On 25 August 2017 at 21:20, Emil Velikov <[email protected]> wrote:
> From: Emil Velikov <[email protected]>
>
> Inlining the implementation does not cause additional overhead in
> terms of build time while the binary is increased only marginally (~1k)
>
> At the same time the compiler should be able to optimise better, although
> this is not a path where we'll notice much difference.
>
> Use a local u_is_power_of_two to avoid pulling u_math.h. Latter of
> which would require updating 5+ locations to have an extra -I directive.
>
> Doing this will allow us to address a unrelated issue, as mentioned in
> the report below.
>
> Cc: "17.2" <[email protected]>
> Cc: Daniel Stone <[email protected]>
> Cc: Jason Ekstrand <[email protected]>
> Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=101851
> Signed-off-by: Emil Velikov <[email protected]>
Note this patch and 2/2 are effectively superseded by the following.

https://patchwork.freedesktop.org/patch/173778/

-Emil
_______________________________________________
mesa-dev mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/mesa-dev

Reply via email to