
6-24
MPC8240 Integrated Processor User's Manual
MOTOROLA
FPM or EDO DRAM Interface Operation
6.3.8.1 RMW Parity Latency Considerations
When RMW parity is enabled, the time required to read, modify, and write increases
latency for some transactions.
For local processor single-beat writes to system memory, the MPC8240 latches the data,
performs a double-word read from system memory (checking parity), and then merges the
write data from the processor with the data read from memory. The MPC8240 then
generates new parity bits for the merged double word and writes the data and parity to
memory. The read-modify-write process adds six clock cycles to a single-beat write
operation. If page-mode retention is enabled (PGMAX > 0), then the MPC8240 keeps the
memory in page-mode for the read-modify-write sequence. Figure 6-16 shows
FPM or EDO timing for a local processor single-beat write operation with RMW parity
enabled.
For PCI writes to system memory with RMW parity enabled, the MPC8240 latches the data
in the internal PCI-to-system-memory-write buffer (PCMWB). If the PCI master writes
complete double words to system memory, the MPC8240 generates the parity bits when the
PCMWB is ushed to memory. However, if the PCI master writes 32-, 16-, or 8-bit data
that cannot be gathered into a complete double word in the PCMWB, a read-modify-write
operation is required. The MPC8240 performs a double-word read from system memory
(checking parity), and then merges the write data from the PCI master with the data read
from memory. The MPC8240 then generates new parity for the merged double word and
writes the data and parity to memory. If page mode retention is enabled (PGMAX > 0), the
MPC8240 keeps the memory in page mode for the read-modify-write sequence.
Because the local processor drives all eight parity bits during burst writes to system
memory, these transactions go directly to the DRAMs with no performance penalty. All
other transactions are unaffected and operate as in normal parity mode.
6.3.9 FPM or EDO ECC
As an alternative to simple parity, the MPC8240 supports ECC for the data-path between
the MPC8240 and system memory. ECC not only allows the MPC8240 to detect errors in
the memory data-path but also to correct single-bit errors in the 64-bit data-path. The ECC
logic in the MPC8240 detects and corrects all single-bit errors and detects all double-bit
errors and all errors within a nibble. Other errors may be detected but are not guaranteed to
be either detected or corrected. Multiple-bit errors are always reported when detected.
However, when a single-bit error occurs, the single-bit error counter register is
incremented, and its value is compared to the single-bit error trigger register. If the values
are not equal, no error is reported; if the values are equal, then an error is reported. Thus,
the single-bit error registers may be programmed so that minor faults with memory are
corrected and ignored, but a catastrophic memory failure generates an interrupt. The
syndrome equations for the ECC code are shown in Table 6-10. For supported
conTgurations, see Table 6-4.
The MPC8240 supports concurrent ECC for the FPM or EDO data-path and parity for the