Skip to content

Latest commit

 

History

History
114 lines (74 loc) · 7.53 KB

README.md

File metadata and controls

114 lines (74 loc) · 7.53 KB
Raw

Requirements

  • Linux kernel version 5.0.0 or higher
  • Python 3.x
  • Python's scipy module installed with "pip3 install scipy" command
  • Python's termcolor module installed with "pip3 install termcolor" command
  • To run perf_event_open system call without having to use sudo access, set the value of perf_event_paranoid to -1 by typing the following command: sudo sysctl -w kernel.perf_event_paranoid=-1

Installation

To install ReuseTracker and all of its dependences, type "make install".

Reproducing Results presented in The Paper

  1. To reproduce the results presented in the "Accuracy without Invalidation" and "Accuracy with Invalidation" subsections in the paper, ensure that you have enough maximum stack size by typing "ulimit -s 4194304" and run the experiment by typing "make accuracy". The resulting histograms can be found in each *_output folder with file name reuse-invalidation-PID.reuse.hpcrun.
  2. To reproduce the results presented in the "Reuse Distances of PARSEC Benchmarks" section, type "make parsec". The resulting histograms are in the following files.
    • parsec_experiment/blackscholes_rd_l1_output/blackscholes-PID.reuse.hpcrun.bin
    • parsec_experiment/blackscholes_rd_spatial_l1_output/blackscholes-PID.reuse.hpcrun.bin
    • parsec_experiment/bodytrack_rd_l1_output/bodytrack-PID.reuse.hpcrun.bin
    • parsec_experiment/bodytrack_rd_spatial_l1_output/bodytrack-PID.reuse.hpcrun.bin
    • parsec_experiment/streamcluster_rd_l3_output/streamcluster-PID.reuse.hpcrun.bin
      • parsec_experiment/streamcluster_rd_spatial_l1_output/streamcluster-PID.reuse.hpcrun.bin
      • parsec_experiment/freqmine_rd_l1_output/freqmine-PID.reuse.hpcrun.bin
      • parsec_experiment/freqmine_rd_spatial_l1_output/freqmine-PID.reuse.hpcrun.bin

Usage on Intel

  • To profile the private cache temporal reuse distance of an application, run the following command.

HPCRUN_WP_REUSE_PROFILE_TYPE="TEMPORAL" HPCRUN_PROFILE_L3=false HPCRUN_WP_REUSE_BIN_SCHEME=4000,2 HPCRUN_WP_CACHELINE_INVALIDATION=true HPCRUN_WP_DONT_FIX_IP=true HPCRUN_WP_DONT_DISASSEMBLE_TRIGGER_ADDRESS=true reusetracker-bin/bin/hpcrun -e WP_REUSETRACKER -e MEM_UOPS_RETIRED:ALL_LOADS@100000 -e MEM_UOPS_RETIRED:ALL_STORES@100000 <./your_executable> your_args

  • To profile the private cache spatial reuse distance of an application, run the following command.

HPCRUN_WP_REUSE_PROFILE_TYPE="SPATIAL" HPCRUN_PROFILE_L3=false HPCRUN_WP_REUSE_BIN_SCHEME=4000,2 HPCRUN_WP_CACHELINE_INVALIDATION=true HPCRUN_WP_DONT_FIX_IP=true HPCRUN_WP_DONT_DISASSEMBLE_TRIGGER_ADDRESS=true reusetracker-bin/bin/hpcrun -e WP_REUSETRACKER -e MEM_UOPS_RETIRED:ALL_LOADS@100000 -e MEM_UOPS_RETIRED:ALL_STORES@100000 <./your_executable> your_args

  • To profile the shared cache temporal reuse distance of an application, run the following command.

HPCRUN_WP_REUSE_PROFILE_TYPE="TEMPORAL" HPCRUN_PROFILE_L3=true HPCRUN_WP_REUSE_BIN_SCHEME=4000,2 HPCRUN_WP_CACHELINE_INVALIDATION=true HPCRUN_WP_DONT_FIX_IP=true HPCRUN_WP_DONT_DISASSEMBLE_TRIGGER_ADDRESS=true reusetracker-bin/bin/hpcrun -e WP_REUSETRACKER -e MEM_UOPS_RETIRED:ALL_LOADS@100000 -e MEM_UOPS_RETIRED:ALL_STORES@100000 <./your_executable> your_args

  • To profile the shared cache spatial reuse distance of an application, run the following command.

HPCRUN_WP_REUSE_PROFILE_TYPE="SPATIAL" HPCRUN_PROFILE_L3=true HPCRUN_WP_REUSE_BIN_SCHEME=4000,2 HPCRUN_WP_CACHELINE_INVALIDATION=true HPCRUN_WP_DONT_FIX_IP=true HPCRUN_WP_DONT_DISASSEMBLE_TRIGGER_ADDRESS=true reusetracker-bin/bin/hpcrun -e WP_REUSETRACKER -e MEM_UOPS_RETIRED:ALL_LOADS@100000 -e MEM_UOPS_RETIRED:ALL_STORES@100000 <./your_executable> your_args

Usage on AMD

  • To profile the private cache temporal reuse distance of an application, run the following command.

HPCRUN_WP_REUSE_PROFILE_TYPE="TEMPORAL" HPCRUN_PROFILE_L3=false HPCRUN_WP_REUSE_BIN_SCHEME=4000,2 HPCRUN_WP_CACHELINE_INVALIDATION=true HPCRUN_WP_DONT_FIX_IP=true HPCRUN_WP_DONT_DISASSEMBLE_TRIGGER_ADDRESS=true reusetracker-bin/bin/hpcrun -e WP_AMD_REUSETRACKER -e IBS_OP@100000 -e AMD_L1_DATA_ACCESS@100000000 <./your_executable> your_args

  • To profile the private cache spatial reuse distance of an application, run the following command.

HPCRUN_WP_REUSE_PROFILE_TYPE="SPATIAL" HPCRUN_PROFILE_L3=false HPCRUN_WP_REUSE_BIN_SCHEME=4000,2 HPCRUN_WP_CACHELINE_INVALIDATION=true HPCRUN_WP_DONT_FIX_IP=true HPCRUN_WP_DONT_DISASSEMBLE_TRIGGER_ADDRESS=true reusetracker-bin/bin/hpcrun -e WP_AMD_REUSETRACKER -e IBS_OP@100000 -e AMD_L1_DATA_ACCESS@100000000 <./your_executable> your_args

  • To profile the shared cache temporal reuse distance of an application, run the following command.

HPCRUN_WP_REUSE_PROFILE_TYPE="TEMPORAL" HPCRUN_PROFILE_L3=true HPCRUN_WP_REUSE_BIN_SCHEME=4000,2 HPCRUN_WP_CACHELINE_INVALIDATION=true HPCRUN_WP_DONT_FIX_IP=true HPCRUN_WP_DONT_DISASSEMBLE_TRIGGER_ADDRESS=true reusetracker-bin/bin/hpcrun -e WP_AMD_REUSETRACKER -e IBS_OP@100000 -e AMD_L1_DATA_ACCESS@100000000 <./your_executable> your_args

  • To profile the shared cache spatial reuse distance of an application, run the following command.

HPCRUN_WP_REUSE_PROFILE_TYPE="SPATIAL" HPCRUN_PROFILE_L3=true HPCRUN_WP_REUSE_BIN_SCHEME=4000,2 HPCRUN_WP_CACHELINE_INVALIDATION=true HPCRUN_WP_DONT_FIX_IP=true HPCRUN_WP_DONT_DISASSEMBLE_TRIGGER_ADDRESS=true reusetracker-bin/bin/hpcrun -e WP_AMD_REUSETRACKER -e IBS_OP@100000 -e AMD_L1_DATA_ACCESS@100000000 <./your_executable> your_args

Attribution to Locations in Source Code

To attribute the detected uses, reuses and the total of reuse distances captured between each use-reuse pair to their locations in source code lines and program stacks, you need to take the following steps:

  1. Compile the code that you want to profile using "-g" flag to allow for debugging.

  2. To attribute the detected communications to their locations in source code lines and program stacks, you need to take the following steps:

a. Download and extract a binary release of hpcviewer from http://hpctoolkit.org/download/hpcviewer/latest/hpcviewer-linux.gtk.x86_64.tgz

b. Run ReuseTracker on a program to be profiled

In Intel:

HPCRUN_WP_REUSE_PROFILE_TYPE="TEMPORAL" HPCRUN_PROFILE_L3=false HPCRUN_WP_REUSE_BIN_SCHEME=4000,2 HPCRUN_WP_CACHELINE_INVALIDATION=true HPCRUN_WP_DONT_FIX_IP=true HPCRUN_WP_DONT_DISASSEMBLE_TRIGGER_ADDRESS=true reusetracker-bin/bin/hpcrun -e WP_REUSETRACKER -e MEM_UOPS_RETIRED:ALL_LOADS@100000 -e MEM_UOPS_RETIRED:ALL_STORES@100000 <./your_executable> your_args

In AMD:

HPCRUN_WP_REUSE_PROFILE_TYPE="TEMPORAL" HPCRUN_PROFILE_L3=false HPCRUN_WP_REUSE_BIN_SCHEME=4000,2 HPCRUN_WP_CACHELINE_INVALIDATION=true HPCRUN_WP_DONT_FIX_IP=true HPCRUN_WP_DONT_DISASSEMBLE_TRIGGER_ADDRESS=true reusetracker-bin/bin/hpcrun -e WP_AMD_REUSETRACKER -e IBS_OP@100000 -e AMD_L1_DATA_ACCESS@100000000 <./your_executable> your_args

c. Extract the static program structure from the profiled program by using hpcstruct

reusetracker-bin/bin/hpcstruct <./your_executable>

The output of hpcstruct is <./your_executable>.hpcstruct.

d. Generate an experiment result database using hpcprof

reusetracker-bin/bin/hpcprof -S <./your_executable>.hpcstruct -o

The output of hpcprof is a folder named .

e. Use hpcviewer to read the content of the experiment result database in a GUI interface

hpcviewer/hpcviewer

Information on program stack and source code lines is available in the Scope column, and the other columns display numbers of detected reuses, total time distance, and total reuse distance that correspond to the source code lines.