Owner

NVIDIA

2 indexed repositories · View on GitHub

Megatron-LM
Ongoing research training transformer models at scale

Python · 17897 stars
cccl
CUDA Core Compute Libraries

C++ · 2513 stars

Open All

Consider support for segmented reductions and sorts specified by count-value representation Open

thrust

NVIDIA/cccl#676 · 3 comments · 0 reactions · 1 assignee · Indexed Sep 16, 2026
equal_range could use specialization Open

thrust

NVIDIA/cccl#677 · 0 comments · 0 reactions · 0 assignees · Indexed Sep 16, 2026
tagged_iterator does not propagate its tag to its nested pointer & reference types Open

thrust

NVIDIA/cccl#678 · 0 comments · 0 reactions · 0 assignees · Indexed Sep 16, 2026
Use testing iterator wrappers to ensure that algorithms support their stated requirements Open

thrust

NVIDIA/cccl#679 · 2 comments · 0 reactions · 0 assignees · Indexed Sep 16, 2026
add missing easy algorithms Open

thrust

NVIDIA/cccl#680 · 1 comment · 0 reactions · 0 assignees · Indexed Sep 16, 2026
consider adding algorithm for repeating values a specified number of times Open

thrust

NVIDIA/cccl#681 · 1 comment · 0 reactions · 0 assignees · Indexed Sep 16, 2026
add transpose() example Open

thrust

NVIDIA/cccl#682 · 0 comments · 1 reaction · 0 assignees · Indexed Sep 16, 2026
Destructors must not throw exception Open

thrust

NVIDIA/cccl#683 · 0 comments · 1 reaction · 0 assignees · Indexed Sep 16, 2026
Implement thrust::includes Open

thrust

NVIDIA/cccl#684 · 0 comments · 0 reactions · 0 assignees · Indexed Sep 16, 2026
Add missing hard algorithms Open

thrust

NVIDIA/cccl#685 · 3 comments · 0 reactions · 0 assignees · Indexed Sep 16, 2026
optimize sorting tuples of arithmetic types Open

thrust

NVIDIA/cccl#686 · 0 comments · 0 reactions · 0 assignees · Indexed Sep 16, 2026
add nth_element Open

thrust

NVIDIA/cccl#687 · 0 comments · 0 reactions · 0 assignees · Indexed Sep 16, 2026
Consider complementing the set operations with relational algebra algorithms Open

thrust

NVIDIA/cccl#689 · 0 comments · 0 reactions · 0 assignees · Indexed Sep 16, 2026
thrust::reference needs additional operator overloads Open

thrust

NVIDIA/cccl#690 · 0 comments · 0 reactions · 0 assignees · Indexed Sep 16, 2026
Consider swallowing errors in deallocation functions Open

thrust

NVIDIA/cccl#696 · 2 comments · 0 reactions · 0 assignees · Indexed Sep 16, 2026
inclusive_scan produces the wrong result for char types Open

nvbug thrust

NVIDIA/cccl#698 · 5 comments · 0 reactions · 1 assignee · Indexed Sep 16, 2026
Return value of thrust::minmax_element differs from STL's minmax_element Open

thrust

NVIDIA/cccl#699 · 2 comments · 0 reactions · 1 assignee · Indexed Sep 16, 2026
Potential speedup for count_if and suggested reduce_if Open

thrust

NVIDIA/cccl#700 · 0 comments · 1 reaction · 0 assignees · Indexed Sep 16, 2026
generalize get_iterator_value() to respect execution policy Open

thrust

NVIDIA/cccl#701 · 8 comments · 0 reactions · 1 assignee · Indexed Sep 16, 2026
Don't call the predicate twice in remove_if Open

thrust

NVIDIA/cccl#702 · 7 comments · 0 reactions · 1 assignee · Indexed Sep 16, 2026
Update benchmark results in CUB documentation Open

cub

NVIDIA/cccl#849 · 1 comment · 0 reactions · 1 assignee · Indexed Sep 16, 2026
Zip iterators with copy across GPUs gives error Open

thrust

NVIDIA/cccl#703 · 0 comments · 0 reactions · 0 assignees · Indexed Sep 16, 2026
thrust::system::cuda::par stream documentation incorrect Open

nvbug thrust

NVIDIA/cccl#704 · 9 comments · 2 reactions · 1 assignee · Indexed Sep 16, 2026
`iterator_category_to_system` returns `host_system_tag` for `input_device_iterator_tag` instead of `device_system_tag` Open

nvbug thrust

NVIDIA/cccl#705 · 8 comments · 0 reactions · 1 assignee · Indexed Sep 16, 2026
Document and test WarpScan::Scan's treatment of `init` argument Open

cub

NVIDIA/cccl#850 · 2 comments · 0 reactions · 1 assignee · Indexed Sep 16, 2026
Explicitly document synchronization requirements in Warp-level APIs Open

cub

NVIDIA/cccl#853 · 1 comment · 2 reactions · 0 assignees · Indexed Sep 16, 2026
Null references used for type dispatch Open

thrust

NVIDIA/cccl#709 · 1 comment · 0 reactions · 0 assignees · Indexed Sep 16, 2026
__shfl_sync instructions may have wrong member mask Open

cub

NVIDIA/cccl#854 · 4 comments · 1 reaction · 0 assignees · Indexed Sep 16, 2026
Implementation of load balance algorithlms Open

thrust

NVIDIA/cccl#711 · 0 comments · 0 reactions · 0 assignees · Indexed Sep 16, 2026
Parameterize `thrust::*_vector` Resizing Open

thrust

NVIDIA/cccl#714 · 2 comments · 0 reactions · 0 assignees · Indexed Sep 16, 2026
UBSan complains about device_vector::operator[] calls from host Open

thrust

NVIDIA/cccl#715 · 1 comment · 0 reactions · 0 assignees · Indexed Sep 16, 2026
Provide a standard base class to inherit from for defining custom allocators Open

thrust

NVIDIA/cccl#716 · 4 comments · 5 reactions · 0 assignees · Indexed Sep 16, 2026
`thrust::all_of` is slower than a naive reduction Open

thrust

NVIDIA/cccl#720 · 9 comments · 1 reaction · 1 assignee · Indexed Sep 16, 2026
cub block reductions fail to compile correctly with nvrtc for certain block sizes Open

cub

NVIDIA/cccl#856 · 6 comments · 0 reactions · 1 assignee · Indexed Sep 16, 2026
Compilation failure in uninitialized_{fill,copy}{_n} for certain classes Open

thrust

NVIDIA/cccl#723 · 0 comments · 0 reactions · 0 assignees · Indexed Sep 16, 2026
Examples can (and should) be exemplary C++ Open

good first issue thrust

NVIDIA/cccl#724 · 5 comments · 4 reactions · 0 assignees · Indexed Sep 16, 2026
[Bug?] WarpReduce: Unexpected results with logical warp size < 32 Open

cub

NVIDIA/cccl#858 · 2 comments · 0 reactions · 0 assignees · Indexed Sep 16, 2026
Misaligned access in `thrust::mr::new_delete_resource` Open

thrust

NVIDIA/cccl#726 · 0 comments · 0 reactions · 0 assignees · Indexed Sep 16, 2026
[BUG] `std::string` + `thrust::tuple` on host fails to compile Open

nvbug thrust

NVIDIA/cccl#728 · 3 comments · 0 reactions · 0 assignees · Indexed Sep 16, 2026
Converting constructors and assignment operators on Thrust vector types are inconsistent and dangerous. Open

thrust

NVIDIA/cccl#729 · 3 comments · 0 reactions · 0 assignees · Indexed Sep 16, 2026
Deprecate / Remove old allocators Open

thrust

NVIDIA/cccl#730 · 0 comments · 1 reaction · 0 assignees · Indexed Sep 16, 2026
Documentation Request: Allocators / Iterators Open

thrust

NVIDIA/cccl#731 · 0 comments · 0 reactions · 0 assignees · Indexed Sep 16, 2026
Add test coverage for per-thread default streams Open

thrust

NVIDIA/cccl#732 · 5 comments · 0 reactions · 1 assignee · Indexed Sep 16, 2026
[Documentation] Behavior of block scans with initial value param Open

cub

NVIDIA/cccl#859 · 5 comments · 0 reactions · 1 assignee · Indexed Sep 16, 2026
Intermittent test failures in `test.allocator` (MSVC, OMP host) Open

thrust

NVIDIA/cccl#737 · 0 comments · 0 reactions · 1 assignee · Indexed Sep 16, 2026
Fix assignments to uninitialized memory Open

cub

NVIDIA/cccl#860 · 3 comments · 0 reactions · 1 assignee · Indexed Sep 16, 2026
FileCheck validation fails for some examples when building with OMP/TBB host systems. Open

thrust

NVIDIA/cccl#740 · 1 comment · 0 reactions · 1 assignee · Indexed Sep 16, 2026
Unclear documentation Open

cub

NVIDIA/cccl#861 · 0 comments · 0 reactions · 0 assignees · Indexed Sep 16, 2026
DeviceHistogram fails to compile trying to use atomicAdd on unsupported types Open

cub

NVIDIA/cccl#862 · 8 comments · 0 reactions · 0 assignees · Indexed Sep 16, 2026
[NV 2716223] `adjacent_difference` generates a warning (warning: calling a __host__ function from a __host__ __device__ function) with the OpenMP backend Open

nvbug thrust

NVIDIA/cccl#743 · 3 comments · 0 reactions · 0 assignees · Indexed Sep 16, 2026

Showing the newest 100

This page lists what was indexed most recently. Advanced filter has the whole inventory, narrowed by language, difficulty and how long a task takes.

Open advanced filter

NVIDIA

Showing the newest 100

Get new issues in your inbox