On 6 June 2016 at 05:40, Dave Airlie wrote:
> From: Dave Airlie
>
> The clamping code always clamps to 0..1, which for SNORM is
> incorrect. This adds a bit to say that clamping is for
> an snorm and to use the correct range.
>
> This fixes a number of SNORM cases in:
> GL33-CTS.texture_size_prom
From: Dave Airlie
The clamping code always clamps to 0..1, which for SNORM is
incorrect. This adds a bit to say that clamping is for
an snorm and to use the correct range.
This fixes a number of SNORM cases in:
GL33-CTS.texture_size_promotion.functional
v2: handle the case where the format is a
Patch looks good. I don't have the CTS test accessible, so I couldn't
confirm myself that it is fixed. If that's confirmed, patch is:
Reviewed-by: Eduardo Lima Mitev
Left a few long-line nitpicks below.
Eduardo
On 05/10/2016 05:59 AM, Dave Airlie wrote:
> From: Dave Airlie
>
> The clamping c
ping?
On 10 May 2016 at 13:59, Dave Airlie wrote:
> From: Dave Airlie
>
> The clamping code always clamps to 0..1, which for SNORM is
> incorrect. This adds a bit to say that clamping is for
> an snorm and to use the correct range.
>
> This fixes a number of SNORM cases in:
> GL33-CTS.texture_s
From: Dave Airlie
The clamping code always clamps to 0..1, which for SNORM is
incorrect. This adds a bit to say that clamping is for
an snorm and to use the correct range.
This fixes a number of SNORM cases in:
GL33-CTS.texture_size_promotion.functional
Signed-off-by: Dave Airlie
---
src/mesa