GitHub / pytorch/pytorch issues and pull requests
Labelled with: module: cpu
#149673 - Extract reusable portions of elu_kernel into header
Pull Request -
State: open - Opened by swolchok 9 months ago
- 1 comment
Labels: module: cpu, topic: not user facing
#149638 - [Release/2.6] Pin requirements
Pull Request -
State: closed - Opened by ethanwee1 9 months ago
- 1 comment
Labels: oncall: distributed, module: cpu, release notes: releng, module: inductor, module: dynamo
#149631 - [Intel GPU][PT2E] bugfix: use zero-point to decide conv src zp mask
Pull Request -
State: closed - Opened by pytorchbot 9 months ago
- 1 comment
Labels: module: cpu, open source
#149613 - Combine win and win-arm64 templates
Pull Request -
State: closed - Opened by iremyux 9 months ago
- 2 comments
Labels: oncall: distributed, module: cpu, open source, ciflow/binaries, release notes: build, topic: not user facing, ciflow/mps, module: inductor, module: dynamo, ciflow/inductor, release notes: distributed (checkpoint)
#149600 - Fix ModularIndexing simplification
Pull Request -
State: closed - Opened by bobrenjc93 9 months ago
- 1 comment
Labels: module: cpu, topic: not user facing, module: inductor, ciflow/inductor
#149505 - Parallelize sort
Pull Request -
State: closed - Opened by annop-w 9 months ago
- 16 comments
Labels: module: cpu, open source, Merged, Reverted, topic: not user facing, ciflow/inductor, ci-no-td
#149505 - Parallelize sort
Pull Request -
State: open - Opened by annop-w 9 months ago
- 4 comments
Labels: module: cpu, open source, topic: not user facing
#149498 - Fix ValueError issue
Pull Request -
State: closed - Opened by FlintWangacc 9 months ago
- 4 comments
Labels: module: cpu, open source, module: amp (automated mixed precision), release notes: quantization, module: dynamo
#149498 - Fix ValueError issue
Pull Request -
State: open - Opened by FlintWangacc 9 months ago
- 4 comments
Labels: module: cpu, open source, module: amp (automated mixed precision), release notes: quantization, module: dynamo
#149473 - [Intel GPU][PT2E] bugfix: use zero-point to decide conv src zp mask
Pull Request -
State: closed - Opened by ZhiweiYan-96 9 months ago
- 6 comments
Labels: module: cpu, open source, Merged, ciflow/trunk, topic: not user facing, merging, ciflow/xpu
#149473 - [Intel GPU][PT2E] bugfix: use zero-point to decide conv src zp mask
Pull Request -
State: open - Opened by ZhiweiYan-96 9 months ago
- 4 comments
Labels: module: cpu, open source, ciflow/trunk, topic: not user facing, ciflow/xpu
#149435 - Enable fast path for qlinear (static/dynamic) and qadd for AArch64 though ACL directly.
Pull Request -
State: closed - Opened by fadara01 9 months ago
- 9 comments
Labels: module: cpu, open source, module: arm, release notes: quantization, ciflow/linux-aarch64, arm priority
#149435 - Enable fast path for qlinear (static/dynamic) and qadd for AArch64 though ACL directly.
Pull Request -
State: open - Opened by fadara01 9 months ago
- 5 comments
Labels: module: cpu, open source, module: arm, release notes: quantization, ciflow/linux-aarch64, arm priority
#149417 - [Build] Guard per-op headers in ACLUtils.cpp
Pull Request -
State: closed - Opened by malfet 9 months ago
- 5 comments
Labels: module: cpu, fb-exported, Merged, ciflow/trunk, release notes: quantization, merging
#149362 - Add x86-simd-sort accelerated sorting
Pull Request -
State: open - Opened by sterrettm2 9 months ago
- 8 comments
Labels: module: cpu, triaged, open source, ciflow/trunk, topic: not user facing, module: inductor, ciflow/inductor
#149331 - Migrate to new theme
Pull Request -
State: closed - Opened by svekars 9 months ago
- 7 comments
Labels: oncall: distributed, module: docs, module: cpu, Merged, ciflow/trunk, topic: docs, ciflow/mps, module: inductor, module: dynamo, ciflow/inductor, release notes: distributed (checkpoint), merging, suppress-bc-linter
#149331 - Migrate to new theme
Pull Request -
State: open - Opened by svekars 9 months ago
- 5 comments
Labels: oncall: distributed, module: docs, module: cpu, topic: docs, ciflow/mps, module: inductor, module: dynamo, ciflow/inductor, release notes: distributed (checkpoint), suppress-bc-linter
#149268 - Fix mps scaled dot attention
Pull Request -
State: closed - Opened by rakshekaraj 9 months ago
- 6 comments
Labels: module: cpu, triaged, open source, module: amp (automated mixed precision), Stale, release notes: quantization, release notes: mps
#149164 - [ATen-CPU] Add `math.h` for Gelu
Pull Request -
State: closed - Opened by SS-JIA 9 months ago
- 10 comments
Labels: module: cpu, Merged, ciflow/trunk, topic: not user facing, merging
#149148 - Fix printing INT64_MIN
Pull Request -
State: closed - Opened by isuruf 9 months ago
- 6 comments
Labels: module: cpu, open source, Merged, ciflow/trunk, module: inductor, ciflow/inductor, release notes: dynamo, merging
#149122 - Update the heuristic for AArch64 bmm/baddbmm
Pull Request -
State: open - Opened by michalowski-arm 9 months ago
- 12 comments
Labels: module: cpu, triaged, open source, module: arm, Merged, Reverted, ciflow/trunk, release notes: linalg_frontend, ci-no-td
#149122 - Update the heuristic for AArch64 bmm/baddbmm
Pull Request -
State: open - Opened by michalowski-arm 9 months ago
- 8 comments
Labels: module: cpu, triaged, open source, module: arm, ciflow/trunk, release notes: linalg_frontend, merging
#149114 - [Intel GPU] Allow XPU backend in Depthwise_conv2d&3d operators
Pull Request -
State: open - Opened by yucai-intel 9 months ago
- 19 comments
Labels: module: cpu, open source, Merged, Reverted, ciflow/trunk, ciflow/xpu, release notes: xpu, module: xpu, ci-no-td
#149114 - [Intel GPU] Allow XPU backend in Depthwise_conv2d&3d operators
Pull Request -
State: open - Opened by yucai-intel 9 months ago
- 18 comments
Labels: module: cpu, open source, Merged, Reverted, ciflow/trunk, ciflow/xpu, release notes: xpu, module: xpu, ci-no-td
#149046 - Enable modernize-use-default-member-init and related fixes
Pull Request -
State: closed - Opened by cyyever 9 months ago
- 10 comments
Labels: module: cpu, triaged, module: mkldnn, open source, Merged, ciflow/trunk, release notes: quantization, topic: not user facing, module: inductor, ciflow/inductor, ciflow/linux-aarch64
#149019 - Avoid oneDNN primitives when GradMode is enabled on avx2_vnni_2
Pull Request -
State: closed - Opened by CaoE 9 months ago
- 1 comment
Labels: module: cpu, open source, topic: not user facing
#148996 - [codemod][lowrisk] Fix deprecated use of 0/NULL in caffe2/aten/src/ATen/native/quantized/cpu/qnnpack/src/fc-unpack.cc + 1
Pull Request -
State: closed - Opened by r-barnes 9 months ago
- 7 comments
Labels: module: cpu, fb-exported, Merged, ciflow/trunk, release notes: quantization, release notes: cpp, topic: improvements, topic: not user facing, merging
#148887 - Vincent/rebase 2.5
Pull Request -
State: closed - Opened by vincent-tr 9 months ago
- 2 comments
Labels: oncall: distributed, oncall: jit, module: rocm, module: cpu, release notes: releng, fx, module: inductor, module: dynamo, release notes: distributed (checkpoint)
#148878 - Add Half support for weight_norm on CPU
Pull Request -
State: closed - Opened by CaoE 9 months ago
- 10 comments
Labels: module: cpu, open source, module: half, Merged, ciflow/trunk, release notes: nn, ciflow/inductor, merging
#148876 - Use device agnostic APIs and variable names for dtensor
Pull Request -
State: closed - Opened by amathewc 9 months ago
- 23 comments
Labels: oncall: distributed, module: cpu, triaged, module: mkldnn, open source, module: amp (automated mixed precision), NNC, release notes: quantization, topic: not user facing, module: inductor, module: dynamo, release notes: distributed (checkpoint), module: compiled autograd
#148757 - Fix Wc++98-compat-extra-semi
Pull Request -
State: closed - Opened by cyyever 9 months ago
- 4 comments
Labels: module: cpu, open source, Merged, ciflow/trunk, topic: not user facing, merging
#148727 - Add ccode for FloorDiv
Pull Request -
State: closed - Opened by kalpit-meta-1 9 months ago
- 13 comments
Labels: module: cpu, fb-exported, Merged, ciflow/trunk, topic: not user facing, ciflow/inductor, merging
#148653 - Enable qint8 and quint8 add for AArch64 using ACL directly
Pull Request -
State: closed - Opened by fadara01 9 months ago
- 7 comments
Labels: module: cpu, open source, module: arm, Merged, release notes: quantization, merging, ciflow/linux-aarch64, arm priority
#148640 - [Intel GPU][quant] Refine zero-point memory creation
Pull Request -
State: closed - Opened by ZhiweiYan-96 9 months ago
- 10 comments
Labels: module: cpu, open source, Merged, ciflow/trunk, topic: not user facing, ciflow/inductor, keep-going, merging, ciflow/xpu
#148638 - Remove cppcoreguidelines-pro-type-member-init_fix suppression
Pull Request -
State: closed - Opened by cyyever 9 months ago
- 6 comments
Labels: oncall: jit, module: cpu, triaged, open source, Merged, ciflow/trunk, release notes: jit, module: dynamo, ciflow/inductor
#148638 - Remove cppcoreguidelines-pro-type-member-init_fix suppression
Pull Request -
State: closed - Opened by cyyever 9 months ago
- 6 comments
Labels: oncall: jit, module: cpu, triaged, open source, Merged, ciflow/trunk, release notes: jit, module: dynamo, ciflow/inductor, merging
#148585 - Enable fast qlinear static/dynamic path for AArch64 through ACL directly
Pull Request -
State: closed - Opened by fadara01 9 months ago
- 8 comments
Labels: module: cpu, open source, module: arm, Merged, release notes: quantization, merging, ciflow/linux-aarch64, arm priority
#148583 - Enable fast qlinear static/dynamic path for AArch64 through ACL directly
Pull Request -
State: closed - Opened by fadara01 9 months ago
- 2 comments
Labels: module: cpu, open source, release notes: quantization
#148542 - Enable Direct Use of Arm Compute Library (ACL) in ATen
Pull Request -
State: closed - Opened by fadara01 9 months ago
- 7 comments
Labels: module: cpu, triaged, open source, module: arm, topic: not user facing, ciflow/linux-aarch64, arm priority
#148542 - Enable Direct Use of Arm Compute Library (ACL) in ATen
Pull Request -
State: open - Opened by fadara01 9 months ago
- 6 comments
Labels: module: cpu, open source, module: arm, topic: not user facing, ciflow/linux-aarch64, arm priority
#148529 - Fix clang-tidy bugprone* warnings
Pull Request -
State: closed - Opened by cyyever 9 months ago
- 9 comments
Labels: oncall: distributed, module: cpu, triaged, open source, Merged, ciflow/trunk, release notes: quantization, topic: not user facing, module: dynamo, ciflow/inductor, module: compiled autograd, release notes: inductor (aoti)
#148529 - Fix clang-tidy bugprone* warnings
Pull Request -
State: closed - Opened by cyyever 9 months ago
- 9 comments
Labels: oncall: distributed, module: cpu, triaged, open source, Merged, ciflow/trunk, release notes: quantization, topic: not user facing, module: dynamo, ciflow/inductor, merging, module: compiled autograd, release notes: inductor (aoti)
#148522 - [Intel GPU][pt2e] Enable quantized grouped convolution at XPU
Pull Request -
State: open - Opened by ZhiweiYan-96 9 months ago
- 1 comment
Labels: module: cpu, open source, ciflow/trunk, topic: not user facing, module: inductor, ciflow/inductor, ciflow/xpu
#148423 - [Intel GPU][pt2e]: Collapse 3D input to 2D for matmul in qlinear_pointwise_binary fusion
Pull Request -
State: open - Opened by ZhiweiYan-96 9 months ago
- 3 comments
Labels: module: cpu, open source, ciflow/trunk, topic: not user facing, module: inductor, ciflow/inductor, keep-going, ciflow/xpu
#148407 - Enable ASAN on inductor CUDA tests
Pull Request -
State: closed - Opened by cyyever 9 months ago
- 1 comment
Labels: module: cpu, triaged, open source, topic: not user facing, module: inductor, ciflow/inductor
#148407 - Enable ASAN on inductor CUDA tests
Pull Request -
State: closed - Opened by cyyever 9 months ago
- 1 comment
Labels: module: cpu, triaged, open source, topic: not user facing, module: inductor, ciflow/inductor
#148362 - Fix condition for `CONVERT_NON_VECTORIZED_INIT` invocation
Pull Request -
State: closed - Opened by malfet 9 months ago
- 3 comments
Labels: module: cpu, Merged, ciflow/trunk, release notes: build, topic: bug fixes, topic: build, merging
#148354 - [BE] Use `C10_DIAGNOSTIC_PUSH_AND_IGNORED_IF_DEFINED`
Pull Request -
State: closed - Opened by malfet 9 months ago
- 4 comments
Labels: module: cpu, Merged, release notes: build, topic: bug fixes, topic: build, merging
#148346 - Symmetrization of Cholesky backward gradient
Pull Request -
State: closed - Opened by ayghri 9 months ago
- 3 comments
Labels: oncall: distributed, module: cpu, triaged, module: mkldnn, open source, module: amp (automated mixed precision), release notes: quantization, release notes: releng, module: inductor, module: dynamo, release notes: distributed (checkpoint)
#148284 - [BE] Fix extra semicolon warning
Pull Request -
State: closed - Opened by malfet 9 months ago
- 6 comments
Labels: module: cpu, better-engineering, Merged, ciflow/trunk, topic: not user facing, merging
#148066 - use identity op for alpha=inf in torch.celu and quantized_celu
Pull Request -
State: open - Opened by redwrasse 9 months ago
- 2 comments
Labels: module: cpu, triaged, open source, Stale, release notes: quantization
#148049 - Fix `torch.nn.functional.hardswish` gradients corner case
Pull Request -
State: closed - Opened by zeshengzong 9 months ago
- 32 comments
Labels: module: autograd, module: cpu, triaged, open source, Merged, Reverted, ciflow/trunk, release notes: nn, merging, ci-no-td
#148049 - Fix `torch.nn.functional.hardswish` gradients corner case
Pull Request -
State: open - Opened by zeshengzong 9 months ago
- 1 comment
Labels: module: autograd, module: cpu, open source, release notes: nn
#147969 - [Intel GPU] Avoid including CPU oneDNN header files for Intel GPU
Pull Request -
State: open - Opened by EikanWang 9 months ago
- 1 comment
Labels: module: cpu, module: mkldnn, open source, ciflow/trunk, topic: not user facing, keep-going, ciflow/xpu, ciflow/linux-aarch64
#147964 - [test][do not merge] test on 90e3a3d86d6139a7b00bdf56bdfe0f63ad18e980
Pull Request -
State: closed - Opened by yanbing-j 9 months ago
- 1 comment
Labels: module: cpu, module: mkldnn, open source, ciflow/binaries, ciflow/trunk, topic: not user facing, intel, ciflow/linux-aarch64
#147951 - [WIP][Intel GPU][do not merge] Enable SDPA on XPU
Pull Request -
State: closed - Opened by DDEle 9 months ago
- 5 comments
Labels: module: cpu, module: mkldnn, open source, ciflow/trunk, topic: not user facing, module: inductor, module: dynamo, ciflow/inductor, ciflow/xpu, ciflow/linux-aarch64
#147951 - [WIP][Intel GPU][do not merge] Enable SDPA on XPU
Pull Request -
State: open - Opened by DDEle 9 months ago
- 5 comments
Labels: module: cpu, module: mkldnn, open source, ciflow/trunk, topic: not user facing, module: inductor, module: dynamo, ciflow/inductor, ciflow/xpu, ciflow/linux-aarch64
#147864 - Parallelize bf16->f32 conversion for gemm(bf16:bf16->bf16)
Pull Request -
State: closed - Opened by aditew01 10 months ago
- 2 comments
Labels: module: cpu, triaged, open source, module: arm, topic: not user facing
#147864 - Parallelize bf16->f32 conversion for gemm(bf16:bf16->bf16)
Pull Request -
State: closed - Opened by aditew01 10 months ago
- 2 comments
Labels: module: cpu, triaged, open source, module: arm, topic: not user facing
#147807 - [AOTI][refactor] Fix a typo
Pull Request -
State: closed - Opened by desertfire 10 months ago
- 4 comments
Labels: module: cpu, Merged, ciflow/trunk, topic: not user facing, module: inductor, ciflow/inductor
#147766 - [Inductor-CPU] Memory allocator lock contention is slowing down templated GEMMs
Issue -
State: closed - Opened by sanchitintel 10 months ago
- 1 comment
Labels: module: performance, module: cpu, oncall: cpu inductor
#147693 - [Intel GPU] OneDNN primitive cache support for Int4 WOQ gemm on XPU
Pull Request -
State: open - Opened by baodii 10 months ago
- 37 comments
Labels: module: cpu, module: mkldnn, open source, ciflow/trunk, topic: not user facing, keep-going, ciflow/xpu, release notes: xpu, module: xpu
#147629 - torch.sort: Optimize memory usage with (dtype_indices: ScalarType, dynamic_indices_dtype: bool) options
Pull Request -
State: open - Opened by voidbag 10 months ago
- 15 comments
Labels: module: cpu, triaged, open source, Stale, release notes: mps, module: inductor
#147614 - [Intel GPU] Enable SDPA on XPU
Pull Request -
State: closed - Opened by DDEle 10 months ago
- 20 comments
Labels: module: cpu, open source, Merged, ciflow/trunk, topic: not user facing, module: inductor, keep-going, merging, ciflow/xpu
#147592 - Fix log2, PowByNatural printing
Pull Request -
State: closed - Opened by isuruf 10 months ago
- 6 comments
Labels: module: cpu, open source, Merged, ciflow/trunk, module: inductor, ciflow/inductor, release notes: inductor, merging
#147588 - Also support non-contiguous activation for torch._weight_int8pack_mm on CPU
Pull Request -
State: closed - Opened by sanchitintel 10 months ago
- 9 comments
Labels: module: cpu, open source, Merged, ciflow/trunk, topic: not user facing, intel, merging
#147556 - [caffe2] Ignore compiler option when building using clang
Pull Request -
State: closed - Opened by Nicoshev 10 months ago
- 10 comments
Labels: module: cpu, fb-exported, Merged, ciflow/trunk, topic: not user facing, merging
#147501 - removed zero dim cpu logic from fake_tensor.py
Pull Request -
State: open - Opened by zero000064 10 months ago
- 31 comments
Labels: oncall: distributed, module: cpu, triaged, open source, module: amp (automated mixed precision), NNC, release notes: quantization, topic: not user facing, module: inductor, module: dynamo, release notes: distributed (checkpoint), module: compiled autograd
#147466 - add the `torch.float8_e8m0fnu` dtype to PyTorch
Pull Request -
State: closed - Opened by vkuzo 10 months ago
- 6 comments
Labels: module: cpu, Merged, ciflow/trunk, release notes: quantization, merging
#147462 - add the torch.float8_e8m0fnu` dtype to PyTorch
Pull Request -
State: closed - Opened by vkuzo 10 months ago
- 4 comments
Labels: module: cpu, ciflow/trunk, release notes: quantization
#147367 - Force build to conform C++ standard on windows by adding `/permissive-` flag
Pull Request -
State: closed - Opened by Stonepia 10 months ago
- 26 comments
Labels: oncall: distributed, oncall: jit, module: windows, module: cpu, module: mkldnn, open source, NNC, release notes: jit, module: inductor, module: dynamo, release notes: distributed (checkpoint), module: compiled autograd, module: xpu
#147349 - Refine XPU oneDNN context manager API
Pull Request -
State: open - Opened by guangyey 10 months ago
- 20 comments
Labels: module: cpu, open source, ciflow/trunk, topic: improvements, ciflow/xpu, release notes: xpu
#147337 - Enable a fast path for (static) qlinear for AArch64 through ACL directly.
Pull Request -
State: open - Opened by fadara01 10 months ago
- 7 comments
Labels: module: cpu, triaged, open source, module: arm, release notes: quantization, release notes: releng, ciflow/linux-aarch64, arm priority
#147337 - Enable a fast path for (static) qlinear for AArch64 through ACL directly.
Pull Request -
State: closed - Opened by fadara01 10 months ago
- 8 comments
Labels: module: cpu, triaged, open source, module: arm, release notes: quantization, release notes: releng, ciflow/linux-aarch64, arm priority
#147322 - Add NEON implementation for 8 bit quantized embedding bag on aarch64
Pull Request -
State: closed - Opened by annop-w 10 months ago
- 6 comments
Labels: module: cpu, open source, module: arm, Merged, ciflow/trunk, release notes: quantization, topic: performance, merging, ciflow/linux-aarch64, arm priority
#147303 - fp16 channels_last created Nan in batchnorm backward
Issue -
State: closed - Opened by jthakurH 10 months ago
- 1 comment
Labels: module: cpu, triaged, bug
#147292 - Fix arvr macOS buck pytorch builds
Pull Request -
State: closed - Opened by stepanhruda 10 months ago
- 7 comments
Labels: module: cpu, fb-exported, Merged, ciflow/trunk, release notes: quantization, merging
#147119 - [Edited] Add docstring to improve documentation
Pull Request -
State: closed - Opened by MayureshMore 10 months ago
- 3 comments
Labels: oncall: distributed, oncall: jit, module: rocm, module: cpu, module: mkldnn, open source, release notes: quantization, release notes: releng, fx, module: inductor, module: dynamo
#147072 - [Inductor] Set prop_kind to forward_inference when grad is not needed for mkldnn_linear_pointwise and mkldnn_convolution_pointwise
Pull Request -
State: closed - Opened by jiayisunx 10 months ago
- 3 comments
Labels: module: cpu, open source, ciflow/trunk, ciflow/inductor, release notes: inductor, merging
#147072 - [Inductor] Set prop_kind to forward_inference when grad is not needed for mkldnn_linear_pointwise and mkldnn_convolution_pointwise
Pull Request -
State: open - Opened by jiayisunx 10 months ago
- 1 comment
Labels: module: cpu, open source, ciflow/trunk, ciflow/inductor, release notes: inductor
#147068 - [Inductor][CPP] Add transposed B matrix support for CppMicroGemmFP32Vec
Pull Request -
State: closed - Opened by CaoE 10 months ago
- 9 comments
Labels: module: cpu, open source, Merged, ciflow/trunk, topic: not user facing, module: inductor, ciflow/inductor, merging
#147068 - [Inductor][CPP]Add transposed B matrix support for CppMicroGemmFP32Vec
Pull Request -
State: open - Opened by CaoE 10 months ago
- 4 comments
Labels: module: cpu, open source, ciflow/trunk, topic: not user facing, module: inductor, ciflow/inductor
#147067 - Separate transpose from memory load/store and add load size support for convert_to_int32
Pull Request -
State: closed - Opened by CaoE 10 months ago
- 3 comments
Labels: module: cpu, open source, Merged, ciflow/trunk, topic: not user facing, ciflow/inductor, merging
#147054 - Fix for issue #142834, Segmentation fault in replication_pad2d_backward
Pull Request -
State: closed - Opened by AmalDevHaridevan 10 months ago
- 3 comments
Labels: module: cpu, triaged, open source, Stale
#146989 - [BE]: Try to remove unused type ignores - attempt 1
Pull Request -
State: open - Opened by Skylion007 10 months ago
- 1 comment
Labels: oncall: distributed, oncall: jit, module: rocm, module: cpu, open source, module: amp (automated mixed precision), release notes: quantization, release notes: distributed (c10d), fx, ciflow/mps, module: inductor, module: dynamo, ciflow/inductor, module: compiled autograd, oncall: distributed checkpointing
#146942 - [Inductor] FX backend via Wrapper IR
Pull Request -
State: open - Opened by blaine-rister 10 months ago
- 12 comments
Labels: module: cpu, Merged, Reverted, ciflow/trunk, module: inductor, ciflow/inductor, release notes: inductor, ci-no-td, release notes: inductor (aoti)
#146942 - [Inductor] FX backend via Wrapper IR
Pull Request -
State: open - Opened by blaine-rister 10 months ago
- 7 comments
Labels: module: cpu, ciflow/trunk, module: inductor, ciflow/inductor, release notes: inductor, release notes: inductor (aoti)
#146937 - [draft] ROCm MX-FP8 Scale_mm() Support
Pull Request -
State: closed - Opened by petrex 10 months ago
- 2 comments
Labels: module: rocm, module: cpu, open source, release notes: quantization
#146929 - Support QNX SDP 8.0 in Pytorch Mobile
Pull Request -
State: closed - Opened by eleir9268 10 months ago
- 3 comments
Labels: module: cpu, triaged, open source, oncall: mobile, Stale, release notes: quantization
#146880 - [XPU] Align XPU convolution_backward output layout between fake tensor and real output tensor.
Pull Request -
State: closed - Opened by etaf 10 months ago
- 1 comment
Labels: module: cpu, open source, Merged, topic: not user facing
#146843 - [inductor][cpu] Move VNNI weight packing into AMX GEMM kernel for contiguous BMM weights
Pull Request -
State: closed - Opened by frost-intel 10 months ago
- 13 comments
Labels: module: cpu, open source, Merged, ciflow/trunk, topic: not user facing, module: inductor, ciflow/inductor, merging
#146826 - add mkldnn maxpool support on CPU dispatch
Pull Request -
State: closed - Opened by CaoE 10 months ago
- 5 comments
Labels: module: cpu, module: mkldnn, open source, ciflow/trunk, topic: not user facing, ciflow/inductor, ciflow/linux-aarch64
#146826 - add mkldnn_max_pool2d support on CPU dispatch
Pull Request -
State: open - Opened by CaoE 10 months ago
- 2 comments
Labels: module: cpu, open source, ciflow/trunk, topic: not user facing
#146823 - Use mkldnn_max_pool2d for max_pool2d when indices is not needed
Pull Request -
State: open - Opened by CaoE 10 months ago
- 3 comments
Labels: module: cpu, open source, ciflow/trunk, ciflow/periodic, module: inductor, ciflow/inductor, release notes: inductor
#146823 - Use mkldnn_max_pool2d for max_pool2d when indices is not needed
Pull Request -
State: closed - Opened by CaoE 10 months ago
- 3 comments
Labels: module: cpu, module: mkldnn, open source, ciflow/trunk, ciflow/periodic, module: inductor, ciflow/inductor, release notes: inductor, ciflow/linux-aarch64
#146812 - fix #145064 , added error checking for empty tensor in _pdist_forward
Pull Request -
State: closed - Opened by AmalDevHaridevan 10 months ago
- 5 comments
Labels: oncall: distributed, module: cpu, triaged, module: mkldnn, open source, NNC, ciflow/trunk, release notes: quantization, topic: not user facing, module: inductor, module: dynamo, module: distributed_checkpoint, module: compiled autograd
#146781 - [Inductor-CPU] FP16 X int8 WoQ GEMM for M <= 4 with FP16 accum & compute
Pull Request -
State: closed - Opened by sanchitintel 10 months ago
- 3 comments
Labels: module: cpu, open source, Stale, module: inductor, module: dynamo, ciflow/inductor
#146777 - Enable explicitly vectorized `_weight_int8pack_mm` op for FP16 dtype on x86_64 CPU
Pull Request -
State: closed - Opened by sanchitintel 10 months ago
- 8 comments
Labels: module: cpu, triaged, open source, Stale, ciflow/trunk, intel, release notes: intel
#146777 - Enable vectorized `_weight_int8pack_mm` op on CPU for FP16
Pull Request -
State: open - Opened by sanchitintel 10 months ago
- 1 comment
Labels: module: cpu, open source, ciflow/trunk, release notes: performance_as_product, intel
#146690 - Enable pt2e quantization path for arm
Pull Request -
State: open - Opened by choudhary-devang 10 months ago
- 56 comments
Labels: module: cpu, triaged, open source, module: arm, release notes: quantization, release notes: AO frontend