[PATCH] drm/radeon: avoid UVD corruptions on AGP cards

Dieter Nützel dieter at nuetzel-hh.de
Sun Sep 15 16:44:42 PDT 2013


Am 15.09.2013 13:31, schrieb Christian König:
> From: Christian König <christian.koenig at amd.com>
> 
> Putting everything into VRAM seems to help.
> 
> Signed-off-by: Christian König <christian.koenig at amd.com>
> ---
>  drivers/gpu/drm/radeon/radeon_cs.c | 8 +++++---
>  1 file changed, 5 insertions(+), 3 deletions(-)
> 
> diff --git a/drivers/gpu/drm/radeon/radeon_cs.c
> b/drivers/gpu/drm/radeon/radeon_cs.c
> index 27ea004..ac6ece6 100644
> --- a/drivers/gpu/drm/radeon/radeon_cs.c
> +++ b/drivers/gpu/drm/radeon/radeon_cs.c
> @@ -81,9 +81,11 @@ static int radeon_cs_parser_relocs(struct
> radeon_cs_parser *p)
>  		p->relocs[i].lobj.bo = p->relocs[i].robj;
>  		p->relocs[i].lobj.written = !!r->write_domain;
> 
> -		/* the first reloc of an UVD job is the
> -		   msg and that must be in VRAM */
> -		if (p->ring == R600_RING_TYPE_UVD_INDEX && i == 0) {
> +		/* the first reloc of an UVD job is the msg and that must be in
> +		   VRAM, also but everything into VRAM on AGP cards to avoid
> +		   image corruptions */
> +		if (p->ring == R600_RING_TYPE_UVD_INDEX &&
> +		    (i == 0 || p->rdev->flags & RADEON_IS_AGP)) {
>  			/* TODO: is this still needed for NI+ ? */
>  			p->relocs[i].lobj.domain =
>  				RADEON_GEM_DOMAIN_VRAM;

Go a head - direct way to stable!!!
What a nice weekend present Christian ;-)
Works like a charm.

2 simultaneous mplayer videos (1280x720 H.264, 854x480 H.264) plus ~35 % 
idle on my poor Duron 1800.

Chen Jie's test.avi (1920x1080 H.264) on my 1920x1080 desktop at ~35% 
idle, too.

Cheers,
   Dieter
-------------- next part --------------
An embedded and charset-unspecified text was scrubbed...
Name: dmesg-drm-fixes-3.12-AGP-patch-dpm-1.log
URL: <http://lists.freedesktop.org/archives/dri-devel/attachments/20130916/2e662c7b/attachment-0001.ksh>


More information about the dri-devel mailing list