|
CUDA | Avoid divide by zero in kernel using assume()
|
|
10
|
290
|
October 7, 2025
|
|
Slow matrix multiplication in CUBLAS.gemm_strided_batched with ComplexF64
|
|
1
|
174
|
October 7, 2025
|
|
KernelAbstractions + Enzyme - how to do GPU-side autodiff?
|
|
1
|
156
|
September 25, 2025
|
|
RCCL wrapping
|
|
4
|
190
|
September 20, 2025
|
|
CUDA.jl: Unexpected `mapreduce` error: threads per block exceed GPU limit (640 > 512
|
|
9
|
312
|
September 18, 2025
|
|
CUDA.jl: Warning about loading library from system path
|
|
4
|
210
|
August 30, 2025
|
|
cuSOLVER: two calls to cusolverDnDgesvdj_bufferSize, one via Juila, the other via CUDA yield (very) different results
|
|
2
|
96
|
August 22, 2025
|
|
What is the correct way to use multiple GPUs in Slurm cluster?
|
|
0
|
324
|
August 20, 2025
|
|
Trying to parallelize using CUSOLVERRF.jl with @threads
|
|
7
|
200
|
August 19, 2025
|
|
Metal.jl does not speed up FFT
|
|
8
|
2175
|
August 13, 2025
|
|
UndefVarError: cuda_version in Google Colab with CUDA.jl
|
|
2
|
78
|
August 12, 2025
|
|
Using getrf_batched to find matrix inverses
|
|
2
|
106
|
August 7, 2025
|
|
Sparse matrix multiplication for Metal
|
|
15
|
533
|
July 31, 2025
|
|
DiffEqGPU Trajectory Failure Handling and Heterogeneous Trajectories
|
|
4
|
159
|
July 22, 2025
|
|
Does AMDGPU.jl support integrated graphics?
|
|
3
|
309
|
July 19, 2025
|
|
Kernel with dynamic parallelism seems to be calling CPU functions
|
|
4
|
201
|
July 19, 2025
|
|
Out of dynamic GPU memory?
|
|
8
|
1607
|
July 16, 2025
|
|
Batched Hessian-Vector Product (on the GPU)
|
|
0
|
61
|
July 1, 2025
|
|
Relation between KernelAbstractions and Adapt
|
|
1
|
134
|
June 30, 2025
|
|
Cannot manage to use CUDA.atomic_add!
|
|
4
|
110
|
June 30, 2025
|
|
Heterogeneous random seeding
|
|
1
|
91
|
June 25, 2025
|
|
AMDGPU on AI HX370 versioninfo() crashes
|
|
4
|
396
|
June 8, 2025
|
|
CUDA | custom structs
|
|
3
|
190
|
June 6, 2025
|
|
Why is my GPU kernel an order of magnitude slower than my CPU function?
|
|
8
|
466
|
June 4, 2025
|
|
KernelAbstractions.get_backend(::BitArray) causes StackOverflowError
|
|
1
|
54
|
June 2, 2025
|
|
Porting cuda example to rocm amdgpu
|
|
0
|
58
|
June 2, 2025
|
|
`check-bounds=no` causes illegal memory access when using `rand()` in CUDA kernel
|
|
3
|
149
|
May 31, 2025
|
|
Current state of Metal.jl for ML and SciML
|
|
2
|
378
|
May 30, 2025
|
|
CUDA suddenly crashes with check-bounds=no, used to work fine
|
|
1
|
107
|
May 30, 2025
|
|
Unusually Slow First Device-to-Host Copy on A100 GPU
|
|
6
|
283
|
May 27, 2025
|