amdgpu with 8+ cards for GPU mining?

[Christian König]

Hi Joseph,

as a band aid you can try the attached patch. It should at least fix the 
crash at hand and allow amdgpu to continue with the boot process.

Regards,
Christian.

Am 19.02.2018 um 14:13 schrieb Christian König:
> Hi Joseph,
>
> and here is the root cause of the problem:
>> 0b:00.0 VGA compatible controller: Advanced Micro Devices, Inc. 
>> [AMD/ATI] Ellesmere [Radeon RX 470/480/570/580] (rev ef) (prog-if 00 
>> [VGA controller])
>>     Subsystem: Advanced Micro Devices, Inc. [AMD/ATI] Device 0b31
>>     Control: I/O- Mem- BusMaster- SpecCycle- MemWINV- VGASnoop- 
>> ParErr- Stepping- SERR- FastB2B- DisINTx-
>>     Status: Cap+ 66MHz- UDF- FastB2B- ParErr- DEVSEL=fast >TAbort- 
>> <TAbort- <MAbort- >SERR- <PERR- INTx-
>>     Interrupt: pin A routed to IRQ 11
>>     Region 0: Memory at <ignored> (64-bit, prefetchable) [disabled]
>>     Region 2: Memory at b0000000 (64-bit, prefetchable) [disabled] 
>> [size=2M]
>
> The BIOS is not able to assign resources to one of the VGA adapters 
> when there are more than eight installed.
>
> You could try with pci=realloc, but I doubt that there is much we can 
> do in the operating system when the BIOS messed things up like that.
>
> What we should do is to prevent amdgpu from crashing so badly, e.g. 
> allow to cleanly continue with the working hardware even when one of 
> the devices doesn't work.
>
>> when I load in amdgpu, everything froze, so I don't have the log.
> You can work around that using netconsole, see 
> Documentation/networking/netconsole.txt.
>
> Going to try to fix that by just using the screen shot you send 
> earlier, but it would be better if I can get a full log.
>
> Regards,
> Christian.
>
> Am 19.02.2018 um 12:55 schrieb Joseph Wang:
>> Here is the lspci without amdgpu loaded.  when I load in amdgpu, 
>> everything froze, so I don't have the log.
>>
>>
>> _______________________________________________
>> amd-gfx mailing list
>> amd-gfx at lists.freedesktop.org
>> https://lists.freedesktop.org/mailman/listinfo/amd-gfx
>

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.freedesktop.org/archives/amd-gfx/attachments/20180219/3b24a9e5/attachment.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: 0001-PCI-stop-crashing-in-pci_release_resource.patch
Type: text/x-patch
Size: 1093 bytes
Desc: not available
URL: <https://lists.freedesktop.org/archives/amd-gfx/attachments/20180219/3b24a9e5/attachment.bin>