issues
search
eyalroz
/
cuda-kat
CUDA kernel author's tools
BSD 3-Clause "New" or "Revised" License
104
stars
8
forks
source link
issues
Newest
Newest
Most commented
Recently updated
Oldest
Least commented
Least recently updated
strf dependency
#49
codecircuit
closed
4 years ago
6
Make kat::tuple compatible with std::tuple
#48
eyalroz
opened
4 years ago
0
Should we use CUDA's implicit host-device move and forward?
#47
eyalroz
opened
4 years ago
0
Add a tuple type
#46
eyalroz
closed
4 years ago
1
Explicitly use std::size_t as kat::size_t
#45
eyalroz
closed
4 years ago
0
Introduce a lane_id_t/lane_t type
#44
eyalroz
opened
4 years ago
1
Use "xxx_index" in dimensioned contexts and "xxx_id" in linearized ones
#43
eyalroz
opened
4 years ago
0
Add <algorithm> and <numeric> functions as thread-level primitives?
#42
eyalroz
opened
4 years ago
0
Use the test fixtures in the collaborative primitives in other tests
#41
eyalroz
opened
4 years ago
0
Implement kat::linear_grid::collaborative::block::at_warp_stride()
#40
eyalroz
opened
4 years ago
0
Inappropriate return type and incorrect calculation in grid_info.cuh
#39
eyalroz
closed
4 years ago
0
Specialize functions with many reads/writes for sub-4-byte element types
#38
eyalroz
opened
4 years ago
0
Should we drop support for CUDA 8.x?
#37
eyalroz
opened
4 years ago
0
Add support for lane masks in `collaboration::warp::` methods
#36
eyalroz
closed
4 years ago
0
Add wrappers (and builtins?) for more PTX instructions
#35
eyalroz
opened
4 years ago
0
Distinguish between PTX builtins and SASS builtins
#34
eyalroz
opened
4 years ago
0
Consider using sized integer types for the builtins
#33
eyalroz
opened
4 years ago
0
Rearrange math builtins
#32
eyalroz
opened
4 years ago
0
Development
#31
eyalroz
closed
4 years ago
0
A device-side ostream with `printf()` as its back-end
#30
eyalroz
closed
4 years ago
0
Periodic merge of development branch work
#29
eyalroz
closed
5 years ago
0
Add shuffle tests with non-power-of-2 sizes
#28
eyalroz
opened
5 years ago
0
Add the average-computing functions to builtins.cuh
#27
eyalroz
closed
5 years ago
0
Add wrappers for mathmatical type conversion functionality
#26
eyalroz
opened
5 years ago
0
Merge development branch work
#25
eyalroz
closed
5 years ago
0
Cover all functionality with basic unit tests
#24
eyalroz
opened
5 years ago
1
Bring some order to `_safe` vs `_unsafe`, constexpr vs non-constexpr math functions
#23
eyalroz
closed
5 years ago
0
Cull some of the printing.cuh code
#22
eyalroz
closed
4 years ago
1
Drop __fd__ , __fhd__ etc. in favor of something more palletable
#21
eyalroz
closed
4 years ago
1
Use functions instead of macros in on_device/printing.cuh
#20
eyalroz
closed
4 years ago
1
Implement the non-linear-grid variants of all grid_info functions
#19
eyalroz
closed
5 years ago
0
Add missing <algorithm> functions to the warp and block sequence-ops
#18
eyalroz
opened
5 years ago
0
Use spans as parameters where relevant
#17
eyalroz
opened
5 years ago
0
Semantics of atomic::increment() and atomic::decrement() wrong
#16
eyalroz
closed
5 years ago
0
Pass arguments by value to atomic functions
#15
eyalroz
closed
5 years ago
0
Merge development work
#14
eyalroz
closed
5 years ago
0
Place (essentially) all code within the kat namespace
#13
eyalroz
closed
5 years ago
0
Adapt include guards to this library + drop their use of block comment
#12
eyalroz
closed
5 years ago
0
Support string.h functionality on the device side
#11
eyalroz
closed
5 years ago
0
Unaligned access is missing the align_down() function
#10
eyalroz
opened
5 years ago
0
Support atomicCAS() for all types
#9
eyalroz
closed
5 years ago
0
Merge development work
#8
eyalroz
closed
5 years ago
0
Simplify some of the numeric functions in math.cuh and constexpr.cuh
#7
eyalroz
closed
5 years ago
0
Add a device-side-enabled version of gsl::span or std::span
#6
eyalroz
closed
4 years ago
0
Device-side function for pretty-printing (part of a) column
#5
eyalroz
opened
5 years ago
0
Untangle the mess in primitives/
#4
eyalroz
closed
4 years ago
1
Block-level conjuction and disjunction
#3
eyalroz
opened
5 years ago
0
Add testing instrumentation
#2
eyalroz
closed
5 years ago
0
Doxygen-comment the code
#1
eyalroz
opened
5 years ago
0
Previous