Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

increase number of threads in benchmarks #1166

Merged
merged 1 commit into from
Mar 18, 2025
Merged

increase number of threads in benchmarks #1166

merged 1 commit into from
Mar 18, 2025

Conversation

lplewa
Copy link
Contributor

@lplewa lplewa commented Mar 6, 2025

Description

Checklist

  • Code compiles without errors locally
  • All tests pass locally
  • CI workflows execute properly
  • CI workflows, not executed per PR (e.g. Nightly), execute properly
  • New tests added, especially if they will fail without my changes
  • Added/extended example(s) to cover this functionality
  • Extended the README/documentation
  • All newly added source files have a license
  • All newly added source files are referenced in CMake files
  • Logger (with debug/info/... messages) is used
  • All API changes are reflected in docs and def/map files, and are tested

@github-actions GitHub Actions
Copy link

github-actions bot commented Mar 6, 2025

Compute Benchmarks run (with params: ):
https://github.com/oneapi-src/unified-memory-framework/actions/runs/13699381103

@github-actions GitHub Actions
Copy link

github-actions bot commented Mar 6, 2025

Compute Benchmarks run ():
https://github.com/oneapi-src/unified-memory-framework/actions/runs/13699381103
Job status: success. Test status: success.

Summary

(Emphasized values are the best results)

Regressed 1 (threshold 2.00%)
Benchmark This PR baseline Change
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:1 umfProxy 727.582 ns 699.655000 ns -3.84%

Performance change in benchmark groups

UMF
Relative perf in group multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:12 (3)
Benchmark This PR baseline Change
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:12 umfProxy 17025.100000 ns -
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:12 jemalloc 751.344000 ns -
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:12 tbbProxy 2769.980000 ns -
Relative perf in group multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:1 (11)
Benchmark This PR baseline Change
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:1 umfProxy 1057.030 ns 1050.370000 ns -0.63%
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:1 tbbProxy 498.867 ns 493.991000 ns -0.98%
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:1 jemalloc 73.749 ns 72.919900 ns -1.12%
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:1 glibc - 364.241000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:1 disjoint_pool<os_provider> - 1053.310000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:1 jemalloc_pool<os_provider> - 1654.200000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:1 scalable_pool<os_provider> - 617.139000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:1 proxy_pool<fixed_provider> - 3899.000000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:1 fixed_provider - 2131.040000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:1 disjoint_pool<fixed_provider> - 1123.420000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:1 scalable_pool<fixed_provider> - 602.672000 ns
Relative perf in group multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:12 (3)
Benchmark This PR baseline Change
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:12 umfProxy 15885.000000 ns -
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:12 jemalloc 952.756000 ns -
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:12 tbbProxy 3827.250000 ns -
Relative perf in group multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:1 (9)
Benchmark This PR baseline Change
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:1 jemalloc 155.623000 ns 157.304 ns 1.08%
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:1 umfProxy 1073.710000 ns 1079.910 ns 0.58%
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:1 tbbProxy 555.920 ns 554.097000 ns -0.33%
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:1 glibc - 842.093000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:1 disjoint_pool<os_provider> - 2509.890000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:1 jemalloc_pool<os_provider> - 1130.680000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:1 scalable_pool<os_provider> - 632.484000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:1 disjoint_pool<fixed_provider> - 1758.630000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:1 scalable_pool<fixed_provider> - 635.950000 ns
Relative perf in group multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:12 (3)
Benchmark This PR baseline Change
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:12 umfProxy 14539.300000 ns -
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:12 jemalloc 833.372000 ns -
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:12 tbbProxy 2361.240000 ns -
Relative perf in group multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:1 (9)
Benchmark This PR baseline Change
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:1 tbbProxy 338.451000 ns 338.512 ns 0.02%
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:1 jemalloc 94.765 ns 94.553300 ns -0.22%
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:1 umfProxy 727.582 ns 699.655000 ns -3.84%
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:1 glibc - 177.347000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:1 disjoint_pool<os_provider> - 888.120000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:1 jemalloc_pool<os_provider> - 699.271000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:1 scalable_pool<os_provider> - 380.833000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:1 disjoint_pool<fixed_provider> - 888.670000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:1 scalable_pool<fixed_provider> - 379.256000 ns
Relative perf in group multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:4 (9)
Benchmark This PR baseline Change
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:4 glibc - 2055.620000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:4 disjoint_pool<os_provider> - 39975.000000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:4 jemalloc_pool<os_provider> - 48999.300000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:4 scalable_pool<os_provider> - 1160.690000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:4 disjoint_pool<fixed_provider> - 35462.000000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:4 scalable_pool<fixed_provider> - 1806.580000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:4 umfProxy - 5278.960000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:4 jemalloc - 418.094000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:4 tbbProxy - 1825.310000 ns
Relative perf in group multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:4 (8)
Benchmark This PR baseline Change
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:4 glibc - 1561.100000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:4 disjoint_pool<os_provider> - 26531.100000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:4 jemalloc_pool<os_provider> - 8951.480000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:4 scalable_pool<os_provider> - 2211.190000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:4 scalable_pool<fixed_provider> - 1705.380000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:4 umfProxy - 5001.010000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:4 jemalloc - 493.417000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:4 tbbProxy - 1972.650000 ns
Relative perf in group multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:4 (8)
Benchmark This PR baseline Change
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:4 glibc - 528.910000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:4 disjoint_pool<os_provider> - 12425.400000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:4 jemalloc_pool<os_provider> - 9020.480000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:4 scalable_pool<os_provider> - 1776.320000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:4 scalable_pool<fixed_provider> - 1392.420000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:4 umfProxy - 4445.350000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:4 jemalloc - 444.436000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:4 tbbProxy - 1342.930000 ns
Relative perf in group multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:50000/threads:1 (2)
Benchmark This PR baseline Change
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:50000/threads:1 proxy_pool<os_provider> - 11589.300000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:50000/threads:1 os_provider - 9974.260000 ns

Details

Benchmark details - environment, command...
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:12 umfProxy

Command:

/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/benchmark/umf-benchmark --benchmark_format=csv --benchmark_filter=glibc

Environment Variables:

LD_PRELOAD=/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/lib/libumf_proxy.so

multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:1 umfProxy

Command:

/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/benchmark/umf-benchmark --benchmark_format=csv --benchmark_filter=glibc

Environment Variables:

LD_PRELOAD=/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/lib/libumf_proxy.so

multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:12 umfProxy

Command:

/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/benchmark/umf-benchmark --benchmark_format=csv --benchmark_filter=glibc

Environment Variables:

LD_PRELOAD=/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/lib/libumf_proxy.so

multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:1 umfProxy

Command:

/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/benchmark/umf-benchmark --benchmark_format=csv --benchmark_filter=glibc

Environment Variables:

LD_PRELOAD=/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/lib/libumf_proxy.so

multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:12 umfProxy

Command:

/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/benchmark/umf-benchmark --benchmark_format=csv --benchmark_filter=glibc

Environment Variables:

LD_PRELOAD=/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/lib/libumf_proxy.so

multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:1 umfProxy

Command:

/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/benchmark/umf-benchmark --benchmark_format=csv --benchmark_filter=glibc

Environment Variables:

LD_PRELOAD=/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/lib/libumf_proxy.so

multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:12 jemalloc

Command:

/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/benchmark/umf-benchmark --benchmark_format=csv --benchmark_filter=glibc

Environment Variables:

LD_PRELOAD=libjemalloc.so

multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:1 jemalloc

Command:

/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/benchmark/umf-benchmark --benchmark_format=csv --benchmark_filter=glibc

Environment Variables:

LD_PRELOAD=libjemalloc.so

multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:12 jemalloc

Command:

/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/benchmark/umf-benchmark --benchmark_format=csv --benchmark_filter=glibc

Environment Variables:

LD_PRELOAD=libjemalloc.so

multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:1 jemalloc

Command:

/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/benchmark/umf-benchmark --benchmark_format=csv --benchmark_filter=glibc

Environment Variables:

LD_PRELOAD=libjemalloc.so

multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:12 jemalloc

Command:

/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/benchmark/umf-benchmark --benchmark_format=csv --benchmark_filter=glibc

Environment Variables:

LD_PRELOAD=libjemalloc.so

multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:1 jemalloc

Command:

/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/benchmark/umf-benchmark --benchmark_format=csv --benchmark_filter=glibc

Environment Variables:

LD_PRELOAD=libjemalloc.so

multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:12 tbbProxy

Command:

/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/benchmark/umf-benchmark --benchmark_format=csv --benchmark_filter=glibc

Environment Variables:

LD_PRELOAD=libtbbmalloc_proxy.so

multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:1 tbbProxy

Command:

/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/benchmark/umf-benchmark --benchmark_format=csv --benchmark_filter=glibc

Environment Variables:

LD_PRELOAD=libtbbmalloc_proxy.so

multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:12 tbbProxy

Command:

/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/benchmark/umf-benchmark --benchmark_format=csv --benchmark_filter=glibc

Environment Variables:

LD_PRELOAD=libtbbmalloc_proxy.so

multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:1 tbbProxy

Command:

/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/benchmark/umf-benchmark --benchmark_format=csv --benchmark_filter=glibc

Environment Variables:

LD_PRELOAD=libtbbmalloc_proxy.so

multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:12 tbbProxy

Command:

/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/benchmark/umf-benchmark --benchmark_format=csv --benchmark_filter=glibc

Environment Variables:

LD_PRELOAD=libtbbmalloc_proxy.so

multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:1 tbbProxy

Command:

/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/benchmark/umf-benchmark --benchmark_format=csv --benchmark_filter=glibc

Environment Variables:

LD_PRELOAD=libtbbmalloc_proxy.so

@PatKamin PatKamin self-requested a review March 7, 2025 07:46
@github-actions GitHub Actions
Copy link

Compute Benchmarks run (with params: --timeout 1000):
https://github.com/oneapi-src/unified-memory-framework/actions/runs/13792334137

@github-actions GitHub Actions
Copy link

Compute Benchmarks run (--timeout 1000):
https://github.com/oneapi-src/unified-memory-framework/actions/runs/13792334137
Job status: failure. Test status: skipped.

@github-actions GitHub Actions
Copy link

Compute Benchmarks run (with params: --timeout 1000):
https://github.com/oneapi-src/unified-memory-framework/actions/runs/13793461044

@github-actions GitHub Actions
Copy link

Compute Benchmarks run (--timeout 1000):
https://github.com/oneapi-src/unified-memory-framework/actions/runs/13793461044
Job status: success. Test status: success.

Summary

(Emphasized values are the best results)

Improved 1 (threshold 2.00%)
Benchmark This PR baseline Change
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:1 jemalloc 61.976800 ns 94.234 ns 52.05%
Regressed 7 (threshold 2.00%)
Benchmark This PR baseline Change
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:1 umfProxy 3002.660 ns 737.609000 ns -75.43%
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:1 umfProxy 3156.380 ns 1057.290000 ns -66.50%
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:1 umfProxy 3167.010 ns 1084.960000 ns -65.74%
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:1 jemalloc 199.953 ns 154.769000 ns -22.60%
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:1 jemalloc 80.535 ns 72.616100 ns -9.83%
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:1 tbbProxy 521.332 ns 493.090000 ns -5.42%
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:1 tbbProxy 577.735 ns 553.380000 ns -4.22%

Performance change in benchmark groups

UMF
Relative perf in group multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:12 (3)
Benchmark This PR baseline Change
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:12 umfProxy 85163.700000 ns -
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:12 jemalloc 938.864000 ns -
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:12 tbbProxy 3326.550000 ns -
Relative perf in group multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:1 (11)
Benchmark This PR baseline Change
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:1 tbbProxy 521.332 ns 493.090000 ns -5.42%
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:1 jemalloc 80.535 ns 72.616100 ns -9.83%
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:1 umfProxy 3156.380 ns 1057.290000 ns -66.50%
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:1 glibc - 365.935000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:1 disjoint_pool<os_provider> - 983.969000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:1 jemalloc_pool<os_provider> - 1636.530000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:1 scalable_pool<os_provider> - 615.714000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:1 proxy_pool<fixed_provider> - 3816.850000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:1 fixed_provider - 2108.790000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:1 disjoint_pool<fixed_provider> - 1009.500000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:1 scalable_pool<fixed_provider> - 598.478000 ns
Relative perf in group multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:12 (3)
Benchmark This PR baseline Change
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:12 umfProxy 85486.700000 ns -
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:12 jemalloc 1240.420000 ns -
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:12 tbbProxy 2850.600000 ns -
Relative perf in group multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:1 (9)
Benchmark This PR baseline Change
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:1 tbbProxy 577.735 ns 553.380000 ns -4.22%
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:1 jemalloc 199.953 ns 154.769000 ns -22.60%
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:1 umfProxy 3167.010 ns 1084.960000 ns -65.74%
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:1 glibc - 842.621000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:1 disjoint_pool<os_provider> - 2368.080000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:1 jemalloc_pool<os_provider> - 1122.540000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:1 scalable_pool<os_provider> - 632.277000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:1 disjoint_pool<fixed_provider> - 1517.770000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:1 scalable_pool<fixed_provider> - 633.428000 ns
Relative perf in group multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:12 (3)
Benchmark This PR baseline Change
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:12 umfProxy 84430.700000 ns -
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:12 jemalloc 619.139000 ns -
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:12 tbbProxy 1900.070000 ns -
Relative perf in group multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:1 (9)
Benchmark This PR baseline Change
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:1 jemalloc 61.976800 ns 94.234 ns 52.05%
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:1 tbbProxy 343.814 ns 338.568000 ns -1.53%
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:1 umfProxy 3002.660 ns 737.609000 ns -75.43%
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:1 glibc - 177.660000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:1 disjoint_pool<os_provider> - 585.446000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:1 jemalloc_pool<os_provider> - 696.137000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:1 scalable_pool<os_provider> - 378.742000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:1 disjoint_pool<fixed_provider> - 583.510000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:1 scalable_pool<fixed_provider> - 378.047000 ns
Relative perf in group multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:4 (9)
Benchmark This PR baseline Change
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:4 glibc - 1307.920000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:4 disjoint_pool<os_provider> - 33228.300000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:4 jemalloc_pool<os_provider> - 47688.300000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:4 scalable_pool<os_provider> - 1819.090000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:4 disjoint_pool<fixed_provider> - 31103.100000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:4 scalable_pool<fixed_provider> - 1646.290000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:4 umfProxy - 5655.020000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:4 jemalloc - 529.774000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:4 tbbProxy - 1674.320000 ns
Relative perf in group multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:4 (9)
Benchmark This PR baseline Change
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:4 glibc - 1553.730000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:4 disjoint_pool<os_provider> - 24999.700000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:4 jemalloc_pool<os_provider> - 9687.120000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:4 scalable_pool<os_provider> - 1281.010000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:4 disjoint_pool<fixed_provider> - 17286.200000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:4 scalable_pool<fixed_provider> - 1798.970000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:4 umfProxy - 5422.240000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:4 jemalloc - 592.748000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:4 tbbProxy - 1706.820000 ns
Relative perf in group multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:4 (9)
Benchmark This PR baseline Change
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:4 glibc - 525.424000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:4 disjoint_pool<os_provider> - 10014.800000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:4 jemalloc_pool<os_provider> - 8986.560000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:4 scalable_pool<os_provider> - 1011.560000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:4 disjoint_pool<fixed_provider> - 10256.800000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:4 scalable_pool<fixed_provider> - 1362.120000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:4 umfProxy - 5020.300000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:4 jemalloc - 540.544000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:4 tbbProxy - 1234.260000 ns
Relative perf in group multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:50000/threads:1 (2)
Benchmark This PR baseline Change
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:50000/threads:1 proxy_pool<os_provider> - 10974.000000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:50000/threads:1 os_provider - 9355.770000 ns

Details

Benchmark details - environment, command...
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:12 umfProxy

Command:

/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/benchmark/umf-benchmark --benchmark_format=csv --benchmark_filter=glibc

Environment Variables:

LD_PRELOAD=/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/lib/libumf_proxy.so

multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:1 umfProxy

Command:

/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/benchmark/umf-benchmark --benchmark_format=csv --benchmark_filter=glibc

Environment Variables:

LD_PRELOAD=/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/lib/libumf_proxy.so

multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:12 umfProxy

Command:

/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/benchmark/umf-benchmark --benchmark_format=csv --benchmark_filter=glibc

Environment Variables:

LD_PRELOAD=/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/lib/libumf_proxy.so

multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:1 umfProxy

Command:

/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/benchmark/umf-benchmark --benchmark_format=csv --benchmark_filter=glibc

Environment Variables:

LD_PRELOAD=/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/lib/libumf_proxy.so

multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:12 umfProxy

Command:

/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/benchmark/umf-benchmark --benchmark_format=csv --benchmark_filter=glibc

Environment Variables:

LD_PRELOAD=/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/lib/libumf_proxy.so

multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:1 umfProxy

Command:

/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/benchmark/umf-benchmark --benchmark_format=csv --benchmark_filter=glibc

Environment Variables:

LD_PRELOAD=/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/lib/libumf_proxy.so

multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:12 jemalloc

Command:

/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/benchmark/umf-benchmark --benchmark_format=csv --benchmark_filter=glibc

Environment Variables:

LD_PRELOAD=libjemalloc.so

multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:1 jemalloc

Command:

/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/benchmark/umf-benchmark --benchmark_format=csv --benchmark_filter=glibc

Environment Variables:

LD_PRELOAD=libjemalloc.so

multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:12 jemalloc

Command:

/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/benchmark/umf-benchmark --benchmark_format=csv --benchmark_filter=glibc

Environment Variables:

LD_PRELOAD=libjemalloc.so

multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:1 jemalloc

Command:

/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/benchmark/umf-benchmark --benchmark_format=csv --benchmark_filter=glibc

Environment Variables:

LD_PRELOAD=libjemalloc.so

multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:12 jemalloc

Command:

/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/benchmark/umf-benchmark --benchmark_format=csv --benchmark_filter=glibc

Environment Variables:

LD_PRELOAD=libjemalloc.so

multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:1 jemalloc

Command:

/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/benchmark/umf-benchmark --benchmark_format=csv --benchmark_filter=glibc

Environment Variables:

LD_PRELOAD=libjemalloc.so

multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:12 tbbProxy

Command:

/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/benchmark/umf-benchmark --benchmark_format=csv --benchmark_filter=glibc

Environment Variables:

LD_PRELOAD=libtbbmalloc_proxy.so

multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:1 tbbProxy

Command:

/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/benchmark/umf-benchmark --benchmark_format=csv --benchmark_filter=glibc

Environment Variables:

LD_PRELOAD=libtbbmalloc_proxy.so

multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:12 tbbProxy

Command:

/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/benchmark/umf-benchmark --benchmark_format=csv --benchmark_filter=glibc

Environment Variables:

LD_PRELOAD=libtbbmalloc_proxy.so

multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:1 tbbProxy

Command:

/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/benchmark/umf-benchmark --benchmark_format=csv --benchmark_filter=glibc

Environment Variables:

LD_PRELOAD=libtbbmalloc_proxy.so

multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:12 tbbProxy

Command:

/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/benchmark/umf-benchmark --benchmark_format=csv --benchmark_filter=glibc

Environment Variables:

LD_PRELOAD=libtbbmalloc_proxy.so

multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:1 tbbProxy

Command:

/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/benchmark/umf-benchmark --benchmark_format=csv --benchmark_filter=glibc

Environment Variables:

LD_PRELOAD=libtbbmalloc_proxy.so

@github-actions GitHub Actions
Copy link

Compute Benchmarks run (with params: --timeout 1000):
https://github.com/oneapi-src/unified-memory-framework/actions/runs/13811216426

@github-actions GitHub Actions
Copy link

Compute Benchmarks run (--timeout 1000):
https://github.com/oneapi-src/unified-memory-framework/actions/runs/13811216426
Job status: success. Test status: success.

Summary

(Emphasized values are the best results)

Improved 2 (threshold 2.00%)
Benchmark This PR baseline Change
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:1 umfProxy 2917.110000 ns 3003.920 ns 2.98%
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:1 umfProxy 3052.360000 ns 3133.670 ns 2.66%

Performance change in benchmark groups

UMF
Relative perf in group multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:12 (3)
Benchmark This PR baseline Change
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:12 umfProxy 80660.200000 ns -
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:12 jemalloc 1048.250000 ns -
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:12 tbbProxy 2719.390000 ns -
Relative perf in group multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:1 (11)
Benchmark This PR baseline Change
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:1 umfProxy 3009.700000 ns 3057.580 ns 1.59%
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:1 jemalloc 79.896 ns 79.889000 ns -0.01%
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:1 tbbProxy 524.098 ns 519.525000 ns -0.87%
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:1 glibc - 347.396000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:1 disjoint_pool<os_provider> - 902.917000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:1 jemalloc_pool<os_provider> - 1643.490000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:1 scalable_pool<os_provider> - 503.102000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:1 proxy_pool<fixed_provider> - 4811.200000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:1 fixed_provider - 2400.600000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:1 disjoint_pool<fixed_provider> - 901.185000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:1 scalable_pool<fixed_provider> - 481.997000 ns
Relative perf in group multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:12 (3)
Benchmark This PR baseline Change
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:12 umfProxy 81029.300000 ns -
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:12 jemalloc 1078.500000 ns -
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:12 tbbProxy 3742.910000 ns -
Relative perf in group multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:1 (9)
Benchmark This PR baseline Change
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:1 umfProxy 3052.360000 ns 3133.670 ns 2.66%
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:1 jemalloc 199.443 ns 198.255000 ns -0.60%
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:1 tbbProxy 580.598 ns 577.115000 ns -0.60%
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:1 glibc - 838.328000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:1 disjoint_pool<os_provider> - 4553.000000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:1 jemalloc_pool<os_provider> - 1101.230000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:1 scalable_pool<os_provider> - 565.600000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:1 disjoint_pool<fixed_provider> - 1702.050000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:1 scalable_pool<fixed_provider> - 565.912000 ns
Relative perf in group multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:12 (3)
Benchmark This PR baseline Change
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:12 umfProxy 80786.500000 ns -
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:12 jemalloc 541.234000 ns -
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:12 tbbProxy 2619.530000 ns -
Relative perf in group multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:1 (9)
Benchmark This PR baseline Change
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:1 umfProxy 2917.110000 ns 3003.920 ns 2.98%
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:1 jemalloc 62.246600 ns 62.772 ns 0.84%
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:1 tbbProxy 344.654 ns 344.416000 ns -0.07%
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:1 glibc - 173.701000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:1 disjoint_pool<os_provider> - 642.213000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:1 jemalloc_pool<os_provider> - 670.466000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:1 scalable_pool<os_provider> - 346.455000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:1 disjoint_pool<fixed_provider> - 579.266000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:1 scalable_pool<fixed_provider> - 346.582000 ns
Relative perf in group multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:4 (9)
Benchmark This PR baseline Change
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:4 glibc - 1686.930000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:4 disjoint_pool<os_provider> - 30322.700000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:4 jemalloc_pool<os_provider> - 47098.800000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:4 scalable_pool<os_provider> - 1139.100000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:4 disjoint_pool<fixed_provider> - 26081.800000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:4 scalable_pool<fixed_provider> - 1109.850000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:4 umfProxy - 25254.100000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:4 jemalloc - 564.857000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:4 tbbProxy - 1646.720000 ns
Relative perf in group multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:4 (9)
Benchmark This PR baseline Change
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:4 glibc - 2024.250000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:4 disjoint_pool<os_provider> - 38196.800000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:4 jemalloc_pool<os_provider> - 9682.560000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:4 scalable_pool<os_provider> - 2368.340000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:4 disjoint_pool<fixed_provider> - 18896.400000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:4 scalable_pool<fixed_provider> - 1033.440000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:4 umfProxy - 25707.200000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:4 jemalloc - 698.220000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:4 tbbProxy - 1766.030000 ns
Relative perf in group multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:4 (9)
Benchmark This PR baseline Change
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:4 glibc - 968.870000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:4 disjoint_pool<os_provider> - 10356.800000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:4 jemalloc_pool<os_provider> - 9065.690000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:4 scalable_pool<os_provider> - 1868.500000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:4 disjoint_pool<fixed_provider> - 10411.900000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:4 scalable_pool<fixed_provider> - 764.805000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:4 umfProxy - 24688.600000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:4 jemalloc - 366.993000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:4 tbbProxy - 1224.070000 ns
Relative perf in group multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:50000/threads:1 (2)
Benchmark This PR baseline Change
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:50000/threads:1 proxy_pool<os_provider> - 28637.900000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:50000/threads:1 os_provider - 25570.300000 ns

Details

Benchmark details - environment, command...
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:12 umfProxy

Command:

/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/benchmark/umf-benchmark --benchmark_format=csv --benchmark_filter=glibc

Environment Variables:

LD_PRELOAD=/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/lib/libumf_proxy.so

multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:1 umfProxy

Command:

/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/benchmark/umf-benchmark --benchmark_format=csv --benchmark_filter=glibc

Environment Variables:

LD_PRELOAD=/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/lib/libumf_proxy.so

multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:12 umfProxy

Command:

/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/benchmark/umf-benchmark --benchmark_format=csv --benchmark_filter=glibc

Environment Variables:

LD_PRELOAD=/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/lib/libumf_proxy.so

multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:1 umfProxy

Command:

/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/benchmark/umf-benchmark --benchmark_format=csv --benchmark_filter=glibc

Environment Variables:

LD_PRELOAD=/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/lib/libumf_proxy.so

multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:12 umfProxy

Command:

/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/benchmark/umf-benchmark --benchmark_format=csv --benchmark_filter=glibc

Environment Variables:

LD_PRELOAD=/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/lib/libumf_proxy.so

multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:1 umfProxy

Command:

/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/benchmark/umf-benchmark --benchmark_format=csv --benchmark_filter=glibc

Environment Variables:

LD_PRELOAD=/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/lib/libumf_proxy.so

multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:12 jemalloc

Command:

/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/benchmark/umf-benchmark --benchmark_format=csv --benchmark_filter=glibc

Environment Variables:

LD_PRELOAD=libjemalloc.so

multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:1 jemalloc

Command:

/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/benchmark/umf-benchmark --benchmark_format=csv --benchmark_filter=glibc

Environment Variables:

LD_PRELOAD=libjemalloc.so

multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:12 jemalloc

Command:

/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/benchmark/umf-benchmark --benchmark_format=csv --benchmark_filter=glibc

Environment Variables:

LD_PRELOAD=libjemalloc.so

multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:1 jemalloc

Command:

/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/benchmark/umf-benchmark --benchmark_format=csv --benchmark_filter=glibc

Environment Variables:

LD_PRELOAD=libjemalloc.so

multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:12 jemalloc

Command:

/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/benchmark/umf-benchmark --benchmark_format=csv --benchmark_filter=glibc

Environment Variables:

LD_PRELOAD=libjemalloc.so

multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:1 jemalloc

Command:

/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/benchmark/umf-benchmark --benchmark_format=csv --benchmark_filter=glibc

Environment Variables:

LD_PRELOAD=libjemalloc.so

multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:12 tbbProxy

Command:

/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/benchmark/umf-benchmark --benchmark_format=csv --benchmark_filter=glibc

Environment Variables:

LD_PRELOAD=libtbbmalloc_proxy.so

multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:1 tbbProxy

Command:

/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/benchmark/umf-benchmark --benchmark_format=csv --benchmark_filter=glibc

Environment Variables:

LD_PRELOAD=libtbbmalloc_proxy.so

multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:12 tbbProxy

Command:

/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/benchmark/umf-benchmark --benchmark_format=csv --benchmark_filter=glibc

Environment Variables:

LD_PRELOAD=libtbbmalloc_proxy.so

multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:1 tbbProxy

Command:

/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/benchmark/umf-benchmark --benchmark_format=csv --benchmark_filter=glibc

Environment Variables:

LD_PRELOAD=libtbbmalloc_proxy.so

multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:12 tbbProxy

Command:

/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/benchmark/umf-benchmark --benchmark_format=csv --benchmark_filter=glibc

Environment Variables:

LD_PRELOAD=libtbbmalloc_proxy.so

multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:1 tbbProxy

Command:

/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/benchmark/umf-benchmark --benchmark_format=csv --benchmark_filter=glibc

Environment Variables:

LD_PRELOAD=libtbbmalloc_proxy.so

@github-actions GitHub Actions
Copy link

Compute Benchmarks run (with params: --timeout=1000):
https://github.com/oneapi-src/unified-memory-framework/actions/runs/13811987045

@github-actions GitHub Actions
Copy link

Compute Benchmarks run (--timeout=1000):
https://github.com/oneapi-src/unified-memory-framework/actions/runs/13811987045
Job status: success. Test status: success.

Summary

(Emphasized values are the best results)

Improved 1 (threshold 2.00%)
Benchmark This PR baseline Change
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:1 umfProxy 2942.730000 ns 3003.920 ns 2.08%

Performance change in benchmark groups

UMF
Relative perf in group multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:12 (3)
Benchmark This PR baseline Change
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:12 umfProxy 80154.800000 ns -
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:12 jemalloc 979.794000 ns -
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:12 tbbProxy 3112.730000 ns -
Relative perf in group multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:1 (11)
Benchmark This PR baseline Change
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:1 tbbProxy 517.712000 ns 519.525 ns 0.35%
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:1 jemalloc 80.026 ns 79.889000 ns -0.17%
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:1 umfProxy 3092.040 ns 3057.580000 ns -1.11%
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:1 glibc - 347.396000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:1 disjoint_pool<os_provider> - 902.917000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:1 jemalloc_pool<os_provider> - 1643.490000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:1 scalable_pool<os_provider> - 503.102000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:1 proxy_pool<fixed_provider> - 4811.200000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:1 fixed_provider - 2400.600000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:1 disjoint_pool<fixed_provider> - 901.185000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:1 scalable_pool<fixed_provider> - 481.997000 ns
Relative perf in group multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:12 (3)
Benchmark This PR baseline Change
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:12 umfProxy 80661.300000 ns -
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:12 jemalloc 1138.960000 ns -
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:12 tbbProxy 3157.080000 ns -
Relative perf in group multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:1 (9)
Benchmark This PR baseline Change
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:1 umfProxy 3117.050000 ns 3133.670 ns 0.53%
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:1 jemalloc 197.853000 ns 198.255 ns 0.20%
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:1 tbbProxy 577.783 ns 577.115000 ns -0.12%
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:1 glibc - 838.328000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:1 disjoint_pool<os_provider> - 4553.000000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:1 jemalloc_pool<os_provider> - 1101.230000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:1 scalable_pool<os_provider> - 565.600000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:1 disjoint_pool<fixed_provider> - 1702.050000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:1 scalable_pool<fixed_provider> - 565.912000 ns
Relative perf in group multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:12 (3)
Benchmark This PR baseline Change
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:12 umfProxy 79933.500000 ns -
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:12 jemalloc 565.195000 ns -
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:12 tbbProxy 2072.530000 ns -
Relative perf in group multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:1 (9)
Benchmark This PR baseline Change
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:1 umfProxy 2942.730000 ns 3003.920 ns 2.08%
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:1 tbbProxy 344.861 ns 344.416000 ns -0.13%
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:1 jemalloc 63.049 ns 62.772200 ns -0.44%
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:1 glibc - 173.701000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:1 disjoint_pool<os_provider> - 642.213000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:1 jemalloc_pool<os_provider> - 670.466000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:1 scalable_pool<os_provider> - 346.455000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:1 disjoint_pool<fixed_provider> - 579.266000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:1 scalable_pool<fixed_provider> - 346.582000 ns
Relative perf in group multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:4 (9)
Benchmark This PR baseline Change
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:4 glibc - 1686.930000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:4 disjoint_pool<os_provider> - 30322.700000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:4 jemalloc_pool<os_provider> - 47098.800000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:4 scalable_pool<os_provider> - 1139.100000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:4 disjoint_pool<fixed_provider> - 26081.800000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:4 scalable_pool<fixed_provider> - 1109.850000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:4 umfProxy - 25254.100000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:4 jemalloc - 564.857000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:4 tbbProxy - 1646.720000 ns
Relative perf in group multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:4 (9)
Benchmark This PR baseline Change
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:4 glibc - 2024.250000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:4 disjoint_pool<os_provider> - 38196.800000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:4 jemalloc_pool<os_provider> - 9682.560000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:4 scalable_pool<os_provider> - 2368.340000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:4 disjoint_pool<fixed_provider> - 18896.400000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:4 scalable_pool<fixed_provider> - 1033.440000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:4 umfProxy - 25707.200000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:4 jemalloc - 698.220000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:4 tbbProxy - 1766.030000 ns
Relative perf in group multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:4 (9)
Benchmark This PR baseline Change
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:4 glibc - 968.870000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:4 disjoint_pool<os_provider> - 10356.800000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:4 jemalloc_pool<os_provider> - 9065.690000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:4 scalable_pool<os_provider> - 1868.500000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:4 disjoint_pool<fixed_provider> - 10411.900000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:4 scalable_pool<fixed_provider> - 764.805000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:4 umfProxy - 24688.600000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:4 jemalloc - 366.993000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:4 tbbProxy - 1224.070000 ns
Relative perf in group multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:50000/threads:1 (2)
Benchmark This PR baseline Change
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:50000/threads:1 proxy_pool<os_provider> - 28637.900000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:50000/threads:1 os_provider - 25570.300000 ns

Details

Benchmark details - environment, command...
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:12 umfProxy

Command:

/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/benchmark/umf-benchmark --benchmark_format=csv --benchmark_filter=glibc

Environment Variables:

LD_PRELOAD=/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/lib/libumf_proxy.so

multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:1 umfProxy

Command:

/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/benchmark/umf-benchmark --benchmark_format=csv --benchmark_filter=glibc

Environment Variables:

LD_PRELOAD=/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/lib/libumf_proxy.so

multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:12 umfProxy

Command:

/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/benchmark/umf-benchmark --benchmark_format=csv --benchmark_filter=glibc

Environment Variables:

LD_PRELOAD=/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/lib/libumf_proxy.so

multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:1 umfProxy

Command:

/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/benchmark/umf-benchmark --benchmark_format=csv --benchmark_filter=glibc

Environment Variables:

LD_PRELOAD=/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/lib/libumf_proxy.so

multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:12 umfProxy

Command:

/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/benchmark/umf-benchmark --benchmark_format=csv --benchmark_filter=glibc

Environment Variables:

LD_PRELOAD=/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/lib/libumf_proxy.so

multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:1 umfProxy

Command:

/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/benchmark/umf-benchmark --benchmark_format=csv --benchmark_filter=glibc

Environment Variables:

LD_PRELOAD=/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/lib/libumf_proxy.so

multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:12 jemalloc

Command:

/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/benchmark/umf-benchmark --benchmark_format=csv --benchmark_filter=glibc

Environment Variables:

LD_PRELOAD=libjemalloc.so

multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:1 jemalloc

Command:

/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/benchmark/umf-benchmark --benchmark_format=csv --benchmark_filter=glibc

Environment Variables:

LD_PRELOAD=libjemalloc.so

multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:12 jemalloc

Command:

/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/benchmark/umf-benchmark --benchmark_format=csv --benchmark_filter=glibc

Environment Variables:

LD_PRELOAD=libjemalloc.so

multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:1 jemalloc

Command:

/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/benchmark/umf-benchmark --benchmark_format=csv --benchmark_filter=glibc

Environment Variables:

LD_PRELOAD=libjemalloc.so

multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:12 jemalloc

Command:

/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/benchmark/umf-benchmark --benchmark_format=csv --benchmark_filter=glibc

Environment Variables:

LD_PRELOAD=libjemalloc.so

multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:1 jemalloc

Command:

/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/benchmark/umf-benchmark --benchmark_format=csv --benchmark_filter=glibc

Environment Variables:

LD_PRELOAD=libjemalloc.so

multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:12 tbbProxy

Command:

/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/benchmark/umf-benchmark --benchmark_format=csv --benchmark_filter=glibc

Environment Variables:

LD_PRELOAD=libtbbmalloc_proxy.so

multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:1 tbbProxy

Command:

/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/benchmark/umf-benchmark --benchmark_format=csv --benchmark_filter=glibc

Environment Variables:

LD_PRELOAD=libtbbmalloc_proxy.so

multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:12 tbbProxy

Command:

/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/benchmark/umf-benchmark --benchmark_format=csv --benchmark_filter=glibc

Environment Variables:

LD_PRELOAD=libtbbmalloc_proxy.so

multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:1 tbbProxy

Command:

/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/benchmark/umf-benchmark --benchmark_format=csv --benchmark_filter=glibc

Environment Variables:

LD_PRELOAD=libtbbmalloc_proxy.so

multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:12 tbbProxy

Command:

/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/benchmark/umf-benchmark --benchmark_format=csv --benchmark_filter=glibc

Environment Variables:

LD_PRELOAD=libtbbmalloc_proxy.so

multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:1 tbbProxy

Command:

/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/benchmark/umf-benchmark --benchmark_format=csv --benchmark_filter=glibc

Environment Variables:

LD_PRELOAD=libtbbmalloc_proxy.so

@lplewa lplewa force-pushed the benchmarks_threads branch from c9e59f8 to 66e8451 Compare March 12, 2025 14:24
@github-actions GitHub Actions
Copy link

Compute Benchmarks run (with params: --timeout 1000):
https://github.com/oneapi-src/unified-memory-framework/actions/runs/13813775258

@github-actions GitHub Actions
Copy link

Compute Benchmarks run (--timeout 1000):
https://github.com/oneapi-src/unified-memory-framework/actions/runs/13813775258
Job status: success. Test status: success.

Summary

(Emphasized values are the best results)
No diffs to calculate performance change

Performance change in benchmark groups

UMF
Relative perf in group multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:12 (3)
Benchmark This PR baseline Change
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:12 umfProxy 85625.200000 ns -
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:12 jemalloc 990.117000 ns -
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:12 tbbProxy 3103.210000 ns -
Relative perf in group multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:1 (11)
Benchmark This PR baseline Change
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:1 umfProxy 3059.940 ns 3057.580000 ns -0.08%
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:1 tbbProxy 520.333 ns 519.525000 ns -0.16%
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:1 jemalloc 80.982 ns 79.889000 ns -1.35%
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:1 glibc - 347.396000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:1 disjoint_pool<os_provider> - 902.917000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:1 jemalloc_pool<os_provider> - 1643.490000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:1 scalable_pool<os_provider> - 503.102000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:1 proxy_pool<fixed_provider> - 4811.200000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:1 fixed_provider - 2400.600000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:1 disjoint_pool<fixed_provider> - 901.185000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:1 scalable_pool<fixed_provider> - 481.997000 ns
Relative perf in group multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:12 (3)
Benchmark This PR baseline Change
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:12 umfProxy 86177.500000 ns -
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:12 jemalloc 1203.690000 ns -
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:12 tbbProxy 3110.430000 ns -
Relative perf in group multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:1 (9)
Benchmark This PR baseline Change
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:1 umfProxy 3132.690000 ns 3133.670 ns 0.03%
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:1 tbbProxy 579.979 ns 577.115000 ns -0.49%
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:1 jemalloc 199.270 ns 198.255000 ns -0.51%
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:1 glibc - 838.328000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:1 disjoint_pool<os_provider> - 4553.000000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:1 jemalloc_pool<os_provider> - 1101.230000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:1 scalable_pool<os_provider> - 565.600000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:1 disjoint_pool<fixed_provider> - 1702.050000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:1 scalable_pool<fixed_provider> - 565.912000 ns
Relative perf in group multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:12 (3)
Benchmark This PR baseline Change
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:12 umfProxy 85456.900000 ns -
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:12 jemalloc 623.625000 ns -
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:12 tbbProxy 2096.120000 ns -
Relative perf in group multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:1 (9)
Benchmark This PR baseline Change
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:1 umfProxy 2980.300000 ns 3003.920 ns 0.79%
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:1 jemalloc 62.471300 ns 62.772 ns 0.48%
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:1 tbbProxy 344.218000 ns 344.416 ns 0.06%
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:1 glibc - 173.701000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:1 disjoint_pool<os_provider> - 642.213000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:1 jemalloc_pool<os_provider> - 670.466000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:1 scalable_pool<os_provider> - 346.455000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:1 disjoint_pool<fixed_provider> - 579.266000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:1 scalable_pool<fixed_provider> - 346.582000 ns
Relative perf in group multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:4 (9)
Benchmark This PR baseline Change
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:4 glibc - 1686.930000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:4 disjoint_pool<os_provider> - 30322.700000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:4 jemalloc_pool<os_provider> - 47098.800000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:4 scalable_pool<os_provider> - 1139.100000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:4 disjoint_pool<fixed_provider> - 26081.800000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:4 scalable_pool<fixed_provider> - 1109.850000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:4 umfProxy - 25254.100000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:4 jemalloc - 564.857000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:4 tbbProxy - 1646.720000 ns
Relative perf in group multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:4 (9)
Benchmark This PR baseline Change
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:4 glibc - 2024.250000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:4 disjoint_pool<os_provider> - 38196.800000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:4 jemalloc_pool<os_provider> - 9682.560000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:4 scalable_pool<os_provider> - 2368.340000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:4 disjoint_pool<fixed_provider> - 18896.400000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:4 scalable_pool<fixed_provider> - 1033.440000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:4 umfProxy - 25707.200000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:4 jemalloc - 698.220000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:4 tbbProxy - 1766.030000 ns
Relative perf in group multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:4 (9)
Benchmark This PR baseline Change
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:4 glibc - 968.870000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:4 disjoint_pool<os_provider> - 10356.800000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:4 jemalloc_pool<os_provider> - 9065.690000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:4 scalable_pool<os_provider> - 1868.500000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:4 disjoint_pool<fixed_provider> - 10411.900000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:4 scalable_pool<fixed_provider> - 764.805000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:4 umfProxy - 24688.600000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:4 jemalloc - 366.993000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:4 tbbProxy - 1224.070000 ns
Relative perf in group multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:50000/threads:1 (2)
Benchmark This PR baseline Change
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:50000/threads:1 proxy_pool<os_provider> - 28637.900000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:50000/threads:1 os_provider - 25570.300000 ns

Details

Benchmark details - environment, command...
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:12 umfProxy

Command:

/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/benchmark/umf-benchmark --benchmark_format=csv --benchmark_filter=glibc

Environment Variables:

LD_PRELOAD=/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/lib/libumf_proxy.so

multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:1 umfProxy

Command:

/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/benchmark/umf-benchmark --benchmark_format=csv --benchmark_filter=glibc

Environment Variables:

LD_PRELOAD=/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/lib/libumf_proxy.so

multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:12 umfProxy

Command:

/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/benchmark/umf-benchmark --benchmark_format=csv --benchmark_filter=glibc

Environment Variables:

LD_PRELOAD=/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/lib/libumf_proxy.so

multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:1 umfProxy

Command:

/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/benchmark/umf-benchmark --benchmark_format=csv --benchmark_filter=glibc

Environment Variables:

LD_PRELOAD=/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/lib/libumf_proxy.so

multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:12 umfProxy

Command:

/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/benchmark/umf-benchmark --benchmark_format=csv --benchmark_filter=glibc

Environment Variables:

LD_PRELOAD=/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/lib/libumf_proxy.so

multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:1 umfProxy

Command:

/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/benchmark/umf-benchmark --benchmark_format=csv --benchmark_filter=glibc

Environment Variables:

LD_PRELOAD=/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/lib/libumf_proxy.so

multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:12 jemalloc

Command:

/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/benchmark/umf-benchmark --benchmark_format=csv --benchmark_filter=glibc

Environment Variables:

LD_PRELOAD=libjemalloc.so

multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:1 jemalloc

Command:

/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/benchmark/umf-benchmark --benchmark_format=csv --benchmark_filter=glibc

Environment Variables:

LD_PRELOAD=libjemalloc.so

multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:12 jemalloc

Command:

/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/benchmark/umf-benchmark --benchmark_format=csv --benchmark_filter=glibc

Environment Variables:

LD_PRELOAD=libjemalloc.so

multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:1 jemalloc

Command:

/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/benchmark/umf-benchmark --benchmark_format=csv --benchmark_filter=glibc

Environment Variables:

LD_PRELOAD=libjemalloc.so

multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:12 jemalloc

Command:

/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/benchmark/umf-benchmark --benchmark_format=csv --benchmark_filter=glibc

Environment Variables:

LD_PRELOAD=libjemalloc.so

multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:1 jemalloc

Command:

/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/benchmark/umf-benchmark --benchmark_format=csv --benchmark_filter=glibc

Environment Variables:

LD_PRELOAD=libjemalloc.so

multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:12 tbbProxy

Command:

/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/benchmark/umf-benchmark --benchmark_format=csv --benchmark_filter=glibc

Environment Variables:

LD_PRELOAD=libtbbmalloc_proxy.so

multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:1 tbbProxy

Command:

/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/benchmark/umf-benchmark --benchmark_format=csv --benchmark_filter=glibc

Environment Variables:

LD_PRELOAD=libtbbmalloc_proxy.so

multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:12 tbbProxy

Command:

/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/benchmark/umf-benchmark --benchmark_format=csv --benchmark_filter=glibc

Environment Variables:

LD_PRELOAD=libtbbmalloc_proxy.so

multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:1 tbbProxy

Command:

/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/benchmark/umf-benchmark --benchmark_format=csv --benchmark_filter=glibc

Environment Variables:

LD_PRELOAD=libtbbmalloc_proxy.so

multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:12 tbbProxy

Command:

/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/benchmark/umf-benchmark --benchmark_format=csv --benchmark_filter=glibc

Environment Variables:

LD_PRELOAD=libtbbmalloc_proxy.so

multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:1 tbbProxy

Command:

/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/benchmark/umf-benchmark --benchmark_format=csv --benchmark_filter=glibc

Environment Variables:

LD_PRELOAD=libtbbmalloc_proxy.so

@github-actions GitHub Actions
Copy link

Compute Benchmarks run (with params: --timeout 1000):
https://github.com/oneapi-src/unified-memory-framework/actions/runs/13845968570

@github-actions GitHub Actions
Copy link

Compute Benchmarks run (--timeout 1000):
https://github.com/oneapi-src/unified-memory-framework/actions/runs/13845968570
Job status: success. Test status: success.

Summary

(Emphasized values are the best results)
No diffs to calculate performance change

Performance change in benchmark groups

UMF
Relative perf in group multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:12 (3)
Benchmark This PR baseline Change
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:12 umfProxy 80902.600000 ns -
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:12 jemalloc 979.550000 ns -
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:12 tbbProxy 2884.530000 ns -
Relative perf in group multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:1 (11)
Benchmark This PR baseline Change
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:1 jemalloc 80.843 ns 80.672000 ns -0.21%
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:1 umfProxy 3090.140 ns 3078.530000 ns -0.38%
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:1 tbbProxy 529.941 ns 523.721000 ns -1.17%
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:1 glibc - 355.101000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:1 disjoint_pool<os_provider> - 943.389000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:1 jemalloc_pool<os_provider> - 1631.010000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:1 scalable_pool<os_provider> - 499.427000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:1 proxy_pool<fixed_provider> - 4657.830000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:1 fixed_provider - 2447.950000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:1 disjoint_pool<fixed_provider> - 908.664000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:1 scalable_pool<fixed_provider> - 481.767000 ns
Relative perf in group multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:12 (3)
Benchmark This PR baseline Change
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:12 umfProxy 81083.200000 ns -
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:12 jemalloc 1163.120000 ns -
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:12 tbbProxy 3558.650000 ns -
Relative perf in group multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:1 (9)
Benchmark This PR baseline Change
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:1 jemalloc 199.687 ns 197.715000 ns -0.99%
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:1 tbbProxy 585.485 ns 578.929000 ns -1.12%
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:1 umfProxy 3167.560 ns 3128.120000 ns -1.25%
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:1 glibc - 842.360000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:1 disjoint_pool<os_provider> - 4536.590000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:1 jemalloc_pool<os_provider> - 1096.570000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:1 scalable_pool<os_provider> - 565.826000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:1 disjoint_pool<fixed_provider> - 1642.820000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:1 scalable_pool<fixed_provider> - 564.697000 ns
Relative perf in group multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:12 (3)
Benchmark This PR baseline Change
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:12 umfProxy 80580.000000 ns -
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:12 jemalloc 571.193000 ns -
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:12 tbbProxy 2489.000000 ns -
Relative perf in group multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:1 (9)
Benchmark This PR baseline Change
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:1 jemalloc 62.238500 ns 63.346 ns 1.78%
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:1 umfProxy 3013.960 ns 3006.410000 ns -0.25%
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:1 tbbProxy 345.332 ns 344.434000 ns -0.26%
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:1 glibc - 174.045000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:1 disjoint_pool<os_provider> - 611.892000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:1 jemalloc_pool<os_provider> - 663.450000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:1 scalable_pool<os_provider> - 344.975000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:1 disjoint_pool<fixed_provider> - 575.710000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:1 scalable_pool<fixed_provider> - 345.592000 ns
Relative perf in group multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:4 (9)
Benchmark This PR baseline Change
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:4 glibc - 1826.460000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:4 disjoint_pool<os_provider> - 32068.300000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:4 jemalloc_pool<os_provider> - 51701.900000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:4 scalable_pool<os_provider> - 1056.040000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:4 disjoint_pool<fixed_provider> - 31481.300000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:4 scalable_pool<fixed_provider> - 1603.000000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:4 umfProxy - 24958.000000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:4 jemalloc - 562.061000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:4 tbbProxy - 1646.770000 ns
Relative perf in group multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:4 (9)
Benchmark This PR baseline Change
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:4 glibc - 2253.580000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:4 disjoint_pool<os_provider> - 37574.100000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:4 jemalloc_pool<os_provider> - 9642.330000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:4 scalable_pool<os_provider> - 1095.730000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:4 disjoint_pool<fixed_provider> - 18181.200000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:4 scalable_pool<fixed_provider> - 1716.890000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:4 umfProxy - 25711.500000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:4 jemalloc - 702.341000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:4 tbbProxy - 1747.890000 ns
Relative perf in group multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:4 (9)
Benchmark This PR baseline Change
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:4 glibc - 906.764000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:4 disjoint_pool<os_provider> - 10579.500000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:4 jemalloc_pool<os_provider> - 8982.490000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:4 scalable_pool<os_provider> - 836.281000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:4 disjoint_pool<fixed_provider> - 10713.900000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:4 scalable_pool<fixed_provider> - 1213.180000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:4 umfProxy - 24628.400000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:4 jemalloc - 367.533000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:4 tbbProxy - 1199.690000 ns
Relative perf in group multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:50000/threads:1 (2)
Benchmark This PR baseline Change
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:50000/threads:1 proxy_pool<os_provider> - 28503.400000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:50000/threads:1 os_provider - 25211.800000 ns

Details

Benchmark details - environment, command...
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:12 umfProxy

Command:

/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/benchmark/umf-benchmark --benchmark_format=csv --benchmark_filter=glibc

Environment Variables:

LD_PRELOAD=/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/lib/libumf_proxy.so

multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:1 umfProxy

Command:

/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/benchmark/umf-benchmark --benchmark_format=csv --benchmark_filter=glibc

Environment Variables:

LD_PRELOAD=/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/lib/libumf_proxy.so

multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:12 umfProxy

Command:

/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/benchmark/umf-benchmark --benchmark_format=csv --benchmark_filter=glibc

Environment Variables:

LD_PRELOAD=/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/lib/libumf_proxy.so

multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:1 umfProxy

Command:

/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/benchmark/umf-benchmark --benchmark_format=csv --benchmark_filter=glibc

Environment Variables:

LD_PRELOAD=/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/lib/libumf_proxy.so

multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:12 umfProxy

Command:

/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/benchmark/umf-benchmark --benchmark_format=csv --benchmark_filter=glibc

Environment Variables:

LD_PRELOAD=/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/lib/libumf_proxy.so

multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:1 umfProxy

Command:

/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/benchmark/umf-benchmark --benchmark_format=csv --benchmark_filter=glibc

Environment Variables:

LD_PRELOAD=/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/lib/libumf_proxy.so

multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:12 jemalloc

Command:

/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/benchmark/umf-benchmark --benchmark_format=csv --benchmark_filter=glibc

Environment Variables:

LD_PRELOAD=libjemalloc.so

multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:1 jemalloc

Command:

/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/benchmark/umf-benchmark --benchmark_format=csv --benchmark_filter=glibc

Environment Variables:

LD_PRELOAD=libjemalloc.so

multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:12 jemalloc

Command:

/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/benchmark/umf-benchmark --benchmark_format=csv --benchmark_filter=glibc

Environment Variables:

LD_PRELOAD=libjemalloc.so

multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:1 jemalloc

Command:

/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/benchmark/umf-benchmark --benchmark_format=csv --benchmark_filter=glibc

Environment Variables:

LD_PRELOAD=libjemalloc.so

multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:12 jemalloc

Command:

/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/benchmark/umf-benchmark --benchmark_format=csv --benchmark_filter=glibc

Environment Variables:

LD_PRELOAD=libjemalloc.so

multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:1 jemalloc

Command:

/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/benchmark/umf-benchmark --benchmark_format=csv --benchmark_filter=glibc

Environment Variables:

LD_PRELOAD=libjemalloc.so

multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:12 tbbProxy

Command:

/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/benchmark/umf-benchmark --benchmark_format=csv --benchmark_filter=glibc

Environment Variables:

LD_PRELOAD=libtbbmalloc_proxy.so

multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:1 tbbProxy

Command:

/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/benchmark/umf-benchmark --benchmark_format=csv --benchmark_filter=glibc

Environment Variables:

LD_PRELOAD=libtbbmalloc_proxy.so

multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:12 tbbProxy

Command:

/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/benchmark/umf-benchmark --benchmark_format=csv --benchmark_filter=glibc

Environment Variables:

LD_PRELOAD=libtbbmalloc_proxy.so

multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:1 tbbProxy

Command:

/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/benchmark/umf-benchmark --benchmark_format=csv --benchmark_filter=glibc

Environment Variables:

LD_PRELOAD=libtbbmalloc_proxy.so

multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:12 tbbProxy

Command:

/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/benchmark/umf-benchmark --benchmark_format=csv --benchmark_filter=glibc

Environment Variables:

LD_PRELOAD=libtbbmalloc_proxy.so

multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:1 tbbProxy

Command:

/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/benchmark/umf-benchmark --benchmark_format=csv --benchmark_filter=glibc

Environment Variables:

LD_PRELOAD=libtbbmalloc_proxy.so

@github-actions GitHub Actions
Copy link

Compute Benchmarks run (with params: --timeout 2000):
https://github.com/oneapi-src/unified-memory-framework/actions/runs/13848117213

@github-actions GitHub Actions
Copy link

Compute Benchmarks run (--timeout 2000):
https://github.com/oneapi-src/unified-memory-framework/actions/runs/13848117213
Job status: success. Test status: success.

Summary

(Emphasized values are the best results)

Improved 1 (threshold 2.00%)
Benchmark This PR baseline Change
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:1 jemalloc 61.690900 ns 63.346 ns 2.68%
Regressed 6 (threshold 2.00%)
Benchmark This PR baseline Change
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:1 disjoint_pool<fixed_provider> 1038.430 ns 908.664000 ns -12.50%
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:1 proxy_pool<fixed_provider> 5153.350 ns 4657.830000 ns -9.62%
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:1 disjoint_pool<fixed_provider> 1809.680 ns 1642.820000 ns -9.22%
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:1 disjoint_pool<os_provider> 1034.730 ns 943.389000 ns -8.83%
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:1 disjoint_pool<os_provider> 4878.200 ns 4536.590000 ns -7.00%
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:1 scalable_pool<fixed_provider> 354.970 ns 345.592000 ns -2.64%

Performance change in benchmark groups

UMF
Relative perf in group multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:12 (9)
Benchmark This PR baseline Change
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:12 glibc 4246.590000 ns -
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:12 disjoint_pool<os_provider> 128330.000000 ns -
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:12 jemalloc_pool<os_provider> 388281.000000 ns -
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:12 scalable_pool<os_provider> 2784.520000 ns -
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:12 disjoint_pool<fixed_provider> 174117.000000 ns -
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:12 scalable_pool<fixed_provider> 2876.510000 ns -
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:12 umfProxy 85248.300000 ns -
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:12 jemalloc 935.138000 ns -
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:12 tbbProxy 3109.300000 ns -
Relative perf in group multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:1 (11)
Benchmark This PR baseline Change
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:1 fixed_provider 2408.220000 ns 2447.950 ns 1.65%
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:1 umfProxy 3045.720000 ns 3078.530 ns 1.08%
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:1 tbbProxy 520.097000 ns 523.721 ns 0.70%
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:1 jemalloc 80.141000 ns 80.672 ns 0.66%
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:1 glibc 353.104000 ns 355.101 ns 0.57%
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:1 jemalloc_pool<os_provider> 1625.890000 ns 1631.010 ns 0.31%
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:1 scalable_pool<fixed_provider> 483.753 ns 481.767000 ns -0.41%
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:1 scalable_pool<os_provider> 502.265 ns 499.427000 ns -0.57%
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:1 disjoint_pool<os_provider> 1034.730 ns 943.389000 ns -8.83%
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:1 proxy_pool<fixed_provider> 5153.350 ns 4657.830000 ns -9.62%
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:1 disjoint_pool<fixed_provider> 1038.430 ns 908.664000 ns -12.50%
Relative perf in group multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:12 (9)
Benchmark This PR baseline Change
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:12 glibc 3030.200000 ns -
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:12 disjoint_pool<os_provider> 120829.000000 ns -
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:12 jemalloc_pool<os_provider> 44971.600000 ns -
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:12 scalable_pool<os_provider> 3754.290000 ns -
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:12 disjoint_pool<fixed_provider> 67106.700000 ns -
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:12 scalable_pool<fixed_provider> 3100.860000 ns -
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:12 umfProxy 86450.400000 ns -
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:12 jemalloc 1265.650000 ns -
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:12 tbbProxy 3080.910000 ns -
Relative perf in group multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:1 (9)
Benchmark This PR baseline Change
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:1 umfProxy 3121.950000 ns 3128.120 ns 0.20%
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:1 glibc 840.733000 ns 842.360 ns 0.19%
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:1 scalable_pool<fixed_provider> 566.235 ns 564.697000 ns -0.27%
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:1 scalable_pool<os_provider> 568.102 ns 565.826000 ns -0.40%
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:1 tbbProxy 581.494 ns 578.929000 ns -0.44%
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:1 jemalloc_pool<os_provider> 1101.640 ns 1096.570000 ns -0.46%
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:1 jemalloc 199.334 ns 197.715000 ns -0.81%
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:1 disjoint_pool<os_provider> 4878.200 ns 4536.590000 ns -7.00%
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:1 disjoint_pool<fixed_provider> 1809.680 ns 1642.820000 ns -9.22%
Relative perf in group multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:12 (9)
Benchmark This PR baseline Change
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:12 glibc 1539.800000 ns -
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:12 disjoint_pool<os_provider> 29987.100000 ns -
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:12 jemalloc_pool<os_provider> 36852.600000 ns -
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:12 scalable_pool<os_provider> 2968.720000 ns -
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:12 disjoint_pool<fixed_provider> 30567.300000 ns -
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:12 scalable_pool<fixed_provider> 2359.750000 ns -
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:12 umfProxy 85274.200000 ns -
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:12 jemalloc 640.572000 ns -
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:12 tbbProxy 2068.450000 ns -
Relative perf in group multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:1 (9)
Benchmark This PR baseline Change
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:1 jemalloc 61.690900 ns 63.346 ns 2.68%
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:1 umfProxy 2951.080000 ns 3006.410 ns 1.87%
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:1 glibc 172.757000 ns 174.045 ns 0.75%
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:1 scalable_pool<os_provider> 344.861000 ns 344.975 ns 0.03%
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:1 disjoint_pool<os_provider> 614.099 ns 611.892000 ns -0.36%
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:1 jemalloc_pool<os_provider> 666.193 ns 663.450000 ns -0.41%
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:1 tbbProxy 346.512 ns 344.434000 ns -0.60%
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:1 disjoint_pool<fixed_provider> 579.994 ns 575.710000 ns -0.74%
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:1 scalable_pool<fixed_provider> 354.970 ns 345.592000 ns -2.64%
Relative perf in group multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:50000/threads:1 (2)
Benchmark This PR baseline Change
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:50000/threads:1 proxy_pool<os_provider> 28146.300000 ns 28503.400 ns 1.27%
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:50000/threads:1 os_provider 25133.500000 ns 25211.800 ns 0.31%
Relative perf in group multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:4 (9)
Benchmark This PR baseline Change
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:4 glibc - 1826.460000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:4 disjoint_pool<os_provider> - 32068.300000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:4 jemalloc_pool<os_provider> - 51701.900000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:4 scalable_pool<os_provider> - 1056.040000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:4 disjoint_pool<fixed_provider> - 31481.300000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:4 scalable_pool<fixed_provider> - 1603.000000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:4 umfProxy - 24958.000000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:4 jemalloc - 562.061000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:4 tbbProxy - 1646.770000 ns
Relative perf in group multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:4 (9)
Benchmark This PR baseline Change
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:4 glibc - 2253.580000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:4 disjoint_pool<os_provider> - 37574.100000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:4 jemalloc_pool<os_provider> - 9642.330000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:4 scalable_pool<os_provider> - 1095.730000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:4 disjoint_pool<fixed_provider> - 18181.200000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:4 scalable_pool<fixed_provider> - 1716.890000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:4 umfProxy - 25711.500000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:4 jemalloc - 702.341000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:4 tbbProxy - 1747.890000 ns
Relative perf in group multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:4 (9)
Benchmark This PR baseline Change
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:4 glibc - 906.764000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:4 disjoint_pool<os_provider> - 10579.500000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:4 jemalloc_pool<os_provider> - 8982.490000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:4 scalable_pool<os_provider> - 836.281000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:4 disjoint_pool<fixed_provider> - 10713.900000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:4 scalable_pool<fixed_provider> - 1213.180000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:4 umfProxy - 24628.400000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:4 jemalloc - 367.533000 ns
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:4 tbbProxy - 1199.690000 ns

Details

Benchmark details - environment, command...
multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:12 glibc

Command:

/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/benchmark/umf-benchmark --benchmark_format=csv

multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:1 glibc

Command:

/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/benchmark/umf-benchmark --benchmark_format=csv

multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:12 glibc

Command:

/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/benchmark/umf-benchmark --benchmark_format=csv

multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:1 glibc

Command:

/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/benchmark/umf-benchmark --benchmark_format=csv

multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:12 glibc

Command:

/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/benchmark/umf-benchmark --benchmark_format=csv

multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:1 glibc

Command:

/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/benchmark/umf-benchmark --benchmark_format=csv

multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:50000/threads:1 proxy_pool

Command:

/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/benchmark/umf-benchmark --benchmark_format=csv

multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:50000/threads:1 os_provider

Command:

/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/benchmark/umf-benchmark --benchmark_format=csv

multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:12 disjoint_pool

Command:

/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/benchmark/umf-benchmark --benchmark_format=csv

multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:1 disjoint_pool

Command:

/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/benchmark/umf-benchmark --benchmark_format=csv

multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:12 disjoint_pool

Command:

/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/benchmark/umf-benchmark --benchmark_format=csv

multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:1 disjoint_pool

Command:

/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/benchmark/umf-benchmark --benchmark_format=csv

multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:12 disjoint_pool

Command:

/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/benchmark/umf-benchmark --benchmark_format=csv

multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:1 disjoint_pool

Command:

/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/benchmark/umf-benchmark --benchmark_format=csv

multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:12 jemalloc_pool

Command:

/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/benchmark/umf-benchmark --benchmark_format=csv

multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:1 jemalloc_pool

Command:

/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/benchmark/umf-benchmark --benchmark_format=csv

multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:12 jemalloc_pool

Command:

/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/benchmark/umf-benchmark --benchmark_format=csv

multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:1 jemalloc_pool

Command:

/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/benchmark/umf-benchmark --benchmark_format=csv

multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:12 jemalloc_pool

Command:

/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/benchmark/umf-benchmark --benchmark_format=csv

multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:1 jemalloc_pool

Command:

/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/benchmark/umf-benchmark --benchmark_format=csv

multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:12 scalable_pool

Command:

/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/benchmark/umf-benchmark --benchmark_format=csv

multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:1 scalable_pool

Command:

/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/benchmark/umf-benchmark --benchmark_format=csv

multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:12 scalable_pool

Command:

/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/benchmark/umf-benchmark --benchmark_format=csv

multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:1 scalable_pool

Command:

/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/benchmark/umf-benchmark --benchmark_format=csv

multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:12 scalable_pool

Command:

/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/benchmark/umf-benchmark --benchmark_format=csv

multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:1 scalable_pool

Command:

/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/benchmark/umf-benchmark --benchmark_format=csv

multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:1 proxy_pool

Command:

/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/benchmark/umf-benchmark --benchmark_format=csv

multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:1 fixed_provider

Command:

/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/benchmark/umf-benchmark --benchmark_format=csv

multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:12 disjoint_pool

Command:

/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/benchmark/umf-benchmark --benchmark_format=csv

multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:1 disjoint_pool

Command:

/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/benchmark/umf-benchmark --benchmark_format=csv

multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:12 disjoint_pool

Command:

/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/benchmark/umf-benchmark --benchmark_format=csv

multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:1 disjoint_pool

Command:

/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/benchmark/umf-benchmark --benchmark_format=csv

multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:12 disjoint_pool

Command:

/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/benchmark/umf-benchmark --benchmark_format=csv

multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:1 disjoint_pool

Command:

/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/benchmark/umf-benchmark --benchmark_format=csv

multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:12 scalable_pool

Command:

/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/benchmark/umf-benchmark --benchmark_format=csv

multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:1 scalable_pool

Command:

/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/benchmark/umf-benchmark --benchmark_format=csv

multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:12 scalable_pool

Command:

/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/benchmark/umf-benchmark --benchmark_format=csv

multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:1 scalable_pool

Command:

/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/benchmark/umf-benchmark --benchmark_format=csv

multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:12 scalable_pool

Command:

/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/benchmark/umf-benchmark --benchmark_format=csv

multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:1 scalable_pool

Command:

/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/benchmark/umf-benchmark --benchmark_format=csv

multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:12 umfProxy

Command:

/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/benchmark/umf-benchmark --benchmark_format=csv --benchmark_filter=glibc

Environment Variables:

LD_PRELOAD=/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/lib/libumf_proxy.so

multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:1 umfProxy

Command:

/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/benchmark/umf-benchmark --benchmark_format=csv --benchmark_filter=glibc

Environment Variables:

LD_PRELOAD=/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/lib/libumf_proxy.so

multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:12 umfProxy

Command:

/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/benchmark/umf-benchmark --benchmark_format=csv --benchmark_filter=glibc

Environment Variables:

LD_PRELOAD=/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/lib/libumf_proxy.so

multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:1 umfProxy

Command:

/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/benchmark/umf-benchmark --benchmark_format=csv --benchmark_filter=glibc

Environment Variables:

LD_PRELOAD=/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/lib/libumf_proxy.so

multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:12 umfProxy

Command:

/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/benchmark/umf-benchmark --benchmark_format=csv --benchmark_filter=glibc

Environment Variables:

LD_PRELOAD=/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/lib/libumf_proxy.so

multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:1 umfProxy

Command:

/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/benchmark/umf-benchmark --benchmark_format=csv --benchmark_filter=glibc

Environment Variables:

LD_PRELOAD=/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/lib/libumf_proxy.so

multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:12 jemalloc

Command:

/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/benchmark/umf-benchmark --benchmark_format=csv --benchmark_filter=glibc

Environment Variables:

LD_PRELOAD=libjemalloc.so

multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:1 jemalloc

Command:

/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/benchmark/umf-benchmark --benchmark_format=csv --benchmark_filter=glibc

Environment Variables:

LD_PRELOAD=libjemalloc.so

multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:12 jemalloc

Command:

/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/benchmark/umf-benchmark --benchmark_format=csv --benchmark_filter=glibc

Environment Variables:

LD_PRELOAD=libjemalloc.so

multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:1 jemalloc

Command:

/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/benchmark/umf-benchmark --benchmark_format=csv --benchmark_filter=glibc

Environment Variables:

LD_PRELOAD=libjemalloc.so

multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:12 jemalloc

Command:

/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/benchmark/umf-benchmark --benchmark_format=csv --benchmark_filter=glibc

Environment Variables:

LD_PRELOAD=libjemalloc.so

multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:1 jemalloc

Command:

/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/benchmark/umf-benchmark --benchmark_format=csv --benchmark_filter=glibc

Environment Variables:

LD_PRELOAD=libjemalloc.so

multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:12 tbbProxy

Command:

/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/benchmark/umf-benchmark --benchmark_format=csv --benchmark_filter=glibc

Environment Variables:

LD_PRELOAD=libtbbmalloc_proxy.so

multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/size:4096/iterations:500000/threads:1 tbbProxy

Command:

/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/benchmark/umf-benchmark --benchmark_format=csv --benchmark_filter=glibc

Environment Variables:

LD_PRELOAD=libtbbmalloc_proxy.so

multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:12 tbbProxy

Command:

/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/benchmark/umf-benchmark --benchmark_format=csv --benchmark_filter=glibc

Environment Variables:

LD_PRELOAD=libtbbmalloc_proxy.so

multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:4096/granularity:8/iterations:500000/threads:1 tbbProxy

Command:

/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/benchmark/umf-benchmark --benchmark_format=csv --benchmark_filter=glibc

Environment Variables:

LD_PRELOAD=libtbbmalloc_proxy.so

multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:12 tbbProxy

Command:

/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/benchmark/umf-benchmark --benchmark_format=csv --benchmark_filter=glibc

Environment Variables:

LD_PRELOAD=libtbbmalloc_proxy.so

multiple_malloc_free/max_allocs:10000/thread_local_allocations:1/min_size:8/max_size:128/granularity:8/iterations:500000/threads:1 tbbProxy

Command:

/home/test-user/actions-runners/umf-perf-runner/_work/unified-memory-framework/unified-memory-framework/umf-repo/build/benchmark/umf-benchmark --benchmark_format=csv --benchmark_filter=glibc

Environment Variables:

LD_PRELOAD=libtbbmalloc_proxy.so

@lplewa lplewa force-pushed the benchmarks_threads branch from 66e8451 to e316fdc Compare March 17, 2025 16:59
@lplewa lplewa marked this pull request as ready for review March 17, 2025 16:59
@lplewa lplewa requested a review from a team as a code owner March 17, 2025 16:59
@lukaszstolarczuk lukaszstolarczuk merged commit 4c54366 into oneapi-src:main Mar 18, 2025
249 of 251 checks passed
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Reviewers

@KFilipek KFilipek KFilipek approved these changes

@ldorau ldorau ldorau approved these changes

@bratpiorka bratpiorka bratpiorka approved these changes

@PatKamin PatKamin Awaiting requested review from PatKamin

Assignees
No one assigned
Labels
None yet
Projects
None yet
Milestone
No milestone
Development

Successfully merging this pull request may close these issues.
5 participants
@lplewa @KFilipek @ldorau @bratpiorka @lukaszstolarczuk