ETC TMS320VC5421-200

TMS320VC5421 Fixed-Point
Digital Signal Processor
Data Manual
Literature Number: SPRS098C
December 1999 – Revised November 2001
PRODUCTION DATA information is current as of publication date.
Products conform to specifications per the terms of Texas Instruments
standard warranty. Production processing does not necessarily include
testing of all parameters.
Printed on Recycled Paper
IMPORTANT NOTICE
Texas Instruments Incorporated and its subsidiaries (TI) reserve the right to make corrections, modifications,
enhancements, improvements, and other changes to its products and services at any time and to discontinue
any product or service without notice. Customers should obtain the latest relevant information before placing
orders and should verify that such information is current and complete. All products are sold subject to TI’s terms
and conditions of sale supplied at the time of order acknowledgment.
TI warrants performance of its hardware products to the specifications applicable at the time of sale in
accordance with TI’s standard warranty. Testing and other quality control techniques are used to the extent TI
deems necessary to support this warranty. Except where mandated by government requirements, testing of all
parameters of each product is not necessarily performed.
TI assumes no liability for applications assistance or customer product design. Customers are responsible for
their products and applications using TI components. To minimize the risks associated with customer products
and applications, customers should provide adequate design and operating safeguards.
TI does not warrant or represent that any license, either express or implied, is granted under any TI patent right,
copyright, mask work right, or other TI intellectual property right relating to any combination, machine, or process
in which TI products or services are used. Information published by TI regarding third–party products or services
does not constitute a license from TI to use such products or services or a warranty or endorsement thereof.
Use of such information may require a license from a third party under the patents or other intellectual property
of the third party, or a license from TI under the patents or other intellectual property of TI.
Reproduction of information in TI data books or data sheets is permissible only if reproduction is without
alteration and is accompanied by all associated warranties, conditions, limitations, and notices. Reproduction
of this information with alteration is an unfair and deceptive business practice. TI is not responsible or liable for
such altered documentation.
Resale of TI products or services with statements different from or beyond the parameters stated by TI for that
product or service voids all express and any implied warranties for the associated TI product or service and
is an unfair and deceptive business practice. TI is not responsible or liable for any such statements.
Mailing Address:
Texas Instruments
Post Office Box 655303
Dallas, Texas 75265
Copyright  2001, Texas Instruments Incorporated
REVISION HISTORY
REVISION
DATE
PRODUCT STATUS
HIGHLIGHTS
*
April 1999
Product Preview
Original
A
January 2000
Product Preview
Revised HPI timing and switching characteristics data.
B
August 2000
Production Data
Converted to data manual format and revised to include
production characteristics data.
C
November 2001
Production Data
Removed all references to an industrial part temperature
range.
iii
Contents
Contents
Section
Page
1
TMS320VC5421 Features . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
1
2
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.1
Description . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.2
Migration From the 5420 to the 5421 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.3
Pin Assignments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.3.1
Pin Assignments for the PGE Package . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.3.2
Terminal Assignments for the GGU Package . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.4
Signal Descriptions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2
2
2
3
3
6
9
3
Functional Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
3.1
Memory . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
3.1.1
On-Chip Dual-Access RAM (DARAM) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
3.1.2
On-Chip Single-Access RAM (SARAM) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
3.1.3
On-Chip Two-Way Shared RAM (DARAM) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
3.1.4
On-Chip Boot ROM . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
3.1.5
Extended Program Memory . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
3.1.6
Program Memory . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
3.1.7
Data Memory . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
3.1.8
I/O Memory . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
3.2
Multicore Reset Signals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
3.3
Bootloader . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
3.4
External Interface (XIO) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
3.5
On-Chip Peripherals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
3.5.1
Software-Programmable Wait-State Generators . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
3.5.2
Programmable Bank-Switching . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
3.5.3
Parallel I/O Ports . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
3.6
16-Bit Bidirectional Host-Port Interface (HPI16) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
3.6.1
HPI16 Memory Map . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
3.6.2
HPI Features . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
3.6.3
HPI Multiplexed Mode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
3.6.4
Host/DSP Interrupts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
3.6.5
Emulation Considerations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
3.6.6
HPI Nonmultiplexed Mode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
3.6.7
Other HPI16 System Considerations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
3.7
Multichannel Buffered Serial Port (McBSP) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
3.7.1
Emulation Considerations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
3.8
Direct Memory Access (DMA) Controller . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
3.8.1
DMA Controller Features . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
3.8.2
DMA Accesses to External Memory . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
3.8.3
DMA Controller Synchronization Events . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
3.8.4
DMA Channel Interrupt Selection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
3.8.5
DMA in Autoinitialization Mode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
3.8.6
Subsystem Communications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
3.8.7
Chip Subsystem ID Register . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
17
18
18
19
19
19
19
20
20
20
20
20
21
21
22
23
24
25
25
26
26
26
26
26
27
28
31
32
34
34
36
36
37
37
38
December 1999 – Revised November 2001
SPRS098C
v
Contents
Section
3.9
Page
General-Purpose I/O . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
3.9.1
Hardware Timer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
3.9.2
Software-Programmable Phase-Locked Loop (PLL) . . . . . . . . . . . . . . . . . . . . . . . . . .
3.9.3
PLL Clock Programmable Timer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Memory-Mapped Registers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
McBSP Control Registers and Subaddresses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
DMA Subbank Addressed Registers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Interrupts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
IDLE3 Power-Down Mode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Emulating the 5421 Device . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
38
39
39
41
42
44
45
47
49
49
4
Documentation Support . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
50
5
Electrical Specifications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
5.1
Absolute Maximum Ratings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
5.2
Recommended Operating Conditions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
5.3
Electrical Characteristics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
5.4
Package Thermal Resistance Characteristics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
5.5
Timing Parameter Symbology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
5.6
Clock Options . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
5.6.1
Divide-By-Two, Divide-By-Four, and Bypass Clock Option (PLL Disabled) . . . . . . .
5.6.2
Multiply-By-N Clock Option (PLL Enabled) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
5.7
External Memory Interface Timing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
5.7.1
Memory Read . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
5.7.2
Memory Write . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
5.8
Ready Timing For Externally Generated Wait States . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
5.9
Parallel I/O Interface Timing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
5.9.1
Parallel I/O Port Read . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
5.9.2
Parallel I/O Port Write . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
5.10
Externally Generated Wait States . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
5.10.1
I/O Port Read and Write With Externally Generated Wait States . . . . . . . . . . . . . . .
5.11
Reset, BIO, Interrupt, and MP/MC Timings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
5.12
HOLD and HOLDA Timings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
5.13
External Flag (XF) and TOUT Timings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
5.14
General-Purpose I/O Timing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
5.15
Multichannel Buffered Serial Port (McBSP) Timing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
5.15.1
McBSP Transmit and Receive Timings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
5.15.2
McBSP Transmit and Receive Timing Using CLKR/X as a Clock Source Input to
the Sample Rate Generator (SRGR) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
5.15.3
McBSP General-Purpose I/O Timing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
5.15.4
McBSP as SPI Master or Slave Timing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
5.16
Host-Port Interface Timing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
51
51
51
52
53
53
54
54
55
56
56
58
59
61
61
62
63
63
65
67
69
70
71
71
Mechanical Data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
6.1
Ball Grid Array Mechanical Data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
6.2
Low Profile Quad Flatpack Mechanical Data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
88
88
89
3.10
3.11
3.12
3.13
3.14
3.15
6
vi
SPRS098C
74
76
77
81
December 1999 – Revised November 2001
Figures
List of Figures
Figure
Page
2–1
2–2
144-Pin Low-Profile Flatpack Pin Assignments (PGE – Top View) . . . . . . . . . . . . . . . . . . . . . . . . . . .
144-Ball MicroStar BGA Pin Assignments (GGU – Bottom View) . . . . . . . . . . . . . . . . . . . . . . . . . . . .
3
6
3–1
3–2
3–3
3–4
3–5
3–6
3–7
3–8
3–9
3–10
3–11
3–12
3–13
3–14
3–15
3–16
3–17
3–18
3–19
3–20
3–21
TMS320VC5421 Functional Block Diagram . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Memory Map Relative to CPU Subsystems A and B . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Software Wait-State Register (SWWSR) [Memory-Mapped Register (MMR) Address 0028h] . . .
Software Wait-State Control Register (SWCR) [MMR Address 002Bh] . . . . . . . . . . . . . . . . . . . . . . .
BSCR Register Bit Layout for Each DSP Subsystem . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Memory Map Relative to Host-Port Interface HPI16 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Interfacing to the HPI-16 in Non-Multiplexed Mode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Pin Control Register (PCR) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Multichannel Control Register 2x (MCR2x) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Multichannel Control Register 1x (MCR1x) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Receive Channel Enable Registers Bit Layout for Partitions A to H . . . . . . . . . . . . . . . . . . . . . . . . . .
Transmit Channel Enable Registers Bit Layout for Partitions A to H . . . . . . . . . . . . . . . . . . . . . . . . .
On-Chip Memory Map Relative to DMA (DLAXS/SLAXS = 0) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
DMA External Program Memory Map . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Arbitration Between XIO and xDMA for External Access . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
DMA Transfer Mode Control Register (DMMCRn) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
DMPREC Register . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Chip Subsystem ID Register . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
General-Purpose I/O Control Register . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Clock Mode Register (CLKMD) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Bit Layout of the IMR and IFR Registers for Subsystems A and B . . . . . . . . . . . . . . . . . . . . . . . . . . .
17
18
22
23
23
25
27
29
29
30
30
30
32
33
35
35
37
38
38
40
48
5–1
5–2
5–3
5–4
5–5
5–6
5–7
5–8
5–9
5–10
5–11
5–12
5–13
5–14
5–15
3.3-V Test Load Circuit . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
External Divide-by-Two Clock Timing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
External Multiply-by-One Clock Timing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Memory Read (MSTRB = 0) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Memory Write (MSTRB = 0) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Memory Read With Externally Generated Wait States . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Memory Write With Externally Generated Wait States . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Parallel I/O Port Read (IOSTRB=0) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Parallel I/O Port Write (IOSTRB=0) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
I/O Port Read With Externally Generated Wait States . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
I/O Port Write With Externally Generated Wait States . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Reset and BIO Timings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Interrupt Timing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
XIO Timing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
HOLD and HOLDA Timings (HM = 1) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
52
54
55
57
58
59
60
61
62
63
64
65
66
66
68
December 1999 – Revised November 2001
SPRS098C
vii
Figures
Figure
Page
5–16
5–17
5–18
5–19
5–20
5–21
5–22
5–23
5–24
5–25
5–26
5–27
5–28
5–29
5–30
5–31
5–32
5–33
5–34
External Flag (XF) Timing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Timer (TOUT) Timing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
GPIO Timings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
McBSP Receive Timings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
McBSP Transmit Timings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
McBSP Sample Rate Generator Timings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
McBSP General-Purpose I/O Timings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
McBSP Timing as SPI Master or Slave: CLKSTP = 10b, CLKXP = 0 . . . . . . . . . . . . . . . . . . . . . . . .
McBSP Timing as SPI Master or Slave: CLKSTP = 11b, CLKXP = 0 . . . . . . . . . . . . . . . . . . . . . . . .
McBSP Timing as SPI Master or Slave: CLKSTP = 10b, CLKXP = 1 . . . . . . . . . . . . . . . . . . . . . . . .
McBSP Timing as SPI Master or Slave: CLKSTP = 11b, CLKXP = 1 . . . . . . . . . . . . . . . . . . . . . . . .
Multiplexed Read Timings Using HAS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Multiplexed Read Timings With HAS Held High . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Multiplexed Write Timings Using HAS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Multiplexed Write Timings With HAS Held High . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Nonmultiplexed Read Timings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Nonmultiplexed Write Timings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
HRDY and HINT Relative to CLKOUT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
SELA/B Timing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
69
69
70
73
73
75
76
77
78
79
80
83
84
85
86
86
87
87
87
6–1
6–2
MicroStar BGA Package . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Low-Profile Quad Flatpack . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
88
89
viii
SPRS098C
December 1999 – Revised November 2001
Tables
List of Tables
Table
Page
2–1
2–2
2–3
Pin Assignments for the 144-Pin Low-Profile Quad Flatpack . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Terminal Assignments for the 144-Pin MicroStar BGA . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Signal Descriptions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4
7
9
3–1
3–2
3–3
3–4
3–5
3–6
3–7
3–8
3–9
3–10
3–11
3–12
3–13
3–14
3–15
3–16
3–17
3–18
3–19
3–20
3–21
3–22
3–23
3–24
XIO/ROMEN Modes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Bootloader Operating Modes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
XIO/HPI Modes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Software Wait-State Register (SWWSR) Bit Fields . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Software Wait-State Control Register (SWCR) Bit Fields . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
BSCR Register Bit Functions for Each DSP Subsystem . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Sample Rate Generator Clock Source Selection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Receive Channel Enable Registers for Partitions A to H . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Transmit Channel Enable Registers for Partitions A to H . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
DMA Synchronization Events . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
DMA Channel Interrupt Selection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
DMA Global Reload Register Selection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Chip Subsystem ID Register Bit Functions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
General-Purpose I/O Control Register Bit Functions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Clock Mode Register (CLKMD) Bit Functions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Multiplier Related to PLLNDIV, PLLDIV, and PLLMUL . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
VCO Truth Table . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
VCO Lockup Time . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Processor Memory-Mapped Registers for Each DSP Subsystem . . . . . . . . . . . . . . . . . . . . . . . . . .
Peripheral Memory-Mapped Registers for Each DSP Subsystem . . . . . . . . . . . . . . . . . . . . . . . . . .
McBSP Control Registers and Subaddresses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
DMA Subbank Addressed Registers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
5421 Interrupt Locations and Priorities for Each DSP Subsystem . . . . . . . . . . . . . . . . . . . . . . . . . . .
Bit Functions for IMR and IFR Registers for Each DSP Subsystem . . . . . . . . . . . . . . . . . . . . . . . .
19
20
21
22
23
24
29
30
31
36
36
37
38
39
40
41
41
41
42
43
44
45
47
48
5–1
5–2
5–3
5–4
5–5
5–6
5–7
5–8
5–9
5–10
5–11
5–12
5–13
5–14
5–15
5–16
Recommended Operating Conditions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Electrical Characteristics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Thermal Resistance Characteristics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Divide-By-2 and Divide-by-4 Clock Options Timing Requirements . . . . . . . . . . . . . . . . . . . . . . . . . .
Divide-By-2 and Divide-by-4 Clock Options Switching Characteristics . . . . . . . . . . . . . . . . . . . . . . .
Multiply-By-N Clock Option Timing Requirements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Multiply-By-N Clock Option Switching Characteristics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Memory Read Timing Requirements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Memory Read Switching Characteristics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Memory Write Switching Characteristics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Ready Timing Requirements for Externally Generated Wait States . . . . . . . . . . . . . . . . . . . . . . . . .
Parallel I/O Port Read Timing Requirements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Parallel I/O Port Read Switching Characteristics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Parallel I/O Port Write Switching Characteristics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Externally Generated Wait States Timing Requirements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Reset, BIO, Interrupt, and MP/MC Timing Requirements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
51
52
53
54
54
55
55
56
56
58
59
61
61
62
63
65
December 1999 – Revised November 2001
SPRS098C
ix
Tables
Table
5–17
5–18
5–19
5–20
5–21
5–22
5–23
5–24
5–25
5–26
5–27
5–28
5–29
5–30
5–31
5–32
5–33
5–34
5–35
5–36
5–37
x
Page
HOLD and HOLDA Timing Requirements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
HOLD and HOLDA Switching Characteristics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
External Flag (XF) and TOUT Switching Characteristics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
General-Purpose I/O Timing Requirements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
General-Purpose I/O Switching Characteristics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
McBSP Transmit and Receive Timing Requirements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
McBSP Transmit and Receive Switching Characteristics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
McBSP Sample Rate Generator Timing Requirements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
McBSP Sample Rate Generator Switching Characteristics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
McBSP General-Purpose I/O Timing Requirements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
McBSP General-Purpose I/O Switching Characteristics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
McBSP as SPI Master or Slave Timing Requirements (CLKSTP = 10b, CLKXP = 0) . . . . . . . . . .
McBSP as SPI Master or Slave Switching Characteristics (CLKSTP = 10b, CLKXP = 0) . . . . . .
McBSP as SPI Master or Slave Timing Requirements (CLKSTP = 11b, CLKXP = 0) . . . . . . . . . .
McBSP as SPI Master or Slave Switching Characteristics (CLKSTP = 11b, CLKXP = 0) . . . . . . .
McBSP as SPI Master or Slave Timing Requirements (CLKSTP = 10b, CLKXP = 1) . . . . . . . . . .
McBSP as SPI Master or Slave Switching Characteristics (CLKSTP = 10b, CLKXP = 1) . . . . . .
McBSP as SPI Master or Slave Timing Requirements (CLKSTP = 11b, CLKXP = 1) . . . . . . . . . .
McBSP as SPI Master or Slave Switching Characteristics (CLKSTP = 11b, CLKXP = 1) . . . . . . .
HPI16 Mode Timing Requirements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
HPI16 Mode Switching Characteristics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
SPRS098C
67
67
69
70
70
71
72
74
74
76
76
77
77
78
78
79
79
80
80
81
82
December 1999 – Revised November 2001
Features
1
TMS320VC5421 Features
200-MIPS Dual-Core DSP Consisting of Two
Independent Subsystems
Each Core Has an Advanced Multibus
Architecture With Three Separate 16-Bit
Data Memory Buses and One Program Bus
40-Bit Arithmetic Logic Unit (ALU)
Including a 40-Bit Barrel-Shifter and Two
40-Bit Accumulators Per Core
Each Core Has a 17-Bit × 17-Bit Parallel
Multiplier Coupled to a 40-Bit Adder for
Non-Pipelined Single-Cycle Multiply/
Accumulate (MAC) Operations
Each Core Has a Compare, Select, and
Store Unit (CSSU) for the Add/Compare
Selection of the Viterbi Operator
Each Core Has an Exponent Encoder to
Compute an Exponent Value of a 40-Bit
Accumulator Value in a Single Cycle
Each Core Has Two Address Generators
With Eight Auxiliary Registers and Two
Auxiliary Register Arithmetic Units
(ARAUs)
16-Bit Data Bus With Data Bus Holder
Feature
512K-Word × 16-Bit Extended Program
Address Space
Total of 256K-Word × 16-Bit Dual- and
Single-Access On-Chip RAM (128K-Word x
16-Bit Two-Way Shared Memory)
Single-Instruction Repeat and
Block-Repeat Operations
Instructions With 32-Bit-Long Word
Operands
Instructions With Two or Three Operand
Reads
Fast Return From Interrupts
Arithmetic Instructions With Parallel Store
and Parallel Load
Conditional Store Instructions
Output Control of CLKOUT
Output Control of TOUT
Power Consumption Control With IDLE1,
IDLE2, and IDLE3 Instructions
Dual 1.8-V (Core) and 3.3-V (I/O) Power
Supplies for Low-Power, Fast Operations
10-ns Single-Cycle Fixed-Point Instruction
Interprocessor Communication via Two
Internal 8-Element FIFOs
Twelve Channels of Direct Memory Access
(DMA) for Data Transfers With No CPU
Loading (Six Channels Per Subsystem With
External Access)
Six Multichannel Buffered Serial Ports
(McBSPs) With 128-Channel Selection
Capability (Three McBSPs per Subsystem)
16-Bit Host-Port Interface (HPI) Multiplexed
With External Memory Interface Pins
Software-Programmable Phase-Locked
Loop (APLL) Provides Several Clocking
Options (Requires External Oscillator)
On-Chip Scan-Based Emulation Logic,
IEEE Standard 1149-1 (JTAG) BoundaryScan Logic
Two Software-Programmable Timers
(One Per Subsystem)
Software-Programmable Wait-State
Generator (14 Wait States Maximum)
Provided in 144-pin MicroStar BGA Ball
Grid Array (GGU Suffix) and 144-pin
Low-Profile Quad Flatpack (LQFP) (PGE
Suffix) Packages
MicroStar BGA is a trademark of Texas Instruments.
‡ IEEE Standard 1149.1-1990 Standard-Test-Access Port and Boundary Scan Architecture.
December 1999 – Revised November 2001
SPRS098C
1
Introduction
2
Introduction
This section describes the main features, gives a brief functional overview of the TMS320VC5421, lists the
pin assignments, and provides a signal description table. This data manual also provides a detailed
description section, electrical specifications, parameter measurement information, and mechanical data
about the available packaging.
NOTE: This data manual is designed to be used in conjunction with the TMS320C54x DSP Functional
Overview (literature number SPRU307).
2.1
Description
The TMS320VC5421 fixed-point digital signal processor (DSP) is a dual-core solution running at 200-MIPS
performance. The 5421 consists of two DSP subsystems capable of core-to-core communications and a
128K-word zero-wait-state on-chip program memory shared by the two DSP subsystems. Each subsystem
consists of one 54x DSP core, 32K-word program/data DARAM, 32K-word data SARAM, 2K-word ROM, three
multichannel serial interfaces, xDMA logic, one timer, one APLL, and other miscellaneous circuitry.
The 5421 also contains a host-port interface (HPI) that allows the 5421 to be viewed as a memory-mapped
peripheral to a host processor. The 5421 is pin-compatible with the TMS320VC5420.
Each subsystem has its separate program and data spaces, allowing simultaneous accesses to program
instructions and data. Two read operations and one write operation can be performed in one cycle. Instructions
with parallel store and application-specific instructions can fully utilize this architecture. Furthermore, data can
be transferred between program and data spaces. Such parallelism supports a powerful set of arithmetic,
logic, and bit-manipulation operations that can all be performed in a single machine cycle. The 5421 includes
the control mechanisms to manage interrupts, repeated operations, and function calls. In addition, the 5421
has 128K words of on-chip program memory that can be shared between the two subsystems.
The 5421 is intended as a high-performance, low-cost, high-density DSP for remote data access or voice-over
IP subsystems. It is designed to maintain the current modem architecture with minimal hardware and software
impacts, thus maximizing reuse of existing modem technologies and development efforts.
2.2
Migration From the 5420 to the 5421
Customers migrating from the 5420 to the 5421 need to take into account the following:
•
•
•
•
•
•
•
•
The memory structure of the 5421 has been changed to incorporate 128K x 16-bit words of two-way
shared memory.
The DMA of the 5421 has been enhanced to provide access to external, as well as internal memory.
The HPI and DMA memory maps have been changed to incorporate the new memory 5421.
2K x 16-bit words of ROM have been added to the 5421 for bootloading purposes only.
The VCO pin on the 5420 has been replaced with the HOLDA pin on the 5421 and the HOLD pin was
added to the 5421 at a previously unused pin location.
The McBSPs have been updated with a new mode that allows 128-channel selection capability.
McBSP CLKX/R pins can be used as inputs to internal clock rate generator for CLKS-like function without
the penalty of extra pins.
The SELA/B pin on 5421 is changed to type I/O/Z for added functionality.
NOTE:
For additional information, see TMS320VC5420 to TMS320VC5421 DSP Migration (literature
number SPRA621).
TMS320C54x is a trademark of Texas Instruments.
2
SPRS098C
December 1999 – Revised November 2001
Introduction
2.3
Pin Assignments
Figure 2–1 provides the pin assignments for the 144-pin low-profile quad flatpack (LQFP) package.
Figure 2–2 illustrates the ball locations for the 144-pin ball grid array (BGA) package and is used in conjunction
with Table 2–2 to locate signal names and ball grid numbers.
2.3.1 Pin Assignments for the PGE Package
109
111
110
112
113
114
115
116
117
119
118
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
75
35
74
36
73
PPA14
PPA15
VSS
PPA16
PPA17
B_INT0
B_INT1
B_NMI
IS
B_GPIO2/BIO
B_GPIO1
B_GPIO0
B_BFSR1
B_BDR1
CVDD
VSS
B_BCLKR1
B_BFSX1
VSS
B_BDX1
B_BCLKX1
CVDD
VSS
TEST
XIO
B_RS
B_XF
B_CLKOUT
HMODE
HPIRS
PPA13
PPA12
VSS
DVDD
PPA11
PPA10
VSS
PPD15
PPD14
VSS
PPD13
PPD12
A_BFSR0
A_BDR0
A_BCLKR0
A_BFSX0
VSS
CVDD
A_BDX0
A_BCLKX0
MSTRB
DS
PS
B_BCLKX0
B_BDX0
DV DD
V SS
B_BFSX0
B_BCLKR0
B_BDR0
CVDD
V SS
B_BFSR0
R/W
PPA2
PPA3
SELA/B
PPD8
PPD9
PPD10
PPD11
VSS
72
76
34
71
77
33
70
78
32
69
79
31
68
80
30
67
81
29
66
82
28
65
83
27
64
84
26
63
85
25
62
86
24
61
87
23
60
88
22
59
89
21
58
90
20
57
91
19
56
92
18
55
93
17
54
94
16
53
95
15
52
96
14
51
97
13
50
98
12
49
99
11
48
100
10
47
101
9
46
102
8
45
103
7
44
104
6
43
105
5
42
106
4
41
3
40
107
39
108
2
38
1
37
PPD7
PPA8
PPA0
DVDD
PPA9
PPD1
A_INT1
A_NMI
IOSTRB
A_GPIO2/BIO
A_GPIO1
A_RS
A_GPIO0
VSS
VSS
CVDD
A_BFSR1
A_BDR1
A_BCLKR1
A_BFSX1
CVDD
VSS
A_BDX1
A_BCLKX1
A_XF
A_CLKOUT
HOLDA
TCK
TMS
TDI
TRST
EMU1/OFF
DVDD
A_INT0
EMU0
TDO
143
144
VSS
PPD0
PPD5
PPD4
PPD6
A_BFSX2
A_BDX2
A_BFSR2
A_BDR2
A_BCLKR2
VSS
CV DD
A_BCLKX2
READY
DVDD
CLKIN
HOLD
VSSA
AV DD
VSS
B_BCLKX2
B_BDX2
B_BFSX2
B_BCLKR2
CVDD
VSS
B_BDR2
B_BFSR2
PPD2
PPD3
PPA1
PPA5
DV DD
PPA4
PPA6
PPA7
The TMS320VC5421PGE 144-pin low-profile quad flatpack (LQFP) is footprint- and pin-compatible with the
5420. Table 2–1 lists the pin number and associated signal name for both the multiplexed mode and the
nonmultiplexed mode.
NOTES: A. DVDD is the power supply for the I/O pins while CVDD is the power supply for the core CPU. VSS is the ground for both the I/O pins
and the core CPU.
B. Pin configuration shown for nonmultiplexed mode only. See the pin assignments table for the TMS320VC5421PGE for multiplexed
functions of specific pins and for specific pin numbers.
Figure 2–1. 144-Pin Low-Profile Flatpack Pin Assignments (PGE – Top View)
December 1999 – Revised November 2001
SPRS098C
3
Introduction
Table 2–1. Pin Assignments for the 144-Pin Low-Profile Quad Flatpack
SIGNAL NAME
(NONMULTIPLEXED)
SIGNAL NAME
(MULTIPLEXED)
SIGNAL NAME
(NONMULTIPLEXED)
PPD7
HD7
1
PPA8
PPA0
A_HINT/HA0
3
DVDD
PPA9
HA9
5
PPD1
A_INT1
SIGNAL NAME
(MULTIPLEXED)
HA8
PIN
NO.
2
4
HD1
6
7
A_NMI
8
A_GPIO3/A_TOUT
9
A_GPIO2/BIO
10
11
A_RS
12
A_ROMEN
13
VSS
14
VSS
15
CVDD
16
A_BFSR1
17
A_BDR1
18
A_BCLKR1
19
A_BFSX1
20
CVDD
21
VSS
22
A_BDX1
23
A_BCLKX1
24
A_XF
25
A_CLKOUT
26
HOLDA
27
TCK
28
TMS
29
TDI
30
TRST
31
EMU1/OFF
32
DVDD
33
A_INT0
34
EMU0
35
TDO
36
IOSTRB
A_GPIO1
A_GPIO0
VSS
37
PPD15
PPD14
HD14
39
VSS
PPD13
HD13
41
PPD12
A_BFSR0
43
A_BDR0
44
A_BCLKR0
45
A_BFSX0
46
VSS
47
CVDD
48
A_BDX0
49
A_BCLKX0
MSTRB
HCS
51
DS
PS
HDS1
HD15
38
40
HD12
42
50
HDS2
52
53
B_BCLKX0
54
B_BDX0
55
DVDD
56
VSS
57
B_BFSX0
58
B_BCLKR0
59
B_BDR0
60
CVDD
61
VSS
62
B_BFSR0
63
R/W
HR/W
64
PPA2
HCNTL1/HA2
65
PPA3
HCNTL0/HA3
66
SELA/B
PPA18
67
PPD8
HD8
68
PPD9
HD9
69
PPD10
HD10
70
PPD11
HD11
71
VSS
PPA10
HA10
73
PPA11
75
VSS
77
PPA13
HPIRS
79
HMODE
80
B_CLKOUT
81
B_XF
82
B_RS
83
XIO
84
TEST
85
VSS
86
DVDD
PPA12
4
PIN
NO.
SPRS098C
HA12
72
HA11
74
76
HA13
78
December 1999 – Revised November 2001
Introduction
Table 2–1. Pin Assignments for the 144-Pin Low-Profile Quad Flatpack (Continued)
SIGNAL NAME
(NONMULTIPLEXED)
SIGNAL NAME
(MULTIPLEXED)
PIN
NO.
SIGNAL NAME
(NONMULTIPLEXED)
SIGNAL NAME
(MULTIPLEXED)
PIN
NO.
CVDD
87
B_BCLKX1
88
B_BDX1
89
VSS
90
B_BFSX1
91
B_BCLKR1
92
VSS
93
CVDD
94
B_BDR1
95
B_BFSR1
96
97
B_GPIO1
98
B_GPIO0
B_ROMEN
B_GPIO2/BIO
99
IS
B_NMI
101
B_INT1
B_INT0
B_GPIO3/B_TOUT
100
102
103
PPA17
PPA16
HA16
105
VSS
PPA15
HA15
107
PPA14
HA14
108
PPA7
HA7
109
PPA6
HA6
110
PPA4
HAS/HA4
111
DVDD
PPA5
HA5
113
PPA1
B_HINT/HA1
114
PPD3
HD3
115
PPD2
HD2
116
117
B_BDR2
118
VSS
119
CVDD
120
B_BCLKR2
121
B_BFSX2
122
B_BDX2
123
B_BCLKX2
124
VSS
125
AVDD
126
VSSA
127
HOLD
128
129
DVDD
130
131
A_BCLKX2
132
CVDD
133
VSS
134
A_BCLKR2
135
A_BDR2
136
A_BFSR2
137
A_BDX2
138
B_BFSR2
CLKIN
READY
HRDY
A_BFSX2
HA17
104
106
112
139
PPD6
HD6
140
PPD4
HD4
141
PPD5
HD5
142
PPD0
HD0
143
VSS
December 1999 – Revised November 2001
144
SPRS098C
5
Introduction
2.3.2 Terminal Assignments for the GGU Package
Table 2–2 lists each ball number and its associated signal name for the TMS320VC5421GGU 144-ball BGA
package, which is footprint- and pin-compatible with the 5420.
13 12 11 10 9
8
7
6
5
4
3
2
1
A
B
C
D
E
F
G
H
J
K
L
M
N
†
To locate the A1 reference maker, see package top view in Figure 6–1.
Figure 2–2. 144-Ball MicroStar BGA Pin Assignments (GGU – Bottom View)
6
SPRS098C
December 1999 – Revised November 2001
Introduction
Table 2–2. Terminal Assignments for the 144-Pin MicroStar BGA
SIGNAL NAME
SIGNAL NAME
BALL
SIGNAL NAME
SIGNAL NAME
BALL
(NONMULTIPLEXED)
(MULTIPLEXED)
NO.
(NONMULTIPLEXED)
(MULTIPLEXED)
NO.
PPD7
HD7
A1
PPA8
DVDD
C1
A_NMI
D1
A_RS
E1
CVDD
F1
A_BDR1
G1
CVDD
H1
A_XF
J1
TMS
K1
EMU1/OFF
L1
EMU0
M1
VSS
N1
PPD0
HD0
A2
VSS
B2
PPA0
A_HINT/HA0
C2
A_INT1
D2
A_GPIO1
E2
VSS
F2
A_BFSR1
G2
VSS
H2
A_CLKOUT
J2
TDI
K2
DVDD
L2
TDO
HA8
B1
M2
PPD15
HD15
N2
PPD6
HD6
A3
PPD4
HD4
B3
PPD5
HD5
C3
PPD1
HD1
D3
A_GPIO2/BIO
E3
VSS
F3
A_BCLKR1
G3
A_BDX1
H3
HOLDA
J3
TRST
A_INT0
L3
PPD14
VSS
N3
A_BFSR2
A4
A_BDX2
B4
A_BFSX2
C4
D4
IOSTRB
PPA9
HA9
A_GPIO0
A_ROMEN
K3
HD14
M3
A_GPIO3/A_TOUT
E4
F4
A_BFSX1
G4
H4
TCK
J4
K4
PPD12
A_BFSR0
M4
A_BDR0
N4
CVDD
A5
VSS
B5
A_BCLKR2
C5
A_BDR2
D5
A_BCLKR0
K5
A_BFSX0
L5
VSS
M5
CVDD
N5
CLKIN
A6
DVDD
B6
HRDY
C6
A_BCLKX2
D6
K6
A_BCLKX0
HCS
M6
DS
AVDD
A7
VSS
B7
VSSA
C7
HOLD
D7
DVDD
K7
B_BDX0
L7
M7
B_BCLKX0
N7
A_BCLKX1
PPD13
READY
HD13
A_BDX0
MSTRB
PS
HDS1
December 1999 – Revised November 2001
HD12
L4
L6
HDS2
N6
SPRS098C
7
Introduction
Table 2–2. Terminal Assignments for the 144-Pin MicroStar BGA (Continued)
SIGNAL NAME
SIGNAL NAME
BALL
SIGNAL NAME
SIGNAL NAME
BALL
(NONMULTIPLEXED)
(MULTIPLEXED)
NO.
(NONMULTIPLEXED)
(MULTIPLEXED)
NO.
B_BCLKX2
A8
B_BDX2
B8
B_BFSX2
C8
B_BCLKR2
D8
B_BDR0
K8
B_BCLKR0
L8
B_BFSX0
M8
VSS
N8
CVDD
A9
VSS
B9
B_BDR2
C9
B_BFSR2
D9
HR/W
K9
B_BFSR0
L9
M9
CVDD
PPD2
HD2
A10
PPD3
HD3
B10
PPA1
B_HINT/HA1
C10
PPA5
HA5
D10
IS
B_GPIO3/B_TOUT
E10
B_BFSR1
B_BCLKR1
G10
TEST
B_CLKOUT
J10
PPA12
HA12
K10
HCNTL0/HA3
M10
R/W
VSS
N9
F10
H10
SELA/B
PPA18
L10
PPA3
PPA2
HCNTL1/HA2
N10
DVDD
A11
PPA4
HAS/HA4
B11
VSS
C11
D11
B_GPIO2/BIO
E11
B_INT0
B_BDR1
F11
VSS
B_BFSX1
H11
B_XF
G11
J11
PPA13
HA13
K11
PPD10
HD10
L11
PPD9
HD9
M11
PPD8
HD8
N11
PPA6
HA6
A12
PPA14
HA14
PPA16
HA16
C12
B_INT1
D12
B_GPIO1
E12
CVDD
F12
B_BDX1
G12
CVDD
H12
B_RS
J12
HPIRS
K12
DVDD
L12
VSS
M12
N12
PPA7
HA7
A13
PPD11
HD11
PPA15
HA15
B12
B13
PPA17
HA17
C13
B_NMI
D13
B_GPIO0
B_ROMEN
E13
VSS
F13
VSS
G13
B_BCLKX1
H13
XIO
J13
HMODE
K13
VSS
L13
M13
PPA10
PPA11
8
SPRS098C
HA11
HA10
N13
December 1999 – Revised November 2001
Introduction
2.4
Signal Descriptions
Table 2–3 lists each signal, function, and operating mode(s) grouped by function. See pin assignments section
for exact pin locations based on package type.
Table 2–3. Signal Descriptions
PIN NAME
TYPE†
DESCRIPTION
DATA SIGNALS
PPA18 (MSB)
Parallel port address bus. The DSP can access the external memory locations by way of the external
memory interface using PPA[18:0] in external memory interface (EMIF) mode when the XIO pin is logic
high. PPA18 is a secondary output function of the SELA/B pin.
PPA17
PPA16
PPA15
PPA12
The PPA[17:0] pins are also multiplexed with the HPI interface. In HPI mode (XIO pin is low), the external
address pins PPA[17:0] are used by a host processor for access to the memory map by way of the on-chip
HPI. Refer to the Host-Port Interface (HPI) Signals section of this table for details on the secondary
functions of these pins.
PPA11
These pins are placed into the high-impedance state when OFF is low.
PPA14
PPA13
PPA10
PPA9
I/O/Z
PPA8
PPA7
PPA6
PPA5
PPA4‡§
PPA3
PPA2
PPA1
PPA0 (LSB)
PPD15 (MSB)
Parallel port data bus. The DSP uses this bidirectional data bus to access external memory when the
device is in external memory interface (EMIF) mode (the XIO pin is logic high).
PPD14
PPD13
This data bus is also multiplexed with the 16-bit HPI data bus. When in HPI mode, the bus is used to transfer
data between the host processor and internal DSP memory via the HPI. Refer to the HPI section of this
table for details on the secondary functions of these pins.
PPD12
PPD11
PPD10
PPD9
PPD8
PPD7
PPD6
I/O/Z¶
The data bus includes bus holders to reduce power dissipation caused by floating, unused pins. The bus
holders also eliminate the need for external pullup resistors on unused pins. When the data bus is not being
driven by the 5421, the bus holders keep data pins at the last driven logic level. The data bus keepers are
disabled at reset and can be enabled/disabled via the BH bit of the BSCR register.
PPD5
PPD4
These pins are placed into high-impedance state when OFF is low.
PPD3
PPD2
PPD1
PPD0 (LSB)
†
I = Input, O = Output, S = Supply, Z = High Impedance
This pin has an internal pullup resistor.
§ These pins are Schmitt triggered inputs.
¶ This pin has an internal bus holder controlled by way of the BSCR register in 54x cLEAD core of DSP subsystem A .
# This pin is used by Texas Instruments for device testing and should be left unconnected.
|| This pin has an internal pulldown resistor.
‡
December 1999 – Revised November 2001
SPRS098C
9
Introduction
Table 2–3. Signal Descriptions (Continued)
PIN NAME
TYPE†
DESCRIPTION
DATA SIGNALS (CONTINUED)
A_INT0§
B_INT0§
A_INT1§
I
External user interrupts. A_INT0–B_INT0 are prioritized and are maskable by the interrupt mask register
(IMR) and the interrupt mode bit. A_INT1 –B_INT1 can be polled and reset by way of the interrupt flag
register (IFR).
I
Nonmaskable interrupt. NMI is an external interrupt that cannot be masked by way of the INTM or the IMR.
When NMI is activated, the processor traps to the appropriate vector location.
I
Reset. RS causes the digital signal processor (DSP) to terminate execution and causes a reinitialization
of the CPU and peripherals. When RS is brought to a high level, execution begins at location 0FF80h of
program memory. RS affects various registers and status bits.
B_INT1§
INITIALIZATION, INTERRUPT, AND RESET OPERATIONS
A_NMI§
B_NMI§
A_RS§
B_RS§
The XIO pin is used to configure the parallel port as a host-port interface (HPI mode when XIO pin is low),
or as an asynchronous memory interface (EMIF mode when XIO pin is high).
XIO
I
NOTE: Because the XIO signal is asynchronous, caution must be taken when changing the state of the
XIO pin to ensure the current cycle is properly ended.
At device reset, the XIO pin level determines the initialization value of the MP/MC bit (a bit in the processor
mode status (PMST) register). Refer to the memory section for details.
GENERAL-PURPOSE I/O PINS
A_XF
O/Z
B_XF
A_GPIO0
B_GPIO0
A_GPIO1
B_GPIO1
A_GPIO2/BIO
B_GPIO2/BIO
External flag output (latched software-programmable output-only signal). Bit-addressable. A_XF and
B_XF are placed into the high-impedance state when OFF is low.
A_ROMEN
I/O/Z
B_ROMEN
I
General-purpose I/O pins. The secondary function of these pins. In XIO mode, the
ROM enable (ROMEN) pins are used to enable the applicable on-chip ROM after
reset.
I/O/Z
General-purpose I/O pins (software-programmable I/O signal). Values can be latched (output) by writing into the GPIO register. The states of GPIO pins (inputs) can be read by reading the GPIO register.
The GPIO direction is also programmable by way of the DIRn field in the GPIO register.
I/O/Z
General-purpose I/O. These pins can be configured like GPIO0–GPIO1; however, as an input, the pins
operate as the traditional branch control bit (BIO). If application code does not perform BIO-conditional
instructions, these pins operate as general inputs.
PRIMARY
IOSTRB
A_GPIO3 (A_TOUT)
B_GPIO3 (B_TOUT)
I/O/Z
O
IS
When the device is in HPI mode and HMODE = 0 (multiplexed), these pins act
according to the general-purpose I/O control register. TOUT bit must be set to “1”
to drive the timer output on the pin. IF TOUT = 0, then these pins are
general-purpose I/Os. In EMIF mode (XIO = 1), these signals are active during I/O
space accesses.
†
I = Input, O = Output, S = Supply, Z = High Impedance
This pin has an internal pullup resistor.
§ These pins are Schmitt triggered inputs.
¶ This pin has an internal bus holder controlled by way of the BSCR register in 54x cLEAD core of DSP subsystem A .
# This pin is used by Texas Instruments for device testing and should be left unconnected.
|| This pin has an internal pulldown resistor.
‡
10
SPRS098C
December 1999 – Revised November 2001
Introduction
Table 2–3. Signal Descriptions (Continued)
PIN NAME
TYPE†
DESCRIPTION
MEMORY CONTROL SIGNALS
Program space select signal. The PS signal is asserted during external program space accesses. This pin
is placed into the high-impedance state when OFF is low.
This pin is also multiplexed with the HPI, and functions as the HDS1 data strobe input signal in HPI mode.
Refer to the HPI section of this table for details on the secondary function of this pin.
Data space select signal. The DS signal is asserted during external data space accesses. This pin is
placed into the high-impedance state when OFF is low.
PS‡§
DS‡§
O/Z
IS
This pin is also multiplexed with the HPI, and functions as the HDS2 data strobe input signal in HPI mode.
Refer to the HPI section of this table for details on the secondary function of this pin.
I/O space select signal. The IS signal is asserted during external I/O space accesses. This pin is placed
into the high-impedance state when OFF is low.
This pin is also multiplexed with the general-purpose I/O feature, and functions as the B_GPIO3 (B_TOUT)
input/output signal in HPI mode. Refer to the General-Purpose I/O section of this table for details on the
secondary function of this pin.
MSTRB‡§
READY
O/Z
I
Program and data memory strobe (active in EMIF mode). This pin is placed into the high-impedance state
when OFF is low.
Data-ready input signal. READY indicates that the external device is prepared for a bus transaction to be
completed. If the device is not ready (READY = 0), the processor waits one cycle and checks READY
again. The processor performs the READY detection if at least two software wait states are programmed.
This pin is also multiplexed with the HPI, and functions as the host-port data ready (output) in HPI mode.
Refer to the HPI section of this table for details on the secondary function of this pin.
Read/write output signal. R/W indicates transfer direction during communication to an external device.
R/W is normally in the read mode (high), unless it is asserted low when the DSP performs a write operation.
R/W
O/Z
This pin is also multiplexed with the HPI, and functions as the host-port read/write input in HPI mode. Refer
to the HPI section of this table for details on the secondary function of this pin.
This pin is placed into the high-impedance state when OFF is low.
I/O space memory strobe. External I/O space is accessible by the CPU and not the direct memory access
(DMA) controller. The DMA has its own dedicated I/O space that is not accessible by the CPU.
IOSTRB
O/Z
This pin is also multiplexed with the general-purpose I/O feature, and functions as the A_GPIO3 (A_TOUT)
signal in HPI mode. Refer to the General Purpose I/O section of this table for details on the secondary
function of this pin.
This pin is placed into the high-impedance state when OFF is low.
†
I = Input, O = Output, S = Supply, Z = High Impedance
‡ This pin has an internal pullup resistor.
§ These pins are Schmitt triggered inputs.
¶ This pin has an internal bus holder controlled by way of the BSCR register in 54x cLEAD core of DSP subsystem A .
# This pin is used by Texas Instruments for device testing and should be left unconnected.
|| This pin has an internal pulldown resistor.
December 1999 – Revised November 2001
SPRS098C
11
Introduction
Table 2–3. Signal Descriptions (Continued)
PIN NAME
TYPE†
DESCRIPTION
MEMORY CONTROL SIGNALS (CONTINUED)
PRIMARY
For HPI access (XIO=0), SELA/B is an input.
See Table 3–3 for a truth table of SELA/B, HMODE, and XIO pins and functionality.
PPA18
O/Z
SELA/B
I
For external memory accesses (XIO=1), SELA/B is multiplexed as output PPA18.
See the PPA signal descriptions. These pins are placed into the high-impedance
state when OFF is low.
HOLD‡
I
Hold. HOLD is asserted to request control of the address, data, and control lines. When acknowledged,
these lines go into the high-impedance state.
HOLDA
O/Z
Hold acknowledge. HOLDA indicates to the external circuitry that the processor is in a hold state and that
the address, data, and control lines are in the high-impedance state, allowing them to be available to the
external circuitry. HOLDA also goes into the high-impedance state when OFF is low.
CLOCKING SIGNALS
A_CLKOUT
B_CLKOUT
CLKIN§
O/Z
I
Master clock output signal. CLKOUT cycles at the machine-cycle rate of the CPU. The internal machine
cycle is bounded by the falling edges of this signal. The CLKOUT pin can be turned off by writing a “1” to
the CLKOFF bit of the PMST register. CLKOUT goes into the high-impedance state when EMU1/OFF is
low.
Input clock to the device. CLKIN connects to an oscillator circuit/device (PLL).
MULTICHANNEL BUFFERED SERIAL PORT 0, 1, AND 2 SIGNALS
A_BCLKR0‡§
Receive clocks. BCLKR serves as the serial shift clock for the buffered serial-port receiver. Input from an
external clock source for clocking data into the McBSP. When not being used as a clock, these pins can
be used as general-purpose I/O by setting RIOEN = 1.
B_BCLKR0‡§
A_BCLKR1‡§
B_BCLKR1‡§
I/O/Z
A_BCLKR2‡§
BCLKR can be configured as an output by the way of the CLKRM bit in the PCR register.
B_BCLKR2‡§
These pins are placed into the high-impedance state when OFF is low.
A_BCLKX0‡§
B_BCLKX0‡§
A_BCLKX1‡§
B_BCLKX1‡§
I/O/Z
Transmit clocks. Clock signal used to clock data from the transmit register. This pin can also be configured
as an input by setting the CLKXM = 0 in the PCR register. BCLKX can be sampled as an input by way of
the IN1 bit in the SPC register. When not being used as a clock, these pins can be used as general-purpose
I/O by setting XIOEN = 1.
A_BCLKX2‡§
These pins are placed into the high-impedance state when OFF is low.
B_BCLKX2‡§
A_BDR0
B_BDR0
A_BDR1
B_BDR1
I
Buffered serial data receive (input) pin. When not being used as data-receive pins, these pins can be used
as general-purpose I/O by setting RIOEN = 1.
A_BDR2
B_BDR2
†
I = Input, O = Output, S = Supply, Z = High Impedance
This pin has an internal pullup resistor.
§ These pins are Schmitt triggered inputs.
¶ This pin has an internal bus holder controlled by way of the BSCR register in 54x cLEAD core of DSP subsystem A .
# This pin is used by Texas Instruments for device testing and should be left unconnected.
|| This pin has an internal pulldown resistor.
‡
12
SPRS098C
December 1999 – Revised November 2001
Introduction
Table 2–3. Signal Descriptions (Continued)
PIN NAME
TYPE†
DESCRIPTION
MULTICHANNEL BUFFERED SERIAL PORT 0, 1, AND 2 SIGNALS (CONTINUED)
A_BDX0
B_BDX0
A_BDX1
B_BDX1
O/Z
Buffered serial-port transmit (output) pin. When not being used as data-transmit pins, these pins can be
used as general-purpose I/O by setting XIOEN = 1. These pins are placed into the high-impedance state
when OFF is low.
I/O/Z
Frame synchronization pin for buffered serial-port input data. The BFSR pulse initiates the receive-data
process over the BDR pin. When not being used as data-receive synchronization pins, these pins can be
used as general-purpose I/O by setting RIOEN = 1. These pins are placed into the high-impedance state
when OFF is low.
I/O/Z
Buffered serial-port frame synchronization pin for transmitting data. The BFSX pulse initiates the
transmit-data process over the BDX pin. If RS is asserted when BFSX is configured as output, then BFSX
is turned into input mode by the reset operation. When not being used as data-transmit synchronization
pins, these pins can be used as general-purpose I/O by setting XIOEN = 1. These pins are placed into the
high-impedance state when OFF is low.
A_BDX2
B_BDX2
A_BFSR0
B_BFSR0
A_BFSR1
B_BFSR1
A_BFSR2
B_BFSR2
A_BFSX0
B_BFSX0
A_BFSX1
B_BFSX1
A_BFSX2
B_BFSX2
HOST-PORT INTERFACE (HPI) SIGNALS
PRIMARY
These pins are multiplexed with the external interface pins and are used by the HPI
when the subsystem is in HPI mode (XIO = 0, MP/MC = 0).
HA[17:0]
I
PPA[17:0]
O/Z
See the PPA signal descriptions. These pins are placed into the high-impedance
state when OFF is low.
NOTE: HA4 has a pullup and a Schmitt trigger buffer.
PRIMARY
Parallel bidirectional data bus. These pins are multiplexed with the external
interface pins and are used as an HPI interface when XIO = 0.
HD[15:0]
I/O/Z
PPD[15:0]
I/O/Z
These pins include bus holders to reduce power dissipation caused by floating,
unused inputs. The bus holders also eliminate the need for external pullup resistors
on unused inputs. In multiplexed address/data mode (HMODE = 0), when the data
bus is not being driven by the 5421, the bus holders keep the multiplexed address
inputs on these pins at the last logic level driven by the host. The data bus holders
are disabled at reset and can be enabled/disabled via the BH bit of the BSCR
register.
See the PPD signal descriptions. These pins are placed into the high-impedance
state when OFF is low.
†
I = Input, O = Output, S = Supply, Z = High Impedance
This pin has an internal pullup resistor.
§ These pins are Schmitt triggered inputs.
¶ This pin has an internal bus holder controlled by way of the BSCR register in 54x cLEAD core of DSP subsystem A .
# This pin is used by Texas Instruments for device testing and should be left unconnected.
|| This pin has an internal pulldown resistor.
‡
December 1999 – Revised November 2001
SPRS098C
13
Introduction
Table 2–3. Signal Descriptions (Continued)
PIN NAME
TYPE†
DESCRIPTION
HOST-PORT INTERFACE (HPI) SIGNALS (CONTINUED)
HCNTL0
HCNTL1
HAS‡§
I
I
PPA3
PPA2
PPA4‡§
O/Z
O/Z
HPI control inputs. Use PPA3 and PPA2 for the HCNTL0 and HCNTL1 values during
the HPI HPIC, HPIA, and HPID reads/writes. Only used in multiplexed address/data
mode (HMODE = 0).
These pins are shared with the external memory interface and are only used by the
HPI when the interface is in HPI mode (XIO pin is low). These pins are placed into
the high-impedance state when OFF is low.
Address strobe input. Hosts with multiplexed address and data pins require HAS
to latch the address in the HPIA register. This signal is only used in HPI multiplexed
address/data mode (HMODE pin is low).
This pin is shared with the external memory interface and is only used by the HPI
when the interface is in HPI mode (XIO pin is low). This pin is placed into the
high-impedance state when OFF is low.
HPI chip-select signal. This signal must be active during HPI transfers, and can
remain active between concurrent transfers.
HCS‡§
HDS1‡§
HDS2‡§
I
I
MSTRB‡§
PS‡§
DS‡§
O/Z
This pin is shared with the external memory interface and is only used by the HPI
when the interface is in HPI mode (XIO pin is low). This pin is placed into the
high-impedance state when OFF is low.
HPI data strobes. HDS1 and HDS2 are driven by the host read and write strobes
to control HPI transfers.
O/Z
These pins are shared with the external memory interface and are only used by the
HPI when the interface is in HPI mode (XIO pin is low).
These pins are placed into the high-impedance state when OFF is low.
HPI read/write signal. This signal is used by the host to control the direction of an
HPI transfer.
HR/W
I
R/W
O/Z
This pin is shared with the external memory interface and is only used by the HPI
when the interface is in HPI mode (XIO pin is low).
This pin is placed into the high-impedance state when OFF is low.
HPI data-ready output. The ready output informs the host when the HPI is ready for
the next transfer.
HRDY
O/Z
READY
I
This pin is shared with the external memory interface and is only used by the HPI
when the interface is in HPI mode (XIO pin is low). HRDY is placed into the
high-impedance state when OFF is low.
PRIMARY
A_HINT
B_HINT
HPIRS§
O/Z
PPA0
PPA1
I
O/Z
Host interrupt pin. HPI can interrupt the host by asserting this low. The host can clear
this interrupt by writing a “1” to the HINT bit of the HPIC register. Only supported in
HPI multiplexed address/data mode (HMODE pin is low). These pins are placed into
the high-impedance state when OFF is low.
Host-port interface (HPI) reset pin. This signal resets the host port interface and both subsystems.
†
I = Input, O = Output, S = Supply, Z = High Impedance
‡ This pin has an internal pullup resistor.
§ These pins are Schmitt triggered inputs.
¶ This pin has an internal bus holder controlled by way of the BSCR register in 54x cLEAD core of DSP subsystem A .
# This pin is used by Texas Instruments for device testing and should be left unconnected.
|| This pin has an internal pulldown resistor.
14
SPRS098C
December 1999 – Revised November 2001
Introduction
Table 2–3. Signal Descriptions (Continued)
PIN NAME
TYPE†
DESCRIPTION
HOST-PORT INTERFACE (HPI) SIGNALS (CONTINUED)
Host mode select. When this pin is low, it selects the HPI multiplexed address/data mode. The multiplexed
address/data mode allows hosts with multiplexed address/data lines access to the HPI registers HPIC,
HPIA, and HPID. Host-to-DSP and DSP-to-host interrupts are supported in this mode.
HMODE
I
When HMODE is high, it selects the HPI nonmultiplexed mode. HPI nonmultiplexed mode allows hosts
with separate address/data buses to access the HPI address range by way of the 18-bit address bus and
the HPI data (HPID) register via the 16-bit data bus. Host-to-DSP and DSP-to-host interrupts are not
supported in this mode.
SUPPLY PINS
AVDD
S
Dedicated power supply that powers the PLL. AVDD = 1.8 V. AVDD can be connected to CVDD.
CVDD
S
Dedicated “clean” power supply that powers the core CPUs. CVDD = 1.8 V
DVDD
S
Dedicated “dirty” power supply that powers the I/O pins. DVDD = 3.3 V
VSS
S
Digital ground. Dedicated ground plane for the device.
VSSA
S
Analog ground. Dedicated ground for the PLL. VSSA can be connected to VSS if digital and analog grounds
are not separated.
TEST PIN
TEST#
No connection
EMULATION/TEST PINS
TCK‡§
I
Standard test clock. This is normally a free-running clock signal with a 50% duty cycle. Changes on the
test access port (TAP) of input signals TMS and TDI are clocked into the TAP controller, instruction register,
or selected test-data register on the rising edge of TCK. Changes at the TAP output signal (TDO) occur
on the falling edge of TCK.
TDI‡
I
Test data input. Pin with an internal pullup device. TDI is clocked into the selected register (instruction or
data) on a rising edge of TCK.
TDO
O/Z
Test data output. The contents of the selected register is shifted out of TDO on the falling edge of TCK.
TDO is in high-impedance state except when the scanning of data is in progress. These pins are
placed into high-impedance state when OFF is low.
TMS‡
I
Test mode select. Pin with internal pullup device. This serial control input is clocked into the TAP controller
on the rising edge of TCK.
TRST||
I
Test reset. When high, TRST gives the scan system control of the operations of the device. If TRST is
driven low, the device operates in its functional mode and the IEEE 1149.1 signals are ignored. Pin with
internal pulldown device.
†
I = Input, O = Output, S = Supply, Z = High Impedance
This pin has an internal pullup resistor.
§ These pins are Schmitt triggered inputs.
¶ This pin has an internal bus holder controlled by way of the BSCR register in 54x cLEAD core of DSP subsystem A .
# This pin is used by Texas Instruments for device testing and should be left unconnected.
|| This pin has an internal pulldown resistor.
‡
December 1999 – Revised November 2001
SPRS098C
15
Introduction
Table 2–3. Signal Descriptions (Continued)
PIN NAME
TYPE†
DESCRIPTION
EMULATION/TEST PINS (CONTINUED)
EMU0
I/O/Z
Emulator interrupt 0 pin. When TRST is driven low, EMU0 must be high for the activation of the EMU1/OFF
condition. When TRST is driven high, EMU0 is used as an interrupt to or from the emulator system and
is defined as I/O.
Emulator interrupt 1 pin. When TRST is driven high, EMU1/OFF is used as an interrupt to or from the
emulator system and is defined as I/O. When TRST transitions from high to low, then EMU1 operates as
OFF. EMU/OFF = 0 puts all output drivers into the high-impedance state.
EMU1/OFF
I/O/Z
Note that OFF is used exclusively for testing and emulation purposes (and not for multiprocessing
applications). Therefore, for the OFF condition, the following conditions apply:
TRST = 0, EMU0 = 1, EMU1 = 0
†
I = Input, O = Output, S = Supply, Z = High Impedance
This pin has an internal pullup resistor.
§ These pins are Schmitt triggered inputs.
¶ This pin has an internal bus holder controlled by way of the BSCR register in 54x cLEAD core of DSP subsystem A .
# This pin is used by Texas Instruments for device testing and should be left unconnected.
|| This pin has an internal pulldown resistor.
‡
16
SPRS098C
December 1999 – Revised November 2001
Functional Overview
3
Functional Overview
Pbus
Dbus
Ebus
32K RAM
Dual Access
Program/Data
32K RAM
Single Access
Data
54X cLEAD
(Core A)
Cbus
Pbus
Ebus
Cbus
Dbus
Dbus
Ebus
Cbus
Pbus
P, C, D, E Buses and Control Signals
2K Program
ROM
MBus
GPIO
RHEA
Bridge
TI BUS
RHEA Bus
McBSP1
DSP Subsystem A
16 HPI
xDMA
logic
McBSP2
MBus
P bus
RHEA bus
Arbitrator
McBSP3
RHEAbus
TIMER
APLL
Arbitrator
128K
Dual
Access
PRAM
Core-to-Core
FIFO Interface
MBus
P
Interprocessor
IRQs
Cycle
Arrangmnt
XIO
16HPI
JTAG
Clocks
MBus
Pbus
MBus
Ebus
Dbus
32K RAM
Dual Access
Program/Data
32K RAM
Single Access
Data
54X cLEAD
(Core B)
Cbus
Pbus
Ebus
Dbus
Cbus
Ebus
Dbus
Cbus
Pbus
P, C, D, E Buses and Control Signals
2K
Program ROM
MBus
GPIO
TI Bus
RHEA
Bridge
RHEA Bus
McBSP1
16 HPI
Host Access Bus
xDMA
Logic
McBSP2
MBus
RHEA bus
Arbitrator
McBSP3
MBus
TIMER
JTAG
DSP Subsystem B
Figure 3–1. TMS320VC5421 Functional Block Diagram
December 1999 – Revised November 2001
SPRS098C
17
Functional Overview
3.1
Memory
Each 5421 DSP subsystem maintains the peripheral register memory map and interrupt location/priorities of
the standard 5420. Figure 3–2 shows the size of the required memory blocks and their link map within the
program and data space of the cLEAD core. The total on-chip memory for the 5421 devices is 256K-word
data/program.
Hex
00 0000
00 005F
00 0060
Data
MemoryMapped
Registers
Hex
Hex
01 005F
01 0060
Program Page 2
Hex
Program Page 3
03 0000
Reserved
02 005F
02 0060
Reserved
03 005F
03 0060
On-Chip
DARAM A/B§
(32K Words)
Prog/Data
(OVLY=1)
On-Chip
DARAM A/B§
(32K Words)
Prog/Data
(OVLY=1)
External
(OVLY=0)
External
(OVLY=0)
External
(OVLY=0)
External
(OVLY=0)
On-Chip
two-way
shared
DARAM 0¶
(24K Words)
Prog Only
02 7FFF
02 8000
On-Chip
two-way
shared
DARAM 1¶
(32K Words)
Prog Only
Shared 0
Reserved
External‡
0n 7FFF
0n 8000
03 7FFF
03 8000
On-Chip
two-way
shared
DARAM 3#
(32K Words)
Prog Only
On-Chip
two-way
shared
DARAM 2#
(32K Words)
Prog Only
Program Page n
0n 005F
0n 0060
On-Chip
DARAM A/B§
(32K Words)
Prog/Data
(OVLY=1)
01 7FFF
01 8000
Hex
0n 0000
On-Chip
DARAM A/B§
(32K Words)
Prog/Data
(OVLY=1)
00 7FFF
00 8000
External
(DROM=0)
Hex
02 0000
Reserved
Reserved
On-Chip
SARAM A/B
(32K Words)
Data Only
(DROM=1)
Program Page 1
01 0000
00 005F
00 0060
On-Chip
DARAM A/B§
(32K Words)
Prog/Data
00 7FFF
00 8000
Program Page 0
00 0000
External‡
00 DFFF
00 E000
Reserved
00 F7FF
00 F800
00 FFFF
00 FFFF
ROM
(ROMEN=1)†
Shared 2
Shared 1
01 FFFF
02 FFFF
(extended)
Shared 3
0n FFFF
03 FFFF
(extended)
(extended)
(n = 4 – 127)
†
ROM enabled after reset.
When CPU PMST register bit MP/MC=0 and an address is generated outside the on-chip memory bound or the address reach, i.e.,
XPC > 3h, access is always external, if XIO = 1. Pages 8–127 are mapped over pages 4–7. When XIO = 1 and MP/MC = 1, program pages 0,
1, 2, and 3 are external. Pages 4–127 are mapped over pages 0–3.
§ On-chip DARAM A and SARAM A are for subsystem A. Likewise, on-chip DARAM B and SARAM B are for subsystem B.
¶ On-chip DRAM 0 and DRAM 1 are owned by subsystem A and shared with subsystem B.
# On-chip DRAM 2 and DRAM 3 are owned by subsystem B and shared with subsystem A.
NOTES: A. Clearing the ROMEN bit (GPIO[7]) enables an 8K-word block (0E000h – 0FFFFh) of DARAM .
B. All external accesses require the XIO pin to be high.
C. CPU I/O space is a single page of 64K words. Access is always external.
D. All internal memory is divided into 8K blocks.
‡
Figure 3–2. Memory Map Relative to CPU Subsystems A and B
3.1.1 On-Chip Dual-Access RAM (DARAM)
The 5421 subsystems A and B each have 32K 16-bit words of on-chip DARAM (4 blocks of 8K words). Each
of these DARAM blocks can be accessed twice per machine cycle. This memory is intended primarily to store
data values; however, it can be used to store program as well. At reset, the DARAM is mapped into data
memory space. The DARAM can be mapped into program/data memory space by setting the OVLY bit in the
processor-mode status (PMST) register of the 54x CPU in each DSP subsystem.
18
SPRS098C
December 1999 – Revised November 2001
Functional Overview
3.1.2 On-Chip Single-Access RAM (SARAM)
The 5421 subsystems A and B each have 32K 16-bit words of on-chip SARAM (4 blocks of 8K words). Each
of these SARAM blocks can be accessed once per machine cycle. This memory is intended to store data
values only. At reset, the SARAM is disabled. The SARAM can be enabled in data memory space by setting
the DROM bit in the PMST register.
3.1.3 On-Chip Two-Way Shared RAM (DARAM)
The 5421 has 128K 16-bit words of on-chip DARAM (16 blocks of 8K words) that is shared between the two
DSP subsystems. This memory is intended to store program only. Each subsystem is able to make one
instruction fetch from any location in two-way shared memory each cycle. Neither subsystem CPU can write
to the two-way shared memory as only the DMA can write to two-way shared memory.
3.1.4 On-Chip Boot ROM
The 5421 subsystems A and B each have 2K 16-bit words of on-chip ROM. This ROM is used for bootloading
functions only. Enabling the ROM maps out one 8K-word block of the shared program memory. The ROM can
be disabled by clearing bit 7 (ROMEN) of the general-purpose I/O (GPIO) register. Table 3–1 shows the
XIO/ROMEN modes. The ROM is enabled or disabled at reset for each subsystem depending on the state
of the GPIO0 pin for that subsystem.
Table 3–1. XIO/ROMEN Modes
XIO
ROMEN/GPIO0
0
x
MODE
Fetch internal from RAM
1
0
Fetch external
1
1
ROM enabled
3.1.5 Extended Program Memory
The program memory space on the 5421 device addresses up to 512K 16-bit words. The 5421 device uses
a paged extended memory scheme in program space to allow access of up to 512K of program memory . This
extended program memory (each subsystem) is organized into eight pages (0–7), pages 0–3 are internal,
pages 4–7 are external, each 64K in length. (Pages 8–127 as defined by the program counter extension
register (XPC) are aliases for pages 4–7.) Access to the extended program memory is similar to the 5420. To
implement the extended program memory scheme, the 5421 device includes the following feature:
•
Two 54x instructions are extended to use the additional two bits in the 5421 device.
–
–
READA – Read program memory addressed by accumulator A and store in data memory
WRITA – Write data to program memory addressed by accumulator A
(Writes not allowed for CPUs to shared program memory)
December 1999 – Revised November 2001
SPRS098C
19
Functional Overview
3.1.6 Program Memory
The program memory is accessible on multiple pages, depending on the XPC value. Within these pages,
memory is accessible, depending on the address range.
•
Access in the lower 32K of each page is dependent on the state of OVLY.
–
–
•
OVLY = 0 – Program memory is accessed externally for all values of XPC.
OVLY = 1 – Program memory is accessed from local data/program DARAM for all values of XPC.
Access in the upper 32K of each page is dependent on the state of MP/MC and the value of XPC.
–
–
MP/MC = 0 – Program memory is accessed internally from two-way shared DARAM for XPC = 0–3.
Program memory is accessed externally for XPC = 4–127.
MP/MC = 1 – Program memory is accessed externally for all values of XPC.
3.1.7 Data Memory
The data memory space is a single page of 64K. Access is dependent on the address range. Access in the
lower 32K of data memory is always from local DARAM.
Access in the upper 32K of data memory is dependent on the state of DROM.
•
•
DROM = 0 – Data memory is accessed externally
DROM = 1 – Data memory is accessed internally from local SARAM
3.1.8 I/O Memory
The I/O space is a single page of 64K. Access is always external.
When XIO = 0 and an access to external memory is attempted, any write is ignored and any read is an unknown
value.
3.2
Multicore Reset Signals
The 5421 device includes three reset signals: A_RS, B_RS, and HPIRS. The A_RS and B_RS pins function
as the CPU reset signal for subsystem A and subsystem B, respectively. These signals reset the state of the
CPU registers and upon release, initiate the reset function. Additionally, the A_RS signal resets the on-chip
PLL and initializes the CLKMD register to bypass mode.
The HPI reset signal (HPIRS) places the HPI peripheral into a reset state. It is necessary to wait three clock
cycles after the rising edge of HPIRS before performing an HPI access. The HPIRS signal also resets the PLL
by turning off the PLL and initializing the CLKMD register to bypass mode.
3.3
Bootloader
The on-chip bootloader is used to automatically transfer user code from an external source to anywhere in
program memory after reset. The XIO pin is sampled during a hardware reset and the results indicate the
operating mode as shown in Table 3–2.
Table 3–2. Bootloader Operating Modes
XIO
AFTER RESET
HPI mode, bootload is controlled by host. The external host holds the 5421 in reset while it loads the on-chip
memory of one or both subsystems as determined by the SELA/B pin.
The host can release the 5421 from reset by either of the following methods:
1.
If the A_RS/B_RS pins are held low while HPIRS transitions from low to high, the subsystem cores reset will
be controlled by the A_RS/B_RS pins. When the host has finished downloading code, it drives A_RS/B_RS
high to release the cores from reset.
2.
If the A_RS/B_RS pins are held high while HPIRS transitions from low to high, the subsystems stay in reset
until a HPI data write to address 0x2F occurs. This means the host can download code to subsystem A and
then release core A from reset by writing any data to core A address 0x2F via the HPI. The host can then repeat
the sequence for core B. This mode allows the host to control the 5421 reset without additional hardware.
0
1
20
SPRS098C
XIO mode. ROM is mapped in, if ROMEN pin = 1 during reset.
December 1999 – Revised November 2001
Functional Overview
The 5421 bootloader provides the following options for the source of code to download:
•
•
Parallel from 8-bit or 16-bit-wide EPROM
Serial boot from McBSPs, 8-bit mode
GPIO register bit 7 (ROMEN) is used to enable/disable the ROM after reset. The ROMEN bit reflects the status
of the ROMEN/GPIO0 pin for each core. ROMEN = 1 indicates that the ROM and the 8K-word program
memory block (00 E000h–00 FFFFh) are not available for a CPU write. When ROMEN = 0, this 8K-word
program memory is available and the ROM is disabled.
A combination of interrupt flags and the bit values of an external memory location determine the selection of
the various boot options.
3.4
External Interface (XIO)
The external interface (XIO) supports the 5421 master boot modes and other external accesses. Its features
include:
•
•
•
•
•
Multiplexed with the HPI pins
Selection of XIO or HPI mode is determined by a dedicated pin (XIO)
Provides 512K words of external program space, 64K words of external data space, and 64K words of
external I/O space.
Different boot modes are selectable by the XIO, HMODE, and A_RS/B_RS pins.
After reset, the control register bit ROMEN is always preset to 1.
While XIO = 0 during reset, host HPI mode is on, the host sees all RAM, and ROM is disabled. A host write
to 002Fh releases the CPUs from reset; the 002Fh write by the host clears the ROMEN bit in the GPIO register.
While XIO = 1 and ROMEN = 1 during reset, the CPU starts from ROM (0FF80h) to do boot selection. After
branching to non-ROM area, the code changes the ROMEN bit to enable the RAM area occupied by ROM.
While XIO = 1 and ROMEN = 0 during reset, the CPU starts from external (0FF80h) to do boot selection.
Table 3–3 provides a complete description of HMODE, SELA/B, and XIO pin functionality.
Table 3–3. XIO/HPI Modes
3.5
HMODE
SELA/B
0
0
HPI muxed address/data subsystem A slave to host
HPI MODES (XIO = 0)
SELA/B pin is multiplexed as PPA18 output.
XIO MODES (XIO = 1)
0
1
HPI muxed address/data subsystem B slave to host
SELA/B pin is multiplexed as PPA18 output.
1
0
HPI non-muxed address/data subsystem A slave to host
SELA/B pin is multiplexed as PPA18 output.
1
1
HPI non-muxed address/data subsystem B slave to host
SELA/B pin is multiplexed as PPA18 output.
On-Chip Peripherals
All the 54x devices have the same CPU structure; however, they have different on-chip peripherals connected
to their CPUs. The on-chip peripheral options provided are:
•
•
•
•
•
•
Software-programmable wait-state generator
Programmable bank-switching
Parallel I/O ports
Multichannel buffered serial ports (McBSPs)
A hardware timer
A software-programmable clock generator using a phase-locked loop (PLL)
December 1999 – Revised November 2001
SPRS098C
21
Functional Overview
3.5.1 Software-Programmable Wait-State Generators
The software-programmable wait-state generator can be used to extend external bus cycles up to fourteen
machine cycles to interface with slower off-chip memory and I/O devices. The software wait-state register
(SWWSR) controls the operation of the wait-state generator. The SWWSR of a particular DSP subsystem
(A or B) is used for the external memory interface, depending on the state of the xDMA/XIO arbitration logic
(see Direct Memory Access (DMA) Controller section 3.8 and Table 3–4. The 14 least significant bits (LSBs)
of the SWWSR specify the number of wait states (0–7) to be inserted for external memory accesses to five
separate address ranges. This allows a different number of wait states for each of the five address ranges.
Additionally, the software wait-state multiplier (SWSM) bit of the software wait-state control register (SWCR)
defines a multiplication factor of 1 or 2 for the number of wait states. At reset, the wait-state generator is
initialized to provide seven wait states on all external memory accesses. The SWWSR bit fields are shown
in Figure 3–3 and described in Table 3–4.
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
XPA
I/O
Data
Data
Program
Program
R/W=0
R/W=111
R/W=111
R/W=111
R/W=111
R/W=111
0
LEGEND: R = Read, W = Write
Figure 3–3. Software Wait-State Register (SWWSR) [Memory-Mapped Register (MMR) Address 0028h]
Table 3–4. Software Wait-State Register (SWWSR) Bit Fields
BIT
NO.
NAME
RESET
VALUE
15
XPA
0
Extended program address control bit. XPA is used in conjunction with the program space fields
(bits 0 through 5) to select the address range for program space wait states.
14–12
I/O
1
I/O space. The field value (0–7) corresponds to the base number of wait states for I/O space accesses
within addresses 0000–FFFFh. The SWSM bit of the SWCR defines a multiplication factor of 1 or 2 for
the base number of wait states.
11–9
Data
1
Upper data space. The field value (0–7) corresponds to the base number of wait states for external data
space accesses within addresses 8000–FFFFh. The SWSM bit of the SWCR defines a multiplication
factor of 1 or 2 for the base number of wait states.
8–6
Data
1
Lower data space. The field value (0–7) corresponds to the base number of wait states for external data
space accesses within addresses 0000–7FFFh. The SWSM bit of the SWCR defines a multiplication
factor of 1 or 2 for the base number of wait states.
FUNCTION
Upper program space. The field value (0–7) corresponds to the base number of wait states for external
program space accesses within the following addresses:
5–3
Program
1
XPA = 0: x8000–xFFFFh
XPA = 1: The upper program space bit field has no effect on wait states.
The SWSM bit of the SWCR defines a multiplication factor of 1 or 2 for the base number of wait states.
Program space. The field value (0–7) corresponds to the base number of wait states for external program
space accesses within the following addresses:
2–0
Program
1
XPA = 0: x0000–x7FFFh
XPA = 1: 00000–3FFFFh
The SWSM bit of the SWCR defines a multiplication factor of 1 or 2 for the base number of wait states.
22
SPRS098C
December 1999 – Revised November 2001
Functional Overview
The software wait-state multiplier bit of the software wait-state control register (SWCR) is used to extend the
base number of wait states selected by the SWWSR. The SWCR bit fields are shown in Figure 3–4 and
described in Table 3–5.
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
0
Reserved
SWSM
R/W=0
R/W=0
LEGEND: R = Read, W = Write
Figure 3–4. Software Wait-State Control Register (SWCR) [MMR Address 002Bh]
Table 3–5. Software Wait-State Control Register (SWCR) Bit Fields
PIN
NO.
NAME
RESET
VALUE
15–1
Reserved
0
FUNCTION
These bits are reserved and are unaffected by writes.
Software wait-state multiplier. Used to multiply the number of wait states defined in the SWWSR by a factor
of 1 or 2.
0
SWSM
0
SWSM = 0: wait-state base values are unchanged (multiplied by 1).
SWSM = 1: wait-state base values are multiplied by 2 for a maximum of 14 wait states.
3.5.2 Programmable Bank-Switching
Programmable bank-switching can be used to insert one cycle automatically when crossing memory-bank
boundaries inside program memory or data memory space. One cycle can also be inserted when crossing
from program-memory space to data-memory space (54x) or one program memory page to another program
memory page. This extra cycle allows memory devices to release the bus before other devices start driving
the bus, thereby avoiding bus contention. The size of the memory bank for the bank-switching is defined by
the bank-switching control register (BSCR), as shown in Figure 3–5. The BSCR of a particular DSP subsystem
(A or B) is used for the external memory interface based on the xDMA/XIO arbitration logic. The BSCR bit
fields are described in Table 3–6.
15
14
13
12
11
10
9
8
BNKCMP
PS-DS
Reserved
IPIRQ
R/W
R/W
R/W
R/W
7
6
5
Reserved
4
3
2
1
0
BH
Reserved
EXIO
R/W
R/W
LEGEND: R = Read, W = Write
Figure 3–5. BSCR Register Bit Layout for Each DSP Subsystem
December 1999 – Revised November 2001
SPRS098C
23
Functional Overview
Table 3–6. BSCR Register Bit Functions for Each DSP Subsystem
BIT
NO.
BIT
NAME
RESET
VALUE
FUNCTION
15–12
BNKCMP
1111
Bank compare. BNKCMP determines the external memory-bank size. BNKCMP is used to mask the four
most significant bits (MSBs) of an address. For example, if BNKCMP = 1111b, the four MSBs (bits 12–15)
are compared, resulting in a bank size of 4K words. Bank sizes of 4K words to 64K words are allowed.
Program read – data read access. PS-DS inserts an extra cycle between consecutive accesses of
program read and data read or data read and program read.
11
PS-DS
1
PS-DS = 0
No extra cycles are inserted by this feature.
PS-DS = 1
One extra cycle is inserted between consecutive data and program reads.
10–9
Reserved
0
These bits are reserved and are unaffected by writes.
8
IPIRQ
0
The IPIRQ bit is used to send an interprocessor interrupt to the other subsystem. IPIRQ=1 sends the
interrupt. IPIRQ must be cleared before subsequent interrupts can be made. Refer to the interrupts section
for more details.
7–3
Reserved
0
These bits are reserved and are unaffected by writes.
2
BH
0
Bus holder. BH controls the bus holder feature: BH is cleared to 0 at reset.
BH = 0
The bus holder is disabled.
BH = 1
The bus holder is enabled. When not driven, PPD[15:0] pins are held at the previous logic
level.
1
Reserved
0
This bit is reserved and is unaffected by writes.
External bus interface off. The EXIO bit controls the external bus-off function.
EXIO = 0
The external bus interface functions as usual.
0
EXIO
0
EXIO = 1
The address bus, data bus, and control signals become inactive after completing the
current bus cycle. Note that the DROM, MP/MC, and OVLY bits in the PMST and the
HM bit of ST1 cannot be modified when the interface is disabled.
3.5.3 Parallel I/O Ports
The 5421 has a total of 64K words of I/O port address space. These ports can be addressed by PORTR and
PORTW. The IS signal indicates the read/write access through an I/O port. The devices can interface easily
with external devices through the I/O ports while requiring minimal off-chip address-decoding logic. The
SELA/B pin selects which subsystem is accessing the external I/O space.
24
SPRS098C
December 1999 – Revised November 2001
Functional Overview
3.6
16-Bit Bidirectional Host-Port Interface (HPI16)
The HPI16 is an enhanced 16-bit version of the TMS320C54x DSP 8-bit host-port interface (HPI). The HPI16
is designed to allow a 16-bit host to access the DSP on-chip memory, with the host acting as the master of
the interface.
3.6.1 HPI16 Memory Map
Figure 3–6 illustrates the available memory accessible by the HPI. Neither the CPU nor DMA I/O spaces can
be accessed using the host-port interface.
Hex
Page 0
Hex
00 0000
Page 1
01 0000
Hex
Reserved
00 001F
00 0020
00 005F
00 0060
McBSP
DXR/DRR
MMRegs Only
02 005F
02 0060
On-Chip
Two-Way
Shared
DARAM 0
(32K Words)
Program Only
Subsystem A
01 7FFF
01 8000
On-Chip
SARAM A
(32K Words)
Data Only
00 FFFF
Hex
Page 3
03 0000
Reserved
02 001F
02 0020
On-Chip
DARAM A
(32K Words)
Prog/Data
00 7FFF
00 8000
Page 2
02 0000
Subsystem A
Shared 0
Shared 1
On-Chip
Two-Way
Shared
DARAM 2
(32K Words)
Program Only
On-Chip
DARAM B
(32K Words)
Prog/Data
02 7FFF
02 8000
Subsystem B
03 7FFF
03 8000
On-Chip
SARAM B
(32K Words)
Data Only
On-Chip
Two-Way
Shared
DARAM 1
(32K Words)
Program Only
01 FFFF
McBSP
DXR/DRR
MMRegs Only
02 FFFF
Subsystem B
Shared 2
On-Chip
Two-Way
Shared
DARAM 3
(32K Words)
Program Only
03 FFFF
Shared 3
NOTES: A. All local memory is available to the HPI
B. The encoder maps CPU A Data Page 0 into the HPI Page 0. CPU B Data Page 0 is mapped into the HPI Page 2. Pages 1 and 3
are the on-chip shared program memory.
C. In pages 00 and 02, in the range of 0020–005F, only the following memory mapped registers are accessible: 20,21,30,31,40,41 (read
only), 22,23,32,33,42,43 (write only).
Figure 3–6. Memory Map Relative to Host-Port Interface HPI16
December 1999 – Revised November 2001
SPRS098C
25
Functional Overview
3.6.2 HPI Features
Some of the features of the HPI16 include:
•
•
•
•
•
•
•
•
•
16-bit bidirectional data bus
Multiple data strobes and control signals to allow glueless interfacing to a variety of hosts
Multiplexed and nonmultiplexed address/data modes
18-bit address bus used in nonmultiplexed mode to allow access to all internal memory (including internal
extended address pages)
18-bit address register used in multiplexed mode. Includes address autoincrement feature for faster
accesses to sequential addresses
Interface to on-chip DMA module to allow access to entire internal memory space
HRDY signal to hold off host accesses due to DMA latency
Control register available in multiplexed mode only. Accessible by either host or DSP to provide host/DSP
interrupts, extended addressing, and data prefetch capability
Maximum data rate of 33 megabytes per second (MBps) at 100-MHz DSP clock rate (no other DMA
channels active)
The HPI16 acts as a slave to a 16-bit host processor and allows access to the on-chip memory of the DSP.
There are two modes of operation as determined by the HMODE signal: multiplexed mode and nonmultiplexed
mode.
3.6.3 HPI Multiplexed Mode
In multiplexed mode, HPI16 operation is very similar to that of the standard 8-bit HPI, which is available with
other C54x DSP products. A host with a multiplexed address/data bus can access the HPI16 data register
(HPID), address register (HPIA), or control register (HPIC) via the HD bidirectional data bus. The host initiates
the access with the strobe signals (HDS1, HDS2, HCS) and controls the type of access with the HCNTL,
HR/W, and HAS signals. The DSP can interrupt the host via the HINT signal, and can stall host accesses via
the HRDY signal.
3.6.4 Host/DSP Interrupts
In multiplexed mode, the HPI16 offers the capability for the host and DSP to interrupt each other through the
HPIC register.
For host-to-DSP interrupts, the host must write a “1” to the DSPINT bit of the HPIC register. This generates
an interrupt to the DSP. This interrupt can also be used to wake the DSP from any of the IDLE 1,2, or 3 states.
Note that the DSPINT bit is always read as “0” by both the host and DSP. The DSP cannot write to this bit (see
Figure 3–7).
For DSP-to-host interrupts, the DSP must write a “1” to the HINT bit of the HPIC register to interrupt the host
via the HINT pin. The host acknowledges and clears this interrupt by also writing a “1” to the HINT bit of the
HPIC register. Note that writing a “0” to the HINT bit by either host or DSP has no effect.
3.6.5 Emulation Considerations
The HPI16 can continue operation even when the DSP CPU is halted due to debugger breakpoints or other
emulation events.
3.6.6 HPI Nonmultiplexed Mode
In nonmultiplexed mode, a host with separate address/data buses can access the HPI16 data register (HPID)
via the HD 16-bit bidirectional data bus, and the address register (HPIA) via the 18-bit HA address bus. The
host initiates the access with the strobe signals (HDS1, HDS2, HCS) and controls the direction of the access
with the HR/W signal. The HPI16 can stall host accesses via the HRDY signal. Note that the HPIC register
is not available in nonmultiplexed mode since there are no HCNTL signals available. All host accesses initiate
a DMA read or write access. Figure 3–7 shows a block diagram of the HPI16 in nonmultiplexed mode.
C54x is a trademark of Texas Instruments.
26
SPRS098C
December 1999 – Revised November 2001
Functional Overview
HOST
HD[15:0]
Data[15:0]
HPID[15:0]
Address[n:0]
HA[n :0]
R/W
Data strobes
Ready
HRDY
HR/W
HDS1, HDS2, HCS
DMA
ÎÎ
ÎÎ
ÎÎ
ÎÎ
ÎÎ
ÎÎ
ÎÎ
ÎÎ
Internal
memory
HPI-16
54x
CPU
Figure 3–7. Interfacing to the HPI-16 in Non-Multiplexed Mode
3.6.7 Other HPI16 System Considerations
3.6.7.1
Operation During IDLE
The HPI16 can continue to operate during IDLE1 or IDLE2 by using special clock management logic that turns
on relevant clocks to perform a synchronous memory access, and then turns the clocks back off to save power.
The DSP CPU does not wake up from the IDLE mode during this process.
3.6.7.2
Downloading Code During Reset
The HPI16 can download code while the DSP is in reset. However, the system provides a pin (HPIRS) that
provides a way to take the HPI16 module out of reset while leaving the DSP in reset. The maximum HPI16
data rate is 33 MBps assuming no other DMA activity (100-MIPS DSP subsystem).
3.6.7.3
Performance Issues
On the 5421, the use of SELA/B is optional for access to all on-chip memory. However, with both the 5420 and
5421 implementation using two separate subsystems (subchips), the SELA/B pin is used to select the specific
HPI16 used to access memory.
SELA/B PIN
SUBSYSTEM
0
A
1
B
Accesses to memory contained inside the same subsystem as the selected HPI16 will be faster. For accesses
to an HPI16 in a subsystem different than the memory being addressed, reads take an additional six cycles
and writes an extra five cycles. Therefore, for performance reasons, it is best to additionally decode SELA/B.
December 1999 – Revised November 2001
SPRS098C
27
Functional Overview
3.7
Multichannel Buffered Serial Port (McBSP)
The 5421 device provides high-speed, full-duplex serial ports that allow direct interface to other C54x/LC54x
devices, codecs, and other devices in a system. There are six multichannel buffered serial ports (McBSPs)
on board (three per subsystem).
The McBSP provides:
•
•
•
Full-duplex communication
Double-buffer data registers, which allow a continuous data stream
Independent framing and clocking for receive and transmit
In addition, the McBSP has the following capabilities:
•
Direct interface to:
–
–
–
–
–
•
•
•
•
•
T1/E1 framers
MVIP switching-compatible and ST-BUS compliant devices
IOM-2 compliant device
AC97-compliant device
Serial peripheral interface (SPI)
Multichannel transmit and receive of up to 128 channels
A wide selection of data sizes, including: 8, 12, 16, 20, 24, or 32 bits
µ-law and A-law companding
Programmable polarity for both frame synchronization and data clocks
Programmable internal clock and frame generation
The 5421 McBSPs have been enhanced to provide more flexibility in the choice of the sample rate generator
input clock source. On previous TMS320C5000 DSP platform devices, the McBSP sample rate input clock
can be driven from one of two possible choices: the internal CPU clock , or the external CLKS pin. However,
most C5000 DSP devices have only the internal CPU clock as a possible source because the CLKS pin is
not implemented on most device packages.
To accommodate applications that require an external reference clock for the sample rate generator, the 5421
McBSPs allow either the receive clock pin (BCLKR) or the transmit clock pin (BCLKX) to be configured as the
input clock to the sample rate generator. This enhancement is enabled through two register bits: pin control
register (PCR) bit 7 – enhanced sample clock mode (SCLKME), and sample rate generator register 2
(SRGR2) bit 13 – McBSP sample rate generator clock mode (CLKSM). SCLKME is an addition to the PCR
contained in the McBSPs on previous C5000 devices. The new bit layout of the PCR is shown in Figure 3–8.
For a description of the remaining bits, see TMS320C54x DSP Reference Set, Volume 5: Enhanced
Peripherals (literature number SPRU302).
TMS320C5000 and C5000 are trademarks of Texas Instruments.
28
SPRS098C
December 1999 – Revised November 2001
Functional Overview
15
14
13
12
11
10
9
8
Reserved
XIOEN
RIOEN
FSXM
FSRM
CLKXM
CLKRM
R,+0
RW,+0
RW,+0
RW,+0
RW,+0
RW,+0
RW,+0
7
6
5
4
3
2
1
0
SCLKME
CLKS_STAT
DX_STAT
DR_STAT
FSXP
FSRP
CLKXP
CLKRP
RW,+0
R,+0
R,+0
R,+0
RW,+0
RW,+0
RW,+0
RW,+0
Note:
R = Read, W = Write, +0 = Value at reset
Figure 3–8. Pin Control Register (PCR)
The selection of the sample rate generator (SRG) clock input source is made by the combination of the CLKSM
and SCLKME bit values as shown in Table 3–7.
Table 3–7. Sample Rate Generator Clock Source Selection
SRG Clock Source
SCLKME
CLKSM
0
0
CLKS (not available as a pin on 5421)
0
1
CPU clock
1
0
BCLKR pin
1
1
BCLKX pin
When either of the bidirectional pins, BCLKR or BCLKX, is configured as the clock input, its output buffer is
automatically disabled. For example, with SCLKME = 1 and CLKSM = 0, the BCLKR pin is configured as the
SRG input. In this case, both the transmitter and receiver circuits can be synchronized to the SRG output by
setting the PCR bits (9:8) for CLKXM = 1 and CLKRM = 1. However, the SRG output is only driven onto the
BCLKX pin because the BCLKR output is automatically disabled.
The McBSP supports independent selection of multiple channels for the transmitter and receiver. When
multiple channels are selected, each frame represents a time-division multiplexed (TDM) data stream. In using
time-division multiplexed data streams, the CPU may only need to process a few of them. Thus, to save
memory and bus bandwidth, multichannel selection allows independent enabling of particular channels for
transmission and reception. Up to a maximum of 128 channels in a bit stream can be enabled or disabled.
The 5421 McBSPs have two working modes that are selected by setting the RMCME and XMCME bits in the
multichannel control registers (MCR1x and MCR2x, respectively). See Figure 3–9 and Figure 3–10. For a
description of the remaining bits, see TMS320C54x DSP Reference Set, Volume 5: Enhanced Peripherals
(literature number SPRU302).
•
15
Note:
14
In the first mode, when RMCME = 0 and XMCME = 0, there are two partitions (A and B), with each
containing 16 channels as shown in Figure 3–9 and Figure 3–10. This is compatible with the McBSPs
used in the 5420, where only 32-channel selection is enabled (default).
13
12
11
10
9
8
7
6
5
4
3
2
1
0
Reserved
XMCME
XPBBLK
XPABLK
XCBLK
XMCM
R,+0
RW,+0
RW,+0
RW,+0
R,+0
RW,+0
R = Read, W = Write, +0 = Value at reset; x = McBSP 0,1, or 2
Figure 3–9. Multichannel Control Register 2x (MCR2x)
December 1999 – Revised November 2001
SPRS098C
29
Functional Overview
15
14
Note:
13
12
11
10
9
8
7
6
5
4
3
2
1
0
Reserved
RMCME
RPBBLK
RPABLK
RCBLK
RMCM
R,+0
RW,+0
RW,+0
RW,+0
R,+0
RW,+0
R = Read, W = Write, +0 = Value at reset; x = McBSP 0,1, or 2
Figure 3–10. Multichannel Control Register 1x (MCR1x)
•
In the second mode, with RMCME = 1 and XMCME = 1, the McBSPs have 128 channel selection
capability. Twelve new registers (RCERCx–RCERHx and XCERCx–XCERHx) are used to enable the
128 channel selection. The subaddresses of the new registers are shown in Table 3–21. These new
registers, functionally equivalent to the RCERA0–RCERB1 and XCERA0–XCERB1 registers in the 5420,
are used to enable/disable the transmit and receive of additional channel partitions (C,D,E,F,G, and H)
in the 128 channel stream. For example, XCERH1 is the transmit enable for channel partition H (channels
112 to 127) of MCBSP1 for each DSP subsystem. See Figure 3–11, Table 3–8, Figure 3–12, and
Table 3–9 for bit layout and function of the receive and transmit registers .
15
14
13
12
11
10
9
8
RCERyz15
RCERyz14
RCERyz13
RCERyz12
RCERyz11
RCERyz10
RCERyz9
RCERyz8
RW,+0
RW,+0
RW,+0
RW,+0
RW,+0
RW,+0
RW,+0
RW,+0
7
6
5
4
3
2
1
0
RCERyz7
RCERyz6
RCERyz5
RCERy4
RCERyz3
RCERyz2
RCERyz1
RCERyz0
RW,+0
RW,+0
RW,+0
RW,+0
RW,+0
RW,+0
RW,+0
RW,+0
Note:
R = Read, W = Write, +0 = Value at reset; y = Partition A,B,C,D,E,F,G, or H; z = McBSP 0,1, or 2
Figure 3–11. Receive Channel Enable Registers Bit Layout for Partitions A to H
Table 3–8. Receive Channel Enable Registers for Partitions A to H
Bit
15–0
Note:
Name
Function
RCERyz(15:0)
Receive Channel Enable Register
RCERyz n = 0
Disables reception of nth channel in partition y.
RCERyz n = 1
Enables reception of nth channel in partition y.
y = Partition A,B,C,D,E,F,G, or H; z = McBSP 0,1, or 2; n = bit 15–0
15
14
13
12
11
10
9
8
XCERyz15
XCERyz14
XCERyz13
XCERyz12
XCERyz11
XCERyz10
XCERyz9
XCERyz8
RW,+0
RW,+0
RW,+0
RW,+0
RW,+0
RW,+0
RW,+0
RW,+0
7
6
5
4
3
2
1
0
XCERyz7
XCERyz6
XCERyz5
XCERy4
XCERyz3
XCERyz2
XCERyz1
XCERyz0
RW,+0
RW,+0
RW,+0
RW,+0
RW,+0
RW,+0
RW,+0
RW,+0
Note:
R = Read, W = Write, +0 = Value at reset; y = Partition A,B,C,D,E,F,G, or H; z = McBSP 0,1, or 2
Figure 3–12. Transmit Channel Enable Registers Bit Layout for Partitions A to H
30
SPRS098C
December 1999 – Revised November 2001
Functional Overview
Table 3–9. Transmit Channel Enable Registers for Partitions A to H
Bit
15–0
Note:
Name
Function
XCERyz(15:0)
Transmit Channel Enable Register
XCERyz n = 0
Disables transmit of nth channel in partition y.
XCERyz n = 1
Enables transmit of nth channel in partition y.
y = Partition A,B,C,D,E,F,G, or H; z = McBSP 0,1, or 2; n = bit 15–0
The clock stop mode (CLKSTP) in the McBSP provides compatibility with the serial port interface (SPI)
protocol. Clock stop mode works with only single-phase frames and one word per frame. The word sizes
supported by the McBSP are programmable for 8-, 12-, 16-, 20-, 24-, or 32-bit operation. When the McBSP
is configured to operate in SPI mode, both the transmitter and the receiver operate together as a master or
as a slave.
The McBSP is fully static and operates at arbitrarily low clock frequencies. The maximum McBSP multichannel
operating frequency on the 5421 is 9 MBps. Nonmultichannel operation is limited to 38 MBps.
3.7.1 Emulation Considerations
The McBSP can continue operation even when the DSP CPU is halted due to debugger breakpoints or other
emulation events.
December 1999 – Revised November 2001
SPRS098C
31
Functional Overview
3.8
Direct Memory Access (DMA) Controller
The 5421 includes two 6-channel direct memory access (DMA) controllers for performing data transfers
independent of the CPU, one for each subsystem. The DMA controller controls accesses to off-chip
program/data/IO and internal data/program memory. The primary function of the 5421 DMA controller is to
provide code overlays and manage data transfers between on-chip memory, the peripherals, and off-chip
memory.
In the background of CPU operation, the 5421 DMA allows movement of data between internal and external
program/data memory, and internal peripherals, such as the McBSPs and the HPI. Each subsystem has its
own independent DMA with six programmable channels, which allows for six different contexts for DMA
operation. The HPI has a dedicated auxiliary DMA channel. Figure 3–13 illustrates the memory map
accessible by the DMA.
Hex
00 0000
‡
Data†
Reserved
00 001F
00 0020
00 005F
00 0060
McBSP
DXR/DRR
MMRegs Only
Program Page 0
Hex
00 0000
Reserved
00 001F
00 0020
00 005F
00 0060
00 7FFF
00 8000
00 FFFF
On-Chip
Two-Way
Shared
DARAM 0
(32K Words)
Program
Only
Subsystem A
01 7FFF
01 8000
On-Chip
SARAM A
(32K Words)
Data Only
On-Chip
SARAM A/B
(32K Words)
Data Only
00 FFFF
Program Page 1‡ Hex
McBSP
DXR/DRR
MMRegs Only
On-Chip
DARAM A
(32K Words)
Prog/Data
On-Chip
DARAM A/B
(32K Words)
Prog/Data
00 7FFF
00 8000
Hex
01 0000
Subsystem A
Shared 0
Program Page 2‡ Hex Program Page 3‡
02 0000
03 0000
Reserved
02 001F
02 0020
McBSP
DXR/DRR
02 005F MMRegs Only
On-Chip
02 0060
Two-Way
Shared
On-Chip
DARAM 2
DARAM B
(32K Words)
(32K Words)
Program
Prog/Data
Only
02 7FFF Subsystem B
02 8000
On-Chip
SARAM B
(32K Words)
Data Only
On-Chip
Two-Way
Shared
DARAM 1
(32K Words)
Program
Only
01 FFFF
Shared 1
03 7FFF
03 8000
02 FFFF
Subsystem B
Shared 2
On-Chip
Two-Way
Shared
DARAM 3
(32K Words)
Program
Only
03 FFFF
Shared 3
†
DMD/DMS = 01
DMD/DMS = 00
NOTES: A. All local memory is available to the DMA.
B. All I/O memory accesses by the DMA (DMD/DMS = 10) are mapped to the core-to-core FIFO.
C. In pages 00 and 02, in the range of 0020–005F, only the following memory mapped registers are accessible: 20,21,30,31,40,41
(read only), 22,23,32,33,42,43 (write only).
‡
Figure 3–13. On-Chip Memory Map Relative to DMA (DLAXS/SLAXS = 0)
32
SPRS098C
December 1999 – Revised November 2001
Functional Overview
xDMA External Program Memory Map†
00 0000
01 0000
02 0000
Page 1
Lower
32K
External
Page 0
Lower
32K
External
...
03 0000
Page 2
Lower
32K
External
07 0000
Page 3
Lower
32K
External
Page7
Lower
32K
External
07 7FFF
00 7FFF
01 7FFF
02 7FFF
00 8000
01 8000
02 8000
Page 2
Upper
32K
External
Page 1
Upper
32K
External
Page 0
Upper
32K
External
03 7FFF
03 8000
...
...
07 8000
Page 3
Upper
32K
External
Page 7
Upper
32K
External
...
00 FFFF
03 FFFF
02 FFFF
01 FFFF
07 FFFF
xDMA External Data Memory Map†
01 0000
00 0000
02 0000
03 0000
07 0000
..
Page 1
Lower
32K
External
Page 0
Lower
32K
External
Page 2
Lower
32K
External
Page 3
Lower
32K
External
00 7FFF
01 7FFF
02 7FFF
03 7FFF
00 8000
01 8000
02 8000
03 8000
Page 2
Upper
32K
External
Page 1
Upper
32K
External
Page 0
Upper
32K
External
Page 7
Lower
32K
External
...
...
07 7FFF
07 8000
Page 3
Upper
32K
External
Page 7
Upper
32K
External
...
00 FFFF
†
02 FFFF
01 FFFF
03 FFFF
07 FFFF
Pages 8 – 127 are overlaid over pages 0 – 7.
Figure 3–14. DMA External Program Memory Map
December 1999 – Revised November 2001
SPRS098C
33
Functional Overview
3.8.1 DMA Controller Features
The 5421 DMA has the following features:
•
•
•
•
•
•
•
•
•
•
•
•
The DMA operates independently of the CPU.
The DMA has six channels. The DMA can keep track of the contexts of six independent block transfers.
Two DMA channels are available for external accesses: one for reads and one for writes.
The DMA has higher priority than the CPU for internal accesses.
Each channel has independently programmable priorities.
Each channel’s source and destination address registers include configurable indexing modes. The
address can be held constant, postincremented, postdecremented, or adjusted by a programmable
value.
For internal accesses, each read or write transfer can be initialized by selected events.
Supports 32-bit transfers for internal accesses only.
Single-word (16-bit) transfers are supported for external accesses.
The DMA does not support transfers from peripherals to external memory.
The DMA does not support transfers from external memory to the peripherals.
The DMA does not support external to external transfers.
A 16-bit DMA transfer requires four CPU clock cycles to complete — two cycles for reads and two cycles for
writes. This gives a maximum DMA throughput of 50 MBps. Since the DMA controller shares the DMA bus
with the HPI module, the DMA access rate is reduced when the HPI is active.
3.8.2 DMA Accesses to External Memory
The 5421 DMAs supports external accesses to extended program, extended data, and extended I/O memory.
These overlay pages are only visible to the DMA controller. A maximum of two channels (one for reads, one
for writes) per DMA can be used for external memory accesses. The DMA external accesses require 9 cycles
(minimum) for external writes and 13 cycles (minimum) for external reads.
The control of the bus is arbitrated between the two CPUs and the two DMAs. While one DMA or CPU is in
control of the external bus, the other three components will be held off (via wait-states) until the current transfer
is complete. The DMA takes precedence over XIO requests. The HOLD/HOLDA feature of the 5421 affects
external CPU transfers, as well as external DMA transfers. When an external processor asserts the HOLD
pin to gain control of the memory interface, the HOLDA signal is not asserted until all pending DMA transfers
are completed. To prevent a DMA from blocking out the CPUs or HOLD/HOLDA feature from accessing the
external bus, uninterrupted burst transfers are not supported by the DMAs. Subsequently, CPU and DMA
arbitration testing is performed for each external bus cycle, regardless of the bus activity. With the completion
of each block, the highest priority will be swapped.
For arbitration at the DSP subsystem level, the DMA requests (DMA_REQ_A or DMA_REQ_B) from either
DMA will be sent to both CPUs as shown in Figure 3–15. Regardless of which CPU controls the external pin
interface (XIO), both CPUs must send a grant (GRANT_A, GRANT_B) for control of the bus to be released
to the DMAs.
Arbitration between CPUs is done using a request/grant scheme. Prior to accessing XIO of one of the CPUs,
software is responsible for asserting a request for access to the device pins and polling grant status until the
pins are granted to the requestor. If both CPUs request the bus simultaneously, subsystem A is granted priority.
For details on memory-mapped register bits pertaining to CPU XIO arbitration, see the general-purpose I/O
control register bits [6:4] (CORE SEL, XIO GRANT, XIO REQ) in Table 3–14.
At reset, the default is that subsystem A has access to the device pins. Accesses without a grant will be
allowed, but do not show up on the device pins.
34
SPRS098C
December 1999 – Revised November 2001
Functional Overview
DMA_REQUEST
DMA_REQUEST
54x cLEAD CPU
(DSP Subsystem B)
54x cLEAD CPU
(DSP Subsystem B)
GRANT_B
GRANT_A
XIO
Grant
XIO
Req
Core
SEL=0
GPIO Control
Register
(DSP Subsystem B)
GPIO Control
Register
(DSP Subsystem A)
XIO
Grant
XIO
Req
Core
SEL=1
VCC
CPU_ARB
EMIF
Controller
DMA_REQ_A
DMA_REQ_B
XDMA
(DSP
Subsystem A)
XDMA
(DSP
Subsystem B)
DMA_ARB
DMA_GRANT_A
DMA_GRANT_B
GRANT
XDMA
XDMA
XCPU
XCPU
SEL
XHOLDA
XHOLD
XIO
Figure 3–15. Arbitration Between XIO and xDMA for External Access
The HM bit in the ST1 indicates whether the processor continues internal execution when acknowledging an
active HOLD signal.
•
•
HM = 0, the processor continues execution from internal program memory but places its external interface
in the high-impedance state.
When HM = 1, the processor halts internal execution.
To ensure that proper arbitration occurs, the HM bit should be set to 0 in the memory-mapped ST1 registers
for both CPUs.
To allow the DMA access to extended data pages, the SLAXS and DLAXS bits are added to the DMMCRn
registers. For a description of the remaining bits, see TMS320C54x DSP Reference Set, Volume 5: Enhanced
Peripherals (literature number SPRU302).
15
14
13
AUTO
INIT
DINM
IMOD
12
CT
MOD
11
SLAXS
10
9
SIND
8
7
6
DMS
5
DLAXS
4
3
DIND
2
1
0
DMD
Figure 3–16. DMA Transfer Mode Control Register (DMMCRn)
December 1999 – Revised November 2001
SPRS098C
35
Functional Overview
These new bit fields were created to allow the user to define the space-select for the DMA (internal/external).
Also, a new extended destination data page (XDSTDP[6:0], subaddress 029h) and extended source data
page (XSRCDP[6:0], subaddress 028h) have been created.
DLAXS(DMMCRn[5])
Destination
0 = No external access (default internal)
SLAXS(DMMCRn[11])
Source
0 = No external access (default internal)
1 = External access
1 = External access
For the CPU external access, software can configure the memory cells to reside inside or outside the program
address map. When the cells are mapped into program space, the device automatically accesses them when
their addresses are within bounds. When the program address generation (PAGEN) logic generates an
address outside its bounds, the device automatically generates an external access. All DMA I/O space
accesses are mapped to the core-to-core FIFO.
3.8.3 DMA Controller Synchronization Events
The transfers associated with each DMA channel can be synchronized to one of several events. The DSYN
bit field of the DMA channel x sync select and frame count (DMSFCx) register selects the synchronization
event for a channel. The list of possible events and the DSYN values are shown in Table 3–10.
Table 3–10. DMA Synchronization Events
DSYN VALUE
DMA SYNCHRONIZATION EVENT
0000b
No synchronization used
0001b
McBSP0 Receive Event
0010b
McBSP0 Transmit Event
0011b
McBSP2 Receive Event
0100b
McBSP2 Transmit Event
0101b
McBSP1 Receive Event
0110b
McBSP1 Transmit Event
0111b
FIFO Receive Buffer Not Empty Event
1000b
FIFO Transmit Buffer Not Full Event
1001b – 1111b
Reserved
3.8.4 DMA Channel Interrupt Selection
The DMA controller can generate a CPU interrupt for each of the six channels. However, channels 0, 1, 2,
and 3 are multiplexed with other interrupt sources. DMA channels 0 and 1 share an interrupt line with the
receive and transmit portions of McBSP2 (IMR/IFR bits 6 and 7), and DMA channels 2 and 3 share an interrupt
line with the receive and transmit portions of McBSP1 (IMR/IFR bits 10 and 11). When the 5421 is reset, the
interrupts from these four DMA channels are deselected. The INTSEL bit field in the DMA channel priority and
enable control (DMPREC) register can be used to select these interrupts, as shown in Table 3–11.
Table 3–11. DMA Channel Interrupt Selection
INTSEL Value
IMR/IFR[6]
IMR/IFR[7]
IMR/IFR[10]
IMR/IFR[11]
00b (reset)
BRINT2
BXINT2
BRINT1
BXINT1
01b
BRINT2
BXINT2
DMAC2
DMAC3
10b
DMAC0
DMAC1
DMAC2
DMAC3
11b
36
SPRS098C
Reserved
December 1999 – Revised November 2001
Functional Overview
3.8.5 DMA in Autoinitialization Mode
The DMA can automatically reinitialize itself after completion of a block transfer. Some of the DMA registers
can be preloaded for the next block transfer through the DMA global reload registers (DMGSA, DMGDA,
DMGCR, and DMGFR). Autoinitialization allows:
•
•
Continuous operation: Normally, the CPU would have to reinitialize the DMA immediately after the
completion of the current block transfers, but with the global reload registers, it can reinitialize these values
for the next block transfer any time after the current block transfer begins.
Repetitive operation: The CPU does not preload the global reload register with new values for each block
transfer but only loads them on the first block transfer.
The 5421 DMA has been enhanced to expand the DMA global reload register sets. Each DMA channel now
has its own DMA global reload register set. For example, the DMA global reload register set for channel 0 has
DMGSA0, DMGDA0, DMGCR0, and DMGFR0 while DMA channel 1 has DMGSA1, DMGDA1, DMGCR1,
and DMGFR1, etc.
To utilize the additional DMA global reload registers, the AUTOIX bit is added to the DMPREC register as
shown in Figure 3–17.
15
14
13
FREE
AUTOIX
12
11
10
9
DPRC[5:0]
8
7
6
5
4
3
IOSEL
2
1
0
DE[5:0]
Figure 3–17. DMPREC Register
Table 3–12. DMA Global Reload Register Selection
AUTOIX
DMA GLOBAL RELOAD REGISTER USAGE IN AUTO INIT MODE
0 (default)
All DMA channels use DMGSA0, DMGDA0, DMGCR0 and DMGFR0
1
Each DMA channel uses its own set of global reload registers
3.8.6 Subsystem Communications
The 5421 device provides two options for efficient core-to-core communications:
•
•
3.8.6.1
Core-to-core FIFO communications
DMA global memory transfer
FIFO Data Communications
The subsystems’ FIFO communications interface is shown in the 5421 functional block diagram (Figure 3–1).
Two unidirectional 8-word-deep FIFOs are available in the device for efficient interprocessor communication:
one configured for core A-to-core B data transfers, and the other configured for core B-to-core A data transfers.
Each subsystem, by way of DMA control, can write to its respective output data FIFO and read from its
respective input data FIFO. The FIFOs are accessed using the DMAs I/O space, which is completely
independent of the CPU I/O space. The DMA transfers to or from the FIFOs can be synchronized to “receive
FIFO not empty” and “transmit FIFO not full” events, providing protection from overflow and underflow.
Subsystems can interrupt each other to flag when the FIFOs are either full or empty. The interprocessor
interrupt request bit (IPIRQ) (bit 8 in the BSCR register (BSCR.8)) is set to 1 to generate a PINT in the other
subsystem’s IFR.14. See the Interrupts section (Section 3.13) for more information.
3.8.6.2
DMA Global Memory Transfers
The 5421 enables each subsystem to transfer data directly between the memories that are CPU local via DMA
global memory transfers. The DMA global memory map is shown in Figure 3–13.
December 1999 – Revised November 2001
SPRS098C
37
Functional Overview
3.8.7 Chip Subsystem ID Register
The chip subsystem ID Register (CSIDR) is a read-only memory-mapped register located at 3Eh within each
DSP subsystem. This register contains three elements for electrically readable device identification. The
ChipID bits identify the type of 54x device (21h for 5421). The ChipRev bits contain the revision number of
the device. Lastly, the SubSysID contains a unique subsystem identifier.
15
14
13
12
11
10
9
8
7
Chip ID
6
5
4
3
Chip Rev
2
1
0
SubSysID
Figure 3–18. Chip Subsystem ID Register
Table 3–13. Chip Subsystem ID Register Bit Functions
BIT
NO.
BIT FIELD
NAME
15–8
Chip ID
FUNCTION
54x device type. Contains 21h for 5421.
7–4
Chip Rev
Revision number of device (i.e., 0h for revision 0).
3–0
SubSysID
Identifier for DSP subsystem: A = 0h, B = 1h.
3.9
General-Purpose I/O
In addition to the A_XF and B_XF pins, the 5421 has eight general-purpose I/O pins. These pins are:
A_GPIO0, A_GPIO1, A_GPIO2, A_GPIO3
B_GPIO0, B_GPIO1, B_GPIO2, B_GPIO3
Four general-purpose I/O pins are available to each core. Each GPIO pin can be individually selected as either
an input or an output. Additionally, the timer output is selectable on GPIO pin 3. At core reset, all GPIO pins
are configured as inputs. GPIO data and control bits are accessible through a memory-mapped register at
3Ch with the format shown in Figure 3–19.
15
TOUT
14
13
Reserved
R/W+0
12
11
10
9
8
7
6
5
4
3
2
1
0
GPIO
DIR3
GPIO
DIR2
GPIO
DIR1
GPIO
DIR0
ROM
EN
CORE
SEL
XIO
XIO
GRANT
REQ
GPIO
DAT3
GPIO
DAT2
GPIO
DAT1
GPIO
DAT0
R/W+0
R/W+0
R/W+0
R/W+0
R/W†
R
R+0
R/W+0
R/W+0
R/W+0
R/W+0
R/W+0
† 1 denotes XIO = 1, 0 denotes XIO = 0
Note: R = Read, W = Write, +0 = Value at reset
Figure 3–19. General-Purpose I/O Control Register
38
SPRS098C
December 1999 – Revised November 2001
Functional Overview
Table 3–14. General-Purpose I/O Control Register Bit Functions
BIT
NO.
BIT
NAME
15
TOUT
14-12
Reserved
GPIO
0
GPIOn pin is used as an input.
DIRn†
1
GPIOn pin is used as an output.
0
ROM is mapped out (value at reset if XIO = 0)
11–8
7
6
5
4
3–0
ROMEN‡
BIT
VALUE
FUNCTION
0
Timer output disable. Uses GPIO3 as general-purpose I/O.
1
Timer output enable. Overrides DIR3. Timer output is driven on GPIO3 and readable in DAT3.
X
Register bit is reserved. Read 0, write has no effect.
1
ROM is mapped in (value at reset if XIO = 1)
CORE
0
cLEAD core A is selected for XIO REQ bit. DSP subsystem A is tied low internally for this bit.
SEL
1
cLEAD core B is selected for XIO REQ bit. DSP subsystem B is tied high internally for this bit.
XIO
GRANT
0
EMIF is not available to the cLEAD core determined by the CORE SEL bit.
1
EMIF is granted to the cLEAD core determined by the CORE SEL bit.
XIO
0
EMIF is not requested for the cLEAD core indicated by the CORE SEL bit.
REQ
1
Request EMIF for the cLEAD core indicated by the CORE SEL bit.
GPIO
DATn†
0
GPIOn is driven with a 0 (DIRn = 1). GPIOn is read as 0 (DIRn = 0).
1
GPIOn is driven with a 1 (DIRn = 1). GPIOn is read as 1 (DIRn = 0).
†
n = 3, 2, 1, or 0
‡ 1 denotes XIO = 1, 0 denotes XIO = 0
Register bit 7 is used as ROMEN to enable and disable ROM space. In XIO mode, ROM enable (ROMEN)
reflects the state of the A_GPIO0 and B_GPIO0 pins (GPIODAT0 input) to enable the applicable on-chip ROM
after reset. Register bits (6:4) are used for XIO arbitration of external memory interface (EMIF) control between
DSP subsystems. The timer out (TOUT) bit is used to multiplex the output of the timer and GPIO3. All GPIO
pins are programmable as an input or output by the direction bit (DIRn). Data is either driven or read from the
data bit field (DATn). DIR3 has no affect when TOUT = 1.
GPIO2 is a special case where the logic level determines the operation of BIO-conditional instructions on the
CPU. GPIO2 is always mapped as a general-purpose I/O, but the BIO function exists when this pin is
configured as an input.
3.9.1 Hardware Timer
The 54x devices feature a 16-bit timing circuit with a 4-bit prescaler. The timer counter decrements by one at
every CLKOUT cycle. Each time the counter decrements to zero, a timer interrupt is generated. The timer can
be stopped, restarted, reset, or disabled by specific status bits. The timer output pulse is driven on GPIO3
when the TOUT bit is set to one in the general-purpose I/O control register. The device must be in HPI mode
(XIO = 0) to drive TOUT on the GPIO3 pin.
3.9.2 Software-Programmable Phase-Locked Loop (PLL)
The clock generator provides clocks to the 5421 device, and consists of a phase-locked loop (PLL) circuit. The
clock generator requires a reference clock input, which must be provided by using an external clock source.
The reference clock input is then divided by two (DIV mode) to generate clocks for the 5421 device. Alternately,
the PLL circuit can be used (PLL mode) to generate the device clock by multiplying the reference clock
frequency by a scale factor, allowing use of a clock source with a lower frequency than that of the CPU. Bypass
(multiply by 1) is the default mode at reset. The PLL is an adaptive circuit that, once synchronized, locks onto
and tracks an input clock signal. When the PLL is initially started, it enters a transitional mode during which
the PLL acquires lock with the input signal. Once the PLL is locked, it continues to track and maintain
synchronization with the input signal. Then, other internal clock circuitry allows the synthesis of new clock
frequencies for use as master clock for the 5421 device. Only subsystem A controls the PLL. Subsystem B
cannot access the PLL registers.
December 1999 – Revised November 2001
SPRS098C
39
Functional Overview
The software-programmable PLL features a high level of flexibility, and includes a clock scaler that provides
various clock multiplier ratios, capability to directly enable and disable the PLL, and a PLL lock timer that can
be used to delay switching to PLL clocking mode of the device until lock is achieved. Devices that have a
built-in software-programmable PLL can be configured in one of two clock modes:
•
PLL mode. The input clock (CLKIN) is multiplied by 1 of 31 possible ratios. These ratios are achieved using
the PLL circuitry.
•
DIV (divider) mode. The input clock is divided by 2 or 4. Note that when DIV mode is used, the PLL can
be completely disabled in order to minimize power dissipation.
The software-programmable PLL is controlled using the 16-bit memory-mapped (address 0058h) clock mode
register (CLKMD). The CLKMD register is used to define the clock configuration of the PLL clock module.
Figure 3–20 shows the bit layout of the clock mode register and Table 3–15 describes the bit functions.
15
12
11
10
3
2
1
0
PLLMUL†
PLLDIV†
PLLCOUNT†
PLLON/OFF†
PLLNDIV
STATUS
R/W
R/W
R/W
R/W
R/W
R/W
† When in DIV mode (PLLSTATUS is low), PLLMUL, PLLDIV, PLLCOUNT, and PLLON/OFF are don’t cares, and their contents are indeterminate.
LEGEND: R = Read, W = Write
Figure 3–20. Clock Mode Register (CLKMD)
Table 3–15. Clock Mode Register (CLKMD) Bit Functions
BIT
NO.
BIT
NAME
15–12
PLLMUL†
11
PLLDIV†
FUNCTION
PLL multiplier. PLLMUL defines the frequency multiplier in conjunction with PLLDIV and PLLNDIV. See
Table 3–16.
PLL divider. PLLDIV defines the frequency multiplier in conjunction with PLLMUL and PLLNDIV. See Table 3–16.
10–3
PLLCOUNT†
PLLDIV = 0
Means that an integer multiply factor is used
PLLDIV = 1
Means that a noninteger multiply factor is used
PLL counter value. PLLCOUNT specifies the number of input clock cycles (in increments of 16 cycles) for the
PLL lock timer to count before the PLL begins clocking the processor after the PLL is started. The PLL counter
is a down-counter, which is driven by the input clock divided by 16; therefore, for every 16 input clocks, the PLL
counter decrements by one.
The PLL counter can be used to ensure that the processor is not clocked until the PLL is locked, so that only valid
clock signals are sent to the device.
2
PLLON/OFF†
PLL on/off. PLLON/OFF enables or disables the PLL part of the clock generator in conjunction with the PLLNDIV
bit (see Table 3–17). Note that PLLON/OFF and PLLNDIV can both force the PLL to run; when PLLON/OFF is
high, the PLL runs independently of the state of PLLNDIV.
1
PLLNDIV
PLLNDIV configures PLL mode when high or DIV mode when low. PLLNDIV defines the frequency multiplier in
conjunction with PLLDIV and PLLMUL. See Table 3–16.
Indicates the PLL mode.
0
STATUS
STATUS = 0
Indicates DIV mode
STATUS = 1 Indicates PLL mode
† When in DIV mode (PLLSTATUS is low), PLLMUL, PLLDIV, PLLCOUNT, and PLLON/OFF are don’t cares, and their contents are indeterminate.
40
SPRS098C
December 1999 – Revised November 2001
Functional Overview
Table 3–16. Multiplier Related to PLLNDIV, PLLDIV, and PLLMUL
MULTIPLIER†
PLLNDIV
PLLDIV
PLLMUL
0
x
0–14
0.5
0
x
15
0.25
1
0
0–14
PLLMUL + 1
1
0
15
bypass (multiply by 1)‡
1
1
0 or even
(PLLMUL + 1)/2
1
1
odd
PLLMUL/4
†
CLKOUT = CLKIN * Multiplier
‡ Indicates the default clock mode after reset
Table 3–17. VCO Truth Table
PLLON/OFF
PLLNDIV
0
0
VCO STATE
off
1
0
on
0
1
on
1
1
on
3.9.3 PLL Clock Programmable Timer
During the lockup period, the PLL should not be used to clock the 5421. The PLLCOUNT programmable lock
timer provides a convenient method of automatically delaying clocking of the device by the PLL until lock is
achieved.
The PLL lock timer is a counter, loaded from the PLLCOUNT field in the CLKMD register, that decrements from
its preset value to 0. The timer can be preset to any value from 0 to 255, and its input clock is CLKIN divided
by 16. The resulting lockup delay can therefore be set from 0 to 255 × 16 CLKIN cycles.
The lock timer is activated when the operating mode of the clock generator is switched from DIV to PLL. During
the lockup period, the clock generator continues to operate in DIV mode; after the PLL lock timer decrements
to zero, the PLL begins clocking the 5421.
Accordingly, the value loaded into PLLCOUNT is chosen based on the following formula:
PLLCOUNT +
Lockup Time
16
T CLKIN
where TCLKIN is the input reference clock period and lockup time is the required VCO lockup time, as shown
in Table 3–18.
Table 3–18. VCO Lockup Time
CLKOUT FREQUENCY (MHz)
LOCKUP TIME (µs)
December 1999 – Revised November 2001
5
23
10
17
20
16
40
19
60
24
80
29
100
35
SPRS098C
41
Functional Overview
3.10 Memory-Mapped Registers
The 5421 has 27 memory-mapped CPU registers, which are mapped in data memory space address 0h to
1Fh. Each 5421 device also has a set of memory-mapped registers associated with peripherals. Table 3–19
gives a list of CPU memory-mapped registers (MMRs) available. Table 3–20 shows additional peripheral
MMRs associated with the 5421.
Table 3–19. Processor Memory-Mapped Registers for Each DSP Subsystem
NAME
IMR
IFR
0
DESCRIPTION
0
Interrupt Mask Register
Interrupt Flag Register
1
1
2–5
2–5
ST0
6
6
Status Register 0
ST1
7
7
Status Register 1
AL
8
8
Accumulator A Low Word (15–0)
AH
9
9
Accumulator A High Word (31–16)
AG
10
A
Accumulator A Guard Bits (39–32)
BL
11
B
Accumulator B Low Word (15–0)
BH
12
C
Accumulator B High Word (31–16)
BG
13
D
Accumulator B Guard Bits (39–32)
TREG
14
E
Temporary Register
TRN
15
F
Transition Register
AR0
16
10
Auxiliary Register 0
AR1
17
11
Auxiliary Register 1
AR2
18
12
Auxiliary Register 2
AR3
19
13
Auxiliary Register 3
AR4
20
14
Auxiliary Register 4
AR5
21
15
Auxiliary Register 5
AR6
22
16
Auxiliary Register 6
AR7
23
17
Auxiliary Register 7
SP
24
18
Stack Pointer
BK
25
19
Circular Buffer Size Register
BRC
26
1A
Block-Repeat Counter
RSA
27
1B
Block-Repeat Start Address
REA
28
1C
Block-Repeat End Address
PMST
29
1D
Processor Mode Status Register
XPC
30
1E
Extended Program Counter
—
31
1F
Reserved
—
42
ADDRESS
DEC
HEX
SPRS098C
Reserved for testing
December 1999 – Revised November 2001
Functional Overview
Table 3–20. Peripheral Memory-Mapped Registers for Each DSP Subsystem
NAME
32
20
McBSP 0 Data Receive Register 2
DRR10
33
21
McBSP 0 Data Receive Register 1
DXR20
34
22
McBSP 0 Data Transmit Register 2
DXR10
35
23
McBSP 0 Data Transmit Register 1
TIM
36
24
Timer Register
PRD
37
25
Timer Period Register
TCR
38
26
Timer Control Register
—
39
27
Reserved
SWWSR
40
28
Software Wait-State Register
BSCR
41
29
Bank-Switching Control Register
—
42
2A
Reserved
SWCR
43
2B
Software Wait-State Control Register
HPIC
44
2C
HPI Control Register (HMODE=0 only)
45–47
2D–2F
DRR22
48
30
McBSP 2 Data Receive Register 2
DRR12
49
31
McBSP 2 Data Receive Register 1
DXR22
50
32
McBSP 2 Data Transmit Register 2
DXR12
51
33
McBSP 2 Data Transmit Register 1
SPSA2
52
34
McBSP 2 Subbank Address Register†
SPSD2
53
35
McBSP 2 Subbank Data Register†
—
Reserved
54–55
36–37
SPSA0
56
38
McBSP 0 Subbank Address Register†
SPSD0
57
39
McBSP 0 Subbank Data Register†
—
Reserved
58–59
3A–3B
GPIO
60
3C
General-Purpose I/O Register
—
61
3D
Reserved
CSIDR
62
3E
Chip Subsystem ID register
—
63
3F
Reserved
DRR21
64
40
McBSP 1 Data Receive Register 2
DRR11
65
41
McBSP 1 Data Receive Register 1
DXR21
66
42
McBSP 1 Data Transmit Register 2
DXR11
67
43
McBSP 1 Data Transmit Register 1
68–71
44–47
SPSA1
72
48
McBSP 1 Subbank Address Register†
SPSD1
73
49
McBSP 1 Subbank Data Register†
—
—
Reserved
Reserved
74–83
4A–53
DMPREC
84
54
DMA Priority and Enable Control Register
DMSA
85
55
DMA Subbank Address Register‡
DMSDI
86
56
DMA Subbank Data Register with Autoincrement‡
DMSDN
87
57
DMA Subbank Data Register‡
88
58
Clock Mode Register (CLKMD)
89–95
59–5F
CLKMD
—
‡
DESCRIPTION
DRR20
—
†
ADDRESS
DEC
HEX
Reserved
Reserved
See Table 3–21 for a detailed description of the McBSP control registers and their subaddresses.
See Table 3–22 for a detailed description of the DMA subbank addressed registers.
December 1999 – Revised November 2001
SPRS098C
43
Functional Overview
3.11 McBSP Control Registers and Subaddresses
The control registers for the multichannel buffered serial port (McBSP) are accessed using the subbank
addressing scheme. This allows a set or subbank of registers to be accessed through a single memory
location. The McBSP subbank address register (SPSA) is used as a pointer to select a particular register within
the subbank. The McBSP data register (SPSDx) is used to access (read or write) the selected register.
Table 3–21 shows the McBSP control registers and their corresponding subaddresses.
Table 3–21. McBSP Control Registers and Subaddresses
McBSP0
McBSP1
McBSP2
NAME
ADDRESS
NAME
ADDRESS
SUBADDRESS
39h
SPCR11
49h
SPCR12
35h
00h
Serial port control register 1
39h
SPCR21
49h
SPCR22
35h
01h
Serial port control register 2
39h
RCR11
49h
RCR12
35h
02h
Receive control register 1
39h
RCR21
49h
RCR22
35h
03h
Receive control register 2
XCR10
39h
XCR11
49h
XCR12
35h
04h
Transmit control register 1
ÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁ
ÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁ
ÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁ
ÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁ
ÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁ
ÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁ
ÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁ
ÁÁÁÁÁÁÁÁÁÁ
ÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁ
ÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁ
ÁÁÁÁ
ÁÁÁÁÁ
ÁÁÁÁ
ÁÁÁÁ
ÁÁÁÁ
ÁÁÁÁ
ÁÁÁÁ
ÁÁÁÁÁÁÁÁÁÁÁÁ
ÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁ
ÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁ
ÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁ
ÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁ
ÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁ
ÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁ
ÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁ
ÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁ
ÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁ
ÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁ
ÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁ
ÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁ
ÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁ
ÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁ
ÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁ
NAME
ADDRESS
SPCR10
SPCR20
RCR10
RCR20
DESCRIPTION
XCR20
39h
XCR21
49h
XCR22
35h
05h
Transmit control register 2
SRGR10
39h
SRGR11
49h
SRGR12
35h
06h
Sample rate generator register 1
SRGR20
39h
SRGR21
49h
SRGR22
35h
07h
Sample rate generator register 2
MCR10
39h
MCR11
49h
MCR12
35h
08h
Multichannel register 1
MCR20
39h
MCR21
49h
MCR22
35h
09h
Multichannel register 2
RCERA0
39h
RCERA1
49h
RCERA2
35h
0Ah
Receive channel enable register partition A
RCERB0
39h
RCERB1
49h
RCERB2
35h
0Bh
Receive channel enable register partition B
XCERA0
39h
XCERA1
49h
XCERA2
35h
0Ch
Transmit channel enable register partition A
XCERB0
39h
XCERB1
49h
XCERB2
35h
0Dh
Transmit channel enable register partition B
PCR0
39h
PCR1
49h
PCR2
35h
0Eh
Pin control register
RCERC0
39h
RCERC1
49h
RCERC2
35h
010h
Receive channel enable register partition C
RCERD0
39h
RCERD1
49h
RCERD2
35h
011h
Receive channel enable register partition D
XCERC0
39h
XCERC1
49h
XCERC2
35h
012h
Transmit channel enable register partition C
XCERD0
39h
XCERD1
49h
XCERD2
35h
013h
Transmit channel enable register partition D
RCERE0
39h
RCERE1
49h
RCERE2
35h
014h
Receive channel enable register partition E
RCERF0
39h
RCERF1
49h
RCERF2
35h
015h
Receive channel enable register partition F
XCERE0
39h
XCERE1
49h
XCERE2
35h
016h
Transmit channel enable register partition E
XCERF0
39h
XCERF1
49h
XCERF2
35h
017h
Transmit channel enable register partition F
RCERG0
39h
RCERG1
49h
RCERG2
35h
018h
Receive channel enable register partition G
RCERH0
39h
RCERH1
49h
RCERH2
35h
019h
Receive channel enable register partition H
XCERG0
39h
XCERG1
49h
XCERG2
35h
01Ah
Transmit channel enable register partition G
XCERH0
39h
XCERH1
49h
XCERH2
35h
01Bh
Transmit channel enable register partition H
44
SPRS098C
December 1999 – Revised November 2001
Functional Overview
3.12 DMA Subbank Addressed Registers
The direct memory access (DMA) controller has several control registers associated with it. The main control
register (DMPREC) is a standard memory-mapped register. However, the other registers are accessed using
the subbank addressing scheme. This allows a set or subbank of registers to be accessed through a single
memory location. The DMA subbank address (DMSA) register is used as a pointer to select a particular
register within the subbank, while the DMA subbank data (DMSD) register or the DMA subbank data register
with autoincrement (DMSDI) is used to access (read or write) the selected register.
When the DMSDI register is used to access the subbank, the subbank address is automatically
postincremented so that a subsequent access affects the next register within the subbank. This autoincrement
feature is intended for efficient, successive accesses to several control registers. If the autoincrement feature
is not required, the DMSDN register should be used to access the subbank. Table 3–22 shows the DMA
controller subbank addressed registers and their corresponding subaddresses.
Table 3–22. DMA Subbank Addressed Registers
ÁÁÁÁÁ
ÁÁÁÁÁ
ÁÁÁÁÁ
ÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁ
ÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁ
ÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁ
ÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁ
ÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁ
ÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁ
ÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁ
ÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁ
ÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁ
ÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁ
ÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁ
ÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁ
ÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁ
ÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁ
ÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁ
ÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁ
ÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁ
ÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁ
ÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁ
ÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁ
ÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁ
ÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁ
ÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁ
ÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁ
ÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁ
ÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁ
ADDRESS
SUBADDRESS
DMSRC0
56h/57h
00h
DMA channel 0 source address register
DMDST0
56h/57h
01h
DMA channel 0 destination address register
DMCTR0
56h/57h
02h
DMA channel 0 element count register
DMSFC0
56h/57h
03h
DMA channel 0 sync event and frame count register
DMMCR0
56h/57h
04h
DMA channel 0 transfer mode control register
DMSRC1
56h/57h
05h
DMA channel 1 source address register
DMDST1
56h/57h
06h
DMA channel 1 destination address register
DMCTR1
56h/57h
07h
DMA channel 1 element count register
DMSFC1
56h/57h
08h
DMA channel 1 sync event and frame count register
DMMCR1
56h/57h
09h
DMA channel 1 transfer mode control register
DMSRC2
56h/57h
0Ah
DMA channel 2 source address register
DMDST2
56h/57h
0Bh
DMA channel 2 destination address register
DMCTR2
56h/57h
0Ch
DMA channel 2 element count register
DMSFC2
56h/57h
0Dh
DMA channel 2 sync event and frame count register
DMMCR2
56h/57h
0Eh
DMA channel 2 transfer mode control register
DMSRC3
56h/57h
0Fh
DMA channel 3 source address register
DMDST3
56h/57h
10h
DMA channel 3 destination address register
DMCTR3
56h/57h
11h
DMA channel 3 element count register
DMSFC3
56h/57h
12h
DMA channel 3 sync event and frame count register
DMMCR3
56h/57h
13h
DMA channel 3 transfer mode control register
DMSRC4
56h/57h
14h
DMA channel 4 source address register
DMDST4
56h/57h
15h
DMA channel 4 destination address register
DMCTR4
56h/57h
16h
DMA channel 4 element count register
DMSFC4
56h/57h
17h
DMA channel 4 sync event and frame count register
DMMCR4
56h/57h
18h
DMA channel 4 transfer mode control register
DMSRC5
56h/57h
19h
DMA channel 5 source address register
DMDST5
56h/57h
1Ah
DMA channel 5 destination address register
DMCTR5
56h/57h
1Bh
DMA channel 5 element count register
DMSFC5
56h/57h
1Ch
DMA channel 5 sync event and frame count register
DMMCR5
56h/57h
1Dh
DMA channel 5 transfer mode control register
DMSRCP
56h/57h
1Eh
DMA source program page address (common channel)
NAME
December 1999 – Revised November 2001
DESCRIPTION
SPRS098C
45
Functional Overview
Table 3–22. DMA Subbank Addressed Registers (Continued)
ÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁ
ÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁ
ÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁ
ÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁ
ÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁ
ÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁ
ÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁ
ÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁ
ÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁ
ÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁ
ÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁ
ÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁ
ÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁ
ÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁ
ÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁ
ÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁ
ÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁ
ÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁ
ÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁ
ÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁ
ÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁ
ÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁ
ÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁ
ÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁ
ÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁ
ÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁ
ÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁ
ÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁ
ÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁ
ÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁ
ÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁ
ÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁ
ÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁ
ÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁ
ÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁ
ADDRESS
SUBADDRESS
DMDSTP
56h/57h
1Fh
DMA destination program page address (common channel)
DMIDX0
56h/57h
20h
DMA element index address register 0
DMIDX1
56h/57h
21h
DMA element index address register 1
DMFRI0
56h/57h
22h
DMA frame index register 0
DMFRI1
56h/57h
23h
DMA frame index register 1
DMGSA0
56h/57h
24h
DMA channel 0 global source address reload register
DMGDA0
56h/57h
25h
DMA channel 0 global destination address reload register
DMGCR0
56h/57h
26h
DMA channel 0 global count reload register
DMGFR0
56h/57h
27h
DMA channel 0 global frame count reload register
XSRCDP
56h/57h
28h
DMA extended source data page
XDSTDP
56h/57h
29h
DMA extended destination data page
DMGSA1
56h/57h
2Ah
DMA channel 1 global source address reload register
DMGDA1
56h/57h
2Bh
DMA channel 1 global destination address reload register
DMGCR1
56h/57h
2Ch
DMA channel 1 global count reload register
DMGFR1
56h/57h
2Dh
DMA channel 1 global frame count reload register
DMGSA2
56h/57h
2Eh
DMA channel 2 global source address reload register
DMGDA2
56h/57h
2Fh
DMA channel 2 global destination address reload register
DMGCR2
56h/57h
30h
DMA channel 2 global count reload register
DMGFR2
56h/57h
31h
DMA channel 2 global frame count reload register
DMGSA3
56h/57h
32h
DMA channel 3 global source address reload register
DMGDA3
56h/57h
33h
DMA channel 3 global destination address reload register
DMGCR3
56h/57h
34h
DMA channel 3 global count reload register
DMGFR3
56h/57h
35h
DMA channel 3 global frame count reload register
DMGSA4
56h/57h
36h
DMA channel 4 global source address reload register
DMGDA4
56h/57h
37h
DMA channel 4 global destination address reload register
DMGCR4
56h/57h
38h
DMA channel 4 global count reload register
DMGFR4
56h/57h
39h
DMA channel 4 global frame count reload register
DMGSA5
56h/57h
3Ah
DMA channel 5 global source address reload register
DMGDA5
56h/57h
3Bh
DMA channel 5 global destination address reload register
DMGCR5
56h/57h
3Ch
DMA channel 5 global count reload register
DMGFR5
56h/57h
3Dh
DMA channel 5 global frame count reload register
NAME
46
SPRS098C
DESCRIPTION
December 1999 – Revised November 2001
Functional Overview
3.13 Interrupts
Vector-relative locations and priorities for all internal and external interrupts are shown in Table 3–23.
Table 3–23. 5421 Interrupt Locations and Priorities for Each DSP Subsystem
NAME
LOCATION
PRIORITY
FUNCTION
DECIMAL
HEX
RS, SINTR
0
00
1
Reset (Hardware and Software Reset)
NMI, SINT16
4
04
2
Nonmaskable Interrupt
SINT17
8
08
—
Software Interrupt #17
SINT18
12
0C
—
Software Interrupt #18
SINT19
16
10
—
Software Interrupt #19
SINT20
20
14
—
Software Interrupt #20
SINT21
24
18
—
Software Interrupt #21
SINT22
28
1C
—
Software Interrupt #22
SINT23
32
20
—
Software Interrupt #23
SINT24
36
24
—
Software Interrupt #24
SINT25
40
28
—
Software Interrupt #25
SINT26
44
2C
—
Software Interrupt #26
SINT27
48
30
—
Software Interrupt #27
SINT28
52
34
—
Software Interrupt #28
SINT29
56
38
—
Software Interrupt #29
SINT30
60
3C
—
Software Interrupt #30
INT0, SINT0
64
40
3
External User Interrupt #0
INT1, SINT1
68
44
4
External User Interrupt #1
INT2, SINT2
72
48
5
Reserved
TINT, SINT3
76
4C
6
External Timer Interrupt
BRINT0, SINT4
80
50
7
BSP #0 Receive Interrupt
BXINT0, SINT5
84
54
8
BSP #0 Transmit Interrupt
BRINT2, DMAC0
88
58
9
BSP #2 Receive Interrupt or DMA Channel 0
BXINT2, DMAC1
92
5C
10
BSP #2 Receive Interrupt or DMA Channel 1
INT3, SINT8
96
60
11
Reserved
HPINT, SINT9
100
64
12
HPI Interrupt (from DSPINT in HPIC)
BRINT1, DMAC2
104
68
13
BSP #1 Receive Interrupt or DMA Channel 2
BXINT1, DMAC3
108
6C
14
BSP #1 transmit Interrupt or DMA channel 3
DMAC4, SINT12
112
70
15
DMA Channel 4
DMAC5, SINT13
116
74
16
DMA Channel 5
120
78
17
Interprocessor Interrupt
124–127
7C–7F
—
Reserved
IPINT, SINT14
—
The interprocessor interrupt (IPINT) bit of the interrupt mask register (IMR) and the interrupt flag register (IFR)
allows the subsystem to perform interrupt service routines based on the other subsystem activity. Incoming
IPINT interrupts are latched in IFR.14. Generating an interprocessor interrupt is performed by writing a “1” to
the IPIRQ field of the bank-switching control register (BSCR). Subsequent interrupts must first clear the
interrupt by writing “0” to the IPIRQ field. Figure 3–21 shows the bit layout of the IMR and the IFR. Table 3–24
describes the bit functions.
For example, if subsystem A is required to notify subsystem B of a completed task, subsystem A must write
a “1” to the IPIRQ field to generate a IPINT interrupt on subsystem B. On subsystem B, the IPINT interrupt
is latched in IFR.14. Figure 5 shows the bit layout of the BSCR and Table 6 describes the bit functions.
December 1999 – Revised November 2001
SPRS098C
47
Functional Overview
15
14
13
12
11
10
9
8
Reserved
IPINT
DMAC5
DMAC4
XINT1 or
DMAC3
RINT1 or
DMAC2
HPINT
Reserved
R/W
R/W
R/W
R/W
R/W
7
6
5
4
3
2
1
0
XINT2 or
DMAC1
RINT2 or
DMAC0
XINT0
RINT0
TINT
Reserved
INT1
INT0
R/W
R/W
R/W
R/W
R/W
R/W
R/W
LEGEND: R = Read, W = Write
Figure 3–21. Bit Layout of the IMR and IFR Registers for Subsystems A and B
Table 3–24. Bit Functions for IMR and IFR Registers for Each DSP Subsystem
BIT
NO.
BIT
NAME
BIT
VALUE
15
Reserved
X
Register bit is reserved.
0
IFR/IMR: Interprocessor IRQ has no interrupt pending/is disabled (masked).
14
IPINT
1
IFR/IMR: Interprocessor IRQ has an interrupt pending/is enabled.
0
IFR/IMR: DMA Channel 5 has no interrupt pending/is disabled (masked).
13
DMAC5
1
IFR/IMR: DMA Channel 5 has an interrupt pending/is enabled.
0
IFR/IMR: DMA Channel 4 has no interrupt pending/is disabled (masked).
1
IFR/IMR: DMA Channel 4 has an interrupt pending/is enabled.
0
IFR/IMR: McBSP_1 has no transmit interrupt pending/is disabled (masked).
1
IFR/IMR: McBSP_1 has a transmit interrupt pending/is enabled.
0
IFR/IMR: DMA Channel 3 has no interrupt pending/is disabled (masked).
1
IFR/IMR: DMA Channel 3 has an interrupt pending/is enabled.
0
IFR/IMR: McBSP_1 has no receive interrupt pending/is disabled (masked).
12
DMAC4
XINT1
11
DMAC3
RINT1
10
DMAC2
9
HPINT
8
Reserved
XINT2
7
DMAC1
RINT2
6
DMAC0
48
5
XINT0
4
RINT0
SPRS098C
FUNCTION
1
IFR/IMR: McBSP_1 has a receive interrupt pending/is enabled.
0
IFR/IMR: DMA Channel 2 has no interrupt pending/is disabled (masked).
1
IFR/IMR: DMA Channel 2 has an interrupt pending/is enabled.
0
IFR/IMR: Host-port interface has no DSPINT interrupt pending/is disabled (masked).
1
IFR/IMR: Host-port interface has an DSPINT interrupt pending/is enabled.
X
Register bit is reserved.
0
IFR/IMR: McBSP_2 has no transmit interrupt pending/is disabled (masked).
1
IFR/IMR: McBSP_2 has a transmit interrupt pending/is enabled.
0
IFR/IMR: DMA Channel 1 has no interrupt pending/is disabled (masked).
1
IFR/IMR: DMA Channel 1 has an interrupt pending/is enabled.
0
IFR/IMR: McBSP_2 has no receive interrupt pending/is disabled (masked).
1
IFR/IMR: McBSP_2 has a receive interrupt pending/is enabled.
0
IFR/IMR: DMA Channel 0 has no interrupt pending/is disabled (masked).
1
IFR/IMR: DMA Channel 0 has an interrupt pending/is enabled.
0
IFR/IMR: McBSP_0 has no receive interrupt pending/is disabled (masked).
1
IFR/IMR: McBSP_0 has a receive interrupt pending/is enabled.
0
IFR/IMR: McBSP_0 has no receive interrupt pending/is disabled (masked).
1
IFR/IMR: McBSP_0 has a receive interrupt pending/is enabled.
December 1999 – Revised November 2001
Functional Overview
Table 3–24. Bit Functions for IMR and IFR Registers for Each DSP Subsystem (Continued)
BIT
NO.
BIT
NAME
3
TINT
2
Reserved
1
0
INT1
INT0
BIT
VALUE
FUNCTION
0
IFR/IMR: Timer has no interrupt pending/is disabled (masked).
1
IFR/IMR: Timer has an interrupt pending/is enabled.
X
Register bit is reserved.
0
IFR/IMR: Ext user interrupt pin 1 has no interrupt pending/is disabled (masked).
1
IFR/IMR: Ext user interrupt pin 1 has an interrupt pending/is enabled.
0
IFR/IMR: Ext user interrupt pin 0 has no interrupt pending/is disabled (masked).
1
IFR/IMR: Ext user interrupt pin 0 has an interrupt pending/is enabled.
3.14 IDLE3 Power-Down Mode
The IDLE1 and IDLE2 power-down modes operate as described in the TMS320C54x DSP Reference Set,
Volume 1: CPU and Peripherals (literature number SPRU131). The IDLE3 mode is special in that the clocking
circuitry is shut off to conserve power. The 5421 cannot enter an IDLE3 mode unless both subsystems execute
an IDLE3 instruction. The power-reduced benefits of IDLE3 cannot be realized until both subsystems enter
the IDLE3 state and the internal clocks are automatically shut off. The order in which subsystems enter IDLE3
does not matter.
3.15 Emulating the 5421 Device
The 5421 is a single device, but actually consists of two independent subboundary systems that contain
register/status information used by the emulator tools. The emulator tools must be informed of the multicore
device by modifying the board.cfg file. The board.cfg file is an ASCII file that can be modified with most editors.
This provides the emulator with a description of the JTAG chain. The board.cfg file must identify two
processors when using the 5421. The file contents would look something like this:
“CPU_B” TI320C5xx
“CPU_A” TI320C5xx
Use Code Composer Studio to convert this file into a binary file (board.dat), readable by the emulation tools.
Place the board.dat file in the directory that contains the emulator software.
The subsystems are serially connected together internally. Emulation information is serially transmitted into
the device using the TDI pin. The device responds with serial scan information transmitted out the TDO pin.
December 1999 – Revised November 2001
SPRS098C
49
Documentation Support
4
Documentation Support
Extensive documentation supports all TMS320 DSP family generations of devices from product
announcement through applications development. The following types of documentation are available to
support the design and use of the TMS320C5000 family of DSPs:
•
•
•
•
•
TMS320C54x DSP Functional Overview (literature number SPRU307)
Device-specific data sheets
Complete User Guides
Development-support tools
Hardware and software application reports
The five-volume TMS320C54x DSP Reference Set (literature number SPRU210) consists of:
•
•
•
•
•
Volume 1: CPU and Peripherals (literature number SPRU131)
Volume 2: Mnemonic Instruction Set (literature number SPRU172)
Volume 3: Algebraic Instruction Set (literature number SPRU179)
Volume 4: Applications Guide (literature number SPRU173)
Volume 5: Enhanced Peripherals (literature number SPRU302)
The reference set describes in detail the TMS320C54x DSP family of products currently available and the
hardware and software applications, including algorithms, for fixed-point TMS320 DSP family of devices.
A series of DSP textbooks is published by Prentice-Hall and John Wiley & Sons to support digital signal
processing research and education. The TMS320 DSP family newsletter, Details on Signal Processing, is
published quarterly and distributed to update TMS320 customers on product information.
Information regarding TI DSP products is also available on the Worldwide Web at http://www.ti.com uniform
resource locator (URL).
TMS320 is a trademark of Texas Instruments.
50
SPRS098C
December 1999 – Revised November 2001
Electrical Specifications
5
Electrical Specifications
This section provides the absolute maximum ratings and the recommended operating conditions for the
TMS320VC5421 DSP.
5.1
Absolute Maximum Ratings
The list of absolute maximum ratings are specified over operating case temperature. Stresses beyond those
listed under “absolute maximum ratings” may cause permanent damage to the device. These are stress
ratings only, and functional operation of the device at these or any other conditions beyond those indicated
under “recommended operating conditions” is not implied. Exposure to absolute-maximum-rated conditions
for extended periods may affect device reliability. All voltage values are with respect to VSS.
Supply voltage I/O range, DVDD . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . – 0.5 V to 4.0 V
Supply voltage core range, CVDD . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . – 0.5 V to 2.4 V
Supply voltage analog PLL range, AVDD . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . – 0.5 V to 2.4 V
Input voltage range, VI . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . – 0.5 V to DVDD + 0.5 V
Output voltage range, Vo . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . – 0.5 V to DVDD + 0.5 V
Operating case temperature range, TC . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 0°C to 85°C
Storage temperature range Tstg . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . – 65C to 150C
5.2
Recommended Operating Conditions
The device recommended operating conditions are supplied in Table 5–1 and the electrical characteristics
over recommended operating case temperature range (unless otherwise noted) are listed in Table 5–2.
Figure 5–1 provides the test load circuit values for a 3.3-V device.
Table 5–1. Recommended Operating Conditions
MIN
NOM
MAX
UNIT
DVDD
Device supply voltage, I/O
3
3.3
3.6
V
CVDD
Device supply voltage, core
1.75
1.80
1.98
V
AVDD
Device supply voltage, PLL
1.75
1.80
1.98
V
VSS
Supply voltage, GND
0
Schmitt triggered inputs
VIH
High-level input voltage, I/O
0.7DVDD
DVDD
2
DVDD
DVDD = 3.3 ± 0.3 V
0
0.3DVDD
All other inputs
0
0.8
DVDD = 3.3 ± 0.3 V
All other inputs
Schmitt triggered inputs
VIL
Low-level input voltage, I/O
IOH
High-level output current
IOL
Low-level output current
TC
Operating case temperature
December 1999 – Revised November 2001
V
0
V
V
–300
µA
1.5
mA
85
°C
SPRS098C
51
Electrical Specifications
5.3
Electrical Characteristics
Table 5–2 describes the electrical characteristics over recommended operating case temperature range
(unless otherwise noted).
Table 5–2. Electrical Characteristics
TEST CONDITIONS†
PARAMETER
VOH
High-level output
VOL
Low-level output voltage§
IIZ
Input current in high impedance
II
Input current
(VI = VSS to DVDD)
MIN
DVDD = 3.3 ± 0.3 V, IOH = MAX
voltage§
TYP‡
MAX
2.4
V
IOL = MAX
0.4
V
µA
DVDD = MAX, VI = VSS to DVDD
–10
10
TRST
With internal pulldown
–10
35
See pin descriptions
With internal pullups
PPD[15:0]
Bus holders enabled, DVDD = MAX
All other input-only pins
IDDC
Supply current, both core CPUs
CVDD = 1.8 V, fx = 100 MHz¶,
TC = 25°C
IDDP
Supply current, pins
DVDD = 3.3 V, fclock = 100 MHz#,
TC = 25°C||
IDDA
Supply current, PLL
UNIT
–35
10
–200
200
–10
10
µA
90#
mA
54
mA
5
mA
IDLE2
PLL × n mode, 20 MHz input
2
mA
IDLE3
PLL x n mode, 20 MHz input
IDDC
Supply current, standby
600
µA
Ci
Input capacitance
10
pF
Co
Output capacitance
10
pF
†
For test conditions shown as MIN, MAX, or NOM, use the appropriate value specified in the recommended operating conditions table.
All values are typical unless otherwise specified.
§ All input and output voltage levels except RS, INT0, INT1, NMI, CLKIN, BCLKX, BCLKR, HAS, HCS, HDS1, HDS2, and HPIRS are
LVTTL-compatible.
¶ Clock mode: PLL × 1 with external source
# This value is based on 50% usage of MAC and 50% usage of NOP instructions. Actual operating current varies with the program being executed.
|| This value was obtained using the following conditions: external memory writes at a rate of 20 million writes per second, CLKOFF = 0, full-duplex
operation of all six McBSPs at a rate of 10 million bits per second each, and 15-pF loads on all outputs. For more details on how this calculation
is performed, refer to the Calculation of TMS320LC54x Power Dissipation Application Report (literature number SPRA164).
VIL(MIN) ≤ VI ≤ VIL(MAX) or VIH(MIN) ≤ VI ≤ VIH(MAX)
‡
IOL
50 Ω
Tester Pin
Electronics
VLoad
CT
Output
Under
Test
IOH
Where:
IOL
IOH
VLoad
CT
=
=
=
=
1.5 mA (all outputs)
300 µA (all outputs)
1.5 V
40 pF typical load circuit capacitance
Figure 5–1. 3.3-V Test Load Circuit
52
SPRS098C
December 1999 – Revised November 2001
Electrical Specifications
5.4
Package Thermal Resistance Characteristics
Table 5–3 provides the thermal resistance characteristics for the recommended package types used on the
TMS320VC5421 DSP.
Table 5–3. Thermal Resistance Characteristics
5.5
PARAMETER
GGU
PACKAGE
PGE
PACKAGE
UNIT
RΘJA
38
56
°C/W
RΘJC
5
5
°C/W
Timing Parameter Symbology
Timing parameter symbols used in the timing requirements and switching characteristics tables are created
in accordance with JEDEC Standard 100. To shorten the symbols, some of the pin names and other related
terminology have been abbreviated as follows:
Lowercase subscripts and their meanings:
Letters and symbols and their meanings:
a
access time
H
High
c
cycle time (period)
L
Low
d
delay time
V
Valid
dis
disable time
Z
High impedance
en
enable time
f
fall time
h
hold time
r
rise time
su
setup time
t
transition time
v
valid time
w
pulse duration (width)
X
Unknown, changing, or don’t care level
December 1999 – Revised November 2001
SPRS098C
53
Electrical Specifications
5.6
Clock Options
The frequency of the reference clock provided at the CLKIN pin can be divided by a factor of two or four to
generate the internal machine cycle. The selection of the clock mode is described in the
software-programmable phase-locked loop (PLL) section.
5.6.1 Divide-By-Two, Divide-By-Four, and Bypass Clock Option (PLL Disabled)
The frequency of the reference clock provided at the CLKIN pin can be divided by a factor of two or four to
generate the internal machine cycle. The selection of the clock mode is described in the
software-programmable phase-locked loop (PLL) section.
The following timing requirements and switching characteristics tables assume testing over recommended
operating conditions and H = 0.5tc(CO) (see Figure 5–2).
Table 5–4. Divide-By-2 and Divide-by-4 Clock Options Timing Requirements
†
MIN
MAX
UNIT
20
†
ns
8
ns
8
ns
tc(CI)
Cycle time, CLKIN
tf(CI)
Fall time, CLKIN
tr(CI)
Rise time, CLKIN
tw(CIL)
Pulse duration, CLKIN low
5
ns
tw(CIH)
Pulse duration, CLKIN high
5
ns
This device utilizes a fully static design and therefore can operate with tc(CI) approaching ∞. The device is characterized at frequencies
approaching 0 Hz.
Table 5–5. Divide-By-2 and Divide-by-4 Clock Options Switching Characteristics
PARAMETER
†
MIN
TYP
MAX
UNIT
tc(CO)
Cycle time, CLKOUT
40
2tc(CI)
†
tc(CO)
Cycle time, CLKOUT – bypass mode
40
2tc(CI)
†
ns
td(CIH-CO)
Delay time, CLKIN high to CLKOUT high/low
3
6
10
ns
tf(CO)
Fall time, CLKOUT
2
tr(CO)
Rise time, CLKOUT
2
tw(COL)
Pulse duration, CLKOUT low
H–2
H–1
H+2
ns
tw(COH)
Pulse duration, CLKOUT high
H–2
H–1
H+2
ns
ns
ns
ns
This device utilizes a fully static design and therefore can operate with tc(CI) approaching ∞. The device is characterized at frequencies
approaching 0 Hz.
tr(CI)
tw(CIH)
tf(CI)
tw(CIL)
tc(CI)
CLKIN
tc(CO)
td(CIH-CO)
tw(COH)
tf(CO)
tr(CO)
tw(COL)
CLKOUT
Figure 5–2. External Divide-by-Two Clock Timing
54
SPRS098C
December 1999 – Revised November 2001
Electrical Specifications
5.6.2 Multiply-By-N Clock Option (PLL Enabled)
The frequency of the reference clock provided at the CLKIN pin can be multiplied by a factor of N to generate
the internal machine cycle. The selection of the clock mode and the value of N is described in the
software-programmable phase-locked loop (PLL) section.
The following timing requirements and switching characteristics tables assume testing over recommended
operating conditions and H = 0.5tc(CO) (see Figure 5–3).
Table 5–6. Multiply-By-N Clock Option Timing Requirements
MIN
MAX
20‡
200
PLL multiplier N = x.5†
20‡
100
PLL multiplier N = x.25, x.75†
20‡
50
Integer PLL multiplier N (N =
tc(CI)
Cycle time, CLKIN
1–15)†
UNIT
ns
tf(CI)
Fall time, CLKIN
8
ns
tr(CI)
Rise time, CLKIN
tw(CIL)
Pulse duration, CLKIN low
5
8
ns
ns
tw(CIH)
Pulse duration, CLKIN high
5
ns
†
N = Multiplication factor
‡ The multiplication factor and minimum CLKIN cycle time should be chosen such that the resulting CLKOUT cycle time is within the specified range
(tc(CO))
Table 5–7. Multiply-By-N Clock Option Switching Characteristics
PARAMETER
†
MIN
TYP
10
tc(CI)/N†
4
10
MAX
UNIT
tc(CO)
Cycle time, CLKOUT
td(CI-CO)
Delay time, CLKIN high/low to CLKOUT high/low
ns
tf(CO)
Fall time, CLKOUT
2
ns
tr(CO)
Rise time, CLKOUT
2
ns
tw(COL)
Pulse duration, CLKOUT low
H–2
H–1
H+2
ns
tw(COH)
Pulse duration, CLKOUT high
H–2
H–1
H+2
ns
tp
Transitory phase, PLL lock up time
30
ms
16
ns
N = Multiplication factor
tw(CIH)
tc(CI)
tw(CIL)
tf(CI)
tr(CI)
CLKIN
td(CI-CO)
tc(CO)
tp
tw(COH)
tf(CO)
tw(COL)
tr(CO)
Unstable
CLKOUT
Figure 5–3. External Multiply-by-One Clock Timing
December 1999 – Revised November 2001
SPRS098C
55
Electrical Specifications
5.7
External Memory Interface Timing
5.7.1 Memory Read
External memory reads can be performed in consecutive or nonconsecutive mode under control of the
CONSEC bit in the BSCR. The following timing requirements and switching characteristics tables assume
testing over recommended operating conditions with MSTRB = 0 and H = 0.5tc(CO) (see Figure 5–4).
Table 5–8. Memory Read Timing Requirements
MIN
†
MAX
UNIT
ta(A)M
Access time, read data access from address valid†
ta(MSTRBL)
Access time, read data access from MSTRB low
tsu(D)R
Setup time, read data before CLKOUT low
9
ns
th(D)R
Hold time, read data after CLKOUT low
0
ns
th(A-D)R
Hold time, read data after address invalid
0
ns
th(D)MSTRBH
Hold time, read data after MSTRB high
0
ns
2H–12
ns
2H–11
ns
Address, PS, and DS timings are all included in timings referenced as address.
Table 5–9. Memory Read Switching Characteristics
MIN
MAX
td(CLKL-A)
Delay time, CLKOUT low to address valid†‡
PARAMETER
–1
5
ns
td(CLKH-A)
Delay time, CLKOUT high (transition) to address valid†§
–1
6
ns
td(CLKL-MSL)
Delay time, CLKOUT low to MSTRB low
–1
4
ns
td(CLKL-MSH)
Delay time, CLKOUT low to MSTRB high
–1
4
ns
–1
5§
ns
–1
6§
ns
low†‡
th(CLKL-A)R
Hold time, address valid after CLKOUT
th(CLKH-A)R
Hold time, address valid after CLKOUT high†§
UNIT
†
Address, PS, and DS timings are all included in timings referenced as address.
‡ In the case of a memory read preceded by a memory read
§ In the case of a memory read preceded by a memory write
56
SPRS098C
December 1999 – Revised November 2001
Electrical Specifications
CLKOUT
td(CLKL-A)
th(CLKL-A)R
PPA[18:0]
th(A-D)R
tsu(D)R
ta(A)M
th(D)R
PPD[15:0]
th(D)MSTRBH
td(CLKL-MSL)
td(CLKL-MSH)
ta(MSTRBL)
MSTRB
R/W
PS, DS
Figure 5–4. Memory Read (MSTRB = 0)
December 1999 – Revised November 2001
SPRS098C
57
Electrical Specifications
5.7.2 Memory Write
The following switching characteristics table assumes testing over recommended operating conditions with
MSTRB = 0 and H = 0.5tc(CO) (see Figure 5–5).
Table 5–10. Memory Write Switching Characteristics
MIN
MAX
td(CLKH-A)
Delay time, CLKOUT high to address valid†‡
PARAMETER
–1
6
ns
td(CLKL-A)
Delay time, CLKOUT low to address valid†§
–1
5
ns
td(CLKL-MSL)
Delay time, CLKOUT low to MSTRB low
–1
4
ns
td(CLKL-D)W
Delay time, CLKOUT low to data valid
0
7
ns
td(CLKL-MSH)
Delay time, CLKOUT low to MSTRB high
–1
4
ns
td(CLKH-RWL)
Delay time, CLKOUT high to R/W low
0
4
ns
td(CLKH-RWH)
Delay time, CLKOUT high to R/W high
0
4
ns
td(RWL-MSTRBL)
Delay time, R/W low to MSTRB low
H–2
H+2
ns
th(A)W
Hold time, address valid after CLKOUT high†‡
–1
6
ns
+3§
ns
th(D)MSH
Hold time, write data valid after MSTRB high
H–3
tw(SL)MS
Pulse duration, MSTRB low§
2H–4
low†
tsu(A)W
Setup time, address valid before MSTRB
tsu(D)MSH
Setup time, write data valid before MSTRB high
H
UNIT
ns
2H–4
ns
2H+5§
2H–5
ns
†
Address, PS, and DS timings are all included in timings referenced as address.
In the case of a memory write preceded by a memory write
§ In the case of a memory write preceded by an I/O cycle
‡
CLKOUT
td(CLKH-A)
td(CLKL-A)
th(A)W
PPA[18:0]
td(CLKL-D)W
th(D)MSH
tsu(D)MSH
PPD[15:0]
td(CLKL-MSL)
td(CLKL-MSH)
tsu(A)W
MSTRB
td(CLKH-RWH)
td(CLKH-RWL)
tw(SL)MS
td(RWL-MSTRBL)
R/W
PS, DS
Figure 5–5. Memory Write (MSTRB = 0)
58
SPRS098C
December 1999 – Revised November 2001
Electrical Specifications
5.8
Ready Timing For Externally Generated Wait States
The following timing requirements table assumes testing over recommended operating conditions and
H = 0.5tc(CO) (see Figure 5–6 and Figure 5–7).
Table 5–11. Ready Timing Requirements for Externally Generated Wait States†
MIN
MAX
UNIT
tsu(RDY)
Setup time, READY before CLKOUT low
8
ns
th(RDY)
Hold time, READY after CLKOUT low
0
ns
low‡
tv(RDY)MSTRB
Valid time, READY after MSTRB
th(RDY)MSTRB
Hold time, READY after MSTRB low‡
2H–8
2H
ns
ns
†
The hardware wait states can be used only in conjunction with the software wait states to extend the bus cycles. To generate wait states by READY,
at least two software wait states must be programmed. READY is not sampled until the completion of the internal software wait states.
‡ These timings are included for reference only. The critical timings for READY are those referenced to CLKOUT
CLKOUT
PPA[18:0]
tsu(RDY)
th(RDY)
READY
tv(RDY)MSTRB
th(RDY)MSTRB
MSTRB
Wait States
Generated Internally
Wait State
Generated
by READY
Figure 5–6. Memory Read With Externally Generated Wait States
December 1999 – Revised November 2001
SPRS098C
59
Electrical Specifications
CLKOUT
PPA[18:0]
PPD[15:0]
th(RDY)
tsu(RDY)
READY
tv(RDY)MSTRB
th(RDY)MSTRB
MSTRB
Wait States
Generated Internally
Wait State Generated
by READY
Figure 5–7. Memory Write With Externally Generated Wait States
60
SPRS098C
December 1999 – Revised November 2001
Electrical Specifications
5.9
Parallel I/O Interface Timing
5.9.1 Parallel I/O Port Read
The following timing requirements and switching characteristics tables assume testing over recommended
operating conditions with IOSTRB = 0 and H = 0.5tc(CO) (see Figure 5–8).
Table 5–12. Parallel I/O Port Read Timing Requirements
MIN
†
MAX
UNIT
ta(A)IO
Access time, read data access from address valid†
ta(ISTRBL)IO
Access time, read data access from IOSTRB low
tsu(D)IOR
Setup time, read data before CLKOUT high
9
ns
th(D)IOR
Hold time, read data after CLKOUT high
0
ns
th(ISTRBH-D)R
Hold time, read data after IOSTRB high
0
ns
3H–12
ns
2H–11
ns
Address and IS timings are included in timings referenced as address.
Table 5–13. Parallel I/O Port Read Switching Characteristics
PARAMETER
†
MIN
MAX
td(CLKL-A)
Delay time, CLKOUT low to address valid†
–1
5
UNIT
ns
td(CLKH-ISTRBL)
Delay time, CLKOUT high to IOSTRB low
0
5
ns
td(CLKH-ISTRBH)
Delay time, CLKOUT high to IOSTRB high
0
5
ns
th(A)IOR
Hold time, address after CLKOUT low†
–1
5
ns
Address and IS timings are included in timings referenced as address.
CLKOUT
th(A)IOR
td(CLKL–A)
PPA[18:0]
th(D)IOR
tsu(D)IOR
ta(A)IO
PPD[15:0]
ta(ISTRBL)IO
td(CLKH–ISTRBL)
th(ISTRBH–D)R
td(CLKH–ISTRBH)
IOSTRB
R/W
IS
Figure 5–8. Parallel I/O Port Read (IOSTRB=0)
December 1999 – Revised November 2001
SPRS098C
61
Electrical Specifications
5.9.2 Parallel I/O Port Write
The following switching characteristics table assumes testing over recommended operating conditions with
IOSTRB = 0 and H = 0.5tc(CO) (see Figure 5–9).
Table 5–14. Parallel I/O Port Write Switching Characteristics
PARAMETER
†
valid†
MIN
MAX
UNIT
–1
5
ns
0
5
ns
td(CLKL-A)
Delay time, CLKOUT low to address
td(CLKH-ISTRBL)
Delay time, CLKOUT high to IOSTRB low
td(CLKH-D)IOW
Delay time, CLKOUT high to write data valid
H–5
H+5
ns
td(CLKH-ISTRBH)
Delay time, CLKOUT high to IOSTRB high
0
5
ns
td(CLKL-RWL)
Delay time, CLKOUT low to R/W low
0
4
ns
td(CLKL-RWH)
Delay time, CLKOUT low to R/W high
th(A)IOW
Hold time, address valid after CLKOUT low†
th(D)IOW
tsu(D)IOSTRBH
tsu(A)IOSTRBL
0
4
ns
–1
5
ns
Hold time, write data after IOSTRB high
H–3
H+7
ns
Setup time, write data before IOSTRB high
H–5
H+1
ns
Setup time, address valid before IOSTRB low†
H–5
H+3
ns
Address and IS timings are included in timings referenced as address.
CLKOUT
tsu(A)IOSTRBL
th(A)IOW
td(CLKL-A)
PPA[18:0]
td(CLKH-D)IOW
th(D)IOW
PPD[15:0]
td(CLKH-ISTRBL)
td(CLKH-ISTRBH)
tsu(D)IOSTRBH
IOSTRB
td(CLKL-RWH)
td(CLKL-RWL)
R/W
IS
Figure 5–9. Parallel I/O Port Write (IOSTRB=0)
62
SPRS098C
December 1999 – Revised November 2001
Electrical Specifications
5.10 Externally Generated Wait States
5.10.1
I/O Port Read and Write With Externally Generated Wait States
The following timing requirements table assumes testing over recommended operating conditions and
H = 0.5tc(CO) (see Figure 5–10 and Figure 5–11).
Table 5–15. Externally Generated Wait States Timing Requirements†
MIN
MAX
UNIT
tsu(RDY)
Setup time, READY before CLKOUT low
8
ns
th(RDY)
Hold time, READY after CLKOUT low
0
ns
low‡
tv(RDY)IOSTRB
Valid time, READY after IOSTRB
th(RDY)IOSTRB
Hold time, READY after IOSTRB low‡
3H–9
3H
ns
ns
†
The hardware wait states can be used only in conjunction with the software wait states to extend the bus cycles. To generate wait states using
READY, at least two software wait states must be programmed.
‡ These timings are included for reference only. The critical timings for READY are those referenced to CLKOUT.
CLKOUT
PPA[18:0]
th(RDY)
tsu(RDY)
READY
tv(RDY)IOSTRB
th(RDY)IOSTRB
IOSTRB
Wait
States
Generated
Internally
Wait State Generated
by READY
Figure 5–10. I/O Port Read With Externally Generated Wait States
December 1999 – Revised November 2001
SPRS098C
63
Electrical Specifications
CLKOUT
PPA[18:0]
PPD[15:0]
th(RDY)
tsu(RDY)
READY
tv(RDY)IOSTRB
th(RDY)IOSTRB
IOSTRB
Wait States
Generated
Internally
Wait State Generated
by READY
Figure 5–11. I/O Port Write With Externally Generated Wait States
64
SPRS098C
December 1999 – Revised November 2001
Electrical Specifications
5.11 Reset, BIO, Interrupt, and MP/MC Timings
The following timing requirements table assumes testing over recommended operating conditions and
H = 0.5tc(CO) (see Figure 5–12, Figure 5–13, and Figure 5–14).
Table 5–16. Reset, BIO, Interrupt, and MP/MC Timing Requirements
MIN
MAX
UNIT
th(RS)
Hold time, RS after CLKOUT low
0
ns
th(BIO)
Hold time, BIO after CLKOUT low
0
ns
low†
th(INT)
Hold time, INTn, NMI, after CLKOUT
tw(RSL)
Pulse duration, RS low‡§
tw(BIO)A
Pulse duration, BIO low, asynchronous†
0
ns
4H+5
ns
5H
ns
4H
ns
4H
ns
8
ns
(asynchronous)†
tw(INTH)A
Pulse duration, INTn, NMI high
tw(INTL)A
Pulse duration, INTn, NMI low (asynchronous)†
tw(INTL)WKP
Pulse duration, INTn, NMI low for IDLE2/IDLE3
tw(XIO)
Pulse duration, XIO switched
ten(XIO)
Enable time, after XIO switched
wakeup†
4H
ns
4H+10
low§
5
ns
tsu(RS)
Setup time, RS before CLKIN
tsu(BIO)
Setup time, BIO before CLKOUT low
9
12
ns
ns
tsu(INT)
Setup time, INTn, NMI, RS before CLKOUT low
9
13
ns
†
The external interrupts (INT0–INT1, NMI) are synchronized to the core CPU by way of a two flip-flop synchronizer which samples these inputs
with consecutive falling edges of CLKOUT. The input to the interrupt pins is required to represent a 1-0-0 sequence at the timing that is
corresponding to a three-CLKOUT sampling sequence.
‡ If the PLL mode is selected, then at power-on sequence, or at wakeup from IDLE3, RS must be held low for at least 50 µs to ensure synchronization
and lock-in of the PLL.
§ RS can cause a change in clock frequency, changing the value of H (see the software-programmable phase-locked loop (PLL) section).
CLKIN
tsu(RS)
tw(RSL)
A_RS, B_RS,
INTn, NMI
tsu(INT)
th(RS)
CLKOUT
tsu(BIO)
th(BIO)
BIO
tw(BIO)A
Figure 5–12. Reset and BIO Timings
December 1999 – Revised November 2001
SPRS098C
65
Electrical Specifications
CLKOUT
tsu(INT)
tsu(INT)
th(INT)
INTn, NMI
tw(INTH)A
tw(INTL)A
Figure 5–13. Interrupt Timing
tw(XIO)
XIO
ten(XIO)
A[17:0]
D[15:0]
IS, DS, PS
MSTRB
IOSTRB
R/W
Figure 5–14. XIO Timing
66
SPRS098C
December 1999 – Revised November 2001
Electrical Specifications
5.12 HOLD and HOLDA Timings
The following timing requirements and switching characteristics tables assume testing over recommended
operating conditions and H = 0.5tc(CO) (see Figure 5–15).
Table 5–17. HOLD and HOLDA Timing Requirements
MIN
tw(HOLD)
Pulse duration, HOLD low
tsu(HOLD)
Setup time, HOLD low/high before CLKOUT low
MAX
UNIT
4H+10
ns
8
ns
Table 5–18. HOLD and HOLDA Switching Characteristics
PARAMETER
MIN
MAX
UNIT
tdis(CLKL-A)
Disable time, address, PS, DS, IS high impedance from CLKOUT high
5
ns
tdis(CLKL-RW)
Disable time, R/W high impedance from CLKOUT high
5
ns
tdis(CLKL-S)
Disable time, MSTRB, IOSTRB high impedance from CLKOUT high
5
ns
tdis(CLKL-D)
Disable time, Data from CLKOUT high
5
ns
ten(CLKL-A)
Enable time, address, PS, DS, IS from CLKOUT high
2H+5
ns
ten(CLKL-D)
Enable time, Data from CLKOUT high
2H+5
ns
ten(CLKL-RW)
Enable time, R/W enabled from CLKOUT high
2H+5
ns
ten(CLKL-S)
Enable time, MSTRB, IOSTRB enabled from CLKOUT high
1
2H+5
ns
td(HOLDAL)
Delay time, HOLDA low after CLKOUT high
0
11H+5
ns
td(HOLDAH)
Delay time, HOLDA high after CLKOUT high
0
5
ns
tw(HOLDA)
Pulse duration, HOLDA low duration
December 1999 – Revised November 2001
2H–3
SPRS098C
ns
67
Electrical Specifications
CLKOUT
tsu(HOLD)
tsu(HOLD)
tw(HOLD)
HOLD
td(HOLDAH)
td(HOLDAL)
tw(HOLDA)
HOLDA
tdis(CLKL-A)
ten(CLKL-A)
A[17:0]
PS, DS, IS
tdis(CLKL-D)
ten(CLKL–D)
D[15:0]
tdis(CLKL-RW)
ten(CLKL-RW)
tdis(CLKL-S)
ten(CLKL-S)
tdis(CLKL-S)
ten(CLKL-S)
R/W
MSTRB
IOSTRB
NOTE A: A[17:16] apply to DMA accesses to extended DATA and PROGRAM memory. The CPU has access to only extended
PROGRAM memory.
Figure 5–15. HOLD and HOLDA Timings (HM = 1)
68
SPRS098C
December 1999 – Revised November 2001
Electrical Specifications
5.13 External Flag (XF) and TOUT Timings
The following switching characteristics table assumes testing over recommended operating conditions and
H = 0.5tc(CO) (see Figure 5–16 and Figure 5–17).
Table 5–19. External Flag (XF) and TOUT Switching Characteristics
PARAMETER
td(XF)
MIN
MAX
Delay time, CLKOUT low to XF high
–1
4
Delay time, CLKOUT low to XF low
0
4
UNIT
ns
td(TOUTH)
Delay time, CLKOUT high to TOUT high
–1
5
ns
td(TOUTL)
Delay time, CLKOUT high to TOUT low
–1
5
ns
tw(TOUT)
Pulse duration, TOUT
2H–5
2H+2
ns
CLKOUT
td(XF)
XF
Figure 5–16. External Flag (XF) Timing
CLKOUT
td(TOUTL)
td(TOUTH)
TOUT
tw(TOUT)
Figure 5–17. Timer (TOUT) Timing
December 1999 – Revised November 2001
SPRS098C
69
Electrical Specifications
5.14 General-Purpose I/O Timing
The following timing requirements and switching characteristics tables assume testing over recommended
operating conditions (see Figure 5–18).
Table 5–20. General-Purpose I/O Timing Requirements
MIN
MAX
UNIT
tsu(GPIO-COH)
Setup time, GPIOx input valid before CLKOUT high, GPIOx configured as
general-purpose input.
7
ns
th(GPIO-COH)
Hold time, GPIOx input valid after CLKOUT high, GPIOx configured as general-purpose
input.
0
ns
Table 5–21. General-Purpose I/O Switching Characteristics
PARAMETER
td(COH-GPIO)
Delay time, CLKOUT high to GPIOx output change. GPIOx configured as
general-purpose output.
MIN
MAX
UNIT
–1
5
ns
CLKOUT
tsu(GPIO-COH)
th(GPIO-COH)
GPIOx Input Mode
td(COH-GPIO)
GPIOx Output Mode
Figure 5–18. GPIO Timings
70
SPRS098C
December 1999 – Revised November 2001
Electrical Specifications
5.15 Multichannel Buffered Serial Port (McBSP) Timing
5.15.1
McBSP Transmit and Receive Timings
The following timing requirements and switching characteristics tables assume testing over recommended
operating conditions and H = 0.5tc(CO) (see Figure 5–19 and Figure 5–20).
Table 5–22. McBSP Transmit and Receive Timing Requirements†
MIN
†
MAX
UNIT
tc(BCKRX)
Cycle time, BCLKR/X
BCLKR/X ext
4H
ns
tw(BCKRX)
Pulse duration, BCLKR/X low or BCLKR/X high
BCLKR/X ext
6
ns
BCLKR int
0
BCLKR ext
4
th(BCKRL-BFRH)
Hold time, external BFSR high after BCLKR low
th(BCKRL-BDRV)
Hold time, BDR valid after BCLKR low
th(BCKXL-BFXH)
Hold time, external BFSX high after BCLKX low
tsu(BFRH-BCKRL)
Setup time, external BFSR high before BCLKR low
tsu(BDRV-BCKRL)
Setup time, BDR valid before BCLKR low
tsu(BFXH-BCKXL)
Setup time, external BFSX high before BCLKX low
tr(BCKRX)
Rise time, BCLKR/X
BCLKR/X ext
8
ns
tf(BCKRX)
Fall time, BCLKR/X
BCLKR/X ext
8
ns
BCLKR int
0
BCLKR ext
5
BCLKX int
0
BCLKX ext
4
BCLKR int
10
BCLKR ext
4
BCLKR int
10
BCLKR ext
3
BCLKX int
10
BCLKX ext
6
ns
ns
ns
ns
ns
ns
Polarity bits CLKRP = CLKXP = FSRP = FSXP = 0. If the polarity of any of the signals is inverted, then the timing references of that signal are
also inverted.
December 1999 – Revised November 2001
SPRS098C
71
Electrical Specifications
Table 5–23. McBSP Transmit and Receive Switching Characteristics†
MIN
PARAMETER
MAX
UNIT
tc(BCKRX)
Cycle time, BCLKR/X
BCLKR/X int
4H
tw(BCKRXH)
Pulse duration, BCLKR/X high
BCLKR/X int
D–4‡
D+1‡
ns
ns
C–4‡
C+1‡
ns
ns
tw(BCKRXL)
Pulse duration, BCLKR/X low
BCLKR/X int
td(BCKRH-BFRV)
Delay time, BCLKR high to internal BFSR valid
BCLKR int
–3
3
BCLKX int
–3
8
BCLKX ext
2
15
td(BCKXH-BFXV)
Delay time, BCLKX high to internal BFSX valid
tdis(BCKXH-BDXHZ)
Disable time, BCLKX high to BDX high impedance following last data bit
Delay time, BCLKX high to BDX valid. This applies to all bits except the first
bit transmitted.
td(BCKXH-BDXV)
Delay time, BCLKX high to BDX valid.§
DXENA = 0
Only applies to first bit transmitted when in Data Delay 1
or 2 (XDATDLY=01b or 10b) modes
DXENA = 1
Enable time, BCLKX high to BDX driven.§
ten(BCKXH-BDX)
Only applies to first bit transmitted when in Data Delay 1
or 2 (XDATDLY=01b or 10b) modes
DXENA = 0
DXENA = 1
BCLKX int
–8
3
BCLKX ext
1
12
BCLKX int
–1
11
BCLKX ext
4
20
BCLKX int
11
BCLKX ext
20
BCLKX int
4H+6
BCLKX ext
td(BFXH-BDXV)
Only applies to first bit transmitted when in Data Delay 0
(XDATDLY=00b) mode.
Enable time, BFSX high to BDX driven.§
ten(BFXH-BDX)
Only applies to first bit transmitted when in Data Delay 0
(XDATDLY=00b) mode
DXENA = 0
DXENA = 1
DXENA = 0
DXENA = 1
ns
ns
4H+15
BCLKX int
5
BCLKX ext
16
BCLKX int
4H
BCLKX ext
4H+12
BFSX int
Delay time, BFSX high to BDX valid.§
ns
ns
9
BFSX ext
15
BFSX int
4H
BFSX ext
ns
4H+15
BFSX int
2
BFSX ext
14
BFSX int
4H–1
BFSX ext
2H+5
ns
†
Polarity bits CLKRP = CLKXP = FSRP = FSXP = 0. If the polarity of any of the signals is inverted, then the timing references of that signal are
also inverted.
‡ T=BCLKRX period = (1 + CLKGDV) * 2H
C=BCLKRX low pulse width = T/2 when CLKGDV is odd or zero and = (CLKGDV/2) * 2H when CLKGDV is even
D=BCLKRX high pulse width = T/2 when CLKGDV is odd or zero and = (CLKGDV/2 + 1) * 2H when CLKGDV is even
§ See the TMS320C54x DSP Reference Set, Volume 5: Enhanced Peripherals (literature number SPRU302) for a description of the DX enable
(DXENA) and data delay features of the McBSP.
72
SPRS098C
December 1999 – Revised November 2001
Electrical Specifications
tc(BCKRX)
tw(BCKRXH)
tr(BCKRX)
tw(BCKRXL)
BCLKR
td(BCKRH–BFRV)
td(BCKRH–BFRV)
tr(BCKRX)
BFSR (int)
tsu(BFRH–BCKRL)
th(BCKRL–BFRH)
BFSR (ext)
th(BCKRL–BDRV)
tsu(BDRV–BCKRL)
BDR
(RDATDLY=00b)
Bit (n–1)
(n–2)
tsu(BDRV–BCKRL)
(n–3)
(n–4)
th(BCKRL–BDRV)
BDR
(RDATDLY=01b)
Bit (n–1)
(n–2)
tsu(BDRV–BCKRL)
BDR
(RDATDLY=10b)
(n–3)
th(BCKRL–BDRV)
Bit (n–1)
(n–2)
Figure 5–19. McBSP Receive Timings
tc(BCKRX)
tw(BCKRXH)
tr(BCKRX)
tw(BCKRXL)
tf(BCKRX)
BCLKX
td(BCKXH–BFXV)
td(BCKXH–BFXV)
BFSX (int)
tsu(BFXH–BCKXL)
th(BCKXL–BFXH)
BFSX (ext)
ten(BDFXH–BDX)
BDX
(XDATDLY=00b)
Bit 0
td(BFXH–BDXV)
Bit (n–1)
td(BCKXH–BDXV)
(n–2)
ten(BCKXH–BDX)
BDX
(XDATDLY=01b)
Bit 0
Bit (n–1)
(n–4)
td(BCKXH–BDXV)
(n–2)
(n–3)
td(BCKXH–BDXV)
tdis(BCKXH–BDXHZ)
BDX
(XDATDLY=10b)
(n–3)
ten(BCKXH–BDX)
Bit 0
Bit (n–1)
(n–2)
Figure 5–20. McBSP Transmit Timings
December 1999 – Revised November 2001
SPRS098C
73
Electrical Specifications
5.15.2
McBSP Transmit and Receive Timing Using CLKR/X as a Clock Source Input to
the Sample Rate Generator (SRGR)
The 5421 McBSP has been enhanced to allow the use of an external clock source as an input to the sample
rate generator (SRGR). This capability is enabled by reconfiguring either the transmit shift clock (BCLKX), or
the receive shift clock (BCLKR) to function as the input clock to the SRGR. When the McBSP is used in this
mode, the output of the SRGR is then used as a common shift clock for both the receive and transmit sections
of the serial port. This clock is output on the other of these two pins. Therefore, if BCLKX is reconfigured as
the SRGR input, then BCLKR is used as the shift clock for both the transmit and receive sections of the McBSP.
If BCLKR is reconfigured as the SRGR input, then BCLKX is used as the shift clock for both the transmit and
receive sections of the McBSP. The relevant timings for this mode of operation are depicted in Figure 5–21.
The other timings for serial port operations are the same as when using an internal clock source as described
in the standard McBSP transmit and receive timings presented in section 5.15.1.
The following timing requirements and switching characteristics tables assume testing over recommended
operating conditions and H = 0.5tc(CO) (see Figure 5–21).
Table 5–24. McBSP Sample Rate Generator Timing Requirements
MIN
MAX
2H
UNIT
tc(BCKS)
Cycle time, SRGR clock input
tw(BCKSH)
Pulse duration, SRGR clock input high
H–4
H+1
ns
ns
tw(BCKSL)
Pulse duration, SRGR clock input low
H–4
H+1
ns
tr(BCKS)
Rise time, SRGR clock input
8
ns
tf(BCKS)
Fall time, SRGR clock input
8
ns
Table 5–25. McBSP Sample Rate Generator Switching Characteristics
PARAMETER
td(BCKSH-BCLKRXH)
74
SPRS098C
Delay time, from SRGR clock input to SRGR output
MIN
MAX
3
13
UNIT
ns
December 1999 – Revised November 2001
Electrical Specifications
tc(BCKS)
tw(BCKSH)
tr(BCKS)
tw(BCKSL)
SRGR Input
(BCLKX/BCLKR)
td(BCKSH–BCKRXH)
tf(BCKS)
SRGR Output
(BCLKR/BCLKX)
Receive Signals Referenced to Sample Rate Generator Output
BFSR
BDR
Bit (n–1)
(n–2)
(n–3)
(n–4)
Transmit Signals Referenced to Sample Rate Generator Output
BFSX
BDX
Bit 0
Bit (n–1)
(n–2)
(n–3)
(n–4)
Figure 5–21. McBSP Sample Rate Generator Timings
December 1999 – Revised November 2001
SPRS098C
75
Electrical Specifications
5.15.3
McBSP General-Purpose I/O Timing
The following timing requirements and switching characteristics tables assume testing over recommended
operating conditions (see Figure 5–22).
Table 5–26. McBSP General-Purpose I/O Timing Requirements
MIN
†
high†
tsu(BGPIO-COH)
Setup time, BGPIOx input mode before CLKOUT
th(COH-BGPIO)
Hold time, BGPIOx input mode after CLKOUT high†
MAX
UNIT
9
ns
0
ns
BGPIOx refers to BCLKRx, BFSRx, BDRx, BCLKXx, or BFSXx when configured as a general-purpose input.
Table 5–27. McBSP General-Purpose I/O Switching Characteristics
PARAMETER
Delay time, CLKOUT high to BGPIOx output mode‡
td(COH-BGPIO)
‡
MIN
MAX
–5
5
UNIT
ns
BGPIOx refers to BCLKRx, BFSRx, BCLKXx, BFSXx, or BDXx when configured as a general-purpose output.
tsu(BGPIO-COH)
td(COH-BGPIO)
CLKOUT
th(COH-BGPIO)
BGPIOx Input
Mode†
BGPIOx Output
Mode‡
†
‡
BGPIOx refers to BCLKRx, BFSRx, BDRx, BCLKXx, or BFSXx when configured as a general-purpose input.
BGPIOx refers to BCLKRx, BFSRx, BCLKXx, BFSXx, or BDXx when configured as a general-purpose output.
Figure 5–22. McBSP General-Purpose I/O Timings
76
SPRS098C
December 1999 – Revised November 2001
Electrical Specifications
5.15.4
McBSP as SPI Master or Slave Timing
The following timing requirements and switching characteristics tables assume testing over recommended
operating conditions and H = 0.5tc(CO) (see Figure 5–23, Figure 5–24, Figure 5–25, and Figure 5–26).
Table 5–28. McBSP as SPI Master or Slave Timing Requirements (CLKSTP = 10b, CLKXP = 0)†
MASTER
MIN
†
tsu(BDRV-BCKXL)
Setup time, BDR valid before BCLKX low
th(BCKXL-BDRV)
Hold time, BDR valid after BCLKX low
tsu(BFXL-BCKXH)
Setup time, BFSX low before BCLKX high
tc(BCKX)
Cycle time, BCLKX
SLAVE
MAX
MIN
MAX
UNIT
12
2 – 12H
ns
4
6 + 12H
ns
10
ns
32H
ns
12H
For all SPI slave modes, CLKG is programmed as 1/2 of the CPU clock by setting CLKSM = CLKGDV = 1.
Table 5–29. McBSP as SPI Master or Slave Switching Characteristics (CLKSTP = 10b, CLKXP = 0)†
MASTER‡
PARAMETER
MIN
SLAVE
MAX
MIN
MAX
UNIT
th(BCKXL-BFXL)
Hold time, BFSX low after BCLKX low§
T–5
T+6
td(BFXL-BCKXH)
Delay time, BFSX low to BCLKX high¶
C–5
C+5
td(BCKXH-BDXV)
Delay time, BCLKX high to BDX valid
–3
12
tdis(BCKXL-BDXHZ)
Disable time, BDX high impedance following last data bit from BCLKX
low
C–6
C +10
tdis(BFXH-BDXHZ)
Disable time, BDX high impedance following last data bit from BFSX
high
4H+ 4
8H + 17
ns
td(BFXL-BDXV)
Delay time, BFSX low to BDX valid
4H + 4
8H + 17
ns
ns
ns
6H + 4
10H + 19
ns
ns
†
For all SPI slave modes, CLKG is programmed as 1/2 of the CPU clock by setting CLKSM = CLKGDV = 1.
T = BCLKX period = (1 + CLKGDV) * 2H
C = BCLKX low pulse width = T/2 when CLKGDV is odd or zero and = (CLKGDV/2) * 2H when CLKGDV is even
§ FSRP = FSXP = 1. As a SPI master, BFSX is inverted to provide active-low slave-enable output. As a slave, the active-low signal input on BFSX
and BFSR is inverted before being used internally.
CLKXM = FSXM = 1, CLKRM = FSRM = 0 for master McBSP
CLKXM = CLKRM = FSXM = FSRM = 0 for slave McBSP
¶ BFSX should be low before the rising edge of clock to enable slave devices and then begin a SPI transfer at the rising edge of the master clock
(BCLKX).
‡
LSB
tsu(BFXL-BCKXH)
tc(BCKX)
MSB
BCLKX
th(BCKXL-BFXL)
td(BFXL-BCKXH)
BFSX
tdis(BFXH-BDXHZ)
td(BFXL-BDXV)
td(BCKXH-BDXV)
tdis(BCKXL-BDXHZ)
BDX
Bit 0
Bit(n-1)
tsu(BDRV-BCKXL)
BDR
Bit 0
(n-2)
(n-3)
(n-4)
th(BCKXL-BDRV)
Bit(n-1)
(n-2)
(n-3)
(n-4)
Figure 5–23. McBSP Timing as SPI Master or Slave: CLKSTP = 10b, CLKXP = 0
December 1999 – Revised November 2001
SPRS098C
77
Electrical Specifications
Table 5–30. McBSP as SPI Master or Slave Timing Requirements (CLKSTP = 11b, CLKXP = 0)†
MASTER
MIN
†
tsu(BDRV-BCKXH)
Setup time, BDR valid before BCLKX high
th(BCKXH-BDRV)
Hold time, BDR valid after BCLKX high
tsu(BFXL-BCKXH)
Setup time, BFSX low before BCLKX high
tc(BCKX)
Cycle time, BCLKX
SLAVE
MAX
MIN
MAX
UNIT
12
2 – 12H
ns
4
6 +12H
ns
10
ns
32H
ns
12H
For all SPI slave modes, CLKG is programmed as 1/2 of the CPU clock by setting CLKSM = CLKGDV = 1.
Table 5–31. McBSP as SPI Master or Slave Switching Characteristics (CLKSTP = 11b, CLKXP = 0)†
MASTER‡
PARAMETER
MIN
MAX
SLAVE
MIN
MAX
UNIT
th(BCKXL-BFXL)
Hold time, BFSX low after BCLKX low§
C–5
C+6
td(BFXL-BCKXH)
Delay time, BFSX low to BCLKX high¶
T–5
T+5
td(BCKXL-BDXV)
Delay time, BCLKX low to BDX valid
–3
12
6H + 4
10H + 19
ns
tdis(BCKXL-BDXHZ)
Disable time, BDX high impedance following last data bit from BCLKX
low
–6
10
6H + 4
10H + 17
ns
td(BFXL-BDXV)
Delay time, BFSX low to BDX valid
D–2
D +10
4H + 4
8H + 17
ns
ns
ns
†
For all SPI slave modes, CLKG is programmed as 1/2 of the CPU clock by setting CLKSM = CLKGDV = 1.
‡ T = BCLKX period = (1 + CLKGDV) * 2H
C = BCLKX low pulse width = T/2 when CLKGDV is odd or zero and = (CLKGDV/2) * 2H when CLKGDV is even
D = BCLKX high pulse width = T/2 when CLKGDV is odd or zero and = (CLKGDV/2 + 1) * 2H when CLKGDV is even
§ FSRP = FSXP = 1. As a SPI master, BFSX is inverted to provide active-low slave-enable output. As a slave, the active-low signal input on BFSX
and BFSR is inverted before being used internally.
CLKXM = FSXM = 1, CLKRM = FSRM = 0 for master McBSP
CLKXM = CLKRM = FSXM = FSRM = 0 for slave McBSP
¶ BFSX should be low before the rising edge of clock to enable slave devices and then begin a SPI transfer at the rising edge of the master clock
(BCLKX).
tsu(BFXL-BCKXH)
LSB
tc(BCKX)
MSB
BCLKX
th(BCKXL-BFXL)
td(BFXL-BCKXH)
BFSX
tdis(BCKXL-BDXHZ)
BDX
td(BCKXL-BDXV)
td(BFXL-BDXV)
Bit 0
Bit(n-1)
tsu(BDRV-BCKXH)
BDR
Bit 0
(n-2)
(n-3)
(n-4)
th(BCKXH-BDRV)
Bit(n-1)
(n-2)
(n-3)
(n-4)
Figure 5–24. McBSP Timing as SPI Master or Slave: CLKSTP = 11b, CLKXP = 0
78
SPRS098C
December 1999 – Revised November 2001
Electrical Specifications
Table 5–32. McBSP as SPI Master or Slave Timing Requirements (CLKSTP = 10b, CLKXP = 1)†
MASTER
MIN
†
tsu(BDRV-BCKXH)
Setup time, BDR valid before BCLKX high
th(BCKXH-BDRV)
Hold time, BDR valid after BCLKX high
tsu(BFXL-BCKXL)
Setup time, BFSX low before BCLKX low
tc(BCKX)
Cycle time, BCLKX
SLAVE
MAX
MIN
MAX
UNIT
12
2 – 12H
ns
4
6 + 12H
ns
10
ns
32H
ns
12H
For all SPI slave modes, CLKG is programmed as 1/2 of the CPU clock by setting CLKSM = CLKGDV = 1.
Table 5–33. McBSP as SPI Master or Slave Switching Characteristics (CLKSTP = 10b, CLKXP = 1)†
MASTER‡
PARAMETER
MIN
MAX
SLAVE
MIN
MAX
UNIT
th(BCKXH-BFXL)
Hold time, BFSX low after BCLKX high§
T–5
T+6
td(BFXL-BCKXL)
Delay time, BFSX low to BCLKX low¶
D–5
D+5
td(BCKXL-BDXV)
Delay time, BCLKX low to BDX valid
–3
12
tdis(BCKXH-BDXHZ)
Disable time, BDX high impedance following last data bit from BCLKX
high
D–6
D +10
tdis(BFXH-BDXHZ)
Disable time, BDX high impedance following last data bit from BFSX
high
4H + 4
8H + 17
ns
td(BFXL-BDXV)
Delay time, BFSX low to BDX valid
4H +4
8H + 17
ns
ns
ns
6H + 4
10H + 19
ns
ns
†
For all SPI slave modes, CLKG is programmed as 1/2 of the CPU clock by setting CLKSM = CLKGDV = 1.
T = BCLKX period = (1 + CLKGDV) * 2H
D = BCLKX high pulse width = T/2 when CLKGDV is odd or zero and = (CLKGDV/2 + 1) * 2H when CLKGDV is even
§ FSRP = FSXP = 1. As a SPI master, BFSX is inverted to provide active-low slave-enable output. As a slave, the active-low signal input on BFSX
and BFSR is inverted before being used internally.
CLKXM = FSXM = 1, CLKRM = FSRM = 0 for master McBSP
CLKXM = CLKRM = FSXM = FSRM = 0 for slave McBSP
¶ BFSX should be low before the rising edge of clock to enable slave devices and then begin a SPI transfer at the rising edge of the master clock
(BCLKX).
‡
LSB
tsu(BFXL-BCKXL)
tc(BCKX)
MSB
BCLKX
th(BCKXH-BFXL)
td(BFXL-BCKXL)
BFSX
td(BFXL-BDXV)
tdis(BFXH-BDXHZ)
td(BCKXL-BDXV)
tdis(BCKXH-BDXHZ)
BDX
Bit 0
Bit(n-1)
tsu(BDRV-BCKXH)
BDR
Bit 0
(n-2)
(n-3)
(n-4)
th(BCKXH-BDRV)
Bit(n-1)
(n-2)
(n-3)
(n-4)
Figure 5–25. McBSP Timing as SPI Master or Slave: CLKSTP = 10b, CLKXP = 1
December 1999 – Revised November 2001
SPRS098C
79
Electrical Specifications
Table 5–34. McBSP as SPI Master or Slave Timing Requirements (CLKSTP = 11b, CLKXP = 1)†
MASTER
MIN
†
tsu(BDRV-BCKXL)
Setup time, BDR valid before BCLKX low
th(BCKXL-BDRV)
Hold time, BDR valid after BCLKX low
tsu(BFXL-BCKXL)
Setup time, BFSX low before BCLKX low
tc(BCKX)
Cycle time, BCLKX
SLAVE
MAX
MIN
UNIT
MAX
12
2 – 12H
ns
4
6 + 12H
ns
10
ns
32H
ns
12H
For all SPI slave modes, CLKG is programmed as 1/2 of the CPU clock by setting CLKSM = CLKGDV = 1.
Table 5–35. McBSP as SPI Master or Slave Switching Characteristics (CLKSTP = 11b, CLKXP = 1)†
MASTER‡
PARAMETER
MIN
MAX
SLAVE
MIN
UNIT
MAX
th(BCKXH-BFXL)
Hold time, BFSX low after BCLKX high§
D–5
D+6
td(BFXL-BCKXL)
Delay time, BFSX low to BCLKX low¶
T–5
T+5
td(BCKXH-BDXV)
Delay time, BCLKX high to BDX valid
–3
12
6H + 4
10H + 19
ns
tdis(BCKXH-BDXHZ)
Disable time, BDX high impedance following last data bit from BCLKX
high
–6
10
6H + 4
10H + 17
ns
td(BFXL-BDXV)
Delay time, BFSX low to BDX valid
C–2
C +10
4H + 4
8H + 17
ns
ns
ns
†
For all SPI slave modes, CLKG is programmed as 1/2 of the CPU clock by setting CLKSM = CLKGDV = 1.
‡ T = BCLKX period = (1 + CLKGDV) * 2H
C = BCLKX low pulse width = T/2 when CLKGDV is odd or zero and = (CLKGDV/2) * 2H when CLKGDV is even
D = BCLKX high pulse width = T/2 when CLKGDV is odd or zero and = (CLKGDV/2 + 1) * 2H when CLKGDV is even
§ FSRP = FSXP = 1. As a SPI master, BFSX is inverted to provide active-low slave-enable output. As a slave, the active-low signal input on BFSX
and BFSR is inverted before being used internally.
CLKXM = FSXM = 1, CLKRM = FSRM = 0 for master McBSP
CLKXM = CLKRM = FSXM = FSRM = 0 for slave McBSP
¶ BFSX should be low before the rising edge of clock to enable slave devices and then begin a SPI transfer at the rising edge of the master clock
(BCLKX).
tsu(BFXL-BCKXL)
LSB
MSB
tc(BCKX)
BCLKX
th(BCKXH-BFXL)
td(BFXL-BCKXL)
BFSX
tdis(BCKXH-BDXHZ)
BDX
td(BCKXH-BDXV)
td(BFXL-BDXV)
Bit 0
Bit(n-1)
tsu(BDRV-BCKXL)
BDR
Bit 0
(n-2)
(n-3)
(n-4)
th(BCKXL-BDRV)
Bit(n-1)
(n-2)
(n-3)
(n-4)
Figure 5–26. McBSP Timing as SPI Master or Slave: CLKSTP = 11b, CLKXP = 1
80
SPRS098C
December 1999 – Revised November 2001
Electrical Specifications
5.16 Host-Port Interface Timing
The following timing requirements and switching characteristics tables assume testing over recommended
operating conditions and H = 0.5tc(CO) (see Figure 5–27 through Figure 5–34). In the following tables, DS
refers to the logical OR of HCS, HDS1, and HDS2, and HD refers to any of the HPI data bus pins (HD0, HD1,
HD2, etc.).
Table 5–36. HPI16 Mode Timing Requirements
MIN
edge†‡
tsu(HBV-DSL)
Setup time, HAD valid before DS falling
th(DSL-HBV)
Hold time, HAD valid after DS falling edge†‡
edge†
MAX
UNIT
5
ns
5
ns
tsu(HBV-HSL)
Setup time, HAD valid before HAS falling
5
ns
th(HSL-HBV)
Hold time, HAD valid after HAS falling edge†
5
ns
tsu(HAV-DSH)
Setup time, address valid before DS rising edge (nonmultiplexed write)‡
5
ns
tsu(HAV-DSL)
Setup time, address valid before DS falling edge (nonmultiplexed
read)‡
–(4H + 5)
ns
th(DSH-HAV)
Hold time, address valid after DS rising edge (nonmultiplexed mode)‡
1
ns
edge‡
tsu(HSL-DSL)
Setup time, HAS low before DS falling
5
ns
th(HSL-DSL)
Hold time, HAS low after DS falling edge‡
2
ns
tw(DSL)
Pulse duration, DS low‡
30
ns
10
ns
tw(DSH)
tc(DSH-DSH)§
tsu(HDV-DSH)W
Pulse duration, DS
high‡
Nonmultiplexed or multiplexed mode (no
increment) memory accesses (or writes to
the FETCH bit) with no DMA activity.
Reads
10H + 30
Writes
10H + 10
Nonmultiplexed or multiplexed mode (no
Cycle time, DS rising edge to
increment) memory accesses (or writes to
‡
next DS rising edge
the FETCH bit) with 16-bit DMA activity.
Reads
16H + 30
Writes
16H + 10
Nonmultiplexed or multiplexed mode (no
increment) memory accesses (or writes to
the FETCH bit) with 32-bit DMA activity.
Reads
24H + 30
Writes
24H + 10
ns
ns
ns
Multiplexed (autoincrement) memory accesses (or
Cycle time, DS rising edge to
writes to the FETCH bit) with no DMA activity.
‡
next DS rising edge
Multiplexed (autoincrement) memory accesses (or
(In
autoincrement
mode, writes to the FETCH bit) with 16-bit DMA activity.
WRITE timings are the same as Multiplexed (autoincrement) memory accesses (or
READ timings.)
writes to the FETCH bit) with 32-bit DMA activity.
10H + 10
ns
16H + 10
ns
24H + 10
ns
Cycle time, DS rising edge to next DS rising edge for writes to DSPINT and HINT‡
8H
ns
Cycle time, DS rising edge to next DS rising edge for HPIC reads, HPIC XADD bit
writes, and address register reads and writes‡
40
ns
Setup time, HD valid before DS rising edge‡
10
ns
write‡
th(DSH-HDV)W
Hold time, HD valid after DS rising edge,
1
ns
tsu(SELV-DSL)
Setup time, SELA/B valid before DS falling edge‡
5
ns
th(DSH-SELV)
Hold time, SELA/B valid after DS rising edge‡
0
ns
†
HAD stands for HCNTL0, HCNTL1, and HR/W.
DS refers to either HCS or HDS, whichever is controlling the transfer. Refer to the TMS320C54x DSP Reference Set, Volume 5: Enhanced
Peripherals (literature number SPRU302) for information regarding logical operation of the HPI16. These timings are shown assuming that HDS
is the signal controlling the transfer.
§ These timings are for HPI accesses which do not cross from one subsystem to the other. For accesses which do cross from one subsystem to
the other, additional cycles are required. A detailed description of these considerations is provided in the application note Memory Transfers with
TMS320VC5420 and TMS320VC5421 DSPs (literature number SPRA620).
‡
December 1999 – Revised November 2001
SPRS098C
81
Electrical Specifications
Table 5–37. HPI16 Mode Switching Characteristics
PARAMETER
td(DSL-HDD)
MIN
MAX
UNIT
3
20
ns
Case 1a: Memory accesses initiated immediately following a write
when DMAC is active in 16-bit mode and tw(DSH) was < 18H
32H+20 – tw(DSH)
ns
Case 1b: Memory accesses initiated by an autoincrement when
DMAC is active in 16-bit mode and tw(DSH) was < 18H
16H+20 – tw(DSH)
ns
Delay time, DS low to HD
Delay time, DS
low to HD valid
td(DSL-HDV1)#
for first word of
an HPI read†
driven†
Case 1c: Memory accesses not initiated by an autoincrement (or not
immediately following a write) when DMAC is active in 16-bit mode
16H+20
ns
Case 1d: Memory accesses initiated by an autoincrement when
DMAC is active in 16-bit mode and tw(DSH) was ≥ 18H
20
Case 1e: Memory accesses initiated immediately following a write
when DMAC is active in 32-bit mode and tw(DSH) was < 26H
48H+20 – tw(DSH)
Case 1f: Memory access initiated by an autoincrement when DMAC
is active in 32-bit mode and tw(DSH) was < 26H
24H+20 – tw(DSH)
Case 1g: Memory access not initiated by an autoincrement (or not
immediately following a write) when DMAC is active in 32-bit mode
24H+20
Case 1h: Memory access initiated by an autoincrement when DMAC
is active in 32-bit mode and tw(DSH) was ≥ 26H
Delay time,
HAS low to HD
td(HSL-HDV1)# valid for first
word of an HPI
read
20H+20 – tw(DSH)
Case 2b: Memory accesses initiated by an autoincrement when
DMAC is inactive and tw(DSH) was < 10H
10H+20 – tw(DSH)
Case 2c: Memory accesses not initiated by an autoincrement (or not
immediately following a write) when DMAC is inactive
Case 3: HPIC/HPIA reads
20
td(DSH-HYH)#
tv(HYH-HDV)
Valid time, HD valid after HRDY high
th(DSH-HDV)R
Hold time, HD valid after DS rising edge, read‡
td(COH-HYH)
Delay time, CLKOUT rising edge to HRDY high
td(DSL-HYL)
td(DSH-HYL)
ns
20
Memory accesses (or writes to the FETCH bit) when no DMA is
Delay time, DS active
high to HRDY Memory accesses (or writes to the FETCH bit) with one or more
high§
16-bit DMA channels active
(writes
and
Memory accesses (or writes to the FETCH bit) with one or more
autoincrement
32-bit DMA channels active
reads)
Writes to DSPINT and HINT‡
Delay time, DS low to HRDY
low¶
Delay time, DS high to HRDY
low¶
ns
10H+20
Case 2d: Memory accesses initiated by an autoincrement when
DMAC is inactive and tw(DSH) was ≥ 10H
Multiplexed reads with autoincrement. Prefetch completed.
ns
20
Case 2a: Memory accesses initiated immediately following a write
when DMAC is inactive and tw(DSH) was < 10H
td(DSL-HDV2)
ns
3
20
ns
10H+5
16H+5
ns
24H+5
4H + 5
0
7
ns
10
ns
5
ns
12
ns
12
ns
†
HAD stands for HCNTL0, HCNTL1, and HR/W.
‡ HDS refers to either HDS1 or HDS2.
§ DS refers to either HCS or HDS, whichever is controlling the transfer. Refer to the TMS320C54x DSP Reference Set, Volume 5: Enhanced
Peripherals (literature number SPRU302) for information regarding logical operation of the HPI16. These timings are shown assuming that
HDS is the signal controlling the transfer.
¶ HRDY does not go low for other register accesses.
# These timings are for HPI accesses which do not cross from one subsystem to the other. For accesses which do cross from one subsystem
to the other, additional cycles are required. A detailed description of these considerations is provided in the application note Memory Transfers
with TMS320VC5420 and TMS320VC5421 DSPs (literature number SPRA620).
82
SPRS098C
December 1999 – Revised November 2001
Electrical Specifications
Table 5–37. HPI16 Mode Switching Characteristics (Continued)
PARAMETER
MIN
MAX
UNIT
td(HSL-HYL)
Delay time, HAS low to HRDY low, read
12
ns
td(COH–HTX)
Delay time, CLKOUT rising edge to HINT change
5
ns
†
HAD stands for HCNTL0, HCNTL1, and HR/W.
HDS refers to either HDS1 or HDS2.
§ DS refers to either HCS or HDS, whichever is controlling the transfer. Refer to the TMS320C54x DSP Reference Set, Volume 5: Enhanced
Peripherals (literature number SPRU302) for information regarding logical operation of the HPI16. These timings are shown assuming that
HDS is the signal controlling the transfer.
¶ HRDY does not go low for other register accesses.
# These timings are for HPI accesses which do not cross from one subsystem to the other. For accesses which do cross from one subsystem
to the other, additional cycles are required. A detailed description of these considerations is provided in the application note Memory Transfers
with TMS320VC5420 and TMS320VC5421 DSPs (literature number SPRA620).
‡
HCS
tsu(HSL–DSL)
th(HSL–DSL)
HAS
tc(DSH–DSH)
tsu(HBV–HSL)
HDS
tw(DSH)
th(HSL–HBV)
tw(DSL)
HR/W
01
HCNTL[1:0]
01
th(DSH–HDV)R
td(HSL–HDV1)
td(DSL–HDV2)
Data 1
HD[15:0]
td(DSL–HDD)
PF Data
td(DSH–HYL)†
HRDY
td(HSL–HYL)
td(DSH–HYH)†
tv(HYH–HDV)
†
HRDY goes low at these times only after autoincrement reads.
Figure 5–27. Multiplexed Read Timings Using HAS
December 1999 – Revised November 2001
SPRS098C
83
Electrical Specifications
HCS
tsu(HBV–DSL)
tc(DSH–DSH)
HDS
tw(DSH)
th(DSL–HBV)
tw(DSL)
HR/W
01
HCNTL[1:0]
01
th(DSH–HDV)R
td(DSL–HDV1)
td(DSL–HDV2)
PF Data
Data 1
HD[15:0]
td(DSL–HDD)
td(DSH–HYL)†
Á
Á
HRDY
td(DSL–HYL)
td(DSH–HYH)†
tv(HYH–HDV)
†
HRDY goes low at these times only after autoincrement reads.
Figure 5–28. Multiplexed Read Timings With HAS Held High
84
SPRS098C
December 1999 – Revised November 2001
Electrical Specifications
HCS
tsu(HBV–HSL)
th(HSL–DSL)
HAS
tsu(HSL–DSL)
HR/W
th(HSL–HBV)
HCNTL[1:0]
01
01
tc(DSH–DSH)
HDS
tw(DSH)
tw(DSL)
tsu(HDV–DSH)W
HD[15:0]
Data 1
Data 2
th(DSH–HDV)W
HRDY
td(DSH–HYL)
td(DSH–HYH)
Figure 5–29. Multiplexed Write Timings Using HAS
December 1999 – Revised November 2001
SPRS098C
85
Electrical Specifications
HCS
tc(DSH–DSH)
tw(DSH)
HDS
tw(DSL)
tsu(HBV–DSL)
HR/W
th(DSL–HBV)
HCNTL[1:0]
01
01
tsu(HDV–DSH)W
th(DSH–HDV)W
Data 1
HD[15:0]
Data 2
td(DSH–HYL)
HRDY
td(DSH–HYH)
Figure 5–30. Multiplexed Write Timings With HAS Held High
HCS
tw(DSH)
tc(DSH–DSH)
HDS
tsu(HBV–DSL)
tw(DSL)
tsu(HBV–DSL)
th(DSL–HBV)
th(DSL–HBV)
HR/W
tsu(HAV–DSL)
th(DSH–HAV)
HA[17:0]
Valid Address
Valid Address
th(DSH–HDV)R
td(DSL–HDV1)
td(DSL–HDV1)
th(DSH–HDV)R
Data
HD[15:0]
td(DSL–HDD)
tv(HYH–HDV)
Data
td(DSL–HDD)
tv(HYH–HDV)
HRDY
td(DSL–HYL)
td(DSL–HYL)
Figure 5–31. Nonmultiplexed Read Timings
86
SPRS098C
December 1999 – Revised November 2001
Electrical Specifications
HCS
tw(DSH)
tc(DSH–DSH)
HDS
tsu(HBV–DSL)
tsu(HBV–DSL)
th(DSL–HBV)
th(DSL–HBV)
HR/W
tsu(HAV–DSH)
tw(DSL)
th(DSH–HAV)
Valid Address
HA[15:0]
Valid Address
tsu(HDV–DSH)W
tsu(HDV–DSH)W
th(DSH–HDV)W
th(DSH–HDV)W
Data Valid
HD[15:0]
Data Valid
td(DSH–HYH)
HRDY
td(DSH–HYL)
Figure 5–32. Nonmultiplexed Write Timings
HRDY
td(COH–HYH)
CLKOUT
td(COH–HTX)
HINT
Figure 5–33. HRDY and HINT Relative to CLKOUT
HCS
tsu(SELV–DSL)
th(DSH–SELV)
SELA/B
HDS
ÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁ
ÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁ
Figure 5–34. SELA/B Timing
December 1999 – Revised November 2001
SPRS098C
87
Mechanical Data
6
Mechanical Data
6.1
Ball Grid Array Mechanical Data
GGU (S-PBGA-N144)
PLASTIC BALL GRID ARRAY
12,10
SQ
11,90
13 12 11 10 9 8 7 6 5 4 3 2 1
A
B
C
D
E
F
G
H
J
K
L
M
N
0,80
0,80
0,95
0,85
0,12
0,08
1,40 MAX
0,55
0,45
0,08 M
0,45
0,35
0,10
4073221/A 11/96
NOTES: A. All linear dimensions are in millimeters.
B. This drawing is subject to change without notice.
Figure 6–1. MicroStar BGA Package
MicroStar BGA is a trademark of Texas Instruments.
88
SPRS098C
December 1999 – Revised November 2001
Mechanical Data
6.2
Low Profile Quad Flatpack Mechanical Data
PGE (S-PQFP-G144)
Low-Profile Quad Flatpack
108
73
109
72
0,27
0,17
0,08 M
0,50
144
0,13 NOM
37
1
36
Gage Plane
17,50 TYP
20,20 SQ
19,80
22,20
SQ
21,80
0,25
0,05 MIN
0°–ā7°
0,75
0,45
1,45
1,35
Seating Plane
0,08
1,60 MAX
4040147/C 10/96
NOTES: A. All linear dimensions are in millimeters.
B. This drawing is subject to change without notice.
C. Falls within JEDEC MO-136
Figure 6–2. Low-Profile Quad Flatpack
December 1999 – Revised November 2001
SPRS098C
89