|
How to report a bug
|
|
2
|
20586
|
May 27, 2024
|
|
RTX PRO 6000 Blackwell WS: whole-host hard lock during vLLM CUDA-graph
|
|
7
|
421
|
October 17, 2026
|
|
Faster code emulation on CUDA
|
|
9
|
67
|
October 16, 2026
|
|
Bypassing the Cloud: Deploying a Heterogeneous Multi-GPU Private Inference Cluster on Bare-Metal Debian
|
|
0
|
33
|
October 15, 2026
|
|
PRPLL NTT now supports both CUDA and OpenCL
|
|
4
|
571
|
October 15, 2026
|
|
How to stridedly read data for Tensor core?
|
|
3
|
334
|
October 15, 2026
|
|
Penny Royal 3.0: Qwen3.8 on RTX PRO 6000 — adaptive MTP, 524K context, and external container configuration, HiCache, NIXL
|
|
0
|
67
|
October 13, 2026
|
|
Possible NVCC Warp-Aggregated Atomic Issue
|
|
5
|
87
|
October 13, 2026
|
|
CUDA graph hangs with cross-green-context edges on R610 driver; works on R580
|
|
0
|
29
|
October 13, 2026
|
|
Share/Copy memory between two CUDA contexts in the same GPU and process
|
|
0
|
57
|
October 11, 2026
|
|
A correctly rounded implementation of the half-precision function hexpm1()
|
|
0
|
40
|
October 10, 2026
|
|
A correctly rounded implementation of the half-precision arctangent function hatan()
|
|
0
|
45
|
October 5, 2026
|
|
cudaHostRegister for read-only shared memory on GeForce RTX: support and alternatives
|
|
2
|
87
|
October 1, 2026
|
|
Performance tuning the half-precision hyperbolic tangent function htanh()
|
|
0
|
41
|
September 30, 2026
|
|
DMA-BUF not supported on RTX 5090 with open kernel module?
|
|
0
|
78
|
September 30, 2026
|
|
Performance tuning the half-precision exponential function hexp()
|
|
1
|
62
|
September 29, 2026
|
|
How should errors reported through an mbarrier be associated with individual fabric operations?
|
|
1
|
52
|
September 29, 2026
|
|
Follow-up: lost 289th bit in the reduction mod 2^256-2^32-977 after nj_umul256wide - correct fix costs 1.2 % on sm_89, cheaper way?
|
|
2
|
101
|
September 25, 2026
|
|
Implementation of CUDA extended precision floating point operations
|
|
11
|
163
|
September 24, 2026
|
|
CUDA Tiling API and optimal tile shapes
|
|
2
|
64
|
September 23, 2026
|
|
cudaGraphLaunch crashes or reports a launch failure for multi-million-node graphs
|
|
0
|
45
|
September 23, 2026
|
|
The tractor driver will return...this time with V-PACK
|
|
4
|
132
|
September 22, 2026
|
|
Announcement: opencl-api-cpp - A 'sister' library to cuda-api-wrappers, but for OpenCL
|
|
0
|
66
|
September 21, 2026
|
|
Prefetch.tensormap: elected lane vs full warp using CUTLASS/CuTe on SM90 and SM100
|
|
1
|
91
|
September 21, 2026
|
|
CUDA Graph updates report success but retain stale cluster dimensions on driver 580.159.03
|
|
0
|
55
|
September 18, 2026
|
|
Deep-fold: NF4 GEMV and CUDA graphs on RTX 3080 - feedback on performance and portability
|
|
0
|
58
|
September 18, 2026
|
|
Thin SVD Support for Polar SVD algorithm (cusolverDnXgesvdp)
|
|
4
|
128
|
September 16, 2026
|
|
Documentation or specifications on ULP precision for half (FP16) and nv_bfloat16 (BF16)?
|
|
7
|
198
|
September 15, 2026
|
|
RTX 4060 QLoRA training: 38% higher throughput but 7°C higher temperature what measurements should I collect?
|
|
5
|
152
|
September 15, 2026
|
|
Pennyroyal v2.5: Qwen3.8 on RTX PRO 6000 with SGLang — online FP8, NVMe PLE and Docker
|
|
0
|
347
|
September 15, 2026
|