Issues
- 2
Why remove the neondot and neoni8mm support in QU8? The performance drop a lot on Cortex-A55.
#7605 opened by feng-nxp - 0
Why does the benchmark program run with only one thread?
#7554 opened by mszhanyi - 5
- 2
ARM build on gcc-13 failed with src/reference/unary-elementwise.cc:125:14: error: invalid ‘static_cast’ from type ‘xnn_bfloat16’ to type ‘_Float16’
#7489 opened by xwang233 - 0
- 4
Dynamic shape support follow-up
#6423 opened by adonnini - 2
- 1
how to build with enabled sme arm kleidai
#7324 opened by Junhyuk - 0
b30583647bcc061186f1b1da202ebfcbe197f442
#7368 opened by aswin-devil - 4
`xnn_weights_cache_provider` look_up doesn't work?
#6257 opened by mcr229 - 3
Plans for ARM SME support?
#7121 opened by UsmanNadeem - 7
xnn_table_exp2minus_k_over_16 has inconsistent types
#6806 opened by aiusepsi - 6
Unable to Enable Debug Logging in XNNPACK Despite Setting `XNN_LOG_LEVEL=4` and `-DCMAKE_BUILD_TYPE=Debug`
#7206 opened by wewe5215 - 0
QB4W Development
#6502 opened by digantdesai - 1
Add benchmarks for vbinary microkernels
#6396 opened by fbarchard - 5
unsupported instruction `vpdpbusd'
#6389 opened by AndrewSmithWalter - 0
- 0
f32-spmm missing tests
#7062 opened by oliIMG - 3
- 3
RVV fp16arith testing
#6865 opened by oliIMG - 5
Failed to compile XNNPACK on WoA(Windows on ARM) device.
#6558 opened by zhanweiw - 2
How can I parallelize the execution of this benchmark? (https://github.com/google/XNNPACK/blob/master/bench/spmm-benchmark.h)
#6268 opened by AnonymousYWL - 0
Possible null pointer dereference in logging
#6413 opened by apach301 - 2
Low XNNPACK speedup for ARM Cortex A73
#6653 opened by CrimsonRadiator - 5
ARMv7 (with NEON) can not support on Linux but only support ARMv7 (with NEON) on Android
#6210 opened by cv-on-device - 4
Why is Signal 7 reporting an error on the armv7a platform TEST (F16_VCMUL_NEONFP16ARITH_U8, batch_lt_8) ?
#6201 opened by JamesWang2007 - 2
XNN_FLAG_KEEP_DIM not backwards compatible
#6090 opened by mcr229 - 1
Enable WASM build on GitHub Actions
#6453 opened by jiepan-intel - 2
`f32-vgelu` for HVX build fails
#6683 opened by ejparkqc - 0
Int 8 GEMM kernel with Int32 Accumulator
#6688 opened by FabianSchuetze - 1
4x16s4 fp32-gemm kernel have better performance than default(5x16) kernel for meteor lake
#6480 opened by xujuntwt95329 - 0
Enable QC8/QS8 GEMM/IGEMM for Wasm relaxed integer dot product instruction on x64
#6454 opened by fanchenkong1 - 0
- 1
- 3
- 0
- 1
test/sigmoid_nc_test fails on Hexagon simulator
#6305 opened by steven-johnson - 0
- 0
Work with the gvisor team on this
#6240 opened by justzh - 0
scripts/build-android-armv7.sh fails with NDK 21
#6246 opened by fbarchard - 10
Xnnpack still builds with `+dotprod` and `+fp16` with `-DXNNPACK_ENABLE_ARM_DOTPROD=OFF -DXNNPACK_ENABLE_ARM_FP16_SCALAR=OFF -DXNNPACK_ENABLE_ARM_FP16_VECTOR=OFF`
#6165 opened by misterBart - 0
Enable HEXAGON to build XNNPack
#6232 opened by fbarchard - 2
When running `build -c opt --config android_arm64 :end2end_bench` under bench, `sys/system_properties.h` is missing
#6146 opened by Ciao-Wen-Chen - 1
A segment error occurred while running test case static-reshape-test on the armv7a platform
#6208 opened by JamesWang2007 - 1
Is running TEST(CONVERT_NC_F16_QD8, unit_batch) failed because it does not support armv7a ?
#6191 opened by JamesWang2007 - 1
Fallthroughs should be explicit
#6047 opened by r-barnes - 3
Build Android arm-v7 faild
#6164 opened by zihaomu - 2
XNNPACK on LicheePi Console 4A
#6074 opened by alexshendi - 2
XNNPACK make error scc1: error: invalid feature modifier 'i8mm' in '-march=armv8.2-a+i8mm+fp16'
#5958 opened by qiyuefun - 3