask-out the BST circuitry, refer to the IEEE 1149.1
(JTAG) Boundary-Scan Testing in Stratix III Devices chapter in volume 1 of the Stratix III
Device Handbook.
f For more information about using the IEEE Std.1149.1 circuitry for device
configuration, refer to the Configuration, Design Security, Remote System Upgrades in
Stratix IV Devices chapter.
f If you must perform BST for configured devices, you must use the Quartus II software
version 8.1 and onwards to generate the design-specific boundary-scan description
language (BSDL) files. For the procedure to generate post-configured BSDL files using
the Quartus II software, refer to the BSDL Files Generation in Quartus II on the Altera
website.
BSDL Support
BSDL, a subset of VHDL, provides a syntax that allows you to describe the features of
an IEEE Std. 1149.1 BST-capable device that can be tested.
f For more information about BSDL files for IEEE Std. 1149.1-compliant Stratix IV
devices, refer to the Stratix IV BSDL Files on the Altera website.
f BSDL files for IEEE std. 1149.1-compliant Stratix IV devices can also be generated
using the Quartus II software version 8.1 and onwards. For more information about
the procedure to generate BSDL files using the Quartus II software, refer to the BSDL
Files Generation in Quartus II on the Altera website.
Stratix IV Device Handbook
Volume 1
February 2011 Altera Corporation
Chapter 12: JTAG Boundary-Scan Testing in Stratix IV Devices
BSDL Support
12–5
Document Revision History
Table 12–3 lists the revision history for this chapter.
Table 12–3. Document Revision History
Date
Version
February 2011
March 2010
3.2
3.1
November 2009
3.0
Changes
■
Applied new template.
■
Minor text edits.
■
Updated the hand note in the “BST Operation Control” section.
■
Changed “IDCODE JTAG Instruction” to read “IDCODE” as needed.
■
Minor text edits
■
Updated Table 12–1 and Table 12–2.
■
Minor text edits.
■
Added an introductory paragraph to increase search ability.
■
Removed the Conclusion section.
■
Minor text edits.
■
Updated Table 12–1.
■
Updated Table 12–1 and Table 12–2.
■
Removed “Referenced Documents” section.
June 2009
2.3
April 2009
2.2
March 2009
2.1
November 2008
2.0
Minor text edits.
April 2010
1.0
Initial release.
February 2011
Altera Corporation
Stratix IV Device Handbook
Volume 1
12–6
Stratix IV Device Handbook
Volume 1
Chapter 12: JTAG Boundary-Scan Testing in Stratix IV Devices
BSDL Support
February 2011 Altera Corporation
13. Power Management in Stratix IV
Devices
February 2011
SIV51013-3.2
SIV51013-3.2
This chapter describes power management in Stratix® IV devices. Stratix IV devices
offer programmable power technology options for low-power operation. You can use
these options, along with speed grade choices, in different permutations to give the
best power and performance combination. For thermal management, use the
Stratix IV internal temperature sensing device (TSD) with built-in analog-to-digital
converter (ADC) circuitry or external TSD with an external temperature sensor to
easily incorporate this feature in your designs. Being able to monitor the junction
temperature of the device at any time also offers the ability to control air flow to the
device and save power for the whole system.
Overview
Stratix IV FPGAs deliver a breakthrough level of system bandwidth and power
efficiency for high-end applications, allowing you to innovate without compromise.
Stratix IV devices use advanced power management techniques to enable both
density and performance increases while simultaneously reducing power dissipation.
The total power of an FPGA includes static and dynamic power.
■
Static power is the power consumed by the FPGA when it is configured but no
clocks are operating.
■
Dynamic power is the switching power when the device is configured and
running. You configure dynamic power with the equation shown in
Equation 13–1.
Equation 13–1. Dynamic Power Equation
(1)
2
1
P = --- CV frequency
2
Note to Equation 13–1:
(1) P = power; C = load capacitance; and V = supply voltage level.
Equation 13–1 shows that frequency is design dependant. However, you can vary the
voltage to lower dynamic power consumption by the square value of the voltage
difference. Stratix IV devices minimize static and dynamic power with advanced
process optimizations and programmable power technology. These technologies
enable Stratix IV designs to optimally meet design-specific performance requirements
with the lowest possible power.
The Quartus® II software optimizes all designs with Stratix IV power technology to
ensure performance is met at the lowest power consumption. This automatic process
allows you to concentrate on the functionality of the design instead of the power
consumption of the design.
© 2011 Altera Corporation. All rights reserved. ALTERA, ARRIA, CYCLONE, HARDCOPY, MAX, MEGACORE, NIOS, QUARTUS and STRATIX words and logos
are trademarks of Altera Corporation and registered in the U.S. Patent and Trademark Office and in other countries. All other words and logos identified as
trademarks or service marks are the property of their respective holders as described at www.altera.com/common/legal.html. Altera warrants performance of its
semiconductor products to current specifications in accordance with Altera's standard warranty, but reserves the right to make changes to any products and
services at any time without notice. Altera assumes no responsibility or liability arising out of the application or use of any information, product, or service
described herein except as expressly agreed to in writing by Altera. Altera customers are advised to obtain the latest version of device specifications before relying
on any published information and before placing orders for products or services.
ISO
9001:2008
Registered
Stratix IV Device Handbook
Volume 1
February 2011
Feedback Subscribe
13–2
Chapter 13: Power Management in Stratix IV Devices
Stratix IV Power Technology
Power consumption also affects thermal management. Stratix IV devices offer a TSD
feature that self-monitors the device junction temperature and can be used with
external circuitry for other activities, such as controlling air flow to the Stratix IV
FPGA.
This chapter contains the following sections:
■
“Stratix IV Power Technology”
■
“Stratix IV External Power Supply Requirements”
■
“Temperature Sensing Diode”
Stratix IV Power Technology
The following sections describe Stratix IV programmable power technology.
Programmable Power Technology
Stratix IV devices offer the ability to configure portions of the core, called tiles, for
high-speed or low-power mode of operation performed by the Quartus II software
without user intervention. Setting a tile to high-speed or low-power mode is
accomplished with on-chip circuitry and does not require extra power supplies
brought into the Stratix IV device. In a design compilation, the Quartus II software
determines whether a tile must be in high-speed or low-power mode based on the
timing constraints of the design.
f For more information about how the Quartus II software uses programmable power
technology when compiling a design, refer to AN 514: Power Optimization in Stratix IV
FPGAs.
A Stratix IV tile can consist of the following:
■
Memory logic array block (MLAB)/logic array block (LAB) pairs with routing to
the pair
■
MLAB/LAB pairs with routing to the pair and to adjacent digital signal processing
(DSP)/memory block routing
■
TriMatrix memory blocks
■
DSP blocks
All blocks and routing associated with the tile share the same setting of either
high-speed or low-power mode. By default, tiles that include DSP blocks or memory
blocks are set to high-speed mode for optimum performance. Unused DSP blocks and
memory blocks are set to low-power mode to minimize static power. Clock networks
do not support programmable power technology.
With programmable power technology, faster speed grade FPGAs may require less
power because there are fewer high-speed MLAB and LAB pairs, when compared
with slower speed grade FPGAs. The slower speed grade device may have to use
more high-speed MLAB and LAB pairs to meet performance requirements, while the
faster speed grade device can meet performance requirements with MLAB and LAB
pairs in low-power mode.
Stratix IV Device Handbook
Volume 1
February 2011 Altera Corporation
Chapter 13: Power Management in Stratix IV Devices
Stratix IV External Power Supply Requirements
13–3
The Quartus II software sets unused device resources in the design to low-power
mode to reduce static and dynamic power. It also sets the following resources to
low-power mode when they are not used in the design:
■
LABs and MLABs
■
TriMatrix memory blocks
■
DSP blocks
If a phase-locked loop (PLL) is instantiated in the design, asserting the areset pin
high keeps the PLL in low-power mode.
Table 13–1 lists the available Stratix IV programmable power capabilities. Speed grade
considerations can add to the permutations to give you flexibility in designing your
system.
Table 13–1. Programmable Power Capabilities in Stratix IV Devices
Feature
Programmable Power Technology
LAB
Yes
Routing
Yes
Memory Blocks
Fixed setting
(1)
DSP Blocks
Fixed setting
(1)
Global Clock Networks
No
Note to Table 13–1:
(1) Tiles with DSP blocks and memory blocks that are used in the design are always set to high-speed mode. By
default, unused DSP blocks and memory blocks are set to low-power mode.
Stratix IV External Power Supply Requirements
This section describes the different external power supplies required to power
Stratix IV devices. You can supply some of the power supply pins with the same
external power supply, provided they have the same voltage level.
f For power supply pin connection guidelines and power regulator sharing, refer to the
Stratix IV GX and Stratix IV E Device Family Pin Connection Guidelines.
f For each Altera recommended power supply’s operating conditions, refer to the DC
and Switching Characteristics for Stratix IV Devices chapter.
February 2011
Altera Corporation
Stratix IV Device Handbook
Volume 1
13–4
Chapter 13: Power Management in Stratix IV Devices
Temperature Sensing Diode
Temperature Sensing Diode
The Stratix IV TSD uses the characteristics of a PN junction diode to determine die
temperature. Knowing the junction temperature is crucial for thermal management.
Historically, junction temperature is calculated using ambient or case temperature,
junction-to-ambient (ja) or junction to-case (jc) thermal resistance, and device power
consumption. Stratix IV devices can either monitor its die temperature with the
internal TSD with built-in ADC circuitry or the external TSD with an external
temperature sensor. This allows you to control the air flow to the device.
You can use the Stratix IV internal TSD in two different modes of operation—
power-up mode and user mode. For power-up mode, the internal TSD reads the die’s
temperature during configuration if the ALTTEMP_SENSE megafunction is enabled
in your design. The ALTTEMP_SENSE megafunction allows temperature sensing
during device user mode by asserting the clken signal to the internal TSD circuitry. To
reduce device static power, disable the internal TSD with built-in ADC circuitry when
not in use.
f For more information about using the ALTTEMP_SENSE megafunction, refer to the
Thermal Sensor (ALTTEMP_SENSE) Megafunction User Guide.
The external temperature sensor steers bias current through the Stratix IV external
TSD, which measures forward voltage and converts this reading to temperature in the
form of an 8-bit signed number (7 bits plus sign). The 8-bit output represents the
junction temperature of the Stratix IV device and can be used for intelligent power
management.
External Pin Connections
The Stratix IV external TSD requires two pins for voltage reference. Figure 13–1 shows
how to connect the external TSD with an external temperature sensor device. As an
example, external temperature sensing devices, such as MAX1619, MAX1617A,
MAX6627, and ADT 7411, can be connected to the two external TSD pins for
temperature reading.
Figure 13–1. TSD External Pin Connections in Stratix IV Devices
External TSD
TEMPDIODEP
External
Temperature
Sensor
Stratix IV Device
TEMPDIODEN
f For more information about the external TSD specification, refer to the DC and
Switching Characteristics for Stratix IV Devices chapter.
Stratix IV Device Handbook
Volume 1
February 2011 Altera Corporation
Chapter 13: Power Management in Stratix IV Devices
Temperature Sensing Diode
13–5
The TSD is a very sensitive circuit that can be influenced by noise coupled from other
traces on the board and possibly within the device package itself, depending on your
device usage. The interfacing device registers’ temperature is based on millivolts
(mV) of difference, as seen at the external TSD pins. Switching the I/O near the TSD
pins can affect the temperature reading. Altera recommends taking temperature
readings during periods of inactivity in the device or use the internal TSD with
built-in ADC circuitry.
The following are board connection guidelines for the TSD external pin connections:
■
The maximum trace lengths for the TEMPDIODEP/TEMPDIODEN traces must be
less than eight inches.
■
Route both traces in parallel and place them close to each other with grounded
guard tracks on each side.
■
Altera recommends 10-mils width and space for both traces.
■
Route traces through a minimum number of vias and crossunders to minimize the
thermocouple effects.
■
Ensure that the number of vias are the same on both traces.
■
Ensure both traces are approximately the same length.
■
Avoid coupling with toggling signals (for example, clocks and I/O) by having the
GND plane between the diode traces and the high frequency signals.
■
For high-frequency noise filtering, place an external capacitor (close to the external
chip) between the TEMPDIODEP/TEMPDIODEN trace.
■
For Maxim devices, use an external capacitor between 2200 pF to 3300 pF.
■
Place a 0.1 uF bypass capacitor close to the external device.
■
You can use internal TSD with built-in ADC circuitry and external TSD at the same
time.
■
If you only use internal ADC circuitry, the external TSD pins
(TEMPDIODEP/TEMPDIODEN) can connect these pins to GND because the
external TSD pins are not used.
f For more information about the TEMPDIODEP/TEMPDIODEN pin
connection when you are not using an external TSD, refer to the
Stratix IV GX and Stratix IV E Pin Connection Guidelines.
f For device specification and connection guidelines, refer to the external temperature
sensor device data sheet from the device manufacturer.
February 2011
Altera Corporation
Stratix IV Device Handbook
Volume 1
13–6
Chapter 13: Power Management in Stratix IV Devices
Temperature Sensing Diode
Document Revision History
Table 13–2 lists the revision history for this chapter.
Table 13–2. Document Revision History
Date
Version
February 2011
March 2010
November 2009
June 2009
March 2009
3.2
3.1
3.0
2.2
2.1
Changes
■
Applied new template.
■
Minor text edits.
■
Updated the “External Pin Connections” section.
■
Minor text edits.
■
Updated the “Temperature Sensing Diode” and “External Pin Connections” sections.
■
Updated Equation 13–1.
■
Removed Table 13-2: Stratix IV External Power Supply Pins.
■
Minor text edits.
■
Updated the “External Pin Connections” section.
■
Added an introductory paragraph to increase search ability.
■
Removed the Conclusion section.
■
Updated “Temperature Sensing Diode” and “External Pin Connections” sections.
■
Updated Figure 13–1.
■
Removed “Referenced Documents” section.
November 2008
2.0
Minor text edits.
May 2008
1.0
Initial release.
Stratix IV Device Handbook
Volume 1
February 2011 Altera Corporation
Additional Information
About this Handbook
This chapter provides additional information about the document and Altera.
How to Contact Altera
To locate the most up-to-date information about Altera products, refer to the
following table.
Contact (1)
Technical support
Technical training
Product literature
Contact Method
Address
Website
www.altera.com/support
Website
www.altera.com/training
Email
[email protected]
Website
www.altera.com/literature
Nontechnical support (general)
Email
[email protected]
(software licensing)
Email
[email protected]
Note to Table:
(1) You can also contact your local Altera sales office or sales representative.
Typographic Conventions
The following table shows the typographic conventions this document uses.
Visual Cue
Meaning
Bold Type with Initial Capital
Letters
Indicate command names, dialog box titles, dialog box options, and other GUI
labels. For example, Save As dialog box. For GUI elements, capitalization matches
the GUI.
bold type
Indicates directory names, project names, disk drive names, file names, file name
extensions, software utility names, and GUI labels. For example, \qdesigns
directory, D: drive, and chiptrip.gdf file.
Italic Type with Initial Capital Letters
Indicate document titles. For example, Stratix IV Design Guidelines.
Indicates variables. For example, n + 1.
italic type
Variable names are enclosed in angle brackets (< >). For example, <file name> and
<project name>.pof file.
Initial Capital Letters
Indicate keyboard keys and menu names. For example, the Delete key and the
Options menu.
“Subheading Title”
Quotation marks indicate references to sections in a document and titles of
Quartus II Help topics. For example, “Typographic Conventions.”
September 2012
Altera Corporation
Stratix IV Device Handbook
Volume 1
Info–2
Additional Information
Typographic Conventions
Visual Cue
Meaning
Indicates signal, port, register, bit, block, and primitive names. For example, data1,
tdi, and input. The suffix n denotes an active-low signal. For example, resetn.
Courier type
Indicates command line commands and anything that must be typed exactly as it
appears. For example, c:\qdesigns\tutorial\chiptrip.gdf.
Also indicates sections of an actual file, such as a Report File, references to parts of
files (for example, the AHDL keyword SUBDESIGN), and logic function names (for
example, TRI).
r
An angled arrow instructs you to press the Enter key.
1., 2., 3., and
a., b., c., and so on
Numbered steps indicate a list of items when the sequence of the items is important,
such as the steps listed in a procedure.
■ ■
Bullets indicate a list of items when the sequence of the items is not important.
■
1
The hand points to information that requires special attention.
h
The question mark directs you to a software help system with related information.
f
The feet direct you to another document or website with related information.
m
The multimedia icon directs you to a related multimedia presentation.
c
A caution calls attention to a condition or possible situation that can damage or
destroy the product or your work.
w
A warning calls attention to a condition or possible situation that can cause you
injury.
The envelope links to the Email Subscription Management Center page of the Altera
website, where you can sign up to receive update notifications for Altera documents.
The feedback icon allows you to submit feedback to Altera about the document.
Methods for collecting feedback vary as appropriate for each document.
The social media icons allow you to inform others about Altera documents. Methods
for submitting information vary as appropriate for each medium.
Stratix IV Device Handbook
Volume 1
September 2012 Altera Corporation