Nouveau - Dec 2017 - [PATCH] dri3: remove bogus condition for creating pixmap

Not clear what the depth % 8 was trying to protect against, but it was
breaking 30bpp visuals with DRI3.

Signed-off-by: Ilia Mirkin <imirkin at alum.mit.edu>
---
 src/nouveau_dri2.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/src/nouveau_dri2.c b/src/nouveau_dri2.c
index cbb7b2a..07b6022 100644
--- a/src/nouveau_dri2.c
+++ b/src/nouveau_dri2.c
@@ -1076,7 +1076,7 @@ static PixmapPtr nouveau_dri3_pixmap_from_fd(ScreenPtr
screen, int fd, CARD16 wi
 	struct nouveau_bo *bo = NULL;
 	struct nouveau_pixmap *nvpix;
 
-	if (depth < 8 || depth > 32 || depth % 8)
+	if (depth < 8 || depth > 32)
 		return NULL;
 
 	pixmap = screen->CreatePixmap(screen, 0, 0, depth, 0);
-- 
2.13.6

On Sun, Dec 31, 2017 at 2:23 AM, Ilia Mirkin <imirkin at alum.mit.edu>
wrote:
> Not clear what the depth % 8 was trying to protect against, but it was
> breaking 30bpp visuals with DRI3.
Erm, actually switching to my test config which doesn't enable dri3 is
what "fixed" things. DRI3 is still busted. Ignore this patch until
that's properly figured out for 30bpp.
>
> Signed-off-by: Ilia Mirkin <imirkin at alum.mit.edu>
> ---
>  src/nouveau_dri2.c | 2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/src/nouveau_dri2.c b/src/nouveau_dri2.c
> index cbb7b2a..07b6022 100644
> --- a/src/nouveau_dri2.c
> +++ b/src/nouveau_dri2.c
> @@ -1076,7 +1076,7 @@ static PixmapPtr
nouveau_dri3_pixmap_from_fd(ScreenPtr screen, int fd, CARD16 wi
>         struct nouveau_bo *bo = NULL;
>         struct nouveau_pixmap *nvpix;
>
> -       if (depth < 8 || depth > 32 || depth % 8)
> +       if (depth < 8 || depth > 32)
>                 return NULL;
>
>         pixmap = screen->CreatePixmap(screen, 0, 0, depth, 0);
> --
> 2.13.6
>

On Sun, Dec 31, 2017 at 2:35 AM, Ilia Mirkin <imirkin at alum.mit.edu>
wrote:
> On Sun, Dec 31, 2017 at 2:23 AM, Ilia Mirkin <imirkin at
alum.mit.edu> wrote:
>> Not clear what the depth % 8 was trying to protect against, but it was
>> breaking 30bpp visuals with DRI3.
>
> Erm, actually switching to my test config which doesn't enable dri3 is
> what "fixed" things. DRI3 is still busted. Ignore this patch
until
> that's properly figured out for 30bpp.
So actually it does seem like DRI3 works fine now. After a lot of
tracing, looks like bitsPerPixel will still get set to 32 in this
case, so a lot of the code that does like bitsPerPixel / 8 should
remain OK. I'm following up with a patch which also ensures that the
pixmap->drawable.bitsPerPixel % 8 == 0. Presumably that's what was
trying to be checked earlier.
>
>>
>> Signed-off-by: Ilia Mirkin <imirkin at alum.mit.edu>
>> ---
>>  src/nouveau_dri2.c | 2 +-
>>  1 file changed, 1 insertion(+), 1 deletion(-)
>>
>> diff --git a/src/nouveau_dri2.c b/src/nouveau_dri2.c
>> index cbb7b2a..07b6022 100644
>> --- a/src/nouveau_dri2.c
>> +++ b/src/nouveau_dri2.c
>> @@ -1076,7 +1076,7 @@ static PixmapPtr
nouveau_dri3_pixmap_from_fd(ScreenPtr screen, int fd, CARD16 wi
>>         struct nouveau_bo *bo = NULL;
>>         struct nouveau_pixmap *nvpix;
>>
>> -       if (depth < 8 || depth > 32 || depth % 8)
>> +       if (depth < 8 || depth > 32)
>>                 return NULL;
>>
>>         pixmap = screen->CreatePixmap(screen, 0, 0, depth, 0);
>> --
>> 2.13.6
>>