[PATCH] drm/amdgpu: Add kernel parameter to manage memory error handling.

[Deucher, Alexander]

> -----Original Message-----
> From: Michel Dänzer [mailto:michel at daenzer.net]
> Sent: Wednesday, April 12, 2017 9:17 PM
> To: Panariti, David
> Cc: Deucher, Alexander; amd-gfx at lists.freedesktop.org
> Subject: Re: [PATCH] drm/amdgpu: Add kernel parameter to manage
> memory error handling.
> 
> On 13/04/17 02:38 AM, Panariti, David wrote:
> >> From: Michel Dänzer [mailto:michel at daenzer.net]
> >>
> >>> @@ -212,6 +213,9 @@ module_param_named(cg_mask,
> >> amdgpu_cg_mask, uint,
> >>> 0444);  MODULE_PARM_DESC(pg_mask, "Powergating flags mask (0 =
> >> disable
> >>> power gating)");  module_param_named(pg_mask, amdgpu_pg_mask,
> >> uint,
> >>> 0444);
> >>>
> >>> +MODULE_PARM_DESC(ecc_mask, "ECC/EDC flags mask (0 = disable
> >>> +ECC/EDC)");
> >>
> >> "0 = disable ECC/EDC" implies that they're enabled by default? Was
> >> that already the case before this patch?
> >
> > [davep] Yes it was, and there was actually a problem in some cases
> > where the CZ would hang which is why I added the param. I was
> > wondering if it would be better to default to them being off, but I
> > wasn't sure how important maintaining original behavior is
> > considered. Actually, there are some bugs in the workaround function
> > as it is, so it really should default to off.
> 
> I agree. There have been some bug reports about Carrizo hangs, I wonder
> if any of those might be related to this.

Only the embedded SKUs support EDC.  If they are embedded parts, it could be related.

Alex