XNNPACK

XNNPACK copied to clipboard

Hello. We're testing convolution performances of XNNPACK on Google Pixel 4. (Android 11, CPU only, 4 threads) We've found that XNNPACK's throughput drops quite significantly in the deeper convolution layers...

cakeng

Dear all, I have been running some benchmarks in the following computer: - Amazon EC2 x86 T2.Large with Ubuntu 18.04. DNN Models: - MobileNet V1 - AlexNet For the following...

pablogh-2000

Remove NEON-FMA checks from scalar direct HWC convolution microkernels The checks were added by mistake

copybara-service[bot]

Added create and setup function for mod kernel

8

> I made a P.R for mod kernel in the [include file](https://github.com/google/XNNPACK/blob/master/include/xnnpack.h) for which I create an issue[ #1612](https://github.com/google/XNNPACK/issues/1612)

carrycooldude

CMake build error when using XNNPACK_USE_SYSTEM_LIBS=ON

8

I'm trying to write a [Spack](https://spack.io) package for XNNPACK. Specifically, I'm trying to allow the build to work using the `-DXNNPACK_USE_SYSTEM_LIBS=ON` flag. The exact command looks like: ```console $ 'cmake'...

adamjstewart

Cache blocking on last-level cache

xnnpack-bot

macOS_arm64 build

5

hi y'all with the aim of compiling the `benchmark_model` in TF (as this depends on XNNPACK) on commit [c2db3a8fae0f6558e9dbdee79e67e74c1e95981c](https://github.com/tensorflow/tensorflow/commit/c2db3a8fae0f6558e9dbdee79e67e74c1e95981c) I was trying to build the end2end_bench using `bazel 4.0.0` (ARM64)...

simonmaurer

WAsm SIMD QS8 GEMM/IGEMM microkernels using ExtMul and ExtAddPair instructions

copybara-service[bot]

Leverage experimental WebAssembly SIMD Prefetch instructions

copybara-service[bot]

TODO: Do we depend on XNNPACK's cpuinfo or do we get the version with RUY. Needs some mods to RUY cmake.

powderluv