Reviewed-by: Bas Nieuwenhuizen
On Thu, May 4, 2017 at 2:52 AM, Dave Airlie wrote:
> From: Dave Airlie
>
> There is no reason to advertise transfer ability for formats we can't
> use for anything else. This stops some CTS tests hitting internal
> error for 64-bit types when they see the transfer
---
src/mapi/glapi/gen/ARB_direct_state_access.xml | 2 +-
src/mapi/glapi/gen/ARB_map_buffer_range.xml| 2 +-
src/mesa/main/bufferobj.c | 25 +
src/mesa/main/bufferobj.h | 7 ++-
4 files changed, 33 insertions(+), 3 delet
This will allow us to add KHR_no_error support for *BufferRange
functions.
---
src/mesa/main/bufferobj.c | 77 ++-
1 file changed, 50 insertions(+), 27 deletions(-)
diff --git a/src/mesa/main/bufferobj.c b/src/mesa/main/bufferobj.c
index 76e035f..797b6f
---
src/mesa/main/bufferobj.c | 27 +--
src/mesa/main/bufferobj.h | 6 --
2 files changed, 13 insertions(+), 20 deletions(-)
diff --git a/src/mesa/main/bufferobj.c b/src/mesa/main/bufferobj.c
index 961871c..76e035f 100644
--- a/src/mesa/main/bufferobj.c
+++ b/src/mesa
---
src/mesa/main/bufferobj.c | 15 +++
src/mesa/main/bufferobj.h | 5 -
2 files changed, 7 insertions(+), 13 deletions(-)
diff --git a/src/mesa/main/bufferobj.c b/src/mesa/main/bufferobj.c
index 0a8f030..6161271 100644
--- a/src/mesa/main/bufferobj.c
+++ b/src/mesa/main/bufferob
---
src/mesa/main/bufferobj.c | 33 +
src/mesa/main/bufferobj.h | 9 -
2 files changed, 13 insertions(+), 29 deletions(-)
diff --git a/src/mesa/main/bufferobj.c b/src/mesa/main/bufferobj.c
index 9e96f31..326e901 100644
--- a/src/mesa/main/bufferobj.c
+++ b
---
src/mapi/glapi/gen/ARB_direct_state_access.xml | 2 +-
src/mapi/glapi/gen/gl_API.xml | 2 +-
src/mesa/main/bufferobj.c | 19 +++
src/mesa/main/bufferobj.h | 4
4 files changed, 25 insertions(+), 2 deletions(-)
---
src/mapi/glapi/gen/ARB_direct_state_access.xml | 4 +-
src/mapi/glapi/gen/ARB_map_buffer_range.xml| 2 +-
src/mapi/glapi/gen/gl_API.xml | 2 +-
src/mesa/main/bufferobj.c | 53 ++
src/mesa/main/bufferobj.h
---
src/mesa/main/bufferobj.c | 31 ---
src/mesa/main/bufferobj.h | 3 ---
2 files changed, 34 deletions(-)
diff --git a/src/mesa/main/bufferobj.c b/src/mesa/main/bufferobj.c
index 6161271..2944dce 100644
--- a/src/mesa/main/bufferobj.c
+++ b/src/mesa/main/bufferobj.c
After calling this we were then overriding all the functions with
st versions.
---
src/mesa/state_tracker/st_cb_bufferobjects.c | 3 ---
1 file changed, 3 deletions(-)
diff --git a/src/mesa/state_tracker/st_cb_bufferobjects.c
b/src/mesa/state_tracker/st_cb_bufferobjects.c
index 5911d1e..c1994d5
This will allow us to implement KHR_no_error support for unmap
functions.
---
src/mesa/main/bufferobj.c | 29 +
1 file changed, 17 insertions(+), 12 deletions(-)
diff --git a/src/mesa/main/bufferobj.c b/src/mesa/main/bufferobj.c
index f6822c7..0fdb623 100644
--- a/src/
---
src/mesa/main/bufferobj.c | 18 --
src/mesa/main/bufferobj.h | 7 ---
2 files changed, 8 insertions(+), 17 deletions(-)
diff --git a/src/mesa/main/bufferobj.c b/src/mesa/main/bufferobj.c
index 326e901..0a8f030 100644
--- a/src/mesa/main/bufferobj.c
+++ b/src/mesa/main/buf
---
src/mesa/main/bufferobj.c | 10 +-
src/mesa/main/bufferobj.h | 4
2 files changed, 5 insertions(+), 9 deletions(-)
diff --git a/src/mesa/main/bufferobj.c b/src/mesa/main/bufferobj.c
index d1726d6..f6822c7 100644
--- a/src/mesa/main/bufferobj.c
+++ b/src/mesa/main/bufferobj.c
@@
---
src/mesa/main/bufferobj.c | 18 +-
src/mesa/main/bufferobj.h | 6 --
2 files changed, 9 insertions(+), 15 deletions(-)
diff --git a/src/mesa/main/bufferobj.c b/src/mesa/main/bufferobj.c
index 6fa32e9..75526b5 100644
--- a/src/mesa/main/bufferobj.c
+++ b/src/mesa/main/buff
---
src/mesa/main/bufferobj.c | 8
1 file changed, 4 insertions(+), 4 deletions(-)
diff --git a/src/mesa/main/bufferobj.c b/src/mesa/main/bufferobj.c
index 2944dce..6825030 100644
--- a/src/mesa/main/bufferobj.c
+++ b/src/mesa/main/bufferobj.c
@@ -649,23 +649,23 @@ buffer_sub_data_fallba
---
src/mesa/main/texobj.c | 9 -
1 file changed, 4 insertions(+), 5 deletions(-)
diff --git a/src/mesa/main/texobj.c b/src/mesa/main/texobj.c
index af9baa9..868e4eb 100644
--- a/src/mesa/main/texobj.c
+++ b/src/mesa/main/texobj.c
@@ -322,27 +322,27 @@ _mesa_initialize_texture_object( str
---
src/mesa/main/bufferobj.c | 34 ++
1 file changed, 10 insertions(+), 24 deletions(-)
diff --git a/src/mesa/main/bufferobj.c b/src/mesa/main/bufferobj.c
index f4597ed..44618a3 100644
--- a/src/mesa/main/bufferobj.c
+++ b/src/mesa/main/bufferobj.c
@@ -1176,23 +11
---
src/mapi/glapi/gen/ARB_copy_buffer.xml | 2 +-
src/mapi/glapi/gen/ARB_direct_state_access.xml | 2 +-
src/mesa/main/bufferobj.c | 33 ++
src/mesa/main/bufferobj.h | 9 ++-
4 files changed, 43 insertions(+), 3 dele
---
src/mesa/main/texparam.c | 13 ++---
1 file changed, 6 insertions(+), 7 deletions(-)
diff --git a/src/mesa/main/texparam.c b/src/mesa/main/texparam.c
index 25165e4..0156bbd 100644
--- a/src/mesa/main/texparam.c
+++ b/src/mesa/main/texparam.c
@@ -1398,30 +1398,29 @@ get_tex_level_param
On Wed, 2017-04-26 at 07:35 -0700, Jason Ekstrand wrote:
> Previously, the maximum size of a state that could be allocated from a
> state pool was a block. However, this has caused us various issues
> particularly with shaders which are potentially very large. We've also
> hit issues with render
Reviewed-by: Juan A. Suarez Romero
On Wed, 2017-04-26 at 10:01 -0700, Jason Ekstrand wrote:
> ---
> src/intel/vulkan/anv_allocator.c | 10 +++---
> 1 file changed, 7 insertions(+), 3 deletions(-)
>
> diff --git a/src/intel/vulkan/anv_allocator.c
> b/src/intel/vulkan/anv_allocator.c
> inde
On Tue, 2017-05-02 at 06:44 -0700, Jason Ekstrand wrote:
> Juan,
>
> Were you planning to review the rest of the series? Just wondering.
>
Sorry, it took a bit more to review the remaining patches.
Now it is done.
J.A.
> --Jason
>
>
> On April 26, 2017 7:35:29 AM Jason Ekstrand
On Wed, 2017-05-03 at 14:37 +0100, Emil Velikov wrote:
> On 3 May 2017 at 14:26, Samuel Iglesias Gonsálvez om> wrote:
> > On Wed, 2017-05-03 at 14:15 +0100, Emil Velikov wrote:
> > > On 3 May 2017 at 12:33, Samuel Iglesias Gonsálvez > > ia.c
> > > om> wrote:
> > > > On Wed, 2017-05-03 at 11:50 +0
On Wed, 2017-05-03 at 09:44 -0700, Jason Ekstrand wrote:
> ---
> src/intel/vulkan/anv_allocator.c | 11 +--
> 1 file changed, 5 insertions(+), 6 deletions(-)
>
> diff --git a/src/intel/vulkan/anv_allocator.c
> b/src/intel/vulkan/anv_allocator.c
> index d7b86c8..40913ec 100644
> --- a/src
On Mon, May 01, 2017 at 06:43:23PM -0700, Rafael Antognolli wrote:
> Some code that was placed in brw_draw_upload.c and exported to be used
> by gen8+ was also moved to genX_state_upload, and the respective symbols
> are not exported anymore.
>
> v2:
>- Remove code from brw_draw_upload too
>
On 05/04/2017 02:36 AM, Timothy Arceri wrote:
This and the tgsi copy_propagation pass are very slow, I'd really like
it if we both didn't require the pass for things to work and also didn't
make it any slower. perf is showing these 2 passes at ~11% during Deus
Ex start-up (cold cache).
This
On Mon, May 01, 2017 at 06:43:25PM -0700, Rafael Antognolli wrote:
> Signed-off-by: Rafael Antognolli
Reviewed-by: Topi Pohjolainen
> ---
> src/mesa/drivers/dri/i965/Makefile.sources| 1 +-
> src/mesa/drivers/dri/i965/brw_state.h | 1 +-
> src/mesa/drivers/dri/i965/gen6_cc.c
On Wed, 2017-05-03 at 16:47 -0700, Francisco Jerez wrote:
> Samuel Iglesias Gonsálvez writes:
>
> > Reorder the uniforms to load first the dvec4-aligned variables
> > in the push constant buffer and then push the vec4-aligned ones.
> >
> > This fixes a bug were the dvec3/4 might be loaded one pa
On Mon, May 01, 2017 at 06:43:26PM -0700, Rafael Antognolli wrote:
> On this patch, we port:
>- brw_polygon_stipple
>- brw_polygon_stipple_offset
>- brw_line_stipple
>- brw_drawing_rect
>
> v2:
>- Also emit states for gen4-5 with this code.
> v3:
>- Style fixes and remove e
> hopefully this[1] will eventually happen, which would make this less
> of an issue :-)
>
> [1] http://lists.llvm.org/pipermail/llvm-dev/2017-May/112538.html
Indeed! I have seen that thread and made sure to subscribe to the ML not to
miss anything there.
_
The TGSI DCE pass doesn't eliminate dead assignments like
MOV TEMP[0], TEMP[1] in presence of loops because it assumes
that the visitor doesn't emit dead code. This assumption is
actually wrong and this situation happens.
However, it appears that the merge_registers() pass accidentally
takes care
Use a helper function and struct to convert between a modifier and
tiling mode, so we can use it later for a tiling -> modifier lookup.
Signed-off-by: Daniel Stone
Cc: Ben Widawsky
---
src/mesa/drivers/dri/i965/intel_screen.c | 49 +---
1 file changed, 32 insertions(
When a buffer is being created from FD or GEM flink import, the current
API makes no provision for passing modifier information along with this.
Set the modifier for such images to DRM_FORMAT_MOD_INVALID.
Also preserve the modifier when duplicating an image, as will be done by
GBM when importing f
Calling random callbacks on the display's event queue is hostile, as
we may call into client code when it least expects it. Create our own
event queue, one per wsi_wl_display, and use that for the registry.
Signed-off-by: Daniel Stone
---
src/vulkan/wsi/wsi_common_wayland.c | 44
There's no need to call wl_display_roundtrip() after trying to create a
buffer through wl_drm; if it succeeds then everything is fine, and if it
fails, then we get a fatal protocol error so can't recover anyway.
Additionally, doing a roundtrip on the default / main application queue,
is destructiv
Though most swapchain operations used a queue, they were racy in that
the object was created with the queue only set later, meaning that its
event could potentially be dispatched from the default queue in between
these two steps.
Use proxy wrappers to avoid this race, also assigning wl_buffers cre
Untangle the exit cleanup paths so we don't try to use the registry
variable before it's been initialised.
Signed-off-by: Daniel Stone
---
src/vulkan/wsi/wsi_common_wayland.c | 9 +
1 file changed, 5 insertions(+), 4 deletions(-)
diff --git a/src/vulkan/wsi/wsi_common_wayland.c
b/src/v
On 03.05.2017 18:37, Samuel Pitoiset wrote:
The spec allows memory qualifiers to be used with image variables,
buffers variables and shader storage blocks. This patch also fixes
validate_memory_qualifier_for_type().
Fixes the following ARB_uniform_buffer_object test:
uniform-block-memory-qualif
On 04.05.2017 11:04, Samuel Pitoiset wrote:
The TGSI DCE pass doesn't eliminate dead assignments like
MOV TEMP[0], TEMP[1] in presence of loops because it assumes
that the visitor doesn't emit dead code. This assumption is
actually wrong and this situation happens.
However, it appears that the m
On 04.05.2017 00:02, Marek Olšák wrote:
On Wed, May 3, 2017 at 6:04 PM, Nicolai Hähnle wrote:
On 01.05.2017 14:53, Marek Olšák wrote:
From: Marek Olšák
This is the best place to do it. Now drivers without u_vbuf don't have to
do it.
---
src/mesa/state_tracker/st_atom_array.c | 56
+
Hi Gregory,
On 3 May 2017 at 17:34, Gregory Hainaut wrote:
> I extended the struct __DRIbackgroundCallableExtensionRec because
> the other function pointer is already related for glthread.
>
> DRI2/DRI3 glx code path check that display can be locked (basically
> XInitThread was called)
>
How abou
Fixes returned value changed by b546c9d.
Fixes: b546c9d ("anv: anv_gem_mmap() returns MAP_FAILED as mapping error")
Signed-off-by: Samuel Iglesias Gonsálvez
Cc: "17.0 17.1"
---
src/intel/vulkan/anv_image.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/src/intel/vulkan/anv_
On 3 May 2017 at 17:34, Gregory Hainaut wrote:
> Otherwise print a warning
>
s/print a warning/$something_bad_that_can_happen/
Print an error message for the user if the requirement is not met, or
we're not thread safe.
> v2: based on Nicolai feedback
> Check the DRI extension version
>
> Signed
malloc can return valid pointer for zero size allocation,
which causes OOB access later on
---
src/mesa/main/shaderapi.c | 3 +++
1 file changed, 3 insertions(+)
diff --git a/src/mesa/main/shaderapi.c b/src/mesa/main/shaderapi.c
index c41f006eb7..36cff0ca6e 100644
--- a/src/mesa/main/shaderapi.c
In some cases gl_marshal.py generates entries with "const const" leading
to a compiler warning. This strips the superfluous const from the type
string.
Signed-off-by: Marc Dietrich
---
src/mapi/glapi/gen/gl_marshal.py | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/src/mapi/g
On 04/05/17 18:22, Samuel Pitoiset wrot
On 05/04/2017 02:36 AM, Timothy Arceri wrote:
This and the tgsi copy_propagation pass are very slow, I'd really like
it if we both didn't require the pass for things to work and also
didn't make it any slower. perf is showing these 2 passes at ~11%
durin
I filed a bug on this, my attempt caused some compilation warnings, see:
https://bugs.freedesktop.org/show_bug.cgi?id=100227
On 05/04/2017 01:16 PM, Marc Dietrich wrote:
In some cases gl_marshal.py generates entries with "const const" leading
to a compiler warning. This strips the superfluous co
On 28.04.2017 23:42, Marek Olšák wrote:
From: Marek Olšák
at least I think this is correct.
Yes, LLVM does the endian conversion.
Patches 1-7:
Reviewed-by: Nicolai Hähnle
---
src/gallium/drivers/radeonsi/si_shader.c | 15 ---
1 file changed, 8 insertions(+), 7 deletions(-)
On 28.04.2017 23:42, Marek Olšák wrote:
From: Marek Olšák
It works now.
Nice! Patches 8-14 are also:
Reviewed-by: Nicolai Hähnle
---
src/gallium/drivers/radeonsi/si_state_shaders.c | 10 --
1 file changed, 10 deletions(-)
diff --git a/src/gallium/drivers/radeonsi/si_state_shade
On 03.05.2017 18:59, Li, Samuel wrote:
Thanks for the comments. I was thinking probably marketing names only make
sense for amdgpu, but I am OK to add a pointer to radeon_winsys. Will send a v2
later.
We won't have marketing names when using radeon. The point is that the
code still needs to
Am Donnerstag, 4. Mai 2017, 13:00:38 CEST schrieben Sie:
> I filed a bug on this, my attempt caused some compilation warnings, see:
> https://bugs.freedesktop.org/show_bug.cgi?id=100227
I don't see warnings with my patch. However, I also don't understand the
origin of the problem ;-)
Marc
>
>
Patches 3-7:
Reviewed-by: Nicolai Hähnle
On 02.05.2017 22:53, Samuel Pitoiset wrote:
In plain GLSL, sampler and image types can only be declared
uniform-qualified global variables or 'in' function parameters.
Setting the read_only flag seems quite useless because other
checks will prevent sam
On 3 May 2017 at 16:57, Adam Jackson wrote:
> Signed-off-by: Adam Jackson
Can we have some commit message? Feel free to copy the following.
Introduce _egl_display::Options::Platforms for private storage.
For X11 platforms we can use it for the screen number as set by
EGL_PLATFORM_X11_SCREEN_EXT.
On 3 May 2017 at 16:57, Adam Jackson wrote:
> If the gbm_create_device() call here actually did fail, any subsequent
> eglTerminate on the display would segfault.
>
We do have a couple of checks after we set own_device, although both
are close to impossible to trigger.
Cc:
Reviewed-by: Emil Veli
On 4 May 2017 at 12:58, Emil Velikov wrote:
> On 3 May 2017 at 16:57, Adam Jackson wrote:
>> Signed-off-by: Adam Jackson
> Can we have some commit message? Feel free to copy the following.
>
> Introduce _egl_display::Options::Platforms for private storage.
> For X11 platforms we can use it for t
On 4 May 2017 at 12:12, Nicolai Hähnle wrote:
> On 03.05.2017 18:59, Li, Samuel wrote:
>>
>> Thanks for the comments. I was thinking probably marketing names only make
>> sense for amdgpu, but I am OK to add a pointer to radeon_winsys. Will send a
>> v2 later.
>
>
> We won't have marketing names w
On 05/04/2017 12:26 PM, Timothy Arceri wrote:
On 04/05/17 18:22, Samuel Pitoiset wrot
On 05/04/2017 02:36 AM, Timothy Arceri wrote:
This and the tgsi copy_propagation pass are very slow, I'd really
like it if we both didn't require the pass for things to work and
also didn't make it any slow
On 4 May 2017 at 10:33, Daniel Stone wrote:
> Though most swapchain operations used a queue, they were racy in that
> the object was created with the queue only set later, meaning that its
> event could potentially be dispatched from the default queue in between
> these two steps.
>
> Use proxy wr
On 02.05.2017 22:53, Samuel Pitoiset wrote:
From section 4.3.7 of the ARB_bindless_texture spec:
"(remove the following bullet from the last list on p. 39, thereby
permitting sampler types in interface blocks; image types are also
permitted in blocks by this extension)"
* sampler
On 4 May 2017 at 10:14, Daniel Stone wrote:
> Use a helper function and struct to convert between a modifier and
> tiling mode, so we can use it later for a tiling -> modifier lookup.
>
I think that this patch leads to a functionality change. See below for more.
If that's intentional please mentio
Patches 13 & 14:
Reviewed-by: Nicolai Hähnle
On 02.05.2017 22:53, Samuel Pitoiset wrote:
ARB_bindless_texture allows to declare images inside structures
which means that qualifiers like writeonly should be allowed.
I have a got a confirmation from Jeff Bolz (one author of the spec),
because
On 05/04/2017 02:13 PM, Samuel Pitoiset wrote:
On 05/04/2017 12:26 PM, Timothy Arceri wrote:
On 04/05/17 18:22, Samuel Pitoiset wrot
On 05/04/2017 02:36 AM, Timothy Arceri wrote:
This and the tgsi copy_propagation pass are very slow, I'd really
like it if we both didn't require the pass fo
On 02.05.2017 22:53, Samuel Pitoiset wrote:
From section 4.1.7 of the ARB_bindless_texture spec:
"Samplers aggregated into arrays within a shader (using square
brackets []) can be indexed with arbitrary integer expressions."
v3: - update spec comment formatting
Signed-off-by: Samuel Pit
On 02.05.2017 22:53, Samuel Pitoiset wrote:
From section 4.1.7 of the ARB_bindless_texture spec:
"Samplers can be used as l-values, so can be assigned into and
used as "out" and "inout" function parameters."
From section 4.1.X of the ARB_bindless_texture spec:
"Images can be used as l
On 02.05.2017 22:53, Samuel Pitoiset wrote:
v3: - rebase (and remove (sampler) ? 1 : vector_elements)
Signed-off-by: Samuel Pitoiset
Patches 27-28:
Reviewed-by: Nicolai Hähnle
---
src/compiler/glsl/lower_packed_varyings.cpp | 14 ++
1 file changed, 14 insertions(+)
diff --g
Hi Chad,
On 4 May 2017 at 00:47, Chad Versace wrote:
> From: Chad Versace
>
> This ensures that future calls to eglSwapBuffers and eglMakeCurrent emit
> an error.
>
> This patch is part of a series for fixing
> android.hardware.camera2.cts.RobustnessTest#testAbandonRepeatingRequestSurface
> on C
On 02.05.2017 22:53, Samuel Pitoiset wrote:
In a situation like:
(tex vec4 (record_ref (var_ref f) tex) (constant vec2 (0.00 0.00)) 0
1 () )
The sampler needs to be lowered, otherwise this ends up with
"ir_dereference_variable @ 0x229a100 specifies undeclared variable
`ubo_load_temp
https://bugs.freedesktop.org/show_bug.cgi?id=100928
Bug ID: 100928
Summary: exported opengl function in generated dll was mangled
when building with msvc 2015 Update 3
Product: Mesa
Version: 17.1
Hardware: x86 (IA32)
On 05/04/2017 02:38 PM, Nicolai Hähnle wrote:
On 02.05.2017 22:53, Samuel Pitoiset wrote:
From section 4.1.7 of the ARB_bindless_texture spec:
"Samplers can be used as l-values, so can be assigned into and
used as "out" and "inout" function parameters."
From section 4.1.X of the ARB_bi
On 4 May 2017 at 11:01, Samuel Iglesias Gonsálvez wrote:
> Fixes returned value changed by b546c9d.
>
According to the spec we get VK_ERROR_OUT_OF_HOST_MEMORY or
VK_ERROR_OUT_OF_DEVICE_MEMORY on vkBindImageMemory failure.
I should have explicitly checked it closer :-\
> Fixes: b546c9d ("anv: anv_
On Thu, May 4, 2017 at 8:45 AM, Nicolai Hähnle wrote:
> On 02.05.2017 22:53, Samuel Pitoiset wrote:
>>
>> In a situation like:
>>
>> (tex vec4 (record_ref (var_ref f) tex) (constant vec2 (0.00
>> 0.00)) 0 1 () )
>>
>> The sampler needs to be lowered, otherwise this ends up with
>> "ir_d
On 05/04/2017 02:45 PM, Nicolai Hähnle wrote:
On 02.05.2017 22:53, Samuel Pitoiset wrote:
In a situation like:
(tex vec4 (record_ref (var_ref f) tex) (constant vec2 (0.00
0.00)) 0 1 () )
The sampler needs to be lowered, otherwise this ends up with
"ir_dereference_variable @ 0x22
https://bugs.freedesktop.org/show_bug.cgi?id=100928
--- Comment #1 from jon yang ---
Im not sure if this could be resolved by using this
#pragma comment(linker, "/EXPORT:" __FUNCTION__"=" __FUNCDNAME__)
Quick google search found this someone answer
http://stackoverflow.com/questions/2804893/c-dl
On Thu, 2017-05-04 at 14:03 +0100, Emil Velikov wrote:
> On 4 May 2017 at 11:01, Samuel Iglesias Gonsálvez om> wrote:
> > Fixes returned value changed by b546c9d.
> >
>
> According to the spec we get VK_ERROR_OUT_OF_HOST_MEMORY or
> VK_ERROR_OUT_OF_DEVICE_MEMORY on vkBindImageMemory failure.
> I
From: Nicolai Hähnle
There are a bunch of piglit fast clear tests that regressed on SI, for
example ./bin/ext_framebuffer_multisample-fast-clear single-sample.
The problem is that a texture is bound as a framebuffer, cleared, and
then rendered from in a loop that loops through different clear co
On 04.05.2017 15:06, Samuel Pitoiset wrote:
On 05/04/2017 02:45 PM, Nicolai Hähnle wrote:
On 02.05.2017 22:53, Samuel Pitoiset wrote:
In a situation like:
(tex vec4 (record_ref (var_ref f) tex) (constant vec2 (0.00
0.00)) 0 1 () )
The sampler needs to be lowered, otherwise this e
On 04.05.2017 14:08, Emil Velikov wrote:
On 4 May 2017 at 12:12, Nicolai Hähnle wrote:
On 03.05.2017 18:59, Li, Samuel wrote:
Thanks for the comments. I was thinking probably marketing names only make
sense for amdgpu, but I am OK to add a pointer to radeon_winsys. Will send a
v2 later.
We
On 05/04/2017 03:25 PM, Nicolai Hähnle wrote:
On 04.05.2017 15:06, Samuel Pitoiset wrote:
On 05/04/2017 02:45 PM, Nicolai Hähnle wrote:
On 02.05.2017 22:53, Samuel Pitoiset wrote:
In a situation like:
(tex vec4 (record_ref (var_ref f) tex) (constant vec2 (0.00
0.00)) 0 1 () )
Remove some boilerplate and check more specifically for all the entry-
points that different modules need.
Signed-off-by: Rob Clark
---
Ok, I guess it isn't likely for a driver to expose eglCreateImageKHR
but not eglDeleteImageKHR.. but otoh kmscube is useful for bringing
up drivers.
common.h
With atomic kms backend, userspace can get further ahead of the gpu. In
the decoder, we unref the previous frame when retrieving the current
frame. If userspace gets too far ahead, this can happen while the gpu
is still sampling from the previous frame.
Simple solution is add a fence.
Signed-of
Hi Emil,
On 4 May 2017 at 13:27, Emil Velikov wrote:
>> @@ -581,21 +600,17 @@ intel_create_image_common(__DRIscreen *dri_screen,
>> assert(!(use && count));
>>
>> uint64_t modifier = select_best_modifier(&screen->devinfo, modifiers,
>> count);
>> - switch (modifier) {
>> - case I915_
On Thu, Apr 27, 2017 at 10:58:42AM -0700, Lionel Landwerlin wrote:
> On 27/04/17 08:20, Eric Anholt wrote:
> > Emil Velikov writes:
> >
> > > On 25 April 2017 at 23:56, Lionel Landwerlin
> > > wrote:
> > > > Hi,
> > > >
> > > > While working with changes that span from kernel to user space, I'v
On 4 May 2017 at 14:43, Daniel Stone wrote:
> Hi Emil,
>
> On 4 May 2017 at 13:27, Emil Velikov wrote:
>>> @@ -581,21 +600,17 @@ intel_create_image_common(__DRIscreen *dri_screen,
>>> assert(!(use && count));
>>>
>>> uint64_t modifier = select_best_modifier(&screen->devinfo, modifiers,
>
On 04.05.2017 05:34, Timothy Arceri wrote:
On 04/05/17 13:31, Dave Airlie wrote:
+/* The ARB_separate_shader_object spec says:
+ *
+ * "The executable code for an individual shader stage is taken
from
+ * the current program for that stage. If there is a current
program
+ * object
On Wed, May 03, 2017 at 05:07:32PM -0700, Rafael Antognolli wrote:
> On Wed, May 03, 2017 at 12:22:13PM +0300, Topi Pohjolainen wrote:
> > Patches 1-17 are revision that
> >
> > - rework hiz on gen6 to use on-demand offset calculator allowing
> > one to drop dependency to miptree structure a
On 27 April 2017 at 16:20, Eric Anholt wrote:
> Emil Velikov writes:
>
>> On 25 April 2017 at 23:56, Lionel Landwerlin
>> wrote:
>>> Hi,
>>>
>>> While working with changes that span from kernel to user space, I've
>>> been wondering whether we need to depend on libdrm at all for the anv
>>> & i9
On Thursday, 2017-05-04 09:39:15 -0400, Rob Clark wrote:
> Remove some boilerplate and check more specifically for all the entry-
> points that different modules need.
>
> Signed-off-by: Rob Clark
> ---
> Ok, I guess it isn't likely for a driver to expose eglCreateImageKHR
> but not eglDeleteImag
On Thu, May 4, 2017 at 1:00 AM, Juan A. Suarez Romero
wrote:
> On Wed, 2017-04-26 at 07:35 -0700, Jason Ekstrand wrote:
> > Previously, the maximum size of a state that could be allocated from a
> > state pool was a block. However, this has caused us various issues
> > particularly with shaders
On 4 May 2017 at 14:46, Daniel Vetter wrote:
> On Thu, Apr 27, 2017 at 10:58:42AM -0700, Lionel Landwerlin wrote:
>> On 27/04/17 08:20, Eric Anholt wrote:
>> > Emil Velikov writes:
>> >
>> > > On 25 April 2017 at 23:56, Lionel Landwerlin
>> > > wrote:
>> > > > Hi,
>> > > >
>> > > > While working
On Thu, May 4, 2017 at 1:17 AM, Juan A. Suarez Romero
wrote:
> On Wed, 2017-05-03 at 09:44 -0700, Jason Ekstrand wrote:
> > ---
> > src/intel/vulkan/anv_allocator.c | 11 +--
> > 1 file changed, 5 insertions(+), 6 deletions(-)
> >
> > diff --git a/src/intel/vulkan/anv_allocator.c b/src/i
https://bugs.freedesktop.org/show_bug.cgi?id=100613
--- Comment #21 from Ray Strode [halfline] ---
Unless I'm misunderstanding something, I think this comment in u_format.h is
the crux of the issue:
* If each channel is accessed as an individual N-byte value, X is always•
* at the lowest
From: Emil Velikov
Signed-off-by: Emil Velikov
---
src/intel/vulkan/anv_gem.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/src/intel/vulkan/anv_gem.c b/src/intel/vulkan/anv_gem.c
index 185086fefcc..68c5e611342 100644
--- a/src/intel/vulkan/anv_gem.c
+++ b/src/intel/vulkan
From: Emil Velikov
The function cannot return NULL, update the comment accordingly.
Fixes: b546c9d ("anv: anv_gem_mmap() returns MAP_FAILED as mapping error")
Cc: Samuel Iglesias Gonsálvez
Signed-off-by: Emil Velikov
---
src/intel/vulkan/anv_image.c | 2 +-
1 file changed, 1 insertion(+), 1 d
On 3 May 2017 at 19:16, Andres Gomez wrote:
> v2: Tapani as main contact and Mauro just for help with
> debugging/building (Mauro).
>
> Signed-off-by: Andres Gomez
> Cc: Emil Velikov
> Cc: Tapani Pälli
> Cc: Mauro Rossi
> Acked-by: Tapani Pälli
As long as the people mentioned are fine wi
On 3 May 2017 at 20:35, Rob Herring wrote:
> From: Mauro Rossi
>
> Necessary to fix the following radeonsi building errors:
>
> In file included from external/mesa/src/gallium/drivers/radeonsi/si_blit.c:24:
> In file included from external/mesa/src/gallium/drivers/radeonsi/si_pipe.h:29:
> In file
BLORP doesn't program 3DSTATE_VF, since it doesn't use index buffers,
making the setting irrelevant. So there's no need to re-emit it after
a BLORP operation - the old setting will still be in place.
---
src/mesa/drivers/dri/i965/genX_state_upload.c | 3 +--
1 file changed, 1 insertion(+), 2 dele
We can use a simple shift equation rather than a switch statement.
---
src/mesa/main/varray.c | 12 ++--
1 file changed, 2 insertions(+), 10 deletions(-)
diff --git a/src/mesa/main/varray.c b/src/mesa/main/varray.c
index eda86ec6a82..9497090e88a 100644
--- a/src/mesa/main/varray.c
+++ b/s
The whole "it might be used for non-indexed draws" thing is no longer
true - it turns out this was a mistake, and removed in OpenGL 4.5.
(See Marek's commit 96cbc1ca29e0b1f4f4d6c868b844aecb9080.) So
we can simplify this and just program 0 for non-indexed draws.
We can also use #if blocks to r
On 3 May 2017 at 20:35, Rob Herring wrote:
> A build of only swrast is broken as the Android EGL now depends on
> libdrm as does GBM. While we could make EGL conditionally depend on
> libdrm, we probably want to enable kms_dri winsys as well and that will
> need libdrm enabled. So just always enab
It fixes occasional crashes when app exits
and glthread is still processing commands.
---
src/mesa/main/context.c | 3 +++
1 file changed, 3 insertions(+)
diff --git a/src/mesa/main/context.c b/src/mesa/main/context.c
index 3570f94f5a..c3c4095329 100644
--- a/src/mesa/main/context.c
+++ b/src/mes
1 - 100 of 240 matches
Mail list logo
