[PATCH 2/3] exa/mixed: be more thorough about setting fb_pitch when needed

[Michel Dänzer]

Actually looking at this again, I do have some doubts.

On Thu, 2009-11-12 at 20:11 +0100, Maarten Maathuis wrote: 
> 2009/11/12 Michel Dänzer <michel at daenzer.net>:
> > On Sat, 2009-11-07 at 13:32 +0100, Maarten Maathuis wrote:
> >> Signed-off-by: Maarten Maathuis <madman2003 at gmail.com>
> >> ---
> >>  exa/exa_migration_mixed.c |    6 +++++-
> >>  1 files changed, 5 insertions(+), 1 deletions(-)
> >>
> >> diff --git a/exa/exa_migration_mixed.c b/exa/exa_migration_mixed.c
> >> index 7458e3c..4ea9823 100644
> >> --- a/exa/exa_migration_mixed.c
> >> +++ b/exa/exa_migration_mixed.c
> >> @@ -98,10 +98,11 @@ exaDoMigration_mixed(ExaMigrationPtr pixmaps, int npixmaps, Bool can_accel)
> >>       if (!pExaPixmap->driverPriv)
> >>           exaCreateDriverPixmap_mixed(pPixmap);
> >>
> >> +     pPixmap->devKind = pExaPixmap->fb_pitch;
> >> +
> >>       if (pExaPixmap->pDamage && exaPixmapIsOffscreen(pPixmap)) {
> >>           ExaScreenPriv(pPixmap->drawable.pScreen);
> >>
> >> -         pPixmap->devKind = pExaPixmap->fb_pitch;
> >>           exaCopyDirtyToFb(pixmaps + i);
> >>
> >>           if (pExaScr->deferred_mixed_pixmap == pPixmap)

Doesn't this need to be guarded at least by if
(exaPixmapIsOffscreen(pPixmap))? Is pExaPixmap->fb_pitch valid at all
otherwise?

> >> @@ -197,6 +198,9 @@ exaPrepareAccessReg_mixed(PixmapPtr pPixmap, int index, RegionPtr pReg)
> >>       pPixmap->devPrivate.ptr = pExaPixmap->sys_ptr;
> >>       pPixmap->devKind = pExaPixmap->sys_pitch;
> >>       pExaPixmap->offscreen = FALSE;
> >> +    } else {
> >> +     ExaPixmapPriv(pPixmap);
> >> +     pPixmap->devKind = pExaPixmap->fb_pitch;
> >>      }
> >>  }

Would be cleaner to move the ExaPixmapPriv(pPixmap) to the beginning of
the function.


-- 
Earthling Michel Dänzer           |                http://www.vmware.com
Libre software enthusiast         |          Debian, X and DRI developer