SGUS055 − SEPTEMBER 2004 D Controlled Baseline D D D D D D D D L1/L2 Memory Architecture − One Assembly/Test Site, One Fabrication Site Enhanced Diminishing Manufacturing Sources (DMS) Support Enhanced Product-Change Notification Qualification Pedigree† Low-Price/High-Performance Floating-Point Digital Signal Processors (DSPs): 320C67x (SM320C6712, C6712C, C6712D) − Eight 32-Bit Instructions/Cycle − 100-, 167-MHz Clock Rates − 10-, 6-ns Instruction Cycle Times − 600, 1000 MFLOPS Advanced Very Long Instruction Word (VLIW) C67x DSP Core − Eight Highly Independent Functional Units: − Four ALUs (Floating- and Fixed-Point) − Two ALUs (Fixed-Point) − Two Multipliers (Floating- and Fixed-Point) − Load-Store Architecture With 32 32-Bit General-Purpose Registers − Instruction Packing Reduces Code Size − All Instructions Conditional Instruction Set Features − Hardware Support for IEEE Single-Precision and Double-Precision Instructions − Byte-Addressable (8-, 16-, 32-Bit Data) − 8-Bit Overflow Protection − Saturation − Bit-Field Extract, Set, Clear − Bit-Counting − Normalization Device Configuration − Boot Mode: 8- and 16-Bit ROM Boot − Endianness: Little Endian (12/12C) Little Endian, Big Endian (12D) D D D D D D D D D − 32K-Bit (4K-Byte) L1P Program Cache (Direct Mapped) − 32K-Bit (4K-Byte) L1D Data Cache (2-Way Set-Associative) − 512K-Bit (64K-Byte) L2 Unified Mapped RAM/Cache (Flexible Data/Program Allocation) Enhanced Direct-Memory-Access (EDMA) Controller (16 Independent Channels) 16-Bit External Memory Interface (EMIF) − Glueless Interface to Asynchronous Memories: SRAM and EPROM − Glueless Interface to Synchronous Memories: SDRAM and SBSRAM − 256M-Byte Total Addressable External Memory Space Two Multichannel Buffered Serial Ports (McBSPs) − Direct Interface to T1/E1, MVIP, SCSA Framers − ST-Bus-Switching Compatible − Up to 256 Channels Each − AC97-Compatible − Serial-Peripheral-Interface (SPI) Compatible (Motorola) Two 32-Bit General-Purpose Timers Flexible Phase-Locked-Loop (PLL) Clock Generator [C6712] Flexible Software-Configurable PLL-Based Clock Generator Module [C6712C/C6712D] A Dedicated General-Purpose Input/Output (GPIO) Module With 5 Pins [12C/12D] IEEE-1149.1 (JTAG‡) Boundary-Scan-Compatible CMOS Technology − 0.13-µm/6-Level Copper Metal Process (C6712C/C6712D) − 0.18-µm/5-Level Metal Process (C6712) Please be aware that an important notice concerning availability, standard warranty, and use in critical applications of Texas Instruments semiconductor products and disclaimers thereto appears at the end of this data sheet. 320C67x and C67x are trademarks of Texas Instruments. Motorola is a trademark of Motorola, Inc. Other trademarks are the property of their respective owners. † Component qualification in accordance with JEDEC and industry standards to ensure reliable operation over an extended temperature range. This includes, but is not limited to, Highly Accelerated Stress Test (HAST) or biased 85/85, temperature cycle, autoclave or unbiased HAST, electromigration, bond intermetallic life, and mold compound life. Such qualification testing should not be viewed as justifying use of this component beyond specified performance and environmental limits. ‡ IEEE Standard 1149.1-1990 Standard-Test-Access Port and Boundary Scan Architecture. Copyright 2004, Texas Instruments Incorporated !"# "#$" "%&$#" " '&# " &! #$" "! '$! % !(!)'!"#* ! #$# % !$ !(! "$#! " #! '$+!,'!%."+ # !)!#&$) $&$#!&#* POST OFFICE BOX 1443 • HOUSTON, TEXAS 77251−1443 1 SGUS055 − SEPTEMBER 2004 Table of Contents GFN BGA package (bottom view) [C6712 only] . . . . . . . . . . 3 GDP BGA package (bottom view) [C6712C/12D only] . . . . 3 description . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 device characteristics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 device compatibility . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6 functional block and CPU (DSP core) diagram . . . . . . . . . . . 7 CPU (DSP core) description . . . . . . . . . . . . . . . . . . . . . . . . . . 8 memory map summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10 peripheral register descriptions . . . . . . . . . . . . . . . . . . . . . . . 11 signal groups description . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16 device configurations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18 terminal functions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20 development support . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33 documentation support . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36 CPU CSR register description . . . . . . . . . . . . . . . . . . . . . . . . 37 cache configuration (CCFG) register description (12D) . . . 39 interrupt sources and interrupt selector [C6712 only] . . . . 40 interrupt sources and interrupt selector [12C/12D only] . . 41 EDMA channel synchronization events [C6712 only] . . . . 42 EDMA module and EDMA selector [12C/12D only] . . . . . . 43 clock PLL [C6712 only] . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45 PLL and PLL controller [C6712C/C6712D only] . . . . . . . . . 47 general-purpose input/output (GPIO) . . . . . . . . . . . . . . . . . . 54 power-down mode logic . . . . . . . . . . . . . . . . . . . . . . . . . . . . . power-supply sequencing . . . . . . . . . . . . . . . . . . . . . . . . . . . . power-supply decoupling . . . . . . . . . . . . . . . . . . . . . . . . . . . . IEEE 1149.1 JTAG compatibility statement . . . . . . . . . . . . . 2 POST OFFICE BOX 1443 55 57 59 59 EMIF device speed . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . EMIF big endian mode correctness [C6712D only] . . . bootmode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . absolute maximum ratings over operating case temperature range . . . . . . . . . . . . . . . . . . . . . . . . . . recommended operating conditions . . . . . . . . . . . . . . . . electrical characteristics over recommended ranges of supply voltage and operating case temperature . 59 60 61 61 62 63 parameter measurement information . . . . . . . . . . . . . . . 64 signal transition levels . . . . . . . . . . . . . . . . . . . . . . . . . . 65 timing parameters and board routing analysis . . . . . . 65 input and output clocks . . . . . . . . . . . . . . . . . . . . . . . . . . . 67 asynchronous memory timing . . . . . . . . . . . . . . . . . . . . . 72 synchronous-burst memory timing . . . . . . . . . . . . . . . . . 75 synchronous DRAM timing . . . . . . . . . . . . . . . . . . . . . . . . 77 HOLD/HOLDA timing . . . . . . . . . . . . . . . . . . . . . . . . . . . . 83 BUSREQ timing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 84 reset timing [C6712] . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85 reset timing [C6712C/C6712D] . . . . . . . . . . . . . . . . . . . . 87 external interrupt timing . . . . . . . . . . . . . . . . . . . . . . . . . . 89 multichannel buffered serial port timing . . . . . . . . . . . . . 90 timer timing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 104 general-purpose input/output (GPIO) port timing [C6712C/C6712D only] . . . . . . . . . . . . . . . . . . . . . 105 JTAG test-port timing . . . . . . . . . . . . . . . . . . . . . . . . . . . 106 mechanical data [C6712 only] . . . . . . . . . . . . . . . . . . . . 107 mechanical data [C6712C/C6712D only] . . . . . . . . . . . 108 • HOUSTON, TEXAS 77251−1443 SGUS055 − SEPTEMBER 2004 GFN BGA package (bottom view) [C6712 only] GFN 256-PIN BALL GRID ARRAY (BGA) PACKAGE ( BOTTOM VIEW ) Y W V U T R P N M L K J H G F E D C B A 1 3 2 5 4 7 6 9 8 11 10 13 12 15 14 17 16 19 18 20 GDP BGA package (bottom view) [C6712C/12D only] GDP 272-PIN BALL GRID ARRAY (BGA) PACKAGE ( BOTTOM VIEW ) Y W V U T R P N M L K J H G F E D C B A 3 1 2 5 4 7 6 9 8 POST OFFICE BOX 1443 11 13 15 17 19 10 12 14 16 18 20 • HOUSTON, TEXAS 77251−1443 3 SGUS055 − SEPTEMBER 2004 description The 320C67x DSPs (including the SM320C6712-EP, SM320C6712C-EP, SM320C6712D-EP devices†) are members of the floating-point DSP family in the TMS320C6000 DSP platform. The C6712, C6712C, and C6712D devices are based on the high-performance, advanced very-long-instruction-word (VLIW) architecture developed by Texas Instruments (TI), making these DSPs an excellent choice for multichannel and multifunction applications. With performance of up to 1000 million floating-point operations per second (MFLOPS) at a clock rate of 167 MHz, the C6712C/C6712D device is the lowest-cost DSP in the C6000 DSP platform. The C6712C/C6712D DSP possesses the operational flexibility of high-speed controllers and the numerical capability of array processors. This processor has 32 general-purpose registers of 32-bit word length and eight highly independent functional units. The eight functional units provide four floating-/fixed-point ALUs, two fixed-point ALUs, and two floating-/fixed-point multipliers. The C6712C/C6712D can produce two MACs per cycle for a total of 300 MMACS. With performance of up to 600 million floating-point operations per second (MFLOPS) at a clock rate of 100 MHz, the C6712 device also offers cost-effective solutions to high-performance DSP programming challenges. The C6712 DSP possesses the operational flexibility of high-speed controllers and the numerical capability of array processors. This processor has 32 general-purpose registers of 32-bit word length and eight highly independent functional units. The eight functional units provide four floating-/fixed-point ALUs, two fixed-point ALUs, and two floating-/fixed-point multipliers. The C6712 can produce two multiply-accumulates (MACs) per cycle for a total of 200 million MACs per second (MMACS). The C6712/C6712C/C6712D uses a two-level cache-based architecture and has a powerful and diverse set of peripherals. The Level 1 program cache (L1P) is a 32-Kbit direct mapped cache and the Level 1 data cache (L1D) is a 32-Kbit 2-way set-associative cache. The Level 2 memory/cache (L2) consists of a 512-Kbit memory space that is shared between program and data space. L2 memory can be configured as mapped memory, cache, or combinations of the two. The peripheral set includes two multichannel buffered serial ports (McBSPs), two general-purpose timers, and a glueless 16-bit external memory interface (EMIF) capable of interfacing to SDRAM, SBSRAM, and asynchronous peripherals. The C6712C device also includes a dedicated general-purpose input/output (GPIO) peripheral module. The C6712/C6712C/C6712D DSPs also have application-specific hardware logic, on-chip memory, and additional on-chip peripherals. The C6712/C6712C/C6712D has a complete set of development tools which includes: a new C compiler, an assembly optimizer to simplify programming and scheduling, and a Windows debugger interface for visibility into source code execution. TMS320C6000 and C6000 are trademarks of Texas Instruments. Windows is a registered trademark of the Microsoft Corporation. † Throughout the remainder of this document, the SM320C6712-EP, SM320C6712C-EP, and SM320C6712D-EP shall be referred to as 320C67x or C67x where generic, and where specific, their individual full device part numbers will be used or abbreviated as C6712, C6712C, C6712D, 12, 12C, or 12D, etc. 4 POST OFFICE BOX 1443 • HOUSTON, TEXAS 77251−1443 SGUS055 − SEPTEMBER 2004 device characteristics Table 1 provides an overview of the C6712/C6712C/C6712D DSPs. The table shows significant features of each device, including the capacity of on-chip RAM, the peripherals, the execution time, and the package type with pin count. For more details on the C6000 DSP device part numbers and part numbering, see Table 17 and Figure 5. Table 1. Characteristics of the C6712, C6712C, and C6712D Processors INTERNAL CLOCK SOURCE HARDWARE FEATURES C6712 (FLOATING-POINT DSP) ECLKIN EMIF Peripherals 1 SYSCLK3 or ECLKIN EDMA 1 CPU clock frequency McBSPs 32-Bit Timers GPIO Module 1 1 CPU/2 clock frequency 2 — SYSCLK2 — 2 CPU/4 clock frequency 2 — 1/2 of SYSCLK2 — 2 SYSCLK2 — 1 72K 72K Size (Bytes) On-Chip Memory 4K-Byte (4KB) L1 Program (L1P) Cache 4KB L1 Data (L1D) Cache 64KB Unified Mapped RAM/Cache (L2) Organization CPU ID+ CPU Rev ID Control Status Register (CSR.[31:16]) Frequency MHz Cycle Time ns Voltage C6712C/C6712D (FLOATING-POINT DSPs) 0x0202 0x0203 100 167 10 ns (C6712-100) 6 ns (C6712D-167) 6 ns (C6712C-167) Core (V) 1.8 1.20‡ I/O (V) 3.3 3.3 PLL Options CLKIN frequency multiplier Bypass (x1), x4 − Clock Generator Options Prescaler Multiplier Postscaler — /1, /2, /3, ..., /32 x4, x5, x6, ..., x25 /1, /2, /3, ..., /32 BGA Package 27 x 27 mm 256-Pin BGA (GFN) 272-Pin BGA (GDP) Process Technology µm 0.18 µm 0.13 µm PP† PP (C6712C)† PD (C6712D)† Product Status Product Preview (PP) Advance Information (AI) Production Data (PD) † PRODUCT PREVIEW information concerns products in the formative or design phase of development. Characteristic data and other specifications are design goals. Texas Instruments reserves the right to change or discontinue these products without notice. ADVANCE INFORMATION concerns new products in the sampling or preproduction phase of development. Characteristic data and other specifications are subject to change without notice. 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. ‡ This value is compatible with existing 1.26V designs. POST OFFICE BOX 1443 • HOUSTON, TEXAS 77251−1443 5 SGUS055 − SEPTEMBER 2004 device compatibility The 320C6712 and C6211/C6711 devices are pin-compatible; thus, making new system designs easier and providing faster time to market. The following list summarizes the device characteristic differences among the C6211, C6211B, C6711, C6711B, C6711C, C6711D, C6712, C6712C, and C6712D devices: D The C6211 and C6211B devices have a fixed-point TMS320C62x DSP core (CPU), while the C6711, C6711B, C6711C, C6711D, C6712, C6712C, and C6712D devices have a floating-point C67x CPU. D The C6211, C6211B, C6711, C6711B, C6711C, and C6711D devices have a 32-bit EMIF, while the C6712, C6712C, and C6712D devices have a 16-bit EMIF. D The C6211, C6211B, C6711, C6711B, C6711C, and C6711D devices feature an HPI, while the C6712, C6712C, and C6712D devices do not. D The C6712, C6712C, and C6712D devices have dedicated device configuration pins, BOOTMODE, LENDIAN, and EMIFBE (12D only) that specify the boot-load operation and device endianness, respectively, during reset. On the C6211/C6211B and C6711/C6711B/C6711C/C6711D devices, these configuration pins are integrated with the HPI pins. D The C6211/C6211B device runs at -167 and -150 MHz clock speeds (with a C6211BGFNA extended temperature device that also runs at -150 MHz), while the C6711/C6711B device runs at -150 and -100 MHz (with a C6711BGFNA extended temperature device that also runs at -100 MHz) and the C6711C/C6711D device runs at -200 clock speed (with a C6711CGDPA extended temperature device that also runs at -167 MHz). The C6712 device runs at -100 MHz clock speed and the C6712C/C6712D device runs at -167 MHz clock speed. D The C6211/C6211B, C6711-100, C6711B and C6712 devices have a core voltage of 1.8 V, the C6711-150 device has a core voltage is 1.9 V, and the C6711C/C6711D and C6712C/C6712D devices operate with a core voltage of 1.20† V. D There are several enhancements and features that are only available on the C6711C/C6711D and C6712C/C6712D devices, such as: the CLKOUT3 signal, a software-programmable PLL and PLL Controller, and a GPIO peripheral module. The C6711D and C6712D devices also have additional enhancements such as: EMIF Big Endian mode correctness EMIFBE and the L1D requestor priority to L2 bit [“P” bit] in the cache configuration (CCFG) register. C6712D supports Big Endian mode. D The C6712/C6712C/C6712D is the lowest-cost entry in the TMS320C6000 platform. For a more detailed discussion on the similarities/differences among the C6211, C6711, and C6712 devices, see the How to Begin Development Today with the TMS320C6211 DSP, How to Begin Development with the TMS320C6711 DSP, and How to Begin Development With the TMS320C6712 DSP application reports (literature number SPRA474, SPRA522, and SPRA693, respectively). For a more detailed discussion on the migration of a C6211, C6211B, C6711, or C6711B device to a TMS320C6711C device, see the Migrating from TMS320C6211(B)/6711(B) to TMS320C6711C application report (literature number SPRA837). For a more detailed discussion on the migration of a C6712 device to a TMS320C6712C device, see the Migrating from TMS320C6712 to TMS320C6712C application report (literature number SPRA852). TMS320C62x and C67x are trademarks of Texas Instruments. † This value is compatible with existing 1.26V designs. 6 POST OFFICE BOX 1443 • HOUSTON, TEXAS 77251−1443 SGUS055 − SEPTEMBER 2004 functional block and CPU (DSP core) diagram C6712/C6712C/C6712D Digital Signal Processors SDRAM SBSRAM SRAM 16 External Memory Interface (EMIF) L1P Cache Direct Mapped 4K Bytes Total ROM/FLASH I/O Devices Timer 0 C67x CPU (DSP Core) Timer 1 Framing Chips: H.100, MVIP, SCSA, T1, E1 AC97 Devices, SPI Devices, Codecs Multichannel Buffered Serial Port 1 (McBSP1) Instruction Fetch Enhanced DMA Controller (16 channel) L2 Memory 4 Banks 64K Bytes Total Instruction Dispatch Instruction Decode Data Path A A Register File Multichannel Buffered Serial Port 0 (McBSP0) .L1† .S1† .M1† .D1 Data Path B Control Registers Control Logic Test B Register File In-Circuit Emulation .D2 .M2† .S2† .L2† Interrupt Control L1D Cache 2-Way Set Associative 4K Bytes Total Interrupt Selector PLL‡ GPIO§ Power-Down Logic Boot Configuration † In addition to fixed-point instructions, these functional units execute floating-point instructions. ‡ The C6712C/C6712D device has a software-configurable PLL (with x4 through x25 multiplier and /1 through /32 divider) and a PLL Controller which is different from the hardware PLL peripheral on the C6712 device. § Applicable to the C6712C/C6712D device only POST OFFICE BOX 1443 • HOUSTON, TEXAS 77251−1443 7 SGUS055 − SEPTEMBER 2004 CPU (DSP core) description The CPU fetches advanced very-long instruction words (VLIW) (256 bits wide) to supply up to eight 32-bit instructions to the eight functional units during every clock cycle. The VLIW architecture features controls by which all eight units do not have to be supplied with instructions if they are not ready to execute. The first bit of every 32-bit instruction determines if the next instruction belongs to the same execute packet as the previous instruction, or whether it should be executed in the following clock as a part of the next execute packet. Fetch packets are always 256 bits wide; however, the execute packets can vary in size. The variable-length execute packets are a key memory-saving feature, distinguishing the C67x CPU from other VLIW architectures. The CPU features two sets of functional units. Each set contains four units and a register file. One set contains functional units .L1, .S1, .M1, and .D1; the other set contains units .D2, .M2, .S2, and .L2. The two register files each contain 16 32-bit registers for a total of 32 general-purpose registers. The two sets of functional units, along with two register files, compose sides A and B of the CPU [see the functional block and CPU (DSP Core) diagram and Figure 1]. The four functional units on each side of the CPU can freely share the 16 registers belonging to that side. Additionally, each side features a single data bus connected to all the registers on the other side, by which the two sets of functional units can access data from the register files on the opposite side. While register access by functional units on the same side of the CPU as the register file can service all the units in a single clock cycle, register access using the register file across the CPU supports one read and one write per cycle. The C67x CPU executes all C62x DSP instructions. In addition to C62x fixed-point DSP instructions, the six out of eight functional units (.L1, .M1, .D1, .D2, .M2, and .L2) also execute floating-point instructions. The remaining two functional units (.S1 and .S2) also execute the new LDDW instruction which loads 64 bits per CPU side for a total of 128 bits per cycle. Another key feature of the C67x CPU is the load/store architecture, where all instructions operate on registers (as opposed to data in memory). Two sets of data-addressing units (.D1 and .D2) are responsible for all data transfers between the register files and the memory. The data address driven by the .D units allows data addresses generated from one register file to be used to load or store data to or from the other register file. The C67x CPU supports a variety of indirect addressing modes using either linear- or circular-addressing modes with 5- or 15-bit offsets. All instructions are conditional, and most can access any one of the 32 registers. Some registers, however, are singled out to support specific addressing or to hold the condition for conditional instructions (if the condition is not automatically “true”). The two .M functional units are dedicated for multiplies. The two .S and .L functional units perform a general set of arithmetic, logical, and branch functions with results available every clock cycle. The processing flow begins when a 256-bit-wide instruction fetch packet is fetched from a program memory. The 32-bit instructions destined for the individual functional units are “linked” together by “1” bits in the least significant bit (LSB) position of the instructions. The instructions that are “chained” together for simultaneous execution (up to eight in total) compose an execute packet. A “0” in the LSB of an instruction breaks the chain, effectively placing the instructions that follow it in the next execute packet. If an execute packet crosses the fetch-packet boundary (256 bits wide), the assembler places it in the next fetch packet, while the remainder of the current fetch packet is padded with NOP instructions. The number of execute packets within a fetch packet can vary from one to eight. Execute packets are dispatched to their respective functional units at the rate of one per clock cycle and the next 256-bit fetch packet is not fetched until all the execute packets from the current fetch packet have been dispatched. After decoding, the instructions simultaneously drive all active functional units for a maximum execution rate of eight instructions every clock cycle. While most results are stored in 32-bit registers, they can be subsequently moved to memory as bytes or half-words as well. All load and store instructions are byte-, half-word, or word-addressable. C62x is a trademark of Texas Instruments. 8 POST OFFICE BOX 1443 • HOUSTON, TEXAS 77251−1443 SGUS055 − SEPTEMBER 2004 CPU (DSP core) description (continued) ÁÁÁÁÁ ÁÁÁÁ ÁÁÁÁÁ ÁÁÁÁ ÁÁÁÁÁ ÁÁÁÁÁ ÁÁÁÁÁ ÁÁÁÁÁ ÁÁÁÁ ÁÁÁÁÁ ÁÁÁÁÁ ÁÁÁÁÁ ÁÁÁÁÁ ÁÁÁÁÁ ÁÁÁÁ ÁÁÁÁÁ ÁÁÁÁÁ ÁÁÁÁÁ ÁÁÁÁÁ ÁÁÁÁ ÁÁÁÁÁ ÁÁÁÁÁ ÁÁÁÁÁ ÁÁÁÁÁ ÁÁÁÁÁ ÁÁÁÁ ÁÁÁÁÁ ÁÁÁÁÁ ÁÁÁÁ ÁÁÁÁ ÁÁÁÁ ÁÁÁÁÁ ÁÁÁÁÁ ÁÁÁÁÁ src1 .L1† src2 dst long dst long src LD1 32 MSB ST1 long src long dst dst .S1† src1 Data Path A 8 8 32 32 8 8 Á Á Á Á src2 dst src1 † .M1 src2 LD1 32 LSB DA1 DA2 LD2 32 LSB Á Á Á Á .D1 .D2 dst src1 src2 src2 src2 .S2† LD2 32 MSB ST2 Á Á long src long dst dst .L2† src2 src1 Register File A (A0−A15) 1X .M2† src1 dst Data Path B ÁÁÁÁÁ ÁÁÁÁÁ ÁÁÁÁÁ ÁÁÁÁÁÁ ÁÁÁÁÁ ÁÁÁÁÁÁ ÁÁÁÁÁ ÁÁÁÁÁ ÁÁÁÁÁÁ ÁÁÁÁÁ ÁÁÁÁÁ ÁÁÁÁÁÁ ÁÁÁÁÁ ÁÁÁÁÁÁ ÁÁÁÁÁÁ ÁÁÁÁÁ ÁÁÁÁÁ ÁÁÁÁÁ ÁÁÁÁÁ ÁÁÁÁÁÁ ÁÁÁÁÁÁ ÁÁÁÁÁ ÁÁÁÁÁ ÁÁÁÁÁÁ ÁÁÁÁÁ ÁÁÁÁÁ ÁÁÁÁÁÁ ÁÁÁÁÁ ÁÁÁÁÁÁ ÁÁÁÁÁ ÁÁÁÁÁ ÁÁÁÁÁÁ ÁÁÁÁÁ ÁÁÁÁÁ ÁÁÁÁÁ Á 2X src2 src1 dst src1 dst long dst long src Á Á Á Á 8 Á Á Á Á Register File B (B0−B15) 8 32 32 8 Á Á Á Á 8 † In addition to fixed-point instructions, these functional units execute floating-point instructions. Control Register File Figure 1. 320C67x CPU (DSP Core) Data Paths POST OFFICE BOX 1443 • HOUSTON, TEXAS 77251−1443 9 SGUS055 − SEPTEMBER 2004 memory map summary Table 2 shows the memory map address ranges of the C6712/C6712C/C6712D devices. Internal memory is always located at address 0 and can be used as both program and data memory. The C6712/C6712C/C6712D configuration registers for the common peripherals are located at the same hex address ranges. The external memory address ranges in the C6712/C6712C/C6712D devices begin at the address location 0x8000 0000. Table 2. 320C6712/C6712C/C6712D Memory Map Summary MEMORY BLOCK DESCRIPTION BLOCK SIZE (BYTES) Internal RAM (L2) 64K HEX ADDRESS RANGE 0000 0000 – 0000 FFFF Reserved 24M – 64K 0001 0000 – 017F FFFF External Memory Interface (EMIF) Registers 256K 0180 0000 – 0183 FFFF L2 Registers 256K 0184 0000 – 0187 FFFF Reserved 256K 0188 0000 – 018B FFFF McBSP 0 Registers 256K 018C 0000 – 018F FFFF McBSP 1 Registers 256K 0190 0000 – 0193 FFFF Timer 0 Registers 256K 0194 0000 – 0197 FFFF Timer 1 Registers 256K 0198 0000 – 019B FFFF 019C 0000 – 019C 01FF Interrupt Selector Registers 512 Device Configuration Registers [C6712C/C6712D only] 4 019C 0200 – 019C 0203 Reserved 256K − 516 019C 0204 – 019F FFFF EDMA RAM and EDMA Registers 256K 01A0 0000 – 01A3 FFFF 01A4 0000 – 01AF FFFF Reserved 768K GPIO Registers [C6712C/C6712D only] 16K 01B0 0000 – 01B0 3FFF Reserved 480K 01B0 4000 – 01B7 BFFF PLL Controller Registers [C6712C/C6712D only] 8K 01B7 C000 – 01B7 DFFF Reserved 4M + 520K 01B7 E000 – 01FF FFFF QDMA Registers 52 0200 0000 – 0200 0033 Reserved 736M – 52 0200 0034 – 2FFF FFFF McBSP 0 Data/Peripheral Data Bus 64M 3000 0000 – 33FF FFFF McBSP 1 Data/Peripheral Data Bus 64M 3400 0000 – 37FF FFFF Reserved 64M 3800 0000 – 3BFF FFFF Reserved EMIF CE0† EMIF CE1† 1G + 64M 3C00 0000 – 7FFF FFFF 256M 8000 0000 – 8FFF FFFF 256M 9000 0000 – 9FFF FFFF EMIF CE2† EMIF CE3† 256M A000 0000 – AFFF FFFF 256M B000 0000 – BFFF FFFF Reserved 1G C000 0000 – FFFF FFFF † The number of EMIF address pins (EA[21:2]) limits the maximum addressable memory (SDRAM) to 128MB per CE space. To get 256MB of addressable memory, additional general-purpose output pin or external logic is required. 10 POST OFFICE BOX 1443 • HOUSTON, TEXAS 77251−1443 SGUS055 − SEPTEMBER 2004 peripheral register descriptions Table 3 through Table 13 identify the peripheral registers for the C6712/C6712C/C6712D devices by their register names, acronyms, and hex address or hex address range. For more detailed information on the register contents, bit names, and their descriptions, see the specific peripheral reference guide listed in the TMS320C6000 DSP Peripherals Overview Reference Guide (literature number SPRU190). Table 3. EMIF Registers HEX ADDRESS RANGE ACRONYM 0180 0000 GBLCTL EMIF global control REGISTER NAME 0180 0004 CECTL1 EMIF CE1 space control 0180 0008 CECTL0 EMIF CE0 space control 0180 000C − 0180 0010 CECTL2 Reserved EMIF CE2 space control 0180 0014 CECTL3 EMIF CE3 space control 0180 0018 SDCTL EMIF SDRAM control 0180 001C SDTIM EMIF SDRAM refresh control 0180 0020 SDEXT EMIF SDRAM extension 0180 0024 − 0183 FFFF − Reserved Table 4. L2 Cache Registers HEX ADDRESS RANGE ACRONYM 0184 0000 CCFG REGISTER NAME 0184 4000 L2WBAR L2 writeback base address register 0184 4004 L2WWC L2 writeback word count register 0184 4010 L2WIBAR L2 writeback-invalidate base address register 0184 4014 L2WIWC L2 writeback-invalidate word count register 0184 4020 L1PIBAR L1P invalidate base address register 0184 4024 L1PIWC L1P invalidate word count register 0184 4030 L1DWIBAR L1D writeback-invalidate base address register 0184 4034 L1DWIWC L1D writeback-invalidate word count register 0184 5000 L2WB 0184 5004 L2WBINV 0184 8200 MAR0 Controls CE0 range 8000 0000 − 80FF FFFF 0184 8204 MAR1 Controls CE0 range 8100 0000 − 81FF FFFF 0184 8208 MAR2 Controls CE0 range 8200 0000 − 82FF FFFF 0184 820C MAR3 Controls CE0 range 8300 0000 − 83FF FFFF 0184 8240 MAR4 Controls CE1 range 9000 0000 − 90FF FFFF 0184 8244 MAR5 Controls CE1 range 9100 0000 − 91FF FFFF 0184 8248 MAR6 Controls CE1 range 9200 0000 − 92FF FFFF 0184 824C MAR7 Controls CE1 range 9300 0000 − 93FF FFFF 0184 8280 MAR8 Controls CE2 range A000 0000 − A0FF FFFF 0184 8284 MAR9 Controls CE2 range A100 0000 − A1FF FFFF 0184 8288 MAR10 Controls CE2 range A200 0000 − A2FF FFFF 0184 828C MAR11 Controls CE2 range A300 0000 − A3FF FFFF 0184 82C0 MAR12 Controls CE3 range B000 0000 − B0FF FFFF 0184 82C4 MAR13 Controls CE3 range B100 0000 − B1FF FFFF 0184 82C8 MAR14 Controls CE3 range B200 0000 − B2FF FFFF 0184 82CC MAR15 Controls CE3 range B300 0000 − B3FF FFFF 0184 82D0 − 0187 FFFF − Cache configuration register L2 writeback all register L2 writeback-invalidate all register Reserved POST OFFICE BOX 1443 • HOUSTON, TEXAS 77251−1443 11 SGUS055 − SEPTEMBER 2004 peripheral register descriptions (continued) Table 5. Interrupt Selector Registers HEX ADDRESS RANGE ACRONYM REGISTER NAME COMMENTS 019C 0000 MUXH Interrupt multiplexer high Selects which interrupts drive CPU interrupts 10−15 (INT10−INT15) 019C 0004 MUXL Interrupt multiplexer low Selects which interrupts drive CPU interrupts 4−9 (INT04−INT09) 019C 0008 EXTPOL External interrupt polarity Sets the polarity of the external interrupts (EXT_INT4−EXT_INT7) 019C 000C − 019F FFFF − Reserved Table 6. Device Registers HEX ADDRESS RANGE ACRONYM 019C 0200 DEVCFG 019C 0204 − 019F FFFF − N/A CSR REGISTER DESCRIPTION Device Configuration This C6712C/C6712D-only register allows the user control of the EMIF input clock source. For more detailed information on the device configuration register, see the Device Configurations section of this data sheet. Reserved CPU Control Status Register Identifies which CPU and defines the silicon revision of the CPU. This register also offers the user control of device operation. For more detailed information on the CPU Control Status Register, see the CPU CSR Register Description section of this data sheet. Table 7. EDMA Parameter RAM† HEX ADDRESS RANGE ACRONYM 01A0 0000 − 01A0 0017 − Parameters for Event 0 (6 words) or Reload/Link Parameters for other Event REGISTER NAME 01A0 0018 − 01A0 002F − Parameters for Event 1 (6 words) or Reload/Link Parameters for other Event 01A0 0030 − 01A0 0047 − Parameters for Event 2 (6 words) or Reload/Link Parameters for other Event 01A0 0048 − 01A0 005F − Parameters for Event 3 (6 words) or Reload/Link Parameters for other Event 01A0 0060 − 01A0 0077 − Parameters for Event 4 (6 words) or Reload/Link Parameters for other Event 01A0 0078 − 01A0 008F − Parameters for Event 5 (6 words) or Reload/Link Parameters for other Event 01A0 0090 − 01A0 00A7 − Parameters for Event 6 (6 words) or Reload/Link Parameters for other Event 01A0 00A8 − 01A0 00BF − Parameters for Event 7 (6 words) or Reload/Link Parameters for other Event 01A0 00C0 − 01A0 00D7 − Parameters for Event 8 (6 words) or Reload/Link Parameters for other Event 01A0 00D8 − 01A0 00EF − Parameters for Event 9 (6 words) or Reload/Link Parameters for other Event 01A0 00F0 − 01A0 00107 − Parameters for Event 10 (6 words) or Reload/Link Parameters for other Event 01A0 0108 − 01A0 011F − Parameters for Event 11 (6 words) or Reload/Link Parameters for other Event 01A0 0120 − 01A0 0137 − Parameters for Event 12 (6 words) or Reload/Link Parameters for other Event 01A0 0138 − 01A0 014F − Parameters for Event 13 (6 words) or Reload/Link Parameters for other Event 01A0 0150 − 01A0 0167 − Parameters for Event 14 (6 words) or Reload/Link Parameters for other Event 01A0 0168 − 01A0 017F − Parameters for Event 15 (6 words) or Reload/Link Parameters for other Event 01A0 0180 − 01A0 0197 − Reload/link parameters for Event 0−15 01A0 0198 − 01A0 01AF − Reload/link parameters for Event 0−15 ... ... 01A0 07E0 − 01A0 07F7 − 01A0 07F8 − 01A0 07FF − Reload/link parameters for Event 0−15 Scratch pad area (2 words) † The C6712/C6712C/C6712D device has 85 EDMA parameters total: 16 Event/Reload parameters and 69 Reload-only parameters. 12 POST OFFICE BOX 1443 • HOUSTON, TEXAS 77251−1443 SGUS055 − SEPTEMBER 2004 peripheral register descriptions (continued) For more details on the EDMA parameter RAM 6-word parameter entry structure, see Figure 2. 31 0 EDMA Parameter Word 0 EDMA Channel Options Parameter (OPT) OPT Word 1 EDMA Channel Source Address (SRC) SRC Word 2 Array/Frame Count (FRMCNT) Word 3 Element Count (ELECNT) EDMA Channel Destination Address (DST) CNT DST Word 4 Array/Frame Index (FRMIDX) Element Index (ELEIDX) IDX Word 5 Element Count Reload (ELERLD) Link Address (LINK) RLD Figure 2. EDMA Channel Parameter Entries (6 Words) for Each EDMA Event Table 8. EDMA Registers HEX ADDRESS RANGE ACRONYM 01A0 0800 − 01A0 FEFC − REGISTER NAME 01A0 FF00 ESEL0 EDMA event selector 0 [C6712C/C6712D Only] 01A0 FF04 ESEL1 EDMA event selector 1 [C6712C/C6712D Only] 01A0 FF08 − 01A0 FF0B − 01A0 FF0C ESEL3 01A0 FF1F − 01A0 FFDC − 01A0 FFE0 PQSR Priority queue status register 01A0 FFE4 CIPR Channel interrupt pending register 01A0 FFE8 CIER Channel interrupt enable register 01A0 FFEC CCER Channel chain enable register 01A0 FFF0 ER 01A0 FFF4 EER Event enable register 01A0 FFF8 ECR Event clear register 01A0 FFFC ESR Event set register 01A1 0000 − 01A3 FFFF – Reserved Reserved EDMA event selector 3 [C6712C/C6712D Only] Reserved Event register Reserved Table 9. Quick DMA (QDMA) and Pseudo Registers† HEX ADDRESS RANGE ACRONYM 0200 0000 QOPT QDMA options parameter register REGISTER NAME 0200 0004 QSRC QDMA source address register 0200 0008 QCNT QDMA frame count register 0200 000C QDST QDMA destination address register 0200 0010 QIDX QDMA index register 0200 0014 − 0200 001C − 0200 0020 QSOPT QDMA pseudo options register 0200 0024 QSSRC QDMA pseudo source address register 0200 0028 QSCNT QDMA pseudo frame count register 0200 002C QSDST QDMA pseudo destination address register 0200 0030 QSIDX Reserved QDMA pseudo index register † All the QDMA and Pseudo registers are write-accessible only POST OFFICE BOX 1443 • HOUSTON, TEXAS 77251−1443 13 SGUS055 − SEPTEMBER 2004 peripheral register descriptions (continued) Table 10. PLL Controller Registers [C6712C/C6712D Only] HEX ADDRESS RANGE ACRONYM REGISTER NAME 01B7 C000 PLLPID Peripheral identification register (PID) 01B7 C004 − 01B7 C0FF − 01B7 C100 PLLCSR 01B7 C104 − 01B7 C10F − 01B7 C110 PLLM 01B7 C114 PLLDIV0 PLL controller divider 0 register 01B7 C118 PLLDIV1 PLL controller divider 1 register 01B7 C11C PLLDIV2 PLL controller divider 2 register 01B7 C120 PLLDIV3 PLL controller divider 3 register 01B7 C124 OSCDIV1 Oscillator divider 1 register 01B7 C128 − 01B7 DFFF − [C6712D value: 0x00010801 for PLL Controller] [C6712C value: 0x00010801 for PLL Controller] Reserved PLL control/status register Reserved PLL multiplier control register Reserved Table 11. GPIO Registers [C6712C/C6712D Only] 14 HEX ADDRESS RANGE ACRONYM 01B0 0000 GPEN GPIO enable register REGISTER NAME 01B0 0004 GPDIR GPIO direction register GPIO value register 01B0 0008 GPVAL 01B0 000C − 01B0 0010 GPDH GPIO delta high register 01B0 0014 GPHM GPIO high mask register 01B0 0018 GPDL GPIO delta low register 01B0 001C GPLM GPIO low mask register 01B0 0020 GPGC GPIO global control register 01B0 0024 GPPOL GPIO interrupt polarity register 01B0 0028 − 01B0 3FFF − Reserved Reserved POST OFFICE BOX 1443 • HOUSTON, TEXAS 77251−1443 SGUS055 − SEPTEMBER 2004 peripheral register descriptions (continued) Table 12. Timer 0 and Timer 1 Registers HEX ADDRESS RANGE TIMER 0 TIMER 1 0194 0000 0198 0000 ACRONYM CTLx REGISTER NAME COMMENTS Timer x control register Determines the operating mode of the timer, monitors the timer status, and controls the function of the TOUT pin. 0194 0004 0198 0004 PRDx Timer x period register Contains the number of timer input clock cycles to count. This number controls the TSTAT signal frequency. 0194 0008 0198 0008 CNTx Timer x counter register Contains the current value of the incrementing counter. 0194 000C − 0197 FFFF 0198 000C − 019B FFFF − Reserved − Table 13. McBSP0 and McBSP1 Registers HEX ADDRESS RANGE McBSP0 McBSP1 ACRONYM REGISTER DESCRIPTION McBSPx data receive register via Configuration Bus 018C 0000 0190 0000 DRRx 3000 0000 − 33FF FFFF 3400 0000 − 37FF FFFF DRRx McBSPx data receive register via Peripheral Data Bus 018C 0004 0190 0004 DXRx McBSPx data transmit register via Configuration Bus 3000 0000 − 33FF FFFF 3400 0000 − 37FF FFFF DXRx McBSPx data transmit register via Peripheral Data Bus The CPU and EDMA controller can only read this register; they cannot write to it. 018C 0008 0190 0008 SPCRx 018C 000C 0190 000C RCRx McBSPx receive control register 018C 0010 0190 0010 XCRx McBSPx transmit control register 018C 0014 0190 0014 SRGRx 018C 0018 0190 0018 MCRx McBSPx multichannel control register 018C 001C 0190 001C RCERx McBSPx receive channel enable register 018C 0020 0190 0020 XCERx McBSPx transmit channel enable register 018C 0024 0190 0024 PCRx 018C 0028 − 018F FFFF 0190 0028 − 0193 FFFF − POST OFFICE BOX 1443 McBSPx serial port control register McBSPx sample rate generator register McBSPx pin control register Reserved • HOUSTON, TEXAS 77251−1443 15 SGUS055 − SEPTEMBER 2004 signal groups description CLKIN CLKOUT3† CLKOUT2‡ CLKOUT1§ CLKMODE0 PLLV¶ PLLG¶ PLLF¶ PLLHV† Reset and Interrupts Clock/PLL BIG/LITTLE ENDIAN TMS TDO TDI TCK TRST EMU0 EMU1 EMU2 EMU3 EMU4 EMU5 RESET NMI EXT_INT7# EXT_INT6# EXT_INT5# EXT_INT4# LENDIAN EMIFBE|| RSV RSV IEEE Standard 1149.1 (JTAG) Emulation Reserved • • • RSV RSV BOOTMODE BOOTMODE1 BOOTMODE0 Control/Status 16 ED[15:0] Data CE3 CE2 CE1 CE0 EA[21:2] BE1 BE0 Memory Control ECLKIN ECLKOUT ARE/SDCAS/SSADS AOE/SDRAS/SSOE AWE/SDWE/SSWE ARDY Bus Arbitration HOLD HOLDA BUSREQ Memory Map Space Select 20 Address Byte Enables EMIF (16-bit) (External Memory Interface) † The CLKOUT3 and PLLHV pin functions are applicable to the C6712C/12D device only. ‡ For the C6712C/12D device, the CLKOUT2 pin is multiplexed with the GP[2] pin. Default function is CLKOUT2. To use this pin as GPIO, the GP2EN bit in the GPEN register and the GP2DIR bit in the GPDIR register must be properly configured. § The CLKOUT1 pin function is applicable to the C6712 device only. ¶ These pins apply to the C6712 device only. The C6712C/12D device has a different PLL module and PLL Controller; therefore, the PLLV, PLLG, and PLLF pins are not necessary on the C6712C device. # For the C6712C/12D device, the external interrupts (EXT_INT[7−4]) go through the general-purpose input/output (GPIO) module. When used as interrupt inputs, the GP[7−4] pins must be configured as inputs (via the GPDIR register) and enabled (via the GPEN register) in addition to enabling the interrupts in the interrupt enable register (IER). || This pin functions as the Big Endian mode correctness and is used when Big Endian mode is selected (LENDIAN = 0) [C6712D] Figure 3. CPU (DSP Core) and Peripheral Signals 16 POST OFFICE BOX 1443 • HOUSTON, TEXAS 77251−1443 SGUS055 − SEPTEMBER 2004 signal groups description (continued) TOUT1 Timer 1 Timer 0 TOUT0 TINP0 TINP1 Timers McBSP1 McBSP0 CLKX1 FSX1 DX1 Transmit Transmit CLKX0 FSX0 DX0 CLKR1 FSR1 DR1† Receive Receive CLKR0 FSR0 DR0 CLKS1† Clock Clock CLKS0 McBSPs (Multichannel Buffered Serial Ports) GPIO‡ GP[7](EXT_INT7) GP[6](EXT_INT6) GP[5](EXT_INT5) GP[4](EXT_INT4) CLKOUT2/GP[2] General-Purpose Input/Output (GPIO) Port † For proper C6712C/C6712D device operation, these pins must be externally pulled up with a 10-kΩ resistor. ‡ Only the C6712C/C6712D device supports the general-purpose input/output (GPIO) port peripheral. Figure 4. Peripheral Signals POST OFFICE BOX 1443 • HOUSTON, TEXAS 77251−1443 17 SGUS055 − SEPTEMBER 2004 DEVICE CONFIGURATIONS On the C6712, C6712C, and C6712D devices, bootmode and certain device configurations/peripheral selections are determined at device reset. For the C6712C/C6712D devices only, other device configurations (e.g., EMIF input clock source) are software-configurable via the device configurations register (DEVCFG) [address location 0x019C0200] after device reset. device configurations at device reset Table 14 describes the C6712/12C/12D device configuration pins, which are set up via internal or external pullup/pulldown resistors through the LENDIAN, EMIFBE [12D only], BOOTMODE[1:0], and CLKMODE0 pins. These configuration pins must be in the desired state until reset is released. For more details on these device configuration pins, see the Terminal Functions table of this data sheet. Table 14. Device Configurations Pins at Device Reset (LENDIAN, EMIFBE [12D only], BOOTMODE[1:0], and CLKMODE0) CONFIGURATION PIN GFN and GDP FUNCTIONAL DESCRIPTION EMIF Big Endian mode correctness (EMIFBE) [C6712D only] C15 − “Reserved” pin for C6712/C6712C devices, the C6712/C6712C devices do not support Big Endian mode. When Big Endian mode is selected (LENDIAN = 0), for proper C6712D device operation the EMIFBE pin must be externally pulled low. EMIFBE C15 This enhancement is not supported on the C6712/12C devices. For proper C6712/C6712C device operation, this pin is “Reserved and must be externally pulled high with a 10-kΩ resistor”. This new functionality does not affect systems using the current default value of C15 pin=1. For more detailed information on the Big Endian mode correctness, see the EMIF Big Endian Mode Correctness [C6712D Only] portion of this data sheet. LENDIAN BOOTMODE[1:0] B17 C19, C20 Device Endian mode (LEND) 0 – System operates in Big Endian mode. For the C6712D, the EMIFBE pin must be pulled low. For the C6712 and C6712C, Big Endian mode is not supported. 1 − System operates in Little Endian mode (default) Bootmode Configuration Pins (BOOTMODE) 00 – Emulation boot 01 – CE1 width 8-bit, Asynchronous external ROM boot with default timings (default mode) 10 − CE1 width 16-bit, Asynchronous external ROM boot with default timings 11 − Reserved, do not use For more detailed information on these bootmode configurations, see the bootmode section of this data sheet. For the C6712 device, clock mode select 0 − Bypass mode (x1). CPU clock = CLKIN 1 − PLL mode (x4). CPU clock = 4 x CLKIN [default] CLKMODE0 C4 For the C6712C/C6712D device, clock generator input clock source select 0 – Reserved. Do not use. 1 − CLKIN square wave [default] For proper C6712C/C6712D device operation, this pin must be either left unconnected or externally pulled up with a 1-kΩ resistor. 18 POST OFFICE BOX 1443 • HOUSTON, TEXAS 77251−1443 SGUS055 − SEPTEMBER 2004 DEVICE CONFIGURATIONS (CONTINUED) DEVCFG register description [C6712C/C6712D only] The device configuration register (DEVCFG) allows the user control of the EMIF input clock source for the C6712C/C6712D device only. For more detailed information on the DEVCFG register control bits, see Table 15 and Table 16. Table 15. Device Configuration Register (DEVCFG) [Address location: 0x019C0200 − 0x019C02FF] 31 16 Reserved† RW-0 5 15 4 3 0 Reserved† EKSRC Reserved† RW-0 R/W-0 R/W-0 Legend: R/W = Read/Write; -n = value after reset † Do not write non-zero values to these bit locations. Table 16. Device Configuration (DEVCFG) Register Selection Bit Descriptions BIT # NAME 31:5 Reserved 4 EKSRC 3:0 Reserved DESCRIPTION Reserved. Do not write non-zero values to these bit locations. EMIF input clock source bit. Determines which clock signal is used as the EMIF input clock. 0 = SYSCLK3 (from the clock generator) is the EMIF input clock source (default) 1 = ECLKIN external pin is the EMIF input clock source Reserved. Do not write non-zero values to these bit locations. POST OFFICE BOX 1443 • HOUSTON, TEXAS 77251−1443 19 SGUS055 − SEPTEMBER 2004 TERMINAL FUNCTIONS The terminal functions table identifies the external signal names, the associated pin (ball) numbers along with the mechanical package designator, the pin type (I, O/Z, or I/O/Z), whether the pin has any internal pullup/pulldown resistors and a functional pin description. For more detailed information on device configuration, see the Device Configurations section of this data sheet. 20 POST OFFICE BOX 1443 • HOUSTON, TEXAS 77251−1443 SGUS055 − SEPTEMBER 2004 Terminal Functions SIGNAL NAME PIN NO. GFN GDP TYPE† IPD/ IPU‡ DESCRIPTION CLOCK/PLL CLKIN CLKOUT1 A3 D7 A3 — I O IPU Clock Input IPD Clock output at device speed [C6712 only] The CLK1EN bit in the EMIF GBLCTL register controls the CLKOUT1 pin. CLK1EN = 0: CLKOUT1 is disabled CLK1EN = 1: CLKOUT1 enabled to clock [default] Clock output at half of device speed [C6712 only] CLKOUT2 Y12 Y12 O/Z IPD For the C6712C/12D device, the CLKOUT2 pin is multiplexed with the GP[2] pin. Clock output at half of device speed (O/Z) [default] (SYSCLK2 internal signal from the clock generator) or this pin can be programmed as GP[2] (I/O/Z). When the CLKOUT2 pin is enabled, the CLK2EN bit in the EMIF global control register (GBLCTL) controls the CLKOUT2 pin (All devices). CLK2EN = 0: CLKOUT2 is disabled CLK2EN = 1: CLKOUT2 enabled to clock [default] CLKOUT3 — D10 O IPD Clock output programmable by OSCDIV1 register in the PLL controller. [12C/12D] Clock mode select [C6712] 0 − Bypass mode (x1). CPU clock = CLKIN 1 − PLL mode (x4). CPU clock = 4 x CLKIN [default] CLKMODE0 C4 C4 I IPU Clock generator input clock source select [C6712C/12D] 0 − Reserved. Do not use. 1 − CLKIN square wave [default] For proper C6712C/12D device operation, this pin must be either left unconnected or externally pulled up with a 1-kΩ resistor. PLLV§ PLLG§ A4 — PLL analog VCC connection for the low-pass filter [C6712 only] C6 — A¶ A¶ PLLF B5 — A¶ PLL low-pass filter connection to external components and a bypass capacitor [C6712 only] PLLHV — C5 A¶ Analog power (3.3 V) for PLL [C6712C/C6712D only] TMS B7 B7 I IPU JTAG test-port mode select TDO A8 A8 O/Z IPU JTAG test-port data out TDI A7 A7 I IPU JTAG test-port data in TCK A6 A6 I IPU JTAG test-port clock TRST B6 B6 I IPD JTAG test-port reset. For IEEE 1149.1 JTAG compatibility, see the IEEE 1149.1 JTAG Compatibility Statement section of this data sheet. EMU5 B12 B12 I/O/Z IPU Emulation pin 5. Reserved for future use, leave unconnected. EMU4 C11 C11 I/O/Z IPU Emulation pin 4. Reserved for future use, leave unconnected. EMU3 B10 B10 I/O/Z IPU Emulation pin 3. Reserved for future use, leave unconnected. PLL analog GND connection for the low-pass filter [C6712 only] JTAG EMULATION EMU2 D10 D3 I/O/Z IPU Emulation pin 2. Reserved for future use, leave unconnected. † I = Input, O = Output, Z = High impedance, S = Supply voltage, GND = Ground ‡ For C6712, IPD = Internal pulldown, IPU = Internal pullup. (These IPD/IPU signal pins feature a 30-kΩ IPD or IPU resistor. To pull up a signal to the opposite supply rail, a 1-kΩ resistor should be used.) For C6712C/12D, IPD = Internal pulldown, IPU = Internal pullup. [These IPD/IPU signal pins feature a 13-kΩ resistor (approximate) for the IPD or 18-kΩ resistor (approximate) for the IPU. An external pullup or pulldown resistor no greater than 4.4 kΩ and 2.0 kΩ, respectively, should be used to pull a signal to the opposite supply rail.] § PLLV and PLLG are not part of external voltage supply or ground. See the clock/PLL section for information on how to connect these pins. ¶ A = Analog signal (PLL Filter) POST OFFICE BOX 1443 • HOUSTON, TEXAS 77251−1443 21 SGUS055 − SEPTEMBER 2004 Terminal Functions (Continued) SIGNAL NAME PIN NO. GFN GDP TYPE† IPD/ IPU‡ DESCRIPTION JTAG EMULATION (CONTINUED) Emulation [1:0] pins [C6712]. For the C6712 device, the EMU0 and EMU1 pins are internally pulled up with 30-kΩ resistors. For Emulation and normal operation, no external pullup/pulldown resistors are necessary. However for the Boundary Scan operation, pull down the EMU1 and EMU0 pins with a dedicated 1-kΩ resistor. EMU1 EMU0 B9 D9 B9 D9 I/O/Z IPU Emulation [1:0] pins [C6712C/C6712D]. • Select the device functional mode of operation Operation EMU[1:0] 00 Boundary Scan/Functional Mode (see Note) 01 Reserved 10 Reserved 11 Emulation/Functional Mode [default] (see the IEEE 1149.1 JTAG Compatibility Statement section of this data sheet) The DSP can be placed in Functional mode when the EMU[1:0] pins are configured for either Boundary Scan or Emulation. Note: When the EMU[1:0] pins are configured for Boundary Scan mode, the internal pulldown (IPD) on the TRST signal must not be opposed in order to operate in Functional mode. For the Boundary Scan mode drive EMU[1:0] and RESET pins low [C6712C/12D]. BOOTMODE BOOTMODE1 BOOTMODE0 C19 C20 C19 C20 I IPD Bootmode[1:0] 00 – Emulation boot 01 − CE1 width 8-bit, asynchronous external ROM boot with default timings (default mode) 10 − CE1 width 16-bit, asynchronous external ROM boot with default timings 11 − Reserved, do not use LITTLE/BIG ENDIAN FORMAT LENDIAN B17 B17 I IPU Device Endian mode 0 – System operates in Big Endian mode. For the C6712D, the EMIFBE pin must be pulled low. For the C6712 and C6712C, Big Endian mode is not supported 1 − System operates in Little Endian mode. EMIF Big Endian mode correctness (EMIFBE) [C6712D only] “Reserved” pin for C6712/C6712C devices When Big Endian mode is selected (LENDIAN = 0), for proper C6712D device operation the EMIFBE pin must be externally pulled low. EMIFBE C15 I IPU This enhancement is not supported on the C6712/12C devices. For proper C6712/C6712C device operation, this pin is “Reserved and must be externally pulled low with a with a 10-kΩ resistor”. For more detailed information on the Big Endian mode correctness, see the EMIF Big Endian Mode Correctness [C6712D Only] portion of this data sheet. † I = Input, O = Output, Z = High impedance, S = Supply voltage, GND = Ground ‡ For C6712, IPD = Internal pulldown, IPU = Internal pullup. (These IPD/IPU signal pins feature a 30-kΩ IPD or IPU resistor. To pull up a signal to the opposite supply rail, a 1-kΩ resistor should be used.) For C6712C/12D, IPD = Internal pulldown, IPU = Internal pullup. [These IPD/IPU signal pins feature a 13-kΩ resistor (approximate) for the IPD or 18-kΩ resistor (approximate) for the IPU. An external pullup or pulldown resistor no greater than 4.4 kΩ and 2.0 kΩ, respectively, should be used to pull a signal to the opposite supply rail.] 22 POST OFFICE BOX 1443 • HOUSTON, TEXAS 77251−1443 SGUS055 − SEPTEMBER 2004 Terminal Functions (Continued) SIGNAL NAME PIN NO. GFN GDP TYPE† IPD/ IPU‡ DESCRIPTION RESETS AND INTERRUPTS RESET NMI A13 A13 C13 C13 EXT_INT7 E3 E3 EXT_INT6 D2 D2 EXT_INT5 C1 C1 EXT_INT4 C2 I I IPU Device reset. When using Boundary Scan mode on the C6712C/C6712D device, drive the EMU[1:0] and RESET pins low. For the C6712D device, this pin does not have an IPU. IPD Nonmaskable interrupt • Edge-driven (rising edge) Any noise on the NMI pin may trigger an NMI interrupt; therefore, if the NMI pin is not used, it is recommended that the NMI pin be grounded versus relying on the IPD. External interrupts [C6712] • Edge-driven • Polarity independently selected via the External Interrupt Polarity Register bits (EXTPOL.[3:0]) I IPU General-purpose input/output pins (I/O/Z) which also function as external interrupts (default) [C6712C/C6712D only] • Edge-driven • Polarity independently selected via the External Interrupt Polarity Register C2 bits (EXTPOL.[3:0]), in addition to the GPIO registers. EMIF − CONTROL SIGNALS COMMON TO ALL TYPES OF MEMORY# CE3 V6 V6 O/Z IPU CE2 W6 W6 O/Z IPU CE1 W18 W18 O/Z IPU CE0 V17 V17 O/Z IPU BE1 U19 U19 O/Z IPU BE0 V20 V20 O/Z HOLDA J18 J18 O IPU Hold-request-acknowledge to the host HOLD J17 J17 I IPU Hold request from the host J19 J19 O IPU Bus request output BUSREQ Memory space enables • Enabled by bits 28 through 31 of the word address • Only one asserted during any external data access Byte-enable control • Decoded from the two lowest bits of the internal address • Byte-write enables for most types of memory IPU • Can be directly connected to SDRAM read and write mask signal (SDQM) EMIF − BUS ARBITRATION#w EMIF − ASYNCHRONOUS/SYNCHRONOUS DRAM/SYNCHRONOUS BURST SRAM MEMORY CONTROL# ECLKIN Y11 Y11 I IPD EMIF input clock EMIF output clock (based on ECLKIN) [C6712] ECLKOUT Y10 Y10 O IPD EMIF output clock depends on the EKSRC bit (DEVCFG.[4]) and on EKEN bit (GBLCTL.[5]). [C6712C/C6712D only] EKSRC = 0 – ECLKOUT is based on the internal SYSCLK3 signal from the clock generator (default). EKSRC = 1 – ECLKOUT is based on the the external EMIF input clock source pin (ECLKIN) EKEN = 0 EKEN = 1 – ECLKOUT held low – ECLKOUT enabled to clock (default) † I = Input, O = Output, Z = High impedance, S = Supply voltage, GND = Ground ‡ For C6712, IPD = Internal pulldown, IPU = Internal pullup. (These IPD/IPU signal pins feature a 30-kΩ IPD or IPU resistor. To pull up a signal to the opposite supply rail, a 1-kΩ resistor should be used.) For C6712C/12D, IPD = Internal pulldown, IPU = Internal pullup. [These IPD/IPU signal pins feature a 13-kΩ resistor (approximate) for the IPD or 18-kΩ resistor (approximate) for the IPU. An external pullup or pulldown resistor no greater than 4.4 kΩ and 2.0 kΩ, respectively, should be used to pull a signal to the opposite supply rail.] # To maintain signal integrity for the EMIF signals, serial termination resistors should be inserted into all EMIF output signal lines. POST OFFICE BOX 1443 • HOUSTON, TEXAS 77251−1443 23 SGUS055 − SEPTEMBER 2004 Terminal Functions (Continued) SIGNAL NAME PIN NO. GFN GDP TYPE† IPD/ IPU‡ DESCRIPTION EMIF − ASYNCHRONOUS/SYNCHRONOUS DRAM/SYNCHRONOUS BURST SRAM MEMORY CONTROL (CONTINUED)# ARE/SDCAS/ SSADS V11 V11 O/Z IPU Asynchronous memory read enable/SDRAM column-address strobe/SBSRAM address strobe AOE/SDRAS/ SSOE W10 W10 O/Z IPU Asynchronous memory output enable/SDRAM row-address strobe/SBSRAM output enable AWE/SDWE/ SSWE V12 V12 O/Z IPU Asynchronous memory write enable/SDRAM write enable/SBSRAM write enable ARDY Y5 Y5 I IPU Asynchronous memory ready input EMIF − ADDRESS# EA21 U18 U18 EA20 Y18 Y18 EA19 W17 W17 EA18 Y16 Y16 EA17 V16 V16 EA16 Y15 Y15 EA15 W15 W15 EA14 Y14 Y14 EA13 W14 W14 EA12 V14 V14 EA11 W13 W13 EA10 V10 V10 EA9 Y9 Y9 EA8 V9 V9 EA7 Y8 Y8 EA6 W8 W8 EA5 V8 V8 EA4 W7 W7 EA3 V7 V7 EA2 Y6 Y6 O/Z IPU EMIF external address Note: EMIF address numbering for the C6712, C6712C, and C6712D devices start with EA2 to maintain signal name compatibility with other C671x devices (e.g., C6711, C6713) [see the 16−bit EMIF addressing scheme in the TMS320C6000 DSP External Memory Interface (EMIF) Reference Guide (literature number SPRU266)]. EMIF − DATA# ED15 T19 T19 ED14 T20 T20 ED13 T18 T18 ED12 R20 R20 ED11 R19 R19 ED10 P20 P20 ED9 P18 P18 I/O/Z IPU External data ED8 N20 N20 † I = Input, O = Output, Z = High impedance, S = Supply voltage, GND = Ground ‡ For C6712, IPD = Internal pulldown, IPU = Internal pullup. (These IPD/IPU signal pins feature a 30-kΩ IPD or IPU resistor. To pull up a signal to the opposite supply rail, a 1-kΩ resistor should be used.) For C6712C/12D, IPD = Internal pulldown, IPU = Internal pullup. [These IPD/IPU signal pins feature a 13-kΩ resistor (approximate) for the IPD or 18-kΩ resistor (approximate) for the IPU. An external pullup or pulldown resistor no greater than 4.4 kΩ and 2.0 kΩ, respectively, should be used to pull a signal to the opposite supply rail.] # To maintain signal integrity for the EMIF signals, serial termination resistors should be inserted into all EMIF output signal lines. 24 POST OFFICE BOX 1443 • HOUSTON, TEXAS 77251−1443 SGUS055 − SEPTEMBER 2004 Terminal Functions (Continued) SIGNAL NAME PIN NO. GFN GDP TYPE† IPD/ IPU‡ DESCRIPTION EMIF − DATA (CONTINUED)# ED7 N19 ED6 N18 N19 N18 ED5 M20 M20 ED4 M19 M19 ED3 L19 L19 ED2 L18 L18 ED1 K19 K19 ED0 K18 K18 I/O/Z IPU External data TIMER1 TOUT1 F1 F1 O IPD Timer 1 or general-purpose output TINP1 F2 F2 I IPD Timer 1 or general-purpose input TOUT0 G1 G1 O IPD Timer 0 or general-purpose output TINP0 G2 G2 I IPD Timer 0 or general-purpose input TIMER0 MULTICHANNEL BUFFERED SERIAL PORT 1 (McBSP1) CLKS1 E1 E1 I IPD External clock source (as opposed to internal) On the C6712C/12D device, this pin does not have an internal pulldown (IPD). For proper C6712C/12D device operation, the CLKS1 pin should either be driven externally at all times or be pulled up with a 10-kΩ resistor to a valid logic level. Because it is common for some ICs to 3-state their outputs at times, a 10-kΩ pullup resistor may be desirable even when an external device is driving the pin. CLKR1 M1 M1 I/O/Z IPD Receive clock CLKX1 L3 L3 I/O/Z IPD Transmit clock DR1 M2 M2 I IPU Receive data On the C6712C/12D device, this pin does not have an internal pullup (IPU). For proper C6712C/12D device operation, the DR1 pin should either be driven externally at all times or be pulled up with a 10-kΩ resistor to a valid logic level. Because it is common for some ICs to 3-state their outputs at times, a 10-kΩ pullup resistor may be desirable even when an external device is driving the pin. DX1 L2 L2 O/Z IPU Transmit data FSR1 M3 M3 I/O/Z IPD Receive frame sync FSX1 L1 L1 I/O/Z IPD Transmit frame sync † I = Input, O = Output, Z = High impedance, S = Supply voltage, GND = Ground ‡ For C6712, IPD = Internal pulldown, IPU = Internal pullup. (These IPD/IPU signal pins feature a 30-kΩ IPD or IPU resistor. To pull up a signal to the opposite supply rail, a 1-kΩ resistor should be used.) For C6712C/12D, IPD = Internal pulldown, IPU = Internal pullup. [These IPD/IPU signal pins feature a 13-kΩ resistor (approximate) for the IPD or 18-kΩ resistor (approximate) for the IPU. An external pullup or pulldown resistor no greater than 4.4 kΩ and 2.0 kΩ, respectively, should be used to pull a signal to the opposite supply rail.] # To maintain signal integrity for the EMIF signals, serial termination resistors should be inserted into all EMIF output signal lines. POST OFFICE BOX 1443 • HOUSTON, TEXAS 77251−1443 25 SGUS055 − SEPTEMBER 2004 Terminal Functions (Continued) SIGNAL NAME PIN NO. GFN GDP TYPE† IPD/ IPU‡ DESCRIPTION MULTICHANNEL BUFFERED SERIAL PORT 0 (McBSP0) CLKS0 K3 K3 I IPD External clock source (as opposed to internal) CLKR0 H3 H3 I/O/Z IPD Receive clock CLKX0 G3 G3 I/O/Z IPD Transmit clock DR0 J1 J1 I IPU Receive data DX0 H2 H2 O/Z IPU Transmit data FSR0 J3 J3 I/O/Z IPD Receive frame sync FSX0 H1 H1 I/O/Z IPD Transmit frame sync GENERAL-PURPOSE INPUT/OUTPUT (GPIO) MODULE [C6712C/12D ONLY] Clock output at half of device speed [C6712/12D only] CLKOUT2/GP[2] Y12 Y12 GP[7](EXT_INT7) — E3 GP[6](EXT_INT6) — D2 GP[5](EXT_INT5) — C1 GP[4](EXT_INT4) — C2 I/O/Z IPD For the C6712C/12D device, the CLKOUT2 pin is multiplexed with the GP[2] pin. Clock output at half of device speed (O/Z) [default] (SYSCLK2 internal signal from the clock generator) or this pin can be programmed as GP[2] (I/O/Z). External interrupts [C6712C/12D only] • Edge-driven • Polarity independently selected via the External Interrupt Polarity Register bits (EXTPOL.[3:0]) I/O/Z IPU General-purpose input/output pins (I/O/Z) which also function as external interrupts [C6712C/12D only] • Edge-driven • Polarity independently selected via the External Interrupt Polarity Register bits (EXTPOL.[3:0]), in addition to the GPIO registers. RESERVED FOR TEST Reserved (leave unconnected, do not connect to power or ground) [C6712] Reserved. For the C6712C device, it is recommended that this pin be externally pulled low with a 10-kΩ resistor. RSV C15 C15 RSV C12 C12 IPU O IPU EMIF Big Endian mode correctness (EMIFBE) [C6712D only]. When Big Endian mode is selected, for proper C6712D device operation, this pin must be externally pulled low. (For more detailed information on Big Endian mode correctness, see the Device Configuration section of this data sheet.) Reserved (leave unconnected, do not connect to power or ground). Only the C6712 device has internal pullup (IPU) on this pin. On the C6712C/12D device, this pin does not have an IPU. Only the C6712 device has internal pullups (IPUs). For the C6712, the D12 pin is reserved (leave unconnected, do not connect to power or ground). RSV D12 D12 O IPU On the C6712C/12D device, this pin does not have an IPU. For proper C6712C/12D device operation, the D12 pin must be externally pulled down with a 10-kΩ resistor. † I = Input, O = Output, Z = High impedance, S = Supply voltage, GND = Ground ‡ For C6712, IPD = Internal pulldown, IPU = Internal pullup. (These IPD/IPU signal pins feature a 30-kΩ IPD or IPU resistor. To pull up a signal to the opposite supply rail, a 1-kΩ resistor should be used.) For C6712C, IPD = Internal pulldown, IPU = Internal pullup. [These IPD/IPU signal pins feature a 13-kΩ resistor (approximate) for the IPD or 18-kΩ resistor (approximate) for the IPU. An external pullup or pulldown resistor no greater than 4.4 kΩ and 2.0 kΩ, respectively, should be used to pull a signal to the opposite supply rail.] 26 POST OFFICE BOX 1443 • HOUSTON, TEXAS 77251−1443 SGUS055 − SEPTEMBER 2004 Terminal Functions (Continued) SIGNAL NAME PIN NO. TYPE† IPD/ IPU A5 O IPU — O GFN GDP RSV A5 RSV D3 DESCRIPTION Reserved (leave unconnected, do not connect to power or ground) Reserved (leave unconnected, do not connect to power or ground) Reserved (leave unconnected, do not connect to power or ground) [C6712] RSV N2 N2 O RSV Y20 — O RSV — N1 Reserved. For proper C6712C/12D device operation, this pin must be externally pulled up with a 10-kΩ resistor. RSV — B5 Reserved (leave unconnected, do not connect to power or ground) RSV — D7 RSV — A12 Reserved (leave unconnected, do not connect to power or ground) RSV — B11 Reserved (leave unconnected, do not connect to power or ground) A15 A15 A16 A16 A18 A18 B14 B14 B16 B16 Reserved. For proper C6712C/12D device operation, this pin must be externally pulled up with a 10-kΩ resistor. Reserved (leave unconnected, do not connect to power or ground) IPD Reserved (leave unconnected, do not connect to power or ground) ADDITIONAL RESERVED FOR TEST RSV B18 B18 C14 C14 C16 C16 C17 C17 D18 D18 D20 D20 E18 E18 E19 E19 E20 E20 F18 F18 F20 F20 G18 G18 G19 G19 G20 G20 H19 H19 H20 H20 J20 J20 N3 N3 P1 P1 P2 P2 Reserved (leave unconnected, do not connect to power or ground) P3 P3 † I = Input, O = Output, Z = High impedance, S = Supply voltage, GND = Ground POST OFFICE BOX 1443 • HOUSTON, TEXAS 77251−1443 27 SGUS055 − SEPTEMBER 2004 Terminal Functions (Continued) SIGNAL NAME PIN NO. GFN GDP TYPE† IPD/ IPU DESCRIPTION ADDITIONAL RESERVED FOR TEST RSV R2 R2 R3 R3 T1 T1 T2 T2 U1 U1 U2 U2 U3 U3 V1 V1 V2 V2 V4 V4 V5 V5 W4 W4 Y3 Y3 Reserved (leave unconnected, do not connect to power or ground) Y4 Y4 † I = Input, O = Output, Z = High impedance, S = Supply voltage, GND = Ground 28 POST OFFICE BOX 1443 • HOUSTON, TEXAS 77251−1443 SGUS055 − SEPTEMBER 2004 Terminal Functions (Continued) SIGNAL NAME PIN NO. GFN GDP TYPE† DESCRIPTION SUPPLY VOLTAGE PINS DVDD CVDD A17 A17 B3 B3 B8 B8 B13 B13 C5 — C10 C10 D1 D1 D16 D16 D19 D19 F3 F3 H18 H18 J2 J2 M18 M18 N1 — R1 R1 R18 R18 T3 T3 U5 U5 U7 U7 U12 U12 U16 U16 V13 V13 V15 V15 V19 V19 W3 W3 W9 W9 W12 W12 Y7 Y7 Y17 Y17 — A4 A9 A9 A10 A10 A12 — B2 B2 B19 B19 C3 C3 C7 C7 C18 C18 D5 D5 S 3.3-V supply voltage (see the power-supply decoupling portion of this data sheet) S 1.20‡-V supply voltage (C6712C/C6712D) 1.8-V supply voltage (C6712) (see the power-supply decoupling portion of this data sheet) D6 D6 † I = Input, O = Output, Z = High impedance, S = Supply voltage, GND = Ground ‡ This value is compatible with existing 1.26V designs. POST OFFICE BOX 1443 • HOUSTON, TEXAS 77251−1443 29 SGUS055 − SEPTEMBER 2004 Terminal Functions (Continued) SIGNAL NAME PIN NO. GFN GDP TYPE† DESCRIPTION SUPPLY VOLTAGE PINS (CONTINUED) CVDD D11 D11 D14 D14 D15 D15 F4 F4 F17 F17 K1 K1 K4 K4 K17 K17 L4 L4 L17 L17 L20 L20 R4 R4 R17 R17 U6 U6 U10 U10 U11 U11 U14 U14 U15 U15 V3 V3 V18 V18 W2 W2 W19 W19 A1 A1 A2 A2 S 1.20‡-V supply voltage (C6712C/C6712D) 1.8-V supply voltage (C6712) (see the power-supply decoupling portion of this data sheet) GROUND PINS VSS A11 A11 A14 A14 A19 A19 A20 A20 B1 B1 B4 B4 B11 — B15 B15 B20 B20 — C6 C8 C8 C9 C9 D4 D4 D8 D8 GND Ground pins D13 D13 † I = Input, O = Output, Z = High impedance, S = Supply voltage, GND = Ground ‡ This value is compatible with existing 1.26V designs. 30 POST OFFICE BOX 1443 • HOUSTON, TEXAS 77251−1443 SGUS055 − SEPTEMBER 2004 Terminal Functions (Continued) SIGNAL NAME PIN NO. GFN GDP TYPE† DESCRIPTION GROUND PINS (CONTINUED) VSS D17 D17 E2 E2 E4 E4 E17 E17 F19 F19 G4 G4 G17 G17 H4 H4 H17 H17 J4 J4 — J9 — J10 — J11 — J12 K2 K2 — K9 — K10 — K11 — K12 K20 K20 — L9 — L10 — L11 — L12 M4 M4 — M9 — M10 — M11 — M12 M17 M17 N4 N4 N17 N17 P4 P4 P17 P17 P19 P19 T4 T4 T17 T17 U4 U4 U8 U8 GND Ground pins|| The center thermal balls (J9−J12, K9−K12, L9−L12, M9−M12) [shaded] are all tied to ground and act as both electrical grounds and thermal relief (thermal dissipation). U9 U9 † I = Input, O = Output, Z = High impedance, S = Supply voltage, GND = Ground || Shaded pin numbers denote the center thermal balls for the GDP package [C6712C/12D only]. POST OFFICE BOX 1443 • HOUSTON, TEXAS 77251−1443 31 SGUS055 − SEPTEMBER 2004 Terminal Functions (Continued) SIGNAL NAME PIN NO. GFN GDP TYPE† DESCRIPTION GROUND PINS (CONTINUED) VSS U13 U13 U17 U17 U20 U20 W1 W1 W5 W5 W11 W11 W16 W16 W20 W20 Y1 Y1 Y2 Y2 Y13 Y13 Y19 Y19 GND Ground pins — Y20 † I = Input, O = Output, Z = High impedance, S = Supply voltage, GND = Ground 32 POST OFFICE BOX 1443 • HOUSTON, TEXAS 77251−1443 SGUS055 − SEPTEMBER 2004 development support TI offers an extensive line of development tools for the TMS320C6000 DSP platform, including tools to evaluate the performance of the processors, generate code, develop algorithm implementations, and fully integrate and debug software and hardware modules. The following products support development of C6000 DSP-based applications: Software Development Tools: Code Composer Studio Integrated Development Environment (IDE): including Editor C/C++/Assembly Code Generation, and Debug plus additional development tools Scalable, Real-Time Foundation Software (DSP/BIOS), which provides the basic run-time target software needed to support any DSP application. Hardware Development Tools: Extended Development System (XDS) Emulator (supports C6000 DSP multiprocessor system debug) EVM (Evaluation Module) For a complete listing of development-support tools for the TMS320C6000 DSP platform, visit the Texas Instruments web site on the Worldwide Web at http://www.ti.com uniform resource locator (URL). For information on pricing and availability, contact the nearest TI field sales office or authorized distributor. Code Composer Studio, DSP/BIOS, and XDS are trademarks of Texas Instruments. POST OFFICE BOX 1443 • HOUSTON, TEXAS 77251−1443 33 SGUS055 − SEPTEMBER 2004 device and development-support tool nomenclature To designate the stages in the product development cycle, TI assigns prefixes to the part numbers of all TMS320 DSP devices and support tools. Each TMS320 DSP commercial family member has one of three prefixes: SMX, TMP, or SM. Texas Instruments recommends two of three possible prefix designators for support tools: TMDX and TMDS. These prefixes represent evolutionary stages of product development from engineering prototypes (SMX / TMDX) through fully qualified production devices/tools (SM / TMDS). Device development evolutionary flow: SMX Preproduction device that is not necessarily representative of the final device’s electrical specifications TMP Final silicon die that conforms to the device’s electrical specifications but has not completed quality and reliability verification SM Fully qualified production device Support tool development evolutionary flow: TMDX Development-support product that has not yet completed Texas Instruments internal qualification testing. TMDS Fully qualified development-support product SMX and TMP devices and TMDX development-support tools are shipped with appropriate disclaimers describing their limitations and intended uses. Preproduction devices (SMX) may not be representative of a final product and Texas Instruments reserves the right to change or discontinue these products without notice. SM devices and TMDS development-support tools have been characterized fully, and the quality and reliability of the device have been demonstrated fully. TI’s standard warranty applies. Predictions show that preproduction devices (SMX or TMP) have a greater failure rate than the standard production devices. Texas Instruments recommends that these devices not be used in any production system because their expected end-use failure rate still is undefined. Only qualified production devices are to be used. TI device nomenclature also includes a suffix with the device family name. This suffix indicates the package type (for example, GDP), the temperature range (for example, blank is the default commercial temperature range and A is the extended temperature range), and the device speed range in megahertz (for example, 16 is 167 MHz). Table 17 identifies the C6712/12C/12D device part numbers (orderables). For more details and for ordering information, see the TI website (www.ti.com). Figure 5 provides a legend for reading the complete device name for any member of the TMS320C6000 DSP platform. 34 POST OFFICE BOX 1443 • HOUSTON, TEXAS 77251−1443 SGUS055 − SEPTEMBER 2004 device and development-support tool nomenclature (continued) Table 17. 320C6712/C6712C/C6712D Device Part Numbers (P/Ns) and Ordering Information DEVICE SPEED CVDD (CORE VOLTAGE) DVDD (I/O VOLTAGE) OPERATING CASE TEMPERATURE RANGE 167 MHz/1000 MFLOPS 1.20† V 3.3 V −40_C to 105_C DEVICE ORDERABLE P/N C6712D SM32C6712DGDPA16EP SM PREFIX TMX = TMP = TMS = SMJ = SM = 32 C 6712D GDP A 16 DEVICE SPEED RANGE Experimental device Prototype device Qualified device MIL-PRF-38535, QML High Rel (non-38535) 16 = 167 MHz TEMPERATURE RANGE (DEFAULT: 0°C TO 90°C) Blank = 0°C to 90°C, commercial temperature A = −40°C to 105°C, extended temperature DEVICE FAMILY 32 or 320 = TMS320 DSP family PACKAGE TYPE† GDP = 272-pin plastic BGA GFN = 256-pin plastic BGA GGP = 352-pin plastic BGA GJC = 352-pin plastic BGA GJL = 352-pin plastic BGA GLS = 384-pin plastic BGA GLW = 340-pin plastic BGA GNY = 384-pin plastic BGA GNZ = 352-pin plastic BGA GLZ = 532-pin plastic BGA GHK = 288-pin plastic MicroStar BGAt PYP = 208-pin PowerPADt plastic QFP TECHNOLOGY C = CMOS DEVICE C6000 DSPs: C6711D C6712D C6713B † BGA = QFP = Ball Grid Array Quad Flatpack Figure 5. TMS320C6000 DSP Platform Device Nomenclature (Including the SM320C6712, SM320C6712C, and SM320C6712D Devices) MicroStar BGA and PowerPAD are trademarks of Texas Instruments. † This value is compatible with existing 1.26V designs. POST OFFICE BOX 1443 • HOUSTON, TEXAS 77251−1443 35 SGUS055 − SEPTEMBER 2004 documentation support Extensive documentation supports all TMS320 DSP family generations of devices from product announcement through applications development. The types of documentation available include: data sheets, such as this document, with design specifications; complete user’s reference guides for all devices and tools; technical briefs; development-support tools; on-line help; and hardware and software applications. The following is a brief, descriptive list of support documentation specific to the C6000 DSP devices: The TMS320C6000 CPU and Instruction Set Reference Guide (literature number SPRU189) describes the C6000 DSP core (CPU) architecture, instruction set, pipeline, and associated interrupts. The TMS320C6000 DSP Peripherals Overview Reference Guide [hereafter referred to as the C6000 PRG Overview] (literature number SPRU190) provides an overview and briefly describes the functionality of the peripherals available on the C6000 DSP platform of devices. This document also includes a table listing the peripherals available on the C6000 devices along with literature numbers and hyperlinks to the associated peripheral documents. These C6712C/C6712D peripherals, except the PLL, are similar to the peripherals on the TMS320C6712 and TMS320C64x devices; therefore, see the TMS320C6712 (C6711 or C67x) peripheral information, and in some cases, where indicated, see the TMS320C6712 (C6712 or TMS320C67x or C671x or C67x) peripheral information, and in some cases, where indicated, see the C64x information in the TMS320C6000 DSP Peripherals Overview Reference Guide (literature number SPRU190). The TMS320C6000 Technical Brief (literature number SPRU197) gives an introduction to the C62x/C67x DSP devices, associated development tools, and third-party support. TMS320C6000 DSP Software-Programmable Phase-Locked Loop (PLL) Controller Reference Guide (literature number SPRU233) describes the functionality of the PLL peripheral available on the C6712C/12D device. The Migrating from TMS320C6211(B)/6711(B) to TMS320C6711C application report (literature number SPRA837) describes the differences and issues of interest related to migration from the Texas Instruments TMS320C6211, TMS320C6211B, TMS320C6711, and TMS320C6711B devices, GFN packages, to the TMS320C6711C device, GDP package. The Migrating from TMS320C6712 to TMS320C6712C application report (literature number SPRA852) describes the differences and issues of interest related to migration from the Texas Instruments TMS320C6712 device, GFN package, to the TMS320C6712C device, GDP package. The TMS320C6712, TMS320C6712C, TMS320C6712D Digital Signal Processors Silicon Errata (C6712 Silicon Revisions 1.0, 1.2, 1.3, C6712C Silicon Revision 1.1, and C6712D Silicon Revision 2.0) [literature number SPRZ182C or later] categorizes and describes the known exceptions to the functional specifications and usage notes for the TMS320C6712, TMS320C6712C, TMS320C6712D DSP devices. The TMS320C6713/12C/11C Power Consumption Summary application report (literature number SPRA889) discusses the power consumption for user applications with the TMS320C6713, TMS320C6712C, and TMS320C6711C DSP devices. The tools support documentation is electronically available within the Code Composer Studio Integrated Development Environment (IDE). For a complete listing of C6000 DSP latest documentation, visit the Texas Instruments web site on the Worldwide Web at http://www.ti.com uniform resource locator (URL). See the Worldwide Web URL for the application report How To Begin Development with the TMS320C6712 DSP (literature number SPRA693), which describes in more detail the compatibility and similarities/differences between the C6711 and C6712 devices. 36 POST OFFICE BOX 1443 • HOUSTON, TEXAS 77251−1443 SGUS055 − SEPTEMBER 2004 CPU CSR register description The CPU control status register (CSR) contains the CPU ID and CPU Revision ID (bits 16−31) as well as the status of the device power-down modes [PWRD field (bits 15−10)], program and data cache control modes, the endian bit (EN, bit 8) and the global interrupt enable (GIE, bit 0) and previous GIE (PGIE, bit 1). Figure 6 and Table 18 identify the bit fields in the CPU CSR register. For more detailed information on the bit fields in the CPU CSR register, see the TMS320C6000 DSP Peripherals Overview Reference Guide (literature number SPRU190) and the TMS320C6000 CPU and Instruction Set Reference Guide (literature number SPRU189). 31 24 23 15 16 CPU ID REVISION ID R-0x02 R-0x02 [C6712] R-0x03 [C6712C/12D] 10 9 8 7 6 PWRD SAT EN PCC R/W-0 R/C-0 R-1 R/W-0 5 4 2 1 0 DCC PGIE GIE R/W-0 R/W-0 R/W-0 Legend: R = Readable by the MVC instruction, R/W = Readable/Writeable by the MVC instruction; W = Read/write; -n = value after reset, -x = undefined value after reset, C = Clearable by the MVC instruction Figure 6. CPU Control Status Register (CPU CSR) POST OFFICE BOX 1443 • HOUSTON, TEXAS 77251−1443 37 SGUS055 − SEPTEMBER 2004 CPU CSR register description (continued) Table 18. CPU CSR Register Bit Field Description BIT # NAME 31:24 CPU ID 23:16 REVISION ID DESCRIPTION CPU ID + REV ID. Read only. Identifies which CPU is used and defines the silicon revision of the CPU. CPU ID + REVISION ID (31:16) are combined for a value of: 0x0202 for C6712 and 0x0203 for C6712C/C6712D Control power-down modes. The values are always read as zero. 15:10 9 8 7:5 4:2 PWRD 000000 001001 010001 011010 011100 Others = = = = = = no power-down (default) PD1, wake-up by an enabled interrupt PD1, wake-up by an enabled or not enabled interrupt PD2, wake-up by a device reset PD3, wake-up by a device reset Reserved SAT Saturate bit. Set when any unit performs a saturate. This bit can be cleared only by the MVC instruction and can be set only by a functional unit. The set by the a functional unit has priority over a clear (by the MVC instruction) if they occur on the same cycle. The saturate bit is set one full cycle (one delay slot) after a saturate occurs. This bit will not be modified by a conditional instruction whose condition is false. EN Endian bit. This bit is read-only. Depicts the device endian mode. 0 = Big Endian mode. 1 = Little Endian mode [default]. PCC Program Cache control mode. L1D, Level 1 Program Cache 000/010 = Cache Enabled / Cache accessed and updated on reads. All other PCC values reserved. DCC Data Cache control mode. L1D, Level 1 Data Cache 000/010 = Cache Enabled / 2-Way Cache All other DCC values reserved Previous GIE (global interrupt enable); saves the Global Interrupt Enable (GIE) when an interrupt is taken. Allows for proper nesting of interrupts. 1 PGIE 0 = Previous GIE value is 0. (default) 1 = Previous GIE value is 1. Global interrupt enable bit. Enables (1) or disables (0) all interrupts except the reset interrupt and NMI (nonmaskable interrupt). 0 GIE 0 = Disables all interrupts (except the reset interrupt and NMI) [default] 1 = Enables all interrupts (except the reset interrupt and NMI) 38 POST OFFICE BOX 1443 • HOUSTON, TEXAS 77251−1443 SGUS055 − SEPTEMBER 2004 cache configuration (CCFG) register description (12D) The C6712D device includes an enhancement to the cache configuration (CCFG) register. A “P” bit (CCFG.31) allows the programmer to select the priority of accesses to L2 memory originating from the transfer crossbar (TC) over accesses originating from the L1D memory system. An important class of TC accesses is EDMA transfers, which move data to or from the L2 memory. While the EDMA normally has no issue accessing L2 memory due to the high hit rates on the L1D memory system, there are pathological cases where certain CPU behavior could block the EDMA from accessing the L2 memory for long enough to cause a missed deadline when transferring data to a peripheral such as the McASP or McBSP. This can be avoided by setting the P bit to “1” because the EDMA will assume a higher priority than the L1D memory system when accessing L2 memory. For more detailed information on the P-bit function and for silicon advisories concerning EDMA L2 memory accesses blocked, see the TMS320C6712, TMS320C6712C, TMS320C6712D Digital Signal Processors Silicon Errata (literature number SPRZ182C or later). 31 30 10 9 8 7 3 2 0 P† Reserved IP ID Reserved L2MODE R/W-0 R-x W-0 W-0 R-0 0000 R/W-000 Legend: R = Readable; R/W = Readable/Writeable; -n = value after reset; -x = undefined value after reset † Unlike the C6712/12C devices, the C6712D device includes a P bit. Figure 7. Cache Configuration Register (CCFG) Table 19. CCFG Register Bit Field Description BIT # NAME DESCRIPTION 31 P 30:10 Reserved 9 IP Invalidate L1P bit. 0 = Normal L1P operation 1 = All L1P lines are invalidated 8 ID Invalidate L1D bit. 0 = Normal L1D operation 1 = All L1D lines are invalidated 7:3 Reserved L1D requestor priority to L2 bit. P = 0: L1D requests to L2 higher priority than TC requests P = 1: TC requests to L2 higher priority than L1D requests Reserved. Read-only, writes have no effect. Reserved. Read-only, writes have no effect. L2 operation mode bits (L2MODE). 2:0 L2MODE 000b = 001b = 010b = 011b = 111b = All others L2 Cache disabled (All SRAM mode) [64K SRAM] 1-way Cache (16K L2 Cache) / [48K SRAM] 2-way Cache (32K L2 Cache) / [32K SRAM] 3-way Cache (48K L2 Cache) / [16K SRAM] 4-way Cache (64K L2 Cache) / [no SRAM] Reserved POST OFFICE BOX 1443 • HOUSTON, TEXAS 77251−1443 39 SGUS055 − SEPTEMBER 2004 interrupt sources and interrupt selector [C6712 only] The C67x DSP core on the C6712 supports 16 prioritized interrupts, which are listed in Table 20. The highest-priority interrupt is INT_00 (dedicated to RESET) while the lowest-priority interrupt is INT_15. The first four interrupts (INT_00−INT_03) are non-maskable and fixed. The remaining interrupts (INT_04−INT_15) are maskable and default to the interrupt source specified in Table 20. The interrupt source for interrupts 4−15 can be programmed by modifying the selector value (binary value) in the corresponding fields of the Interrupt Selector Control registers: MUXH (address 0x019C0000) and MUXL (address 0x019C0004). Table 20. C6712 DSP Interrupts DEFAULT INTERRUPT NUMBER INTERRUPT SELECTOR CONTROL REGISTER DEFAULT SELECTOR VALUE (BINARY) INTERRUPT EVENT INT_00 − − RESET INT_01 − − NMI INT_02 − − Reserved Reserved. Do not use. INTERRUPT SOURCE INT_03 − − Reserved Reserved. Do not use. INT_04 MUXL[4:0] 00100 EXT_INT4 External interrupt pin 4 INT_05 MUXL[9:5] 00101 EXT_INT5 External interrupt pin 5 INT_06 MUXL[14:10] 00110 EXT_INT6 External interrupt pin 6 INT_07 MUXL[20:16] 00111 EXT_INT7 External interrupt pin 7 INT_08 MUXL[25:21] 01000 EDMA_INT EDMA channel (0 through 15) interrupt INT_09 MUXL[30:26] 01001 Reserved INT_10 MUXH[4:0] 00011 SD_INT None, but programmable EMIF SDRAM timer interrupt INT_11 MUXH[9:5] 01010 Reserved None, but programmable INT_12 MUXH[14:10] 01011 Reserved None, but programmable INT_13 MUXH[20:16] 00000 Reserved None, but programmable INT_14 MUXH[25:21] 00001 TINT0 Timer 0 interrupt INT_15 MUXH[30:26] 00010 TINT1 Timer 1 interrupt − − 01100 XINT0 McBSP0 transmit interrupt − − 01101 RINT0 McBSP0 receive interrupt − − 01110 XINT1 McBSP1 transmit interrupt − − 01111 RINT1 McBSP1 receive interrupt − − 10000 − 11111 Reserved Reserved. Do not use. † Interrupts INT_00 through INT_03 are non-maskable and fixed. ‡ Interrupts INT_04 through INT_15 are programmable by modifying the binary selector values in the Interrupt Selector Control registers fields. Table 20 shows the default interrupt sources for interrupts INT_04 through INT_15. For more detailed information on interrupt sources and selection, see the TMS320C6000 DSP Interrupt Selector Reference Guide (literature number SPRU646). 40 POST OFFICE BOX 1443 • HOUSTON, TEXAS 77251−1443 SGUS055 − SEPTEMBER 2004 interrupt sources and interrupt selector [C6712C/C6712D only] The C67x DSP core on the C6712C/C6712D supports 16 prioritized interrupts, which are listed in Table 21. The highest priority interrupt is INT_00 (dedicated to RESET) while the lowest priority is INT_15. The first four interrupts are non-maskable and fixed. The remaining interrupts (4−15) are maskable and default to the interrupt source listed in Table 21. However, their interrupt source may be reprogrammed to any one of the sources listed in Table 22 (Interrupt Selector). Table 22 lists the selector value corresponding to each of the alternate interrupt sources. The selector choice for interrupts 4−15 is made by programming the corresponding fields (listed in Table 21) in the MUXH (address 0x019C0000) and MUXL (address 0x019C0004) registers. Table 21. DSP Interrupts [C6712C/C6712D] Table 22. Interrupt Selector [12C/12D] DSP INTERRUPT NUMBER INTERRUPT SELECTOR CONTROL REGISTER DEFAULT SELECTOR VALUE (BINARY) DEFAULT INTERRUPT EVENT INTERRUPT SELECTOR VALUE (BINARY) INT_00 − − RESET 00000 − − INT_01 − − NMI 00001 TINT0 Timer 0 INT_02 − − Reserved 00010 TINT1 Timer 1 INT_03 − − 00011 SDINT EMIF INT_04 MUXL[4:0] 00100 Reserved GPINT4† 00100 GPIO INT_05 MUXL[9:5] 00101 INT_06 MUXL[14:10] 00110 GPINT5† GPINT6† GPINT4† GPINT5† GPIO 00101 INTERRUPT EVENT MODULE GPIO INT_07 MUXL[20:16] 00111 GPINT7† 00111 GPINT6† GPINT7† INT_08 MUXL[25:21] 01000 EDMAINT 01000 EDMAINT EDMA INT_09 MUXL[30:26] 01001 EMUDTDMA 01001 EMUDTDMA Emulation INT_10 MUXH[4:0] 00011 SDINT 01010 EMURTDXRX Emulation INT_11 MUXH[9:5] 01010 EMURTDXRX 01011 EMURTDXTX Emulation INT_12 MUXH[14:10] 01011 EMURTDXTX 01100 XINT0 McBSP0 INT_13 MUXH[20:16] 00000 DSPINT 01101 RINT0 McBSP0 INT_14 MUXH[25:21] 00001 TINT0 01110 XINT1 McBSP1 INT_15 MUXH[30:26] 00010 TINT1 01111 RINT1 McBSP1 10000 GPINT0 GPIO 00110 GPIO † Interrupt Events GPINT4, GPINT5, GPINT6, and GPINT7 are outputs from the GPIO module (GP). They originate from the device pins GP[4](EXT_INT4), GP[5](EXT_INT5), GP[6](EXT_INT6), and GP[7](EXT_INT7). These pins can be used as edge-sensitive EXT_INTx with polarity controlled by the External Interrupt Polarity Register (EXTPOL.[3:0]). The corresponding pins must first be enabled in the GPIO module by setting the corresponding enable bits in the GP Enable Register (GPEN.[7:4]), and configuring them as inputs in the GP Direction Register (GPDIR.[7:4]). These interrupts can be controlled through the GPIO module in addition to the simple EXTPOL.[3:0] bits. For more information on interrupt control via the GPIO module, see the TMS320C6000 DSP General-Purpose Input/Output (GPIO) Reference Guide (literature number SPRU584). [C6712C/C6712D only]. POST OFFICE BOX 1443 • HOUSTON, TEXAS 77251−1443 41 SGUS055 − SEPTEMBER 2004 EDMA channel synchronization events [C6712 only] The C67x EDMA on the C6712 device supports up to 16 EDMA channels. Four of the sixteen channels (channels 8−11) are reserved for EDMA chaining, leaving 12 EDMA channels available to service peripheral devices. Table 23 lists the source of synchronization events associated with each of the programmable EDMA channels. For the C6712, the association of an event to a channel is fixed; each of the EDMA channels has one specific event associated with it. For more detailed information on the EDMA module, associated channels, and event-transfer chaining, see the TMS320C6000 DSP Enhanced Direct Memory Access (EDMA) Controller Reference Guide (literature number SPRU234). Table 23. 320C6712 EDMA Channel Synchronization Events EDMA CHANNEL EVENT NAME EVENT DESCRIPTION 0 − 1 TINT0 Reserved Timer 0 interrupt 2 TINT1 Timer 1 interrupt 3 SD_INT 4 EXT_INT4 External interrupt pin 4 5 EXT_INT5 External interrupt pin 5 6 EXT_INT6 External interrupt pin 6 7 8† EXT_INT7 External interrupt pin 7 EMIF SDRAM timer interrupt EDMA_TCC8 EDMA transfer complete code (TCC) 1000b interrupt 9† 10† EDMA_TCC9 EDMA TCC 1001b interrupt EDMA_TCC10 EDMA TCC 1010b interrupt 11† EDMA_TCC11 EDMA TCC 1011b interrupt 12 XEVT0 McBSP0 transmit event 13 REVT0 McBSP0 receive event 14 XEVT1 McBSP1 transmit event 15 REVT1 McBSP1 receive event † EDMA channels 8 through 11 are used for transfer chaining only. For more detailed information on event-transfer chaining, see the TMS320C6000 DSP Enhanced Direct Memory Access (EDMA) Controller Reference Guide (literature number SPRU234). 42 POST OFFICE BOX 1443 • HOUSTON, TEXAS 77251−1443 SGUS055 − SEPTEMBER 2004 EDMA module and EDMA selector [12C/12D only] The C67x EDMA for the C6712C/C6712D device also supports up to 16 EDMA channels. Four of the sixteen channels (channels 8−11) are reserved for EDMA chaining, leaving 12 EDMA channels available to service peripheral devices. On the C6712C/C6712D device, the user, through the EDMA selector registers, can control the EDMA channels servicing peripheral devices. The EDMA selector registers are located at addresses 0x01A0FF00 (ESEL0), 0x01A0FF04 (ESEL1), and 0x01A0FF0C (ESEL3). These EDMA selector registers control the mapping of the EDMA events to the EDMA channels. Each EDMA event has an assigned EDMA selector code (see Table 25). By loading each EVTSELx register field with an EDMA selector code, users can map any desired EDMA event to any specified EDMA channel. Table 24 lists the default EDMA selector value for each EDMA channel. See Table 24 and Table 25 for the EDMA Event Selector registers and their associated bit descriptions. Table 24. EDMA Channels [C6712C/C6712D Only] Table 25. EDMA Selector [12C/12D Only] EDMA CHANNEL EDMA SELECTOR CONTROL REGISTER DEFAULT SELECTOR VALUE (BINARY) DEFAULT EDMA EVENT EDMA SELECTOR CODE (BINARY) 0 ESEL0[5:0] 000000 − 000000 1 ESEL0[13:8] 000001 TINT0 000001 TINT0 TIMER0 2 ESEL0[21:16] 000010 TINT1 000010 TINT1 TIMER1 3 ESEL0[29:24] 000011 SDINT 000011 SDINT EMIF 4 ESEL1[5:0] 000100 GPINT4† 000100 GPINT4† GPIO 5 ESEL1[13:8] 000101 GPINT5† 000101 GPINT5† GPIO 6 ESEL1[21:16] 000110 GPINT6† 000110 GPINT6† GPIO 7 ESEL1[29:24] 000111 GPINT7† 000111 GPINT7† GPIO 8 − − TCC8 (Chaining) 001000 9 − − TCC9 (Chaining) 001001 10 − − TCC10 (Chaining) 001010 11 − − TCC11 (Chaining) 001011 12 ESEL3[5:0] 001000 XEVT0 001100 XEVT0 McBSP0 13 ESEL3[13:8] 001001 REVT0 001101 REVT0 McBSP0 14 ESEL3[21:16] 001010 XEVT1 001110 XEVT1 McBSP1 15 ESEL3[29:24] 001011 REVT1 001111 REVT1 010000−111111 EDMA EVENT MODULE Reserved Reserved Reserved GPINT2 GPIO Reserved McBSP1 Reserved † The GPINT[4−7] interrupt events are sourced from the GPIO module via the external interrupt capable GP[4−7] pins [12C/12D only]. POST OFFICE BOX 1443 • HOUSTON, TEXAS 77251−1443 43 SGUS055 − SEPTEMBER 2004 EDMA module and EDMA selector [12C/12D only] (continued) Table 26. EDMA Event Selector Registers (ESEL0, ESEL1, and ESEL3) ESEL0 Register (0x01A0 FF00) 30 31 29 28 27 24 23 22 21 20 19 Reserved EVTSEL3 Reserved EVTSEL2 R−0 R/W−00 0011b R−0 R/W−00 0010b 14 15 13 12 11 8 7 6 5 4 16 0 3 Reserved EVTSEL1 Reserved EVTSEL0 R−0 R/W−00 0001b R−0 R/W−00 0000b Legend: R = Read only, R/W = Read/Write; -n = value after reset ESEL1 Register (0x01A0 FF04) 30 31 29 28 27 24 23 22 21 20 19 Reserved EVTSEL7 Reserved EVTSEL6 R−0 R/W−00 0111b R−0 R/W−00 0110b 14 15 13 12 11 8 6 5 7 4 16 0 3 Reserved EVTSEL5 Reserved EVTSEL4 R−0 R/W−00 0101b R−0 R/W−00 0100b Legend: R = Read only, R/W = Read/Write; -n = value after reset ESEL3 Register (0x01A0 FF0C) 30 31 29 28 27 24 23 22 21 20 19 Reserved EVTSEL15 Reserved EVTSEL14 R−0 R/W−00 1111b R−0 R/W−00 1110b 14 15 13 12 11 8 7 6 5 4 16 3 0 Reserved EVTSEL13 Reserved EVTSEL12 R−0 R/W−00 1101b R−0 R/W−00 1100b Legend: R = Read only, R/W = Read/Write; -n = value after reset Table 27. EDMA Event Selection Registers (ESEL0, ESEL1, and ESEL3) Description BIT # NAME 31:30 23:22 15:14 7:6 Reserved DESCRIPTION Reserved. Read-only, writes have no effect. EDMA event selection bits for channel x. Allows mapping of the EDMA events to the EDMA channels. 29:24 21:16 13:8 5:0 EVTSELx The EVTSEL0 through EVTSEL15 bits correspond to the channels 0 to 15, respectively. These EVTSELx fields are user−selectable. By configuring the EVTSELx fields to the EDMA selector value of the desired EDMA sync event number (see Table 25), users can map any EDMA event to the EDMA channel. For example, if EVTSEL15 is programmed to 00 0001b (the EDMA selector code for TINT0), then channel 15 is triggered by Timer0 TINT0 events. 44 POST OFFICE BOX 1443 • HOUSTON, TEXAS 77251−1443 SGUS055 − SEPTEMBER 2004 clock PLL [C6712 only] All of the internal C6712 clocks are generated from a single source through the CLKIN pin. This source clock either drives the PLL, which multiplies the source clock in frequency to generate the internal CPU clock, or bypasses the PLL to become the internal CPU clock. To use the PLL to generate the CPU clock, the external PLL filter circuit must be properly designed. Figure 8 shows the external PLL circuitry for either x1 (PLL bypass) or x4 PLL multiply modes. Figure 9 shows the external PLL circuitry for a system with ONLY x1 (PLL bypass) mode. To minimize the clock jitter, a single clean power supply should power both the C6712 device and the external clock oscillator circuit. Noise coupling into PLLF will directly impact PLL clock jitter. The minimum CLKIN rise and fall times should also be observed. For the input clock timing requirements, see the input and output clocks electricals section. Table 28 lists some examples of compatible CLKIN external clock sources. Table 28. Compatible CLKIN External Clock Sources [C6712] COMPATIBLE PARTS FOR EXTERNAL CLOCK SOURCES (CLKIN) PART NUMBER MANUFACTURER JITO-2 Fox Electronix STA series, ST4100 series SaRonix Corporation SG-636 Epson America 342 Corning Frequency Control ICS525-02 Integrated Circuit Systems Oscillators PLL 3.3V EMI Filter PLLV Internal to C6712 PLL CLKMODE0 C3 10 mF PLLMULT C4 0.1 mF PLLCLK CLKIN CLKIN 1 LOOP FILTER 0 CPU CLOCK PLL Multiply Factors CPU Clock Frequency f(CPU CLOCK) 0 x1(BYPASS) 1 x f(CLKIN) 1 x4 4 x f(CLKIN) C2 PLLG CLKMODE0 PLLF Available Multiply Factors (For C1, C2, and R1 values, see Table 29.) C1 R1 NOTES: A. Keep the lead length and the number of vias between the PLLF pin, the PLLG pin, and R1, C1, and C2 to a minimum. In addition, place all PLL external components (R1, C1, C2, C3, C4, and the EMI Filter) as close to the C6000 DSP device as possible. For the best performance, TI recommends that all the PLL external components be on a single side of the board without jumpers, switches, or components other than the ones shown. B. For reduced PLL jitter, maximize the spacing between switching signals and the PLL external components (R1, C1, C2, C3, C4, and the EMI filter). C. The 3.3-V supply for the EMI filter must be from the same 3.3-V power plane supplying the I/O voltage, DVDD. D. EMI filter manufacturer: TDK part number ACF451832-333, 223, 153, 103. Panasonic part number EXCCET103U. Figure 8. External PLL Circuitry for Either PLL x4 Mode or x1 (Bypass) Mode [C6712] POST OFFICE BOX 1443 • HOUSTON, TEXAS 77251−1443 45 SGUS055 − SEPTEMBER 2004 clock PLL [C6712 only] (continued) 3.3V PLLV Internal to C6712 CLKMODE0 PLL PLLMULT PLLCLK CLKIN CLKIN LOOP FILTER 1 CPU CLOCK PLLF PLLG 0 NOTES: A. For a system with ONLY PLL x1 (bypass) mode, short the PLLF terminal to the PLLG terminal. B. The 3.3-V supply for the EMI filter must be from the same 3.3-V power plane supplying the I/O voltage, DVDD. Figure 9. External PLL Circuitry for x1 (Bypass) Mode Only [C6712] Table 29. C6712 PLL Component Selection Table CLKMODE CLKIN RANGE (MHz) CPU CLOCK FREQUENCY (CLKOUT1) RANGE (MHz) CLKOUT2 RANGE (MHz) R1 [±1%] (Ω) C1 [±10%] (nF) C2 [±10%] (pF) TYPICAL LOCK TIME (µs)† x4 16.3−37.5 65−150 32.5−75 60.4 27 560 75 † Under some operating conditions, the maximum PLL lock time may vary as much as 150% from the specified typical value. For example, if the typical lock time is specified as 100 µs, the maximum value may be as long as 250 µs. 46 POST OFFICE BOX 1443 • HOUSTON, TEXAS 77251−1443 SGUS055 − SEPTEMBER 2004 PLL and PLL controller [C6712C/C6712D only] The 320C6712C/C6712D includes a PLL and a flexible PLL controller peripheral consisting of a prescaler (D0) and four dividers (OSCDIV1, D1, D2, and D3). The PLL controller is able to generate different clocks for different parts of the system (i.e., DSP core, Peripheral Data Bus, External Memory Interface, McASP, and other peripherals). Figure 10 illustrates the PLL, the PLL controller, and the clock generator logic. PLLHV +3.3 V C1 EMI filter 10 µF C2 0.1 µF CLKMODE0 PLLOUT CLKIN PLLREF DIVIDER D0 1 0 Reserved /1, /2, ..., /32 ENA PLLEN (PLL_CSR.[0]) PLL x4 to x25 1 0 D1EN (PLLDIV1.[15]) D0EN (PLLDIV0.[15]) OSCDIV1 CLKOUT3 For Use in System D2EN (PLLDIV2.[15]) /1, /2, ..., /32 ENA DIVIDER D1† /1, /2, ..., /32 ENA SYSCLK1 (DSP Core) DIVIDER D2† /1, /2, ..., /32 ENA SYSCLK2 (Peripherals) DIVIDER D3 /1, /2, ..., /32 OD1EN (OSCDIV1.[15]) D3EN (PLLDIV3.[15]) SYSCLK3 ENA ECLKIN (EMIF Clock Input) 1 0 EKSRC Bit (DEVCFG.[4]) EMIF C6712C/C6712D DSPs ECLKOUT † Dividers D1 and D2 must never be disabled. Never write a “0” to the D1EN or D2EN bits in the PLLDIV1 and PLLDIV2 registers. NOTES: A. Place all PLL external components (C1, C2, and the EMI Filter) as close to the C67x DSP device as possible. For the best performance, TI recommends that all the PLL external components be on a single side of the board without jumpers, switches, or components other than the ones shown. B. For reduced PLL jitter, maximize the spacing between switching signals and the PLL external components (C1, C2, and the EMI Filter). C. The 3.3-V supply for the EMI filter must be from the same 3.3-V power plane supplying the I/O voltage, DVDD. D. EMI filter manufacturer TDK part number ACF451832-333, -223, -153, -103. Panasonic part number EXCCET103U. Figure 10. PLL and Clock Generator Logic [C6712C/C6712D Only] POST OFFICE BOX 1443 • HOUSTON, TEXAS 77251−1443 47 SGUS055 − SEPTEMBER 2004 PLL and PLL controller [C6712C/C6712D only] (continued) The PLL Reset Time is the amount of wait time needed when resetting the PLL (writing PLLRST=1), in order for the PLL to properly reset, before bringing the PLL out of reset (writing PLLRST = 0). For the PLL Reset Time value, see Table 30. The PLL Lock Time is the amount of time from when PLLRST = 0 with PLLEN = 0 (PLL out of reset, but still bypassed) to when the PLLEN bit can be safely changed to “1” (switching from bypass to the PLL path), see Table 30 and Figure 10. Under some operating conditions, the maximum PLL Lock Time may vary from the specified typical value. For the PLL Lock Time values, see Table 30. Table 30. PLL Lock and Reset Times (C6712C/C6712D only) MIN PLL Lock Time PLL Reset Time TYP MAX UNIT 75 187.5 µs 125 ns Table 31 shows the C6712C/C6712D device’s CLKOUT signals, how they are derived and by what register control bits, and the default settings. For more details on the PLL, see the PLL and Clock Generator Logic diagram (Figure 10). Table 31. CLKOUT Signals, Default Settings, and Control CLOCK OUTPUT SIGNAL NAME DEFAULT SETTING (ENABLED or DISABLED) CONTROL BIT(s) (Register) CLKOUT2 ON (ENABLED) D2EN = 1 (PLLDIV2.[15]) CK2EN = 1 (EMIF GBLCTL.[3]) CLKOUT3 ON (ENABLED) OD1EN = 1 (OSCDIV1.[15]) DESCRIPTION SYSCLK2 selected [default] Derived from CLKIN SYSCLK3 selected [default]. ECLKOUT ON (ENABLED); derived from SYSCLK3 EKSRC = 0 (DEVCFG.[4]) EKEN = 1 (EMIF GBLCTL.[5]) To select ECLKIN as source: EKSRC = 1 (DEVCFG.[4]) and EKEN = 1 (EMIF GBLCTL.[5]) This input clock is directly available as an internal high-frequency clock source that may be divided down by a programmable divider OSCDIV1 (/1, /2, /3, ..., /32) and output on the CLKOUT3 pin for other use in the system. Figure 10 shows that the input clock source may be divided down by divider PLLDIV0 (/1, /2, ..., /32) and then multiplied up by a factor of x4, x5, x6, and so on, up to x25. Either the input clock (PLLEN = 0) or the PLL output (PLLEN = 1) then serves as the high-frequency reference clock for the rest of the DSP system. The DSP core clock, the peripheral bus clock, and the EMIF clock may be divided down from this high-frequency clock (each with a unique divider) . For example, with a 40-MHz input, if the PLL output is configured for 300 MHz, the DSP core may be operated at 150 MHz (/2) while the EMIF may be configured to operate at a rate of 60 MHz (/5). Note that there is a specific minimum and maximum reference clock (PLLREF) and output clock (PLLOUT) for the block labeled PLL in Figure 10, as well as for the DSP core, peripheral bus, and EMIF. The clock generator must not be configured to exceed any of these constraints (certain combinations of external clock input, internal dividers, and PLL multiply ratios might not be supported). See Table 32 for the PLL clocks input and output frequency ranges. 48 POST OFFICE BOX 1443 • HOUSTON, TEXAS 77251−1443 SGUS055 − SEPTEMBER 2004 PLL and PLL controller [C6712C/C6712D only] (continued) Table 32. PLL Clock Frequency Ranges†‡ GDP16 CLOCK SIGNAL UNIT MIN MAX PLLREF (PLLEN = 1) 12 100 MHz PLLOUT 140 600 MHz SYSCLK1 − Device Speed (DSP Core) MHz SYSCLK3 (EKSRC = 0) − 100 MHz † SYSCLK2 rate must be exactly half of SYSCLK1. ‡ Also see the electrical specification (timing requirements and switching characteristics parameters) in the Input and Output Clocks section of this data sheet. The EMIF itself may be clocked by an external reference clock via the ECLKIN pin or can be generated on-chip as SYSCLK3. SYSCLK3 is derived from divider D3 off of PLLOUT (see Figure 10, PLL and Clock Generator Logic). The EMIF clock selection is programmable via the EKSRC bit in the DEVCFG register. The settings for the PLL multiplier and each of the dividers in the clock generation block may be reconfigured via software at run time. If either the input to the PLL changes due to D0, CLKMODE0, or CLKIN, or if the PLL multiplier is changed, then software must enter bypass first and stay in bypass until the PLL has had enough time to lock (see electrical specifications). For the programming procedure, see the TMS320C6000 DSP Software-Programmable Phase-Locked Loop (PLL) Controller Reference Guide (literature number SPRU233). SYSCLK2 is the internal clock source for peripheral bus control. SYSCLK2 (Divider D2) must be programmed to be half of the SYSCLK1 rate. For example, if D1 is configured to divide-by-2 mode (/2), then D2 must be programmed to divide-by-4 mode (/4). SYSCLK2 is also tied directly to CLKOUT2 pin (see Figure 10). During the programming transition of Divider D1 and Divider D2 (resulting in SYSCLK1 and SYSCLK2 output clocks, see Figure 10), the order of programming the PLLDIV1 and PLLDIV2 registers must be observed to ensure that SYSCLK2 always runs at half the SYSCLK1 rate or slower. For example, if the divider ratios of D1 and D2 are to be changed from /1, /2 (respectively) to /5, /10 (respectively) then, the PLLDIV2 register must be programmed before the PLLDIV1 register. The transition ratios become /1, /2; /1, /10; and then /5, /10. If the divider ratios of D1 and D2 are to be changed from /3, /6 to /1, /2 then, the PLLDIV1 register must be programmed before the PLLDIV2 register. The transition ratios, for this case, become /3, /6; /1, /6; and then /1, /2. The final SYSCLK2 rate must be exactly half of the SYSCLK1 rate. Note that Divider D1 and Divider D2 must always be enabled (i.e., D1EN and D2EN bits are set to “1” in the PLLDIV1 and PLLDIV2 registers). For detailed information on the clock generator (PLL Controller registers) and their associated software bit descriptions, see Table 33 through Table 36. POST OFFICE BOX 1443 • HOUSTON, TEXAS 77251−1443 49 SGUS055 − SEPTEMBER 2004 PLL and PLL controller [C6712C/C6712D only] (continued) PLLCSR Register (0x01B7 C100) 28 31 27 24 23 20 19 16 Reserved R−0 15 12 11 8 7 6 5 4 3 2 1 0 Reserved STABLE Reserved PLLRST Reserved PLLPWRDN PLLEN R−0 R−x R−0 RW−1 R/W−0 R/W−0b RW−0 Legend: R = Read only, R/W = Read/Write; -n = value after reset Table 33. PLL Control/Status Register (PLLCSR) BIT # NAME 31:7 Reserved Reserved. Read-only, writes have no effect. 6 STABLE Clock Input Stable. This bit indicates if the clock input has stabilized. 0 – Clock input not yet stable. Clock counter is not finished counting (default). 1 – Clock input stable. 5:4 Reserved Reserved. Read-only, writes have no effect. 3 PLLRST Asserts RESET to PLL 0 – PLL Reset Released. 1 – PLL Reset Asserted (default). 2 Reserved Reserved. The user must write a “0” to this bit. 1 PLLPWRDN 0 50 PLLEN DESCRIPTION Select PLL Power Down 0 – PLL Operational (default). 1 – PLL Placed in Power-Down State. PLL Mode Enable 0 – Bypass Mode (default). PLL disabled. Divider D0 and PLL are bypassed. SYSCLK1/SYSCLK2/SYSCLK3 are divided down directly from input reference clock. 1 – PLL Enabled. Divider D0 and PLL are not bypassed. SYSCLK1/SYSCLK2/SYSCLK3 are divided down from PLL output. POST OFFICE BOX 1443 • HOUSTON, TEXAS 77251−1443 SGUS055 − SEPTEMBER 2004 PLL and PLL controller [C6712C/C6712D only] (continued) PLLM Register (0x01B7 C110) 24 23 28 27 31 20 19 16 Reserved R−0 15 12 11 8 7 6 5 4 3 2 Reserved PLLM R−0 R/W−0 0111 1 0 Legend: R = Read only, R/W = Read/Write; -n = value after reset Table 34. PLL Multiplier Control Register (PLLM) BIT # NAME 31:5 Reserved 4:0 PLLM DESCRIPTION Reserved. Read-only, writes have no effect. PLL multiply mode [default is x7 (0 0111)]. 00000 = Reserved 10000 = 00001 = Reserved 10001 = 00010 = Reserved 10010 = 00011 = Reserved 10011 = 00100 = x4 10100 = 00101 = x5 10101 = 00110 = x6 10110 = 00111 = x7 10111 = 01000 = x8 11000 = 01001 = x9 11001 = 01010 = x10 11010 = 01011 = x11 11011 = 01100 = x12 11100 = 01101 = x13 11101 = 01110 = x14 11110 = 01111 = x15 11111 = x16 x17 x18 x19 x20 x21 x22 x23 x24 x25 Reserved Reserved Reserved Reserved Reserved Reserved PLLM select values 00000 through 00011 and 11010 through 11111 are not supported. POST OFFICE BOX 1443 • HOUSTON, TEXAS 77251−1443 51 SGUS055 − SEPTEMBER 2004 PLL and PLL controller [C6712C/C6712D only] (continued) PLLDIV0, PLLDIV1, PLLDIV2, and PLLDIV3 Registers (0x01B7 C114, 0x01B7 C118, 0x01B7 C11C, and 0x01B7 C120, respectively) 28 31 24 27 23 20 19 16 Reserved R−0 14 15 12 11 8 7 5 4 3 2 DxEN Reserved PLLDIVx R/W−1 R−0 R/W−x xxxx† 1 0 Legend: R = Read only, R/W = Read/Write; -n = value after reset † Default values for the PLLDIV0, PLLDIV1, PLLDIV2, and PLLDIV3 bits are /1 (0 0000), /1 (0 0000), /2 (0 0001), and /2 (0 0001), respectively. CAUTION: D1 and D2 should never be disabled. D3 should only be disabled if ECLKIN is used. Table 35. PLL Wrapper Divider x Registers (Prescaler Divider D0 and Post-Scaler Dividers D1, D2, and D3)‡ BIT # NAME 31:16 Reserved 15 DxEN 14:5 Reserved DESCRIPTION Reserved. Read-only, writes have no effect. Divider Dx Enable (where x denotes 0 through 3). 0 – Divider x Disabled. No clock output. 1 – Divider x Enabled (default). These divider-enable bits are device-specific and must be set to 1 to enable. Reserved. Read-only, writes have no effect. PLL Divider Ratio [Default values for the PLLDIV0, PLLDIV1, PLLDIV2, and PLLDIV3 bits are /1, /1, /2, and /2, respectively]. 4:0 PLLDIVx 00000 00001 00010 00011 00100 00101 00110 00111 01000 01001 01010 01011 01100 01101 01110 01111 = = = = = = = = = = = = = = = = /1 /2 /3 /4 /5 /6 /7 /8 /9 /10 /11 /12 /13 /14 /15 /16 10000 10001 10010 10011 10100 10101 10110 10111 11000 11001 11010 11011 11100 11101 11110 11111 = = = = = = = = = = = = = = = = /17 /18 /19 /20 /21 /22 /23 /24 /25 /26 /27 /28 /29 /30 /31 /32 ‡ Note that SYSCLK2 must run at half the rate of SYSCLK1. Therefore, the divider ratio of D2 must be two times slower than D1. For example, if D1 is set to /2, then D2 must be set to /4. 52 POST OFFICE BOX 1443 • HOUSTON, TEXAS 77251−1443 SGUS055 − SEPTEMBER 2004 PLL and PLL controller [C6712C/C6712D only] (continued) OSCDIV1 Register (0x01B7 C124) 24 23 28 27 31 20 19 16 Reserved R−0 15 14 12 11 8 7 5 4 3 2 OD1EN Reserved OSCDIV1 R/W−1 R−0 R/W−0 0111 1 0 Legend: R = Read only, R/W = Read/Write; -n = value after reset The OSCDIV1 register controls the oscillator divider 1 for CLKOUT3. The CLKOUT3 signal does not go through the PLL path. Table 36. Oscillator Divider 1 Register (OSCDIV1) BIT # NAME 31:16 Reserved 15 OD1EN 14:5 Reserved DESCRIPTION Reserved. Read-only, writes have no effect. Oscillator Divider 1 Enable. 0 – Oscillator Divider 1 Disabled. 1 – Oscillator Divider 1 Enabled (default). Reserved. Read-only, writes have no effect. Oscillator Divider 1 Ratio [default is /8 (0 0111)]. 4:0 OSCDIV1 00000 00001 00010 00011 00100 00101 00110 00111 01000 01001 01010 01011 01100 01101 01110 01111 = = = = = = = = = = = = = = = = /1 /2 /3 /4 /5 /6 /7 /8 /9 /10 /11 /12 /13 /14 /15 /16 10000 10001 10010 10011 10100 10101 10110 10111 11000 11001 11010 11011 11100 11101 11110 11111 POST OFFICE BOX 1443 = = = = = = = = = = = = = = = = /17 /18 /19 /20 /21 /22 /23 /24 /25 /26 /27 /28 /29 /30 /31 /32 • HOUSTON, TEXAS 77251−1443 53 SGUS055 − SEPTEMBER 2004 general-purpose input/output (GPIO) To use the GP[7:4, 2] software-configurable GPIO pins, the GPxEN bits in the GP Enable (GPEN) Register and the GPxDIR bits in the GP Direction (GPDIR) Register must be properly configured. GPxEN = 1 GP[x] pin is enabled GPxDIR = 0 GP[x] pin is an input GPxDIR = 1 GP[x] pin is an output where “x” represents one of the 7 through 4, or 2 GPIO pins Figure 11 shows the GPIO enable bits in the GPEN register for the C6712C and C6712D devices. To use any of the GPx pins as general-purpose input/output functions, the corresponding GPxEN bit must be set to “1” (enabled). Default values are device-specific, so refer to Figure 11 for the C6712C/C6712D default configuration. 31 24 23 16 Reserved R-0 15 14 13 12 11 10 9 8 7 6 5 4 3 2 Reserved GP7 EN GP6 EN GP5 EN GP4 EN — GP2 EN 1 0 — — R/W-0 R/W-1 R/W-1 R/W-1 R/W-1 R/W-0 R/W-0 R/W-0 R/W-0 Legend: R/W = Readable/Writeable; -n = value after reset, -x = undefined value after reset Figure 11. GPIO Enable Register (GPEN) [Hex Address: 01B0 0000] Figure 12 shows the GPIO direction bits in the GPDIR register. This register determines if a given GPIO pin is an input or an output providing the corresponding GPxEN bit is enabled (set to “1”) in the GPEN register. By default, all the GPIO pins are configured as input pins. 31 24 23 16 Reserved R-0 15 14 13 12 11 10 9 8 Reserved R/W-0 R/W-0 R/W-0 R/W-0 R/W-0 R/W-0 R/W-0 R/W-0 7 6 5 4 3 2 1 0 GP7 DIR GP6 DIR GP5 DIR GP4 DIR — GP2 DIR — — R/W-0 R/W-0 R/W-0 R/W-0 R/W-0 R/W-0 R/W-0 R/W-0 Legend: R/W = Readable/Writeable; -n = value after reset, -x = undefined value after reset Figure 12. GPIO Direction Register (GPDIR) [Hex Address: 01B0 0004] For more detailed information on general-purpose inputs/outputs (GPIOs), see the TMS320C6000 DSP General-Purpose Input/Output (GPIO) Reference Guide (literature number SPRU584). 54 POST OFFICE BOX 1443 • HOUSTON, TEXAS 77251−1443 SGUS055 − SEPTEMBER 2004 power-down mode logic Figure 13 shows the power-down mode logic on the C6712C/12D. CLKOUT1‡ CLKOUT2 Internal Clock Tree Clock Distribution and Dividers PD1 PD2 PowerDown Logic Clock PLL IFR Internal Peripherals IER PWRD CSR CPU PD3 320C6712/12C/12D CLKIN RESET † External input clocks, with the exception of CLKOUT3 [12C/12D only] and CLKIN, are not gated by the power-down mode logic. ‡ CLKOUT1 is applicable on the C6712 device only. Figure 13. Power-Down Mode Logic† triggering, wake-up, and effects The power-down modes and their wake-up methods are programmed by setting the PWRD field (bits 15−10) of the control status register (CSR). The PWRD field of the CSR is shown in Figure 14 and described in Table 37. When writing to the CSR, all bits of the PWRD field should be set at the same time. Logic 0 should be used when “writing” to the reserved bit (bit 15) of the PWRD field. The CSR is discussed in detail in the TMS320C6000 CPU and Instruction Set Reference Guide (literature number SPRU189). POST OFFICE BOX 1443 • HOUSTON, TEXAS 77251−1443 55 SGUS055 − SEPTEMBER 2004 31 16 15 14 13 12 11 10 Reserved Enable or Non-Enabled Interrupt Wake Enabled Interrupt Wake PD3 PD2 PD1 R/W-0 R/W-0 R/W-0 R/W-0 R/W-0 R/W-0 7 9 8 0 Legend: R/W−x = Read/write reset value NOTE: The shadowed bits are not part of the power-down logic discussion and therefore are not covered here. For information on these other bit fields in the CSR register, see the TMS320C6000 CPU and Instruction Set Reference Guide (literature number SPRU189). Figure 14. PWRD Field of the CSR Register A delay of up to nine clock cycles may occur after the instruction that sets the PWRD bits in the CSR before the PD mode takes effect. As best practice, NOPs should be padded after the PWRD bits are set in the CSR to account for this delay. If PD1 mode is terminated by a non-enabled interrupt, the program execution returns to the instruction where PD1 took effect. If PD1 mode is terminated by an enabled interrupt, the interrupt service routine will be executed first, then the program execution returns to the instruction where PD1 took effect. In the case with an enabled, interrupt, the GIE bit in the CSR and the NMIE bit in the interrupt enable register (IER) must also be set in order for the interrupt service routine to execute; otherwise, execution returns to the instruction where PD1 took effect upon PD1 mode termination by an enabled interrupt. PD2 and PD3 modes can only be aborted by device reset. Table 37 summarizes all the power-down modes. 56 POST OFFICE BOX 1443 • HOUSTON, TEXAS 77251−1443 SGUS055 − SEPTEMBER 2004 Table 37. Characteristics of the Power-Down Modes PRWD FIELD (BITS 15−10) POWER-DOWN MODE WAKE-UP METHOD 000000 No power-down — — 001001 PD1 Wake by an enabled interrupt 010001 PD1 Wake by an enabled or non-enabled interrupt 011010 011100 PD2† PD3† EFFECT ON CHIP’S OPERATION CPU halted (except for the interrupt logic) Power-down mode blocks the internal clock inputs at the boundary of the CPU, preventing most of the CPU’s logic from switching. During PD1, EDMA transactions can proceed between peripherals and internal memory. Wake by a device reset Output clock from PLL is halted, stopping the internal clock structure from switching and resulting in the entire chip being halted. All register and internal RAM contents are preserved. All functional I/O “freeze” in the last state when the PLL clock is turned off. Wake by a device reset Input clock to the PLL stops generating clocks. All register and internal RAM contents are preserved. All functional I/O “freeze” in the last state when the PLL clock is turned off. Following reset, the PLL needs time to re-lock, just as it does following power-up. Wake-up from PD3 takes longer than wake-up from PD2 because the PLL needs to be re-locked, just as it does following power-up. All others Reserved — — † When entering PD2 and PD3, all functional I/O remains in the previous state. However, for peripherals which are asynchronous in nature or peripherals with an external clock source, output signals may transition in response to stimulus on the inputs. Under these conditions, peripherals will not operate according to specifications. On C6712D silicon revision 2.0 and C6712C silicon revision 1.1, the device includes a programmable PLL which allows software control of PLL bypass via the PLLEN bit in the PLLCSR register. With this enhanced functionality comes some additional considerations when entering power-down modes. The power-down modes (PD2 and PD3) function by disabling the PLL to stop clocks to the device. However, if the PLL is bypassed (PLLEN = 0), the device will still receive clocks from the external clock input (CLKIN). Therefore, bypassing the PLL makes the power-down modes PD2 and PD3 ineffective. Make sure that the PLL is enabled by writing a “1” to PLLEN bit (PLLCSR.0) before writing to either PD3 (CSR.11) or PD2 (CSR.10) to enter a power-down mode. power-supply sequencing TI DSPs do not require specific power sequencing between the core supply and the I/O supply. However, systems should be designed to ensure that neither supply is powered up for extended periods of time (>1 second) if the other supply is below the proper operating voltage. system-level design considerations System-level design considerations, such as bus contention, may require supply sequencing to be implemented. In this case, for C6712, the core supply should be powered up at the same time as, or prior to (and powered down after) the I/O buffers. For C6712C/12D, the core supply should be powered up prior to (and powered down after), the I/O buffers. This is to ensure that the I/O buffers receive valid inputs from the core before the output buffers are powered up, thus, preventing bus contention with other chips on the board. power-supply design considerations A dual-power supply with simultaneous sequencing can be used to eliminate the delay between core and I/O power up. A Schottky diode can also be used to tie the core rail to the I/O rail (see Figure 15). POST OFFICE BOX 1443 • HOUSTON, TEXAS 77251−1443 57 SGUS055 − SEPTEMBER 2004 I/O Supply DVDD Schottky Diode C6000 DSP Core Supply CVDD VSS GND Figure 15. Schottky Diode Diagram Core and I/O supply voltage regulators should be located close to the DSP (or DSP array) to minimize inductance and resistance in the power delivery path. Additionally, when designing for high-performance applications utilizing the C6000 platform of DSPs, the PC board should include separate power planes for core, I/O, and ground, all bypassed with high-quality low-ESL/ESR capacitors. C6712 device applicable only On systems using C62x and C67x DSPs, like the C6712 device, the core may consume in excess of 2 A per DSP until the I/O supply powers on. This extra current results from uninitialized logic within the DSP(s). A normal current state returns once the I/O power supply turns on and the CPU sees a clock pulse. Decreasing the amount of time between the core supply power-up and the I/O supply power-up reduces the effects of the current draw. If the external supply to the DSP core cannot supply the excess current, the minimum core voltage may not be achieved until after normal current returns. This voltage starvation of the core supply during power up will not affect run-time operation. Voltage starvation can affect power supply systems that gate the I/O supply via the core supply, causing the I/O supply to never turn on. During the transition from excess to normal current, a voltage spike may be seen on the core supply. Care must be taken when designing overvoltage protection circuitry on the core supply to not restart the power sequence due to this spike. Otherwise, the supply may cycle indefinitely. 58 POST OFFICE BOX 1443 • HOUSTON, TEXAS 77251−1443 SGUS055 − SEPTEMBER 2004 power-supply decoupling In order to properly decouple the supply planes from system noise, place as many capacitors (caps) as possible close to the DSP. Assuming 0603 caps, the user should be able to fit a total of 60 caps — 30 for the core supply and 30 for the I/O supply. These caps need to be close (no more than 1.25 cm maximum distance) to the DSP to be effective. Physically smaller caps are better, such as 0402, but the size needs to be evaluated from a yield/manufacturing point-of-view. Parasitic inductance limits the effectiveness of the decoupling capacitors, therefore physically smaller capacitors should be used while maintaining the largest available capacitance value. As with the selection of any component, verification of capacitor availability over the product’s production lifetime needs to be considered. IEEE 1149.1 JTAG compatibility statement The 320C6712/12C/12D DSP requires that both TRST and RESET be asserted upon power up to be properly initialized. While RESET initializes the DSP core, TRST initializes the DSP’s emulation logic. Both resets are required for proper operation. While both TRST and RESET need to be asserted upon power up, only RESET needs to be released for the DSP to boot properly. TRST may be asserted indefinitely for normal operation, keeping the JTAG port interface and DSP’s emulation logic in the reset state. TRST only needs to be released when it is necessary to use a JTAG controller to debug the DSP or exercise the DSP’s boundary scan functionality. For maximum reliability, the 320C6712/12C/12D DSP includes an internal pulldown (IPD) on the TRST pin to ensure that TRST will always be asserted upon power up and the DSP’s internal emulation logic will always be properly initialized. JTAG controllers from Texas Instruments actively drive TRST high. However, some third-party JTAG controllers may not drive TRST high but expect the use of a pullup resistor on TRST. When using this type of JTAG controller, assert TRST to initialize the DSP after powerup and externally drive TRST high before attempting any emulation or boundary scan operations. Following the release of RESET, the low-to-high transition of TRST must be “seen” to latch the state of EMU1 and EMU0. The EMU[1:0] pins configure the device for either Boundary Scan mode or Emulation mode. For more detailed information, see the terminal functions section of this data sheet. EMIF device speed The maximum EMIF speed on the C6712C/C6712D device is 100 MHz. TI recommends utilizing I/O buffer information specification (IBIS) to analyze all AC timings to determine if the maximum EMIF speed is achievable for a given board layout. To properly use IBIS models to attain accurate timing analysis for a given system, see the Using IBIS Models for Timing Analysis application report (literature number SPRA839). For ease of design evaluation, Table 38 contains IBIS simulation results showing the maximum EMIF-SDRAM interface speeds for the given example boards (TYPE) and SDRAM speed grades. Timing analysis should be performed to verify that all AC timings are met for the specified board layout. Other configurations are also possible, but again, timing analysis must be done to verify proper AC timings. To maintain signal integrity, serial termination resistors should be inserted into all EMIF output signal lines (see the Terminal Functions table for the EMIF output signals). POST OFFICE BOX 1443 • HOUSTON, TEXAS 77251−1443 59 SGUS055 − SEPTEMBER 2004 Table 38. C6712C/C6712D Example Boards and Maximum EMIF Speed BOARD CONFIGURATION TYPE 1-Load Short Traces 2-Loads Short Traces 3-Loads Short Traces 3-Loads Long Traces EMIF INTERFACE COMPONENTS One bank of one 32-Bit SDRAM One bank of two 16-Bit SDRAMs One bank of two 32-Bit SDRAMs One bank of buffer One bank of one 32-Bit SDRAM One bank of one 32-Bit SBSRAM One bank of buffer BOARD TRACE 1 to 3-inch traces with proper termination resistors; Trace impedance ~ 50 Ω 1.2 to 3 inches from EMIF to each load, with proper termination resistors; Trace impedance ~ 78 Ω 1.2 to 3 inches from EMIF to each load, with proper termination resistors; Trace impedance ~ 78 Ω 4 to 7 inches from EMIF; Trace impedance ~ 63 Ω SDRAM SPEED GRADE MAXIMUM ACHIEVABLE EMIF-SDRAM INTERFACE SPEED 143 MHz 32-bit SDRAM (−7) 100 MHz 166 MHz 32-bit SDRAM (−6) 200 MHz 32-bit SDRAM (−5) For short traces, SDRAM data output hold time on these SDRAM speed grades cannot meet EMIF input hold time requirement (see NOTE 1). 125 MHz 16-bit SDRAM (−8E) 100 MHz 133 MHz 16-bit SDRAM (−75) 100 MHz 143 MHz 16-bit SDRAM (−7E) 100 MHz 167 MHz 16-bit SDRAM (−6A) 100 MHz 167 MHz 16-bit SDRAM (−6) 100 MHz 125 MHz 16-bit SDRAM (−8E) For short traces, EMIF cannot meet SDRAM input hold requirement (see NOTE 1). 133 MHz 16-bit SDRAM (−75) 100 MHz 143 MHz 16-bit SDRAM (−7E) 100 MHz 167 MHz 16-bit SDRAM (−6A) 100 MHz 167 MHz 16-bit SDRAM (−6) For short traces, EMIF cannot meet SDRAM input hold requirement (see NOTE 1). 143 MHz 32-bit SDRAM (−7) 83 MHz 166 MHz 32-bit SDRAM (−6) 83 MHz 183 MHz 32-bit SDRAM (−55) 83 MHz 200 MHz 32-bit SDRAM (−5) SDRAM data output hold time cannot meet EMIF input hold requirement (see NOTE 1). 183 MHz 32-bit SDRAM (−55) NOTE 1: Results are based on IBIS simulations for the given example boards (TYPE). Timing analysis should be performed to determine if timing requirements can be met for the particular system. EMIF big endian mode correctness [C6712D only] The device Endian mode pin (LENDIAN) selects the endian mode of operation (little endian or big endian) for the C6712D device. Little endian is the default setting. When Big Endian mode is selected (LENDIAN = 0), the EMIF Big Endian mode correctness pin (EMIFBE) must to be pulled low. Figure 16 shows the mapping of 16-bit and 8-bit data for C6712D devices with EMIF endianness correction. EMIF DATA LINES (PINS) WHERE DATA PRESENT ED[15:8] (BE1) ED[7:0] (BE0) 16-Bit Device in Any Endianness Mode 8-Bit Device in Any Endianness Mode † The C6712/C6712C devices support Little Endian mode of operation only. Figure 16. 16/8-Bit EMIF Big Endian Mode Correctness Mapping [C6712D Only]† This new feature does not affect systems operating in Little Endian mode, providing the default value of the C15 pin =1 is used. 60 POST OFFICE BOX 1443 • HOUSTON, TEXAS 77251−1443 SGUS055 − SEPTEMBER 2004 bootmode The C67x device resets using the active-low signal RESET and the internal reset signal (C6712C/C6712D; for the C6712 device, the RESET signal is the same as the internal reset signal). While RESET is low, the internal reset is also asserted and the device is held in reset and is initialized to the prescribed reset state. Refer to reset timing for reset timing characteristics and states of device pins during reset. The release of the internal reset signal (see the Reset Phase 3 discussion in the Reset Timing section of this data sheet) starts the processor running with the prescribed device configuration and boot mode. The C6712/C6712C/C6712D has two type of boot mode: D Emulation boot In Emulation boot mode, it is not necessary to load valid code into internal memory. The emulation driver will release the CPU from the “stalled” state, at which point the CPU will vector to address 0. Prior to beginning execution, the emulator sets a breakpoint at address 0. This prevents the execution of invalid code by halting the CPU prior to executing the first instruction. Emulation boot is a good tool in the debug phase of development. D EMIF boot (using default ROM timings) Upon the release of internal reset, the 1K-Byte ROM code located in the beginning of CE1 is copied to address 0 by the EDMA using the default ROM timings, while the CPU is internally “stalled”. The data should be stored in the endian format that the system is using. The boot process also lets you choose the width of the ROM. In this case, the EMIF automatically assembles consecutive 8-bit bytes or 16-bit half-words to form the 32-bit instruction words to be copied. The transfer is automatically done by the EDMA as a single-frame block transfer from the ROM to address 0. After completion of the block transfer, the CPU is released from the “stalled” state and starts running from address 0. absolute maximum ratings over operating case temperature range (unless otherwise noted)† Supply voltage range, CVDD (see Note 2): (C6712C/C6712D only) . . . . . . . . . . . . . . . . . . . . . − 0.3 V to 1.8 V (C6712) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . − 0.3 V to 2.3 V Supply voltage range, DVDD (see Note 2) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . −0.3 V to 4 V Input voltage ranges: (C6712C/C6712D only) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . −0.3 V to DVDD + 0.5 V (C6712) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . −0.3 V to 4 V Output voltage ranges: (C6712C/C6712D only) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . −0.3 V to DVDD + 0.5 V (C6712) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . −0.3 V to 4 V Operating case temperature range, TC . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . −40_C to 105_C Storage temperature range, Tstg . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . −65_C to 150_C † 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. NOTE 2: All voltage values are with respect to VSS. POST OFFICE BOX 1443 • HOUSTON, TEXAS 77251−1443 61 SGUS055 − SEPTEMBER 2004 recommended operating conditions CVDD Supply voltage, Core‡ DVDD Supply voltage, I/O‡ VSS Supply ground VIH High-level input voltage (C6712C/12D only) MIN 1.14§ NOM 1.20§ MAX UNIT 1.32 V (C6712) 1.71 1.8 1.89 V (C6712C/C6712D only) 3.13 3.3 3.47 V (C6712) 3.14 3.3 3.46 V 0 0 0 V (C6712C/C6712D only) All signals except CLKS1, DR1, and RESET 2 CLKS1, DR1, and RESET 2 High-level input voltage (C6712) VIL Low-level input voltage (C6712C/12D only) 2 All signals except CLKS1, DR1, and RESET CLKS1, DR1, and RESET IOH High-level output current (C6712C)¶ High-level output current (C6712D)¶ Low-level output current (C6712) IOL Low-level output current (C6712C)¶ Low-level output current (C6712D)¶ 0.8 0.3*DVDD 0.8 Low-level input voltage (C6712) High-level output current (C6712) V All signals except CLKOUT1, CLKOUT2, and ECLKOUT –4 CLKOUT1, CLKOUT2, and ECLKOUT –8 All signals except ECLKOUT, CLKOUT2, CLKOUT3, CLKS1, and DR1 –8 ECLKOUT, CLKOUT2, and CLKOUT3 All signals except ECLKOUT, CLKOUT2, CLKS1, and DR1 ECLKOUT and CLKOUT2 V mA –16 –8 mA –16 All signals except CLKOUT1, CLKOUT2, and ECLKOUT 4 CLKOUT1, CLKOUT2, and ECLKOUT 8 All signals except ECLKOUT, CLKOUT2, CLKOUT3, CLKS1, and DR1 8 ECLKOUT, CLKOUT2, and CLKOUT3 mA 16 CLKS1 and DR1 3 All signals except ECLKOUT, CLKOUT2, CLKS1, and DR1 8 ECLKOUT and CLKOUT2 mA mA mA 16 CLKS1 and DR1 3 mA TC Operating case temperature −40 105 _C ‡ For the C6712 device, the core supply should be powered up at the same time as, or prior to (and powered down after), the I/O supply. For the C6712C/12D device, the core supply should be powered up prior to (and powered down after), the I/O supply. Systems should be designed to ensure that neither supply is powered up for an extended period of time if the other supply is below the proper operating voltage. § These values are compatible with existing 1.26V designs. ¶ Refers to DC (or steady state) currents only, actual switching currents are higher. For more details, see the device-specific IBIS models. 62 POST OFFICE BOX 1443 • HOUSTON, TEXAS 77251−1443 SGUS055 − SEPTEMBER 2004 electrical characteristics over recommended ranges of supply voltage and operating case temperature† (unless otherwise noted) PARAMETER VOH VOL II IOZ High-level output voltage Low-level output voltage Input current Off-state output current TEST CONDITIONS MAX UNIT 2.4 V C6712: All signals DVDD = MIN, IOH = MAX 2.4 V 12C/12D: All signals except CLKS1 and DR1 DVDD = MIN, IOL = MAX 12C/12D: CLKS1 and DR1 C6712: All signals DVDD = MIN, IOL = MAX 12C/12D: All signals except CLKS1 and DR1 VI = VSS to DVDD 12C/12D: CLKS1 and DR1 C6712: All signals VI = VSS to DVDD 12C/12D: All signals except CLKS1 and DR1 VO = DVDD or 0 V 12C/12D: CLKS1 and DR1 Supply current, CPU + CPU memory access‡ IDD2V TYP DVDD = MIN, IOH = MAX C6712: All signals IDD2V MIN 12C/12D: All signals except CLKS1 and DR1 Supply current, peripherals‡ Core supply current (C6712C/12D)‡ VO = DVDD or 0 V C6712, CVDD = NOM, CPU clock = 100 MHz 0.4 V 0.4 V 0.4 V ±170 uA ±10 uA ±150 uA ±170 uA ±10 uA ±10 uA 336 mA C6712, CVDD = NOM, CPU clock = 100 MHz 180 mA C6712C, CVDD = 1.26 V, CPU clock = 150 MHz 430 mA C6712D, CVDD = 1.26 V, CPU clock = 167 MHz 475 mA IDD3V Supply current, I/O pins‡ C6712, DVDD = NOM, CPU clock = 100 MHz 50 mA IDD3V I/O supply current (C6712C/12D)‡ DVDD = 3.3 V, EMIF speed = 100 MHz 75 mA Ci Input capacitance Co Output capacitance C6712 7 C6712C/C6712D 7 C6712 7 C6712C/C6712D 7 pF pF † For test conditions shown as MIN, MAX, or NOM, use the appropriate value specified in the recommended operating conditions table. ‡ For the C6712 device, these currents were measured with average activity (50% high/50% low power). For more details on CPU, peripheral, and I/O activity, see the TMS320C62x/C67x Power Consumption Summary application report (literature number SPRA486). For the C6712C/12D device, these currents were measured with average activity (50% high/50% low power) at 25°C case temperature and 100-MHz EMIF. This model represents a device performing high-DSP-activity operations 50% of the time, and the remainder performing low-DSP-activity operations. The high/low-DSP-activity models are defined as follows: High-DSP-Activity Model: CPU: 8 instructions/cycle with 2 LDDW instructions [L1 Data Memory: 128 bits/cycle via LDDW instructions; L1 Program Memory: 256 bits/cycle; L2/EMIF EDMA: 50% writes, 50% reads to/from SDRAM (50% bit-switching)] McBSP: 2 channels at E1 rate Timers: 2 timers at maximum rate Low-DSP-Activity Model: CPU: 2 instructions/cycle with 1 LDH instruction [L1 Data Memory: 16 bits/cycle; L1 Program Memory: 256 bits per 4 cycles; L2/EMIF EDMA: None] McBSP: 2 channels at E1 rate Timers: 2 timers at maximum rate The actual current draw is highly application-dependent. For more details on core and I/O activity, refer to the TMS320C6713/12C/11C Power Consumption Summary application report (literature number SPRA889). POST OFFICE BOX 1443 • HOUSTON, TEXAS 77251−1443 63 SGUS055 − SEPTEMBER 2004 PARAMETER MEASUREMENT INFORMATION IOL Tester Pin Electronics 50 Ω Vcomm Output Under Test CT IOH Where: IOL IOH Vcomm+ CT = = = = 2 mA 2 mA 0.8 V 10−15-pF typical load-circuit capacitance Figure 17. Test Load Circuit for AC Timing Measurements for C6712 Only Tester Pin Electronics 42 Ω 3.5 nH Transmission Line Z0 = 50 Ω (see note) 4.0 pF Data Sheet Timing Reference Point Output Under Test Device Pin (see note) 1.85 pF NOTE: The data sheet provides timing at the device pin. For output timing analysis, the tester pin electronics and its transmission line effects must be taken into account. A transmission line with a delay of 2 ns or longer can be used to produce the desired transmission line effect. The transmission line is intended as a load only. It is not necessary to add or subtract the transmission line delay (2 ns or longer) from the data sheet timings. Input requirements in this data sheet are tested with an input slew rate of < 4 Volts per nanosecond (4 V/ns) at the device pin. Figure 18. Test Load Circuit for AC Timing Measurements for C6712C/C6712D Only 64 POST OFFICE BOX 1443 • HOUSTON, TEXAS 77251−1443 SGUS055 − SEPTEMBER 2004 PARAMETER MEASUREMENT INFORMATION (CONTINUED) signal transition levels All input and output timing parameters are referenced to 1.5 V for both “0” and “1” logic levels. Vref = 1.5 V Figure 19. Input and Output Voltage Reference Levels for ac Timing Measurements All rise and fall transition timing parameters are referenced to VIL MAX and VIH MIN for input clocks, and VOL MAX and VOH MIN for output clocks. Vref = VIH MIN (or VOH MIN) Vref = VIL MAX (or VOL MAX) Figure 20. Rise and Fall Transition Time Voltage Reference Levels timing parameters and board routing analysis The timing parameter values specified in this data sheet do not include delays by board routings. As a good board design practice, such delays must always be taken into account. Timing values may be adjusted by increasing/decreasing such delays. TI recommends utilizing the available I/O buffer information specification (IBIS) models to analyze the timing characteristics correctly. To properly use IBIS models to attain accurate timing analysis for a given system, see the Using IBIS Models for Timing Analysis application report (literature number SPRA839). If needed, external logic hardware such as buffers may be used to compensate any timing differences. For inputs, timing is most impacted by the round-trip propagation delay from the DSP to the external device and from the external device to the DSP. This round-trip delay tends to negatively impact the input setup time margin, but also tends to improve the input hold time margins (see Table 39 and Figure 21). Figure 21 represents a general transfer between the DSP and an external device. The figure also represents board route delays and how they are perceived by the DSP and the external device. POST OFFICE BOX 1443 • HOUSTON, TEXAS 77251−1443 65 SGUS055 − SEPTEMBER 2004 PARAMETER MEASUREMENT INFORMATION (CONTINUED) Table 39. Board-Level Timings Example (see Figure 21) NO. DESCRIPTION 1 Clock route delay 2 Minimum DSP hold time 3 Minimum DSP setup time 4 External device hold time requirement 5 External device setup time requirement 6 Control signal route delay 7 External device hold time 8 External device access time 9 DSP hold time requirement 10 DSP setup time requirement 11 Data route delay ECLKOUT (Output from DSP) 1 ECLKOUT (Input to External Device) Control Signals† (Output from DSP) 2 3 4 5 Control Signals (Input to External Device) 6 7 Data Signals‡ (Output from External Device) 8 10 9 11 Data Signals‡ (Input to DSP) † Control signals include data for Writes. ‡ Data signals are generated during Reads from an external device. Figure 21. Board-Level Input/Output Timings 66 POST OFFICE BOX 1443 • HOUSTON, TEXAS 77251−1443 SGUS055 − SEPTEMBER 2004 INPUT AND OUTPUT CLOCKS timing requirements for CLKIN†‡ (see Figure 22) [C6712] −100 CLKMODE = x4 NO. MIN 1 2 3 CLKMODE = x1 MAX MIN UNIT MAX tc(CLKIN) tw(CLKINH) Cycle time, CLKIN 40 10 ns Pulse duration, CLKIN high 0.4C 0.45C ns tw(CLKINL) tt(CLKIN) Pulse duration, CLKIN low 0.4C 0.45C ns 4 Transition time, CLKIN † The reference points for the rise and fall transitions are measured at VIL MAX and VIH MIN. ‡ C = CLKIN cycle time in ns. For example, when CLKIN frequency is 25 MHz, use C = 40 ns. 5 1 ns timing requirements for CLKIN†‡§ (see Figure 22) [C6712C/C6712D] −167 PLL MODE (PLLEN = 1) NO. 1 2 3 BYPASS MODE (PLLEN = 0) MIN MAX 6 83.3 MIN UNIT MAX tc(CLKIN) tw(CLKINH) Cycle time, CLKIN 6 ns Pulse duration, CLKIN high 0.4C 0.4C ns tw(CLKINL) tt(CLKIN) Pulse duration, CLKIN low 0.4C 0.4C ns 4 Transition time, CLKIN † The reference points for the rise and fall transitions are measured at VIL MAX and VIH MIN. ‡ C = CLKIN cycle time in ns. For example, when CLKIN frequency is 25 MHz, use C = 40 ns. § See the PLL and PLL Controller section of this data sheet. 1 5 5 ns 4 2 CLKIN 3 4 Figure 22. CLKIN Timings POST OFFICE BOX 1443 • HOUSTON, TEXAS 77251−1443 67 SGUS055 − SEPTEMBER 2004 INPUT AND OUTPUT CLOCKS (CONTINUED) switching characteristics over recommended operating conditions for CLKOUT1†‡§ (see Figure 23) [C6712 only] −100 NO. CLKMODE = x4 PARAMETER MIN 1 2 3 4 tc(CKO1) tw(CKO1H) Cycle time, CLKOUT1 tw(CKO1L) tt(CKO1) CLKMODE = x1 MAX MIN MAX P − 0.7 P + 0.7 P − 0.7 P + 0.7 ns Pulse duration, CLKOUT1 high (P/2) − 0.7 (P/2 ) + 0.7 PH − 0.7 PH + 0.7 ns Pulse duration, CLKOUT1 low (P/2) − 0.7 (P/2 ) + 0.7 PL − 0.7 PL + 0.7 ns 0.6 ns Transition time, CLKOUT1 0.6 † The reference points for the rise and fall transitions are measured at VOL MAX and VOH MIN. ‡ PH is the high period of CLKIN in ns and PL is the low period of CLKIN in ns. § P = 1/CPU clock frequency in nanoseconds (ns) 1 4 2 CLKOUT1 3 4 Figure 23. CLKOUT1 Timings [C6712 Only] 68 UNIT POST OFFICE BOX 1443 • HOUSTON, TEXAS 77251−1443 SGUS055 − SEPTEMBER 2004 INPUT AND OUTPUT CLOCKS (CONTINUED) switching characteristics over recommended operating conditions for CLKOUT2†‡ (see Figure 24) [C6712] −100 NO. 1 2 3 4 PARAMETER UNIT MIN MAX 2P − 0.7 2P + 0.7 ns tc(CKO2) tw(CKO2H) Cycle time, CLKOUT2 Pulse duration, CLKOUT2 high P − 0.7 P + 0.7 ns tw(CKO2L) tt(CKO2) Pulse duration, CLKOUT2 low P − 0.7 P + 0.7 ns 0.6 ns Transition time, CLKOUT2 † P = 1/CPU clock frequency in ns ‡ The reference points for the rise and fall transitions are measured at VOL MAX and VOH MIN. switching characteristics over recommended operating conditions for CLKOUT2‡§ (see Figure 24) [C6712C/C6712D] −167 NO. 1 2 3 4 PARAMETER MIN UNIT MAX tc(CKO2) tw(CKO2H) Cycle time, CLKOUT2 C2 − 0.8 C2 + 0.8 ns Pulse duration, CLKOUT2 high (C2/2) − 0.8 (C2/2) + 0.8 ns tw(CKO2L) tt(CKO2) Pulse duration, CLKOUT2 low (C2/2) − 0.8 (C2/2) + 0.8 ns Transition time, CLKOUT2 2 ns ‡ The reference points for the rise and fall transitions are measured at VOL MAX and VOH MIN. § C2 = CLKOUT2 period in ns. CLKOUT2 period is determined by the PLL controller output SYSCLK2 period, which must be set to CPU period divide-by-2. 1 4 2 CLKOUT2 3 4 Figure 24. CLKOUT2 Timings POST OFFICE BOX 1443 • HOUSTON, TEXAS 77251−1443 69 SGUS055 − SEPTEMBER 2004 INPUT AND OUTPUT CLOCKS (CONTINUED) switching characteristics over recommended operating conditions for CLKOUT3†‡ (see Figure 25) [C6712C/C6712D only] 12C−167 NO. 1 PARAMETER MIN 12D−167 MAX MIN UNIT MAX tc(CKO3) tw(CKO3H) Cycle time, CLKOUT3 C3 − 0.6 C3 + 0.6 C3 − 0.9 C3 + 0.9 ns Pulse duration, CLKOUT3 high (C3/2) − 0.6 (C3/2) + 0.6 (C3/2) − 0.9 (C3/2) + 0.9 ns tw(CKO3L) tt(CKO3) Pulse duration, CLKOUT3 low (C3/2) − 0.6 (C3/2) + 0.6 (C3/2) − 0.9 (C3/2) + 0.9 ns 4 3 ns 5 td(CLKINH-CKO3V) 7.5 ns 2 3 Transition time, CLKOUT3 2 Delay time, CLKIN high to CLKOUT3 valid 1.5 6.5 1.5 † The reference points for the rise and fall transitions are measured at VOL MAX and VOH MIN. ‡ C3 = CLKOUT3 period in ns. CLKOUT3 period is a divide-down of the CPU clock, configurable via the RATIO field in the PLLDIV3 register. CLKIN 5 1 5 4 3 CLKOUT3 2 4 NOTE A: For this example, the CLKOUT3 frequency is CLKIN divide-by-2. Figure 25. CLKOUT3 Timings [C6712C/C6712D Only] timing requirements for ECLKIN§ (see Figure 26) −100 NO. 1 2 3 4 MIN MAX MIN MAX UNIT tc(EKI) tw(EKIH) Cycle time, ECLKIN 15 10 ns Pulse duration, ECLKIN high 6.8 4.5 ns tw(EKIL) tt(EKI) Pulse duration, ECLKIN low 6.8 4.5 Transition time, ECLKIN 3 § The reference points for the rise and fall transitions are measured at VIL MAX and VIH MIN. 1 4 2 ECLKIN 3 4 Figure 26. ECLKIN Timings 70 −167 POST OFFICE BOX 1443 • HOUSTON, TEXAS 77251−1443 ns 3 ns SGUS055 − SEPTEMBER 2004 INPUT AND OUTPUT CLOCKS (CONTINUED) switching characteristics over recommended operating conditions for ECLKOUT†‡§ (see Figure 27) −100 NO. 1 2 3 4 5 PARAMETER MIN −167 MAX MIN MAX UNIT tc(EKO) tw(EKOH) Cycle time, ECLKOUT E − 0.7 E + 0.7 E − 0.9 E + 0.9 ns Pulse duration, ECLKOUT high EH − 0.7 EH + 0.7 EH − 0.9 EH + 0.9 ns tw(EKOL) tt(EKO) Pulse duration, ECLKOUT low EL − 0.7 EL + 0.7 EL − 0.9 EL + 0.9 ns 2 ns td(EKIH-EKOH) td(EKIL-EKOL) Delay time, ECLKIN high to ECLKOUT high Transition time, ECLKOUT 0.6 1 7 1 6.5 ns 6 Delay time, ECLKIN low to ECLKOUT low 1 † The reference points for the rise and fall transitions are measured at VOL MAX and VOH MIN. ‡ E = ECLKIN period in ns § EH is the high period of ECLKIN in ns and EL is the low period of ECLKIN in ns. 7 1 6.5 ns ECLKIN 6 1 2 5 3 4 4 ECLKOUT Figure 27. ECLKOUT Timings POST OFFICE BOX 1443 • HOUSTON, TEXAS 77251−1443 71 SGUS055 − SEPTEMBER 2004 ASYNCHRONOUS MEMORY TIMING timing requirements for asynchronous memory cycles†‡ (see Figure 28−Figure 29) −100 NO. 3 4 6 7 MIN tsu(EDV-AREH) th(AREH-EDV) Setup time, EDx valid before ARE high tsu(ARDY-EKOH) th(EKOH-ARDY) −167 MAX MIN UNIT MAX 13 6.5 ns Hold time, EDx valid after ARE high 1 1 ns Setup time, ARDY valid before ECLKOUT high 6 3 ns 1.7 2.3 Hold time, ARDY valid after ECLKOUT high ns † To ensure data setup time, simply program the strobe width wide enough. ARDY is internally synchronized. The ARDY signal is recognized in the cycle for which the setup and hold time is met. To use ARDY as an asynchronous input, the pulse width of the ARDY signal should be wide enough (e.g., pulse width = 2E) to ensure setup and hold time is met. ‡ RS = Read setup, RST = Read strobe, RH = Read hold, WS = Write setup, WST = Write strobe, WH = Write hold. These parameters are programmed via the EMIF CE space control registers. switching characteristics over recommended operating conditions for asynchronous memory cycles‡§¶ (see Figure 28−Figure 29) −100 NO. PARAMETER MIN −167 MAX MIN MAX UNIT 1 tosu(SELV-AREL) Output setup time, select signals valid to ARE low RS * E − 3 RS*E − 1.7 ns 2 toh(AREH-SELIV) Output hold time, ARE high to select signals invalid RH * E − 3 RH*E − 1.7 ns 5 td(EKOH-AREV) Delay time, ECLKOUT high to ARE valid 8 tosu(SELV-AWEL) Output setup time, select signals valid to AWE low 9 toh(AWEH-SELIV) Output hold time, AWE high to select signals and EDx invalid 10 td(EKOH-AWEV) Delay time, ECLKOUT high to AWE valid 11 tosu(EDV-AWEL) Output setup time, ED valid to AWE low 1.5 11 WS * E − 3 WH * E − 3 1.5 (WS−1)*E − 1.7 11 1.5 7 ns WS*E − 1.7 ns WH*E − 1.7 ns 1.5 (WS−1)*E − 1.7 7 ns ns ‡ RS = Read setup, RST = Read strobe, RH = Read hold, WS = Write setup, WST = Write strobe, WH = Write hold. These parameters are programmed via the EMIF CE space control registers. § E = ECLKOUT period in ns ¶ Select signals include: CE[3:0], BE[1:0], EA[21:2], and AOE. 72 POST OFFICE BOX 1443 • HOUSTON, TEXAS 77251−1443 SGUS055 − SEPTEMBER 2004 ASYNCHRONOUS MEMORY TIMING (CONTINUED) Setup = 2 Strobe = 3 Not Ready Hold = 2 ECLKOUT 1 2 CE[3:0] 1 2 BE[1:0] BE 1 2 EA[21:2] Address 3 4 ED[15:0] 1 2 Read Data AOE/SDRAS/SSOE† 5 5 ARE/SDCAS/SSADS† AWE/SDWE/SSWE† 7 6 7 6 ARDY † AOE/SDRAS/SSOE, ARE/SDCAS/SSADS, and AWE/SDWE/SSWE operate as AOE (identified under select signals), ARE, and AWE, respectively, during asynchronous memory accesses. Figure 28. Asynchronous Memory Read Timing POST OFFICE BOX 1443 • HOUSTON, TEXAS 77251−1443 73 SGUS055 − SEPTEMBER 2004 ASYNCHRONOUS MEMORY TIMING (CONTINUED) Setup = 2 Strobe = 3 Hold = 2 Not Ready ECLKOUT 8 9 CEx 8 9 BE[3:0] BE 8 9 EA[21:2] Address 11 9 ED[31:0] Write Data AOE/SDRAS/SSOE† ARE/SDCAS/SSADS† 10 10 AWE/SDWE/SSWE† 7 6 7 6 ARDY † AOE/SDRAS/SSOE, ARE/SDCAS/SSADS, and AWE/SDWE/SSWE operate as AOE (identified under select signals), ARE, and AWE, respectively, during asynchronous memory accesses. Figure 29. Asynchronous Memory Write Timing 74 POST OFFICE BOX 1443 • HOUSTON, TEXAS 77251−1443 SGUS055 − SEPTEMBER 2004 SYNCHRONOUS-BURST MEMORY TIMING timing requirements for synchronous-burst SRAM cycles† (see Figure 30) −100 NO. 6 7 MIN tsu(EDV-EKOH) th(EKOH-EDV) Setup time, read EDx valid before ECLKOUT high Hold time, read EDx valid after ECLKOUT high −167 MAX 6 2.1‡ MIN MAX 1.5 UNIT ns 2.5 ns † The C6712/12C/12D SBSRAM interface takes advantage of the internal burst counter in the SBSRAM. Accesses default to incrementing 4-word bursts, but random bursts and decrementing bursts are done by interrupting bursts in progress. All burst types can sustain continuous data flow. ‡ Make sure the external SBSRAM meets the timing specifications of the C6712 device. Delays or buffers may be needed to compensate for any timing differences. IBIS analysis should be used to correctly model the system interface. switching characteristics over recommended operating conditions for synchronous-burst SRAM cycles†§ (see Figure 30 and Figure 31) −100 NO. 1 2 3 4 5 8 9 10 11 12 PARAMETER MIN td(EKOH-CEV) td(EKOH-BEV) Delay time, ECLKOUT high to CEx valid 1.5 td(EKOH-BEIV) td(EKOH-EAV) Delay time, ECLKOUT high to BEx invalid td(EKOH-EAIV) td(EKOH-ADSV) Delay time, ECLKOUT high to EAx invalid Delay time, ECLKOUT high to ARE/SDCAS/SSADS valid 1.5 td(EKOH-OEV) td(EKOH-EDV) Delay time, ECLKOUT high to, AOE/SDRAS/SSOE valid 1.5 td(EKOH-EDIV) td(EKOH-WEV) Delay time, ECLKOUT high to EDx invalid 1.5 Delay time, ECLKOUT high to AWE/SDWE/SSWE valid 1.5 −167 MAX 11‡ MAX 1.2 7 ns 7 ns 11‡ Delay time, ECLKOUT high to BEx valid 1.5 1.2 11‡ Delay time, ECLKOUT high to EAx valid 1.5 7 ns 7 ns 1.2 7 ns 7 ns 1.2 11‡ ns 1.2 11‡ Delay time, ECLKOUT high to EDx valid ns 1.2 11‡ 11‡ UNIT MIN ns ns † The C6712/12C/12D SBSRAM interface takes advantage of the internal burst counter in the SBSRAM. Accesses default to incrementing 4-word bursts, but random bursts and decrementing bursts are done by interrupting bursts in progress. All burst types can sustain continuous data flow. ‡ Make sure the external SBSRAM meets the timing specifications of the C6712 device. Delays or buffers may be needed to compensate for any timing differences. IBIS analysis should be used to correctly model the system interface. § ARE/SDCAS/SSADS, AOE/SDRAS/SSOE, and AWE/SDWE/SSWE operate as SSADS, SSOE, and SSWE, respectively, during SBSRAM accesses. POST OFFICE BOX 1443 • HOUSTON, TEXAS 77251−1443 1.2 7 75 SGUS055 − SEPTEMBER 2004 SYNCHRONOUS-BURST MEMORY TIMING (CONTINUED) ECLKOUT 1 1 CE[3:0] BE[1:0] 2 BE1 3 BE2 BE3 4 BE4 5 EA[21:2] EA 6 ED[15:0] 7 Q1 Q2 Q3 Q4 8 8 ARE/SDCAS/SSADS† 9 9 AOE/SDRAS/SSOE† AWE/SDWE/SSWE† † ARE/SDCAS/SSADS, AOE/SDRAS/SSOE, and AWE/SDWE/SSWE operate as SSADS, SSOE, and SSWE, respectively, during SBSRAM accesses. Figure 30. SBSRAM Read Timing ECLKOUT 1 1 CE[3:0] BE[1:0] 2 BE1 3 BE2 BE3 5 4 EA[21:2] ED[15:0] BE4 EA 10 Q1 8 11 Q2 Q3 Q4 8 ARE/SDCAS/SSADS† AOE/SDRAS/SSOE† 12 12 AWE/SDWE/SSWE† † ARE/SDCAS/SSADS, AOE/SDRAS/SSOE, and AWE/SDWE/SSWE operate as SSADS, SSOE, and SSWE, respectively, during SBSRAM accesses. Figure 31. SBSRAM Write Timing 76 POST OFFICE BOX 1443 • HOUSTON, TEXAS 77251−1443 SGUS055 − SEPTEMBER 2004 SYNCHRONOUS DRAM TIMING timing requirements for synchronous DRAM cycles† (see Figure 32) −100 NO. 6 MIN tsu(EDV-EKOH) th(EKOH-EDV) Setup time, read EDx valid before ECLKOUT high −167 MAX 6 MIN MAX 1.5 UNIT ns 7 Hold time, read EDx valid after ECLKOUT high 2.1 2.5 ns † The C6712/12C/12D SDRAM interface takes advantage of the internal burst counter in the SDRAM. Accesses default to incrementing 4-word bursts, but random bursts and decrementing bursts are done by interrupting bursts in progress. All burst types can sustain continuous data flow. switching characteristics over recommended operating conditions for synchronous DRAM cycles†‡ (see Figure 32−Figure 38) −100 NO. 1 2 3 4 5 8 9 10 11 PARAMETER −167 UNIT MIN MAX MIN MAX 1.5 11 1.5 7 ns 7 ns td(EKOH-CEV) td(EKOH-BEV) Delay time, ECLKOUT high to CEx valid td(EKOH-BEIV) td(EKOH-EAV) Delay time, ECLKOUT high to BEx invalid td(EKOH-EAIV) td(EKOH-CASV) Delay time, ECLKOUT high to EAx invalid 1.5 Delay time, ECLKOUT high to ARE/SDCAS/SSADS valid 1.5 td(EKOH-EDV) td(EKOH-EDIV) Delay time, ECLKOUT high to EDx valid Delay time, ECLKOUT high to EDx invalid 1.5 td(EKOH-WEV) td(EKOH-RAS) Delay time, ECLKOUT high to AWE/SDWE/SSWE valid 1.5 Delay time, ECLKOUT high to BEx valid 11 1.5 Delay time, ECLKOUT high to EAx valid 1.5 11 ns 7 1.5 11 1.5 11 ns 7 ns 7 ns 1.5 11 1.5 ns ns 7 ns 12 Delay time, ECLKOUT high to, AOE/SDRAS/SSOE valid 1.5 11 1.5 7 ns † The C6712/12C/12D SDRAM interface takes advantage of the internal burst counter in the SDRAM. Accesses default to incrementing 4-word bursts, but random bursts and decrementing bursts are done by interrupting bursts in progress. All burst types can sustain continuous data flow. ‡ ARE/SDCAS/SSADS, AWE/SDWE/SSWE, and AOE/SDRAS/SSOE operate as SDCAS, SDWE, and SDRAS, respectively, during SDRAM accesses. POST OFFICE BOX 1443 • HOUSTON, TEXAS 77251−1443 77 SGUS055 − SEPTEMBER 2004 SYNCHRONOUS DRAM TIMING (CONTINUED) READ ECLKOUT 1 1 CE[3:0] 2 BE1 BE[1:0] EA[21:13] EA[11:2] 4 Bank 5 4 Column 5 4 3 BE2 BE3 BE4 5 EA12 6 D1 ED[15:0] 7 D2 D3 D4 AOE/SDRAS/SSOE† 8 8 ARE/SDCAS/SSADS† AWE/SDWE/SSWE† † ARE/SDCAS/SSADS, AWE/SDWE/SSWE, and AOE/SDRAS/SSOE operate as SDCAS, SDWE, and SDRAS, respectively, during SDRAM accesses. Figure 32. SDRAM Read Command (CAS Latency 3) 78 POST OFFICE BOX 1443 • HOUSTON, TEXAS 77251−1443 SGUS055 − SEPTEMBER 2004 SYNCHRONOUS DRAM TIMING (CONTINUED) WRITE ECLKOUT 1 2 CE[3:0] 2 3 4 BE[1:0] BE1 4 BE2 BE3 BE4 D2 D3 D4 5 Bank EA[21:13] 5 4 Column EA[11:2] 4 5 EA12 9 ED[15:0] 10 9 D1 AOE/SDRAS/SSOE† 8 8 11 11 ARE/SDCAS/SSADS† AWE/SDWE/SSWE† † ARE/SDCAS/SSADS, AWE/SDWE/SSWE, and AOE/SDRAS/SSOE operate as SDCAS, SDWE, and SDRAS, respectively, during SDRAM accesses. Figure 33. SDRAM Write Command POST OFFICE BOX 1443 • HOUSTON, TEXAS 77251−1443 79 SGUS055 − SEPTEMBER 2004 SYNCHRONOUS DRAM TIMING (CONTINUED) ACTV ECLKOUT 1 1 CE[3:0] BE[1:0] 4 Bank Activate 5 EA[21:13] 4 Row Address 5 EA[11:2] 4 Row Address 5 EA12 ED[15:0] 12 12 AOE/SDRAS/SSOE† ARE/SDCAS/SSADS† AWE/SDWE/SSWE† † ARE/SDCAS/SSADS, AWE/SDWE/SSWE, and AOE/SDRAS/SSOE operate as SDCAS, SDWE, and SDRAS, respectively, during SDRAM accesses. Figure 34. SDRAM ACTV Command DCAB ECLKOUT 1 1 4 5 12 12 11 11 CE[3:0] BE[1:0] EA[21:13, 11:2] EA12 ED[15:0] AOE/SDRAS/SSOE† ARE/SDCAS/SSADS† AWE/SDWE/SSWE† † ARE/SDCAS/SSADS, AWE/SDWE/SSWE, and AOE/SDRAS/SSOE operate as SDCAS, SDWE, and SDRAS, respectively, during SDRAM accesses. Figure 35. SDRAM DCAB Command 80 POST OFFICE BOX 1443 • HOUSTON, TEXAS 77251−1443 SGUS055 − SEPTEMBER 2004 SYNCHRONOUS DRAM TIMING (CONTINUED) DEAC ECLKOUT 1 1 CE[3:0] BE[1:0] 4 5 Bank EA[21:13] EA[11:2] 4 5 12 12 11 11 EA12 ED[15:0] AOE/SDRAS/SSOE† ARE/SDCAS/SSADS† AWE/SDWE/SSWE† † ARE/SDCAS/SSADS, AWE/SDWE/SSWE, and AOE/SDRAS/SSOE operate as SDCAS, SDWE, and SDRAS, respectively, during SDRAM accesses. Figure 36. SDRAM DEAC Command REFR ECLKOUT 1 1 12 12 8 8 CE[3:0] BE[1:0] EA[21:2] EA12 ED[15:0] AOE/SDRAS/SSOE† ARE/SDCAS/SSADS† AWE/SDWE/SSWE† † ARE/SDCAS/SSADS, AWE/SDWE/SSWE, and AOE/SDRAS/SSOE operate as SDCAS, SDWE, and SDRAS, respectively, during SDRAM accesses. Figure 37. SDRAM REFR Command POST OFFICE BOX 1443 • HOUSTON, TEXAS 77251−1443 81 SGUS055 − SEPTEMBER 2004 SYNCHRONOUS DRAM TIMING (CONTINUED) MRS ECLKOUT 1 1 4 MRS value 5 12 12 8 8 11 11 CE[3:0] BE[1:0] EA[21:2] ED[15:0] AOE/SDRAS/SSOE† ARE/SDCAS/SSADS† AWE/SDWE/SSWE† † ARE/SDCAS/SSADS, AWE/SDWE/SSWE, and AOE/SDRAS/SSOE operate as SDCAS, SDWE, and SDRAS, respectively, during SDRAM accesses. Figure 38. SDRAM MRS Command 82 POST OFFICE BOX 1443 • HOUSTON, TEXAS 77251−1443 SGUS055 − SEPTEMBER 2004 HOLD/HOLDA TIMING timing requirements for the HOLD/HOLDA cycles† (see Figure 39) −100 −167 NO. MIN 3 th(HOLDAL-HOLDL) † E = ECLKIN period in ns Hold time, HOLD low after HOLDA low UNIT MAX E ns switching characteristics over recommended operating conditions for the HOLD/HOLDA cycles†‡ (see Figure 39) NO. −100 −167 PARAMETER MIN 1 2 4 td(HOLDL-EMHZ) td(EMHZ-HOLDAL) Delay time, HOLD low to EMIF Bus high impedance td(HOLDH-EMLZ) td(EMLZ-HOLDAH) Delay time, HOLD high to EMIF Bus low impedance Delay time, EMIF Bus high impedance to HOLDA low 12D−167 MAX MIN UNIT 2E § 2E MAX § −0.1 2E 0 2E ns 2E 7E 2E 7E ns ns 5 Delay time, EMIF Bus low impedance to HOLDA high −1.5 2E 0 2E ns † E = ECLKIN period in ns ‡ EMIF Bus consists of CE[3:0], BE[1:0], ED[15:0], EA[21:2], ARE/SDCAS/SSADS, AOE/SDRAS/SSOE, and AWE/SDWE/SSWE. § All pending EMIF transactions are allowed to complete before HOLDA is asserted. If no bus transactions are occurring, then the minimum delay time can be achieved. Also, bus hold can be indefinitely delayed by setting NOHOLD = 1. External Requestor Owns Bus DSP Owns Bus DSP Owns Bus 3 HOLD 2 5 HOLDA EMIF Bus† 1 C6712/C6712C/C6712D 4 C6712/C6712C/C6712D † EMIF Bus consists of CE[3:0], BE[1:0], ED[15:0], EA[21:2], ARE/SDCAS/SSADS, AOE/SDRAS/SSOE, and AWE/SDWE/SSWE. Figure 39. HOLD/HOLDA Timing POST OFFICE BOX 1443 • HOUSTON, TEXAS 77251−1443 83 SGUS055 − SEPTEMBER 2004 BUSREQ TIMING switching characteristics over recommended operating conditions for the BUSREQ cycles (see Figure 40) −100 NO. 1 PARAMETER td(EKOH-BUSRV) Delay time, ECLKOUT high to BUSREQ valid MAX MIN MAX 2 10 1.5 7.2 ECLKOUT 1 1 BUSREQ Figure 40. BUSREQ Timing 84 POST OFFICE BOX 1443 • HOUSTON, TEXAS 77251−1443 −167 MIN UNIT ns SGUS055 − SEPTEMBER 2004 RESET TIMING [C6712] timing requirements for reset† (see Figure 41) −100 NO. MIN MAX UNIT Width of the RESET pulse (PLL stable)‡ 10P ns 1 tw(RST) Width of the RESET pulse (PLL needs to sync up)§ 250 µs 14 tsu(BOOT) th(BOOT) Setup time, BOOTMODE[1:0] configuration bits valid before RESET high 2P ns Hold time, BOOTMODE[1:0] configuration bits valid after RESET high 2P 15 ns † P = 1/CPU clock frequency in ns. For example, when running parts at 100 MHz, use P = 10 ns. ‡ This parameter applies to CLKMODE x1 when CLKIN is stable, and applies to CLKMODE x4 when CLKIN and PLL are stable. § This parameter applies to CLKMODE x4 only (it does not apply to CLKMODE x1). The RESET signal is not connected internally to the clock PLL circuit. The PLL, however, may need up to 250 µs to stabilize following device power up or after PLL configuration has been changed. During that time, RESET must be asserted to ensure proper device operation. See the clock PLL section for PLL lock times. switching characteristics over recommended operating conditions during reset†¶# (see Figure 41) −100 NO. 2 3 4 5 6 7 8 9 12 13 PARAMETER MIN MAX UNIT td(RSTL-ECKI) td(RSTH-ECKI) Delay time, RESET low to ECLKIN synchronized internally 2P + 3E 3P + 4E ns Delay time, RESET high to ECLKIN synchronized internally 2P + 3E 3P + 4E ns td(RSTL-EMIFZHZ) td(RSTH-EMIFZV) Delay time, RESET low to EMIF Z group high impedance 2P + 3E td(RSTL-EMIFHIV) td(RSTH-EMIFHV) Delay time, RESET low to EMIF high group invalid td(RSTL-EMIFLIV) td(RSTH-EMIFLV) Delay time, RESET low to EMIF low group invalid td(RSTL-ZHZ) td(RSTH-ZV) Delay time, RESET low to Z group high impedance 2P ns Delay time, RESET high to Z group valid 2P ns Delay time, RESET high to EMIF Z group valid ns 3P + 4E 2P + 3E Delay time, RESET high to EMIF high group valid ns 3P + 4E 2P + 3E Delay time, RESET high to EMIF low group valid ns ns ns 3P + 4E ns † P = 1/CPU clock frequency in ns. For example, when running parts at 100 MHz, use P = 10 ns. ¶ E = ECLKIN period in ns # EMIF Z group consists of: EA[21:2], ED[15:0], CE[3:0], BE[1:0], ARE/SDCAS/SSADS, AWE/SDWE/SSWE, and AOE/SDRAS/SSOE EMIF high group consists of: HOLDA EMIF low group consists of: BUSREQ Z group consists of: CLKX0, CLKX1, FSX0, FSX1, DX0, DX1, CLKR0, CLKR1, FSR0, FSR1, TOUT0, and TOUT1. POST OFFICE BOX 1443 • HOUSTON, TEXAS 77251−1443 85 SGUS055 − SEPTEMBER 2004 RESET TIMING [C6712] (CONTINUED) CLKOUT1 CLKOUT2 1 14 15 RESET 2 3 4 5 6 7 8 9 ECLKIN† EMIF Z Group‡ EMIF High Group‡ EMIF Low Group‡ Z Group‡ 12 13 BOOTMODE[1:0] † ECLKIN should be provided during reset in order to drive EMIF signals to the correct reset values. ECLKOUT continues to clock as long as ECLKIN is provided. ‡ EMIF Z group consists of: EA[21:2], ED[15:0], CE[3:0], BE[1:0], ARE/SDCAS/SSADS, AWE/SDWE/SSWE, and AOE/SDRAS/SSOE EMIF high group consists of: HOLDA EMIF low group consists of: BUSREQ Z group consists of: CLKX0, CLKX1, FSX0, FSX1, DX0, DX1, CLKR0, CLKR1, FSR0, FSR1, TOUT0, and TOUT1. Figure 41. Reset Timing [C6712] 86 POST OFFICE BOX 1443 • HOUSTON, TEXAS 77251−1443 SGUS055 − SEPTEMBER 2004 RESET TIMING [C6712C/C6712D] timing requirements for reset†‡ (see Figure 42) −167 NO. 1 12 MIN tw(RST) tsu(BOOT) 13 Pulse duration, RESET Setup time, boot configuration bits valid before RESET high§ Hold time, boot configuration bits valid after RESET high§ MAX UNIT 100 ns 2P ns th(BOOT) 2P ns † P = 1/CPU clock frequency in ns. For example, when running parts at 167 MHz, use P = 6 ns. ‡ For the C6712C/12D device, the PLL is bypassed immediately after the device comes out of reset. The PLL Controller can be programmed to change the PLL mode in software. For more detailed information on the PLL Controller, see the TMS320C6000 DSP Software-Programmable Phase-Lock Loop (PLL) Controller Reference Guide (literature number SPRU233). § The Boot and device configurations bits are latched asynchronously when RESET is transitioning high. The Boot and device configurations bits consist of: BOOTMODE[1:0] and LENDIAN. switching characteristics over recommended operating conditions during reset¶ (see Figure 42) −167 NO. PARAMETER MIN Delay time, external RESET high to internal reset high and all signal groups valid#|| MAX 512 x CLKIN period 2 td(RSTH-ZV) 3a td(RSTL-ECKOL) td(RSTL-ECKOL) Delay time, RESET low to ECLKOUT low (6712C) 0 ns Delay time, RESET low to ECLKOUT high impedance (6712D) 0 ns td(RSTH-ECKOV) td(RSTL-CKO2IV) Delay time, RESET high to ECLKOUT valid Delay time, RESET low to CLKOUT2 invalid (6712C) 0 td(RSTL-CKO2IV) td(RSTH-CKO2V) Delay time, RESET low to CLKOUT2 high impedance (6712D) 0 td(RSTL-CKO3L) td(RSTH-CKO3V) Delay time, RESET low to CLKOUT3 low td(RSTL-EMIFZHZ) td(RSTL-EMIFLIV) Delay time, RESET low to EMIF Z group high impedance|| 0 ns Delay time, RESET low to EMIF low group (BUSREQ) invalid|| Delay time, RESET low to Z group high impedance|| 0 ns 3b 4 5a 5b 6 7 8 9 10 11 CLKMODE0 = 1 UNIT 6P Delay time, RESET high to CLKOUT2 valid ns ns ns 6P 0 Delay time, RESET high to CLKOUT3 valid ns ns ns 6P ns td(RSTL-Z1HZ) 0 ns ¶ P = 1/CPU clock frequency in ns. Note that while internal reset is asserted low, the CPU clock (SYSCLK1) period is equal to the input clock (CLKIN) period multiplied by 8. For example, if the CLKIN period is 20 ns, then the CPU clock (SYSCLK1) period is 20 ns x 8 = 160 ns. Therefore, P = SYSCLK1 = 160 ns while internal reset is asserted. # The internal reset is stretched exactly 512 x CLKIN cycles if CLKIN is used (CLKMODE0 = 1). If the input clock (CLKIN) is not stable when RESET is deasserted, the actual delay time may vary. || EMIF Z group consists of: EA[21:2], ED[15:0], CE[3:0], BE[1:0], ARE/SDCAS/SSADS, AWE/SDWE/SSWE, AOE/SDRAS/SSOE and HOLDA EMIF low group consists of: BUSREQ Z group consists of: CLKR0, CLKR1, CLKX0, CLKX1, FSR0, FSR1, FSX0, FSX1, DX0, DX1, TOUT0, and TOUT1. POST OFFICE BOX 1443 • HOUSTON, TEXAS 77251−1443 87 SGUS055 − SEPTEMBER 2004 RESET TIMING [C6712C/C6712D] (CONTINUED) Phase 1 Phase 2 Phase 3 CLKIN ECLKIN 1 RESET 2 Internal Reset Internal SYSCLK1 Internal SYSCLK2 Internal SYSCLK3 6712CECLKOUT§ 3 4 5 6 7 8 6712DECLKOUT§ 6712C CLKOUT2§ 6712D CLKOUT2§ CLKOUT3 9 2 10 2 11 2 EMIF Z Group† EMIF Low Group† Z Group† Boot and Device Configuration Pins‡ 13 14 † EMIF Z group consists of: EA[21:2], ED[15:0], CE[3:0], BE[1:0], ARE/SDCAS/SSADS, AWE/SDWE/SSWE, AOE/SDRAS/SSOE and HOLDA EMIF low group consists of: BUSREQ Z group consists of: CLKR0, CLKR1, CLKX0, CLKX1, FSR0, FSR1, FSX0, FSX1, DX0, DX1, TOUT0, and TOUT1. ‡ Boot and device configurations consist of: BOOTMODE[1:0] and LENDIAN. Figure 42. Reset Timing [C6712C/12D] Reset Phase 1: The RESET pin is asserted. During this time, all internal clocks are running at the CLKIN frequency divide-by-8. The CPU is also running at the CLKIN frequency divide-by-8. Reset Phase 2: The RESET pin is deasserted but the internal reset is stretched. During this time, all internal clocks are running at the CLKIN frequency divide-by-8. The CPU is also running at the CLKIN frequency divide-by-8. Reset Phase 3: Both the RESET pin and internal reset are deasserted. During this time, all internal clocks are running at their default divide-down frequency of CLKIN. The CPU clock (SYSCLK1) is running at CLKIN frequency. The peripheral clock (SYSCLK2) is running at CLKIN frequency divide-by-2. The EMIF internal clock source (SYSCLK3) is running at CLKIN frequency divide-by-2. SYSCLK3 is reflected on the ECLKOUT pin (when EKSRC bit = 0 [default]). CLKOUT3 is running at CLKIN frequency divide-by-8. 88 POST OFFICE BOX 1443 • HOUSTON, TEXAS 77251−1443 SGUS055 − SEPTEMBER 2004 EXTERNAL INTERRUPT TIMING timing requirements for external interrupts† (see Figure 43) −100 NO. MIN 1 tw(ILOW) 2 tw(IHIGH) MAX −167 MIN MAX UNIT Width of the NMI interrupt pulse low 2P 2P ns Width of the EXT_INT interrupt pulse low 2P 4P ns Width of the NMI interrupt pulse high 2P 2P ns Width of the EXT_INT interrupt pulse high 2P 4P ns † P = 1/CPU clock frequency in ns. For example, when running parts at 100 MHz, use P = 10 ns. 1 2 EXT_INT, NMI Figure 43. External/NMI Interrupt Timing POST OFFICE BOX 1443 • HOUSTON, TEXAS 77251−1443 89 SGUS055 − SEPTEMBER 2004 MULTICHANNEL BUFFERED SERIAL PORT TIMING timing requirements for McBSP†‡ (see Figure 44) [C6712] −100 NO. 2 3 tc(CKRX) tw(CKRX) Cycle time, CLKR/X CLKR/X ext Pulse duration, CLKR/X high or CLKR/X low CLKR/X ext CLKR int 5 tsu(FRH-CKRL) Setup time, external FSR high before CLKR low 6 th(CKRL-FRH) Hold time, external FSR high after CLKR low 7 tsu(DRV-CKRL) Setup time, DR valid before CLKR low 8 th(CKRL-DRV) Hold time, DR valid after CLKR low 10 tsu(FXH-CKXL) Setup time, external FSX high before CLKX low 11 th(CKXL-FXH) Hold time, external FSX high after CLKX low MIN 2P§ 0.5tc(CKRX) − 1 20 CLKR ext 1 CLKR int 6 CLKR ext 3 CLKR int 22 CLKR ext 3 CLKR int 3 CLKR ext 4 CLKX int 23 CLKX ext 1 CLKX int 6 CLKX ext 3 MAX UNIT ns ns ns ns ns ns ns ns † CLKRP = CLKXP = FSRP = FSXP = 0. If polarity of any of the signals is inverted, then the timing references of that signal are also inverted. ‡ P = 1/CPU clock frequency in ns. For example, when running parts at 100 MHz, use P = 10 ns. § The minimum CLKR/X period is twice the CPU cycle time (2P). This means that the maximum bit rate for communications between the McBSP and other device is 50 Mbps for 100 MHz CPU clock; where the McBSP is either the master or the slave. Care must be taken to ensure that the AC timings specified in this data sheet are met. The maximum bit rate for McBSP-to-McBSP communications is 33 Mbps; therefore, the minimum CLKR/X clock cycle is either twice the CPU cycle time (2P), or 30 ns (33 MHz), whichever value is larger. For example, when running parts at 100 MHz (P = 10 ns), use 30 ns as the minimum CLKR/X clock cycle (by setting the appropriate CLKGDV ratio or external clock source). The maximum bit rate for McBSP-to-McBSP communications applies when the serial port is a master of the clock and frame syncs (with CLKR connected to CLKX, FSR connected to FSX, CLKXM = FSXM = 1, and CLKRM = FSRM = 0) in data delay 1 or 2 mode (R/XDATDLY = 01b or 10b) and the other device the McBSP communicates to is a slave. 90 POST OFFICE BOX 1443 • HOUSTON, TEXAS 77251−1443 SGUS055 − SEPTEMBER 2004 MULTICHANNEL BUFFERED SERIAL PORT TIMING (CONTINUED) timing requirements for McBSP†‡ (see Figure 44) [C6712C/C6712D] −167 NO. 2 3 tc(CKRX) tw(CKRX) Cycle time, CLKR/X Pulse duration, CLKR/X high or CLKR/X low 5 tsu(FRH-CKRL) Setup time, external FSR high before CLKR low 6 th(CKRL-FRH) Hold time, external FSR high after CLKR low 7 tsu(DRV-CKRL) Setup time, DR valid before CLKR low 8 th(CKRL-DRV) Hold time, DR valid after CLKR low 10 tsu(FXH-CKXL) Setup time, external FSX high before CLKX low 11 th(CKXL-FXH) Hold time, external FSX high after CLKX low CLKR/X ext MIN 2P§ CLKR/X ext 0.5 *tc(CKRX) −1¶ CLKR int 9 CLKR ext 1 CLKR int 6 CLKR ext 3 CLKR int 8 CLKR ext 0 CLKR int 3 CLKR ext 4 CLKX int 9 CLKX ext 1 CLKX int 6 CLKX ext 3 MAX UNIT ns ns ns ns ns ns ns ns † CLKRP = CLKXP = FSRP = FSXP = 0. If polarity of any of the signals is inverted, then the timing references of that signal are also inverted. ‡ P = 1/CPU clock frequency in ns. For example, when running parts at 167 MHz, use P = 6 ns. § The minimum CLKR/X period is twice the CPU cycle time (2P) and not faster than 75 Mbps (13.3 ns). This means that the maximum bit rate for communications between the McBSP and other device is 75 Mbps for 150 MHz CPU clock; where the McBSP is either the master or the slave. Care must be taken to ensure that the AC timings specified in this data sheet are met. The maximum bit rate for McBSP-to-McBSP communications is 67 Mbps; therefore, the minimum CLKR/X clock cycle is either twice the CPU cycle time (2P), or 15 ns (67 MHz), whichever value is larger. For example, when running parts at 150 MHz (P = 6.7 ns), use 15 ns as the minimum CLKR/X clock cycle (by setting the appropriate CLKGDV ratio or external clock source). When running parts at 60 MHz (P = 16.67 ns), use 2P = 33 ns (30 MHz) as the minimum CLKR/X clock cycle. The maximum bit rate for McBSP-to-McBSP communications applies when the serial port is a master of the clock and frame syncs (with CLKR connected to CLKX, FSR connected to FSX, CLKXM = FSXM = 1, and CLKRM = FSRM = 0) in data delay 1 or 2 mode (R/XDATDLY = 01b or 10b) and the other device the McBSP communicates to is a slave. ¶ This parameter applies to the maximum McBSP frequency. Operate serial clocks (CLKR/X) in the reasonable range of 40/60 duty cycle. POST OFFICE BOX 1443 • HOUSTON, TEXAS 77251−1443 91 SGUS055 − SEPTEMBER 2004 MULTICHANNEL BUFFERED SERIAL PORT TIMING (CONTINUED) switching characteristics over recommended operating conditions for McBSP†‡ (see Figure 44) [C6712] −100 NO. PARAMETER Delay time, CLKS high to CLKR/X high for internal CLKR/X generated from CLKS input MAX 4 26 2P§¶ C − 1# C + 1# ns ns 1 td(CKSH-CKRXH) 2 Cycle time, CLKR/X CLKR/X int 3 tc(CKRX) tw(CKRX) Pulse duration, CLKR/X high or CLKR/X low CLKR/X int 4 td(CKRH-FRV) Delay time, CLKR high to internal FSR valid CLKR int −11 3 CLKX int −11 3 CLKX ext 3 9 CLKX int −9 4 CLKX ext CLKX int 3 −9+ D1|| 9 7 + D2|| CLKX ext 3 + D1|| 19 + D2|| 9 td(CKXH-FXV) Delay time, CLKX high to internal FSX valid 12 tdis(CKXH-DXHZ) Disable time, DX high impedance following last data bit from CLKX high 13 td(CKXH-DXV) Delay time, CLKX high to DX valid 14 td(FXH-DXV) UNIT MIN ns ns Delay time, FSX high to DX valid FSX int −1 3 ONLY applies when in data delay 0 (XDATDLY = 00b) mode FSX ext 3 9 ns ns ns ns † CLKRP = CLKXP = FSRP = FSXP = 0. If polarity of any of the signals is inverted, then the timing references of that signal are also inverted. ‡ Minimum delay times also represent minimum output hold times. § P = 1/CPU clock frequency in ns. For example, when running parts at 100 MHz, use P = 10 ns. ¶ The minimum CLKR/X period is twice the CPU cycle time (2P). This means that the maximum bit rate for communications between the McBSP and other device is 50 Mbps for 100 MHz CPU clock; where the McBSP is either the master or the slave. Care must be taken to ensure that the AC timings specified in this data sheet are met. The maximum bit rate for McBSP-to-McBSP communications is 33 Mbps; therefore, the minimum CLKR/X clock cycle is either twice the CPU cycle time (2P), or 30 ns (33 MHz), whichever value is larger. For example, when running parts at 100 MHz (P = 10 ns), use 30 ns as the minimum CLKR/X clock cycle (by setting the appropriate CLKGDV ratio or external clock source). The maximum bit rate for McBSP-to-McBSP communications applies when the serial port is a master of the clock and frame syncs (with CLKR connected to CLKX, FSR connected to FSX, CLKXM = FSXM = 1, and CLKRM = FSRM = 0) in data delay 1 or 2 mode (R/XDATDLY = 01b or 10b) and the other device the McBSP communicates to is a slave. # C = H or L S = sample rate generator input clock = 2P if CLKSM = 1 (P = 1/CPU clock frequency) = sample rate generator input clock = P_clks if CLKSM = 0 (P_clks = CLKS period) H = CLKX high pulse width = (CLKGDV/2 + 1) * S if CLKGDV is even = (CLKGDV + 1)/2 * S if CLKGDV is odd or zero L = CLKX low pulse width = (CLKGDV/2) * S if CLKGDV is even = (CLKGDV + 1)/2 * S if CLKGDV is odd or zero CLKGDV should be set appropriately to ensure the McBSP bit rate does not exceed the maximum limit (see ¶ footnote above). || Extra delay from CLKX high to DX valid applies only to the first data bit of a device, if and only if DXENA = 1 in SPCR. If DXENA = 0, then D1 = D2 = 0 If DXENA = 1, then D1 = 2P, D2 = 4P 92 POST OFFICE BOX 1443 • HOUSTON, TEXAS 77251−1443 SGUS055 − SEPTEMBER 2004 MULTICHANNEL BUFFERED SERIAL PORT TIMING (CONTINUED) switching characteristics over recommended operating conditions for McBSP†‡ (see Figure 44) [C6712C/C6712D] 12C−167 NO. PARAMETER 12D−167 MIN MAX MIN MAX 1.8 10 1.8 10 UNIT 1 td(CKSH-CKRXH) Delay time, CLKS high to CLKR/X high for internal CLKR/X generated from CLKS input 2 tc(CKRX) Cycle time, CLKR/X CLKR/X int 2P§¶ 3 tw(CKRX) Pulse duration, CLKR/X high or CLKR/X low CLKR/X int C − 1# C + 1# C − 1# C + 1# ns 4 td(CKRH-FRV) Delay time, CLKR high to internal FSR valid CLKR int −2 3 −2 3 ns −2 3 −2 3 td(CKXH-FXV) Delay time, CLKX high to internal FSX valid CLKX int 9 CLKX ext 2 9 2 9 −1 4 −1 4 tdis(CKXH-DXHZ) Disable time, DX high impedance following last data bit from CLKX high CLKX int 12 CLKX ext 1.5 10 1.5 10 CLKX int td(CKXH-DXV) Delay time, CLKX high to DX valid −3.2 + D1|| 0.5 + D1|| 4 + D2|| 10+ D2|| −3.2 + D1|| 0.5 + D1|| 4 + D2|| 10+ D2|| 13 14 td(FXH-DXV) 2P§¶ ns ns ns ns CLKX ext Delay time, FSX high to DX valid FSX int −1.5 4.5 −1 7.5 ONLY applies when in data delay 0 (XDATDLY = 00b) mode FSX ext 2 9 2 11.5 ns ns † CLKRP = CLKXP = FSRP = FSXP = 0. If polarity of any of the signals is inverted, then the timing references of that signal are also inverted. ‡ Minimum delay times also represent minimum output hold times. § P = 1/CPU clock frequency in ns. For example, when running parts at 167 MHz, use P = 6 ns. ¶ The minimum CLKR/X period is twice the CPU cycle time (2P) and not faster than 75 Mbps (13.3 ns). This means that the maximum bit rate for communications between the McBSP and other device is 75 Mbps for 150 MHz CPU clock; where the McBSP is either the master or the slave. Care must be taken to ensure that the AC timings specified in this data sheet are met. The maximum bit rate for McBSP-to-McBSP communications is 67 Mbps; therefore, the minimum CLKR/X clock cycle is either twice the CPU cycle time (2P), or 15 ns (67 MHz), whichever value is larger. For example, when running parts at 150 MHz (P = 6.7 ns), use 15 ns as the minimum CLKR/X clock cycle (by setting the appropriate CLKGDV ratio or external clock source). When running parts at 60 MHz (P = 16.67 ns), use 2P = 33 ns (30 MHz) as the minimum CLKR/X clock cycle. The maximum bit rate for McBSP-to-McBSP communications applies when the serial port is a master of the clock and frame syncs (with CLKR connected to CLKX, FSR connected to FSX, CLKXM = FSXM = 1, and CLKRM = FSRM = 0) in data delay 1 or 2 mode (R/XDATDLY = 01b or 10b) and the other device the McBSP communicates to is a slave. # C = H or L S = sample rate generator input clock = 2P if CLKSM = 1 (P = 1/CPU clock frequency) = sample rate generator input clock = P_clks if CLKSM = 0 (P_clks = CLKS period) H = CLKX high pulse width = (CLKGDV/2 + 1) * S if CLKGDV is even = (CLKGDV + 1)/2 * S if CLKGDV is odd or zero L = CLKX low pulse width = (CLKGDV/2) * S if CLKGDV is even = (CLKGDV + 1)/2 * S if CLKGDV is odd or zero CLKGDV should be set appropriately to ensure the McBSP bit rate does not exceed the maximum limit (see ¶ footnote above). || Extra delay from CLKX high to DX valid applies only to the first data bit of a device, if and only if DXENA = 1 in SPCR. If DXENA = 0, then D1 = D2 = 0 If DXENA = 1, then D1 = 2P, D2 = 4P POST OFFICE BOX 1443 • HOUSTON, TEXAS 77251−1443 93 SGUS055 − SEPTEMBER 2004 MULTICHANNEL BUFFERED SERIAL PORT TIMING (CONTINUED) CLKS 1 2 3 3 CLKR 4 4 FSR (int) 5 6 FSR (ext) 7 DR 8 Bit(n-1) (n-2) (n-3) 2 3 3 CLKX 9 FSX (int) 11 10 FSX (ext) FSX (XDATDLY=00b) 12 DX Bit 0 14 13 Bit(n-1) 13 (n-2) Figure 44. McBSP Timings 94 POST OFFICE BOX 1443 • HOUSTON, TEXAS 77251−1443 (n-3) SGUS055 − SEPTEMBER 2004 MULTICHANNEL BUFFERED SERIAL PORT TIMING (CONTINUED) timing requirements for FSR when GSYNC = 1 (see Figure 45) −100 −167 NO. MIN 1 2 tsu(FRH-CKSH) th(CKSH-FRH) UNIT MAX Setup time, FSR high before CLKS high 4 ns Hold time, FSR high after CLKS high 4 ns CLKS 1 2 FSR external CLKR/X (no need to resync) CLKR/X (needs resync) Figure 45. FSR Timing When GSYNC = 1 POST OFFICE BOX 1443 • HOUSTON, TEXAS 77251−1443 95 SGUS055 − SEPTEMBER 2004 MULTICHANNEL BUFFERED SERIAL PORT TIMING (CONTINUED) timing requirements for McBSP as SPI master or slave: CLKSTP = 10b, CLKXP = 0†‡ (see Figure 46) [C6712] −100 MASTER NO. MIN 4 tsu(DRV-CKXL) th(CKXL-DRV) Setup time, DR valid before CLKX low SLAVE MAX 26 5 Hold time, DR valid after CLKX low 4 † P = 1/CPU clock frequency in ns. For example, when running parts at 100 MHz, use P = 10 ns. ‡ For all SPI slave modes, CLKG is programmed as 1/2 of the CPU clock by setting CLKSM = CLKGDV = 1. MIN UNIT MAX 2 − 6P ns 6 + 12P ns timing requirements for McBSP as SPI master or slave: CLKSTP = 10b, CLKXP = 0†‡ (see Figure 46) [C6712C/C6712D] −167 NO. 4 5 tsu(DRV-CKXL) th(CKXL-DRV) Setup time, DR valid before CLKX low MASTER SLAVE MIN MIN MAX UNIT MAX 12 2 − 6P ns 4 5 + 12P ns Hold time, DR valid after CLKX low † P = 1/CPU clock frequency in ns. For example, when running parts at 167 MHz, use P = 6 ns. ‡ For all SPI slave modes, CLKG is programmed as 1/2 of the CPU clock by setting CLKSM = CLKGDV = 1. switching characteristics over recommended operating conditions for McBSP as SPI master or slave: CLKSTP = 10b, CLKXP = 0†‡ (see Figure 46) [C6712] −100 NO. MASTER§ PARAMETER 2 th(CKXL-FXL) td(FXL-CKXH) Hold time, FSX low after CLKX low¶ Delay time, FSX low to CLKX high# 3 td(CKXH-DXV) Delay time, CLKX high to DX valid 6 tdis(CKXL-DXHZ) Disable time, DX high impedance following last data bit from CLKX low 7 tdis(FXH-DXHZ) Disable time, DX high impedance following last data bit from FSX high 1 SLAVE MIN UNIT MIN MAX T−9 T+9 ns L−9 L+9 ns −9 9 L−9 L+9 6P + 4 MAX 10P + 20 ns ns 2P + 3 6P + 20 ns 8 td(FXL-DXV) Delay time, FSX low to DX valid 4P + 2 8P + 20 ns † P = 1/CPU clock frequency in ns. For example, when running parts at 100 MHz, use P = 10 ns. ‡ For all SPI slave modes, CLKG is programmed as 1/2 of the CPU clock by setting CLKSM = CLKGDV = 1. § S = Sample rate generator input clock = 2P if CLKSM = 1 (P = 1/CPU clock frequency) = Sample rate generator input clock = P_clks if CLKSM = 0 (P_clks = CLKS period) T = CLKX period = (1 + CLKGDV) * S H = CLKX high pulse width = (CLKGDV/2 + 1) * S if CLKGDV is even = (CLKGDV + 1)/2 * S if CLKGDV is odd or zero L = CLKX low pulse width = (CLKGDV/2) * S if CLKGDV is even = (CLKGDV + 1)/2 * S if CLKGDV is odd or zero ¶ FSRP = FSXP = 1. As a SPI master, FSX is inverted to provide active-low slave-enable output. As a slave, the active-low signal input on FSX and FSR is inverted before being used internally. CLKXM = FSXM = 1, CLKRM = FSRM = 0 for master McBSP CLKXM = CLKRM = FSXM = FSRM = 0 for slave McBSP # FSX 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 (CLKX). 96 POST OFFICE BOX 1443 • HOUSTON, TEXAS 77251−1443 SGUS055 − SEPTEMBER 2004 MULTICHANNEL BUFFERED SERIAL PORT TIMING (CONTINUED) switching characteristics over recommended operating conditions for McBSP as SPI master or slave: CLKSTP = 10b, CLKXP = 0†‡ (see Figure 46) [C6712C/C6712D] 12C−167 NO. MASTER§ PARAMETER 12D−167 MASTER§ SLAVE MIN MAX MIN MAX SLAVE MIN MAX MIN UNIT MAX 1 th(CKXL-FXL) Hold time, FSX low after CLKX low¶ T−2 T+3 T−2 T+3 ns 2 td(FXL-CKXH) Delay time, FSX low to CLKX high# L−2 L+3 L−2 L+3 ns 3 td(CKXH-DXV) Delay time, CLKX high to DX valid −3 4 −3 4 tdis(CKXL-DXHZ) Disable time, DX high impedance following last data bit from CLKX low L−4 L+3 L−2 L+3 7 tdis(FXH-DXHZ) Disable time, DX high impedance following last data bit from FSX high 8 td(FXL-DXV) Delay time, FSX low to DX valid 6 6P + 2 10P + 17 6P + 2 10P + 17 ns ns 2P + 1.5 6P + 17 2P + 3 6P + 17 ns 4P + 2 8P + 17 4P + 2 8P + 17 ns † P = 1/CPU clock frequency in ns. For example, when running parts at 167 MHz, use P = 6 ns. ‡ For all SPI slave modes, CLKG is programmed as 1/2 of the CPU clock by setting CLKSM = CLKGDV = 1. § S = Sample rate generator input clock = 2P if CLKSM = 1 (P = 1/CPU clock frequency) = Sample rate generator input clock = P_clks if CLKSM = 0 (P_clks = CLKS period) T = CLKX period = (1 + CLKGDV) * S H = CLKX high pulse width = (CLKGDV/2 + 1) * S if CLKGDV is even = (CLKGDV + 1)/2 * S if CLKGDV is odd or zero L = CLKX low pulse width = (CLKGDV/2) * S if CLKGDV is even = (CLKGDV + 1)/2 * S if CLKGDV is odd or zero ¶ FSRP = FSXP = 1. As a SPI master, FSX is inverted to provide active-low slave-enable output. As a slave, the active-low signal input on FSX and FSR is inverted before being used internally. CLKXM = FSXM = 1, CLKRM = FSRM = 0 for master McBSP CLKXM = CLKRM = FSXM = FSRM = 0 for slave McBSP # FSX 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 (CLKX). CLKX 1 2 FSX 7 6 DX 8 3 Bit 0 Bit(n-1) 4 DR Bit 0 (n-2) (n-3) (n-4) 5 Bit(n-1) (n-2) (n-3) (n-4) Figure 46. McBSP Timing as SPI Master or Slave: CLKSTP = 10b, CLKXP = 0 POST OFFICE BOX 1443 • HOUSTON, TEXAS 77251−1443 97 SGUS055 − SEPTEMBER 2004 MULTICHANNEL BUFFERED SERIAL PORT TIMING (CONTINUED) timing requirements for McBSP as SPI master or slave: CLKSTP = 11b, CLKXP = 0†‡ (see Figure 47) [C6712] −100 MASTER NO. MIN 4 tsu(DRV-CKXH) th(CKXH-DRV) Setup time, DR valid before CLKX high SLAVE MAX 26 5 Hold time, DR valid after CLKX high 4 † P = 1/CPU clock frequency in ns. For example, when running parts at 100 MHz, use P = 10 ns. ‡ For all SPI slave modes, CLKG is programmed as 1/2 of the CPU clock by setting CLKSM = CLKGDV = 1. MIN UNIT MAX 2 − 6P ns 6 + 12P ns timing requirements for McBSP as SPI master or slave: CLKSTP = 11b, CLKXP = 0†‡ (see Figure 47) [C6712C/C6712D] −167 MASTER NO. MIN 4 5 tsu(DRV-CKXH) th(CKXH-DRV) Setup time, DR valid before CLKX high SLAVE MAX MIN UNIT MAX 12 2 − 6P ns 4 5 + 12P ns Hold time, DR valid after CLKX high † P = 1/CPU clock frequency in ns. For example, when running parts at 167 MHz, use P = 6 ns. ‡ For all SPI slave modes, CLKG is programmed as 1/2 of the CPU clock by setting CLKSM = CLKGDV = 1. switching characteristics over recommended operating conditions for McBSP as SPI master or slave: CLKSTP = 11b, CLKXP = 0†‡ (see Figure 47) [C6712] −100 NO. MASTER§ PARAMETER MIN 2 th(CKXL-FXL) td(FXL-CKXH) Hold time, FSX low after CLKX low¶ Delay time, FSX low to CLKX high# 3 td(CKXL-DXV) 6 tdis(CKXL-DXHZ) 1 SLAVE MAX MIN UNIT MAX L−9 L+9 T−9 T+9 ns Delay time, CLKX low to DX valid −9 9 6P + 4 10P + 20 ns Disable time, DX high impedance following last data bit from CLKX low −9 9 6P + 3 10P + 20 ns ns 7 td(FXL-DXV) Delay time, FSX low to DX valid H−9 H+9 4P + 2 8P + 20 ns † P = 1/CPU clock frequency in ns. For example, when running parts at 100 MHz, use P = 10 ns. ‡ For all SPI slave modes, CLKG is programmed as 1/2 of the CPU clock by setting CLKSM = CLKGDV = 1. § S = Sample rate generator input clock = 2P if CLKSM = 1 (P = 1/CPU clock frequency) = Sample rate generator input clock = P_clks if CLKSM = 0 (P_clks = CLKS period) T = CLKX period = (1 + CLKGDV) * S H = CLKX high pulse width = (CLKGDV/2 + 1) * S if CLKGDV is even = (CLKGDV + 1)/2 * S if CLKGDV is odd or zero L = CLKX low pulse width = (CLKGDV/2) * S if CLKGDV is even = (CLKGDV + 1)/2 * S if CLKGDV is odd or zero ¶ FSRP = FSXP = 1. As a SPI master, FSX is inverted to provide active-low slave-enable output. As a slave, the active-low signal input on FSX and FSR is inverted before being used internally. CLKXM = FSXM = 1, CLKRM = FSRM = 0 for master McBSP CLKXM = CLKRM = FSXM = FSRM = 0 for slave McBSP # FSX 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 (CLKX). 98 POST OFFICE BOX 1443 • HOUSTON, TEXAS 77251−1443 SGUS055 − SEPTEMBER 2004 MULTICHANNEL BUFFERED SERIAL PORT TIMING (CONTINUED) switching characteristics over recommended operating conditions for McBSP as SPI master or slave: CLKSTP = 11b, CLKXP = 0†‡ (see Figure 47) [C6712C/C6712D] 12C−167 NO. MASTER§ PARAMETER 12D−167 MASTER§ SLAVE MIN MAX MIN MAX SLAVE MIN MAX MIN UNIT MAX 1 th(CKXL-FXL) Hold time, FSX low after CLKX low¶ L−2 L+3 L−2 L+3 ns 2 td(FXL-CKXH) Delay time, FSX low to CLKX high# T−2 T+3 T−2 T+3 ns 3 td(CKXL-DXV) Delay time, CLKX low to DX valid −3 4 6P + 2 10P + 17 −3 4 6P + 2 10P + 17 ns 6 tdis(CKXL-DXHZ) Disable time, DX high impedance following last data bit from CLKX low −4 4 6P + 1.5 10P + 17 −2 4 6P + 3 10P + 17 ns 7 td(FXL-DXV) Delay time, FSX low to DX valid H−2 H+4 4P + 2 8P + 17 H−2 H + 6.5 4P + 2 8P + 17 ns † P = 1/CPU clock frequency in ns. For example, when running parts at 167 MHz, use P = 6 ns. ‡ For all SPI slave modes, CLKG is programmed as 1/2 of the CPU clock by setting CLKSM = CLKGDV = 1. § S = Sample rate generator input clock = 2P if CLKSM = 1 (P = 1/CPU clock frequency) = Sample rate generator input clock = P_clks if CLKSM = 0 (P_clks = CLKS period) T = CLKX period = (1 + CLKGDV) * S H = CLKX high pulse width = (CLKGDV/2 + 1) * S if CLKGDV is even = (CLKGDV + 1)/2 * S if CLKGDV is odd or zero L = CLKX low pulse width = (CLKGDV/2) * S if CLKGDV is even = (CLKGDV + 1)/2 * S if CLKGDV is odd or zero ¶ FSRP = FSXP = 1. As a SPI master, FSX is inverted to provide active-low slave-enable output. As a slave, the active-low signal input on FSX and FSR is inverted before being used internally. CLKXM = FSXM = 1, CLKRM = FSRM = 0 for master McBSP CLKXM = CLKRM = FSXM = FSRM = 0 for slave McBSP # FSX 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 (CLKX). CLKX 1 2 6 Bit 0 7 FSX DX 3 Bit(n-1) 4 DR Bit 0 (n-2) (n-3) (n-4) 5 Bit(n-1) (n-2) (n-3) (n-4) Figure 47. McBSP Timing as SPI Master or Slave: CLKSTP = 11b, CLKXP = 0 POST OFFICE BOX 1443 • HOUSTON, TEXAS 77251−1443 99 SGUS055 − SEPTEMBER 2004 MULTICHANNEL BUFFERED SERIAL PORT TIMING (CONTINUED) timing requirements for McBSP as SPI master or slave: CLKSTP = 10b, CLKXP = 1†‡ (see Figure 48) [C6712] −100 MASTER NO. MIN 4 tsu(DRV-CKXH) th(CKXH-DRV) Setup time, DR valid before CLKX high SLAVE MAX 26 5 Hold time, DR valid after CLKX high 4 † P = 1/CPU clock frequency in ns. For example, when running parts at 100 MHz, use P = 10 ns. ‡ For all SPI slave modes, CLKG is programmed as 1/2 of the CPU clock by setting CLKSM = CLKGDV = 1. MIN UNIT MAX 2 − 6P ns 6 + 12P ns timing requirements for McBSP as SPI master or slave: CLKSTP = 10b, CLKXP = 1†‡ (see Figure 48) [C6712C/C6712D] −167 MASTER NO. MIN 4 5 tsu(DRV-CKXH) th(CKXH-DRV) Setup time, DR valid before CLKX high SLAVE MAX MIN UNIT MAX 12 2 − 6P ns 4 5 + 12P ns Hold time, DR valid after CLKX high † P = 1/CPU clock frequency in ns. For example, when running parts at 167 MHz, use P = 6 ns. ‡ For all SPI slave modes, CLKG is programmed as 1/2 of the CPU clock by setting CLKSM = CLKGDV = 1. switching characteristics over recommended operating conditions for McBSP as SPI master or slave: CLKSTP = 10b, CLKXP = 1†‡ (see Figure 48) [C6712] −100 NO. MASTER§ PARAMETER MIN 2 th(CKXH-FXL) td(FXL-CKXL) Hold time, FSX low after CLKX high¶ Delay time, FSX low to CLKX low# 3 td(CKXL-DXV) Delay time, CLKX low to DX valid 6 tdis(CKXH-DXHZ) Disable time, DX high impedance following last data bit from CLKX high 7 tdis(FXH-DXHZ) Disable time, DX high impedance following last data bit from FSX high 1 SLAVE MAX T−9 T+9 H−9 H+9 −9 9 H−9 H+9 MIN UNIT MAX ns ns 6P + 4 10P + 20 ns ns 2P + 3 6P + 20 ns 8 td(FXL-DXV) Delay time, FSX low to DX valid 4P + 2 8P + 20 ns † P = 1/CPU clock frequency in ns. For example, when running parts at 100 MHz, use P = 10 ns. ‡ For all SPI slave modes, CLKG is programmed as 1/2 of the CPU clock by setting CLKSM = CLKGDV = 1. § S = Sample rate generator input clock = 2P if CLKSM = 1 (P = 1/CPU clock frequency) = Sample rate generator input clock = P_clks if CLKSM = 0 (P_clks = CLKS period) T = CLKX period = (1 + CLKGDV) * S H = CLKX high pulse width = (CLKGDV/2 + 1) * S if CLKGDV is even = (CLKGDV + 1)/2 * S if CLKGDV is odd or zero L = CLKX low pulse width = (CLKGDV/2) * S if CLKGDV is even = (CLKGDV + 1)/2 * S if CLKGDV is odd or zero ¶ FSRP = FSXP = 1. As a SPI master, FSX is inverted to provide active-low slave-enable output. As a slave, the active-low signal input on FSX and FSR is inverted before being used internally. CLKXM = FSXM = 1, CLKRM = FSRM = 0 for master McBSP CLKXM = CLKRM = FSXM = FSRM = 0 for slave McBSP # FSX 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 (CLKX). 100 POST OFFICE BOX 1443 • HOUSTON, TEXAS 77251−1443 SGUS055 − SEPTEMBER 2004 MULTICHANNEL BUFFERED SERIAL PORT TIMING (CONTINUED) switching characteristics over recommended operating conditions for McBSP as SPI master or slave: CLKSTP = 10b, CLKXP = 1†‡ (see Figure 48) [C6712C/C6712D] 12C−167 NO. MASTER§ PARAMETER 12D−167 MASTER§ SLAVE MIN MAX MIN MAX SLAVE MIN MAX MIN UNIT MAX 1 th(CKXH-FXL) Hold time, FSX low after CLKX high¶ T−2 T+3 T−2 T+3 ns 2 td(FXL-CKXL) Delay time, FSX low to CLKX low# H−2 H+3 H−2 H+3 ns 3 td(CKXL-DXV) Delay time, CLKX low to DX valid −3 4 −3 4 tdis(CKXH-DXHZ) Disable time, DX high impedance following last data bit from CLKX high H − 3.6 H+3 H−2 H+3 7 tdis(FXH-DXHZ) Disable time, DX high impedance following last data bit from FSX high 8 td(FXL-DXV) Delay time, FSX low to DX valid 6 6P + 2 10P + 17 6P + 2 10P + 17 ns ns 2P + 1.5 6P + 17 2P + 3 6P + 17 ns 4P + 2 8P + 17 4P + 2 8P + 17 ns † P = 1/CPU clock frequency in ns. For example, when running parts at 167 MHz, use P = 6 ns. ‡ For all SPI slave modes, CLKG is programmed as 1/2 of the CPU clock by setting CLKSM = CLKGDV = 1. § S = Sample rate generator input clock = 2P if CLKSM = 1 (P = 1/CPU clock frequency) = Sample rate generator input clock = P_clks if CLKSM = 0 (P_clks = CLKS period) T = CLKX period = (1 + CLKGDV) * S H = CLKX high pulse width = (CLKGDV/2 + 1) * S if CLKGDV is even = (CLKGDV + 1)/2 * S if CLKGDV is odd or zero L = CLKX low pulse width = (CLKGDV/2) * S if CLKGDV is even = (CLKGDV + 1)/2 * S if CLKGDV is odd or zero ¶ FSRP = FSXP = 1. As a SPI master, FSX is inverted to provide active-low slave-enable output. As a slave, the active-low signal input on FSX and FSR is inverted before being used internally. CLKXM = FSXM = 1, CLKRM = FSRM = 0 for master McBSP CLKXM = CLKRM = FSXM = FSRM = 0 for slave McBSP # FSX 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 (CLKX). CLKX 1 2 FSX 7 6 DX 8 3 Bit 0 Bit(n-1) 4 DR Bit 0 (n-2) (n-3) (n-4) 5 Bit(n-1) (n-2) (n-3) (n-4) Figure 48. McBSP Timing as SPI Master or Slave: CLKSTP = 10b, CLKXP = 1 POST OFFICE BOX 1443 • HOUSTON, TEXAS 77251−1443 101 SGUS055 − SEPTEMBER 2004 MULTICHANNEL BUFFERED SERIAL PORT TIMING (CONTINUED) timing requirements for McBSP as SPI master or slave: CLKSTP = 11b, CLKXP = 1†‡ (see Figure 49) [C6712] −100 MASTER NO. MIN 4 tsu(DRV-CKXH) th(CKXH-DRV) Setup time, DR valid before CLKX high SLAVE MAX 26 5 Hold time, DR valid after CLKX high 4 † P = 1/CPU clock frequency in ns. For example, when running parts at 100 MHz, use P = 10 ns. ‡ For all SPI slave modes, CLKG is programmed as 1/2 of the CPU clock by setting CLKSM = CLKGDV = 1. MIN UNIT MAX 2 − 6P ns 6 + 12P ns timing requirements for McBSP as SPI master or slave: CLKSTP = 11b, CLKXP = 1†‡ (see Figure 49) [C6712C/C6712D] −167 MASTER NO. MIN 4 5 tsu(DRV-CKXH) th(CKXH-DRV) Setup time, DR valid before CLKX high SLAVE MAX MIN UNIT MAX 12 2 − 6P ns 4 5 + 12P ns Hold time, DR valid after CLKX high † P = 1/CPU clock frequency in ns. For example, when running parts at 167 MHz, use P = 6 ns. ‡ For all SPI slave modes, CLKG is programmed as 1/2 of the CPU clock by setting CLKSM = CLKGDV = 1. switching characteristics over recommended operating conditions for McBSP as SPI master or slave: CLKSTP = 11b, CLKXP = 1†‡ (see Figure 49) [C6712] −100 NO. MASTER§ PARAMETER SLAVE MIN MAX H−9 H+9 T−9 T+9 MIN UNIT MAX 2 th(CKXH-FXL) td(FXL-CKXL) Hold time, FSX low after CLKX high¶ Delay time, FSX low to CLKX low# 3 td(CKXH-DXV) Delay time, CLKX high to DX valid −9 9 6P + 4 10P + 20 ns 6 tdis(CKXH-DXHZ) Disable time, DX high impedance following last data bit from CLKX high −9 9 6P + 3 10P + 20 ns 1 ns ns 7 td(FXL-DXV) Delay time, FSX low to DX valid L−9 L+9 4P + 2 8P + 20 ns † P = 1/CPU clock frequency in ns. For example, when running parts at 100 MHz, use P = 10 ns. ‡ For all SPI slave modes, CLKG is programmed as 1/2 of the CPU clock by setting CLKSM = CLKGDV = 1. § S = Sample rate generator input clock = 2P if CLKSM = 1 (P = 1/CPU clock frequency) = Sample rate generator input clock = P_clks if CLKSM = 0 (P_clks = CLKS period) T = CLKX period = (1 + CLKGDV) * S H = CLKX high pulse width = (CLKGDV/2 + 1) * S if CLKGDV is even = (CLKGDV + 1)/2 * S if CLKGDV is odd or zero L = CLKX low pulse width = (CLKGDV/2) * S if CLKGDV is even = (CLKGDV + 1)/2 * S if CLKGDV is odd or zero ¶ FSRP = FSXP = 1. As a SPI master, FSX is inverted to provide active-low slave-enable output. As a slave, the active-low signal input on FSX and FSR is inverted before being used internally. CLKXM = FSXM = 1, CLKRM = FSRM = 0 for master McBSP CLKXM = CLKRM = FSXM = FSRM = 0 for slave McBSP # FSX 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 (CLKX). 102 POST OFFICE BOX 1443 • HOUSTON, TEXAS 77251−1443 SGUS055 − SEPTEMBER 2004 MULTICHANNEL BUFFERED SERIAL PORT TIMING (CONTINUED) switching characteristics over recommended operating conditions for McBSP as SPI master or slave: CLKSTP = 11b, CLKXP = 1†‡ (see Figure 49) [C6712C/C6712D] 12C−167 NO. MASTER§ PARAMETER 12D−167 MASTER§ SLAVE MIN MAX MIN MAX SLAVE MIN MAX MIN UNIT MAX 1 th(CKXH-FXL) Hold time, FSX low after CLKX high¶ H−2 H+3 H−2 H+3 ns 2 td(FXL-CKXL) Delay time, FSX low to CLKX low# T−2 T+3 T−2 T+3 ns 3 td(CKXH-DXV) Delay time, CLKX high to DX valid −3 4 6P + 2 10P + 17 −3 4 6P + 2 10P + 17 ns 6 tdis(CKXH-DXHZ) Disable time, DX high impedance following last data bit from CLKX high −3.6 4 6P + 1.5 10P + 17 −2 4 6P + 3 10P + 17 ns 7 td(FXL-DXV) Delay time, FSX low to DX valid L−2 L+4 4P + 2 8P + 17 L−2 L + 6.5 4P + 2 8P + 17 ns † P = 1/CPU clock frequency in ns. For example, when running parts at 167 MHz, use P = 6 ns. ‡ For all SPI slave modes, CLKG is programmed as 1/2 of the CPU clock by setting CLKSM = CLKGDV = 1. § S = Sample rate generator input clock = 2P if CLKSM = 1 (P = 1/CPU clock frequency) = Sample rate generator input clock = P_clks if CLKSM = 0 (P_clks = CLKS period) T = CLKX period = (1 + CLKGDV) * S H = CLKX high pulse width = (CLKGDV/2 + 1) * S if CLKGDV is even = (CLKGDV + 1)/2 * S if CLKGDV is odd or zero L = CLKX low pulse width = (CLKGDV/2) * S if CLKGDV is even = (CLKGDV + 1)/2 * S if CLKGDV is odd or zero ¶ FSRP = FSXP = 1. As a SPI master, FSX is inverted to provide active-low slave-enable output. As a slave, the active-low signal input on FSX and FSR is inverted before being used internally. CLKXM = FSXM = 1, CLKRM = FSRM = 0 for master McBSP CLKXM = CLKRM = FSXM = FSRM = 0 for slave McBSP # FSX 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 (CLKX). CLKX 1 2 FSX 7 6 DX 3 Bit 0 Bit(n-1) 4 DR Bit 0 (n-2) (n-3) (n-4) 5 Bit(n-1) (n-2) (n-3) (n-4) Figure 49. McBSP Timing as SPI Master or Slave: CLKSTP = 11b, CLKXP = 1 POST OFFICE BOX 1443 • HOUSTON, TEXAS 77251−1443 103 SGUS055 − SEPTEMBER 2004 TIMER TIMING timing requirements for timer inputs† (see Figure 50) −100 −167 NO. MIN 1 2 tw(TINPH) tw(TINPL) UNIT MAX Pulse duration, TINP high 2P ns Pulse duration, TINP low 2P ns † P = 1/CPU clock frequency in ns. For example, when running parts at 100 MHz, use P = 10 ns. switching characteristics over recommended operating conditions for timer outputs† (see Figure 50) NO. −100 −167 PARAMETER MIN 3 4 tw(TOUTH) tw(TOUTL) Pulse duration, TOUT high 4P −3 ns Pulse duration, TOUT low 4P −3 ns † P = 1/CPU clock frequency in ns. For example, when running parts at 100 MHz, use P = 10 ns. 2 1 TINPx 4 3 TOUTx Figure 50. Timer Timing 104 UNIT MAX POST OFFICE BOX 1443 • HOUSTON, TEXAS 77251−1443 SGUS055 − SEPTEMBER 2004 GENERAL-PURPOSE INPUT/OUTPUT (GPIO) PORT TIMING [C6712C/C6712D ONLY] timing requirements for GPIO inputs†‡ (see Figure 51) −167 NO. 1 MIN tw(GPIH) tw(GPIL) Pulse duration, GPIx high MAX 4P UNIT ns 2 Pulse duration, GPIx low 4P ns † P = 1/CPU clock frequency in ns. For example, when running parts at 167 MHz, use P = 6 ns. ‡ The pulse width given is sufficient to generate a CPU interrupt or an EDMA event. However, if a user wants to have the DSP recognize the GPIx changes through software polling of the GPIO register, the GPIx duration must be extended to at least 24P to allow the DSP enough time to access the GPIO register through the CFGBUS. switching characteristics over recommended operating conditions for GPIO outputs†§ (see Figure 51) −167 NO. 3 PARAMETER tw(GPOH) tw(GPOL) MIN Pulse duration, GPOx high 12P − 3 MAX UNIT ns 4 Pulse duration, GPOx low 12P − 3 ns † P = 1/CPU clock frequency in ns. For example, when running parts at 167 MHz, use P = 6 ns. § The number of CFGBUS cycles between two back-to-back CFGBUS writes to the GPIO register is 12 SYSCLK1 cycles; therefore, the minimum GPOx pulse width is 12P. 2 1 GPIx 4 3 GPOx Figure 51. GPIO Port Timing POST OFFICE BOX 1443 • HOUSTON, TEXAS 77251−1443 105 SGUS055 − SEPTEMBER 2004 JTAG TEST-PORT TIMING timing requirements for JTAG test port (see Figure 52) −100 NO. 1 MIN −167 MAX MIN MAX UNIT Cycle time, TCK 35 35 ns 3 tc(TCK) tsu(TDIV-TCKH) Setup time, TDI/TMS/TRST valid before TCK high 10 10 ns 4 th(TCKH-TDIV) Hold time, TDI/TMS/TRST valid after TCK high 5 7 ns switching characteristics over recommended operating conditions for JTAG test port (see Figure 52) −100 NO. 2 PARAMETER td(TCKL-TDOV) Delay time, TCK low to TDO valid −167 MIN MAX MIN MAX –3 18 0 15 1 TCK 2 2 TDO 4 3 TDI/TMS/TRST Figure 52. JTAG Test-Port Timing 106 POST OFFICE BOX 1443 • HOUSTON, TEXAS 77251−1443 UNIT ns SGUS055 − SEPTEMBER 2004 MECHANICAL DATA [C6712 ONLY] GFN (S-PBGA-N256) [C6712 only] PLASTIC BALL GRID ARRAY 27,20 SQ 26,80 24,70 SQ 23,80 24,13 TYP 1,27 0,635 A1 Corner 0,635 1,27 Y W V U T R P N M L K J H G F E D C B A 1 3 2 5 4 7 6 9 8 10 11 13 15 17 19 12 14 16 18 20 Bottom View 2,32 MAX 1,17 NOM Seating Plane 0,40 0,30 0,90 0,60 0,15 M 0,70 0,50 0,15 4040185-2/D 02/02 NOTES: A. All linear dimensions are in millimeters. B. This drawing is subject to change without notice. C. Falls within JEDEC MO-151 thermal resistance characteristics (S-PBGA package) [C6712 only] °C/W Air Flow (m/s)† RΘJC RΘJA Junction-to-case 6.4 N/A Junction-to-free air 25.5 0.0 RΘJA RΘJA Junction-to-free air 23.1 0.5 Junction-to-free air 22.3 1.0 RΘJA Junction-to-free air † m/s = meters per second 21.2 2.0 NO 1 2 3 4 5 POST OFFICE BOX 1443 • HOUSTON, TEXAS 77251−1443 107 SGUS055 − SEPTEMBER 2004 MECHANICAL DATA [C6712C/C6712D ONLY] GDP (S−PBGA−N272) [C6712C/12D only] PLASTIC BALL GRID ARRAY 27,20 SQ 26,80 24,20 SQ 23,80 24,13 TYP 1,27 0,635 Y W V U T R P N M L K J H G F E D C B A A1 Corner 1,27 0,635 3 1 2 1,22 1,12 5 4 7 6 9 8 11 13 15 17 19 10 12 14 16 18 20 Bottom View 2,57 MAX Seating Plane 0,65 0,57 0,90 0,60 0,10 0,70 0,50 0,15 4204396/A 04/02 NOTES: A. All linear dimensions are in millimeters. B. This drawing is subject to change without notice. C. Falls within JEDEC MO-151 thermal resistance characteristics (S-PBGA package) [C6712C/12D only] NO °C/W Air Flow (m/s)† Two Signals, Two Planes (4-Layer Board) 1 RΘJC Junction-to-case 9.7 N/A 2 PsiJT Junction-to-package top 1.5 0.0 3 RΘJB RΘJA Junction-to-board 19 N/A Junction-to-free air 22 0.0 RΘJA RΘJA Junction-to-free air 21 0.5 Junction-to-free air 20 1.0 RΘJA RΘJA Junction-to-free air 19 2.0 Junction-to-free air 18 4.0 16 0.0 4 5 6 7 8 9 PsiJB Junction-to-board † m/s = meters per second 108 POST OFFICE BOX 1443 • HOUSTON, TEXAS 77251−1443 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 TI 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. Information of third parties may be subject to additional restrictions. 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. TI products are not authorized for use in safety-critical applications (such as life support) where a failure of the TI product would reasonably be expected to cause severe personal injury or death, unless officers of the parties have executed an agreement specifically governing such use. Buyers represent that they have all necessary expertise in the safety and regulatory ramifications of their applications, and acknowledge and agree that they are solely responsible for all legal, regulatory and safety-related requirements concerning their products and any use of TI products in such safety-critical applications, notwithstanding any applications-related information or support that may be provided by TI. Further, Buyers must fully indemnify TI and its representatives against any damages arising out of the use of TI products in such safety-critical applications. TI products are neither designed nor intended for use in military/aerospace applications or environments unless the TI products are specifically designated by TI as military-grade or "enhanced plastic." Only products designated by TI as military-grade meet military specifications. Buyers acknowledge and agree that any such use of TI products which TI has not designated as military-grade is solely at the Buyer's risk, and that they are solely responsible for compliance with all legal and regulatory requirements in connection with such use. TI products are neither designed nor intended for use in automotive applications or environments unless the specific TI products are designated by TI as compliant with ISO/TS 16949 requirements. Buyers acknowledge and agree that, if they use any non-designated products in automotive applications, TI will not be responsible for any failure to meet such requirements. Following are URLs where you can obtain information on other Texas Instruments products and application solutions: Products Amplifiers Data Converters DSP Clocks and Timers Interface Logic Power Mgmt Microcontrollers RFID RF/IF and ZigBee® Solutions amplifier.ti.com dataconverter.ti.com dsp.ti.com www.ti.com/clocks interface.ti.com logic.ti.com power.ti.com microcontroller.ti.com www.ti-rfid.com www.ti.com/lprf Applications Audio Automotive Broadband Digital Control Medical Military Optical Networking Security Telephony Video & Imaging Wireless www.ti.com/audio www.ti.com/automotive www.ti.com/broadband www.ti.com/digitalcontrol www.ti.com/medical www.ti.com/military www.ti.com/opticalnetwork www.ti.com/security www.ti.com/telephony www.ti.com/video www.ti.com/wireless Mailing Address: Texas Instruments, Post Office Box 655303, Dallas, Texas 75265 Copyright © 2008, Texas Instruments Incorporated