Ecosyste.ms: Issues

An open API service for providing issue and pull request metadata for open source projects.

GitHub / openxla/iree issues and pull requests

#13166 - Use PJRT-based custom call registration to register standard JAX custom calls (e.g. Eigh)

Issue - State: open - Opened by vmurali over 1 year ago
Labels: integrations/stablehlo, p1

#13166 - Use PJRT-based custom call registration to register standard JAX custom calls (e.g. Eigh)

Issue - State: open - Opened by vmurali over 1 year ago
Labels: integrations/stablehlo, p1

#13166 - Use PJRT-based custom call registration to register standard JAX custom calls (e.g. Eigh)

Issue - State: open - Opened by vmurali over 1 year ago
Labels: integrations/stablehlo, p1

#13166 - Use PJRT-based custom call registration to register standard JAX custom calls (e.g. Eigh)

Issue - State: open - Opened by vmurali over 1 year ago
Labels: integrations/stablehlo, p1

#13165 - Create JAX's custom call shim for GPUs

Issue - State: open - Opened by vmurali over 1 year ago
Labels: integrations/stablehlo, p1

#13165 - Create JAX's custom call shim for GPUs

Issue - State: open - Opened by vmurali over 1 year ago
Labels: integrations/stablehlo, p1

#13165 - Create JAX's custom call shim for GPUs

Issue - State: open - Opened by vmurali over 1 year ago
Labels: integrations/stablehlo, p1

#13165 - Create JAX's custom call shim for GPUs

Issue - State: open - Opened by vmurali over 1 year ago
Labels: integrations/stablehlo, p1

#13165 - Create JAX's custom call shim for GPUs

Issue - State: open - Opened by vmurali over 1 year ago
Labels: integrations/stablehlo, p1

#13164 - Update the MHLO Input Conversion pass to create `func.call` for `mhlo.custom_call`

Issue - State: open - Opened by vmurali over 1 year ago - 1 comment
Labels: integrations/stablehlo, p1

#13164 - Update the MHLO Input Conversion pass to create `func.call` for `mhlo.custom_call`

Issue - State: open - Opened by vmurali over 1 year ago - 1 comment
Labels: integrations/stablehlo, p1

#13164 - Update the MHLO Input Conversion pass to create `func.call` for `mhlo.custom_call`

Issue - State: open - Opened by vmurali over 1 year ago - 1 comment
Labels: integrations/stablehlo, p1

#13163 - Move JAX's custom call example to openxla-pjrt-plugin

Issue - State: open - Opened by vmurali over 1 year ago
Labels: integrations/stablehlo, p1

#13163 - Move JAX's custom call example to openxla-pjrt-plugin

Issue - State: open - Opened by vmurali over 1 year ago
Labels: integrations/stablehlo, p1

#13163 - Move JAX's custom call example to openxla-pjrt-plugin

Issue - State: open - Opened by vmurali over 1 year ago
Labels: integrations/stablehlo, p1

#13163 - Move JAX's custom call example to openxla-pjrt-plugin

Issue - State: open - Opened by vmurali over 1 year ago
Labels: integrations/stablehlo, p1

#13163 - Move JAX's custom call example to openxla-pjrt-plugin

Issue - State: open - Opened by vmurali over 1 year ago
Labels: integrations/stablehlo, p1

#13145 - [Epic] Determine workflows for profiling JAX -> IREE + CUDA

Issue - State: closed - Opened by ScottTodd over 1 year ago - 10 comments
Labels: performance ⚡, runtime/tools, integrations/stablehlo, epic, p1

#13145 - [Epic] Determine workflows for profiling JAX -> IREE + CUDA

Issue - State: closed - Opened by ScottTodd over 1 year ago - 10 comments
Labels: performance ⚡, runtime/tools, integrations/stablehlo, epic, p1

#13145 - [Epic] Determine workflows for profiling JAX -> IREE + CUDA

Issue - State: closed - Opened by ScottTodd over 1 year ago - 10 comments
Labels: performance ⚡, runtime/tools, integrations/stablehlo, epic, p1

#13131 - Remove "nokokoro" tag from `dynamic_library_test`.

Pull Request - State: closed - Opened by ScottTodd over 1 year ago - 3 comments
Labels: cleanup 🧹

#13129 - Split benchamrk suite build target into default and long-running

Pull Request - State: closed - Opened by pzread over 1 year ago - 4 comments
Labels: benchmarks:cuda, benchmarks:x86_64, benchmarks:comp-stats, benchmarks:android-cpu, benchmarks:android-gpu, benchmarks:vulkan-nvidia

#13115 - Supporting non aligned reduction splitting

Issue - State: open - Opened by ThomasRaoux over 1 year ago - 22 comments
Labels: enhancement ➕, performance ⚡

#13110 - Introduce a strategy that improves pack(LHS) using transform dialect.

Pull Request - State: closed - Opened by hanhanW over 1 year ago - 7 comments

#13104 - Unaligned matmul work

Issue - State: open - Opened by allieculp over 1 year ago - 11 comments

#13095 - Separate out Cuda tests that depend on specific architecture

Issue - State: closed - Opened by GMNGeoffrey over 1 year ago - 6 comments

#13095 - Separate out Cuda tests that depend on specific architecture

Issue - State: closed - Opened by GMNGeoffrey over 1 year ago - 6 comments

#13095 - Separate out Cuda tests that depend on specific architecture

Issue - State: closed - Opened by GMNGeoffrey over 1 year ago - 6 comments

#13089 - Support multi channels by supporting channel splitting

Pull Request - State: closed - Opened by okkwon over 1 year ago - 10 comments

#13089 - Support multi channels by supporting channel splitting

Pull Request - State: closed - Opened by okkwon over 1 year ago - 10 comments

#13089 - Support multi channels by supporting channel splitting

Pull Request - State: closed - Opened by okkwon over 1 year ago - 10 comments

#13089 - Support multi channels by supporting channel splitting

Pull Request - State: closed - Opened by okkwon over 1 year ago - 10 comments

#13089 - Support multi channels by supporting channel splitting

Pull Request - State: closed - Opened by okkwon over 1 year ago - 10 comments

#13069 - [CI] Revive SHARK pytest benchmarks.

Pull Request - State: closed - Opened by monorimet over 1 year ago - 7 comments

#13069 - [CI] Revive SHARK pytest benchmarks.

Pull Request - State: closed - Opened by monorimet over 1 year ago - 7 comments

#13043 - [Epic] Convert to using TF Python API & reap the benefits in CI and benchmarks

Issue - State: closed - Opened by julianwa over 1 year ago - 12 comments
Labels: epic, p1

#13043 - [Epic] Convert to using TF Python API & reap the benefits in CI and benchmarks

Issue - State: closed - Opened by julianwa over 1 year ago - 12 comments
Labels: epic, p1

#13043 - [Epic] Convert to using TF Python API & reap the benefits in CI and benchmarks

Issue - State: closed - Opened by julianwa over 1 year ago - 12 comments
Labels: epic, p1

#13043 - [Epic] Convert to using TF Python API & reap the benefits in CI and benchmarks

Issue - State: closed - Opened by julianwa over 1 year ago - 12 comments
Labels: epic, p1

#13043 - [Epic] Convert to using TF Python API & reap the benefits in CI and benchmarks

Issue - State: closed - Opened by julianwa over 1 year ago - 12 comments
Labels: epic, p1

#13027 - Track down pmap tests and demo model

Issue - State: open - Opened by julianwa over 1 year ago - 6 comments

#12992 - [TM_TENSOR] Add tm_tensor.sort op and add it's lowering to Linalg_Ext

Pull Request - State: closed - Opened by Abhishek-Varma over 1 year ago - 13 comments

#12992 - [TM_TENSOR] Add tm_tensor.sort op and add it's lowering to Linalg_Ext

Pull Request - State: closed - Opened by Abhishek-Varma over 1 year ago - 13 comments

#12971 - [DNS] Test Android benchmark

Pull Request - State: open - Opened by pzread over 1 year ago - 1 comment
Labels: benchmarks:cuda, benchmarks:x86_64, benchmarks:comp-stats, benchmarks:android-cpu, benchmarks:android-gpu, benchmarks:vulkan-nvidia

#12971 - [DNS] Test Android benchmark

Pull Request - State: open - Opened by pzread over 1 year ago - 1 comment
Labels: benchmarks:cuda, benchmarks:x86_64, benchmarks:comp-stats, benchmarks:android-cpu, benchmarks:android-gpu, benchmarks:vulkan-nvidia

#12971 - [DNS] Test Android benchmark

Pull Request - State: open - Opened by pzread over 1 year ago - 1 comment
Labels: benchmarks:cuda, benchmarks:x86_64, benchmarks:comp-stats, benchmarks:android-cpu, benchmarks:android-gpu, benchmarks:vulkan-nvidia

#12971 - [DNS] Test Android benchmark

Pull Request - State: open - Opened by pzread over 1 year ago - 1 comment
Labels: benchmarks:cuda, benchmarks:x86_64, benchmarks:comp-stats, benchmarks:android-cpu, benchmarks:android-gpu, benchmarks:vulkan-nvidia

#12971 - [DNS] Test Android benchmark

Pull Request - State: open - Opened by pzread over 1 year ago - 1 comment
Labels: benchmarks:cuda, benchmarks:x86_64, benchmarks:comp-stats, benchmarks:android-cpu, benchmarks:android-gpu, benchmarks:vulkan-nvidia

#12952 - Test issue

Issue - State: closed - Opened by aaron-schneider over 1 year ago

#12952 - Test issue

Issue - State: closed - Opened by aaron-schneider over 1 year ago

#12952 - Test issue

Issue - State: closed - Opened by aaron-schneider over 1 year ago

#12949 - [Epic] Componentize XLA SPMD Partitioner

Issue - State: closed - Opened by julianwa over 1 year ago
Labels: epic, p1

#12899 - Create benchmarks for user-specificed matmul shapes for f16 and f32 datatype

Issue - State: open - Opened by manishucsd over 1 year ago - 7 comments
Labels: enhancement ➕, infrastructure/benchmark, benchmarks:cuda, benchmarks:comp-stats

#12873 - [Epic] Add Models in CUDA Benchmark Focus Suite to IREE Benchmarking System

Issue - State: closed - Opened by mariecwhite over 1 year ago - 11 comments
Labels: epic, p1

#12873 - [Epic] Add Models in CUDA Benchmark Focus Suite to IREE Benchmarking System

Issue - State: closed - Opened by mariecwhite over 1 year ago - 11 comments
Labels: epic, p1

#12873 - [Epic] Add Models in CUDA Benchmark Focus Suite to IREE Benchmarking System

Issue - State: closed - Opened by mariecwhite over 1 year ago - 11 comments
Labels: epic, p1

#12873 - [Epic] Add Models in CUDA Benchmark Focus Suite to IREE Benchmarking System

Issue - State: closed - Opened by mariecwhite over 1 year ago - 11 comments
Labels: epic, p1

#12873 - [Epic] Add Models in CUDA Benchmark Focus Suite to IREE Benchmarking System

Issue - State: closed - Opened by mariecwhite over 1 year ago - 11 comments
Labels: epic, p1

#12859 - Implement int4(3, 2) matmul weight quantization for matmul

Issue - State: open - Opened by IanNod over 1 year ago - 59 comments
Labels: enhancement ➕, codegen

#12859 - Support i4/i3/i2 weight quantized matmul

Issue - State: open - Opened by IanNod over 1 year ago - 64 comments
Labels: enhancement ➕, codegen

#12859 - Support i4/i3/i2 weight quantized matmul

Issue - State: open - Opened by IanNod over 1 year ago - 64 comments
Labels: enhancement ➕, codegen

#12859 - Support i4/i3/i2 weight quantized matmul

Issue - State: open - Opened by IanNod over 1 year ago - 64 comments
Labels: enhancement ➕, codegen

#12859 - Support i4/i3/i2 weight quantized matmul

Issue - State: open - Opened by IanNod over 1 year ago - 64 comments
Labels: enhancement ➕, codegen

#12824 - Crash in LLVMGPUTileAndDistribute / Tiling interface

Issue - State: open - Opened by nicolasvasilache over 1 year ago - 13 comments
Labels: bug 🐞, codegen/nvvm

#12807 - Add convolution implicit gemm strategy behind a flag to LLVMGPU

Pull Request - State: closed - Opened by qedawkins over 1 year ago

#12806 - Add a transform dialect strategy for implicit gemm

Pull Request - State: closed - Opened by qedawkins over 1 year ago

#12756 - [NFC] Deprecate LinalgExt TransformOps.

Pull Request - State: closed - Opened by hanhanW over 1 year ago - 2 comments

#12756 - [NFC] Deprecate LinalgExt TransformOps.

Pull Request - State: closed - Opened by hanhanW over 1 year ago - 2 comments

#12733 - Convolution-like generic not fusing with transpose through reshapes

Issue - State: closed - Opened by qedawkins over 1 year ago - 7 comments
Labels: compiler/dialects

#12713 - [metal] Implement a Metal HAL driver

Pull Request - State: open - Opened by antiagainst over 1 year ago - 6 comments
Labels: codegen/spirv, hal/metal

#12709 - Fix mobilenet_v2_int8 model input typo

Pull Request - State: closed - Opened by hcindyl over 1 year ago - 3 comments
Labels: benchmarks:x86_64

#12709 - Fix mobilenet_v2_int8 model input typo

Pull Request - State: closed - Opened by hcindyl over 1 year ago - 3 comments
Labels: benchmarks:x86_64

#12684 - Enable Arm server support in CI, and create corresponding Python package

Issue - State: closed - Opened by freddan80 over 1 year ago - 32 comments
Labels: enhancement ➕, infrastructure

#12684 - Enable Arm server support in CI, and create corresponding Python package

Issue - State: open - Opened by freddan80 over 1 year ago - 28 comments
Labels: enhancement ➕, infrastructure

#12678 - Sever IREE repo connection with mlir-hlo

Issue - State: open - Opened by julianwa over 1 year ago - 5 comments
Labels: infrastructure, integrations/stablehlo

#12678 - Sever IREE repo connection with mlir-hlo

Issue - State: open - Opened by julianwa over 1 year ago - 5 comments
Labels: infrastructure, integrations/stablehlo

#12678 - Sever IREE repo connection with mlir-hlo

Issue - State: open - Opened by julianwa over 1 year ago - 8 comments
Labels: infrastructure, integrations/stablehlo

#12678 - Sever IREE repo connection with mlir-hlo

Issue - State: open - Opened by julianwa over 1 year ago - 5 comments
Labels: infrastructure, integrations/stablehlo

#12678 - Sever IREE repo connection with mlir-hlo

Issue - State: open - Opened by julianwa over 1 year ago - 8 comments
Labels: infrastructure, integrations/stablehlo

#12625 - Adding a local executable plugin mechanism.

Pull Request - State: closed - Opened by benvanik over 1 year ago - 4 comments
Labels: hal/cpu

#12625 - Adding a local executable plugin mechanism.

Pull Request - State: closed - Opened by benvanik over 1 year ago - 4 comments
Labels: hal/cpu

#12625 - Adding a local executable plugin mechanism.

Pull Request - State: closed - Opened by benvanik over 1 year ago - 4 comments
Labels: hal/cpu

#12625 - Adding a local executable plugin mechanism.

Pull Request - State: closed - Opened by benvanik over 1 year ago - 4 comments
Labels: hal/cpu

#12625 - Adding a local executable plugin mechanism.

Pull Request - State: closed - Opened by benvanik over 1 year ago - 4 comments
Labels: hal/cpu

#12595 - CMake configure time has increased a lot

Issue - State: open - Opened by stellaraccident over 1 year ago - 5 comments

#12595 - CMake configure time has increased a lot

Issue - State: open - Opened by stellaraccident over 1 year ago - 5 comments

#12595 - CMake configure time has increased a lot

Issue - State: open - Opened by stellaraccident over 1 year ago - 5 comments

#12536 - RISC-V multi-core computation

Issue - State: closed - Opened by hahayeong over 1 year ago - 4 comments

#12457 - Fix TorchInductor Dependency Issue

Issue - State: open - Opened by mariecwhite over 1 year ago - 1 comment
Labels: infrastructure/shark

#12457 - Fix TorchInductor Dependency Issue

Issue - State: open - Opened by mariecwhite over 1 year ago - 1 comment
Labels: infrastructure/shark

#12457 - Fix TorchInductor Dependency Issue

Issue - State: open - Opened by mariecwhite over 1 year ago - 1 comment
Labels: infrastructure/shark

#12456 - Fix TF GPU Memory Allocation Issue

Issue - State: closed - Opened by mariecwhite over 1 year ago - 1 comment
Labels: infrastructure/shark

#12456 - Fix TF GPU Memory Allocation Issue

Issue - State: closed - Opened by mariecwhite over 1 year ago - 1 comment
Labels: infrastructure/shark

#12456 - Fix TF GPU Memory Allocation Issue

Issue - State: closed - Opened by mariecwhite over 1 year ago - 1 comment
Labels: infrastructure/shark

#12453 - Add ability to use a version of Tensorflow built from source

Issue - State: closed - Opened by mariecwhite over 1 year ago - 1 comment
Labels: infrastructure/shark