Skip to content

test#1

Open
yolo2themoon wants to merge 1 commit intomainfrom
test
Open

test#1
yolo2themoon wants to merge 1 commit intomainfrom
test

Conversation

@yolo2themoon
Copy link
Copy Markdown
Owner

No description provided.

@yolo2themoon
Copy link
Copy Markdown
Owner Author

Examples(CUDA) ad_gravity comet cornell_box fem128 fem99 fractal game_of_life inital_value_problem mpm128 mpm3d mpm88 mpm99 mpm_lagrangian_forces nbody odop_solar pbf2d sdf_renderer simple_uv taichi_logo vortex_rings
kernel_elapsed_time ( current/master ) -4.0689% +0.4925% +0.1737% -0.2683% +1.3329% +0.0023% +1.3061% -0.0604% +0.2095% -0.1286% +0.2169% +0.2303% +1.3744% +0.5612% -1.8322% -0.4912% -0.1844% -0.0031% +0.3755% +0.4838%
end2end_time ( current/master ) +0.4727% +0.4727% +0.4727% +0.4727% +0.4727% +0.4727% +0.4727% +0.4727% +0.4727% +0.4727% +0.4727% +0.4727% +0.4727% +0.4727% +0.4727% +0.4727% +0.4727% +0.4727% +0.4727% +0.4727%

@yolo2themoon
Copy link
Copy Markdown
Owner Author

yolo2themoon commented Dec 18, 2021

current commit hash: 7e83cc583b0338b69c4da58077700bc48f1d7fba

master commit hash: d45680bdfe2001341f463e836700d45688864c8e

Examples(CUDA) ad_gravity comet cornell_box fem128 fem99 fractal game_of_life inital_value_problem mpm128 mpm3d mpm88 mpm99 mpm_lagrangian_forces nbody odop_solar pbf2d sdf_renderer simple_uv taichi_logo vortex_rings
kernel_elapsed_time ( current/master ) +541.6487% +16.8578% +6.0112% +491.0853% +533.7113% -11.8045% +17.9971% +14.1247% +8.996% -26.6261% +29.4145% +25.4136% +459.0929% +385.1374% +5.3686% +96.9254% +96.1534% +5.8248% +7.03% +2.9578%
end2end_time ( current/master ) +0.4727% +0.0782% +0.1483% -0.1772% -0.3345% +0.9384% -0.028% +0.4657% -4.8292% +0.4248% +0.0857% +1.1165% -1.6351% +0.8418% +1.3449% +2.2758% +0.0745% -0.0187% +0.3216% +0.3686%
benchmarks fluctuation threshold 5.0%
kernel improvement mpm3d(-26.6261%), fractal(-11.8045%),
kernel regression odop_solar(+5.3686%), simple_uv(+5.8248%), cornell_box(+6.0112%), taichi_logo(+7.03%), mpm128(+8.996%), inital_value_problem(+14.1247%), comet(+16.8578%), game_of_life(+17.9971%), mpm99(+25.4136%), mpm88(+29.4145%), sdf_renderer(+96.1534%), pbf2d(+96.9254%), nbody(+385.1374%), mpm_lagrangian_forces(+459.0929%), fem128(+491.0853%), fem99(+533.7113%), ad_gravity(+541.6487%),
end2end improvement None
end2end regression None

@yolo2themoon
Copy link
Copy Markdown
Owner Author

current commit hash: 7e83cc583b0338b69c4da58077700bc48f1d7fba

master commit hash: d45680bdfe2001341f463e836700d45688864c8e

Examples(CUDA) ad_gravity comet cornell_box fem128 fem99 fractal game_of_life inital_value_problem mpm128 mpm3d mpm88 mpm99 mpm_lagrangian_forces nbody odop_solar pbf2d sdf_renderer simple_uv taichi_logo vortex_rings
kernel_elapsed_time ( current/master ) -4.0689% +0.4925% +0.1737% -0.2683% +1.3329% +0.0023% +1.3061% -0.0604% +0.2095% -0.1286% +0.2169% +0.2303% +1.3744% +0.5612% -1.8322% -0.4912% -0.1844% -0.0031% +0.3755% +0.4838%
end2end_time ( current/master ) +0.4727% +0.0782% +0.1483% -0.1772% -0.3345% +0.9384% -0.028% +0.4657% -4.8292% +0.4248% +0.0857% +1.1165% -1.6351% +0.8418% +1.3449% +2.2758% +0.0745% -0.0187% +0.3216% +0.3686%
benchmarks fluctuation threshold 5.0%
kernel improvement None
kernel regression None
end2end improvement None
end2end regression None

@yolo2themoon
Copy link
Copy Markdown
Owner Author

current commit hash: 6810930f6b0bbe8a513d1a5369b9c66f2628f2d4

master commit hash: f1dde07398a9e8aaf62f0c75cd81bef1bd82a574

Examples CUDA
benchmarks fluctuation threshold 5.0%
kernel improvement fem128(-5.2555%),
kernel regression None
end2end improvement None
end2end regression None

@yolo2themoon
Copy link
Copy Markdown
Owner Author

current commit hash: 6810930f6b0bbe8a513d1a5369b9c66f2628f2d4

master commit hash: f1dde07398a9e8aaf62f0c75cd81bef1bd82a574

Examples CUDA
benchmarks fluctuation threshold 5.0%
kernel improvement fem128(-5.2555%),
kernel regression None
end2end improvement None
end2end regression None
MicroBenchmarks CUDA
benchmarks fluctuation threshold 5.0%
kernel improvement saxpy_sparse_i64_1MB(-26.4714%), stencil_2d_sparse_i32_1MB(-25.976%), stencil_2d_sparse_i32_4MB(-25.2133%), stencil_2d_field_f32_16KB(-16.9178%), stencil_2d_field_f32_64KB(-16.0393%), stencil_2d_field_f32_4KB(-15.7962%), stencil_2d_field_f32_4MB(-14.2939%), stencil_2d_sparse_i64_16MB(-13.8988%), stencil_2d_ndarray_f64_16MB(-13.6408%), stencil_2d_field_f32_256KB(-13.5404%), stencil_2d_sparse_f64_4MB(-13.5367%), stencil_2d_field_f32_16MB(-13.3652%), stencil_2d_sparse_f64_1MB(-12.4817%), stencil_2d_sparse_f64_64KB(-11.956%), stencil_2d_sparse_f64_256KB(-11.8964%), stencil_2d_field_i64_4KB(-11.416%), stencil_2d_field_f32_1MB(-11.3036%), stencil_2d_ndarray_i64_4MB(-10.433%), stencil_2d_ndarray_f64_4MB(-9.4318%), stencil_2d_ndarray_i64_16MB(-8.6888%), stencil_2d_ndarray_f64_64MB(-8.0516%), stencil_2d_sparse_f32_64MB(-6.4991%), stencil_2d_ndarray_f64_256KB(-6.41%), stencil_2d_ndarray_f64_1MB(-5.7553%), stencil_2d_ndarray_f64_64KB(-5.7009%), stencil_2d_ndarray_f32_64KB(-5.6578%), reduction_dynamic_i64_4KB(-5.099%),
kernel regression stencil_2d_field_i64_256KB(+6.0016%), reduction_dynamic_i64_64KB(+6.2322%), stencil_2d_sparse_f32_1MB(+6.7978%), stencil_2d_sparse_f32_16MB(+6.9065%), stencil_2d_sparse_f32_64KB(+7.013%), stencil_2d_sparse_f32_256KB(+7.9095%), stencil_2d_ndarray_i32_256MB(+8.3201%), stencil_2d_field_f64_1MB(+8.3817%), stencil_2d_field_f64_4MB(+9.3205%), reduction_dynamic_i64_16MB(+9.9471%), stencil_2d_field_i64_4MB(+10.3677%), stencil_2d_ndarray_i32_64MB(+10.6358%), stencil_2d_ndarray_i32_256KB(+10.6831%), stencil_2d_sparse_f32_4MB(+11.7705%), stencil_2d_ndarray_i32_64KB(+12.8105%), stencil_2d_ndarray_i32_1MB(+13.1494%), reduction_dynamic_i64_1MB(+13.3529%), stencil_2d_ndarray_i32_16KB(+13.7897%), reduction_dynamic_i64_256KB(+13.7901%), stencil_2d_ndarray_i32_4KB(+14.0156%), stencil_2d_sparse_i64_64KB(+14.2119%), stencil_2d_sparse_i64_256KB(+14.5987%), stencil_2d_ndarray_i32_4MB(+14.6099%), reduction_dynamic_i64_16KB(+14.9691%), stencil_2d_sparse_i64_4MB(+15.5412%), stencil_2d_field_f64_16MB(+15.9917%), stencil_2d_ndarray_i32_16MB(+16.8132%), stencil_2d_sparse_i64_1MB(+17.4868%), reduction_dynamic_i64_4MB(+17.9089%), reduction_sparse_i32_16KB(+20.977%), stencil_2d_sparse_i32_64MB(+30.127%),

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant