issues
search
google
/
uVkCompute
A micro Vulkan compute pipeline and a collection of benchmarking compute shaders
Apache License 2.0
224
stars
38
forks
source link
issues
Newest
Newest
Most commented
Recently updated
Oldest
Least commented
Least recently updated
Add a one workgroup argmax benchmark
#49
angelz913
closed
2 months ago
2
Add sample argmax kernel for a single subgroup
#47
qedawkins
closed
2 months ago
5
Fix subgroup_arithmetic benchmark for flexible subgroup sizes
#46
dneto0
closed
11 months ago
0
subgroup_arithmetic benchmark fails verification due to unexpected gl_SubgroupSize
#45
dneto0
closed
11 months ago
1
Fix memory benchmarks for unexpected gl_SubgroupSize
#44
dneto0
closed
11 months ago
1
Inconsistent gl_SubgroupSize across different GPUs and Vulkan versions/extensions
#43
dneto0
closed
11 months ago
3
New vector-times-matrix-transposed benchmark fails to run on Nvidia GPUs..
#42
oscarbg
opened
11 months ago
0
Add RenderDoc integration
#41
kuhar
closed
11 months ago
1
Add vector-times-matrix-transposed benchmark (V2)
#40
kuhar
closed
11 months ago
7
Add benchmark sample for vector times matrix transposed
#38
qedawkins
closed
11 months ago
3
Why large loop count will cause problem on integrated gpu?
#37
yangfengzzz
closed
1 year ago
0
Fix instance creation error on VULKAN_SDK >= 1.3.216 by opting-in to extension VK_KHR_PORTABILITY_subset
#35
a-earthperson
opened
1 year ago
2
[mmt] Prefetch LHS and RHS
#34
kuhar
closed
1 year ago
0
[matmul] Add naive mmt benchmark
#33
kuhar
closed
1 year ago
0
[memory] Copy multiple elements per thread
#32
kuhar
closed
1 year ago
0
[matmul] Tweak innerproduct i8->i32 implementation
#31
kuhar
closed
1 year ago
0
[matmul] Add basic i8->i32 matmul tiled for inner product
#30
kuhar
closed
1 year ago
0
Compile shader permutations in parallel
#29
kuhar
closed
1 year ago
0
[matmul] Clear output buffer between benchmark runs
#28
kuhar
closed
1 year ago
0
Fix i32 matmul output check
#27
kuhar
closed
1 year ago
0
Add i32 matmul benchmark variant
#26
kuhar
closed
1 year ago
0
Fix bindings mismatch with f16 matmul
#25
kuhar
closed
1 year ago
0
Bump submodules
#24
kuhar
closed
1 year ago
0
Add i8 tiled matmul benchmark
#23
kuhar
closed
1 year ago
0
[CI] Update lint CI action
#22
kuhar
closed
1 year ago
0
Clean up before introducing i8 benchmarks
#21
kuhar
closed
1 year ago
0
How to run this om Android devices?
#20
SaschaWillems
closed
2 years ago
2
fix gcc7.5 build error
#19
tpoisonooo
closed
2 years ago
1
Benchmark mad crash on Jetson Nano
#18
tpoisonooo
opened
2 years ago
1
Draft: feat(uvkc/vulkan): add validation_layer
#17
tpoisonooo
opened
2 years ago
0
Update README.md
#16
tpoisonooo
closed
2 years ago
2
Build Fail on Jetson Nano
#15
tpoisonooo
closed
2 years ago
4
fails to compile on Linux (clang and gcc)
#14
gdamjan
closed
2 years ago
2
Build Failing @ Ubuntu 18.4
#13
pure-water
opened
3 years ago
10
Use GitHub Actions for Linux and Android build
#12
antiagainst
closed
3 years ago
0
Change matmul benchmark to support both Adreno and Mali Valhall
#11
antiagainst
closed
3 years ago
0
Make matmul workgroup X, Y configurable via CMake
#10
antiagainst
closed
3 years ago
0
Add additional hints for finding spirv-as & glslc.
#9
kdub
closed
3 years ago
0
[macOS] Some small fix-ups on macOS.
#8
ergawy
closed
3 years ago
2
Separate matmul fp16 shader and add support for texture to it
#7
ThomasRaoux
closed
3 years ago
0
Add tree and atomic reduction benchmarks
#6
antiagainst
closed
3 years ago
0
Fix README typo
#5
sofiageo
closed
3 years ago
1
Add Float16 support to Matmul and mad_throughput benchmark
#4
ThomasRaoux
closed
3 years ago
0
Add a tiled conv2d benchmark
#3
antiagainst
closed
3 years ago
0
Add benchmark to measure peak compute throughput
#2
ThomasRaoux
closed
4 years ago
0
Add matmul benchmark using tiled matmul shader
#1
ThomasRaoux
closed
4 years ago
1