p3232(double in)
{
double flr = floor(in);
FP3232 ret;
ret.integral = flr;
ret.frac = (in - flr) * (1ULL << 32);
return ret;
}
--
Earthling Michel Dänzer | http://www.amd.com
Libre software enthusiast | Debian, X and DRI develop
nd the conversion code so verbose as to be borderline
confusing, but I guess the top priority right now is to make it
correct...
I assume there will be followup patches to make the relevant code use
these and fix make check.
--
Earthling Michel Dänzer | http://www.
ot; copy & paste route?
FWIW, EXA already had a copy of miTrapezoids before this change to avoid
GPU<->CPU ping-pong (GPU clear, CPU rasterization, GPU composition).
--
Earthling Michel Dänzer | http://www.amd.com
Libre
er allocate the storage before calling down?
> Yeah, that's a good idea.
That won't work for EXA/mixed, which only allocates devPrivate.ptr on
demand.
Sorry I missed this and the DestroyPixmap ABI change when reviewing the
series.
--
Earthling Michel
From: Michel Dänzer
Make check has been broken since commit
f32c827d513c44f07e1d0fbcc0c96cef18c9a4d9 ('Input: Fix frac calculation on
[Raw]DeviceEvent conversion').
* Use floor() instead of trunc() in order to achieve the inverse of
FP3232.integral + FP3232.frac / (1ULL <
On Mon, 2011-10-03 at 16:16 -0700, Jamey Sharp wrote:
> On Mon, Oct 03, 2011 at 12:03:33PM +0200, Michel Dänzer wrote:
> > On Sam, 2011-10-01 at 23:08 -0700, Jamey Sharp wrote:
> > > In the process, move reference counting into FreePixmap instead of
> > > doing it in
n = ((char *) x)[1];
> > --
> > 1.7.3.4
> >
>
> I recommend writing a few tests in test/misc.c for this. It won't take long,
> they're quite simple to write and may just spot another bug.
FWIW, this broke the existing xi2 tests, at least on powerpc.
I had to find
pixmap is freed immediately. How about something
like 'UnreferencePixmap' or an abbreviation thereof?
--
Earthling Michel Dänzer | http://www.amd.com
Libre software enthusiast | Debian, X and DRI developer
e above, any particular reason why it wouldn't work
with that?
--
Earthling Michel Dänzer | http://www.amd.com
Libre software enthusiast | Debian, X and DRI developer
___
xorg-devel@lists.x.org: X.Org dev
On Fre, 2011-09-16 at 23:36 +1000, Peter Hutterer wrote:
> On Fri, Sep 16, 2011 at 10:50:45AM +0200, Michel Dänzer wrote:
> > > > Speaking as a radeon driver developer, merging the driver into the
> > > > server tree would be unworkable at this point because since t
On Don, 2011-09-15 at 13:46 -0500, Jeremy Huddleston wrote:
> On Sep 15, 2011, at 11:02 AM, Michel Dänzer wrote:
>
> > 3) Out of tree drivers will become second class citizens.
>
> I don't see that as a con. I see that as a benefit. If something is
> not in the
to work in a much
> more enterprise-like distro role before I began at Canonical, and I can
> attest that bisectable sources were a *big* win when a bug was
> reproducible and the sources could actually be bisected. The current X
> approach fails the latter condition.
It's n
opment model" has been in effect, it's not possible to get even
trivial changes into the server tree without a ridiculous amount of
time/effort.
--
Earthling Michel Dänzer | http://www.amd.com
Libre software enthusiast | Debian, X and DR
ing this driver independent is just a must have, and not optional.
Well, what's implemented in the Gallium DRI state tracker is just a de
facto standard which works with the current X drivers. The intention was
certainly for this to be opaque at the DRI2 protocol level. Once there
is divergence bet
ART !\n");
> + WARN(1, "trying to unbind memory to uninitialized GART !\n");
While you're at it, 'unbind from'?
--
Earthling Michel Dänzer | http://www.amd.com
Libre software enthusiast | Debian, X
On Mit, 2011-08-10 at 08:26 -0700, Keith Packard wrote:
> On Wed, 10 Aug 2011 11:36:16 +0200, Michel Dänzer wrote:
>
> > From: Michel Dänzer
> >
> > Otherwise sys_pitch will be stale when a system memory copy is allocated.
> >
> > Fixes https://bugs.freed
From: Michel Dänzer
Otherwise sys_pitch will be stale when a system memory copy is allocated.
Fixes https://bugs.freedesktop.org/show_bug.cgi?id=38322 and a crash when
unlocking the screen with xscreensaver, reported by Janne Huttunen.
Signed-off-by: Michel Dänzer
Tested-by: Janne Huttunen
On Die, 2011-08-09 at 09:23 -0700, Keith Packard wrote:
> On Tue, 09 Aug 2011 17:51:40 +0200, Michel Dänzer wrote:
>
> > It does, in the block this line moved out of. This fix is for the
> > pExaPixmap->sys_ptr == NULL case.
>
> That isn't clearly the case looki
On Die, 2011-08-09 at 08:29 -0700, Keith Packard wrote:
> On Tue, 9 Aug 2011 11:34:26 +0200, Michel Dänzer wrote:
>
> > + pExaPixmap->sys_pitch = PixmapBytePad(width, depth);
> > +
>
> Seems like pExaPixmap->sys_ptr should also get updated here somehow?
I
From: Michel Dänzer
Otherwise sys_pitch will be stale when a system memory copy is allocated.
Fixes a crash when unlocking the screen with xscreensaver, reported by Janne
Huttunen.
Signed-off-by: Michel Dänzer
Tested-by: Janne Huttunen
Signed-off-by: Michel Dänzer
---
exa/exa_mixed.c
) might be the easiest
to fix. Might be as 'simple' as making the GLX visual matching code pay
more attention to component ordering.
--
Earthling Michel Dänzer | http://www.amd.com
Libre software enthusiast | Debian, X and DRI developer
gt; RegionTranslate(®, xoff, yoff);
> pExaScr->prepare_access_reg(pPixmap, EXA_PREPARE_DEST, ®);
Reviewed-by: Michel Dänzer
--
Earthling Michel Dänzer |http://www.vmware.com
Libre software enthusiast | Debian, X and DRI develope
--
> Changes to v2:
> - avoid leaking file descriptors
> (affected are get_fnmode and set_fnmode)
Reviewed-by: Michel Dänzer
Thanks!
--
Earthling Michel Dänzer |http://www.vmware.com
Libre software enthusiast | Debian, X and DRI developer
___
ead of 0/1/-1 where possible
[...]
> +if (write(fd, &mode, 1) != 1)
> +return -1;
[...]
> +if (read(fd, &retvalue, 1) != 1)
> +return FKEYMODE_UNKNOWN;
These would leak fd, wouldn't they?
Looks good to me otherwise.
--
E
On Don, 2011-05-19 at 11:00 +1000, Peter Hutterer wrote:
> On Wed, May 18, 2011 at 10:38:36AM +0200, Michel Dänzer wrote:
> > On Mit, 2011-05-18 at 15:00 +1000, Peter Hutterer wrote:
> > > On Apple keyboards, the multimedia function keys are overlaid with the F
>
On Die, 2011-05-17 at 13:08 -0400, Owen Taylor wrote:
> On Tue, 2011-05-17 at 18:32 +0200, Michel Dänzer wrote:
> > On Die, 2011-05-17 at 11:53 -0400, Owen Taylor wrote:
> > >
> > > If libc rand() is too slow, then some inlined linear-congruential
> > >
From: Michel Dänzer
Inspired by the corresponding intel driver change by Chris Wilson.
Also drop the special case for filling the cache initially, doesn't seem to
buy anything at this point.
v2: Always update eviction position after adding new glyph to cache.
Signed-off-by: Michel D
From: Michel Dänzer
Mostly reducing function call argument passing overhead.
Also remove unused rect member pDst.
Signed-off-by: Michel Dänzer
---
exa/exa_glyphs.c | 87 --
exa/exa_priv.h |1 -
2 files changed, 32 insertions(+), 56
From: Michel Dänzer
Signed-off-by: Michel Dänzer
---
exa/exa_glyphs.c | 45 +
exa/exa_priv.h |3 ---
2 files changed, 21 insertions(+), 27 deletions(-)
diff --git a/exa/exa_glyphs.c b/exa/exa_glyphs.c
index 5c46ec9..66e4abf 100644
--- a/exa
From: Michel Dänzer
We know the mask covers the glyph extents.
Signed-off-by: Michel Dänzer
---
exa/exa_render.c | 74 ++---
1 files changed, 42 insertions(+), 32 deletions(-)
diff --git a/exa/exa_render.c b/exa/exa_render.c
index 6f2af8a
the full functionality. (As you may guess, I'm using 2, so
the function keys produce Fx by default, but I can use the multimedia
functions with the fn key)
I really like the idea behind this patch though. :)
--
Earthling Michel Dänzer |http://www.vmware.com
Libre so
On Die, 2011-05-17 at 11:53 -0400, Owen Taylor wrote:
> On Tue, 2011-05-17 at 15:03 +0200, Michel Dänzer wrote:
> > From: Michel Dänzer
> >
> > This should be just as good on average but is less expensive.
>
> If we're not hitting the cache, isn't the cost
On Die, 2011-05-17 at 15:03 +0200, Michel Dänzer wrote:
> From: Michel Dänzer
>
> Signed-off-by: Michel Dänzer
> ---
> exa/exa_unaccel.c | 11 ++-
> 1 files changed, 10 insertions(+), 1 deletions(-)
>
> diff --git a/exa/exa_unaccel.c b/exa/exa_unaccel.c
From: Michel Dänzer
Signed-off-by: Michel Dänzer
---
exa/exa_unaccel.c | 11 ++-
1 files changed, 10 insertions(+), 1 deletions(-)
diff --git a/exa/exa_unaccel.c b/exa/exa_unaccel.c
index 5d26b77..2bdaede 100644
--- a/exa/exa_unaccel.c
+++ b/exa/exa_unaccel.c
@@ -344,10 +344,19
From: Michel Dänzer
See https://bugs.freedesktop.org/show_bug.cgi?id=35197 .
Signed-off-by: Michel Dänzer
---
exa/exa_unaccel.c | 11 ++-
1 files changed, 10 insertions(+), 1 deletions(-)
diff --git a/exa/exa_unaccel.c b/exa/exa_unaccel.c
index 078b91c..5d26b77 100644
--- a/exa
From: Michel Dänzer
Mostly reducing function call argument passing overhead.
Also remove unused rect member pDst.
Signed-off-by: Michel Dänzer
---
exa/exa_glyphs.c | 87 --
exa/exa_priv.h |1 -
2 files changed, 32 insertions(+), 56
From: Michel Dänzer
Inspired by the corresponding intel driver change by Chris Wilson.
Signed-off-by: Michel Dänzer
---
exa/exa.c|3 +
exa/exa_glyphs.c | 219 ++
exa/exa_priv.h | 20 ++---
3 files changed, 82 insertions
From: Michel Dänzer
Signed-off-by: Michel Dänzer
---
exa/exa_glyphs.c | 43 ---
exa/exa_priv.h |3 ---
2 files changed, 20 insertions(+), 26 deletions(-)
diff --git a/exa/exa_glyphs.c b/exa/exa_glyphs.c
index 28aa01f..5711ca7 100644
--- a/exa
From: Michel Dänzer
This should be just as good on average but is less expensive.
Signed-off-by: Michel Dänzer
---
exa/exa_glyphs.c |4 ++--
exa/exa_priv.h |2 +-
2 files changed, 3 insertions(+), 3 deletions(-)
diff --git a/exa/exa_glyphs.c b/exa/exa_glyphs.c
index 5c46ec9
From: Michel Dänzer
We know the mask covers the glyph extents.
Signed-off-by: Michel Dänzer
---
exa/exa_render.c | 74 ++---
1 files changed, 42 insertions(+), 32 deletions(-)
diff --git a/exa/exa_render.c b/exa/exa_render.c
index 6f2af8a
*glx_dpy = __glXInitialize(dpy);
>
> if (glx_dpy == NULL)
>return False;
Superfluous whitespace-only change.
Looks good to me otherwise, but I'm not really familiar with the
client-side GLX drawable lifetime either.
--
Earthling Michel Dänzer |
On Fre, 2011-05-06 at 18:18 +0300, ville.syrj...@nokia.com wrote:
> From: Ville Syrjälä
>
> Only send invalidate events for drawables if some client has requested
> some buffers.
>
> Signed-off-by: Ville Syrjälä
Reviewed-by: Michel Dänzer
--
Earth
the previous one. This would
only fail if the client misses several wraparounds, in which case
apparently it doesn't care all that much in the first place. :)
Please do this.
--
Earthling Michel Dänzer |http://www.vmware.com
Libre software enthusiast |
+ return FALSE;
> +}
> +(*draw->pScreen->DestroyPixmap)(priv->pixmap);
> +front->pitch = pixmap->devKind;
> +front->cpp = pixmap->drawable.bitsPerPixel / 8;
> +priv->pixmap = pixmap;
> +
> +return TRUE;
> +}
Maybe at least some of this c
gt; likelyhood is still very high. The whole mess looks like a simple driver
> bug to me.
I think the xserver patches are still necessary, otherwise how are the
cached DRI2 pPriv->buffers updated for other windows sharing the same
pixmap?
--
Earthling Michel Dänzer |
On Fre, 2011-04-29 at 08:21 -0700, Jesse Barnes wrote:
> On Fri, 29 Apr 2011 09:23:27 +0200
> Michel Dänzer wrote:
>
> > On Fre, 2011-04-29 at 08:52 +0200, Michel Dänzer wrote:
> > > On Don, 2011-04-28 at 13:27 -0700, Jesse Barnes wrote:
> > > > @@ -1114,7
On Fre, 2011-04-29 at 08:52 +0200, Michel Dänzer wrote:
> On Don, 2011-04-28 at 13:27 -0700, Jesse Barnes wrote:
> > @@ -1114,7 +1169,7 @@ DRI2ScreenInit(ScreenPtr pScreen, DRI2InfoPtr info)
> > ds->ScheduleSwap = info->ScheduleSwap;
> > ds->ScheduleW
;GetMSC;
> - cur_minor = 3;
> + cur_minor = 4;
> } else {
> cur_minor = 1;
> }
This bugfix should probably be separate.
--
Earthling Michel Dänzer |http://www.vmware.com
Libre software enthusiast
an->VDisplay == best->VDisplay &&
> - (fabs(scan->VRefresh - desired->VRefresh) <
> - fabs(best->VRefresh - desired->VRefresh))) {
> + (abs(scan->VRefresh - desired->VRefresh) <
> + abs(best->VRefresh
e glyph cache.
I have patches queued for EXA to do the same.
--
Earthling Michel Dänzer |http://www.vmware.com
Libre software enthusiast | Debian, X and DRI developer
___
xorg-devel@lists.x.org: X.
RI_CORE_VERSION,
> + (void **)&screen->swrast,
> + __DRI_SWRAST, __DRI_SWRAST_VERSION);
> +}
> +if (screen->driver == NULL) {
I think it would be better to install swrastg as swrast_dri.so.
--
Earthling Michel Dänzer |
On Fre, 2011-04-08 at 18:37 +0200, Soeren Sandmann wrote:
> Michel Dänzer writes:
>
> > On Don, 2011-04-07 at 19:41 +0200, Soeren Sandmann wrote:
> >>
> >> The trap offsets change depends on a change to pixman that I will push
> >> as soon as the chan
On Fre, 2011-04-08 at 09:13 +0200, Michel Dänzer wrote:
> On Don, 2011-04-07 at 23:47 +0300, Timo Aaltonen wrote:
> > diff --git a/src/sis_utility.c b/src/sis_utility.c
> > index 64d8919..0892cf5 100644
> > --- a/src/sis_utility.c
> > +++ b/src/sis_uti
s if pixman is updated but not xserver?
--
Earthling Michel Dänzer |http://www.vmware.com
Libre software enthusiast | Debian, X and DRI developer
___
xorg-devel@lists.x.org: X.Org development
Archives: http
> +if((!(newvbflags & CRT2_ENABLE)) && (!(newvbflags & DISPTYPE_CRT1))) {
> if(!quiet) {
>xf86DrvMsg(pScrn->scrnIndex, X_ERROR,
> "CRT2 can't be switched off while CRT1 is off\n");
The outer parens around the ! expr
On Mit, 2011-04-06 at 09:50 -0700, Keith Packard wrote:
> On Wed, 06 Apr 2011 10:19:01 +0200, Michel Dänzer wrote:
>
> > 10 days have passed.
>
> I hoped you would take a few minutes and write a more efficient
> version.
Well, that's not what you said in the parts I
On Son, 2011-03-27 at 23:00 -0700, Keith Packard wrote:
> On Fri, 25 Mar 2011 18:26:56 +0100, Michel Dänzer wrote:
> > From: Michel Dänzer
> >
> > Without this, when a compositing manager unredirects a fullscreen window
> > which
> > uses DRI2 and page flip
&rgnDst, &pWin->borderClip, prgnSrc);
> +#ifdef COMPOSITE
> +if (pPixmap->screen_x || pPixmap->screen_y) {
> + xoff = -pPixmap->screen_x;
> + yoff = -pPixmap->screen_y;
> +}
> +#endif
> +RegionTranslate(&rgnDst, xoff, yoff);
Cou
On Die, 2011-03-29 at 09:48 -0400, Adam Jackson wrote:
> On Tue, 2011-03-29 at 10:09 +0200, Michel Dänzer wrote:
> > On Mon, 2011-03-28 at 12:40 -0400, Adam Jackson wrote:
> > > diff --git a/glx/glxcmds.c b/glx/glxcmds.c
> > > index 9b4bc9e..9d60bdb 100644
> >
of destruction order.
> + */
> +if (drawableId != glxDrawableId && type == GLX_DRAWABLE_WINDOW &&
> !AddResource(pDraw->id, __glXDrawableRes, pGlxDraw)) {
Is pDraw->id == drawableId always true here for windows? That was one of
the issues previously proposed f
this optimization, and the miTrapezoidBounds function did
> sometimes show up on profiles, probably because it does several
> divisions per trapezoid.
>
> V2: Call DamageRegionProcessPending() - pointed out by Michel Dänzer.
> V3: Call DamageRegionProcessPending() *after* renderin
mpositeClip);
> +
> if (!maskFormat)
> {
> int i;
Don't you also need to call DamageRegionProcessPending() after the
operation?
--
Earthling Michel Dänzer |http://www.vmware.com
Libre software enthusiast | Debian, X and DRI deve
On Mon, 2011-03-28 at 08:47 -0700, Keith Packard wrote:
> On Mon, 28 Mar 2011 10:04:23 +0200, Michel Dänzer wrote:
>
> > So once again, a fix is delayed because it's not considered quite
> > perfect yet.
>
> Kristian is the maintainer of the DRI2 code, and so shou
From: Michel Dänzer
The latter calls the former, let's cut the middle man and eliminate a branch
in a hot path.
According to Git history, ExaSetPixmapPriv was never used anywhere, just drop
it.
Signed-off-by: Michel Dänzer
Reviewed-by: Cyril Brulebois
---
exa/exa_priv.h |7 +++--
On Son, 2011-03-27 at 04:41 +0200, Cyril Brulebois wrote:
> (Mode I'm-discovering-the-code on)
>
> Michel Dänzer (26/03/2011):
> > From: Michel Dänzer
> >
> > The latter calls the former, let's cut the middle man and eliminate a branch
> > in a hot
On Son, 2011-03-27 at 23:00 -0700, Keith Packard wrote:
> On Fri, 25 Mar 2011 18:26:56 +0100, Michel Dänzer wrote:
> > From: Michel Dänzer
> >
> > Without this, when a compositing manager unredirects a fullscreen window
> > which
> > uses DRI2 and page flip
On Sam, 2011-03-26 at 11:50 +, Maarten Maathuis wrote:
> 2011/3/26 Michel Dänzer :
> > From: Michel Dänzer
> >
> > The latter calls the former, let's cut the middle man and eliminate a branch
> > in a hot path.
> >
> > Signed-off-by: Miche
From: Michel Dänzer
The latter calls the former, let's cut the middle man and eliminate a branch
in a hot path.
Signed-off-by: Michel Dänzer
---
exa/exa_priv.h |7 +++
1 files changed, 3 insertions(+), 4 deletions(-)
diff --git a/exa/exa_priv.h b/exa/exa_priv.h
index e5d90d4..70
gt; +free(ref);
> return BadAlloc;
Indentation fail? (Looks to be an issue across the series)
Also, if the second AddResource call fails, the effects of the first one
should be reversed as well. But this is certainly an improvement even
without that.
--
Earthling Miche
From: Michel Dänzer
Without this, when a compositing manager unredirects a fullscreen window which
uses DRI2 and page flipping, the DRI2 buffer information for the compositing
manager's output window (typically the Composite Overlay Window or root window)
may become stale, resulting in all
From: Michel Dänzer
Otherwise we might keep stale cached information, e.g. after the driver
performed page flipping.
This is part of the fix for
https://bugs.freedesktop.org/show_bug.cgi?id=35452 .
Signed-off-by: Michel Dänzer
Reviewed-by: Ville Syrjälä
Reviewed-by: Mario Kleiner
---
hw
On Fre, 2011-03-25 at 17:27 +0200, Ville Syrjälä wrote:
> On Fri, Mar 25, 2011 at 03:32:06PM +0100, ext Michel Dänzer wrote:
> > On Fre, 2011-03-25 at 14:47 +0200, Ville Syrjälä wrote:
> > > On Fri, Mar 25, 2011 at 12:35:37PM +0100, ext Michel Dänzer wrote:
> >
On Fre, 2011-03-25 at 15:45 +0100, Mario Kleiner wrote:
> On Mar 25, 2011, at 3:32 PM, Michel Dänzer wrote:
>
> > On Fre, 2011-03-25 at 14:47 +0200, Ville Syrjälä wrote:
> >> On Fri, Mar 25, 2011 at 12:35:37PM +0100, ext Michel Dänzer wrote:
> >>> From: Michel
On Fre, 2011-03-25 at 14:47 +0200, Ville Syrjälä wrote:
> On Fri, Mar 25, 2011 at 12:35:37PM +0100, ext Michel Dänzer wrote:
> > From: Michel Dänzer
> >
> > Without this, when a compositing manager unredirects a fullscreen window
> > which
> > uses DRI2 a
On Fre, 2011-03-25 at 06:10 -0700, Jakob Bornecrantz wrote:
> On Mar 25, 2011, at 13:47, Ville Syrjälä wrote:
> > On Fri, Mar 25, 2011 at 12:35:37PM +0100, ext Michel Dänzer wrote:
> >> From: Michel Dänzer
> >>
> >> Without this, when a compositing man
From: Michel Dänzer
Without this, when a compositing manager unredirects a fullscreen window which
uses DRI2 and page flipping, the DRI2 buffer information for the compositing
manager's output window (typically the Composite Overlay Window or root window)
may become stale, resulting in all
From: Michel Dänzer
If there's a GPU copy and a non-zero devKind was passed in, set the GPU copy
pitch to that instead of to a possibly bogus value derived from the new width.
This is e.g. used by the radeon driver's drmmode_xf86crtc_resize hook, fixes
https://bugs.freedesktop.org/sh
On Fre, 2011-02-18 at 14:23 +0100, Maarten Maathuis wrote:
> 2011/2/11 Maarten Maathuis :
> > 2011/2/11 Michel Dänzer :
> >> On Don, 2011-02-10 at 20:44 +0100, Maarten Maathuis wrote:
> >>> 2011/2/10 Michel Dänzer :
> >>> > On Don, 2011-02-10 at 20:15
r, but it's not worth the loss for a minor
> improvement that may not even exist on someone else's computer.
>
> Signed-off-by: Maarten Maathuis
Reviewed-by: Michel Dänzer
--
Earthling Michel Dänzer |http://www.vmware.com
Libre software enthusias
desktop.org/mesa/mesa/commit/?id=8d62eb45997a199e116661e26217b4d44fb9ba1e
I think this should have gone to the xorg-devel list. :) Moving there.
--
Earthling Michel Dänzer |http://www.vmware.com
Libre software enthusiast | Debian, X and DRI developer
_
On Don, 2011-02-10 at 15:23 -0800, Keith Packard wrote:
> On Thu, 10 Feb 2011 18:56:58 +0100, Michel Dänzer wrote:
>
> > Though I'm afraid I don't have the time or interest for a more complete
> > solution at this point. Can this go in for now, even if it only fixe
On Don, 2011-02-10 at 20:44 +0100, Maarten Maathuis wrote:
> 2011/2/10 Michel Dänzer :
> > On Don, 2011-02-10 at 20:15 +0100, Maarten Maathuis wrote:
> >> - It turns out that part of the problem was actually on the driver side.
> >> - The performance loss is not worth t
-by: Maarten Maathuis
With the above fixed,
Reviewed-by: Michel Dänzer
--
Earthling Michel Dänzer |http://www.vmware.com
Libre software enthusiast | Debian, X and DRI developer
___
xorg-devel@lists.x.org: X.
On Don, 2011-02-10 at 18:46 +0100, Michel Dänzer wrote:
> On Don, 2011-02-10 at 08:42 -0800, Keith Packard wrote:
> > On Thu, 10 Feb 2011 07:41:27 +0100, Michel Dänzer
> > wrote:
> >
> > > Yes, and the patch fixes it at least with the Gallium Xorg state
> >
On Don, 2011-02-10 at 08:42 -0800, Keith Packard wrote:
> On Thu, 10 Feb 2011 07:41:27 +0100, Michel Dänzer wrote:
>
> > Yes, and the patch fixes it at least with the Gallium Xorg state
> > tracker.
>
> Ok. Seems like this patch should include other visual types too;
exaMoveInPixmap_mixed(pPixmap);
> + return;
Indentation is still wrong here.
Looks good otherwise.
--
Earthling Michel Dänzer |http://www.vmware.com
Libre software enthusiast | Debian, X and DRI developer
___
return;
}
}
if (pExaScr->deferred_mixed_pixmap &&
pExaScr->deferred_mixed_pixmap != pPixmap)
[...]
rather than duplicating the else blocks.
(BTW indentation looks wrong again)
Shouldn't pExaScr->last_time_f
On Mit, 2011-02-09 at 12:43 -0800, Keith Packard wrote:
> On Wed, 09 Feb 2011 10:48:24 +0100, Michel Dänzer wrote:
>
> > Anyone? Should the server abort in this case instead? It certainly can't
> > work properly as it is.
>
> Sure looks like any driver that supp
On Fre, 2011-01-21 at 10:15 +0100, Michel Dänzer wrote:
> From: Michel Dänzer
>
> RandR doesn't seem to deal with pseudocolour at all, so without this the
> pseudocolour palette never gets loaded to the hardware.
>
> v2: Move logic into xf86_crtc_supports_gamma().
An
On Mon, 2011-02-07 at 09:34 +0100, Maarten Maathuis wrote:
> On Wed, Feb 2, 2011 at 11:55 PM, Keith Packard wrote:
> > On Sun, 30 Jan 2011 14:06:14 +0100, Maarten Maathuis
> > wrote:
> >
> >> - Not sure if it was causing problems, but you never know.
> &
On Fre, 2011-02-04 at 16:38 +0200, Pauli Nieminen wrote:
> On 04/02/11 10:44 +0100, ext Michel Dänzer wrote:
> > On Don, 2011-02-03 at 19:48 +0200, Pauli wrote:
> > > From: Pauli Nieminen
> > >
> > > If client disconnects and new client gets same id DRI2
event list entries that were never added. (Bug #31086)').
--
Earthling Michel Dänzer |http://www.vmware.com
Libre software enthusiast | Debian, X and DRI developer
___
xorg-devel@lists.x.org: X.Org
On Die, 2011-02-01 at 10:59 +0200, Rami Ylimäki wrote:
> On 01/31/2011 06:08 PM, Michel Dänzer wrote:
> > On Mon, 2011-01-31 at 15:48 +0200, Rami Ylimäki wrote:
> >> Signed-off-by: Rami Ylimäki
> >> ---
> >> do_blt.c | 10 ++
> >> 1
; +return False;
> +}
> +
This will only catch tests which end up calling InitGetImage.
--
Earthling Michel Dänzer |http://www.vmware.com
Libre software enthusiast | Debian, X and DRI developer
maller than the default window size.
The numbers are not comparable if the operation dimensions are
different. It might be better to refuse tests that can't fit.
--
Earthling Michel Dänzer |http://www.vmware.com
Libre software enthusiast | Debian,
mine I didn't even CC the list anymore for
this, YMMV.
--
Earthling Michel Dänzer |http://www.vmware.com
Libre software enthusiast | Debian, X and DRI developer
___
xorg-devel@lists.x.org: X.Org develop
From: Michel Dänzer
RandR doesn't seem to deal with pseudocolour at all, so without this the
pseudocolour palette never gets loaded to the hardware.
v2: Move logic into xf86_crtc_supports_gamma().
Signed-off-by: Michel Dänzer
---
hw/xfree86/modes/xf86Crtc.c |4
1 files chang
On Mit, 2011-01-19 at 18:29 +0100, Maarten Maathuis wrote:
> 2011/1/19 Michel Dänzer :
> > On Mit, 2011-01-19 at 10:09 +0100, Maarten Maathuis wrote:
> >> This is it, 10 ms a short time and even with this value large amounts
> >> of text are still not shown fluentl
ixed(pExaScr->deferred_mixed_pixmap);
I still can't help wondering if we aren't missing something about what
makes the difference between the text being 'shown' or not[0], but this
approach would be getting rather complicated already, and if there's no
timeout period which g
On Die, 2011-01-18 at 23:59 +0100, Maarten Maathuis wrote:
> 2011/1/18 Maarten Maathuis :
> > 2011/1/18 Maarten Maathuis :
> >> 2011/1/18 Michel Dänzer :
> >>> On Mon, 2011-01-17 at 21:45 +0100, Maarten Maathuis wrote:
> >>>> 2010/12/20 Michel Dänzer
901 - 1000 of 1388 matches
Mail list logo
