[Mesa-dev] [PATCH 00/12] i965: Add support for 16x MSAA on SKL+

[Ben Widawsky]

On Thu, Sep 17, 2015 at 05:00:02PM +0100, Neil Roberts wrote:
> I ran the series through Piglit but there are some issues.
> 
> The texelFetch test appears to be broken for sample counts > 10 and
> needs this patch to work:
> 
> http://patchwork.freedesktop.org/patch/59485/
> 
> The accuracy tests are failing but I think the problem is just that it
> is too strict. I've written a page about this with a diagram here:
> 
> http://busydoingnothing.co.uk/accuracy-test-is-broken/
> 
> The following tests are failing but on my SKL device the corresponding
> tests with 8 samples are also failing. As far as I understand these
> aren't known regressions for other people so it may be something to do
> with my device being pre-production. It would be great if someone
> could test on a production device.
> 
>  multisample-formats 16 gl_arb_texture_float 
>  multisample-formats 16 gl_arb_texture_rg
>  multisample-formats 16 gl_arb_texture_rg-float
>  multisample-formats 16 gl_arb_texture_rg-int 
>  multisample-formats 16 gl_ext_packed_float
>  multisample-formats 16 gl_ext_texture_integer
>  multisample-formats 16 gl_ext_texture_snorm
>  multisample-formats 16 gl_ext_texture_srgb 
>  alpha-to-one-dual-src-blend 16 
>  formats 16
>  blit-scaled samples=16
> 
> These two tests are only failing with 16 samples but so far I haven't
> been able to figure out why. I'm hoping that these will magically pass
> on a production device.
> 
>  clip-and-scissor-blit 16 downsample 
>  unaligned-blit 16 color msaa 
> 
> Jenkins is reporting some unstable tests on g45 and g965 but they
> don't seem related so I'm a bit confused.
> 

I can review this finer detail when I return from XDC, other than the first 2
which I reviewed, the series is
Acked-by: Ben Widawsky <ben at bwidawsk.net>

(FWIW, I think a bunch of the last patches should probably be squashed into 1,
but as you like...)

-- 
Ben Widawsky, Intel Open Source Technology Center