According to the Sandybridge PRM, Volume 1, Part 5, Section 1.9.15,
Gen6's blitter supports Y-tiled buffers as well as X-tiled. Pitch is
specified in 512-byte granularity for X-tiled, but 128-byte for Y-tiled.
Gen5 and earlier unfortunately only support X-tiled buffers.
Fixes a software fallback
intelEmitCopyBlit already checks for this, so the check is redundant and
unnecessary. This consolidates the logic (which will soon change).
NOTE: This is a candidate for the 7.10 and 7.11 branches.
Signed-off-by: Kenneth Graunke
---
src/mesa/drivers/dri/intel/intel_tex_copy.c |5 -
1 f
gl_texture_image::InternalFormat is actually the user requested internal
format, not what the texture actually is. Thus, even though we don't
support 32-bit depth buffers, we need to recognize the enumeration here.
Otherwise, it wrongly returns the color read buffer instead of the depth
read buffe
On Sun, Jun 26, 2011 at 10:39 PM, Chia-I Wu wrote:
> From: Chia-I Wu
>
> The idea is that DRI driver, libGL and libOSMesa are libraries that can
> be independently enabled, yet --with-driver does not allow us to easily
> do that, if not impossible. This also matches what
> --enable-{egl,xorg,d3d
Ok in fact there's a gcc bug about memcmp:
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=43052
In short gcc's memcmp builtin is totally lame and loses to glibc's
memcmp (including call overhead, no knowledge about alignment etc.) even
when comparing only very few bytes (and loses BIG time for lots of
https://bugs.freedesktop.org/show_bug.cgi?id=38789
Summary: [bisected] Hacks cubestorm, noof and pipes from
XScreenSaver stopped displaying anything
Product: Mesa
Version: git
Platform: x86-64 (AMD64)
OS/Version: Linux (All)
Commit 6750226e6d915742ebf96bae2cfcdd287b85db35 bumped the base MRF to
m2 instead of m0, but failed to adjust inst->mlen, which was being set
to the highest MRF. Subtracting the base MRF solves the issue.
Signed-off-by: Kenneth Graunke
---
src/mesa/drivers/dri/i965/brw_fs_visitor.cpp | 11 +++
I didn't even look at that was just curious why the memcmp (which is
used a lot in other places) is slow. However, none of the other memcmp
seem to show up prominently (cso functions are quite low in profiles,
_mesa_search_program_cache uses memcmp too but it's not that high
neither). So I guess th
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
On 06/29/2011 07:34 AM, Micael wrote:
> I'm new here, and I'm looking for a way to submit a patch that I have
> created that fixes an "index out of bounds" issue.
> I already attached the patch to the bug report in question
> (https://bugs.freedesktop
Okay, so maybe I'm failing to recognize the exact situation here, but
wouldn't it be possible to mark the FS state with a serial number and
just compare those? Or are these FS states not CSO-cached?
~ C.
On Wed, Jun 29, 2011 at 3:44 PM, Roland Scheidegger wrote:
> Actually I ran some numbers her
On Wed, 15 Jun 2011 00:49:33 -0700, Kenneth Graunke
wrote:
> No MOV is necessary since signed/unsigned integers share the same
> bit-representation; it's simply a question of interpretation. In
> particular, the fs_reg::imm union shouldn't need updating.
>
> Signed-off-by: Kenneth Graunke
> --
Actually I ran some numbers here and tried out a optimized struct compare:
original ipers: 12.1 fps
ajax patch: 15.5 fps
optimized struct compare: 16.8 fps
This is the function I used for that (just enabled in that lp_setup
function):
static INLINE int util_cmp_struct(const void *src1, const voi
From: John
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=30217
---
src/mesa/drivers/dri/intel/intel_context.c |2 ++
1 files changed, 2 insertions(+), 0 deletions(-)
diff --git a/src/mesa/drivers/dri/intel/intel_context.c
b/src/mesa/drivers/dri/intel/intel_context.c
index 91f6c6d.
---
src/mesa/drivers/dri/i915/i830_vtbl.c |2 --
src/mesa/drivers/dri/i915/i915_vtbl.c |2 --
src/mesa/drivers/dri/intel/intel_fbo.c| 13 -
src/mesa/drivers/dri/intel/intel_screen.c |1 -
4 files changed, 0 insertions(+), 18 deletions(-)
diff --git a/src/mesa
This should help us avoid leaking regions in region reference code by
making the API more predictable.
---
src/mesa/drivers/dri/intel/intel_regions.c | 10 ++
1 files changed, 6 insertions(+), 4 deletions(-)
diff --git a/src/mesa/drivers/dri/intel/intel_regions.c
b/src/mesa/drivers/dri
Even if we don't have a current context, if we're freeing the rb we
should free its region (and BO). The renderbuffer unreference checks
appear to be just cargo-cult from the region unreference code.
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=30217
---
src/mesa/drivers/dri/intel/inte
Note that as a result of this cleanup a bug in
intel_process_dri2_buffer_no_separate_stencil() beomes quite apparent.
---
src/mesa/drivers/dri/intel/intel_context.c | 40
src/mesa/drivers/dri/intel/intel_fbo.c | 19 -
src/mesa/drivers/dri/intel/inte
Right back at ya', comments in-line, in response to your in-line comments.
Thanks for the good feedback and review.
cheers, danm
On 06/29/2011 12:53 PM, Kenneth Graunke wrote:
On 06/28/2011 02:48 PM, Dan McCabe wrote:
Up until now modifying the GLSL compiler has been pretty straightforward.
T
On Wed, Jun 29, 2011 at 10:34 AM, Micael wrote:
> Hi
> I'm new here, and I'm looking for a way to submit a patch that I have
> created that fixes an "index out of bounds" issue.
> I already attached the patch to the bug report in question
> (https://bugs.freedesktop.org/show_bug.cgi?id=34495). Is
Hi
I'm new here, and I'm looking for a way to submit a patch that I have
created that fixes an "index out of bounds" issue.
I already attached the patch to the bug report in question (
https://bugs.freedesktop.org/show_bug.cgi?id=34495). Is that enough?
Thanks.
--
Micael Dias
__
On 06/28/2011 02:48 PM, Dan McCabe wrote:
> We now tie the grammar to the ctors of the ASTs they reference.
>
> This requires that we actually have definitions of the ctors.
>
> In addition, we also need to define "print" and "hir" methods for the AST
> classes. The Print methods are pretty simpl
On 06/28/2011 02:48 PM, Dan McCabe wrote:
> Up until now modifying the GLSL compiler has been pretty straightforward.
> This is where things get interesting. But still pretty straightforward.
Dan,
This patch looks good! I found a few small issues...nothing major. I
think one more round and it s
https://bugs.freedesktop.org/show_bug.cgi?id=38716
Alexandre Demers changed:
What|Removed |Added
Status|REOPENED|RESOLVED
Resolution|
Ohh that's interesting, you'd think the comparison shouldn't be that
expensive (though I guess in ipers case the comparison is never true).
memcmp is quite extensively used everywhere. Maybe we could replace that
with something faster (since we only ever care if the blocks are the
same but not care
Create a new GLX drawable struct to track client related info, and add a
wrap counter to it drawable and track it as we receive events. This
allows us to support the full 64 bits of the event structure we pass to
the client even though the server only gives us a 32 bit count.
Reviewed-by: Michel
On Wed, 2011-06-29 at 13:19 -0400, Adam Jackson wrote:
> Perversely, do this by eliminating the comparison between stored and
> current fs state. On ipers, a perf trace showed try_update_scene_state
> using 31% of a CPU, and 98% of that was in 'repz cmpsb', ie, the memcmp.
> Taking that out takes
Use the new swap event type so we get valid SBC values.
Reviewed-by: Ian Romanick
Reviewed-by: Jeremy Huddleston
Signed-off-by: Jesse Barnes
---
configure.ac |4 ++--
src/glx/dri2.c |4 ++--
src/glx/glxext.c |4 ++--
3 files changed, 6 insertions(+), 6 deletions(-)
diff --gi
On Wed, 29 Jun 2011 11:56:29 +0200, Gustaw Smolarczyk
wrote:
> 2011/6/29 Eric Anholt :
> > + if (fabs(test_data[0] - result_data[0]) > epsilon ||
> > + fabs(test_data[1] - result_data[1]) > epsilon ||
> > + fabs(test_data[2] - result_data[2]) > epsilon ||
> > +
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
On 06/28/2011 04:49 PM, Eric Anholt wrote:
> Fixes a bug caught by oglconform, and now piglit
> ARB_vertex_program/getenv4d-with-error. The wrapping of an existing
> GL function made it so that we couldn't distinguish an error in
> looking up our argu
Perversely, do this by eliminating the comparison between stored and
current fs state. On ipers, a perf trace showed try_update_scene_state
using 31% of a CPU, and 98% of that was in 'repz cmpsb', ie, the memcmp.
Taking that out takes try_update_scene_state down to 6.5% of the
profile; more import
On 06/28/2011 10:47 PM, Kenneth Graunke wrote:
On 06/28/2011 02:48 PM, Dan McCabe wrote:
Previously we added productions for:
switch_body
case_label_list
case_statement
case_statement_list
Now add AST structs corresponding to those productions.
Both 1/3 and 3/3 l
2011/6/29 Jose Fonseca :
> Stephane,
>
> It doens't look right to me: we did allocate these (with the
> tgsi_alloc_tokens() call above); and the driver should not delete them (the
> driver will make a copy of the tokens if it needs them, and should not hold
> on to the incoming tokens.
>
> If th
Chad Versace (1):
Add attachment token DRI2BufferHiz
Jesse Barnes (2):
Revert "dri2proto: make DRI2 swap event match GLX spec"
dri2proto: add a new DRI2BufferSwapComplete struct that matches the spec
git tag: dri2proto-2.6
http://xorg.freedesktop.org/archive/individual/proto/dr
On Tue, 28 Jun 2011 21:20:14 -0600, Brian Paul wrote:
> On Tue, Jun 28, 2011 at 5:49 PM, Eric Anholt wrote:
> > This error result doesn't appear in the GL 2.1 or 3.2 compatibility
> > specs, and triggers an unexpected GL error in Intel's oglconform when
> > it tries to reset the feedback state af
https://bugs.freedesktop.org/show_bug.cgi?id=38776
Alex Deucher changed:
What|Removed |Added
Product|Mesa|freedesktop.org
Component|Other
https://bugs.freedesktop.org/show_bug.cgi?id=38776
--- Comment #1 from Vadim 2011-06-29 08:05:16 PDT ---
Created an attachment (id=48555)
--> (https://bugs.freedesktop.org/attachment.cgi?id=48555)
gpg key
--
Configure bugmail: https://bugs.freedesktop.org/userprefs.cgi?tab=email
--- You ar
https://bugs.freedesktop.org/show_bug.cgi?id=38776
Summary: Account request for Vadim Girlin
Product: Mesa
Version: unspecified
Platform: Other
OS/Version: All
Status: NEW
Severity: normal
Priority: medium
On Tue, Jun 28, 2011 at 11:47 AM, Emil Velikov wrote:
> Commit 1a339b6c(st/mesa: prefer native texture formats when possible)
> introduced two new arguments to the st_choose_format() functions.
> This patch fixes the order and passes the correct internal_target
> rather than GL_NONE
>
> NOTE: This
On Mon, Jun 27, 2011 at 9:03 AM, Andre Maasikas wrote:
> fixes assert later on in texcompress2/r600g
Pushed. Thanks!
-Brian
___
mesa-dev mailing list
mesa-dev@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/mesa-dev
On Wed, Jun 29, 2011 at 5:49 AM, Jon TURNEY wrote:
>> Module: Mesa
>> Branch: master
>> Commit: 062a1e291fdc0ef69b6677f8ae0e3471047e281d
>> URL:
>> http://cgit.freedesktop.org/mesa/mesa/commit/?id=062a1e291fdc0ef69b6677f8ae0e3471047e281d
>>
>> Author: Stéphane Marchesin
>> Date: Tue Jun 28
Jon,
Thanks for pointing this out. I apparently missed at least one case.
I'll fix ASAP.
/Thomas
On 06/29/2011 02:20 PM, Jon TURNEY wrote:
On 28/06/2011 13:19, Thomas Hellstrom wrote:
The api and the state tracker manager code as well as the state tracker code
assumed that only a single c
---
There is ~20% fps boost with the etqw timenetdemo after this patch.
src/gallium/drivers/r600/r600.h |1 +
src/gallium/drivers/r600/r600_pipe.c |3 ---
src/gallium/winsys/r600/drm/r600_hw_context.c |4 +++-
3 files changed, 4 insertions(+), 4 deletions(-)
d
On 28/06/2011 13:19, Thomas Hellstrom wrote:
> The api and the state tracker manager code as well as the state tracker code
> assumed that only a single context could be bound to a drawable. That is not
> a valid assumption, since multiple contexts can bind to the same drawable.
>
> Fix this by ma
> Module: Mesa
> Branch: master
> Commit: 062a1e291fdc0ef69b6677f8ae0e3471047e281d
> URL:
> http://cgit.freedesktop.org/mesa/mesa/commit/?id=062a1e291fdc0ef69b6677f8ae0e3471047e281d
>
> Author: Stéphane Marchesin
> Date: Tue Jun 28 00:53:01 2011 -0700
>
> i915g: Improve SIN/COS a bit.
>
>
Stephane,
It doens't look right to me: we did allocate these (with the
tgsi_alloc_tokens() call above); and the driver should not delete them (the
driver will make a copy of the tokens if it needs them, and should not hold on
to the incoming tokens.
If there's a double free, then I suspect the
Looks a good thing to do.
Jose
- Original Message -
> ---
> src/gallium/auxiliary/draw/draw_pipe_aaline.c |8
> src/gallium/auxiliary/draw/draw_pipe_aapoint.c |9 +
> 2 files changed, 17 insertions(+), 0 deletions(-)
>
> diff --git a/src/gallium/auxiliary/draw/
On Wed, Jun 29, 2011 at 4:12 PM, Thomas Hellstrom wrote:
[snipped]
> diff --git a/src/gallium/state_trackers/vega/vg_manager.c
> b/src/gallium/state_trackers/vega/vg_manager.c
> index eeea686..dec1581 100644
> --- a/src/gallium/state_trackers/vega/vg_manager.c
> +++ b/src/gallium/state_trackers/v
- Original Message -
> On 06/29/2011 10:10 AM, Jose Fonseca wrote:
> > - Original Message -
> >
> >> Note that the state trackers themselves don't use atomic stamps on
> >> frame-buffers. Multiple context rendering to the same drawable
> >> should
> >> be protected by the appl
2011/6/29 Eric Anholt :
> ---
> tests/all.tests | 4 +
> tests/spec/CMakeLists.txt | 1 +
> tests/spec/arb_vertex_program/CMakeLists.gl.txt | 16
> tests/spec/arb_vertex_program/CMakeLists.txt | 1 +
> .../spec/ar
On 06/29/2011 10:10 AM, Jose Fonseca wrote:
- Original Message -
Note that the state trackers themselves don't use atomic stamps on
frame-buffers. Multiple context rendering to the same drawable should
be protected by the application.
I known that contexts must not be used by
- Original Message -
> Note that the state trackers themselves don't use atomic stamps on
> frame-buffers. Multiple context rendering to the same drawable should
> be protected by the application.
I known that contexts must not be used by two threads simultaneously, but I
don't recall eve
The api and the state tracker manager code as well as the state tracker code
assumed that only a single context could be bound to a drawable. That is not
a valid assumption, since multiple contexts can bind to the same drawable.
Fix this by making it the state tracker's responsibility to update al
52 matches
Mail list logo
