JuliaGPU/AMDGPU.jl

Issues

• Support for multi-GPU nodes broken in 0.7

  #527 opened a year ago
  3

• Support for ROCm 5.7.1

  #522 opened a year ago
  9

• Multiple workers hang test suite on Julia 1.10

  #521 opened a year ago
  0

• Navi 3 causes `malloc(): unsorted double linked list corrupted`

  #518 opened a year ago
  1

• Investigate GPUArrays tests suite error

  #515 opened a year ago
  1

• Investigate using MIOpen's immediate mode for conv algorithm search

  #507 opened a year ago
  0

• `sincos` intrinsic is broken with GPUCompiler 0.24

  #502 opened a year ago
  1

• Synchronization in REPL

  #500 opened a year ago
  2

• Precompilation fails

  #499 opened a year ago
  2

• [Question] AI accelerators

  #483 opened a year ago
  0

• AMDGPU.jl on rolling release distros (Arch): Libraries unavailable

  #478 opened a year ago
  3

• AMDGPU on windows

  #465 opened a year ago
  10

• Unexpected error: ccall requires compiler when using QR

  #461 opened a year ago
  5

• hipErrorSharedObjectInitFailed

  #451 opened a year ago
  2

• Setting HIP_VISIBLE_DEVICES to an invalid ID fails in an unhelpful way

  #450 opened a year ago
  0

• Segmentation fault on `hipStreamDestroy`

  #449 opened a year ago
  4

• Failure calling upon calling Enzyme `autodiff_deferred`

  #444 opened a year ago
  22

• Update ROCm artifacts to newer version

  #440 opened a year ago
  2

• Cannot `unsafe_wrap` a device array if `lock=false`

  #436 opened a year ago
  1

• WARNING: could not import AMDGPU.device_libs_path into Compiler

  #434 opened a year ago
  0

• Default to `Int32` indices as much as possible

  #432 opened a year ago
  0

• `rand`: MethodError: no method matching `rng_native_52(::GPUArrays.RNG)`

  #429 opened a year ago
  0

• @ROCDynamicLocalArray: add support for dynamic eltype and expressions for dims

  #428 opened a year ago
  3

• Switching to device ≠ 1 hangs on multi-GPU node

  #425 opened a year ago
  2

• First install with `JULIA_AMDGPU_DISABLE_ARTIFACTS` leads to broken config

  #424 opened a year ago
  2

• Make `unsafe_copy3d!` TLS compatible

  #421 opened a year ago
  1

• rocBLAS synchronization issue?

  #418 opened a year ago
  5

• KernelAbstraction printing test fails

  #406 opened 2 years ago
  0

• Optimize `wait!` for HSA kernel launches

  #405 opened a year ago
  1

• Update HIP errors codes

  #404 opened a year ago
  1

• Detect hardware queue limit and use to limit queue pool size

  #403 opened a year ago
  1

• Fails to load on AMD Ryzen 9 7950X integrated graphics

  #401 opened a year ago
  6

• Test failures locally on 1.9.0-beta4 -- Radeon 6800XT

  #400 opened a year ago
  3

• Test fail locally with `AssertionError: AMDGPU.Runtime.LOGGING_STATIC_ENABLED`

  #399 opened 2 years ago
  1

• Update LibHSA wrappers to include SVM APIs

  #397 opened 2 years ago
  0

• AMDGPU Stable Website on Gitlab and JuliaHub Not Updated

  #396 opened 2 years ago
  2

• `@assert` errors on ROCArrays if device ≠ 1

  #388 opened a year ago
  1

• `AMDGPU.launch_configuration` fails on device ≠ 1

  #387 opened a year ago
  5

• HSA memory fault using `AMDGPU.rand()` on device ≠ 1

  #386 opened a year ago
  1

• HSA memory fault upon switching from default device on multi-GPU node

  #385 opened 2 years ago
  1

• rocBLAS: Remove old hand-wrapped code

  #384 opened 2 years ago
  0

• Support for rand from Julia Base on device code

  #378 opened a year ago
  2

• Guidance on randn hanging on MI250X

  #377 opened 2 years ago
  3

• AMDGPU.jl master is broken on Julia 1.7

  #372 opened a year ago
  3

• AMDGPU.jl doesn't seem to work with 7900 series GPUs

  #371 opened a year ago
  22

• Logs may be unpaired when errors occur in runtime

  #360 opened 2 years ago
  0

• `--check-bounds=no` is broken on Julia 1.9.0-beta3

  #354 opened 2 years ago
  19

• Free `KernelState` in finalizer

  #352 opened 2 years ago
  2

• Don't disable ROCm external library type definitions when non-functional

  #350 opened a year ago
  1

• Add function or macro for AMDGPU.jl equivalent to CUDA.CuDynamicSharedArray (and CUDA.CuStaticSharedArray)

  #347 opened 2 years ago
  1