[PATCH] drm/radeon: fix TOPDOWN handling for bo_create

Lauri Kasanen cand at gmx.com
Thu Mar 12 03:21:23 PDT 2015


On Thu, 12 Mar 2015 18:02:56 +0900
Michel Dänzer <michel at daenzer.net> wrote:

> struct ttm_place::lpfn is honoured even with TTM_PL_FLAG_TOPDOWN, so
> latter should work with RADEON_GEM_CPU_ACCESS. It sounds like the
> problem is really that some BOs are expected to be within a certain
> range from the beginning of VRAM, but lpfn isn't set accordingly. It
> would be better to fix that by setting lpfn directly than indirectly via
> RADEON_GEM_CPU_ACCESS.
> 
> 
> Anyway, since this isn't the first bug which prevents
> TTM_PL_FLAG_TOPDOWN from working as intended in the radeon driver, I
> wonder if its performance impact should be re-evaluated. Lauri?

I'm sorry, I'm not in a place where I could spend the time to redo the
benchmarks.

If it causes too many issues it is of course easy to disable, but so
far the issues shown have not been caused by it - it merely exposed
wrong settings/bugs elsewhere. From this POV I would say it's good to
have it enabled, to stress the various parts.

This doesn't warm the heart of the guy with flicker after suspend, so
perhaps a kernel module parameter to disable it (defaulting to enabled)?

- Lauri


More information about the dri-devel mailing list