Sign in
chromium
/
external
/
github.com
/
google
/
XNNPACK
/
HEAD
d7ddd02
Store static_pad buffer at the innermost level.
by Volodymyr Kysenko
· 15 hours ago
upstream/master
52ac311
Reorder operations in YNNPACK's grouped convolution definition.
by Volodymyr Kysenko
· 15 hours ago
4939810
Add scalar support and test coverage of horizontal_sum
by Dillon Sharlet
· 18 hours ago
548e701
Add scalar implementation of simd::vec
by Dillon Sharlet
· 19 hours ago
f5f93c4
Make all floating point sum kernels arithmetically consistent on x86.
by Dillon Sharlet
· 19 hours ago
3e0e760
Fix target processor inference has logic bug
by Frank Barchard
· 2 days ago
92d786a
Add linkopts and malloc settings to new binaries
by Dillon Sharlet
· 2 days ago
d899254
Add max pooling test case to reduce_window test
by Dillon Sharlet
· 2 days ago
1e15a3e
Set output shape too
by Dillon Sharlet
· 2 days ago
0d1b729
[gn] Add support for most of XNNPACK's tests
by Richard Townsend
· 2 days ago
d08f992
More narrowly disable sme2-qemu build
by Dillon Sharlet
· 2 days ago
d698580
Change simd tests to skip in Test::SetUp instead of a custom main
by Dillon Sharlet
· 2 days ago
313afbb
Added qs8_qc2w neondot kernel.
by Misha Gutman
· 2 days ago
fa19b54
Add a test only implementation of reduce_window.
by Alexander Shaposhnikov
· 2 days ago
023f2bd
Force root for pack_b in YNNPACK dot subgraph when num_k_dims > 1.
by Volodymyr Kysenko
· 2 days ago
4e3a0a6
Add subtract_fp32_bf16 kernel
by Marie White
· 2 days ago
102dac3
Temporarily disable sme2-qemu
by Dillon Sharlet
· 2 days ago
675932d
Fix fingerprinting test compilation for C++11.
by Quentin Khan
· 3 days ago
4cad8c8
Temporarily disable sme2-qemu
by Dillon Sharlet
· 3 days ago
4a5d698
Remove WORKSPACE
by Dillon Sharlet
· 3 days ago
cb7ba0c
Update bazel dependencies
by Dillon Sharlet
· 3 days ago
f837e5b
Speculative fix for:
by Dillon Sharlet
· 3 days ago
7cf53d6
Fix workflow trigger paths
by Dillon Sharlet
· 3 days ago
16c0f7d
Temporarily disable sme2-qemu
by Dillon Sharlet
· 3 days ago
b33278f
Add 2-bit QD8_F32_QC2W GEMM for AVX2 - add template and generate 1x8
by Frank Barchard
· 4 days ago
f95d1d3
Make SME2 docker image multi-arch
by Dillon Sharlet
· 4 days ago
46426be
Fix dot_bench argument parsing
by Marie White
· 4 days ago
fd47a05
Remove explicit bazel version
by Dillon Sharlet
· 4 days ago
2ceace7
Fix usage of `std::uniform_int_distribution` with unsupported types.
by Dillon Sharlet
· 4 days ago
a49f92c
Fix incorrect usages of anonymous namespaces
by Dillon Sharlet
· 4 days ago
1b32c8a
Remove slinky integration in XNNPACK
by Dillon Sharlet
· 4 days ago
199903d
Merge pull request #9393 from ken-unger:f32-rsum-rvv
by XNNPACK Team
· 4 days ago
3caec45
Remove 12x4x4 dot kernel from avx512
by Dillon Sharlet
· 4 days ago
b4deb87
Fix flaky failures due to differences in overflow when converting float to bf16
by Dillon Sharlet
· 4 days ago
0a76944
Tweak LUTs so we only need unsigned index kernels.
by Dillon Sharlet
· 4 days ago
4b8cfde
Add convert bf16 to fp32 kernels for AVX2 and AVX512
by Marie White
· 4 days ago
88634cd
Fix 2D lut use cases
by Dillon Sharlet
· 4 days ago
4db8f78
Implement convert f32 to bf16 for AVX2 and AVX512BF
by Marie White
· 5 days ago
e7da7bc
Clean up `get_binary_kernel`
by Dillon Sharlet
· 5 days ago
7e65961
Update KleidiAI dependency
by Dillon Sharlet
· 5 days ago
e9459e3
Update cpuinfo, gtest, KleidiAI dependencies
by Dillon Sharlet
· 5 days ago
8ff3752
Add `//ynnpack/subgraph/test:dot_bench`
by Dillon Sharlet
· 6 days ago
973925d
Simplify cache size comparison in dot kernel scheduling.
by Volodymyr Kysenko
· 8 days ago
e3132f2
increase static-reduce test tolerance to support RVV implementation
by Ken Unger
· 8 days ago
ce92ca4
[gn] Reduce boilerplate in BUILD.gn
by Richard Townsend
· 8 days ago
6555fdf
[gn] Manual DEPS upgrade
by Richard Townsend
· 8 days ago
91fb03b
Make scalar, SSE2 fp32 sum kernels consistent with AVX512
by Alexander Shaposhnikov
· 8 days ago
966ac38
[gn] Add basic Github Actions support
by Richard Townsend
· 8 days ago
3c69df8
Add convert ops for expensive unary integer ops
by Marie White
· 9 days ago
5498cdc
Add tile_k = 1 bf16 dot kernels
by Dillon Sharlet
· 9 days ago
107a4e6
Remove type punning in `get_scalar_value_as_float`
by Dillon Sharlet
· 9 days ago
f8ffb61
Remove unnecessary `xnn_get_value_size_by_id` helper.
by Dillon Sharlet
· 9 days ago
7fc1517
Use `is_static_scalar` where appropriate
by Dillon Sharlet
· 10 days ago
bb9067b
Replace integer unary subgraphs with luts.
by Marie White
· 10 days ago
d573645
Add tile_k = 1 kernels for int8
by Dillon Sharlet
· 10 days ago
0d8008d
Fix dot kernel cost estimation for k2, k3 > 1
by Dillon Sharlet
· 10 days ago
ba52e3b
Add dot_flag::consistent_arithmetic to avx512 int8 kernels
by Dillon Sharlet
· 11 days ago
ae79db0
Switch sme2 job to Docker.
by Alexander Shaposhnikov
· 12 days ago
ea7049c
Add Dockerfile.sme2. The new image will be used by the job testing sme2.
by Alexander Shaposhnikov
· 12 days ago
9f31ff2
Switch bazel builds to opt mode. Adopt -g1 (clang's -gline-tables-only).
by Alexander Shaposhnikov
· 12 days ago
5b1d09c
Update slinky in XNNPACK
by Dillon Sharlet
· 12 days ago
7487f32
Merge pull request #9416 from salmanmkc:upgrade-github-actions-node24-general
by XNNPACK Team
· 12 days ago
e44588c
Disable -Wmaybe-uninitialized via pragma.
by Alexander Shaposhnikov
· 12 days ago
c7cfb30
Switch android jobs to Docker.
by Alexander Shaposhnikov
· 12 days ago
a009c19
Add Dockerfile.android. The new image will be used the job cmake-android.
by Alexander Shaposhnikov
· 13 days ago
54be061
Adjust the range of inputs in consistent_arithmetic_test.
by Alexander Shaposhnikov
· 13 days ago
3188f2d
Switch riscv jobs to Docker.
by Alexander Shaposhnikov
· 13 days ago
e167310
Add Dockerfile.riscv The new image will be used by the both riscv build jobs.
by Alexander Shaposhnikov
· 13 days ago
bfdb346
Fix typo in build.yml.
by Alexander Shaposhnikov
· 2 weeks ago
4b075e7
This CL switches bazel-linux-x86_64-clang-18, bazel-linux-x86_64-gcc-9, bazel-linux-aarch64-gcc-13, bazel-linux-aarch64-clang-18 jobs to Docker.
by Alexander Shaposhnikov
· 2 weeks ago
cd6443e
Add vim to the list of packages in Dockerfile.standard
by Alexander Shaposhnikov
· 2 weeks ago
5d79a4b
[gn] Fix bad BUILD.gn path [2/2]
by Richard Townsend
· 2 weeks ago
a80ebb0
Switch cmake-linux-aarch64, cmake-linux-armhf to Docker.
by Alexander Shaposhnikov
· 2 weeks ago
38aa0b8
Rename mis-named helper function
by Dillon Sharlet
· 2 weeks ago
1343a1e
Fixed zero points naming for 2-bit.
by Misha Gutman
· 2 weeks ago
a7970af
Upgrade GitHub Actions to latest versions
by Salman Muin Kayser Chishti
· 2 weeks ago
762894d
[gn] Fix bad BUILD.gn path [1/2]
by Richard Townsend
· 2 weeks ago
beb3639
Make cmake-linux-x86_64 gh action run inside a docker container.
by Alexander Shaposhnikov
· 2 weeks ago
52c3e47
1. Add Dockerfile that will be used by XNNPACK github actions.
by Alexander Shaposhnikov
· 2 weeks ago
4b87e28
Fix use of deprecated (and now removed) bazel build rules
by Dillon Sharlet
· 2 weeks ago
76c91a5
Use std::nearbyint for rounding
by Dillon Sharlet
· 3 weeks ago
c98965d
[gn] build: add GN files for library and third_party dependencies
by Richard Townsend
· 3 weeks ago
4f52cf0
Set bounds of input buffers
by Dillon Sharlet
· 3 weeks ago
a0958aa
Expose define_lut() in elementwise.h
by Marie White
· 3 weeks ago
546e7f2
[gn] Update DEPS and add documentation for external libraries
by Richard Townsend
· 3 weeks ago
3399031
[gn] build: update build_srcs.bzl so it can be read by GN
by Richard Townsend
· 3 weeks ago
6565815
Clean up handling of global symbols and lets in slinky
by Dillon Sharlet
· 3 weeks ago
0c0b85d
Remove unnecessary dependency on launchpad.net for gcc13
by Dillon Sharlet
· 3 weeks ago
f6613ee
Remove unconditional printing of the random seed
by Dillon Sharlet
· 3 weeks ago
01ec44f
[RVV] add f32-rsum and f32-prelu rvv kernels
by Ken Unger
· 3 weeks ago
a1e5382
Merge pull request #9379 from ken-unger:f32-vrnd-rvv
by XNNPACK Team
· 4 weeks ago
aa9f48a
Don't rewrite transpose_a before stencil_copy if the transpose is of one of the new dimensions created by stencil_copy
by Dillon Sharlet
· 4 weeks ago
9a4ad45
Add Lookup Table (LUT) operation to YNNPACK.
by Marie White
· 4 weeks ago
36c2607
[RVV] f32-vrnd fix for issue #8087
by Ken Unger
· 4 weeks ago
2ef7272
Rewrite sum(x * x) to be sum_squared(x) for int32.
by Marie White
· 4 weeks ago
2ad963a
Add check for floating point types
by Marie White
· 4 weeks ago
32d0c39
Fix bug in rewrite_convert_to_quantize()
by Marie White
· 4 weeks ago
1598bc3
Allocate input buffer to test with XNN_EXTRA_BYTES
by Dillon Sharlet
· 4 weeks ago
7f72120
Merge pull request #9366 from ejparkqc:master
by XNNPACK Team
· 4 weeks ago
b62141c
Fix memory leaks if the model creation fails
by Dillon Sharlet
· 4 weeks ago
Next »