[Mesa-dev] [PATCH 5/5] intel: Remove check for kernel 2.6.29.

[Kenneth Graunke]

On 04/06/2013 05:09 AM, Daniel Vetter wrote:
> On Sat, Apr 06, 2013 at 12:26:40AM -0700, Kenneth Graunke wrote:
>> Now that we require 2.6.39, there's no need to also check for 2.6.29.
>> Calling drm_intel_bufmgr_gem_enable_fenced_relocs() without checking
>> should be safe, as it simply sets a flag.
>>
>> This does remove the check for zero fences available, but that doesn't
>> seem worth checking.
>
> Iirc this can only happen when we have ums and the ddx wants more fences
> than it should ask for. Ums is long gone, too. Last two patches are
>
> Reviewed-by: Daniel Vetter <daniel.vetter at ffwll.ch>

Thanks!

> Aside: We have a bit an accounting bug here in libdrm in that libdrm and
> the kernel don't take into account that the number of pinned fences can
> change while running (e.g. per-crtc fbs or pageflipping temporarily
> pinning more buffers).

Yeah, exactly...that's why I thought a one-time check an context 
creation wasn't worth having.

> Given that another fence undercount corner case bug
> in UXA/libdrm survived for about 4 years until it was reported, I don't
> care one bit ;-)
>
> Cheers, Daniel