search

nktice / AMD-AI

AMD (Radeon GPU) ROCm based setup for popular AI tools on Ubuntu 22.04 / 23.04 / 23.10 / 24.04
Other
136 stars 8 forks source link

Fail on kernel 6.9.1-060901-generic #8

Open kcsf opened 1 week ago

kcsf commented 1 week ago

First, thanks for you very thorough guide. I wish it was working for me. I think the issue is with the kernel. What kernel are you using for ubuntu 24.04 ?

nktice commented 1 week ago

Can you share info on where you're getting crashes? Here's uname info. 

$ uname -a
Linux n 6.8.0-35-generic #35-Ubuntu SMP PREEMPT_DYNAMIC Mon May 20 15:51:52 UTC 2024 x86_64 x86_64 x86_64 GNU/Linux

I did have issues with installing early on in development, where install app crashed before the install process, but that seems fixed now. When did you download installation media that you are using?

kcsf commented 1 week ago

i just wiped my whole system and started from scratch with Ubuntu 24.04. followed all you steps, no joy:

cg@cg-AMD-BC-250:~$ clinfo Number of platforms: 3 Platform Profile: FULL_PROFILE Platform Version: OpenCL 1.1 Mesa 24.0.5-1ubuntu1 Platform Name: Clover Platform Vendor: Mesa Platform Extensions: cl_khr_icd Platform Profile: FULL_PROFILE Platform Version: OpenCL 2.1 AMD-APP (3614.0) Platform Name: AMD Accelerated Parallel Processing Platform Vendor: Advanced Micro Devices, Inc. Platform Extensions: cl_khr_icd cl_amd_event_callback Platform Profile: FULL_PROFILE Platform Version: OpenCL 3.0 Platform Name: rusticl Platform Vendor: Mesa/X.org Platform Extensions: cl_khr_byte_addressable_store cl_khr_create_command_queue cl_khr_expect_assume cl_khr_extended_versioning cl_khr_icd cl_khr_il_program cl_khr_spirv_no_integer_wrap_decoration

Platform Name: Clover Number of devices: 0 Platform Name: AMD Accelerated Parallel Processing Number of devices: 0 Platform Name: rusticl Number of devices: 0

cg@cg-AMD-BC-250:~$ uname -a Linux cg-AMD-BC-250 6.8.0-35-generic #35-Ubuntu SMP PREEMPT_DYNAMIC Mon May 20 15:51:52 UTC 2024 x86_64 x86_64 x86_64 GNU/Linux

kcsf commented 1 week ago

(Mesa Opencl was another attempt I did later...still nothing)

nktice commented 1 week ago

Have you managed to get rocm drivers working? If so can you share the rocminfo output? That may be helpful. What card do you have? and how old is your system? Any hardware notes?
When you try the guide how far into the process before there's issues?