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
Merge the `by_driver_api_module` and `by_runtime_api_module` example program directories
#489
eyalroz
closed
1 year ago
0
cuda::memory::host::allocate should return a region_t
#488
eyalroz
closed
1 year ago
0
Better handling of negative device id's
#487
eyalroz
closed
1 year ago
0
Adopting code to `cuda::kernel::wrap` API changes
#486
fvanmaele
closed
1 year ago
2
Move physical allocation namespace out of `virtual_` and up into `memory`
#485
eyalroz
closed
1 year ago
0
Support setting kernel block cluster dimensions
#484
eyalroz
closed
7 months ago
1
Support converting regions into spans - at least explicitly
#483
eyalroz
closed
1 year ago
0
cuda::span improvements
#482
eyalroz
closed
1 year ago
0
stream_t::enqueue::free() is not const!
#481
eyalroz
closed
1 year ago
0
Try to build an executable for each C++ version after 11 supported by the compiler
#480
eyalroz
closed
1 year ago
0
The IPC pointer import mechanism is unsafe, inconsistent with rest of lib
#479
eyalroz
closed
1 year ago
0
Fixes span usage for both std::span and cuda::span.
#478
raplonu
closed
1 year ago
7
By default, create CUDA C++ programs
#477
eyalroz
closed
1 year ago
0
device::get() marked noexcept
#475
eyalroz
closed
1 year ago
1
devices begin() can throw - don't have it as noexcept
#474
eyalroz
closed
1 year ago
1
Avoid the fancy-shmancy heap allocation for stream callbacks
#473
eyalroz
closed
1 year ago
1
stream_t::enqueue::memset() should support regions
#472
eyalroz
closed
1 year ago
0
Make more comparison operators constexpr and noexcept
#471
eyalroz
closed
1 year ago
0
Support CUDA 12.x texture, surface and tensor "objects"
#470
eyalroz
opened
1 year ago
0
Need a convenient function for getting a kernel from a __global__ function
#469
eyalroz
closed
1 year ago
1
Have memory::type_of treat non-CUDA-allocated memory as a non-error
#468
eyalroz
closed
1 year ago
0
Array-to-raw-mem copy function must either get a context or determine it
#467
eyalroz
closed
1 year ago
0
Make copy_parameters_t user-facing and beef it up
#466
eyalroz
closed
1 year ago
0
update build cmd in readme
#465
qy3u
closed
1 year ago
1
Unable to build examples with instructions in README.md
#464
qy3u
closed
1 year ago
0
Allow obtaining a pointer's device and context without wrapping it
#463
eyalroz
closed
1 year ago
0
Support obtaining a pointer's device in CUDA 9.1 and earlier
#462
eyalroz
closed
1 year ago
0
Reconsider accepting pointers without contexts
#461
eyalroz
opened
1 year ago
0
Support colcon build
#460
wep21
closed
1 year ago
4
cuda::kernel::get() should not take an arbitrary context
#459
eyalroz
closed
1 year ago
0
Support execution graph capture via stream_t's
#458
eyalroz
closed
5 months ago
3
Add support for user objects (in graphs etc.)
#457
eyalroz
opened
1 year ago
0
Support CUDA 12.0 graph API changes
#456
eyalroz
opened
1 year ago
0
Missing definition for allocate_managed() in src/cuda/api/context.hpp
#455
kingcrimsontianyu
closed
1 year ago
1
Support scheduling actual batches of "batch memory operations" on a stream
#454
eyalroz
opened
1 year ago
0
Set flags for the flush-remote-writes stream operation
#453
eyalroz
closed
1 year ago
0
Support the memory barrier "stream memory operation"
#452
eyalroz
closed
1 year ago
0
Stream callbacks should use CUDA_CB, not CUDART_CB
#451
eyalroz
closed
1 year ago
0
Memory leak: Allocation without release in cuda::memory::virtual::set_access_mode
#450
eyalroz
closed
1 year ago
0
`apriori_compiled_kernel_t::get_attribute` should be marked inline
#449
jhol
closed
1 year ago
6
cuda::profiling::mark::range_start and range_end call create_attributions the wrong way
#448
eyalroz
closed
1 year ago
0
Accept "execution affinity" info when creating a context
#447
eyalroz
opened
1 year ago
0
Support obtaining unique stream and context IDs
#446
eyalroz
opened
1 year ago
0
Support use of external semaphores
#445
eyalroz
opened
1 year ago
0
Should we expose pointer-to-const versions of CUDA driver handles?
#444
eyalroz
opened
1 year ago
0
Member initialisation order error.
#443
j-horner-c4x
closed
1 year ago
3
Lingering `link::input_type_t` in `cuda/api/link.hpp`
#442
jhol
closed
1 year ago
2
kernel_t::context() should use wrap() to be noexcept
#441
eyalroz
closed
1 year ago
0
Should wrapper classes throw when failing on destruction?
#440
eyalroz
opened
1 year ago
1
Avoid move'ing compilation_options_t unless necessary
#439
eyalroz
closed
1 year ago
0
Previous
Next