Hi Joerg, now with the really patched kernel 5.17.5 the warning doesn't show up anymore. I did two cold starts and one reboot. There are only two warnings regarding the CPU: [ 2.659141] amdgpu 0000:30:00.0: amdgpu: PSP runtime database doesn't exist [ 3.538925] amdgpu: SRAT table not found Sorry for the confusion and many thanks for your help. Regards, Jörg. [Sorry for re-sending. But somehow the other addressees got lost.] JoergRoedel wrote on 04/05/2022 10:16:> Hi Jörg, > > On Tue, May 03, 2022 at 12:17:40AM +0200, Jörg-Volker Peetz wrote: >> May 2 21:50:27 xxx kernel: WARNING: CPU: 0 PID: 1 at >> drivers/iommu/amd/init.c:851 amd_iommu_enable_interrupts+0x312/0x3f0 > > Are you sure you tested the right kernel? My patch removes that warning, > so it can't trigger anymore. It also adds a new warning, but in > different file and line. > >> In 'kern.log' I also found this: >> >> May 2 21:53:27 xxx kernel: [drm:amdgpu_job_timedout [amdgpu]] *ERROR* ring gfx >> timeout, signaled seq=16, emitted seq=17 > > GPU errors, hard to say what triggered this. Can you please send me your > exact MB and CPU model? There is a chance this is firmware-related. > > Besides that I learned that on some systems this warning only triggers > on resume. So increasing the timeout seems to be the only viable fix. > Can you please try the attached diff? It also prints the time it took to > enable the GA log. > > Regards, > > Joerg
