Skip to content

Commit

Permalink
Update xnnpack (pytorch#4340)
Browse files Browse the repository at this point in the history 
Summary: Pull Request resolved: pytorch#4340

Reviewed By: kirklandsign

Differential Revision: D60093405

Pulled By: digantdesai

fbshipit-source-id: 3e01dc50123fc05cf8a81e04ca976278bdfe6a0a
  • Loading branch information
@digantdesai @facebook-github-bot
digantdesai authored and facebook-github-bot committed Jul 23, 2024
1 parent b7fb9bd commit 908b5a5
Showing 1 changed file with 1 addition and 1 deletion.
2 changes: 1 addition & 1 deletion backends/xnnpack/third-party/XNNPACK
Submodule XNNPACK updated 24 files
+6 −0 CMakeLists.txt
+7 −0 cmake/microkernels.cmake
+7 −0 microkernels.bzl
+9 −0 scripts/generate-qd8-f16-qb4w-gemm.sh
+649 −0 src/amalgam/gen/scalar.c
+9 −2 src/configs/gemm-config.c
+0 −2 src/microparams-init.c
+2 −1 src/operators/fully-connected-nc.c
+10 −29 src/packing.c
+133 −0 src/qd8-f16-qb4w-gemm/gen/qd8-f16-qb4w-gemm-1x2-minmax-scalar.c
+179 −0 src/qd8-f16-qb4w-gemm/gen/qd8-f16-qb4w-gemm-1x4-minmax-scalar.c
+273 −0 src/qd8-f16-qb4w-gemm/gen/qd8-f16-qb4w-gemm-1x8-minmax-scalar.c
+174 −0 src/qd8-f16-qb4w-gemm/gen/qd8-f16-qb4w-gemm-2x2-minmax-scalar.c
+250 −0 src/qd8-f16-qb4w-gemm/gen/qd8-f16-qb4w-gemm-2x4-minmax-scalar.c
+408 −0 src/qd8-f16-qb4w-gemm/gen/qd8-f16-qb4w-gemm-2x8-minmax-scalar.c
+392 −0 src/qd8-f16-qb4w-gemm/gen/qd8-f16-qb4w-gemm-4x4-minmax-scalar.c
+161 −0 src/qs8-gemm/scalar-qd8-f16-qb4w.c.in
+25 −0 src/xnnpack/gemm.h
+1 −3 src/xnnpack/microparams-init.h
+12 −0 test/BUILD.bazel
+11 −8 test/gemm-microkernel-tester.cc
+3,202 −0 test/qd8-f16-qb4w-gemm-minmax.cc
+40 −0 test/qd8-f16-qb4w-gemm-minmax.yaml
+40 −40 tools/generate-gemm-test.py

0 comments on commit 908b5a5

Please sign in to comment.