Fujitsu SPARC JPS1 Implementation Supplement Manual page 171

Difference Between Error Marking on SPARC64 IV and
SPARC64 V
TABLE P-7
SPARC64 V.
Error Marking on SPARC64 IV and SPARC64 V
TABLE P-7
ECC for cacheable data
Trigger of error marking The detection of a raw UE
ERROR_MARK_ID value
Target data of error
marking
Note: (5) is different
The extent of replaced
data at error marking
Error marking on SPARC64 IV and SPARC64 V differs in two ways:
On SPARC64 V, only the doubleword with raw
On SPARC64 IV, the quadword containing the doubleword with raw
replaced with two copies of the ERROR_MARK_ID.
On SPARC64 V, error marking is not applied to incoming interrupt packet data.
On SPARC64 IV, error marking is applied even for incoming interrupt packet
data.
160 SPARC JPS1 Implementation Supplement: Fujitsu SPARC64 V • Release 1.0, 1 July 2002
lists the differences between error marking on SPARC64 IV and
SPARC64 IV
ECC for UPA
Value specified in
TABLE P-6
(1) D1 cache data
(2) U2 cache data
(3) Incoming cacheable data from UPA
(4) Outgoing cacheable data to UPA for
writeback or copyback
(5) Incoming interrupt packet data
from UPA
The quadword (16-byte data on 16-byte
boundary) containing the doubleword
with raw and its two ECCs are
UE
replaced.
The doubleword and ECC specified in
are written to each of the
TABLE P-4
two doublewords in the quadword.
SPARC64 V
ECC for UPA
The detection of a raw
. Value specified in
(1)–(4) as described for SPARC IV
(5) is not applied. For the incoming interrupt
packet data, error marking is not applied
and the incoming data and ECC are directly
set to ASI_INTR_DATA 0:7_R and its ECC
register.
Only the doubleword with the raw
replaced, as specified in
is replaced at error marking.
UE
UE
.
TABLE P-6
is
UE
.
TABLE P-4
is
UE