Software Tuning, Performance Optimization & Platform Monitoring
Discussion around monitoring and software tuning methodologies, Performance Monitoring Unit (PMU) of Intel microprocessors, and platform monitoring
Announcements
This community is designed for sharing of public information. Please do not share Intel or third-party confidential information here.

Problems with Top-down Microarchitecture Analysis Method

Xueqi_L_
Beginner
124 Views

I'm trying to analyse an execution on an Intel IvyBridge CPU (Xeon E5-2620 V2) with the Top-down Microarchitecture Analysis Method, described in Chapters B of the Intel® 64 and IA-32 Architectures Optimization Reference Manual

Therefore I perform performance counter events measurements with Perf (perf stat -e rNNN), There are some results I don’t understand:

  1. CPU_UNHALTED.TREAD_P(0x003C) < CYCLE_ACTIVITY.STALLS_LDM_PENDING(Event A3H, Umask 06H)

       I can not understand which it means?

  1. CYCLE_ACTIVITY.CYCLES_L2_PENDING > CYCLE_ACTIVITY.CYCLES_L1D_PENDING and CYCLE_ACTIVITY.STALLS_L2_PENDING > CYCLE_ACTIVITY.STALLS_L1D_PENDING
0 Kudos
0 Replies
Reply