+- hashcat Forum (https://hashcat.net/forum)
+-- Forum: Support (https://hashcat.net/forum/forum-3.html)
+--- Forum: hashcat (https://hashcat.net/forum/forum-45.html)
+--- Thread: [Solved] A problem with hashcat (/thread-10727.html)
[Solved] A problem with hashcat - Slopy - 04-17-2022
Hey, I recently upgraded my GPU to a nvidia 30 series card and tried to use hashcat however I keep on getting an error
I'm using the latest hashcat I believe v6.2.5 and I installed it from here, the official website
Whenever I try to run hashcat I get this error
Code:
cuLinkAddData(): the provided PTX was compiled with an unsupported toolchain.
* Device #1: Kernel ./OpenCL/shared.cl link failed. Error Log:
ptxas application ptx input, line 9; fatal : Unsupported .version 7.6; current version is '7.4'
* Device #1: Kernel ./OpenCL/shared.cl build failed.I installed this CUDA version
Code:
cuda_11.6.2_511.65_windows.exeAlright thanks to ZerBea, I figured out the problem
I had two GPUs a GT 710 and an rtx
Apparently, nvidia won't update your game ready driver if the GT 710 is active. I tried to disable the GT 710, update the driver then re enable it, however that somehow just automatically rerolled the drive version back to the old version
So you[whoever faces this problem] have two options, either install an older cuda toolkit or get rid of that old gpu
RE: A problem with hashcat - ZerBea - 04-17-2022
Your cuda version doesn't match to your driver version.
https://docs.nvidia.com/cuda/cuda-toolkit-release-notes/index.html#abstract
RE: A problem with hashcat - Slopy - 04-17-2022
(04-17-2022, 06:26 PM)ZerBea Wrote: Your cuda version doesn't match to your driver version.
https://docs.nvidia.com/cuda/cuda-toolkit-release-notes/index.html#abstract
does that mean that I should install an older cuda toolkit?
RE: A problem with hashcat - ZerBea - 04-18-2022
"does that mean that I should install an older cuda toolkit? "
No. Take a look at table 3: CUDA Toolkit and Corresponding Driver Versions
Install latest driver (you haven't mentioned your driver version) and latest cuda toolkit (you have done that: version 11.6.2) as mentioned by the link.
In my case (running Arch Linux):
installed driver 510.60.02 and matching cuda toolkit 11.6.1
as mentioned in table 3:
CUDA 11.6 Update 1 >=510.47.03 >=511.65