Skip to content

Commit

Permalink
update libMLIR_h bindings and fix bindings update step n°2 (#202)
Browse files Browse the repository at this point in the history 
* update libMLIR_h.jl

* fetch lazy artifacts in pre-process step
  • Loading branch information
@Pangoraw
Pangoraw authored Oct 28, 2024
1 parent e084f9e commit 071df34
Show file tree
Hide file tree
Showing 2 changed files with 1,385 additions and 262 deletions.
2 changes: 1 addition & 1 deletion .github/workflows/regenerate-mlir-bindings.yml
View file
Original file line number Diff line number Diff line change
Expand Up @@ -22,7 +22,7 @@ jobs:
run: |
import Pkg
Pkg.add("JuliaFormatter")
- run: julia --project=. -e 'using Pkg; pkg"instantiate"; pkg"update"; pkg"precompile"'
- run: julia --project=. -e 'using Pkg; Pkg.instantiate(); using Clang; Clang.JLLEnvs.get_system_includes()'
working-directory: ./deps/ReactantExtra
env:
JULIA_DEPOT_PATH: ${{ runner.temp }}/julia_depot
Expand Down
Loading

1 comment on commit 071df34

@github-actions
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Reactant.jl Benchmarks

Benchmark suite Current: 071df34 Previous: e084f9e Ratio
ViT base (256 x 256 x 3 x 32)/forward/CUDA/Reactant (optimize = :after_enzyme) 1293777162 ns 1350735972 ns 0.96
ViT base (256 x 256 x 3 x 32)/forward/CUDA/Reactant 1310912331 ns 1204673780 ns 1.09
ViT base (256 x 256 x 3 x 32)/forward/CUDA/Reactant (optimize = :before_enzyme) 1289076756 ns 1222340041 ns 1.05
ViT base (256 x 256 x 3 x 32)/forward/CUDA/Reactant (optimize = :only_enzyme) 2516467034 ns 2330048189 ns 1.08
ViT base (256 x 256 x 3 x 32)/forward/CUDA/Lux 218752184 ns 219309808 ns 1.00
ViT base (256 x 256 x 3 x 32)/forward/CPU/Reactant (optimize = :after_enzyme) 5129809896 ns 5215948763 ns 0.98
ViT base (256 x 256 x 3 x 32)/forward/CPU/Reactant 5430197585 ns 6254101434 ns 0.87
ViT base (256 x 256 x 3 x 32)/forward/CPU/Reactant (optimize = :before_enzyme) 4979540725 ns 5201734622 ns 0.96
ViT base (256 x 256 x 3 x 32)/forward/CPU/Reactant (optimize = :only_enzyme) 6919614889 ns 7460701472 ns 0.93
ViT base (256 x 256 x 3 x 32)/forward/CPU/Lux 38413778097 ns 30735680811 ns 1.25
ViT small (256 x 256 x 3 x 4)/forward/CUDA/Reactant (optimize = :after_enzyme) 1278646924 ns 1332885325 ns 0.96
ViT small (256 x 256 x 3 x 4)/forward/CUDA/Reactant 1277917926 ns 1298553637 ns 0.98
ViT small (256 x 256 x 3 x 4)/forward/CUDA/Reactant (optimize = :before_enzyme) 1257911932 ns 1340798546.5 ns 0.94
ViT small (256 x 256 x 3 x 4)/forward/CUDA/Reactant (optimize = :only_enzyme) 2531611781 ns 2364390446 ns 1.07
ViT small (256 x 256 x 3 x 4)/forward/CUDA/Lux 8801215 ns 8812264.5 ns 1.00
ViT small (256 x 256 x 3 x 4)/forward/CPU/Reactant (optimize = :after_enzyme) 1574880615 ns 1635773422 ns 0.96
ViT small (256 x 256 x 3 x 4)/forward/CPU/Reactant 1547093404 ns 1621494476 ns 0.95
ViT small (256 x 256 x 3 x 4)/forward/CPU/Reactant (optimize = :before_enzyme) 1545080349.5 ns 1609772124 ns 0.96
ViT small (256 x 256 x 3 x 4)/forward/CPU/Reactant (optimize = :only_enzyme) 2756285160 ns 2830061683 ns 0.97
ViT small (256 x 256 x 3 x 4)/forward/CPU/Lux 2982928651 ns 2877521675 ns 1.04
ViT tiny (256 x 256 x 3 x 32)/forward/CUDA/Reactant (optimize = :after_enzyme) 1228305920 ns 1310197829 ns 0.94
ViT tiny (256 x 256 x 3 x 32)/forward/CUDA/Reactant 1266393159.5 ns 1304840979.5 ns 0.97
ViT tiny (256 x 256 x 3 x 32)/forward/CUDA/Reactant (optimize = :before_enzyme) 1237209870.5 ns 1238123602.5 ns 1.00
ViT tiny (256 x 256 x 3 x 32)/forward/CUDA/Reactant (optimize = :only_enzyme) 2559979407 ns 2594487217 ns 0.99
ViT tiny (256 x 256 x 3 x 32)/forward/CUDA/Lux 22504460 ns 22665498 ns 0.99
ViT tiny (256 x 256 x 3 x 32)/forward/CPU/Reactant (optimize = :after_enzyme) 2130810338 ns 2163350638 ns 0.98
ViT tiny (256 x 256 x 3 x 32)/forward/CPU/Reactant 2154516099 ns 2227062239 ns 0.97
ViT tiny (256 x 256 x 3 x 32)/forward/CPU/Reactant (optimize = :before_enzyme) 2156781390 ns 2171553343 ns 0.99
ViT tiny (256 x 256 x 3 x 32)/forward/CPU/Reactant (optimize = :only_enzyme) 3372274619 ns 3558585268 ns 0.95
ViT tiny (256 x 256 x 3 x 32)/forward/CPU/Lux 6137706960 ns 7747467151.5 ns 0.79
ViT tiny (256 x 256 x 3 x 4)/forward/CUDA/Reactant (optimize = :after_enzyme) 1307625738.5 ns 1283318913 ns 1.02
ViT tiny (256 x 256 x 3 x 4)/forward/CUDA/Reactant 1350711438 ns 1216591656.5 ns 1.11
ViT tiny (256 x 256 x 3 x 4)/forward/CUDA/Reactant (optimize = :before_enzyme) 1313441438.5 ns 1370999738 ns 0.96
ViT tiny (256 x 256 x 3 x 4)/forward/CUDA/Reactant (optimize = :only_enzyme) 2565794827 ns 2428504820 ns 1.06
ViT tiny (256 x 256 x 3 x 4)/forward/CUDA/Lux 6571875 ns 7186419 ns 0.91
ViT tiny (256 x 256 x 3 x 4)/forward/CPU/Reactant (optimize = :after_enzyme) 1420391837 ns 1467788340 ns 0.97
ViT tiny (256 x 256 x 3 x 4)/forward/CPU/Reactant 1423519966 ns 1449340906 ns 0.98
ViT tiny (256 x 256 x 3 x 4)/forward/CPU/Reactant (optimize = :before_enzyme) 1414508660 ns 1443567043 ns 0.98
ViT tiny (256 x 256 x 3 x 4)/forward/CPU/Reactant (optimize = :only_enzyme) 2618724343 ns 2702131547 ns 0.97
ViT tiny (256 x 256 x 3 x 4)/forward/CPU/Lux 1347152220 ns 1095865115 ns 1.23
ViT tiny (256 x 256 x 3 x 16)/forward/CUDA/Reactant (optimize = :after_enzyme) 1242600424.5 ns 1210758212 ns 1.03
ViT tiny (256 x 256 x 3 x 16)/forward/CUDA/Reactant 1239539558.5 ns 1233393897 ns 1.00
ViT tiny (256 x 256 x 3 x 16)/forward/CUDA/Reactant (optimize = :before_enzyme) 1252776529 ns 1225142106.5 ns 1.02
ViT tiny (256 x 256 x 3 x 16)/forward/CUDA/Reactant (optimize = :only_enzyme) 2517090435 ns 2374047383 ns 1.06
ViT tiny (256 x 256 x 3 x 16)/forward/CUDA/Lux 12117214 ns 12131784 ns 1.00
ViT tiny (256 x 256 x 3 x 16)/forward/CPU/Reactant (optimize = :after_enzyme) 1692247967 ns 1757223123 ns 0.96
ViT tiny (256 x 256 x 3 x 16)/forward/CPU/Reactant 1691308826 ns 1753678153 ns 0.96
ViT tiny (256 x 256 x 3 x 16)/forward/CPU/Reactant (optimize = :before_enzyme) 1686383037 ns 1747793752 ns 0.96
ViT tiny (256 x 256 x 3 x 16)/forward/CPU/Reactant (optimize = :only_enzyme) 2886906039 ns 3015378976 ns 0.96
ViT tiny (256 x 256 x 3 x 16)/forward/CPU/Lux 3101540732 ns 2910278832 ns 1.07
ViT small (256 x 256 x 3 x 16)/forward/CUDA/Reactant (optimize = :after_enzyme) 1275391377 ns 1322233467 ns 0.96
ViT small (256 x 256 x 3 x 16)/forward/CUDA/Reactant 1280225766.5 ns 1322254383 ns 0.97
ViT small (256 x 256 x 3 x 16)/forward/CUDA/Reactant (optimize = :before_enzyme) 1212838454 ns 1199964457 ns 1.01
ViT small (256 x 256 x 3 x 16)/forward/CUDA/Reactant (optimize = :only_enzyme) 2378532850 ns 2400347071 ns 0.99
ViT small (256 x 256 x 3 x 16)/forward/CUDA/Lux 27266014 ns 27163900 ns 1.00
ViT small (256 x 256 x 3 x 16)/forward/CPU/Reactant (optimize = :after_enzyme) 2178479799 ns 2214089958 ns 0.98
ViT small (256 x 256 x 3 x 16)/forward/CPU/Reactant 2152860611 ns 2239028395 ns 0.96
ViT small (256 x 256 x 3 x 16)/forward/CPU/Reactant (optimize = :before_enzyme) 2150176632 ns 2212157343 ns 0.97
ViT small (256 x 256 x 3 x 16)/forward/CPU/Reactant (optimize = :only_enzyme) 3383375992 ns 3527800382 ns 0.96
ViT small (256 x 256 x 3 x 16)/forward/CPU/Lux 6759656709 ns 5811115543.5 ns 1.16
ViT small (256 x 256 x 3 x 32)/forward/CUDA/Reactant (optimize = :after_enzyme) 1184096696 ns 1226440738 ns 0.97
ViT small (256 x 256 x 3 x 32)/forward/CUDA/Reactant 1201129956 ns 1160529232.5 ns 1.03
ViT small (256 x 256 x 3 x 32)/forward/CUDA/Reactant (optimize = :before_enzyme) 1284430231.5 ns 1284160996.5 ns 1.00
ViT small (256 x 256 x 3 x 32)/forward/CUDA/Reactant (optimize = :only_enzyme) 2495514608 ns 2249108448 ns 1.11
ViT small (256 x 256 x 3 x 32)/forward/CUDA/Lux 52594359 ns 52687663 ns 1.00
ViT small (256 x 256 x 3 x 32)/forward/CPU/Reactant (optimize = :after_enzyme) 2992060164 ns 3019468880 ns 0.99
ViT small (256 x 256 x 3 x 32)/forward/CPU/Reactant 2979508219 ns 3069044359 ns 0.97
ViT small (256 x 256 x 3 x 32)/forward/CPU/Reactant (optimize = :before_enzyme) 2970649475 ns 3044926123 ns 0.98
ViT small (256 x 256 x 3 x 32)/forward/CPU/Reactant (optimize = :only_enzyme) 4313180920 ns 4458047333 ns 0.97
ViT small (256 x 256 x 3 x 32)/forward/CPU/Lux 9775441963 ns 8346217965 ns 1.17
ViT base (256 x 256 x 3 x 16)/forward/CUDA/Reactant (optimize = :after_enzyme) 1216534742 ns 1248064916 ns 0.97
ViT base (256 x 256 x 3 x 16)/forward/CUDA/Reactant 1250449877 ns 1265067809.5 ns 0.99
ViT base (256 x 256 x 3 x 16)/forward/CUDA/Reactant (optimize = :before_enzyme) 1221392049.5 ns 1274798656.5 ns 0.96
ViT base (256 x 256 x 3 x 16)/forward/CUDA/Reactant (optimize = :only_enzyme) 2283650842 ns 2435701548 ns 0.94
ViT base (256 x 256 x 3 x 16)/forward/CUDA/Lux 70872798 ns 70937902 ns 1.00
ViT base (256 x 256 x 3 x 16)/forward/CPU/Reactant (optimize = :after_enzyme) 3168519792 ns 3268589369 ns 0.97
ViT base (256 x 256 x 3 x 16)/forward/CPU/Reactant 3126643460 ns 3233353228 ns 0.97
ViT base (256 x 256 x 3 x 16)/forward/CPU/Reactant (optimize = :before_enzyme) 3149006217 ns 3264816245 ns 0.96
ViT base (256 x 256 x 3 x 16)/forward/CPU/Reactant (optimize = :only_enzyme) 4562237724 ns 4685900819 ns 0.97
ViT base (256 x 256 x 3 x 16)/forward/CPU/Lux 15598056402 ns 15059348114 ns 1.04
ViT base (256 x 256 x 3 x 4)/forward/CUDA/Reactant (optimize = :after_enzyme) 1251135175 ns 1293711217 ns 0.97
ViT base (256 x 256 x 3 x 4)/forward/CUDA/Reactant 1253121520 ns 1320275711 ns 0.95
ViT base (256 x 256 x 3 x 4)/forward/CUDA/Reactant (optimize = :before_enzyme) 1218759552.5 ns 1275143769 ns 0.96
ViT base (256 x 256 x 3 x 4)/forward/CUDA/Reactant (optimize = :only_enzyme) 2266543596 ns 2411331594 ns 0.94
ViT base (256 x 256 x 3 x 4)/forward/CUDA/Lux 20767815 ns 20583569 ns 1.01
ViT base (256 x 256 x 3 x 4)/forward/CPU/Reactant (optimize = :after_enzyme) 1867500661 ns 2119165721 ns 0.88
ViT base (256 x 256 x 3 x 4)/forward/CPU/Reactant 1832460010 ns 2218102344 ns 0.83
ViT base (256 x 256 x 3 x 4)/forward/CPU/Reactant (optimize = :before_enzyme) 1841480266 ns 2186413098 ns 0.84
ViT base (256 x 256 x 3 x 4)/forward/CPU/Reactant (optimize = :only_enzyme) 3075968823 ns 3188249362 ns 0.96
ViT base (256 x 256 x 3 x 4)/forward/CPU/Lux 4287256056 ns 3741883940 ns 1.15

This comment was automatically generated by workflow using github-action-benchmark.

Please sign in to comment.