Package Details: cuda-10.2 10.2.89-3

Git Clone URL: https://aur.archlinux.org/cuda-10.2.git (read-only, click to copy)
Package Base: cuda-10.2
Description: NVIDIA's GPU programming toolkit
Upstream URL: https://developer.nvidia.com/cuda-zone
Licenses: custom:NVIDIA
Provides: cuda-sdk, cuda-toolkit
Submitter: jerry73204
Maintainer: jerry73204
Last Packager: jerry73204
Votes: 1
Popularity: 0.000000
First Submitted: 2020-09-28 21:19 (UTC)
Last Updated: 2020-10-18 20:59 (UTC)

Dependencies (6)

Required by (1)

Sources (3)

Latest Comments

gamezelda commented on 2025-03-30 10:18 (UTC) (edited on 2025-03-30 10:19 (UTC) by gamezelda)

glibc-2.41, which is already in the main Arch repos, breaks building any CUDA programs due to incompatible definitions of some math functions (cospi, ...):

This can be fixed by the following patch from Gentoo: https://gitweb.gentoo.org/repo/gentoo.git/tree/dev-util/nvidia-cuda-toolkit/files/nvidia-cuda-toolkit-glibc-2.41-r1.patch?id=a8ab18e9b1a9957b8cf39d72d808825752ea0e3f

The path needs to be changed from builds/cuda_nvcc/targets/x86_64-linux/include/crt/math_functions.h to builds/cuda-toolkit/targets/x86_64-linux/include/crt/math_functions.h but otherwise the patch applies as-is.

mulletmate commented on 2022-05-16 11:31 (UTC)

Installing this (through an AUR helper) takes a really long time for me. Not sure if this is by design or if it is my end.

arch_tom commented on 2021-08-09 01:54 (UTC)

patching file 'builds/cuda-samples///findgllib.mk' Hunk #1 FAILED at 60. Hunk #2 FAILED at 100. patch: ** Can't reopen file 'builds/cuda-samples///findgllib.mk' : No such file or directory

jypma commented on 2021-03-08 11:40 (UTC)

Just tried to build this, got

gimple-match.c: In function ‘bool gimple_simplify_ROUND_DIV_EXPR(code_helper*, tree_node**, gimple**, tree_node* (*)(tree), code_helper, tree, tree, tree)’:
gimple-match.c:84783:1: internal compiler error: Segmentation fault

while building gcc-8.4.0. Not sure which gcc is the host compiler, I've got gcc (GCC) 10.2.0 installed.

MrMidnightz commented on 2021-02-07 15:37 (UTC)

If you get an error like ERROR: A failure occurred in prepare().

Try doing it through CLI like pamac

jerry73204 commented on 2020-10-15 22:00 (UTC)

Fixed. Thanks ashirviskas.

ashirviskas commented on 2020-10-15 09:35 (UTC)

Could you remove replaces like it is in cuda-10.0? Otherwise you can't have multiple CUDA versions on your machine.