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.

Sign up for GitHub

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

Jump to bottom

Fix typo in parallelise function name #178

Merged
merged 1 commit into from
Nov 21, 2023
Merged

Fix typo in parallelise function name #178

merged 1 commit into from
Nov 21, 2023

Conversation

thomasfaingnaert
Copy link
Member

Fixes #177

@thomasfaingnaert thomasfaingnaert mentioned this pull request Nov 20, 2023
Closed
@maleadt
Copy link
Member

maleadt commented Nov 20, 2023

Benchmark results for commit d29a133 (comparing to 6a8e8cb):

test master PR Δmin
WMMA GEMM Float16*Float16+Float16=Float16 (2048×2048) · (2048×2048) (TN) Block (128, 128, 64) Warps (4, 2) OP (16, 16, 16) 226.0 μs ± 3.66% (207.0 … 235.0 μs) 230.0 μs ± 0.736% (225.0 … 235.0 μs) +8.5% ❌
WMMA GEMM+bias Float16*Float16+Float32=Float32 (128×128) · (128×128) (TN) OP (16, 16, 16) 10.1 μs ± 1.36% (9.78 … 10.5 μs) 8.92 μs ± 1.8% (8.58 … 9.3 μs) -12.2% ✅
WMMA GEMM+bias Float16*Float16+Float32=Float32 (128×128) · (128×128) (NT) OP (16, 16, 16) 10.2 μs ± 1.35% (9.78 … 10.7 μs) 9.08 μs ± 1.88% (8.82 … 9.54 μs) -9.8% ✅
Tropical GEMM Float32*Float32=Float32 (128×256) · (256×128) (NT) OP (8, 16, 2), base shape (4, 8, 1) 125.0 μs ± 4.55% (112.0 … 128.0 μs) 127.0 μs ± 0.146% (127.0 … 128.0 μs) +13.5% ❌
FPU GEMM Float32*Float32=Float32 (128×128) · (128×128) (NN) OP (4, 8, 1), base shape (4, 8, 1) 13.5 μs ± 1.07% (13.1 … 13.8 μs) 15.3 μs ± 1.09% (15.0 … 15.5 μs) +14.5% ❌
Tropical GEMM Float32*Float32=Float32 (256×256) · (256×256) (NN) OP (8, 16, 2), base shape (4, 8, 1) 110.0 μs ± 0.152% (110.0 … 111.0 μs) 125.0 μs ± 0.117% (125.0 … 126.0 μs) +13.7% ❌
FPU GEMM Float32*Float32=Float64 (128×128) · (128×128) (NN) OP (8, 16, 2), base shape (4, 8, 1) 177.0 μs ± 4.17% (157.0 … 181.0 μs) 180.0 μs ± 0.244% (178.0 … 181.0 μs) +13.5% ❌

Comparison with baseline

test GemmKernels Baseline %
FPU GEMM Float32*Float32=Float32 (256×256) · (256×256) (TT) OP (8, 16, 2), base shape (4, 8, 1) 15.0 μs ± 1.13% (14.8 … 15.3 μs) 7.18 μs ± 1.62% (6.91 … 7.63 μs) 46.8
FPU GEMM Float32*Float32=Float32 (128×128) · (128×128) (NT) OP (4, 32, 1), base shape (1, 32, 1) 16.1 μs ± 0.996% (15.7 … 16.5 μs) 4.18 μs ± 3.81% (4.05 … 4.53 μs) 25.8
WMMA GEMM Float16*Float16+Float16=Float16 (2048×2048) · (2048×2048) (TN) Block (128, 128, 64) Warps (4, 2) OP (16, 16, 16) 230.0 μs ± 0.736% (225.0 … 235.0 μs) 190.0 μs ± 2.64% (180.0 … 195.0 μs) 80.1
WMMA GEMM Float16*Float16+Float32=Float32 (128×128) · (128×128) (TN) Block (128, 128, 64) Warps (4, 2) OP (16, 16, 16) 7.95 μs ± 1.78% (7.63 … 8.34 μs) 2.35 μs ± 6.49% (2.15 … 2.62 μs) 28.1
WMMA GEMM Float16*Float16+Float16=Float16 (2048×2048) · (2048×2048) (NT) Block (128, 128, 64) Warps (4, 2) OP (16, 16, 16) 224.0 μs ± 1.71% (204.0 … 229.0 μs) 185.0 μs ± 2.69% (172.0 … 193.0 μs) 84.5
FPU GEMM Float32*Float32=Float32 (128×128) · (128×128) (TT) OP (16, 16, 1), base shape (8, 4, 1) 14.5 μs ± 1.04% (14.3 … 14.8 μs) 4.04 μs ± 3.52% (3.81 … 4.29 μs) 26.7
FPU GEMM Float32*Float32=Float32 (2048×2048) · (2048×2048) (NN) OP (8, 16, 2), base shape (4, 8, 1) 1.84 ms ± 2.86% (1.8 … 2.02 ms) 996.0 μs ± 1.71% (936.0 … 1030.0 μs) 51.9
FPU GEMM Float32*Float32=Float32 (128×128) · (128×128) (TT) OP (32, 4, 1), base shape (32, 1, 1) 13.1 μs ± 0.942% (12.6 … 13.4 μs) 4.03 μs ± 3.49% (3.81 … 4.29 μs) 30.2
FPU GEMM Float32*Float32=Float32 (128×128) · (128×128) (NN) OP (32, 4, 1), base shape (16, 2, 1) 19.1 μs ± 0.896% (18.8 … 19.3 μs) 4.08 μs ± 2.72% (3.81 … 4.29 μs) 20.3
FPU GEMM Float16*Float16=Float32 (2048×2048) · (2048×2048) (TT) OP (8, 16, 2), base shape (4, 8, 1) 1.84 ms ± 2.63% (1.8 … 2.01 ms) 292.0 μs ± 0.333% (290.0 … 293.0 μs) 16.1
FPU GEMM Float64*Float64=Float64 (2048×2048) · (2048×2048) (TN) OP (8, 16, 2), base shape (4, 8, 1) 46.1 ms ± 0.801% (46.0 … 52.4 ms) 41.0 ms ± 0.00254% (41.0 … 41.0 ms) 89.2
FPU GEMM Float32*Float32=Float32 (128×128) · (128×128) (TN) OP (32, 4, 1), base shape (16, 2, 1) 19.1 μs ± 0.897% (18.8 … 19.3 μs) 4.29 μs ± 3.58% (4.05 … 4.53 μs) 21.5
FPU GEMM Float64*Float64=Float64 (256×256) · (256×256) (TN) OP (8, 16, 2), base shape (4, 8, 1) 269.0 μs ± 6.23% (255.0 … 293.0 μs) 109.0 μs ± 0.162% (109.0 … 109.0 μs) 42.7
WMMA GEMM Float16*Float16+Float16=Float16 (256×256) · (256×256) (NT) Block (128, 128, 64) Warps (4, 2) OP (16, 16, 16) 7.42 μs ± 1.97% (7.15 … 7.87 μs) 3.82 μs ± 2.71% (3.58 … 4.29 μs) 50.0
FPU GEMM Float32*Float32=Float32 (128×128) · (128×128) (NN) OP (8, 16, 2), base shape (4, 8, 1) 15.3 μs ± 1.13% (14.8 … 15.7 μs) 4.08 μs ± 2.78% (3.81 … 4.53 μs) 25.8
WMMA GEMM Float16*Float16+Float16=Float16 (512×256) · (256×512) (TT) Block (128, 128, 64) Warps (4, 2) OP (16, 16, 16) 8.71 μs ± 1.9% (8.34 … 9.06 μs) 5.14 μs ± 2.9% (4.77 … 5.48 μs) 57.1
FPU GEMM Float32*Float32=Float32 (128×128) · (128×128) (TN) OP (4, 8, 1), base shape (4, 8, 1) 15.2 μs ± 1.09% (15.0 … 15.5 μs) 4.3 μs ± 3.42% (4.05 … 4.53 μs) 27.0
FPU GEMM Float32*Float32=Float32 (2048×2048) · (2048×2048) (TT) OP (8, 16, 2), base shape (4, 8, 1) 1.88 ms ± 3.25% (1.84 … 2.05 ms) 1.1 ms ± 1.78% (1.02 … 1.15 ms) 55.5
FPU GEMM Float64*Float64=Float64 (256×256) · (256×256) (NN) OP (8, 16, 2), base shape (4, 8, 1) 272.0 μs ± 6.36% (256.0 … 292.0 μs) 109.0 μs ± 0.155% (109.0 … 109.0 μs) 42.6
FPU GEMM Float32*Float32=Float32 (128×128) · (128×128) (TT) OP (4, 8, 1), base shape (4, 8, 1) 15.2 μs ± 1.07% (15.0 … 15.5 μs) 4.03 μs ± 3.57% (3.81 … 4.29 μs) 25.4
WMMA GEMM Float16*Float16+Float32=Float32 (2048×2048) · (2048×2048) (NN) Block (128, 128, 64) Warps (4, 2) OP (16, 16, 16) 359.0 μs ± 4.76% (344.0 … 389.0 μs) 292.0 μs ± 0.314% (290.0 … 294.0 μs) 84.3
FPU GEMM Float32*Float32=Float32 (128×128) · (128×128) (NN) OP (16, 16, 1), base shape (4, 8, 1) 15.3 μs ± 0.903% (15.0 … 15.7 μs) 4.08 μs ± 2.57% (3.81 … 4.29 μs) 25.4
FPU GEMM Float32*Float32=Float32 (2048×2048) · (2048×2048) (TN) OP (8, 16, 2), base shape (4, 8, 1) 1.45 ms ± 1.6% (1.43 … 1.53 ms) 1.11 ms ± 1.81% (1.04 … 1.14 ms) 72.8
WMMA GEMM Float16*Float16+Float32=Float32 (128×128) · (128×128) (NN) Block (128, 128, 64) Warps (4, 2) OP (16, 16, 16) 8.1 μs ± 2.02% (7.87 … 8.34 μs) 2.24 μs ± 7.08% (1.91 … 2.62 μs) 24.2
FPU GEMM Float16*Float16=Float32 (2048×2048) · (2048×2048) (NN) OP (8, 16, 2), base shape (4, 8, 1) 1.84 ms ± 2.46% (1.81 … 2.02 ms) 292.0 μs ± 0.319% (290.0 … 294.0 μs) 16.0
FPU GEMM Float64*Float64=Float64 (128×128) · (128×128) (TT) OP (8, 16, 2), base shape (4, 8, 1) 130.0 μs ± 0.408% (129.0 … 131.0 μs) 28.7 μs ± 0.459% (28.4 … 29.1 μs) 22.0
WMMA GEMM Float16*Float16+Float32=Float32 (128×128) · (128×128) (TT) Block (128, 128, 64) Warps (4, 2) OP (16, 16, 16) 8.02 μs ± 2.03% (7.87 … 8.34 μs) 2.48 μs ± 6.6% (2.15 … 2.62 μs) 27.3
FPU GEMM Float64*Float64=Float64 (2048×2048) · (2048×2048) (TT) OP (8, 16, 2), base shape (4, 8, 1) 46.2 ms ± 0.757% (46.0 … 52.4 ms) 41.0 ms ± 0.00298% (41.0 … 41.0 ms) 89.1
FPU GEMM Float32*Float32=Float32 (128×128) · (128×128) (NT) OP (32, 4, 1), base shape (32, 1, 1) 12.6 μs ± 1.3% (12.4 … 12.9 μs) 4.19 μs ± 3.72% (4.05 … 4.53 μs) 32.7
WMMA GEMM Float16*Float16+Float32=Float32 (2048×2048) · (2048×2048) (NT) Block (128, 128, 64) Warps (4, 2) OP (16, 16, 16) 369.0 μs ± 5.15% (350.0 … 397.0 μs) 291.0 μs ± 0.164% (289.0 … 293.0 μs) 82.6
FPU GEMM Float16*Float16=Float32 (256×256) · (256×256) (TN) OP (8, 16, 2), base shape (4, 8, 1) 13.0 μs ± 1.12% (12.6 … 13.1 μs) 3.79 μs ± 3.16% (3.58 … 4.29 μs) 28.3
WMMA GEMM Float16*Float16+Float16=Float16 (256×256) · (256×256) (NN) Block (128, 128, 64) Warps (4, 2) OP (16, 16, 16) 8.43 μs ± 1.91% (8.11 … 8.82 μs) 3.71 μs ± 3.8% (3.58 … 4.05 μs) 44.1
WMMA GEMM Float16*Float16+Float32=Float32 (2048×2048) · (2048×2048) (TT) Block (128, 128, 64) Warps (4, 2) OP (16, 16, 16) 365.0 μs ± 5.04% (343.0 … 389.0 μs) 292.0 μs ± 0.187% (290.0 … 294.0 μs) 84.6
FPU GEMM Float16*Float16=Float32 (2048×2048) · (2048×2048) (NT) OP (8, 16, 2), base shape (4, 8, 1) 2.31 ms ± 3.51% (2.26 … 2.57 ms) 291.0 μs ± 0.325% (289.0 … 293.0 μs) 12.8
FPU GEMM Float16*Float16=Float32 (128×128) · (128×128) (TN) OP (8, 16, 2), base shape (4, 8, 1) 6.88 μs ± 1.25% (6.44 … 6.91 μs) 2.35 μs ± 6.39% (2.15 … 2.62 μs) 33.3
WMMA GEMM Float16*Float16+Float32=Float32 (256×128) · (128×256) (TN) Block (128, 128, 64) Warps (4, 2) OP (16, 16, 16) 9.05 μs ± 1.86% (8.82 … 9.3 μs) 2.88 μs ± 5.08% (2.62 … 3.1 μs) 29.7
WMMA GEMM Float16*Float16+Float32=Float32 (128×256) · (256×128) (TN) Block (128, 128, 64) Warps (4, 2) OP (16, 16, 16) 12.3 μs ± 1.15% (12.2 … 12.6 μs) 2.9 μs ± 5.15% (2.62 … 3.1 μs) 21.6
FPU GEMM Float32*Float32=Float32 (128×128) · (128×128) (NT) OP (4, 8, 2), base shape (4, 8, 1) 16.0 μs ± 1.05% (15.7 … 16.5 μs) 4.18 μs ± 3.64% (4.05 … 4.53 μs) 25.8
FPU GEMM Float32*Float32=Float32 (128×128) · (128×128) (NT) OP (8, 16, 2), base shape (4, 8, 1) 16.4 μs ± 0.963% (16.0 … 16.7 μs) 4.19 μs ± 3.82% (4.05 … 4.53 μs) 25.4
FPU GEMM Float32*Float32=Float32 (128×128) · (128×128) (NN) OP (32, 4, 1), base shape (32, 1, 1) 31.6 μs ± 0.473% (31.5 … 31.9 μs) 4.08 μs ± 2.83% (3.81 … 4.29 μs) 12.1
FPU GEMM Float32*Float32=Float32 (128×128) · (128×128) (TT) OP (8, 16, 2), base shape (4, 8, 1) 13.6 μs ± 0.914% (13.1 … 13.8 μs) 4.04 μs ± 3.51% (3.81 … 4.53 μs) 29.1
FPU GEMM Float32*Float32=Float32 (128×128) · (128×128) (NN) OP (4, 32, 1), base shape (1, 32, 1) 15.7 μs ± 1.06% (15.5 … 16.0 μs) 4.09 μs ± 2.6% (3.81 … 4.29 μs) 24.6
FPU GEMM Float16*Float16=Float32 (256×256) · (256×256) (NT) OP (8, 16, 2), base shape (4, 8, 1) 18.1 μs ± 0.879% (17.9 … 18.4 μs) 3.89 μs ± 3.67% (3.58 … 4.29 μs) 20.0
WMMA GEMM Float16*Float16+Float16=Float16 (2048×2048) · (2048×2048) (TT) Block (128, 128, 64) Warps (4, 2) OP (16, 16, 16) 224.0 μs ± 2.38% (205.0 … 229.0 μs) 188.0 μs ± 2.07% (178.0 … 197.0 μs) 86.6
WMMA GEMM Float16*Float16+Float16=Float16 (2048×2048) · (2048×2048) (NN) Block (128, 128, 64) Warps (4, 2) OP (16, 16, 16) 221.0 μs ± 3.83% (208.0 … 237.0 μs) 185.0 μs ± 2.82% (175.0 … 191.0 μs) 83.9
FPU GEMM Float32*Float32=Float32 (128×128) · (128×128) (TN) OP (4, 32, 1), base shape (2, 16, 1) 21.9 μs ± 0.761% (21.7 … 22.2 μs) 4.3 μs ± 3.37% (4.05 … 4.53 μs) 18.7
WMMA GEMM Float16*Float16+Float32=Float32 (128×256) · (256×128) (NN) Block (128, 128, 64) Warps (4, 2) OP (16, 16, 16) 12.1 μs ± 1.39% (11.9 … 12.4 μs) 3.08 μs ± 3.12% (2.62 … 3.34 μs) 22.0
FPU GEMM Float64*Float64=Float64 (128×128) · (128×128) (NT) OP (8, 16, 2), base shape (4, 8, 1) 140.0 μs ± 6.43% (129.0 … 149.0 μs) 29.5 μs ± 0.492% (29.1 … 29.8 μs) 22.5
FPU GEMM Float32*Float32=Float32 (128×128) · (128×128) (NN) OP (16, 16, 1), base shape (8, 4, 1) 14.5 μs ± 1.1% (14.3 … 14.8 μs) 4.08 μs ± 2.89% (3.81 … 4.29 μs) 26.7
FPU GEMM Float32*Float32=Float32 (128×128) · (128×128) (TT) OP (4, 8, 2), base shape (4, 8, 1) 12.4 μs ± 1.41% (12.2 … 12.6 μs) 4.03 μs ± 3.59% (3.81 … 4.29 μs) 31.4
FPU GEMM Float32*Float32=Float32 (128×128) · (128×128) (TT) OP (16, 16, 1), base shape (4, 8, 1) 15.4 μs ± 1.07% (15.3 … 15.7 μs) 4.03 μs ± 3.48% (3.81 … 4.29 μs) 25.0
WMMA GEMM Float16*Float16+Float16=Float16 (256×512) · (512×256) (NN) Block (128, 128, 64) Warps (4, 2) OP (16, 16, 16) 12.0 μs ± 1.34% (11.7 … 12.4 μs) 5.21 μs ± 3.26% (5.01 … 5.72 μs) 42.9
WMMA GEMM Float16*Float16+Float16=Float16 (512×512) · (512×512) (NT) Block (128, 128, 64) Warps (4, 2) OP (16, 16, 16) 12.4 μs ± 1.48% (11.9 … 12.6 μs) 7.25 μs ± 2.0% (6.91 … 7.63 μs) 58.0
WMMA GEMM Float16*Float16+Float16=Float16 (512×256) · (256×512) (NN) Block (128, 128, 64) Warps (4, 2) OP (16, 16, 16) 8.68 μs ± 2.17% (8.34 … 9.06 μs) 4.96 μs ± 3.45% (4.77 … 5.25 μs) 57.1
FPU GEMM Float64*Float64=Float64 (128×128) · (128×128) (NN) OP (8, 16, 2), base shape (4, 8, 1) 148.0 μs ± 0.162% (147.0 … 149.0 μs) 32.6 μs ± 0.519% (32.4 … 32.9 μs) 22.0
WMMA GEMM Float16*Float16+Float32=Float32 (256×256) · (256×256) (TN) Block (128, 128, 64) Warps (4, 2) OP (16, 16, 16) 12.4 μs ± 1.37% (12.2 … 12.6 μs) 3.8 μs ± 3.34% (3.58 … 4.29 μs) 29.4
WMMA GEMM Float16*Float16+Float32=Float32 (256×256) · (256×256) (TT) Block (128, 128, 64) Warps (4, 2) OP (16, 16, 16) 12.2 μs ± 1.32% (11.9 … 12.4 μs) 3.93 μs ± 3.91% (3.58 … 4.29 μs) 30.0
FPU GEMM Float32*Float32=Float32 (128×128) · (128×128) (TT) OP (4, 32, 1), base shape (2, 16, 1) 21.9 μs ± 0.747% (21.7 … 22.2 μs) 4.04 μs ± 3.54% (3.81 … 4.29 μs) 17.6
FPU GEMM Float32*Float32=Float32 (128×128) · (128×128) (TT) OP (32, 4, 1), base shape (16, 2, 1) 19.1 μs ± 0.872% (18.8 … 19.3 μs) 4.03 μs ± 3.43% (3.81 … 4.29 μs) 20.3
FPU GEMM Float32*Float32=Float32 (128×128) · (128×128) (TN) OP (4, 8, 2), base shape (4, 8, 1) 11.8 μs ± 1.42% (11.7 … 12.2 μs) 4.28 μs ± 3.58% (4.05 … 4.77 μs) 34.7
WMMA GEMM Float16*Float16+Float16=Float16 (256×512) · (512×256) (TT) Block (128, 128, 64) Warps (4, 2) OP (16, 16, 16) 12.2 μs ± 1.43% (11.9 … 12.4 μs) 5.17 μs ± 2.97% (4.77 … 5.48 μs) 40.0
WMMA GEMM Float16*Float16+Float16=Float16 (512×256) · (256×512) (TN) Block (128, 128, 64) Warps (4, 2) OP (16, 16, 16) 8.72 μs ± 2.01% (8.34 … 9.06 μs) 5.14 μs ± 2.84% (4.77 … 5.48 μs) 57.1
FPU GEMM Float32*Float32=Float32 (128×128) · (128×128) (TN) OP (4, 16, 1), base shape (4, 8, 1) 15.2 μs ± 1.07% (15.0 … 15.5 μs) 4.3 μs ± 3.43% (4.05 … 4.53 μs) 27.0
WMMA GEMM Float16*Float16+Float32=Float32 (256×256) · (256×256) (NN) Block (128, 128, 64) Warps (4, 2) OP (16, 16, 16) 12.2 μs ± 1.27% (11.9 … 12.4 μs) 3.94 μs ± 3.71% (3.81 … 4.29 μs) 32.0
FPU GEMM Float64*Float64=Float64 (2048×2048) · (2048×2048) (NT) OP (8, 16, 2), base shape (4, 8, 1) 46.4 ms ± 0.882% (46.4 … 52.8 ms) 41.0 ms ± 0.00232% (41.0 … 41.0 ms) 88.5
FPU GEMM Float32*Float32=Float32 (128×128) · (128×128) (NN) OP (8, 8, 1), base shape (4, 8, 1) 15.4 μs ± 0.853% (15.0 … 15.7 μs) 4.08 μs ± 2.71% (3.81 … 4.29 μs) 25.4
WMMA GEMM Float16*Float16+Float16=Float16 (512×256) · (256×512) (NT) Block (128, 128, 64) Warps (4, 2) OP (16, 16, 16) 8.78 μs ± 2.11% (8.34 … 9.06 μs) 4.92 μs ± 3.3% (4.77 … 5.25 μs) 57.1
WMMA GEMM Float16*Float16+Float16=Float16 (512×512) · (512×512) (TN) Block (128, 128, 64) Warps (4, 2) OP (16, 16, 16) 12.4 μs ± 1.44% (12.2 … 12.9 μs) 7.41 μs ± 1.61% (7.15 … 7.87 μs) 58.8
WMMA GEMM Float16*Float16+Float32=Float32 (2048×2048) · (2048×2048) (TN) Block (128, 128, 64) Warps (4, 2) OP (16, 16, 16) 373.0 μs ± 5.01% (350.0 … 397.0 μs) 293.0 μs ± 0.293% (291.0 … 295.0 μs) 83.2
WMMA GEMM Float16*Float16+Float16=Float16 (512×512) · (512×512) (NN) Block (128, 128, 64) Warps (4, 2) OP (16, 16, 16) 12.3 μs ± 1.38% (11.9 … 12.6 μs) 7.31 μs ± 2.03% (7.15 … 7.63 μs) 60.0
FPU GEMM Float32*Float32=Float32 (128×128) · (128×128) (NN) OP (4, 8, 1), base shape (4, 8, 1) 15.3 μs ± 1.09% (15.0 … 15.5 μs) 4.08 μs ± 2.68% (3.81 … 4.53 μs) 25.4
FPU GEMM Float16*Float16=Float32 (128×128) · (128×128) (NT) OP (8, 16, 2), base shape (4, 8, 1) 9.19 μs ± 1.55% (8.82 … 9.54 μs) 2.04 μs ± 6.3% (1.67 … 2.38 μs) 18.9
WMMA GEMM Float16*Float16+Float32=Float32 (256×128) · (128×256) (NN) Block (128, 128, 64) Warps (4, 2) OP (16, 16, 16) 9.31 μs ± 1.75% (9.06 … 9.54 μs) 2.93 μs ± 5.21% (2.62 … 3.1 μs) 28.9
WMMA GEMM Float16*Float16+Float32=Float32 (128×128) · (128×128) (NT) Block (128, 128, 64) Warps (4, 2) OP (16, 16, 16) 8.03 μs ± 2.09% (7.87 … 8.34 μs) 2.31 μs ± 5.21% (2.15 … 2.62 μs) 27.3
FPU GEMM Float32*Float32=Float32 (128×128) · (128×128) (NT) OP (16, 16, 1), base shape (8, 4, 1) 14.6 μs ± 0.891% (14.3 … 14.8 μs) 4.18 μs ± 3.6% (4.05 … 4.53 μs) 28.3
FPU GEMM Float32*Float32=Float32 (128×128) · (128×128) (TT) OP (4, 32, 1), base shape (1, 32, 1) 33.0 μs ± 0.412% (32.7 … 33.4 μs) 4.02 μs ± 3.46% (3.81 … 4.29 μs) 11.7
FPU GEMM Float32*Float32=Float32 (128×128) · (128×128) (TT) OP (4, 16, 1), base shape (4, 8, 1) 15.2 μs ± 1.11% (15.0 … 15.5 μs) 4.04 μs ± 3.58% (3.81 … 4.29 μs) 25.4
FPU GEMM Float32*Float32=Float32 (128×128) · (128×128) (NT) OP (32, 4, 1), base shape (16, 2, 1) 19.2 μs ± 0.802% (18.8 … 19.6 μs) 4.18 μs ± 3.67% (4.05 … 4.53 μs) 21.5
FPU GEMM Float32*Float32=Float32 (2048×2048) · (2048×2048) (NT) OP (8, 16, 2), base shape (4, 8, 1) 2.35 ms ± 2.62% (2.32 … 2.63 ms) 867.0 μs ± 0.867% (818.0 … 877.0 μs) 35.3
FPU GEMM Float64*Float64=Float64 (256×256) · (256×256) (TT) OP (8, 16, 2), base shape (4, 8, 1) 277.0 μs ± 6.35% (256.0 … 295.0 μs) 109.0 μs ± 0.154% (109.0 … 110.0 μs) 42.6
FPU GEMM Float32*Float32=Float32 (128×128) · (128×128) (TT) OP (8, 8, 1), base shape (4, 8, 1) 15.3 μs ± 1.0% (15.0 … 15.5 μs) 4.03 μs ± 3.43% (3.81 … 4.29 μs) 25.4
WMMA GEMM Float16*Float16+Float16=Float16 (256×256) · (256×256) (TT) Block (128, 128, 64) Warps (4, 2) OP (16, 16, 16) 7.47 μs ± 1.75% (7.15 … 7.87 μs) 3.64 μs ± 3.24% (3.58 … 4.05 μs) 50.0
WMMA GEMM Float16*Float16+Float32=Float32 (128×256) · (256×128) (TT) Block (128, 128, 64) Warps (4, 2) OP (16, 16, 16) 12.2 μs ± 1.4% (11.9 … 12.4 μs) 3.0 μs ± 4.69% (2.62 … 3.34 μs) 22.0
FPU GEMM Float16*Float16=Float32 (128×128) · (128×128) (TT) OP (8, 16, 2), base shape (4, 8, 1) 8.17 μs ± 1.61% (7.87 … 8.34 μs) 2.17 μs ± 8.11% (1.91 … 2.38 μs) 24.2
WMMA GEMM Float16*Float16+Float16=Float16 (512×512) · (512×512) (TT) Block (128, 128, 64) Warps (4, 2) OP (16, 16, 16) 12.4 μs ± 1.44% (11.9 … 12.9 μs) 7.27 μs ± 1.97% (7.15 … 7.63 μs) 60.0
WMMA GEMM Float16*Float16+Float32=Float32 (256×256) · (256×256) (NT) Block (128, 128, 64) Warps (4, 2) OP (16, 16, 16) 12.4 μs ± 1.37% (12.2 … 12.6 μs) 3.91 μs ± 3.63% (3.58 … 4.29 μs) 29.4
FPU GEMM Float32*Float32=Float32 (256×256) · (256×256) (NN) OP (8, 16, 2), base shape (4, 8, 1) 15.3 μs ± 1.1% (15.0 … 15.5 μs) 6.8 μs ± 1.97% (6.44 … 6.91 μs) 42.9
FPU GEMM Float32*Float32=Float32 (128×128) · (128×128) (NT) OP (8, 8, 1), base shape (4, 8, 1) 15.4 μs ± 1.07% (15.3 … 15.7 μs) 4.19 μs ± 3.84% (4.05 … 4.53 μs) 26.6
FPU GEMM Float64*Float64=Float64 (256×256) · (256×256) (NT) OP (8, 16, 2), base shape (4, 8, 1) 260.0 μs ± 1.43% (256.0 … 266.0 μs) 109.0 μs ± 0.156% (109.0 … 109.0 μs) 42.6
FPU GEMM Float32*Float32=Float32 (128×128) · (128×128) (NT) OP (4, 16, 1), base shape (4, 8, 1) 15.3 μs ± 0.954% (15.0 … 15.7 μs) 4.19 μs ± 3.64% (4.05 … 4.53 μs) 27.0
WMMA GEMM Float16*Float16+Float32=Float32 (256×128) · (128×256) (TT) Block (128, 128, 64) Warps (4, 2) OP (16, 16, 16) 9.15 μs ± 1.76% (8.82 … 9.3 μs) 2.89 μs ± 4.9% (2.62 … 3.1 μs) 29.7
FPU GEMM Float16*Float16=Float32 (128×128) · (128×128) (NN) OP (8, 16, 2), base shape (4, 8, 1) 9.22 μs ± 1.37% (8.82 … 9.54 μs) 2.21 μs ± 7.74% (1.91 … 2.38 μs) 21.6
FPU GEMM Float32*Float32=Float32 (128×128) · (128×128) (TN) OP (8, 8, 1), base shape (4, 8, 1) 15.2 μs ± 1.13% (15.0 … 15.5 μs) 4.28 μs ± 3.57% (4.05 … 4.77 μs) 27.0
FPU GEMM Float16*Float16=Float32 (256×256) · (256×256) (TT) OP (8, 16, 2), base shape (4, 8, 1) 15.8 μs ± 1.11% (15.5 … 16.0 μs) 3.94 μs ± 3.94% (3.58 … 4.29 μs) 23.1
FPU GEMM Float32*Float32=Float32 (128×128) · (128×128) (NT) OP (16, 16, 1), base shape (4, 8, 1) 15.3 μs ± 0.945% (15.0 … 15.7 μs) 4.18 μs ± 3.69% (4.05 … 4.53 μs) 27.0
WMMA GEMM Float16*Float16+Float32=Float32 (128×256) · (256×128) (NT) Block (128, 128, 64) Warps (4, 2) OP (16, 16, 16) 12.4 μs ± 1.36% (12.2 … 12.6 μs) 2.77 μs ± 4.91% (2.38 … 3.1 μs) 19.6
FPU GEMM Float32*Float32=Float32 (128×128) · (128×128) (NN) OP (4, 8, 2), base shape (4, 8, 1) 15.0 μs ± 1.12% (14.8 … 15.3 μs) 4.09 μs ± 2.63% (3.81 … 4.29 μs) 25.8
FPU GEMM Float32*Float32=Float32 (128×128) · (128×128) (TN) OP (32, 4, 1), base shape (32, 1, 1) 31.6 μs ± 0.46% (31.5 … 31.9 μs) 4.3 μs ± 3.33% (4.05 … 4.53 μs) 12.9
WMMA GEMM Float16*Float16+Float32=Float32 (256×128) · (128×256) (NT) Block (128, 128, 64) Warps (4, 2) OP (16, 16, 16) 9.16 μs ± 1.66% (8.82 … 9.3 μs) 2.92 μs ± 5.35% (2.62 … 3.1 μs) 29.7
FPU GEMM Float32*Float32=Float32 (128×128) · (128×128) (NT) OP (4, 32, 1), base shape (2, 16, 1) 21.9 μs ± 0.744% (21.7 … 22.2 μs) 4.19 μs ± 3.78% (4.05 … 4.53 μs) 18.7
FPU GEMM Float16*Float16=Float32 (256×256) · (256×256) (NN) OP (8, 16, 2), base shape (4, 8, 1) 15.7 μs ± 1.13% (15.5 … 16.0 μs) 3.94 μs ± 3.84% (3.81 … 4.29 μs) 24.6
FPU GEMM Float16*Float16=Float32 (2048×2048) · (2048×2048) (TN) OP (8, 16, 2), base shape (4, 8, 1) 1.54 ms ± 2.65% (1.52 … 1.67 ms) 292.0 μs ± 0.265% (290.0 … 293.0 μs) 19.1
FPU GEMM Float64*Float64=Float64 (2048×2048) · (2048×2048) (NN) OP (8, 16, 2), base shape (4, 8, 1) 46.3 ms ± 0.891% (46.3 … 52.7 ms) 40.8 ms ± 0.00535% (40.8 … 40.8 ms) 88.2
WMMA GEMM Float16*Float16+Float16=Float16 (256×512) · (512×256) (TN) Block (128, 128, 64) Warps (4, 2) OP (16, 16, 16) 12.1 μs ± 1.38% (11.9 … 12.4 μs) 5.11 μs ± 3.28% (4.77 … 5.72 μs) 40.0
FPU GEMM Float32*Float32=Float32 (128×128) · (128×128) (NN) OP (4, 32, 1), base shape (2, 16, 1) 22.0 μs ± 0.616% (21.7 … 22.2 μs) 4.07 μs ± 2.75% (3.81 … 4.29 μs) 17.6
WMMA GEMM Float16*Float16+Float16=Float16 (256×512) · (512×256) (NT) Block (128, 128, 64) Warps (4, 2) OP (16, 16, 16) 12.0 μs ± 1.45% (11.7 … 12.4 μs) 4.49 μs ± 3.65% (4.29 … 4.77 μs) 36.7
FPU GEMM Float32*Float32=Float32 (256×256) · (256×256) (NT) OP (8, 16, 2), base shape (4, 8, 1) 18.2 μs ± 0.769% (17.9 … 18.6 μs) 6.74 μs ± 1.91% (6.44 … 6.91 μs) 36.0
FPU GEMM Float32*Float32=Float32 (128×128) · (128×128) (NN) OP (4, 16, 1), base shape (4, 8, 1) 15.2 μs ± 1.13% (15.0 … 15.5 μs) 4.08 μs ± 2.68% (3.81 … 4.53 μs) 25.4
FPU GEMM Float32*Float32=Float32 (128×128) · (128×128) (NT) OP (4, 8, 1), base shape (4, 8, 1) 15.3 μs ± 1.07% (15.0 … 15.5 μs) 4.18 μs ± 3.62% (4.05 … 4.53 μs) 27.0
FPU GEMM Float32*Float32=Float32 (128×128) · (128×128) (TN) OP (4, 32, 1), base shape (1, 32, 1) 33.0 μs ± 0.454% (32.7 … 33.4 μs) 4.29 μs ± 3.52% (4.05 … 4.53 μs) 12.4
FPU GEMM Float32*Float32=Float32 (128×128) · (128×128) (TN) OP (8, 16, 2), base shape (4, 8, 1) 12.1 μs ± 1.22% (11.7 … 12.4 μs) 4.29 μs ± 3.45% (4.05 … 4.53 μs) 34.7
FPU GEMM Float64*Float64=Float64 (128×128) · (128×128) (TN) OP (8, 16, 2), base shape (4, 8, 1) 130.0 μs ± 0.277% (129.0 … 131.0 μs) 28.7 μs ± 0.474% (28.4 … 29.1 μs) 22.0
FPU GEMM Float32*Float32=Float32 (128×128) · (128×128) (TN) OP (16, 16, 1), base shape (4, 8, 1) 15.2 μs ± 1.11% (15.0 … 15.5 μs) 4.29 μs ± 3.5% (4.05 … 4.53 μs) 27.0
WMMA GEMM Float16*Float16+Float16=Float16 (256×256) · (256×256) (TN) Block (128, 128, 64) Warps (4, 2) OP (16, 16, 16) 7.42 μs ± 1.48% (7.15 … 7.87 μs) 3.6 μs ± 2.62% (3.1 … 4.05 μs) 43.3
FPU GEMM Float32*Float32=Float32 (128×128) · (128×128) (TN) OP (16, 16, 1), base shape (8, 4, 1) 14.5 μs ± 1.08% (14.3 … 14.8 μs) 4.3 μs ± 3.34% (4.05 … 4.53 μs) 28.3
FPU GEMM Float32*Float32=Float32 (256×256) · (256×256) (TN) OP (8, 16, 2), base shape (4, 8, 1) 12.6 μs ± 1.3% (12.4 … 12.9 μs) 7.15 μs ± 1.21% (6.91 … 7.39 μs) 55.8

@codecov Codecov
Copy link

codecov bot commented Nov 20, 2023

Codecov Report

Attention: 33 lines in your changes are missing coverage. Please review.

Comparison is base (6a8e8cb) 34.23% compared to head (d29a133) 34.23%.

Files Patch % Lines
src/kernel.jl 0.00% 29 Missing ⚠️
src/epilogue.jl 0.00% 4 Missing ⚠️
Additional details and impacted files 
@@           Coverage Diff           @@
##           master     #178   +/-   ##
=======================================
  Coverage   34.23%   34.23%           
=======================================
  Files          11       11           
  Lines         923      923           
=======================================
  Hits          316      316           
  Misses        607      607

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

@thomasfaingnaert thomasfaingnaert merged commit 8207701 into master Nov 21, 2023
1 check passed
@thomasfaingnaert thomasfaingnaert deleted the tf/fix-parallelise-name branch November 21, 2023 13:27
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Reviewers
No reviews
Assignees
No one assigned
Labels
None yet
Projects
None yet
Milestone
No milestone
Development

Successfully merging this pull request may close these issues.

A wrong function name parallellise
2 participants
@thomasfaingnaert @maleadt