JuliaGPU/CUDA.jl issues and pull requests

#1690 - CUDA test failed

Issue - State: closed - Opened by lf28 almost 2 years ago - 1 comment
Labels: bug, needs information

#1690 - CUDA test failed

Issue - State: closed - Opened by lf28 almost 2 years ago - 1 comment
Labels: bug, needs information

#1690 - CUDA test failed

Issue - State: closed - Opened by lf28 almost 2 years ago - 1 comment
Labels: bug, needs information

#1690 - CUDA test failed

Issue - State: closed - Opened by lf28 almost 2 years ago - 1 comment
Labels: bug, needs information

#1687 - Switch secrets management to cryptic.

Pull Request - State: open - Opened by maleadt almost 2 years ago
Labels: ci

#1684 - Precompilation spends significant time in parser

Issue - State: open - Opened by maleadt almost 2 years ago - 2 comments
Labels: performance

#1678 - view(data, idx) boundschecking is disproportionately expensive

Issue - State: closed - Opened by maleadt almost 2 years ago - 2 comments
Labels: performance

#1674 - ReinterpretArray indexing causes dynamic invocation on 1.9

Issue - State: open - Opened by pxl-th about 2 years ago - 3 comments
Labels: bug, cuda kernels

#1644 - Add scoped atomic_thread_fence

Pull Request - State: open - Opened by vchuravy about 2 years ago - 4 comments
Labels: enhancement, cuda kernels

#1633 - device-side rand() are not random between successive kernel launches

Issue - State: closed - Opened by zpeng2 about 2 years ago - 5 comments
Labels: enhancement, good first issue, cuda kernels

#1632 - [CUSPARSE] Add CuMatrix * CuSparseMatrix products

Pull Request - State: closed - Opened by amontoison about 2 years ago - 3 comments
Labels: enhancement, cuda libraries

#1632 - [CUSPARSE] Add CuMatrix * CuSparseMatrix products

Pull Request - State: closed - Opened by amontoison about 2 years ago - 3 comments
Labels: enhancement, cuda libraries

#1632 - [CUSPARSE] Add CuMatrix * CuSparseMatrix products

Pull Request - State: closed - Opened by amontoison about 2 years ago - 3 comments
Labels: enhancement, cuda libraries

#1623 - Update to CUTENSOR 1.6 and cuQuantum 22.07

Pull Request - State: closed - Opened by maleadt about 2 years ago - 4 comments
Labels: installation

#1607 - Option for deterministic sparse matrix vector multiplication

Issue - State: open - Opened by renatobellotti about 2 years ago - 2 comments
Labels: enhancement, good first issue, cuda libraries

#1601 - Failed to compile PTX code when using NSight on Win11

Issue - State: closed - Opened by jamblejoe about 2 years ago - 4 comments
Labels: bug, needs information

#1579 - device_reset! does not seem to work anymore

Issue - State: closed - Opened by maleadt over 2 years ago - 2 comments
Labels: bug, upstream

#1572 - Add eigen for general purpose computation of eigenvectors/eigenvalues

Issue - State: closed - Opened by GVigne over 2 years ago - 1 comment
Labels: enhancement, help wanted, cuda libraries

#1563 - limit csc/csr/bsr sparse conversion index to be cint & fix a few conversion bugs

Pull Request - State: open - Opened by Roger-luo over 2 years ago
Labels: cuda array

#1562 - Cannot reclaim GPU Memory; CUDA.reclaim()

Issue - State: closed - Opened by jackn11 over 2 years ago - 2 comments
Labels: bug

#1540 - Memory allocation becomes very slow when reserved bytes is large

Issue - State: closed - Opened by DrChainsaw over 2 years ago - 3 comments
Labels: bug

#1522 - GC integration: multiple threads fail to free each others objects, lead to OOM

Issue - State: open - Opened by lmh91 over 2 years ago - 13 comments
Labels: bug

#1520 - RFC: Support Adapt.AbstractGPUDevice

Pull Request - State: closed - Opened by oschulz over 2 years ago - 1 comment

#1511 - CUSPARSE: CSR GEMM should dispatch to mm!/cusparseSpGEMM

Issue - State: open - Opened by mashu over 2 years ago - 3 comments
Labels: enhancement, good first issue, cuda libraries

#1492 - Support for more complex associative scans

Issue - State: open - Opened by nathanaelbosch over 2 years ago - 4 comments
Labels: enhancement, good first issue

#1485 - NVTX: consider using Start/End for ranges

Issue - State: closed - Opened by maleadt over 2 years ago - 3 comments
Labels: enhancement, good first issue, cuda libraries

#1472 - Indexing a view uses scalar indexing

Issue - State: closed - Opened by CarloLucibello over 2 years ago - 2 comments
Labels: enhancement, cuda array

#1472 - Indexing a view uses scalar indexing

Issue - State: closed - Opened by CarloLucibello over 2 years ago - 2 comments
Labels: enhancement, cuda array

#1467 - CUSPARSE: Better error msg for unsupported sparse mm

Pull Request - State: open - Opened by Roger-luo over 2 years ago - 5 comments
Labels: cuda libraries

#1426 - WMMA Float64

Pull Request - State: open - Opened by carstenbauer over 2 years ago - 2 comments

#1425 - WMMA BFloat16 (BF16)

Pull Request - State: open - Opened by carstenbauer over 2 years ago - 4 comments

#1419 - WMMA TensorFloat32 (TF32)

Pull Request - State: open - Opened by carstenbauer over 2 years ago - 24 comments
Labels: enhancement, cuda kernels, needs changes

#1406 - CUSPARSE does not support reductions

Issue - State: closed - Opened by CarloLucibello over 2 years ago - 3 comments
Labels: enhancement

#1393 - Support ordering argument for atomics

Pull Request - State: open - Opened by tkf over 2 years ago - 2 comments

#1377 - Cross-device copy of wrapped arrays fails

Issue - State: closed - Opened by glwagner almost 3 years ago - 2 comments
Labels: bug, cuda array

#1371 - UndefVarError: parameters not defined on Windows10

Issue - State: closed - Opened by prog900 almost 3 years ago - 4 comments
Labels: bug

#1363 - mul! does not dispatch on Adjoint

Issue - State: closed - Opened by oashour almost 3 years ago - 1 comment
Labels: bug, good first issue, cuda array

#1322 - Add explicit strides to CuArray and CuDeviceArray.

Pull Request - State: open - Opened by maleadt almost 3 years ago
Labels: cuda array

#1235 - [POC] Support for cuFile.

Pull Request - State: open - Opened by maleadt about 3 years ago - 2 comments

#1215 - improve sparse matrix conversions

Pull Request - State: open - Opened by Roger-luo about 3 years ago - 1 comment
Labels: enhancement, cuda libraries

#1203 - Try running under compute-sanitizer with more than one thread.

Pull Request - State: closed - Opened by maleadt about 3 years ago
Labels: ci

#1197 - 2-arg `show`

Pull Request - State: open - Opened by mcabbott about 3 years ago - 1 comment

#1184 - Calling similar(...) on AbstractCuSparseArray with dims

Pull Request - State: open - Opened by birkmichael about 3 years ago - 11 comments
Labels: cuda array, needs tests, bugfix

#1140 - Add a hostcall interface

Pull Request - State: open - Opened by maleadt about 3 years ago - 6 comments
Labels: enhancement, cuda kernels

#1126 - Multidimensional `reverse`

Issue - State: closed - Opened by piever about 3 years ago - 5 comments
Labels: enhancement, good first issue

#1118 - Fixed reducedim operations for CuQRPackedQ

Pull Request - State: open - Opened by rkube about 3 years ago - 4 comments
Labels: enhancement, cuda array, needs tests, needs changes

#1092 - Register BFloat16

Pull Request - State: open - Opened by DhairyaLGandhi over 3 years ago - 5 comments
Labels: cuda libraries

#1066 - Latency improvements.

Pull Request - State: open - Opened by maleadt over 3 years ago

#1053 - Legacy cuIpc* APIs incompatible with stream-ordered allocator

Issue - State: closed - Opened by christophernhill over 3 years ago - 6 comments
Labels: bug, upstream

#1053 - Legacy cuIpc* APIs incompatible with stream-ordered allocator

Issue - State: closed - Opened by christophernhill over 3 years ago - 6 comments
Labels: bug, upstream

#974 - Int128 support

Issue - State: open - Opened by maleadt over 3 years ago - 2 comments
Labels: enhancement, cuda kernels

#937 - sort! and partialsort! are considerably slower than CPU versions

Issue - State: closed - Opened by rafaqz over 3 years ago - 9 comments
Labels: performance

#918 - updating docs to include matrix-vector multiply example

Pull Request - State: open - Opened by akashkgarg over 3 years ago - 5 comments
Labels: documentation

#912 - Add support for strided cufft

Pull Request - State: open - Opened by N5N3 over 3 years ago - 11 comments
Labels: cuda libraries

#862 - Profiling Julia with Nsight Systems on Windows results in blank window

Issue - State: closed - Opened by anj00 over 3 years ago - 2 comments
Labels: bug, upstream

#850 - Support for `sm_80` `cp.async`: asynchronous on-device copies

Issue - State: closed - Opened by JesseLu over 3 years ago - 14 comments
Labels: enhancement, good first issue, cuda kernels, important

#814 - Added support for kron

Pull Request - State: open - Opened by JATippit over 3 years ago - 8 comments
Labels: enhancement, cuda array, needs changes

#798 - Implement sync_threads using an unaligned barrier.

Pull Request - State: open - Opened by maleadt over 3 years ago - 1 comment
Labels: cuda kernels, speculative

#793 - NVPTX i128 support broken on LLVM 11 / Julia 1.6

Issue - State: closed - Opened by ali-ramadhan over 3 years ago - 6 comments
Labels: bug, upstream

#764 - Improve 'VS C++ redistributable' error message

Issue - State: closed - Opened by anandijain over 3 years ago - 2 comments
Labels: documentation, good first issue

#764 - Improve 'VS C++ redistributable' error message

Issue - State: closed - Opened by anandijain over 3 years ago - 2 comments
Labels: documentation, good first issue

#764 - Improve 'VS C++ redistributable' error message

Issue - State: closed - Opened by anandijain over 3 years ago - 2 comments
Labels: documentation, good first issue

#764 - Improve 'VS C++ redistributable' error message

Issue - State: closed - Opened by anandijain over 3 years ago - 2 comments
Labels: documentation, good first issue

#763 - Track array ownership to avoid illegal memory accesses

Issue - State: closed - Opened by marius311 over 3 years ago - 5 comments
Labels: enhancement

#666 - add custom sum tutorial

Pull Request - State: open - Opened by jw3126 almost 4 years ago - 5 comments
Labels: documentation

#567 - Add GPU to CPU RPC mechanism

Pull Request - State: open - Opened by ajuvercr almost 4 years ago - 2 comments
Labels: enhancement, cuda kernels

#556 - Missing dispatch for indexing of reshaped arrays

Issue - State: closed - Opened by ChrisRackauckas almost 4 years ago - 7 comments
Labels: bug

#552 - Constant memory support

Pull Request - State: open - Opened by S-D-R about 4 years ago - 9 comments
Labels: enhancement, cuda kernels

#544 - TagBot trigger issue

Issue - State: closed - Opened by JuliaTagBot about 4 years ago - 178 comments

#538 - Performance regression on matrix multiplication between CUDA.jl 1.3.3 and 2.1.0/master

Issue - State: closed - Opened by jeremiedb about 4 years ago - 4 comments
Labels: cuda libraries, performance, regression

#538 - Performance regression on matrix multiplication between CUDA.jl 1.3.3 and 2.1.0/master

Issue - State: closed - Opened by jeremiedb about 4 years ago - 4 comments
Labels: cuda libraries, performance, regression

#538 - Performance regression on matrix multiplication between CUDA.jl 1.3.3 and 2.1.0/master

Issue - State: closed - Opened by jeremiedb about 4 years ago - 4 comments
Labels: cuda libraries, performance, regression

#538 - Performance regression on matrix multiplication between CUDA.jl 1.3.3 and 2.1.0/master

Issue - State: closed - Opened by jeremiedb about 4 years ago - 4 comments
Labels: cuda libraries, performance, regression

#491 - copyto!() between cpu and gpu with subarrays

Issue - State: closed - Opened by btmit about 4 years ago - 4 comments

#440 - Hostcall

Issue - State: open - Opened by maleadt about 4 years ago - 1 comment
Labels: enhancement, cuda kernels

#440 - Hostcall

Issue - State: open - Opened by maleadt about 4 years ago - 1 comment
Labels: enhancement, cuda kernels

#440 - Hostcall

Issue - State: open - Opened by maleadt about 4 years ago - 1 comment
Labels: enhancement, cuda kernels

#287 - Prefix scan using warp primitives and fallback for non primitive types

Pull Request - State: open - Opened by Ellipse0934 over 4 years ago - 16 comments
Labels: enhancement, cuda array, performance

#267 - Make Ref pass by-reference

Issue - State: closed - Opened by maleadt over 4 years ago - 5 comments
Labels: enhancement, cuda kernels, speculative

#129 - Exception when using CuArrays with Flux

Issue - State: closed - Opened by NiklasGustafsson over 5 years ago - 6 comments
Labels: bug, cuda array

#129 - Exception when using CuArrays with Flux

Issue - State: closed - Opened by NiklasGustafsson over 5 years ago - 6 comments
Labels: bug, cuda array

#129 - Exception when using CuArrays with Flux

Issue - State: closed - Opened by NiklasGustafsson over 5 years ago - 6 comments
Labels: bug, cuda array

#129 - Exception when using CuArrays with Flux

Issue - State: closed - Opened by NiklasGustafsson over 5 years ago - 6 comments
Labels: bug, cuda array

#127 - Element-wise conversion to Duals

Issue - State: closed - Opened by ChrisRackauckas over 5 years ago - 2 comments
Labels: bug, cuda array

#126 - copy!(dst, src) and copyto!(dst, src) are significantly slower and allocate more memory than copyto!(dest, do, src, so[, N])

Issue - State: closed - Opened by colinxs over 5 years ago - 1 comment
Labels: cuda array, performance

GitHub / JuliaGPU/CUDA.jl issues and pull requests