参数资料
型号: IBM21P100BGC
元件分类: 总线控制器
英文描述: PCI BUS CONTROLLER, PBGA304
封装: 31 X 31 MM, PLASTIC, BGA-304
文件页数: 64/144页
文件大小: 5197K
代理商: IBM21P100BGC
IBM21P100BGC
IBM 133 PCI-X Bridge R2.0
Bus Operation
Page 26 of 144
ppb20_operations.fm.01
October 15, 2001
The bridge translates a PCI-X memory read block command into one of three conventional PCI memory read
commands based on the byte count and starting address. If the starting address and byte count are such that
only a single DWord (or less) is being read, the conventional transaction uses the memory read command. If
the PCI-X transaction reads more than one DWord, but does not cross a cache line boundary (as indicated by
the Cache Line Size register in the conventional Configuration Space header), the conventional transaction
uses the memory read line command. If the PCI-X transaction crosses a cache line boundary, the conven-
tional transaction uses the memory read multiple command.
If a disconnect occurs before the byte count of the PCI-X memory read block command is exhausted, the
bridge continues to issue the command until all the bytes in the count are received. The bridge disconnects
once the buffer is filled and prefetches more data as 128-byte sectors of the buffer become free when split
completion data is returned to the originator, until the byte count is exhausted.
3.3.1.3 PCI to PCI Transactions
This mode does not involve any translation.
If the memory read command targets non-prefetchable memory space, the memory read fetches only the
requested double word. Bits 9:8 of the primary and secondary data buffering control registers control the
prefetching algorithm for the memory read command in prefetchable space. The default value of these bits
indicates that up to one cache line will be prefetched.
For the memory read line command, the prefetching algorithm is controlled by bits 7:6 of the primary and
secondary data buffering control registers. The default value of these bits indicates that up to one cache line
will be prefetched.
For the memory read multiple command, the prefetching algorithm is controlled by bits 5:4 of the primary and
secondary data buffering control registers. The default value of these bits indicates that a full prefetch will be
done, subject to the limit imposed by the maximum memory read byte count value set by bits (14:12) of the
same register. The default value is 512 bytes or an entire read buffer. Data fetching operations will be discon-
nected at all 1 MB boundaries.
3.3.1.4 PCI-X to PCI-X Transactions
This mode does not involve any translation.
The amount of data that is fetched is controlled by the downstream and upstream split transaction control
register. The split transaction capacity and split transaction commitment limit fields control how much data is
requested at any one time.
3.3.2 I/O Read
The I/O Read command is not translated and fetches a DWord of data. The command will either be split in the
PCI-X mode or delayed in the conventional PCI mode.
3.3.3 Configuration Read
3.3.3.1 Type 1 Configuration Read
The Type 1 configuration read command is only accepted on the primary interface. The command will either
be split in the PCI-X mode or delayed in the conventional PCI mode.
相关PDF资料
PDF描述
IBM25403GCX-3JC76C2 RISC PROCESSOR, PQFP16
IBM25405GP-3BA200C2 RISC PROCESSOR, PBGA456
IBM25EMPPC603EFG-100 32-BIT, 100 MHz, RISC PROCESSOR, PQFP240
IBM25EMPPC603EBG-100 32-BIT, 100 MHz, RISC PROCESSOR, CBGA255
IBM25EMPPC740LDBC4000 32-BIT, 400 MHz, RISC PROCESSOR, CBGA255
相关代理商/技术参数
参数描述
IBM24L5086 制造商:AVED MEMORY PRODUCTS 功能描述: 制造商:AVED Memory Products 功能描述:
IBM25403GCX-3BC80C2 制造商:IBM 功能描述:RISC PROCESSOR, 160 Pin Plastic BGA
IBM25403GCX-3JC50C2 制造商:IBM 功能描述:403GCX-3JC50C2
IBM25403GCX-3JC66C2 制造商:IBM 功能描述:
IBM25403GCX3JC76C2 制造商:IBM 功能描述: