The attached patches add two intrinsics to the R600 backend which are
necessary for geometry shader support in the radeonsi driver.
--
Earthling Michel Dänzer| http://www.amd.com
Libre software enthusiast |Mesa and X developer
>From 8feb7201
On 01/23/2014 03:54 PM, Chris Forbes wrote:
> How big is the sampler state pointer on Broadwell?
>
> My understanding was it would have to be >32bits, and this looks like
> it will do the wrong thing if the sampler state table is placed across
> a 4G boundary. Perhaps there is something else that
From: Michel Dänzer
Fixes half a dozen piglit tests with radeonsi.
Signed-off-by: Michel Dänzer
---
lib/Target/R600/SIInstructions.td | 5 +
test/CodeGen/R600/trunc.ll| 10 ++
2 files changed, 15 insertions(+)
diff --git a/lib/Target/R600/SIInstructions.td
b/lib/Target/R
https://bugs.freedesktop.org/show_bug.cgi?id=73999
--- Comment #4 from Michel Dänzer ---
Works for me with
--with-llvm-prefix=
instead of
LLVM_CONFIG=<...>
--
You are receiving this mail because:
You are the assignee for the bug.
___
mesa-dev mai
---
src/mesa/state_tracker/st_cb_fbo.c |6 ++
1 file changed, 6 insertions(+)
diff --git a/src/mesa/state_tracker/st_cb_fbo.c
b/src/mesa/state_tracker/st_cb_fbo.c
index eca04b8..7801484 100644
--- a/src/mesa/state_tracker/st_cb_fbo.c
+++ b/src/mesa/state_tracker/st_cb_fbo.c
@@ -544,6 +54
---
src/gallium/drivers/svga/svga_screen.c | 18 +++---
1 file changed, 11 insertions(+), 7 deletions(-)
diff --git a/src/gallium/drivers/svga/svga_screen.c
b/src/gallium/drivers/svga/svga_screen.c
index 12ea4af..32eec03 100644
--- a/src/gallium/drivers/svga/svga_screen.c
+++ b/src
This fixes a few fbo-clear-formats tests. It avoids needlessly
promoting a L8A8 texture to RGBA.
---
src/gallium/drivers/svga/svga_format.c |2 ++
1 file changed, 2 insertions(+)
diff --git a/src/gallium/drivers/svga/svga_format.c
b/src/gallium/drivers/svga/svga_format.c
index 5a153a8..7631
How big is the sampler state pointer on Broadwell?
My understanding was it would have to be >32bits, and this looks like
it will do the wrong thing if the sampler state table is placed across
a 4G boundary. Perhaps there is something else that makes this not a
factor, though.
On Fri, Jan 24, 2014
On Wednesday 22 January 2014, Brian Paul wrote:
> On 01/21/2014 03:35 PM, Fredrik Höglund wrote:
> > ---
> > src/mesa/main/bufferobj.c | 66
> > +
> > src/mesa/main/bufferobj.h | 14 ++
> > 2 files changed, 80 insertions(+)
> >
> > diff -
---
src/glx/dri_common.c | 60
src/glx/dri_common.h | 7 +++---
2 files changed, 4 insertions(+), 63 deletions(-)
diff --git a/src/glx/dri_common.c b/src/glx/dri_common.c
index 93c45ea..0dd8982 100644
--- a/src/glx/dri_common.c
+++ b/src/glx/d
---
src/loader/loader.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/src/loader/loader.c b/src/loader/loader.c
index 63b977a..f98cdb1 100644
--- a/src/loader/loader.c
+++ b/src/loader/loader.c
@@ -350,7 +350,7 @@ loader_get_driver_for_fd(int fd, unsigned driver_types)
out
Noticed when valgrinding an unrelated bug.
---
src/glx/dri3_glx.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/src/glx/dri3_glx.c b/src/glx/dri3_glx.c
index 9068ff9..3e82965 100644
--- a/src/glx/dri3_glx.c
+++ b/src/glx/dri3_glx.c
@@ -1338,14 +1338,13 @@ dri3_open(Displa
I want to stop trusting the server for the driver name, and instead decide
on our own based on the fd, so I needed this code motion.
---
src/glx/dri2_glx.c | 28 ++--
1 file changed, 14 insertions(+), 14 deletions(-)
diff --git a/src/glx/dri2_glx.c b/src/glx/dri2_glx.c
ind
Since the loader changes, there has been a compiler warning that the
prototype didn't match. It turns out that if a loader error message was
ever thrown, you'd segfault because of trying to use the warning level as
a format string.
---
src/glx/dri3_glx.c | 2 +-
src/glx/dri_common.c | 25 +
Steam links against libudev.so.0, while we're linking against
libudev.so.1. The result is that the symbol names (which are the same in
the two libraries) end up conflicting, and some of the usage of .so.1
calls the .so.0 bits, which have different internal structures, and
segfaults happen.
By usi
https://bugs.freedesktop.org/show_bug.cgi?id=73999
--- Comment #3 from Luzipher ---
Created attachment 92696
--> https://bugs.freedesktop.org/attachment.cgi?id=92696&action=edit
messages produced by compilers while running configure
--
You are receiving this mail because:
You are the assignee
This allows Mesa to choose to rename driver .sos (or split drivers),
without needing a flag day with the corresponding 2D driver.
---
src/glx/dri2_glx.c | 12 +++-
1 file changed, 11 insertions(+), 1 deletion(-)
diff --git a/src/glx/dri2_glx.c b/src/glx/dri2_glx.c
index 4e2c16a..0a0dac9 1
https://bugs.freedesktop.org/show_bug.cgi?id=73999
--- Comment #2 from Luzipher ---
Created attachment 92695
--> https://bugs.freedesktop.org/attachment.cgi?id=92695&action=edit
emerge --info
--
You are receiving this mail because:
You are the assignee for the bug.
___
https://bugs.freedesktop.org/show_bug.cgi?id=73999
--- Comment #1 from Luzipher ---
Created attachment 92694
--> https://bugs.freedesktop.org/attachment.cgi?id=92694&action=edit
emerge -pqv mesa
--
You are receiving this mail because:
You are the assignee for the bug.
https://bugs.freedesktop.org/show_bug.cgi?id=73999
Priority: medium
Bug ID: 73999
Assignee: mesa-dev@lists.freedesktop.org
Summary: mesa 32bit build fails with llvm enabled (cross
compilation)
Severity: normal
Classificat
Kenneth Graunke writes:
> Broadwell requires software to specify QPitch in a bunch of packets,
> so we decided to store it in the miptree. However, when I did that
> refactoring, I missed a subtlety: the hardware expects QPitch to be
> "in units of rows in the uncompressed surface".
>
> This is
On 01/23/2014 03:12 PM, Fredrik Höglund wrote:
On Wednesday 22 January 2014, Brian Paul wrote:
On 01/21/2014 03:35 PM, Fredrik Höglund wrote:
This will be used by glBindTextures() so we don't have to look it up
for each texture.
---
src/mesa/main/mtypes.h |1 +
src/mesa/main/texobj.c |
On Wednesday 22 January 2014, Brian Paul wrote:
> On 01/21/2014 03:35 PM, Fredrik Höglund wrote:
> > This will be used by glBindTextures() so we don't have to look it up
> > for each texture.
> > ---
> > src/mesa/main/mtypes.h |1 +
> > src/mesa/main/texobj.c |2 ++
> > 2 files changed,
https://bugs.freedesktop.org/show_bug.cgi?id=34476
Eric Anholt changed:
What|Removed |Added
Status|NEW |RESOLVED
Resolution|---
I added support for these on Haswell, but forgot to update the Broadwell
code before landing it. Fixes Piglit's max-samplers test.
Signed-off-by: Kenneth Graunke
---
src/mesa/drivers/dri/i965/gen8_vec4_generator.cpp | 25 ---
1 file changed, 22 insertions(+), 3 deletions(-)
I added this assertion in the last revision of the gen8_fs_generator
patch, and apparently didn't test it enough. Many Piglit tests fail
this test. The destination hstride is 1, while the source hstride and
vstride appear to be 0. This sounds like "scalar source" mode, which is
actually supporte
I added support for these on Haswell, but forgot to update the Broadwell
code before landing it. Partially fixes Piglit's max-samplers test.
Signed-off-by: Kenneth Graunke
---
src/mesa/drivers/dri/i965/gen8_fs_generator.cpp | 19 ++-
1 file changed, 18 insertions(+), 1 deletion(
MOV_RAW disables masking, but doesn't force the instruction to be
uncompressed. That needs to be done by hand.
Fixes textureGather and texture offset tests.
Signed-off-by: Kenneth Graunke
---
src/mesa/drivers/dri/i965/gen8_fs_generator.cpp | 5 +
1 file changed, 5 insertions(+)
diff --git
Looks good to me and fixes failing piglit test
(arb_viewport_array-scissor-ignore).
Reviewed-by: Jon Ashburn
On 01/23/2014 01:53 PM, Courtney Goeltzenleuchter wrote:
Fix ScissorArrayv and ScissorIndexed to return after detecting
parameter error.
Signed-off-by: Courtney Goeltzenleuchter
---
Fix ScissorArrayv and ScissorIndexed to return after detecting
parameter error.
Signed-off-by: Courtney Goeltzenleuchter
---
src/mesa/main/scissor.c | 2 ++
1 file changed, 2 insertions(+)
diff --git a/src/mesa/main/scissor.c b/src/mesa/main/scissor.c
index 14c8e8a..83f39e2 100644
--- a/src/mes
Oops -- there's junk in this commit message from before it was pulled
into this series.
On Tue, Jan 21, 2014 at 11:34 PM, Chris Forbes wrote:
> None of the other 3-component 16bpc formats are directly supported, so
> they get promoted to XRGB equivalents. *Not* promoting RGB16F the same
> way mak
On Thu, Jan 23, 2014 at 3:31 PM, Emil Velikov wrote:
> On 23/01/14 20:15, Ilia Mirkin wrote:
>> On Thu, Jan 23, 2014 at 3:11 PM, Emil Velikov
>> wrote:
>>> On 17/01/14 02:23, Ilia Mirkin wrote:
Fixes fbo-drawbuffers-none glClearBuffer piglit test.
Signed-off-by: Ilia Mirkin
On 23/01/14 20:15, Ilia Mirkin wrote:
> On Thu, Jan 23, 2014 at 3:11 PM, Emil Velikov
> wrote:
>> On 17/01/14 02:23, Ilia Mirkin wrote:
>>> Fixes fbo-drawbuffers-none glClearBuffer piglit test.
>>>
>>> Signed-off-by: Ilia Mirkin
>>> ---
>>>
>>> Only tested on nv50, but implementations seem simil
On 23/01/14 19:39, Ilia Mirkin wrote:
>> Ilia,
>> > Do you know why we cannot set the rt height to 64? After all you
>> > explicitly set the format to NONE.
> Because then the RT would have a size, which we don't want. The real
> question is why not set the width to 0. It was that way in Christoph'
On Thu, Jan 23, 2014 at 3:11 PM, Emil Velikov wrote:
> On 17/01/14 02:23, Ilia Mirkin wrote:
>> Fixes fbo-drawbuffers-none glClearBuffer piglit test.
>>
>> Signed-off-by: Ilia Mirkin
>> ---
>>
>> Only tested on nv50, but implementations seem similar enough.
>>
>> src/gallium/drivers/nouveau/nv50
On 17/01/14 02:23, Ilia Mirkin wrote:
> Fixes fbo-drawbuffers-none glClearBuffer piglit test.
>
> Signed-off-by: Ilia Mirkin
> ---
>
> Only tested on nv50, but implementations seem similar enough.
>
> src/gallium/drivers/nouveau/nv50/nv50_surface.c | 17 +
> src/gallium/drivers
On 17/01/14 02:23, Ilia Mirkin wrote:
> This is needed since commit 9baa45f78b (st/mesa: bind NULL colorbuffers
> as specified by glDrawBuffers).
>
> This implementation is highly based on a larger commit by
> Christoph Bumiller in his gallium-nine
> branch.
>
Ilia,
Do you know why we cannot set
On Thu, Jan 23, 2014 at 2:40 PM, Emil Velikov wrote:
> On 17/01/14 02:23, Ilia Mirkin wrote:
>> This is needed since commit 9baa45f78b (st/mesa: bind NULL colorbuffers
>> as specified by glDrawBuffers).
>>
>> This implementation is highly based on a larger commit by
>> Christoph Bumiller in his g
On 01/23/2014 11:19 AM, Mark Mueller wrote:
> On Thu, Jan 23, 2014 at 2:24 AM, Marek Olšák wrote:
>
>> On Thu, Jan 23, 2014 at 4:17 AM, Mark Mueller
>> wrote:
>>> Hi Merek,
>>
>> I'm Marek.
>>
>>>
>>>
>>>
>>>
>>>
>>> On Wed, Jan 22, 2014 at 2:49 PM, Marek Olšák wrote:
Hi Mark,
>>
On Thu, Jan 23, 2014 at 2:24 AM, Marek Olšák wrote:
> On Thu, Jan 23, 2014 at 4:17 AM, Mark Mueller
> wrote:
> > Hi Merek,
>
> I'm Marek.
>
> >
> >
> >
> >
> >
> > On Wed, Jan 22, 2014 at 2:49 PM, Marek Olšák wrote:
> >>
> >> Hi Mark,
> >>
> >
>
[...]
> >
> >
> >> Also, I have a proposal for S
https://bugs.freedesktop.org/show_bug.cgi?id=73956
Brian Paul changed:
What|Removed |Added
Status|NEW |RESOLVED
Resolution|---
https://bugs.freedesktop.org/show_bug.cgi?id=73854
Emil Velikov changed:
What|Removed |Added
Status|REOPENED|RESOLVED
Resolution|---
- Original Message -
> Fixes glean fragProg1 regression caused by commit b9f68d927ea
> (implement TGSI_PROPERTY_FS_COLOR0_WRITES_ALL_CBUFS). This bug
> only appears when the fragment shader emits fragment.Z before
> color outputs. The bug was caused by confusion between register
> index
Signed-off-by: Emil Velikov
---
src/loader/loader.c | 2 ++
1 file changed, 2 insertions(+)
diff --git a/src/loader/loader.c b/src/loader/loader.c
index a5bd769..a6dc40c 100644
--- a/src/loader/loader.c
+++ b/src/loader/loader.c
@@ -307,9 +307,11 @@ loader_get_driver_for_fd(int fd, unsigned driv
The former symbol is never defined within mesa. Based
it seems that the original intent was to use NDEBUG.
Signed-off-by: Emil Velikov
---
src/glsl/ir.h | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/src/glsl/ir.h b/src/glsl/ir.h
index 19e8383..e266328 100644
--- a/src/glsl/
On 22/01/14 00:46, Jordan Justen wrote:
> On Sat, Jan 11, 2014 at 8:54 AM, Emil Velikov
> wrote:
>> All the various window system integration layers duplicate roughly the
>> same code for figuring out device and driver name, pci-id's, etc. Which
>> is sad. So extract it out into a loader util l
Alex Henrie writes:
> 2014/1/22 Eric Anholt :
>> Thanks for the reminder. I was about to commit it, then noticed that
>> there was a placeholder copyright message. Because there's no single
>> license to the project, the current one is ill-defined. If you approve
>> of the following diff to pa
Fixes glean fragProg1 regression caused by commit b9f68d927ea
(implement TGSI_PROPERTY_FS_COLOR0_WRITES_ALL_CBUFS). This bug
only appears when the fragment shader emits fragment.Z before
color outputs. The bug was caused by confusion between register
indexes and semantic indexes.
Also added some
Reviewed-by: Marek Olšák
Marek
On Thu, Jan 23, 2014 at 5:50 PM, Brian Paul wrote:
> Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=73956
> Cc: 10.0
> Tested-by: Ahmed Allam
> ---
> src/mesa/state_tracker/st_cb_fbo.c |3 ++-
> 1 file changed, 2 insertions(+), 1 deletion(-)
>
> dif
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=73956
Cc: 10.0
Tested-by: Ahmed Allam
---
src/mesa/state_tracker/st_cb_fbo.c |3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/src/mesa/state_tracker/st_cb_fbo.c
b/src/mesa/state_tracker/st_cb_fbo.c
index 637f7ee..eca
On 01/23/2014 06:04 PM, Brian Paul wrote:
On 01/23/2014 07:31 AM, Ahmed Allam wrote:
From: Ahmed Allam
Fixes: https://bugs.freedesktop.org/show_bug.cgi?id=73956
Signed-off-by: Ahmed Allam
---
src/mesa/main/buffers.c | 4 +++-
1 file changed, 3 insertions(+), 1 deletion(-)
diff --git a/
On 01/23/2014 07:31 AM, Ahmed Allam wrote:
From: Ahmed Allam
Fixes: https://bugs.freedesktop.org/show_bug.cgi?id=73956
Signed-off-by: Ahmed Allam
---
src/mesa/main/buffers.c | 4 +++-
1 file changed, 3 insertions(+), 1 deletion(-)
diff --git a/src/mesa/main/buffers.c b/src/mesa/main/buff
https://bugs.freedesktop.org/show_bug.cgi?id=73571
Jan Vesely changed:
What|Removed |Added
Status|NEW |RESOLVED
Resolution|---
On Wed, Jan 22, 2014 at 09:13:14PM +0100, Marek Olšák wrote:
> From: Marek Olšák
>
Reviewed-by: Tom Stellard
> ---
> src/gallium/drivers/radeonsi/si_blit.c| 5 ++---
> src/gallium/drivers/radeonsi/si_commands.c| 1 -
> src/gallium/drivers/radeonsi/si_compute.c | 25 +
On Thu, Jan 23, 2014 at 12:48 PM, Ian Romanick wrote:
> From: Siavash Eliasi
>
> Ian manually ran the map_buffer_range* tests and the
> arb_map_buffer_alignment-* tests, but he did not do a full piglit run.
Thanks for going ahead and testing. The patch is
Reviewed-by: Chia-I Wu
> v2 (idr): U
From: Ahmed Allam
Fixes: https://bugs.freedesktop.org/show_bug.cgi?id=73956
Signed-off-by: Ahmed Allam
---
src/mesa/main/buffers.c | 4 +++-
1 file changed, 3 insertions(+), 1 deletion(-)
diff --git a/src/mesa/main/buffers.c b/src/mesa/main/buffers.c
index 2bdbf41..6c846cb 100644
--- a/src/
- Original Message -
> On Thu, Jan 09, 2014 at 11:32:10AM -0800, Jose Fonseca wrote:
> > This series is pretty much what I proposed several months ago, rebased on
> > top of master, with minor if any tweaks.
> >
> > Still to do is remove the deprecated abstractions (which now are mere
> >
Only skimmed through but series looks good to me.
Jose
- Original Message -
> ---
> src/gallium/drivers/svga/svga_tgsi_insn.c | 94
> ++---
> 1 file changed, 47 insertions(+), 47 deletions(-)
>
> diff --git a/src/gallium/drivers/svga/svga_tgsi_insn.c
> b/src/g
For the series:
Reviewed-by: Marek Olšák
Marek
On Thu, Jan 23, 2014 at 3:11 AM, Brian Paul wrote:
> In the _mesa_get_format_color_encoding() and _mesa_get_srgb_format_linear()
> functions.
> ---
> src/mesa/main/formats.c | 12
> 1 file changed, 12 insertions(+)
>
> diff --git a
https://bugs.freedesktop.org/show_bug.cgi?id=73672
Marek Olšák changed:
What|Removed |Added
Status|NEW |RESOLVED
Resolution|---
Not really. I'll change it.
Marek
On Thu, Jan 23, 2014 at 4:46 AM, Michel Dänzer wrote:
> On Mit, 2014-01-22 at 21:13 +0100, Marek Olšák wrote:
>>
>> index 3eab4d9..fd4e26a 100644
>> --- a/src/gallium/drivers/radeonsi/si_state.c
>> +++ b/src/gallium/drivers/radeonsi/si_state.c
>> @@ -166,6 +166,
Yes, but we might need it someday, so I preserved it.
Marek
On Thu, Jan 23, 2014 at 4:48 AM, Michel Dänzer wrote:
> On Mit, 2014-01-22 at 21:13 +0100, Marek Olšák wrote:
>>
>> @@ -81,6 +80,10 @@ struct si_textures_info {
>> unsignedn_samplers;
>> };
>>
>> +struct s
Yeah, talked with them last week, just didn't had time to test/fix the
implementation.
Basically they fixed the map/unmap issue in one of their nightly builds
and now we need to confirm that the patch below with this nightly build
fixes the issue.
Additionally to the patch below which flushe
Reviewed-by: Marek Olšák
Marek
On Thu, Jan 23, 2014 at 3:47 AM, Ilia Mirkin wrote:
> When the min_index is very large (or very negative), the multipliation
> can overflow 32 bits and result in an incorrect map pointer
> modification.
>
> Signed-off-by: Ilia Mirkin
> ---
>
> I tested this by se
On Thu, Jan 23, 2014 at 4:17 AM, Mark Mueller wrote:
> Hi Merek,
I'm Marek.
>
>
>
>
>
> On Wed, Jan 22, 2014 at 2:49 PM, Marek Olšák wrote:
>>
>> Hi Mark,
>>
>> Could you please mention or document somewhere in the code (e.g. in
>> main/formats.h) which _REV formats are incorrect according to y
On Thu, 2014-01-23 at 08:58 +0200, Pohjolainen, Topi wrote:
> On Wed, Jan 22, 2014 at 10:33:01PM +1100, Timothy Arceri wrote:
> > Signed-off-by: Timothy Arceri
> > ---
> > src/glsl/glsl_parser.yy | 128
> > +---
> > 1 file changed, 56 insertions(+), 72
https://bugs.freedesktop.org/show_bug.cgi?id=73956
--- Comment #2 from ahmabd...@hotmail.com ---
Created attachment 92646
--> https://bugs.freedesktop.org/attachment.cgi?id=92646&action=edit
Full backtrace
The source code line in
Ogre::GL3PlusFBOManager::_createTempFramebuffer ()
at
/home/
https://bugs.freedesktop.org/show_bug.cgi?id=73946
--- Comment #7 from David Herrmann ---
@Michel, any idea how to fix that?
The scanout buffers are created via gbm and this falls back to mesa
dri->image->createImage() with __DRI_IMAGE_SCANOUT set. So I guess the radeonsi
driver uses wrong tiling
68 matches
Mail list logo
