On 11/30/2016 10:44 PM, Christian Gmeiner wrote:
> This a very lightweight library to add basic support for
> renderonly GPUs. It does all the magic regarding in/exporting
> buffers etc. This library will likely break android support and
> hopefully will get replaced with a better solution based on
On 08/12/16 12:02 AM, Roland Scheidegger wrote:
> The bug in llvm has been fixed, can you confirm lp_test_format passes again?
Yep, it does, thanks!
--
Earthling Michel Dänzer | http://www.amd.com
Libre software enthusiast | Mesa and X develop
On 08/12/16 11:53 AM, Pierre-Loup A. Griffais wrote:
> On 12/07/2016 06:13 PM, Michel Dänzer wrote:
>> On 08/12/16 09:59 AM, Rob Clark wrote:
>>> +author again.. no idea why list keeps loosing extra cc's..
>>
>> Mailman removes addresses from Cc which are subscribed to the list and
>> have "Avoid d
On 12/07/2016 06:13 PM, Michel Dänzer wrote:
On 08/12/16 09:59 AM, Rob Clark wrote:
+author again.. no idea why list keeps loosing extra cc's..
Mailman removes addresses from Cc which are subscribed to the list and
have "Avoid duplicate copies of messages?" enabled in the list
Subscription O
Reviewed-by: Bruce Cherniak
> On Dec 7, 2016, at 5:55 PM, Rowley, Timothy O
> wrote:
>
> This one-off specialization was missed.
> ---
> src/gallium/drivers/swr/rasterizer/core/backend.h | 4 +-
> .../drivers/swr/rasterizer/memory/StoreTile.h | 46 +++---
> 2 files change
Reviewed-by: Bruce Cherniak
> On Dec 7, 2016, at 7:16 PM, Tim Rowley wrote:
>
> v2: use fmul(1/65536) instead of fdiv(65535)
> ---
> .../drivers/swr/rasterizer/common/formats.cpp | 104 ++---
> .../drivers/swr/rasterizer/common/formats.h| 7 +-
> .../drivers/swr/ra
This handles two cases. The first is where the second ifs
condition is a phi with constant srcs origination from the first if.
The second is where both ifs have the same condtion.
A limitation to merging is that currently this code will only merge
the ifs if there is only a single phi instruction
On 08/12/16 09:59 AM, Rob Clark wrote:
> +author again.. no idea why list keeps loosing extra cc's..
Mailman removes addresses from Cc which are subscribed to the list and
have "Avoid duplicate copies of messages?" enabled in the list
Subscription Options.
--
Earthling Michel Dänzer
https://bugs.freedesktop.org/show_bug.cgi?id=99010
--- Comment #14 from Tobias Droste ---
Oops.
Fixes sent to mailing list:
https://lists.freedesktop.org/archives/mesa-dev/2016-December/137798.html
You now get a output like this:
llvm found: yes
llvm-config: /llvm-conf
This renames llvm_check_version_for to llvm_require_version and let it
set a variable to mark that LLVM will be used.
Use this to make a usefull configure output and to only check if the
libs are found in LLVM if it is actually used.
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=99010
S
Make sure that HAVE_LLVM compiler define is only set if LLVM is
actually used.
Signed-off-by: Tobias Droste
---
configure.ac | 7 ++-
1 file changed, 6 insertions(+), 1 deletion(-)
diff --git a/configure.ac b/configure.ac
index a209d87e62..7362ecdfaf 100644
--- a/configure.ac
+++ b/configur
Make setting LLVM_LIBS a function, put it to the other LLVM related
functions and call it if LLVM is used.
Signed-off-by: Tobias Droste
---
configure.ac | 109 +--
1 file changed, 54 insertions(+), 55 deletions(-)
diff --git a/configure.ac
Gallium code used HAVE_LLVM to check if it needs to compile code for
LLVM in header and source files.
With the new logic HAVE_LLVM is always set. Use extra define to figure
out if LLVM is used.
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=99010
Signed-off-by: Tobias Droste
---
config
This is actually not needed because the version is checked later.
Line 2609:
if test "x$enable_gallium_llvm" == "xyes"; then
llvm_require_version $LLVM_REQUIRED_GALLIUM "gallium"
llvm_add_default_components "gallium"
HAVE_GALLIUM_LLVM=xyes
DEFINES="${DEFINES} -DHAVE_GALLIUM_LLVM"
this renames MESA_LLVM to FOUND_LLVM and updates the config.log report
to say if LLVM is found or not, to make clear that this does not mean
that it is used.
Signed-off-by: Tobias Droste
---
configure.ac | 21 ++---
1 file changed, 10 insertions(+), 11 deletions(-)
diff --git a/
Patches 1 to 4 (hopefully) fix
https://bugs.freedesktop.org/show_bug.cgi?id=99010.
Patch 5 and 6 are cosmetic changes.
I build and run tested this with radeonsi, radv and llvmpipe and build tested
with softpipe.
@Emil:
Please test and push if everything is ok.
Tobias Droste (6):
configure.a
Hi all,
Sorry for the slow response. I think it is fine to drop the driver :(
Not because the driver is currently unmaintained, which is very true
and is a very good reason, but that there is now a Intel Vulkan
driver. Vulkan is somewhat as low-level as Gallium is (or even
lower-level). The dr
On 07/12/16 07:16 PM, Nicolai Hähnle wrote:
> On 07.12.2016 08:50, Michel Dänzer wrote:
>> On 06/12/16 10:24 PM, Marek Olšák wrote:
>>> On Mon, Dec 5, 2016 at 10:05 AM, Michel Dänzer
>>> wrote:
On 03/12/16 05:38 AM, Marek Olšák wrote:
> From: Marek Olšák
>
> This fixes random rad
v2: use fmul(1/65536) instead of fdiv(65535)
---
.../drivers/swr/rasterizer/common/formats.cpp | 104 ++---
.../drivers/swr/rasterizer/common/formats.h| 7 +-
.../drivers/swr/rasterizer/core/format_traits.h| 90 +-
.../drivers/swr/rasterizer/jitt
Reviewed-by: Edward O'Callaghan
On 12/08/2016 11:55 AM, Dave Airlie wrote:
> From: Dave Airlie
>
> We don't use the nir shader in here at all.
>
> Signed-off-by: Dave Airlie
> ---
> src/amd/common/ac_nir_to_llvm.c | 15 ++-
> 1 file changed, 6 insertions(+), 9 deletions(-)
>
> d
+author again.. no idea why list keeps loosing extra cc's.. it is
kinda annoying..
BR,
-R
On Wed, Dec 7, 2016 at 7:28 PM, Roland Scheidegger wrote:
> I haven't seen the driver author's opinion on this yet, so it's probably
> fair to give him some more time to answer. It's not like this is really
From: Dave Airlie
We don't use the nir shader in here at all.
Signed-off-by: Dave Airlie
---
src/amd/common/ac_nir_to_llvm.c | 15 ++-
1 file changed, 6 insertions(+), 9 deletions(-)
diff --git a/src/amd/common/ac_nir_to_llvm.c b/src/amd/common/ac_nir_to_llvm.c
index bb52e8f..f5f7
Signed-off-by: Eric Engestrom
---
src/egl/main/eglapi.c | 18 +-
1 file changed, 9 insertions(+), 9 deletions(-)
diff --git a/src/egl/main/eglapi.c b/src/egl/main/eglapi.c
index 9950a72..925e881 100644
--- a/src/egl/main/eglapi.c
+++ b/src/egl/main/eglapi.c
@@ -890,7 +890,7 @@ eg
Nobody else makes use of this function.
We can always re-export it if someone ever needs it.
Signed-off-by: Eric Engestrom
---
src/egl/main/eglapi.c | 2 +-
src/egl/main/eglapi.h | 3 ---
2 files changed, 1 insertion(+), 4 deletions(-)
diff --git a/src/egl/main/eglapi.c b/src/egl/main/eglapi.c
On 12/08/2016 11:28 AM, Roland Scheidegger wrote:
> I haven't seen the driver author's opinion on this yet, so it's probably
> fair to give him some more time to answer. It's not like this is really
> urgent...
Absolutely!
>
> Roland
>
> Am 08.12.2016 um 01:11 schrieb Edward O'Callaghan:
>> H
No functional change, just rewriting it in an easier-to-understand way.
Signed-off-by: Eric Engestrom
---
src/egl/drivers/dri2/platform_x11.c | 24 ++--
1 file changed, 10 insertions(+), 14 deletions(-)
diff --git a/src/egl/drivers/dri2/platform_x11.c
b/src/egl/drivers/dri2
This series is,
Reviewed-by: Edward O'Callaghan
On 12/08/2016 11:18 AM, Dave Airlie wrote:
> While working on spilling, geom shaders and compute queues, I've built
> up a bunch of cleanups, so best to dequeue them early.
>
> Dave.
>
> ___
> mesa-dev m
I haven't seen the driver author's opinion on this yet, so it's probably
fair to give him some more time to answer. It's not like this is really
urgent...
Roland
Am 08.12.2016 um 01:11 schrieb Edward O'Callaghan:
> Hi all,
>
> So I'll get right to the crux of this; In summary the consensus would
Give explicit names to the various memory heaps, reducing the risk of
bugs when adding or modifying them.
While at it, use a static_assert to make sure we don't have too many
memory heaps.
Signed-off-by: Eric Engestrom
---
I'm sure there are better names than these, but this patch is here to
su
Following the previous patch, give explicit names to the various memory types,
reducing the risk of bugs when adding or modifying them.
While at it, use a static_assert to make sure we don't have too many
memory types.
Signed-off-by: Eric Engestrom
---
This one isn't really useful yet, but I th
From: Dave Airlie
This just passes the subresource info in and uses it.
Signed-off-by: Dave Airlie
---
src/amd/vulkan/radv_meta_copy.c | 32 +---
1 file changed, 13 insertions(+), 19 deletions(-)
diff --git a/src/amd/vulkan/radv_meta_copy.c b/src/amd/vulkan/radv_me
From: Dave Airlie
Signed-off-by: Dave Airlie
---
src/amd/vulkan/radv_meta_blit2d.c | 3 +--
1 file changed, 1 insertion(+), 2 deletions(-)
diff --git a/src/amd/vulkan/radv_meta_blit2d.c
b/src/amd/vulkan/radv_meta_blit2d.c
index 6e92f80..5a42fd8 100644
--- a/src/amd/vulkan/radv_meta_blit2d.c
+
From: Dave Airlie
This makes it easier to add new params before the array ones.
Signed-off-by: Dave Airlie
---
src/amd/common/ac_nir_to_llvm.c | 11 ++-
1 file changed, 6 insertions(+), 5 deletions(-)
diff --git a/src/amd/common/ac_nir_to_llvm.c b/src/amd/common/ac_nir_to_llvm.c
index
From: Dave Airlie
This isn't needed for later things like geom shader copy shaders,
we won't have NIR.
Signed-off-by: Dave Airlie
---
src/amd/common/ac_nir_to_llvm.c | 13 ++---
1 file changed, 6 insertions(+), 7 deletions(-)
diff --git a/src/amd/common/ac_nir_to_llvm.c b/src/amd/comm
From: Dave Airlie
Just noticed this file was missing license and any
explaination of what is in it.
Signed-off-by: Dave Airlie
---
src/amd/vulkan/radv_meta_bufimage.c | 27 +++
1 file changed, 27 insertions(+)
diff --git a/src/amd/vulkan/radv_meta_bufimage.c
b/src/amd
From: Dave Airlie
This code will have use for copy shaders etc.
Signed-off-by: Dave Airlie
---
src/amd/vulkan/radv_pipeline.c | 76 --
1 file changed, 43 insertions(+), 33 deletions(-)
diff --git a/src/amd/vulkan/radv_pipeline.c b/src/amd/vulkan/radv_pi
From: Dave Airlie
Signed-off-by: Dave Airlie
---
src/amd/common/ac_nir_to_llvm.c | 3 ---
1 file changed, 3 deletions(-)
diff --git a/src/amd/common/ac_nir_to_llvm.c b/src/amd/common/ac_nir_to_llvm.c
index 5820e62..c2a74d4 100644
--- a/src/amd/common/ac_nir_to_llvm.c
+++ b/src/amd/common/ac_ni
From: Dave Airlie
Removes some unnecessary functions and pull
some stuff out of the loop.
Signed-off-by: Dave Airlie
---
src/amd/vulkan/radv_meta_bufimage.c | 42 +++--
1 file changed, 7 insertions(+), 35 deletions(-)
diff --git a/src/amd/vulkan/radv_meta_bufim
From: Dave Airlie
This make it easier to add multiple queues later.
Signed-off-by: Dave Airlie
---
src/amd/vulkan/radv_meta_copy.c | 34 +++---
1 file changed, 23 insertions(+), 11 deletions(-)
diff --git a/src/amd/vulkan/radv_meta_copy.c b/src/amd/vulkan/radv_meta
From: Dave Airlie
Signed-off-by: Dave Airlie
---
src/amd/common/ac_nir_to_llvm.c | 14 +-
1 file changed, 5 insertions(+), 9 deletions(-)
diff --git a/src/amd/common/ac_nir_to_llvm.c b/src/amd/common/ac_nir_to_llvm.c
index f644980..0ad9a7a 100644
--- a/src/amd/common/ac_nir_to_llvm
While working on spilling, geom shaders and compute queues, I've built
up a bunch of cleanups, so best to dequeue them early.
Dave.
___
mesa-dev mailing list
mesa-dev@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/mesa-dev
On 08/12/2016 00:38, Dieter Nützel wrote:
Am 07.12.2016 23:54, schrieb Axel Davy:
This patch adds internal multithreading to gallium nine.
The goal is to offload almost all gallium nine calls (and some other
work) to a worker thread.
Hello Axel,
really nice job!
'But' show us the 'numbers'..
Hi all,
So I'll get right to the crux of this; In summary the consensus would
then be to drop ilo?
If so, I am not sure of this communities procedure? However, if it helps
the patch is here:
https://cgit.freedesktop.org/~funfunctor/mesa/log/?h=eol-ilo
Kind Regards,
Edward.
On 12/07/2016 07:08 A
On 12/08/2016 02:53 AM, Emil Velikov wrote:
> On 6 December 2016 at 22:34, Edward O'Callaghan
> wrote:
>> On 12/07/2016 12:36 AM, Emil Velikov wrote:
>>> On 6 December 2016 at 11:30, Edward O'Callaghan
>>> wrote:
As per the C spec, it is illegal to alias pointers to different
types. T
---
.../drivers/swr/rasterizer/common/formats.cpp | 104 ++---
.../drivers/swr/rasterizer/common/formats.h| 7 +-
.../drivers/swr/rasterizer/core/format_traits.h| 90 +-
.../drivers/swr/rasterizer/jitter/fetch_jit.cpp| 12 +++
src/gallium/dr
This one-off specialization was missed.
---
src/gallium/drivers/swr/rasterizer/core/backend.h | 4 +-
.../drivers/swr/rasterizer/memory/StoreTile.h | 46 +++---
2 files changed, 24 insertions(+), 26 deletions(-)
diff --git a/src/gallium/drivers/swr/rasterizer/core/backend.h
Am 07.12.2016 23:54, schrieb Axel Davy:
This patch adds internal multithreading to gallium nine.
The goal is to offload almost all gallium nine calls (and some other
work) to a worker thread.
Hello Axel,
really nice job!
'But' show us the 'numbers'...
...do you have some numbers of hand?
Tha
Count properly the number of backbuffers,
and use the new info to release the correct
number of buffers
Signed-off-by: Axel Davy
---
src/gallium/state_trackers/nine/swapchain9.c | 31 ++--
src/gallium/state_trackers/nine/swapchain9.h | 1 +
2 files changed, 16 insertions
It was unused.
Signed-off-by: Axel Davy
---
src/gallium/state_trackers/nine/nine_state.h | 1 -
src/gallium/state_trackers/nine/stateblock9.c | 5 -
2 files changed, 6 deletions(-)
diff --git a/src/gallium/state_trackers/nine/nine_state.h
b/src/gallium/state_trackers/nine/nine_state.h
ind
pixels stateblocks need to capture these.
Signed-off-by: Axel Davy
---
src/gallium/state_trackers/nine/device9.c | 5 -
1 file changed, 4 insertions(+), 1 deletion(-)
diff --git a/src/gallium/state_trackers/nine/device9.c
b/src/gallium/state_trackers/nine/device9.c
index 5b4d2ae..8b03e9b 1
Apparently specular enable doesn't affect
the alpha channel.
Fixes https://github.com/iXit/Mesa-3D/issues/253
Behaviour comfirmed looking in wine sources.
Signed-off-by: Axel Davy
---
src/gallium/state_trackers/nine/nine_ff.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/
Use the new API for DISCARD.
Signed-off-by: Axel Davy
---
src/gallium/state_trackers/nine/swapchain9.c | 180 ---
1 file changed, 132 insertions(+), 48 deletions(-)
diff --git a/src/gallium/state_trackers/nine/swapchain9.c
b/src/gallium/state_trackers/nine/swapchain9.c
Current is already initialized to Diffuse.
Signed-off-by: Axel Davy
---
src/gallium/state_trackers/nine/nine_ff.c | 5 -
1 file changed, 5 deletions(-)
diff --git a/src/gallium/state_trackers/nine/nine_ff.c
b/src/gallium/state_trackers/nine/nine_ff.c
index 7a09986..c3245a3 100644
--- a/src
This is useful to check what is exactly recorded.
Signed-off-by: Axel Davy
---
src/gallium/state_trackers/nine/stateblock9.c | 5 +
1 file changed, 5 insertions(+)
diff --git a/src/gallium/state_trackers/nine/stateblock9.c
b/src/gallium/state_trackers/nine/stateblock9.c
index a2b281e..fbbd
The computation was wrong. COLOR0's last component
should be equal to the material diffuse w component.
The behaviour was checked with a test app on Windows.
Wine has the same behaviour.
Signed-off-by: Axel Davy
---
src/gallium/state_trackers/nine/nine_ff.c | 31 ++-
We disable the alpha stage if the color stage is disabled.
Signed-off-by: Axel Davy
---
src/gallium/state_trackers/nine/nine_ff.c | 5 +++--
1 file changed, 3 insertions(+), 2 deletions(-)
diff --git a/src/gallium/state_trackers/nine/nine_ff.c
b/src/gallium/state_trackers/nine/nine_ff.c
index
Fix the check for computing ff specular.
This seems to match the opengl behavior,
and give the correct output on windows.
Signed-off-by: Axel Davy
---
src/gallium/state_trackers/nine/nine_ff.c | 9 +++--
1 file changed, 7 insertions(+), 2 deletions(-)
diff --git a/src/gallium/state_tracker
We were advertising support for WFOG (like all win drivers),
but we weren't implementing it.
This patch implements the behaviour. See comments.
Signed-off-by: Axel Davy
---
src/gallium/state_trackers/nine/nine_ff.c | 42 +--
1 file changed, 35 insertions(+), 7 deletio
We were ignoring MULTISAMPLEMASK for non-maskable multisample
modes, but we were missing the non-multisampled case.
Fixes a crash in Halo.
Signed-off-by: Axel Davy
---
src/gallium/state_trackers/nine/device9.c| 4 ++--
src/gallium/state_trackers/nine/nine_state.c | 2 +-
2 files changed, 3
In non-debug build, the variables are unused,
and thus trigger a compilation warning.
Signed-off-by: Axel Davy
---
src/gallium/state_trackers/nine/iunknown.c | 8
src/gallium/state_trackers/nine/surface9.c | 2 ++
src/gallium/state_trackers/nine/volume9.c | 2 ++
3 files changed, 12 in
See the patch for the new controls added.
Signed-off-by: Axel Davy
---
src/gallium/state_trackers/nine/adapter9.h | 2 ++
src/gallium/state_trackers/nine/swapchain9.c| 4 ++--
src/gallium/targets/d3dadapter9/drm.c | 17 +
src/mesa/drivers/dri/common/xmlpool/t
Some states were not properly recorded in pixel stateblocks.
Signed-off-by: Axel Davy
---
src/gallium/state_trackers/nine/device9.c | 5 -
1 file changed, 4 insertions(+), 1 deletion(-)
diff --git a/src/gallium/state_trackers/nine/device9.c
b/src/gallium/state_trackers/nine/device9.c
index
This serie fixes several gallium nine bugs.
There is also a few cleanups.
It also has a new feature: new options are added
to choose the behaviour of the swapchain DISCARD
mode (which is the most common mode).
Basically the default is that when vsync is off,
the frames are presented as soon as pos
Check value validity and mimick Win behaviour.
Signed-off-by: Axel Davy
---
src/gallium/state_trackers/nine/device9.c | 9 ---
src/gallium/state_trackers/nine/nine_limits.h | 36 +++
src/gallium/state_trackers/nine/nine_state.c | 2 ++
src/gallium/state_tracker
The new code properly releases the previous buffers
allocated.
Signed-off-by: Axel Davy
---
src/gallium/state_trackers/nine/nine_ff.c| 22 +++--
src/gallium/state_trackers/nine/nine_state.c | 46
2 files changed, 39 insertions(+), 29 deletions(-)
diff --
SetCursor assumes for now a 32x32 argb cursor with pitch 128.
32x32 argb doesn't have pitch 128 on all hw, thus use a
temporary surface with the correct pitch when needed.
Signed-off-by: Axel Davy
---
src/gallium/state_trackers/nine/device9.c | 20 ++--
src/gallium/state_trackers
In NineDevice9_SetDefaultState we can assume the
implicit swapchain is properly initialized.
Signed-off-by: Axel Davy
---
src/gallium/state_trackers/nine/device9.c | 21 +++--
1 file changed, 7 insertions(+), 14 deletions(-)
diff --git a/src/gallium/state_trackers/nine/device9.c
When a texture is read by a stage for colorop, it should
be disabled, and disable following stages.
When a texture is read for alphaop, 1.0f is read for the input,
which is the behaviour for a dummy texture.
Signed-off-by: Axel Davy
---
src/gallium/state_trackers/nine/nine_ff.c | 22 +++
The check was not catching all possible cases.
NVE4 should be fine.
Signed-off-by: Axel Davy
---
src/gallium/state_trackers/nine/nine_ff.c | 25 +
1 file changed, 1 insertion(+), 24 deletions(-)
diff --git a/src/gallium/state_trackers/nine/nine_ff.c
b/src/gallium/state_
The API will enable better support for the commonly
used DISCARD swapchain parameter.
Signed-off-by: Axel Davy
---
include/d3dadapter/present.h | 24
1 file changed, 24 insertions(+)
diff --git a/include/d3dadapter/present.h b/include/d3dadapter/present.h
index 187843f.
The code was wrongly detecting which texture coordinates
to generate when the coordinate index was different to
the stage index.
Signed-off-by: Axel Davy
---
src/gallium/state_trackers/nine/nine_ff.c | 9 +
1 file changed, 5 insertions(+), 4 deletions(-)
diff --git a/src/gallium/state_t
Add precision on the behaviour of WaitBufferReleased.
All implementers and users of the API were expecting
that behaviour.
Signed-off-by: Axel Davy
---
include/d3dadapter/present.h | 5 -
1 file changed, 4 insertions(+), 1 deletion(-)
diff --git a/include/d3dadapter/present.h b/include/d3da
Some drivers crash if NOP is generated.
Besides there is no point to generate NOP.
Signed-off-by: Axel Davy
---
src/gallium/state_trackers/nine/nine_shader.c | 9 -
1 file changed, 8 insertions(+), 1 deletion(-)
diff --git a/src/gallium/state_trackers/nine/nine_shader.c
b/src/gallium/s
Fixes bad rendering of a test app.
Wine has the same behaviour.
Signed-off-by: Axel Davy
---
src/gallium/state_trackers/nine/nine_ff.c | 4
1 file changed, 4 deletions(-)
diff --git a/src/gallium/state_trackers/nine/nine_ff.c
b/src/gallium/state_trackers/nine/nine_ff.c
index 8b6f442..a1eb
The variable is used only in the hw path.
Signed-off-by: Axel Davy
---
src/gallium/state_trackers/nine/nine_state.c | 2 --
1 file changed, 2 deletions(-)
diff --git a/src/gallium/state_trackers/nine/nine_state.c
b/src/gallium/state_trackers/nine/nine_state.c
index 57f7b2d..69f0cff 100644
---
if first alpha stage is disabled and writes to temp,
diffuse alpha is written to temp.
Last stage always writes to current.
Behaviour was deduced by tests with a test app.
Signed-off-by: Axel Davy
---
src/gallium/state_trackers/nine/nine_ff.c | 20
1 file changed, 20 insert
The cap should be advertised for d3d10 able cards,
but only for Ex contexts.
Unfortunately at this point Mesa has no way to know if
Ex is used or not (the info is got later).
Signed-off-by: Axel Davy
---
src/gallium/state_trackers/nine/adapter9.c | 2 +-
1 file changed, 1 insertion(+), 1 deleti
src was initialized instead of dst.
Signed-off-by: Axel Davy
---
src/gallium/state_trackers/nine/stateblock9.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/src/gallium/state_trackers/nine/stateblock9.c
b/src/gallium/state_trackers/nine/stateblock9.c
index 16e60ef..a2b
Ignore empty draw calls.
Avoid assertion fault when such draw calls happen
in u_upload_mgr.
Signed-off-by: Axel Davy
---
src/gallium/state_trackers/nine/device9.c | 2 ++
1 file changed, 2 insertions(+)
diff --git a/src/gallium/state_trackers/nine/device9.c
b/src/gallium/state_trackers/nine/de
Textures already have the check in BaseTexture9.
Non-Textures cannot be in the MANAGED Pool.
Signed-off-by: Axel Davy
---
src/gallium/state_trackers/nine/surface9.c | 4
1 file changed, 4 deletions(-)
diff --git a/src/gallium/state_trackers/nine/surface9.c
b/src/gallium/state_trackers/nin
DRI_CONF_NINE_OVERRIDEVENDOR was missing gettext for the
description.
Signed-off-by: Axel Davy
---
src/mesa/drivers/dri/common/xmlpool/t_options.h | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/src/mesa/drivers/dri/common/xmlpool/t_options.h
b/src/mesa/drivers/dri/common/xm
We want to use centroid for nonmaskable
multisampling as well.
Signed-off-by: Axel Davy
---
src/gallium/state_trackers/nine/pixelshader9.h | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/src/gallium/state_trackers/nine/pixelshader9.h
b/src/gallium/state_trackers/nine/pixelsh
cursor.image is used for software cursor
emulation. It wasn't released.
Signed-off-by: Axel Davy
---
src/gallium/state_trackers/nine/device9.c | 1 +
1 file changed, 1 insertion(+)
diff --git a/src/gallium/state_trackers/nine/device9.c
b/src/gallium/state_trackers/nine/device9.c
index a78d18e.
The call was not needed.
Signed-off-by: Axel Davy
---
src/gallium/state_trackers/nine/device9.c | 1 -
1 file changed, 1 deletion(-)
diff --git a/src/gallium/state_trackers/nine/device9.c
b/src/gallium/state_trackers/nine/device9.c
index 055e71d..7bdf19f 100644
--- a/src/gallium/state_trackers
To offload mipmap generation as well.
Signed-off-by: Axel Davy
---
src/gallium/state_trackers/nine/nine_state.c | 15 +++
src/gallium/state_trackers/nine/nine_state.h | 7 +++
2 files changed, 22 insertions(+)
diff --git a/src/gallium/state_trackers/nine/nine_state.c
b/src/gal
The last level was not released.
Signed-off-by: Axel Davy
---
src/gallium/state_trackers/nine/volumetexture9.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/src/gallium/state_trackers/nine/volumetexture9.c
b/src/gallium/state_trackers/nine/volumetexture9.c
index 4908ed7..5
Do the upload in the other thread.
Usually managed buffers are used once per frame.
It is then very likely pending_upload is 0 at Lock
time.
Signed-off-by: Axel Davy
---
src/gallium/state_trackers/nine/buffer9.c | 3 +++
src/gallium/state_trackers/nine/buffer9.h | 12 +++-
2 files chan
This new buffer upload path enables to lock
faster than the normal path when using
DISCARD/NOOVERWRITE.
Signed-off-by: Axel Davy
---
src/gallium/state_trackers/nine/Makefile.sources | 2 +
src/gallium/state_trackers/nine/buffer9.c | 75 +-
src/gallium/state_trackers/nine/buffer
The secondary pipe will be used for operations
that don't need synchronization.
Signed-off-by: Axel Davy
---
src/gallium/state_trackers/nine/device9.c | 4 +++-
src/gallium/state_trackers/nine/device9.h | 2 ++
2 files changed, 5 insertions(+), 1 deletion(-)
diff --git a/src/gallium/state_track
Use the acquire/release semantic when we don't need
to wait for any pending command.
Signed-off-by: Axel Davy
---
src/gallium/state_trackers/nine/basetexture9.c | 8 ++--
src/gallium/state_trackers/nine/buffer9.c | 4 +++-
src/gallium/state_trackers/nine/nine_state.c| 16 +++
Will be used to upload buffers.
Signed-off-by: Axel Davy
---
src/gallium/state_trackers/nine/nine_state.c | 11 +++
src/gallium/state_trackers/nine/nine_state.h | 8
2 files changed, 19 insertions(+)
diff --git a/src/gallium/state_trackers/nine/nine_state.c
b/src/gallium/stat
If the volumes (and the texture container) are not referenced,
then they are no pending operations on them. We can lock directly.
Signed-off-by: Axel Davy
---
src/gallium/state_trackers/nine/volume9.c | 9 -
1 file changed, 8 insertions(+), 1 deletion(-)
diff --git a/src/gallium/state_t
If the surfaces (and the texture container) are not referenced,
then they are no pending operations on them. We can lock directly.
Signed-off-by: Axel Davy
---
src/gallium/state_trackers/nine/surface9.c | 12 +++-
1 file changed, 11 insertions(+), 1 deletion(-)
diff --git a/src/gallium/
Use nine_context_box_upload for uploads:
. systemmem surface to default surface
. managed surface internal content to its resource.
Check the uploads are executed before any action
that can alter the data, that is LockRect,
NineSurface9_CopyDefaultToMem and surface destruction.
Signed-off-by: Axe
Next patches will introduce an offset.
Signed-off-by: Axel Davy
---
src/gallium/state_trackers/nine/nine_state.c| 15 +--
src/gallium/state_trackers/nine/vertexbuffer9.c | 3 ++-
src/gallium/state_trackers/nine/vertexbuffer9.h | 2 +-
3 files changed, 12 insertions(+), 8 deleti
Enables to not wait for the worker thread for ColorFill.
Signed-off-by: Axel Davy
---
src/gallium/state_trackers/nine/device9.c| 7 +--
src/gallium/state_trackers/nine/nine_state.c | 17 +
src/gallium/state_trackers/nine/nine_state.h | 9 +
src/gallium/state_tra
Use nine_context_box_upload for uploads:
. systemmem volume to default volume
. managed volume internal content to its resource.
Check the uploads are executed before any action
that can alter the data, that is LockBox and
volume destruction.
Signed-off-by: Axel Davy
---
src/gallium/state_track
BackBuffers can survive swapchain dtor if
the user has a reference on them.
The swapchain itself has no reference on the buffer.
Signed-off-by: Axel Davy
---
src/gallium/state_trackers/nine/swapchain9.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/src/gallium/state_tracke
This patch it to avoid name conflict when device->pipe
will be moved to nine_context.
Signed-off-by: Axel Davy
---
src/gallium/state_trackers/nine/nine_ff.c| 20 +++
src/gallium/state_trackers/nine/nine_state.c | 90 ++--
src/gallium/state_trackers/nine/nine_state
The new arguments enable to reference the objects while
the function hasn't run.
Signed-off-by: Axel Davy
---
src/gallium/state_trackers/nine/device9.c| 14 --
src/gallium/state_trackers/nine/nine_state.c | 10 ++
src/gallium/state_trackers/nine/nine_state.h | 4
3
1 - 100 of 242 matches
Mail list logo