| 数据搜索系统,热门电子元器件搜索 |
|
AM486DX2 数据表(PDF) 32 Page - Advanced Micro Devices |
|
|
|||||||||||||||||||||||||||||
AM486DX2 数据表(HTML) 32 Page - Advanced Micro Devices |
|
32 / 67 page ![]() 32 Am5X86 Microprocessor AMD PRELIMINARY 4.8.7.2 BOFF Write-Back Arbitration Implementation The use of BOFF to perform snooping of the on-chip cache is used in systems where more than one cache- able bus master resides on the microprocessor bus. The BOFF signal forces the microprocessor to relinquish the bus in the following clock cycle, regardless of the type of bus cycle it was performing at the time. Consequently, the use of BOFF as a bus arbitrator should be imple- mented with care to avoid system problems. 4.8.8 BOFF Design Considerations The use of BOFF as a bus arbitration control mechanism is immediate. BOFF forces the microprocessor to abort an access in the following clock cycle after it is asserted. The following design issues must be considered. 4.8.8.1 Cache Line Fills The microprocessor aborts a cache line fill during a burst read if BOFF is asserted during the access. Upon re- gaining the bus, the read access commences where it left off when BOFF was recognized. External buffers should take this cycle continuation into consideration if BOFF is allowed to abort burst read cycles. 4.8.8.2 Cache Line Copy-Backs Similar to the burst read, the burst write also can be aborted at any time with the BOFF signal. Upon regain- ing access to the bus, the write continues from where it was aborted. External buffers and control logic should take into consideration the necessary control, if any, for burst write continuations. 4.8.8.3 Locked Accesses Locked bus cycles occur in various forms. Locked ac- cesses occur during read-modify-write operations, in- terrupt acknowledges, and page table updates. Although asserting BOFF during a locked cycle is per- mitted, extreme care should be taken to ensure data coherency for semaphore updates and proper data or- dering. 4.8.9 BOFF During Write-Back If BOFF is asserted during a write-back, the processor performing the write-back goes off the bus in the next clock cycle. If BOFF is released, the processor restarts that write-back access from the point at which it was aborted. The behavior is identical to the normal BOFF case that includes the abort and restart behavior. 4.8.10 Snooping Characteristics During a Cache Line Fill The microprocessor takes responsibility for responding to snoop cycles for a cache line only during the time that the line is actually in the cache or in a copy-back buffer. There are times during the cache line fill cycle and during the cache replacement cycle when the line is “in transit” and snooping responsibility must be taken by other sys- tem components. The following cases apply if snooping is invoked via AHOLD, and neither HOLD nor BOFF is asserted. s System designers should consider the possibility that a snooping cycle may arrive at the same time as a cache line fill or replacement for the same ad- dress. If a snooping cycle arrives at the same time as a cache line fill with the same address, the CPU uses the cache line fill, but does not place it in the cache. s If a snooping cycle occurs at the same time as a cache line fill with a different address, the cache line fill is placed into the cache unless EADS is recog- nized before the first BRDY but after ADS is assert- ed, or EADS is recognized on the last BRDY of the cache line fill. In these cases, the line is not placed into the cache. 4.8.11 Snooping Characteristics During a Copy-Back If a copy-back is occurring because of a cache line re- placement, the address being replaced can be matched by a snoop until assertion of the last BRDY of the copy- back. This is when the modified line resides in the copy- back buffer. An EADS as late as two clocks before the last BRDY can cause HITM to be asserted. Figure 15 illustrates the microprocessor relinquishing responsibility of recognizing snoops for a line that is copied back. It shows the latest EADS assertion that can cause HITM assertion. HITM remains active for only one clock period in that example. HITM remains active through the last BRDY of the corresponding write-back; in that case, the write-back has already completed. This is the latest point where snooping can start, because two clock cycles later, the final BRDY of the write-back is applied. If a snoop cycle hits the copy-back address after the first BRDY of the copy-back and ADS has been issued, the microprocessor asserts HITM. Keep in mind that the write-back was initiated due to a read miss and not due to a snoop to a modified line. In the second case, no snooping is recognized if a modified line is detected. |
|
链接网址 |
| ALLDATASHEET是否为您带来帮助? [ DONATE ] |
关于 Alldatasheet | 广告服务 | 联系我们 | 隐私政策 | 数据表链接 | 链接交换 | 制造商名单 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 |