Skip to content

Actions: mcx/XNNPACK

Actions

All workflows

Actions

Loading...
Loading

Showing runs from all workflows
362 workflow runs
362 workflow runs

Filter by Event

Filter by Status

Filter by Branch

Filter by Actor

Replace assert with xnn_log_warning. Build and update caches #76: Commit db9c9f5 pushed by pull bot
6h 0m 5s master
master
6h 0m 5s
Don't pack left-hand operand since this is currently only supported f… Build and update caches #75: Commit a4e966a pushed by pull bot
6h 0m 7s master
master
6h 0m 7s
QS8-PACKW AVX2 using vpmaddubsw Build and update caches #74: Commit 3390e6d pushed by pull bot
6h 0m 6s master
master
6h 0m 6s
Fix bazel-linux-aarch64-gcc13 workflow and resolve accompanying bui… Build and update caches #73: Commit 5f23827 pushed by pull bot
6h 0m 6s master
master
6h 0m 6s
QS8-QC4W packw scalar microkernels Build and update caches #72: Commit 98f1c8e pushed by pull bot
6h 0m 4s master
master
6h 0m 4s
Changes to batch_matrix_multiply_bench: Build and update caches #71: Commit 24e2005 pushed by pull bot
6h 0m 15s master
master
6h 0m 15s
Convolution bias does not need to be explicitly converted to f16. Build and update caches #70: Commit cc7c00a pushed by pull bot
1d 12h 17m 32s master
master
1d 12h 17m 32s
Handle f16 GEMM weights and biases when converting to f16. Build and update caches #69: Commit 0b82d1f pushed by pull bot
6h 0m 7s master
master
6h 0m 7s
Merge pull request #7479 from xujuntwt95329:x8_gio_pack Build and update caches #68: Commit ee5baf3 pushed by pull bot
6h 0m 11s master
master
6h 0m 11s
Add a builder for aarch64 under gcc-13 in addition to clang-18 (helps… Build and update caches #67: Commit 0270f91 pushed by pull bot
7h 35m 14s master
master
7h 35m 14s
Speculative fix for #7489 Build and update caches #66: Commit 61ac1d2 pushed by pull bot
16h 0m 11s master
master
16h 0m 11s
Enable X32-GIO-PACKW AVX microkernel Build and update caches #65: Commit e297edd pushed by pull bot
1d 0h 0m 7s master
master
1d 0h 0m 7s
Allow calling fully-connected and (depthwise|de)convolution opera… Build and update caches #64: Commit 52f3146 pushed by pull bot
16h 0m 9s master
master
16h 0m 9s
Some outputs of even split may be optimized away since they are unused. Build and update caches #63: Commit 8c7d7e7 pushed by pull bot
12h 30m 20s master
master
12h 30m 20s
F32-GEMM AVX512 generate up to 16x64 Build and update caches #62: Commit cbe5cdf pushed by pull bot
1d 0h 54m 10s master
master
1d 0h 54m 10s
Fix the template neon-mlal-lane.c.in Build and update caches #61: Commit 05b3026 pushed by pull bot
1d 0h 31m 4s master
master
1d 0h 31m 4s
Remove deprecated call from size test Build and update caches #60: Commit 8731139 pushed by pull bot
1d 23h 23m 40s master
master
1d 23h 23m 40s
Simplify the xnn_pack_f{16,32}_gemm_gio_w functions to use `std::co… Build and update caches #59: Commit 6515679 pushed by pull bot
12h 18m 15s master
master
12h 18m 15s
Add flag to allow avx512vbmi to be disabled Build and update caches #58: Commit d385289 pushed by pull bot
11h 3m 3s master
master
11h 3m 3s
Remove unused variable Build and update caches #57: Commit 6bb94cb pushed by pull bot
15h 47m 52s master
master
15h 47m 52s
Simplify kc remainder processing for qs8-x8c8 avxvnni Build and update caches #56: Commit 2e53ffc pushed by pull bot
1d 5h 22m 19s master
master
1d 5h 22m 19s
Remove operator size test Build and update caches #55: Commit abf7ff2 pushed by pull bot
3h 13m 10s master
master
3h 13m 10s
Mark slow benchmarks with xnnpack_slow_benchmark_tags. Build and update caches #54: Commit ea9018e pushed by pull bot
1d 6h 46m 2s master
master
1d 6h 46m 2s
Add fill_uniform_random_bits and use it to speed up tests/benchmarks. Build and update caches #53: Commit 0e576bd pushed by pull bot
1d 6h 59m 29s master
master
1d 6h 59m 29s
Added wasmrelaxedsimd test for qs8-rsum. Build and update caches #52: Commit b3a94e8 pushed by pull bot
15h 2m 51s master
master
15h 2m 51s
ProTip! You can narrow down the results and go further in time using created:<2024-10-15 or the other filters available.