[PATCH 2/2] drm/radeon/kms/blit: workaround for a possible hardware bug

2012-02-01 Thread Michel Dänzer
On Mit, 2012-02-01 at 11:42 -0500, Ilija Hadzic wrote: > If a blit copy operation specifies a rectangle whose one dimension > is 16384 (max allowed by these chips), the chip will silently > drop all commands. What exactly does 'silently drop all commands' mean? Did you notice the following in i2

[PATCH 2/2] drm/radeon/kms/blit: workaround for a possible hardware bug

2012-02-01 Thread Ilija Hadzic
On Wed, 1 Feb 2012, Michel [ISO-8859-1] D?nzer wrote: > On Mit, 2012-02-01 at 11:42 -0500, Ilija Hadzic wrote: >> If a blit copy operation specifies a rectangle whose one dimension >> is 16384 (max allowed by these chips), the chip will silently >> drop all commands. > > What exactly does 'silen

[PATCH 2/2] drm/radeon/kms/blit: workaround for a possible hardware bug

2012-02-01 Thread Ilija Hadzic
If a blit copy operation specifies a rectangle whose one dimension is 16384 (max allowed by these chips), the chip will silently drop all commands. Fixed by reducing the maximum dimension by one rectangle unit size. In the mainline kernel, the problem is exposed only when buffers are very large (1G

Re: [PATCH 2/2] drm/radeon/kms/blit: workaround for a possible hardware bug

2012-02-01 Thread Ilija Hadzic
On Wed, 1 Feb 2012, Michel [ISO-8859-1] D�nzer wrote: On Mit, 2012-02-01 at 11:42 -0500, Ilija Hadzic wrote: If a blit copy operation specifies a rectangle whose one dimension is 16384 (max allowed by these chips), the chip will silently drop all commands. What exactly does 'silently drop a

Re: [PATCH 2/2] drm/radeon/kms/blit: workaround for a possible hardware bug

2012-02-01 Thread Michel Dänzer
On Mit, 2012-02-01 at 11:42 -0500, Ilija Hadzic wrote: > If a blit copy operation specifies a rectangle whose one dimension > is 16384 (max allowed by these chips), the chip will silently > drop all commands. What exactly does 'silently drop all commands' mean? Did you notice the following in i2

[PATCH 2/2] drm/radeon/kms/blit: workaround for a possible hardware bug

2012-02-01 Thread Ilija Hadzic
If a blit copy operation specifies a rectangle whose one dimension is 16384 (max allowed by these chips), the chip will silently drop all commands. Fixed by reducing the maximum dimension by one rectangle unit size. In the mainline kernel, the problem is exposed only when buffers are very large (1G