Skip to content
Closed
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
2 changes: 1 addition & 1 deletion backends/xnnpack/third-party/XNNPACK
Submodule XNNPACK updated 64 files
+80 −53 BUILD.bazel
+25 −0 CMakeLists.txt
+5 −5 WORKSPACE
+33 −0 bench/f32-raddstoreexpminusmax.cc
+1 −1 bench/gemm-benchmark.h
+37 −2 bench/models/benchmark.cc
+14 −16 bench/models/fp32-attention.cc
+1 −1 bench/models/models.h
+17 −4 bench/qp8-f32-qb4w-gemm.cc
+16 −0 build_params.bzl
+2 −2 cmake/DownloadCpuinfo.cmake
+2 −2 cmake/DownloadKleidiAI.cmake
+4 −0 cmake/gen/avx256skx_microkernels.cmake
+2 −0 cmake/gen/microkernels.cmake
+2 −1 cmake/gen/neoni8mm_microkernels.cmake
+16 −0 cmake/gen/neonsme2_microkernels.cmake
+16 −0 cmake/gen/neonsme_microkernels.cmake
+4 −0 cmake/gen/rvv_microkernels.cmake
+4 −0 gen/avx256skx_microkernels.bzl
+12 −0 gen/microkernels.bzl
+2 −1 gen/neoni8mm_microkernels.bzl
+14 −0 gen/neonsme2_microkernels.bzl
+14 −0 gen/neonsme_microkernels.bzl
+4 −0 gen/rvv_microkernels.bzl
+10 −4 scripts/generate-f32-raddstoreexpminusmax.sh
+8 −0 scripts/generate-qs8-vlrelu.sh
+2 −2 src/configs/gemm-config.c
+7 −4 src/configs/hardware-config.c
+9 −1 src/configs/raddstoreexpminusmax-config.c
+8 −6 src/configs/unary-elementwise-config.c
+78 −48 src/f32-raddstoreexpminusmax/avx2-rr2-p5.c.in
+24 −24 src/f32-raddstoreexpminusmax/avx512f-rr2-p5.c.in
+32 −32 src/f32-raddstoreexpminusmax/gen/f32-raddstoreexpminusmax-avx2-rr2-p5-u16-acc2.c
+51 −51 src/f32-raddstoreexpminusmax/gen/f32-raddstoreexpminusmax-avx2-rr2-p5-u32-acc2.c
+51 −51 src/f32-raddstoreexpminusmax/gen/f32-raddstoreexpminusmax-avx2-rr2-p5-u32-acc4.c
+24 −24 src/f32-raddstoreexpminusmax/gen/f32-raddstoreexpminusmax-avx2-rr2-p5-u8.c
+237 −0 src/f32-raddstoreexpminusmax/gen/f32-raddstoreexpminusmax-avx256skx-rr2-p5-u16-acc2.c
+269 −0 src/f32-raddstoreexpminusmax/gen/f32-raddstoreexpminusmax-avx256skx-rr2-p5-u32-acc2.c
+273 −0 src/f32-raddstoreexpminusmax/gen/f32-raddstoreexpminusmax-avx256skx-rr2-p5-u32-acc4.c
+218 −0 src/f32-raddstoreexpminusmax/gen/f32-raddstoreexpminusmax-avx256skx-rr2-p5-u8.c
+24 −24 src/f32-raddstoreexpminusmax/gen/f32-raddstoreexpminusmax-avx512f-rr2-p5-u16.c
+32 −32 src/f32-raddstoreexpminusmax/gen/f32-raddstoreexpminusmax-avx512f-rr2-p5-u32-acc2.c
+51 −51 src/f32-raddstoreexpminusmax/gen/f32-raddstoreexpminusmax-avx512f-rr2-p5-u64-acc2.c
+51 −51 src/f32-raddstoreexpminusmax/gen/f32-raddstoreexpminusmax-avx512f-rr2-p5-u64-acc4.c
+9 −8 src/operators/transpose-nd.c
+7 −7 src/packing.cc
+31 −0 src/qp8-f32-qb4w-gemm/qp8-f32-qb4w-gemm-minmax-16x4c16s2-mstep4-neoni8mm.c
+51 −0 src/qs8-vlrelu/gen/qs8-vlrelu-rvv-u1v.c
+51 −0 src/qs8-vlrelu/gen/qs8-vlrelu-rvv-u2v.c
+5 −0 src/qs8-vlrelu/qs8-vlrelu.h
+62 −0 src/qs8-vlrelu/rvv.c.in
+54 −0 src/qu8-vlrelu/gen/qu8-vlrelu-rvv-u1v.c
+54 −0 src/qu8-vlrelu/gen/qu8-vlrelu-rvv-u2v.c
+5 −0 src/qu8-vlrelu/qu8-vlrelu.h
+1 −0 src/xnnpack/gemm.h
+14 −11 src/xnnpack/hardware-config.h
+5 −0 src/xnnpack/raddstoreexpminusmax.h
+9 −9 src/xnnpack/reduce.h
+3 −2 src/xnnpack/subgraph.h
+148 −0 test/f32-raddstoreexpminusmax.cc
+6 −0 test/f32-raddstoreexpminusmax.yaml
+24 −2 test/qp8-f32-qb4w-gemm-minmax.cc
+7 −0 test/qp8-f32-qb4w-gemm-minmax.yaml
+2 −0 tools/update-microkernels.py
Loading