issues
search
eyalroz
/
cuda-api-wrappers
Thin, unified, C++-flavored wrappers for the CUDA APIs
BSD 3-Clause "New" or "Revised" License
790
stars
80
forks
source link
issues
Newest
Newest
Most commented
Recently updated
Oldest
Least commented
Least recently updated
Robustify namespace use in our macros
#542
eyalroz
closed
1 year ago
0
Support for additional vendor GPUs
#541
dmcdougall
closed
1 year ago
6
Made some changes to error handling, fixed a compilation error and sanitized macros.
#540
aryan-programmer
closed
1 year ago
2
Compilation logs are returned with the trailing nul as a a valid character
#539
eyalroz
closed
1 year ago
0
Add support for nvJitLink linking
#538
eyalroz
opened
1 year ago
0
Scheduling a memory copy on stream 0x056b5efe0: context is destroyed
#537
thomas-bouvier
closed
1 year ago
3
Calling set_endpoint by mistake in a set_endpoint_untyped method of copy_parameters_t
#535
eyalroz
closed
1 year ago
0
some example program unique_ptr's need to be float[], not float
#534
eyalroz
closed
1 year ago
0
Provide API hooking functionality like in libcuhook
#533
eyalroz
opened
1 year ago
0
Support empty nodes as typed_node's
#532
eyalroz
opened
1 year ago
0
Use CMake 3.27's fatbin generation mechanism
#531
eyalroz
opened
1 year ago
0
program_t::add_registered_globals should be able to take a container of any string-like type
#529
eyalroz
closed
1 year ago
1
mapped memory allocation currently insists on CU_MEMHOSTALLOC_DEVICEMAP
#528
eyalroz
closed
1 year ago
0
Add support for the read-only attribute for cuda::memory::host::register()
#527
eyalroz
closed
1 year ago
0
Use proper C++ random number generators in example programs
#526
eyalroz
closed
1 year ago
0
copy_parameters_t::set_single_context shouldn't take an endpoint_t
#525
eyalroz
closed
1 year ago
0
Spaces -> tabs fixes
#524
eyalroz
closed
1 year ago
0
Pass device_t's by const-ref more
#523
eyalroz
closed
1 year ago
0
Rename synch_ to sync_ on everything
#522
eyalroz
closed
1 year ago
0
Reduce boilerplate + avoid warning when overriding context for the current scope
#521
eyalroz
closed
1 year ago
0
Don't use a context-wrapper scoped-context-override when unnecessary
#520
eyalroz
closed
1 year ago
0
Support use of external memory resources
#519
eyalroz
closed
1 year ago
1
Avoid C-style casts
#518
eyalroz
closed
1 year ago
0
typed_set mistakenly accepts values of size 8
#517
eyalroz
closed
1 year ago
1
Use explicit cast and correct types for write_single_value
#516
eyalroz
closed
1 year ago
0
Missing includes
#515
codecircuit
closed
1 year ago
5
Definitions of cuda::memory::managed::allocate(device, num_bytes)
#514
codecircuit
closed
1 year ago
1
Getting an unknown CUDA error with cuMemPoolImportFromShareableHandle()
#513
eyalroz
opened
1 year ago
0
Error setting reserved virtual mem region permissions
#512
eyalroz
opened
1 year ago
0
Support creation of CUDA runtime error exceptions with fully-overriden what() messages
#511
eyalroz
closed
1 year ago
0
When NVRTC/PTX compiler return an invalid-option error - put the options in the error message
#510
eyalroz
closed
1 year ago
0
kernel compilation option --gpu-architecture can cause NVRTC_ERROR_INVALID_OPTION
#509
DanCher1403
closed
1 year ago
4
With CUDA >= 11.3, don't avoid creating a module just because CUBIN is unavailable
#508
eyalroz
closed
1 year ago
0
Run the simpleStreams example with all possible synch modes
#507
eyalroz
closed
1 year ago
0
GitHub Windows builds fail due to missing cudaProfiler.h
#506
eyalroz
closed
10 months ago
0
Re-establish example program build success on GitHub actions Windows runners
#505
eyalroz
closed
1 year ago
0
Build fails with cooperative_groups on GitHub Actions Windows runners with CUDA >= 11.7
#504
eyalroz
closed
10 months ago
1
Get rid of dependency on helper_string.h in example programs
#503
eyalroz
closed
1 year ago
0
Get rid of dependencies on helper_cuda.hpp
#502
eyalroz
closed
1 year ago
0
Add "unsupported device sync" error from NVRTC 12.1
#501
eyalroz
closed
1 year ago
0
Switch from NVVM to LTO-IR
#500
eyalroz
closed
1 year ago
0
Disable surface-related functionality with CUDA 12.x
#499
eyalroz
closed
1 year ago
0
Support num_blocks() as an alias for grid_size() in the launch config builder
#498
eyalroz
closed
1 year ago
0
Use relative include paths within the library
#497
eyalroz
closed
1 year ago
0
fatbin targets placed in "wrong" target folder
#495
eyalroz
closed
1 year ago
0
MSVC balks at set_access_mode() in vectorAddMMAP.cpp
#494
eyalroz
closed
1 year ago
0
launch config builder's overall_size() should take overall_dimension_t
#493
eyalroz
closed
1 year ago
0
Drop inclusion of cooperative_groups.h
#492
eyalroz
closed
1 year ago
0
Avoid MSVC warnings about standard-library facilities
#491
eyalroz
closed
1 year ago
0
Switch from CUDA::nvToolkitExt to CUDA::nvtx3 (and to CMake 3.25)
#490
eyalroz
closed
1 year ago
0
Previous
Next