| 전자부품 데이터시트 검색엔진 |
|
ADSP-SC596 데이터시트(PDF) 6 Page - Analog Devices |
|
|
|||||||||||||||||||||||||||||
ADSP-SC596 데이터시트(HTML) 6 Page - Analog Devices |
|
6 / 134 page ![]() ADSP-SC595/SC596/SC598 Preliminary Technical Data Rev. PrD | Page 6 of 134 | May 2022 accesses are most efficient when one block stores data using the DM bus for transfers, and the other block stores instructions and data using the PM bus for transfers. Using the DM and PM buses, with each bus dedicated to a memory block, assures single-cycle execution with two data transfers. In this case, the instruction must be available in the cache. The system configuration is flexible, but a typical configuration is 512 kb DM, 128 kb PM, and 128 kb of instruction cache, with the remaining L1 memory configured as SRAM. Each address- able memory space outside the L1 memory can be accessed either directly or via cache. The memory map in Table 2 gives the L1 memory address space and shows multiple L1 memory blocks offering a configurable mix of SRAM and cache. L1 Requester and Completer Ports Each SHARC+ core has two requester/completer ports to and from the system fabric. One requester port fetches instructions. The second requester port drives data to the system world. Completer Port 1 together with Completer Port 2 memory direct memory access (MDMA) run conflict free access to the individual memory blocks. For the completer port address, refer to the L1 memory address map in Table 2. L1 On-Chip Memory Bandwidth The internal memory architecture allows programs to have four accesses at the same time to any of the four blocks, assuming no block conflicts. The total bandwidth is realized using both the DMD and PMD buses (2 × 64-bits CCLK speed and 2 × 32-bit SYSCLK speed). Instruction and Data Cache The ADSP-SC595/SC596/SC598 processors also include a traditional instruction cache (I-cache) and two data caches (D-caches, one each for PM/DM) with parity support for all caches. These caches support one instruction access and two data accesses over the DM and PM buses per CCLK cycle. The cache controllers automatically manage the configured L1 memory. The system can configure part of the L1 memory for automatic management by the cache controllers. The sizes of Figure 3. SHARC+ SIMD Core Block Diagram S SIMD Core 11-STAGE PROGRAM SEQUENCER PM ADDRESS 32 PM DATA 64 DM DATA 64 DM ADDRESS 32 DAG1 16 × 32 MRF 80-BIT ALU MULTIPLIER SHIFTER PEx DATA REGISTER Rx 16 × 40-BIT DMD/PMD 64 ASTATx STYKx ASTATy STYKy PEy DATA REGISTER Sx 16 × 40-BIT MRB 80-BIT MSB 80-BIT MSF 80-BIT DAG2 16 × 32 ALU MULTIPLIER SHIFTER DATA SWAP SYSTEM I/F USTAT PX PM ADDRESS 24 PM DATA 48 DEBUG TRACE FLAGS CEC BTB BP CONFLICT CACHE TO IMIF |
|
링크 URL |
| ALLDATASHEET 가 귀하에 도움이 되셨나요? [ DONATE ] |
Alldatasheet는? | 광고문의 | 운영자에게 연락하기 | 개인정보취급방침 | 링크 투 데이터시트 | 링크교환 | 제조사별 검색 All Rights Reserved©Alldatasheet.com |
| Russian : Alldatasheetru.com | Korean : Alldatasheet.co.kr | Spanish : Alldatasheet.es | French : Alldatasheet.fr | Italian : Alldatasheetit.com Portuguese : Alldatasheetpt.com | Polish : Alldatasheet.pl | Vietnamese : Alldatasheet.vn Indian : Alldatasheet.in | Mexican : Alldatasheet.com.mx | British : Alldatasheet.co.uk | New Zealand : Alldatasheet.co.nz |
|
Family Site : ic2ic.com |
icmetro.com |