L1I_PREFETCHES

Provides information about the number of issued L1 cache

 

 

line prefetch requests (64 bytes/line). The reported number

 

 

includes streaming and non-streaming prefetches. Hits and

 

 

misses in L1 instruction cache are both included.

L1

Instruction Cache Read Misses

Number of L1 instruction cache read misses.

L1

Instruction Cache Demand Miss

Percentage of demand fetch reads that missed.

Percentage

 

Total L1 Instruction Cache References

Sum of demand fetch reads and L1 cache line prefetch

 

 

requests.

Metrics for Integrity Servers Dual-Core Itanium 2 and Itanium 9300 Quad-Core Processor Systems

BACK_END_BUBBLE.ALL

Full Pipe Bubbles in Main Pipe due to all causes.

 

This is the number of cycles lost (stall cycles) due to any of

 

five possible events (FPU/L1D, RSE, EXE, branch/exception,

 

or the front-end).

BACK_END_BUBBLE.FE

Full Pipe Bubbles in Main Pipe due to front end.

 

This is the number of cycles lost (stall cycles) due to

 

instruction cache ITLB and branch execution stalls.

CPU_OP_CYCLES.ALL

IA64_INST_RETIRED

L1I_PREFETCHES

L1I_READS

L2I_DEMAND_READS

L2I_PREFETCHES

L2I_READS.ALL.ALL

L2I_READS.MISS.ALL

L2I_READS.MISS.DMND

%of Cycles lost due to all stalls (lower is better)

%of Cycles lost due to Front end stalls (ICACHE, ITLB and branch execution)

L1 instruction cache references

L1 instruction cache misses

L1 instruction cache miss percentage

Number of elapsed CPU operating cycles.

Number of retired IA–64 instructions. The count includes predicated on and predicated off instructions, and nops, but excludes hardware-inserted RSE operations.

Number of issued L1 cache line prefetch requests (64 bytes/line). For more information, see L1I_PREFETCHES.

Number of demand fetch reads to the L1 instruction cache (32-byte chunks). For more information, see L1I_READS .

Number of instruction requests to L2 instruction cache due to L1 instruction cache demand fetch misses. This includes the number of demand fetches that miss both the L1 instruction cache and the ISB, regardless of whether they hit or miss in the RAB.

Number of prefetch requests issued to the L2 instruction cache. This includes streaming and non-streaming prefetches.

Number of cacheable code reads handled by the L2 instruction cache.

Number of demand fetch and prefetch misses in cacheable code reads handled by the L2 instruction cache. This only includes the primary misses.

Number of demand fetch misses in cacheable code reads handled by the L2 instruction cache. This includes the primary misses.

Percentage of cycles lost due to all stalls.

Percentage of cycles lost due to instruction cache, instruction translation lookaside buffer, and branch execution stalls.

Number of L1 instruction cache references.

Number of L1 instruction cache prefetch and demand fetch misses.

Percentage of L1 instruction demand fetch and prefetch that are misses.

icache Measurement Report Description 205