[Intel-gfx] [PATCH 2/8] drm/i915: add the FBC mutex

Chris Wilson chris at chris-wilson.co.uk
Tue Jun 30 07:25:38 PDT 2015


On Tue, Jun 30, 2015 at 10:53:06AM -0300, Paulo Zanoni wrote:
> From: Paulo Zanoni <paulo.r.zanoni at intel.com>
> 
> Make sure we're not going to have weird races in really weird cases
> where a lot of different CRTCs are doing rendering and modesets at the
> same time.
> 
> With this change, we can start removing the struct_mutex locking we
> have around FBC in the next patches. The only struct_mutex requirement
> will be around intel_fbc_update(), since it's the place that does the
> CFB stolen memory allocations.
> 
> v2:
>  - Rebase (6 months later)
>  - Also lock debugfs and stolen.
> v3:
>  - Don't lock a single value read (Chris).
>  - Replace lockdep assertions with WARNs (Daniel).
>  - Improve commit message.
>  - Don't forget intel_pre_plane_update() locking.
> 
> Signed-off-by: Paulo Zanoni <paulo.r.zanoni at intel.com>

I haven't spotted anything scary, but I only did a cursory scan of the
patch for errors, and not down a delve into each function looking for
trouble.

Can I pretty please ask that you split fbc out of i915_gem_stolen.c
before this patch (rather than teach stolen more about fbc)?
-Chris

-- 
Chris Wilson, Intel Open Source Technology Centre


More information about the Intel-gfx mailing list