Intel ARCHITECTURE IA-32 - Manual (Page 494)

IA-32 Intel® Architecture Optimization
B-40
Reads
Non-prefetch
Full (BSQ) The number of read
(excludes RFOs and
HW|SW prefetches)
transactions to
WB-type memory.
Beware of granularity
issues with this
event. BSQ_allocation 1. REQ_LEN0 |
REQ_LEN1 |
MEM_TYPE1 |
MEM_TYPE2|
REQ_CACHE_TYPE|
REQ_DEM_TYPE
2. Enable edge
filtering 6 in
the CCCR.
Reads
Invalidate Full-
RFO (BSQ) The number of read
invalidate (RFO)
transactions to
WB-type memory. BSQ_allocation 1. REQ_TYPE0 |
REQ_LEN0 |
REQ_LEN1 |
MEM_TYPE1 |
MEM_TYPE2|
REQ_CACHE_TYPE|
REQ_ORD_TYPE|
REQ_DEM_TYPE
2. Enable edge
filtering 6 in
the CCCR.
UC Reads
Chunk (BSQ) The number of
8-byte aligned UC
read transactions.
User note: Read
requests associated
with 16 byte
operands may
under-count. BSQ_allocation 1. REQ_LEN0|
REQ_ORD_TYPE |
REQ_DEM_TYPE
2. Enable edge
filtering 6 in
the CCCR.
UC Reads
Chunk Split
(BSQ) The number of UC
read transactions
that span an 8-byte
boundary. User note:
Read requests may
under-count if the
data chunk straddles
64-byte boundary. BSQ_allocation 1. REQ_LEN0|
REQ_SPLIT_TYPE |
REQ_ORD_TYPE |
REQ_DEM_TYPE
2. Enable edge
filtering 6 in
the CCCR.
continued
Table B-1 Pentium 4 Processor Performance Metrics (continued)
Metric Description
Event Name or Metric
Expression Event Mask Value
Required