Skip to content

Instantly share code, notes, and snippets.

@tanakamura
Created December 11, 2020 17:43
Show Gist options
  • Save tanakamura/64b72dd879a03c37c01567097fdd44c7 to your computer and use it in GitHub Desktop.
Save tanakamura/64b72dd879a03c37c01567097fdd44c7 to your computer and use it in GitHub Desktop.
membw
ostimer: clock_gettime
userland_timer: rdtscp
use_cpucyle: no
==== membw_1t ====
| |MiB/sec
=================================
| simple-long-copy | 8626.01679
---------------------------------
| libc-memcpy |16827.83568
---------------------------------
| gccvec128-copy |15736.69427
---------------------------------
| avx256-copy |12487.78551
---------------------------------
| avx512-copy |15291.04252
---------------------------------
| x86-rep-movs1 |16832.51263
---------------------------------
| x86-rep-movs2 |16825.93715
---------------------------------
| x86-rep-movs4 |16853.70625
---------------------------------
| simple-long-sum | 3514.78857
---------------------------------
| gccvec128-load |11933.91781
---------------------------------
| avx256-load |13151.61736
---------------------------------
| avx512-load |16177.55816
---------------------------------
| x86-rep-scas1 | 1964.07480
---------------------------------
| x86-rep-scas2 | 3908.75604
---------------------------------
| x86-rep-scas4 | 7566.66647
---------------------------------
| libc-memset |18719.22074
---------------------------------
| simple-long-store | 3859.03517
---------------------------------
| gccvec128-store | 8701.54418
---------------------------------
| sse-stream-store | 7810.85641
---------------------------------
| avx256-store | 7437.18614
---------------------------------
|avx256-stream-store |18332.59159
---------------------------------
| avx512-store | 7968.94657
---------------------------------
|avx512-stream-store |18825.27303
---------------------------------
| x86-rep-stos1 |19177.97698
---------------------------------
| x86-rep-stos2 |19175.46311
---------------------------------
| x86-rep-stos4 |19221.25982
---------------------------------
v : test_name
==== membw_mt ====
| |MiB/sec
=================================
| simple-long-copy |11485.98692
---------------------------------
| libc-memcpy |15448.70731
---------------------------------
| gccvec128-copy |19084.46169
---------------------------------
| avx256-copy |16871.95492
---------------------------------
| avx512-copy |16182.95448
---------------------------------
| x86-rep-movs1 |16190.15853
---------------------------------
| x86-rep-movs2 |16177.72331
---------------------------------
| x86-rep-movs4 |16178.24903
---------------------------------
| simple-long-sum |13419.08993
---------------------------------
| gccvec128-load |21375.51947
---------------------------------
| avx256-load |21202.63976
---------------------------------
| avx512-load |21211.32513
---------------------------------
| x86-rep-scas1 | 4831.02138
---------------------------------
| x86-rep-scas2 | 9633.13392
---------------------------------
| x86-rep-scas4 |21350.19603
---------------------------------
| libc-memset |18016.47619
---------------------------------
| simple-long-store | 8465.40324
---------------------------------
| gccvec128-store |18421.63989
---------------------------------
| sse-stream-store | 5183.06645
---------------------------------
| avx256-store |16239.95960
---------------------------------
|avx256-stream-store |14949.69632
---------------------------------
| avx512-store |15665.39353
---------------------------------
|avx512-stream-store |19950.98561
---------------------------------
| x86-rep-stos1 |21323.15696
---------------------------------
| x86-rep-stos2 |21081.28530
---------------------------------
| x86-rep-stos4 |21177.97636
---------------------------------
v : test_name
ostimer: clock_gettime
userland_timer: rdtscp
use_cpucyle: no
==== membw_1t ====
| |MiB/sec
=================================
| simple-long-copy | 8240.20986
---------------------------------
| libc-memcpy |30729.89499
---------------------------------
| gccvec128-copy |15315.74866
---------------------------------
| avx256-copy |16827.42409
---------------------------------
| x86-rep-movs1 |15344.81789
---------------------------------
| x86-rep-movs2 |15390.76913
---------------------------------
| x86-rep-movs4 |15405.88845
---------------------------------
| simple-long-sum | 4552.18162
---------------------------------
| gccvec128-load |15519.26063
---------------------------------
| avx256-load |16973.75039
---------------------------------
| x86-rep-scas1 | 2013.29473
---------------------------------
| x86-rep-scas2 | 3913.59388
---------------------------------
| x86-rep-scas4 | 7131.12876
---------------------------------
| libc-memset |13461.70626
---------------------------------
| simple-long-store | 4497.35390
---------------------------------
| gccvec128-store | 7291.21804
---------------------------------
| sse-stream-store |17334.29030
---------------------------------
| avx256-store |13840.30006
---------------------------------
|avx256-stream-store |18255.29689
---------------------------------
| x86-rep-stos1 |12820.89971
---------------------------------
| x86-rep-stos2 |12804.82399
---------------------------------
| x86-rep-stos4 |12766.54410
---------------------------------
v : test_name
==== membw_mt ====
| |MiB/sec
=================================
| simple-long-copy |21618.74738
---------------------------------
| libc-memcpy |22239.31641
---------------------------------
| gccvec128-copy |21318.69990
---------------------------------
| avx256-copy |21094.55489
---------------------------------
| x86-rep-movs1 |22028.59084
---------------------------------
| x86-rep-movs2 |21984.53335
---------------------------------
| x86-rep-movs4 |21978.43441
---------------------------------
| simple-long-sum |34563.38525
---------------------------------
| gccvec128-load |34331.02449
---------------------------------
| avx256-load |33715.25639
---------------------------------
| x86-rep-scas1 |15622.96985
---------------------------------
| x86-rep-scas2 |30084.77091
---------------------------------
| x86-rep-scas4 |35121.87741
---------------------------------
| libc-memset |15164.92536
---------------------------------
| simple-long-store |15739.06507
---------------------------------
| gccvec128-store |15414.05011
---------------------------------
| sse-stream-store |18448.13995
---------------------------------
| avx256-store |14777.56825
---------------------------------
|avx256-stream-store |18406.64456
---------------------------------
| x86-rep-stos1 |15352.57758
---------------------------------
| x86-rep-stos2 |15478.34531
---------------------------------
| x86-rep-stos4 |15361.13116
---------------------------------
v : test_name
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment