Skip to content

Instantly share code, notes, and snippets.

@tanakamura
Created December 12, 2020 13:21
Show Gist options
  • Save tanakamura/820dd8a7955abdaabe8a7bb4ccf1568c to your computer and use it in GitHub Desktop.
Save tanakamura/820dd8a7955abdaabe8a7bb4ccf1568c to your computer and use it in GitHub Desktop.
membw-release
ostimer: clock_gettime
userland_timer: rdtscp
use_cpucyle: no
==== membw_1t ====
| |MiB/sec
=================================
| simple-long-copy |23061.14028
---------------------------------
| libc-memcpy |30647.25286
---------------------------------
| gccvec128-copy |23222.25079
---------------------------------
| sse-stream-copy |29031.68858
---------------------------------
| avx256-copy |19672.56514
---------------------------------
| avx512-copy |21077.17218
---------------------------------
| x86-rep-movs1 |24709.44269
---------------------------------
| x86-rep-movs2 |24493.94709
---------------------------------
| x86-rep-movs4 |24621.54765
---------------------------------
| simple-long-sum |18099.60396
---------------------------------
| gccvec128-load |17087.87951
---------------------------------
| avx256-load |21639.78757
---------------------------------
| avx512-load |19560.91241
---------------------------------
| x86-rep-scas1 | 1987.68205
---------------------------------
| x86-rep-scas2 | 3980.28054
---------------------------------
| x86-rep-scas4 | 7675.72196
---------------------------------
| libc-memset |32139.14959
---------------------------------
| simple-long-store |32380.32919
---------------------------------
| gccvec128-store |32183.77025
---------------------------------
| sse-stream-store |37651.30423
---------------------------------
| avx256-store |16199.43685
---------------------------------
|avx256-stream-store |36821.98625
---------------------------------
| avx512-store |14486.51131
---------------------------------
|avx512-stream-store |37183.07848
---------------------------------
| x86-rep-stos1 |37690.32077
---------------------------------
| x86-rep-stos2 |37686.77444
---------------------------------
| x86-rep-stos4 |37611.89814
---------------------------------
v : test_name
==== membw_mt ====
| |MiB/sec
=================================
| simple-long-copy |37906.15087
---------------------------------
| libc-memcpy |31606.11660
---------------------------------
| gccvec128-copy |38253.97419
---------------------------------
| sse-stream-copy |32107.87994
---------------------------------
| avx256-copy |34670.56255
---------------------------------
| avx512-copy |29534.36102
---------------------------------
| x86-rep-movs1 |31807.50252
---------------------------------
| x86-rep-movs2 |31950.60432
---------------------------------
| x86-rep-movs4 |31979.60606
---------------------------------
| simple-long-sum |42745.82870
---------------------------------
| gccvec128-load |39412.48345
---------------------------------
| avx256-load |42135.37085
---------------------------------
| avx512-load |45014.30249
---------------------------------
| x86-rep-scas1 | 6315.20725
---------------------------------
| x86-rep-scas2 |13899.63699
---------------------------------
| x86-rep-scas4 |26347.83150
---------------------------------
| libc-memset |35082.87555
---------------------------------
| simple-long-store |35210.08822
---------------------------------
| gccvec128-store |35356.64049
---------------------------------
| sse-stream-store |35357.57442
---------------------------------
| avx256-store |26725.86464
---------------------------------
|avx256-stream-store |35905.61204
---------------------------------
| avx512-store |23499.26808
---------------------------------
|avx512-stream-store |36622.17425
---------------------------------
| x86-rep-stos1 |37021.94285
---------------------------------
| x86-rep-stos2 |36852.71857
---------------------------------
| x86-rep-stos4 |36580.25603
---------------------------------
v : test_name
ostimer: clock_gettime
userland_timer: rdtscp
use_cpucyle: no
==== membw_1t ====
| |MiB/sec
=================================
| simple-long-copy |17121.85149
---------------------------------
| libc-memcpy |33919.57685
---------------------------------
| gccvec128-copy |17346.93274
---------------------------------
| sse-stream-copy |33245.64575
---------------------------------
| avx256-copy |17247.10187
---------------------------------
| x86-rep-movs1 |15237.31231
---------------------------------
| x86-rep-movs2 |15199.66915
---------------------------------
| x86-rep-movs4 |15225.22821
---------------------------------
| simple-long-sum |18873.81914
---------------------------------
| gccvec128-load |24741.18684
---------------------------------
| avx256-load |26101.13705
---------------------------------
| x86-rep-scas1 | 2002.49796
---------------------------------
| x86-rep-scas2 | 3906.55904
---------------------------------
| x86-rep-scas4 | 7367.45993
---------------------------------
| libc-memset |12377.01423
---------------------------------
| simple-long-store |12383.17123
---------------------------------
| gccvec128-store |12375.87737
---------------------------------
| sse-stream-store |24183.44625
---------------------------------
| avx256-store |12521.12146
---------------------------------
|avx256-stream-store |24174.91251
---------------------------------
| x86-rep-stos1 |11809.75082
---------------------------------
| x86-rep-stos2 |11791.23546
---------------------------------
| x86-rep-stos4 |11777.72853
---------------------------------
v : test_name
==== membw_mt ====
| |MiB/sec
=================================
| simple-long-copy |15643.31748
---------------------------------
| libc-memcpy |16310.59826
---------------------------------
| gccvec128-copy |15673.33577
---------------------------------
| sse-stream-copy |31470.43728
---------------------------------
| avx256-copy |16202.02408
---------------------------------
| x86-rep-movs1 |18181.19547
---------------------------------
| x86-rep-movs2 |18204.57788
---------------------------------
| x86-rep-movs4 |18237.49746
---------------------------------
| simple-long-sum |32121.45498
---------------------------------
| gccvec128-load |31886.97492
---------------------------------
| avx256-load |31959.10308
---------------------------------
| x86-rep-scas1 |14706.48811
---------------------------------
| x86-rep-scas2 |29464.08412
---------------------------------
| x86-rep-scas4 |32184.99919
---------------------------------
| libc-memset |11873.63042
---------------------------------
| simple-long-store |11874.02621
---------------------------------
| gccvec128-store |11874.80319
---------------------------------
| sse-stream-store |24032.46191
---------------------------------
| avx256-store |11963.84517
---------------------------------
|avx256-stream-store |24034.70226
---------------------------------
| x86-rep-stos1 |13184.53377
---------------------------------
| x86-rep-stos2 |13076.57022
---------------------------------
| x86-rep-stos4 |13096.20906
---------------------------------
v : test_name
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment