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.

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

Add PGO builds #751

Open
chfast opened this issue Mar 6, 2021 · 2 comments
Open

Add PGO builds #751

chfast opened this issue Mar 6, 2021 · 2 comments
Labels
optimization Performance optimization

Comments

@chfast
Copy link
Collaborator

chfast commented Mar 6, 2021

The PGO builds give good boost to the performance, but we need to select out training set.

Comparing o/master to ../pgouse/o/master
Benchmark                                                     Time             CPU      Time Old      Time New       CPU Old       CPU New
------------------------------------------------------------------------------------------------------------------------------------------
fizzy/execute/blake2b/512_bytes_rounds_1                   -0.1681         -0.1740            99            82            99            82
fizzy/execute/blake2b/512_bytes_rounds_16                  -0.2512         -0.2512          1503          1125          1503          1125
fizzy/execute/ecpairing/onepoint                           -0.1878         -0.1878        473759        384788        473721        384762
fizzy/execute/keccak256/512_bytes_rounds_1                 -0.3023         -0.3023           116            81           116            81
fizzy/execute/keccak256/512_bytes_rounds_16                -0.3153         -0.3152          1713          1173          1713          1173
fizzy/execute/memset/256_bytes                             -0.2147         -0.2147             8             6             8             6
fizzy/execute/memset/60000_bytes                           -0.2239         -0.2239          1659          1287          1659          1287
fizzy/execute/mul256_opt0/input1                           -0.3080         -0.3079            33            23            33            23
fizzy/execute/ramanujan_pi/33_runs                         -0.3250         -0.3250           151           102           151           102
fizzy/execute/sha1/512_bytes_rounds_1                      -0.2244         -0.2244           104            81           104            81
fizzy/execute/sha1/512_bytes_rounds_16                     -0.2124         -0.2124          1443          1137          1443          1137
fizzy/execute/sha256/512_bytes_rounds_1                    -0.2048         -0.2048            97            77            97            77
fizzy/execute/sha256/512_bytes_rounds_16                   -0.1844         -0.1844          1323          1079          1323          1079
fizzy/execute/taylor_pi/pi_1000000_runs                    -0.0811         -0.0812         42727         39260         42724         39257
fizzy/execute/micro/eli_interpreter/exec105                -0.2854         -0.2854             6             4             6             4
fizzy/execute/micro/factorial/20                           -0.1208         -0.1208             1             1             1             1
fizzy/execute/micro/fibonacci/24                           -0.1348         -0.1348          5828          5042          5827          5042
fizzy/execute/micro/host_adler32/1                         -0.1522         -0.1522             0             0             0             0
fizzy/execute/micro/host_adler32/1000                      -0.2068         -0.2066            33            26            33            26
fizzy/execute/micro/icall_hash/1000_steps                  -0.1021         -0.1021            72            65            72            65
fizzy/execute/micro/spinner/1                              -0.1430         -0.1430             0             0             0             0
fizzy/execute/micro/spinner/1000                           -0.2250         -0.2250            10             8            10             8
@axic
Copy link
Member

axic commented Mar 8, 2021

Looks nice! What set did you run PGO on?

@chfast
Copy link
Collaborator Author

chfast commented Mar 8, 2021

Just fizzy-bench.

@axic axic added the optimization Performance optimization label Mar 29, 2021
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
optimization Performance optimization
Projects
None yet
Development

No branches or pull requests

2 participants