Reviewed-by: Jason Ekstrand
On Fri, Feb 6, 2015 at 9:42 AM, Francisco Jerez
wrote:
> Fixes metadata guess when instructions in the program specify a
> destination register with non-zero reg_offset and when the payload of
> a LOAD_PAYLOAD spans several registers.
> ---
> src/m
al type of MOV-to-MRF that can move from
(gN, gN+1) to (mK, mK+4) and we handle that in lower_load_payload right
now. However, it doesn't follow the standard rules.
--Jason
On Fri, Feb 20, 2015 at 2:10 PM, Jason Ekstrand
wrote:
>
>
> On Fri, Feb 20, 2015 at 1:09 PM, Francisco Jerez
On Fri, Feb 20, 2015 at 2:43 PM, Francisco Jerez
wrote:
> Jason Ekstrand writes:
>
> > One more comment here... This particularly regards your plan of
> separating
> > it into "things that match the destination" and "other things" and not
> copy
>
This is mostly correct and it's a good solution. The only problem is that
it doesn't handle the skipRows packing property properly. This property
tells the driver the stride (in rows) between image planes in the data.
Most of the places where depth is used below, we should be using the stride
(in
On Fri, Feb 20, 2015 at 4:30 PM, Laura Ekstrand
wrote:
> Solves bugs related to the driver not setting up the texture miptree
> correctly, leading to faulty PBO uploads and downloads.
> ---
> src/mesa/drivers/common/meta_tex_subimage.c | 13 +
> src/mesa/drivers/dri/i965/intel_tex.c
---
src/glsl/nir/nir_opt_gcm.c | 4
1 file changed, 4 insertions(+)
diff --git a/src/glsl/nir/nir_opt_gcm.c b/src/glsl/nir/nir_opt_gcm.c
index bf565b9..b4f5fd3 100644
--- a/src/glsl/nir/nir_opt_gcm.c
+++ b/src/glsl/nir/nir_opt_gcm.c
@@ -121,9 +121,11 @@ gcm_pin_instructions_block(nir_block *
On Feb 12, 2015 1:28 AM, "Kenneth Graunke" wrote:
>
> On Wednesday, February 11, 2015 04:14:20 PM Ian Romanick wrote:
> > On 02/11/2015 04:05 PM, Matt Turner wrote:
> > > On Wed, Feb 11, 2015 at 3:51 PM, Ian Romanick
wrote:
> > >> On 02/11/2015 02:54 PM, Matt Turner wrote:
> > >>> We propagate ne
On Mon, Feb 23, 2015 at 12:51 AM, Martin Peres wrote:
> On 20/02/15 20:38, Laura Ekstrand wrote:
>
>> Please provide a page number and a section title in your spec comment.
>>
>
> I'm not very fond of giving the page number as Khronos updates the pdf
> quite often
> and the page number may change
On Tue, Feb 24, 2015 at 2:19 AM, Kenneth Graunke
wrote:
> st_glsl_to_tgsi and ir_to_mesa have handled conditional discards for a
> long time; the previous patch added that capability to i965.
>
> i965 (Haswell) shader-db stats:
>
> Without NIR:
> total instructions in shared programs: 5792133 ->
ore I gave an R-B
and scarred them off. I wanted a second set of eyes. So here's mine now.
Reviewed-by: Jason Ekstrand
for the series.
--Jason
> On 02/24/2015 03:20 PM, Laura Ekstrand wrote:
> > This moves the line setting immutability for the texture to after
> > _m
On Tue, Feb 24, 2015 at 7:01 PM, Anuj Phogat wrote:
> On Tue, Feb 24, 2015 at 8:31 AM, Neil Roberts
> wrote:
> > Anuj Phogat writes:
> >
> >> using a flag passed in as function parameter. This will enable
> >> _mesa_meta_pbo_GetTexSubImage to be used for reading in to
> >> non-pbo buffers.
> >>
Reviewed-by: Jason Ekstrand
On Wed, Feb 25, 2015 at 8:02 AM, Neil Roberts wrote:
> A layered PBO image is now interpreted as a single tall 2D image so
> the z argument in _mesa_meta_bind_fbo_image is ignored. Therefore this
> was just redundantly rebinding the same image r
I'll admit that I don't know that portion of the code all that well, but it
looks sensible to me.
Reviewed-by: Jason Ekstrand
On Thu, Feb 26, 2015 at 3:55 PM, Kenneth Graunke
wrote:
> _mesa_choose_tex_format (texformat.c) tries I8_SNORM, L8_SNORM, and
> either L8A8_SNORM or
Cc: 10.5
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=89292
---
src/glsl/nir/nir_lower_vars_to_ssa.c| 34 +
src/mesa/drivers/common/meta_tex_subimage.c | 2 +-
2 files changed, 26 insertions(+), 10 deletions(-)
diff --git a/src/glsl/nir/nir_lower_va
Cc: 10.5
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=89292
---
src/mesa/drivers/common/meta_tex_subimage.c | 8
1 file changed, 4 insertions(+), 4 deletions(-)
diff --git a/src/mesa/drivers/common/meta_tex_subimage.c
b/src/mesa/drivers/common/meta_tex_subimage.c
index 5ae12c
NEVER MIND THIS PATCH. I had a git-send-email fail
On Fri, Feb 27, 2015 at 12:26 PM, Jason Ekstrand
wrote:
> Cc: 10.5
> Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=89292
> ---
> src/glsl/nir/nir_lower_vars_to_ssa.c| 34
> +
>
Alternatively, it's probably safe (and faster) to just save/restore the
scissor in meta_begin/end. The _mesa_meta_BlitFramebuffer implementation
stashes everything else anyway.
--Jason
On Fri, Feb 27, 2015 at 12:29 PM, Jason Ekstrand
wrote:
> Cc: 10.5
> Bugzilla: https://bugs.free
Thanks for figuring out how to do this properly.
Reviewed-by: Jason Ekstrand
On Fri, Feb 27, 2015 at 12:06 AM, Kenneth Graunke
wrote:
> We used to loop over all color attachments, and emit FB writes for each
> one, even if the shader didn't write to a corresponding output variab
Reviewed-by: Jason Ekstrand
On Fri, Feb 27, 2015 at 2:19 PM, Ian Romanick wrote:
> From: Ian Romanick
>
> It appears that all the other instructions that need it already use it.
> This one just got missed.
>
> Signed-off-by: Ian Romanick
> Cc: "10.5"
> Cc:
On Fri, Feb 27, 2015 at 6:04 AM, Jose Fonseca wrote:
> On 26/02/15 18:07, Brian Paul wrote:
>
>> On 02/26/2015 09:51 AM, Jose Fonseca wrote:
>>
>>> This is to enable the code to build with -Werror=vla in the short term,
>>> and enable the code to build with MSVC2013 soon after.
>>> ---
>>> incl
From: Jason Ekstrand
This takes "fbo-stencil blit GL_STENCIL_INDEX1/4/16" from crash to pass on
BDW.
---
src/mesa/main/teximage.c | 3 +++
1 file changed, 3 insertions(+)
diff --git a/src/mesa/main/teximage.c b/src/mesa/main/teximage.c
index 6e45cc9..a68574d 100644
--- a/src
Previously, the nir_if control-flow node had a source built straight into
it that was the if condition. This has been the source of a lot of
edge-case headaches due to, in particular, the two different use sets that
we were carrying around. This patch changes it to have a special jump
instruction
On Feb 27, 2015 9:17 PM, "Ben Widawsky" wrote:
>
> On Fri, Feb 27, 2015 at 07:24:16PM +, Neil Roberts wrote:
> > The vertical alignment is ignored in the surface state for 1D array
> > textures so we can tightly pack them.
> >
> > I've run this through Piglit and it doesn't cause any regressio
On Feb 28, 2015 4:55 AM, "Thomas Helland" wrote:
>
> So here comes my hash-table series mentioned earlier.
>
> So, first of all, there's some issues.
> I've been strugling with hitting assertion failures.
> The table returns null at times when it apparently should not.
> It occurs after patch 1, a
Both patches are
Reviewed-by: Jason Ekstrand
On Sat, Feb 28, 2015 at 11:32 AM, Thomas Helland
wrote:
> Shader-db i965 instructions:
> total instructions in shared programs: 1711180 -> 1711159 (-0.00%)
> instructions in affected programs: 825 -> 804 (-
Yeah, this is probably good enough for now.
Reviewed-by: Jason Ekstrand
On Feb 28, 2015 1:42 PM, "Matt Turner" wrote:
> We could do better by tracking scratch reads and writes.
>
> Cc: 10.5
> Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=88793
> ---
>
On Sat, Feb 28, 2015 at 10:33 AM, Connor Abbott wrote:
> On Fri, Feb 27, 2015 at 8:13 PM, Jason Ekstrand
> wrote:
> > Previously, the nir_if control-flow node had a source built straight into
> > it that was the if condition. This has been the source of a lot of
> > ed
Not sure why you Cc'd me, but whatever. Seems obvious enough.
Reviewed-by: Jason Ekstrand
On 02/27/2015 06:50 PM, Ian Romanick wrote:
From: Ian Romanick
Unused since b18fd23.
brw_fs.cpp:2878:44: warning: unused parameter 'dispatch_width'
[-Wunused-parameter]
clear_de
I don't know if you got everything but I didn't notice anything missing and
what's there looks correct to me.
Reviewed-by: Jason Ekstrand
On Mar 1, 2015 7:08 PM, "Ilia Mirkin" wrote:
> Signed-off-by: Ilia Mirkin
> ---
> src/mapi/glapi/gen/es_EXT.xml | 6 ++
Ping. I don't want this one to get lost in the header file churn.
--Jason
On Feb 27, 2015 12:33 PM, "Jason Ekstrand" wrote:
> Alternatively, it's probably safe (and faster) to just save/restore the
> scissor in meta_begin/end. The _mesa_meta_BlitFramebuffer implement
a
>
> On Fri, Feb 27, 2015 at 6:55 PM, Jason Ekstrand
> wrote:
> > From: Jason Ekstrand
> >
> > This takes "fbo-stencil blit GL_STENCIL_INDEX1/4/16" from crash to pass
> on
> > BDW.
> > ---
> > src/mesa/main/teximage.c | 3 +++
> > 1 f
On Mon, Mar 2, 2015 at 11:33 AM, Ilia Mirkin wrote:
> On Mon, Mar 2, 2015 at 2:32 PM, Jason Ekstrand
> wrote:
> >
> >
> > On Mon, Mar 2, 2015 at 11:18 AM, Ilia Mirkin
> wrote:
> >>
> >> Hmmm... I was just looking at this code in connection to attepm
On Mon, Mar 2, 2015 at 8:02 AM, Jose Fonseca wrote:
> On 27/02/15 16:15, Brian Paul wrote:
>
>> On 02/27/2015 09:10 AM, Ian Romanick wrote:
>>
>>> On 02/26/2015 10:07 AM, Brian Paul wrote:
>>>
On 02/26/2015 09:51 AM, Jose Fonseca wrote:
> This is to enable the code to build with -We
---
src/glsl/nir/nir_worklist.c | 10 +-
src/glsl/nir/nir_worklist.h | 4 ++--
2 files changed, 7 insertions(+), 7 deletions(-)
diff --git a/src/glsl/nir/nir_worklist.c b/src/glsl/nir/nir_worklist.c
index a8baae9..3087a1d 100644
--- a/src/glsl/nir/nir_worklist.c
+++ b/src/glsl/nir/nir_wo
LGTM. I would still rather do the allocation in live_variables not on the
stack, but I can clean that up later.
Reviewed-by: Jason Ekstrand
On Tue, Mar 3, 2015 at 5:19 AM, Jose Fonseca wrote:
> v2:
> - Single statement, by using memset return value as suggested by Ian
> Romanic
On Tue, Mar 3, 2015 at 10:07 AM, Chad Versace
wrote:
> On 02/23/2015 06:32 AM, Jonny Lamb wrote:
> > Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=67676
> > ---
> > include/EGL/eglmesaext.h| 5 +
> > src/egl/drivers/dri2/platform_wayland.c | 9 +++--
> > src/eg
On Tue, Mar 3, 2015 at 1:00 PM, Ilia Mirkin wrote:
> On Tue, Mar 3, 2015 at 3:58 PM, Kenneth Graunke
> wrote:
> > On Monday, March 02, 2015 02:33:31 PM Ilia Mirkin wrote:
> >> On Mon, Mar 2, 2015 at 2:32 PM, Jason Ekstrand
> wrote:
> >> >
> >> &
On Wed, Mar 4, 2015 at 8:45 AM, Ian Romanick wrote:
> On 03/01/2015 05:19 PM, Jason Ekstrand wrote:
> > Not sure why you Cc'd me, but whatever. Seems obvious enough.
>
> Because b18fd23 was your commit. Sometimes left over, unused parameters
> are the result of a mistak
}
> xtiled_to_linear(x0, x1, x2, x3, y0, y1,
> dst, src, dst_pitch, swizzle_bit, mem_copy);
> @@ -465,16 +532,26 @@ ytiled_to_linear_faster(uint32_t x0, uint32_t x1,
> uint32_t x2, uint32_t x3,
>if (mem_copy == memcpy)
> return ytiled_to_linea
/bugs.freedesktop.org/show_bug.cgi?id=89416
> Tested-by: Uriy Zhuravlev [v1]
> Reviewed-by: Jason Ekstrand
> ---
> Benchmarks I should run?
>
I'm not worried about it. Maybe changing from aligned to unaligned reads
makes more or less of a difference than changing writes.
Series is
Reviewed-by: Jason Ekstrand
On Wed, Mar 4, 2015 at 3:15 PM, Emil Velikov
wrote:
> On 4 March 2015 at 17:22, Neil Roberts wrote:
> > This reverts commit 546aba143d13ba3f993ead4cc30b2404abfc0202.
> >
> > I think the changes to the calls to glBlitFramebuffer from
On Wed, Mar 4, 2015 at 5:28 PM, Matt Turner wrote:
> On Wed, Mar 4, 2015 at 5:07 PM, Jason Ekstrand
> wrote:
> > On Wed, Mar 4, 2015 at 4:56 PM, Matt Turner wrote:
> >> @@ -357,16 +394,22 @@ linear_to_xtiled_faster(uint32_t x0, uint32_t x1,
> >> uint32_t x
Fine by me
Reviewed-by: Jason Ekstrand
On Thu, Mar 5, 2015 at 8:26 PM, Kenneth Graunke
wrote:
> Being able to see both location and driver_location can be useful when
> debugging IO mistakes.
>
> Signed-off-by: Kenneth Graunke
> ---
> src/glsl/nir/nir_print.c | 2 +-
&g
On Thu, Mar 5, 2015 at 5:36 PM, Thomas Helland
wrote:
> 2015-02-28 17:05 GMT+01:00 Jason Ekstrand :
> >
> > On Feb 28, 2015 4:55 AM, "Thomas Helland"
> wrote:
> >>
> >> So here comes my hash-table series mentioned earlier.
> >>
>
This looks fine to me. I just kicked off a build on our test farm and,
assuming that looks good (I'll send another e-mail in the morning if it
does),
Reviewed-by: Jason Ekstrand
I ran shader-db on the change and I was kind of surprised to see that it
doesn't really do anythin
Acked-by: Jason Ekstrand
On Mar 6, 2015 2:18 AM, "Kenneth Graunke" wrote:
> The code in glsl_to_nir is entirely dead, as we translate from GLSL to
> NIR at link time, when there isn't a _mesa_glsl_parse_state to pass,
> so every caller passes NULL.
>
> glsl_to_nir
I gave you an back on 3; I'll let Eric actually review it. The rest are
Reviewed-by: Jason Ekstrand
On Mar 6, 2015 2:18 AM, "Kenneth Graunke" wrote:
> Piglit's spec/glsl-1.20/compiler/structure-and-array-operations/
> array-selection.vert test contains the followin
On Mar 7, 2015 10:57 AM, "Sean Burke" wrote:
>
> The memory layout of compatible internal formats may differ in bytes per
> block, so TexFormat is not a reliable measure of compatibility.
Additionally,
> the current check allows compressed textures of the same block size to be
used,
> which is in
On Sat, Mar 7, 2015 at 12:16 PM, Sean Burke wrote:
>
>
> On Sat, Mar 7, 2015 at 12:06 PM, Jason Ekstrand
> wrote:
>
>>
>> On Mar 7, 2015 10:57 AM, "Sean Burke" wrote:
>> >
>> > The memory layout of compatible internal formats may differ
On Thu, Mar 5, 2015 at 9:39 PM, Jason Ekstrand wrote:
> This looks fine to me. I just kicked off a build on our test farm and,
> assuming that looks good (I'll send another e-mail in the morning if it
> does),
>
> Reviewed-by: Jason Ekstrand
>
Jenkins results look go
On Sat, Mar 7, 2015 at 9:58 AM, Emil Velikov
wrote:
> On 27 February 2015 at 08:06, Kenneth Graunke
> wrote:
> > We used to loop over all color attachments, and emit FB writes for each
> > one, even if the shader didn't write to a corresponding output variable.
> > Those color attachments would
t_compatible(struct gl_context *ctx,
> +GLenum srcFormat, GLenum dstFormat)
> +{
> + /*
>
+* For the purposes of CopyImageSubData, two internal formats
> +* are considered compatible if any of the following conditions are
> +
ed, view-compatible formats return the
> correct result, and make style fixes. Original commit message amended
> for clarity.
> v4: Reformatted spec citations.
>
> Reviewed-by: Jason Ekstrand
> ---
> src/mesa/main/copyimage.c | 151
> +
On Mar 7, 2015 10:33 PM, "Kenneth Graunke" wrote:
>
> On Saturday, March 07, 2015 10:09:15 PM Jason Ekstrand wrote:
> > LGTM.
>
> Jason,
>
> Sean doesn't have commit access (it's actually his first patch) - if
> this is good to go, would you mind pu
LGTM
Reviewed-by: Jason Ekstrand
On Mon, Mar 9, 2015 at 1:58 AM, Kenneth Graunke
wrote:
> Array variable copy splitting generates a bunch of stuff we want to
> clean up before proceeding.
>
> Signed-off-by: Kenneth Graunke
> Cc: Jason Ekstrand
> ---
> src/
the only point of the hash table was to avoid inserting
> > the same node multiple times for different dereferences. We never
> > actually searched the hash table! This patch uses an intrusive
> > linked list instead. Since exec_list uses head and tail sentinels,
> > chec
On Mon, Mar 9, 2015 at 7:24 PM, Matt Turner wrote:
> On Mon, Mar 9, 2015 at 6:36 PM, Kenneth Graunke
> wrote:
> > From: Jason Ekstrand
> >
> > __next and __prev are pointers to the structure containing the exec_node
> > link, not the embedded exec_node. NUL
On Mon, Mar 9, 2015 at 7:28 PM, Matt Turner wrote:
> On Mon, Mar 9, 2015 at 7:04 PM, Jason Ekstrand
> wrote:
> > Push it!
>
> Our policy is to wait a day for most things.
>
Sure. Not really arguing for early pushing. Mostly just surprised that
Connor picked up on
On Mon, Mar 9, 2015 at 7:48 PM, Connor Abbott wrote:
> On Mon, Mar 9, 2015 at 10:35 PM, Matt Turner wrote:
> > On Mon, Mar 9, 2015 at 7:24 PM, Matt Turner wrote:
> >> On Mon, Mar 9, 2015 at 6:36 PM, Kenneth Graunke
> wrote:
> >>> From: Jason Ekstrand
On Mon, Mar 9, 2015 at 7:58 PM, Matt Turner wrote:
> On Mon, Mar 9, 2015 at 7:32 PM, Jason Ekstrand
> wrote:
> >
> >
> > On Mon, Mar 9, 2015 at 7:24 PM, Matt Turner wrote:
> >>
> >> On Mon, Mar 9, 2015 at 6:36 PM, Kenneth Graunke
> >> wrote
On Mon, Mar 9, 2015 at 7:59 PM, Connor Abbott wrote:
> On Mon, Mar 9, 2015 at 10:54 PM, Jason Ekstrand
> wrote:
> >
> >
> > On Mon, Mar 9, 2015 at 7:48 PM, Connor Abbott
> wrote:
> >>
> >> On Mon, Mar 9, 2015 at 10:35 PM, Matt Turner
> wrote:
&g
On Mon, Mar 9, 2015 at 11:06 PM, Matt Turner wrote:
> On Mon, Mar 9, 2015 at 6:36 PM, Kenneth Graunke
> wrote:
> > From: Jason Ekstrand
> >
> > __next and __prev are pointers to the structure containing the exec_node
> > link, not the embedded exec_node. NUL
On Tue, Mar 10, 2015 at 4:50 PM, Matt Turner wrote:
> On Tue, Mar 10, 2015 at 4:20 PM, Ian Romanick wrote:
> > On 02/28/2015 12:19 PM, Matt Turner wrote:
> >> On Sat, Feb 28, 2015 at 11:47 AM, Thomas Helland
> >> wrote:
> >>> On Feb 28,
On Tue, Mar 10, 2015 at 5:45 PM, Matt Turner wrote:
> On Tue, Mar 10, 2015 at 1:09 PM, Jason Ekstrand
> wrote:
> > How about we do things slightly differently and check
> "(__node)->field.next
> > != NULL" just like we do on regular versions. Since the check
On Fri, Jan 27, 2017 at 11:22 AM, Emil Velikov
wrote:
> On 26 January 2017 at 17:27, Jason Ekstrand wrote:
> > I'm pretty sure we've kept up with the bug fixes.
> > ---
> > src/intel/vulkan/anv_device.c | 2 +-
> > 1 file changed, 1 insertion(+), 1 delet
I think this is ok,
Reviewed-by: Jason Ekstrand
On Thu, Jan 26, 2017 at 1:27 AM, Kenneth Graunke
wrote:
> At this point, the pitch is in bytes. We haven't yet divided the pitch
> by 4 for tiled surfaces, so abs(pitch) may be larger than 32K. This
> means the bit 15 trick won&#
Cc: "13.0 17.0"
---
src/intel/blorp/blorp_clear.c | 19 +++
1 file changed, 19 insertions(+)
diff --git a/src/intel/blorp/blorp_clear.c b/src/intel/blorp/blorp_clear.c
index afc505d..f8ac6dc 100644
--- a/src/intel/blorp/blorp_clear.c
+++ b/src/intel/blorp/blorp_clear.c
@@ -349,6
This causes hangs on Broadwell if you try to render to it. I have no
idea how we managed to not hit this earlier.
Cc: "13.0 17.0"
---
src/intel/isl/isl_format.c | 5 +++--
1 file changed, 3 insertions(+), 2 deletions(-)
diff --git a/src/intel/isl/isl_format.c b/src/intel/isl/isl_format.c
index
On Fri, Jan 27, 2017 at 2:31 PM, Matt Turner wrote:
> On Fri, Jan 27, 2017 at 2:18 PM, Jason Ekstrand
> wrote:
> > This causes hangs on Broadwell if you try to render to it. I have no
> > idea how we managed to not hit this earlier.
> >
> > Cc: "13.
Reviewed-by: Jason Ekstrand
On Mon, Jan 30, 2017 at 2:37 PM, Nanley Chery wrote:
> Commit 2852efcda40274acf3272611c6a3b7731523a72d moved the location of
> the depth input attachment surface state from the render pass to the
> image view, but failed to update the surface state loca
On Jan 30, 2017 10:03 PM, "Tapani Pälli" wrote:
On 01/30/2017 04:00 PM, Emil Velikov wrote:
> On 30 January 2017 at 11:37, Tapani Pälli wrote:
>
>> fixes to issues spotted by Emil Velikov:
>>
>>- set ANV_TIMESTAMP corretly
>>- fix typo with VULKAN_GEM_FILES
>>
>> v2: update to use Mak
On Jan 30, 2017 4:15 AM, "Lionel Landwerlin"
wrote:
On 27/01/17 09:57, Juan A. Suarez Romero wrote:
> On Fri, 2017-01-27 at 09:46 +, Lionel Landwerlin wrote:
>
>> But what the test does is calling OpSpecConstantOp[2], which is the
>>> operation we are patching here.
>>>
>>> And according to
upstream
>
> Signed-off-by: Topi Pohjolainen
> CC: Kenneth Graunke
> CC: Jason Ekstrand
> ---
> src/mesa/drivers/dri/i965/brw_draw.c | 49 ++
> +++---
> 1 file changed, 46 insertions(+), 3 deletions(-)
>
> diff --git a/src/mesa/dri
It is not clear from the docs exactly how pipelined STATE_BASE_ADDRESS
actually is. Previously, we knew we needed to flush prior to re-emitting
STATE_BASE_ADDRESS on gen8+ but we had never confirmed it on gen7 so we
left it alone and avoided the flush. Recently, Mark found hangs on gen7
which app
context *brw)
>
> dw1 = GEN6_SF_SWIZZLE_ENABLE | num_outputs <<
> GEN6_SF_NUM_OUTPUTS_SHIFT;
> dw2 = GEN6_SF_STATISTICS_ENABLE;
> + dw3 = GEN6_SF_SCISSOR_ENABLE;
> + dw4 = 0;
>
We could, in theory, only scissor if one of our viewports is actually
smaller
Blorp doesn't use uniform push constants anymore so there's no need to
support them in compile_nir_shader.
Signed-off-by: Jason Ekstrand
---
src/intel/blorp/blorp.c | 23 ---
1 file changed, 23 deletions(-)
diff --git a/src/intel/blorp/blorp.c b/src/intel/blo
While we're at it, we rename it to remove the brw_ prefix
Signed-off-by: Jason Ekstrand
---
src/intel/blorp/blorp.c | 26 +-
src/intel/blorp/blorp_blit.c | 2 +-
src/intel/blorp/blorp_clear.c | 2 +-
src/intel/blorp/blorp_genX_exec.h
Reviewed-by: Jason Ekstrand
On Sun, Jan 22, 2017 at 10:42 PM, Kenneth Graunke
wrote:
> This reverts commit 0bac2551e40410e2251daf4fd9faf69310ab34ce.
>
> Now that we position the guardband correctly (applying translations
> in addition to scaling) and made it as large (or larg
On Tue, Jan 31, 2017 at 10:38 AM, Jason Ekstrand
wrote:
>
> On Tue, Jan 31, 2017 at 8:15 AM, Topi Pohjolainen <
> topi.pohjolai...@gmail.com> wrote:
>
>> Current estimate doesn't consider space needed for surface states
>> and it only calculates for one shade
On Mon, Jan 23, 2017 at 10:19 PM, Ben Widawsky wrote:
> v2: Preserve legacy behavior when plane is 0 (Jason Ekstrand)
> EINVAL when input plane is greater than total planes (Jason Ekstrand)
> Don't leak the image after fromPlanar (Daniel)
> Move bo->image check below plane
On Mon, Jan 23, 2017 at 10:19 PM, Ben Widawsky wrote:
> Unlike stride, there was no previous offset getter, so it can be right
> on the first try.
>
> v2: Return EINVAL when plane is greater than total planes to make it
> match the similar APIs.
> Avoid leak after fromPlanar (Daniel)
> Make sure
On Mon, Jan 23, 2017 at 10:21 PM, Ben Widawsky wrote:
> The idea behind modifiers like this is that the user of GBM will have
> some mechanism to query what properties the hardware supports for its BO
> or surface. This information is directly passed in (and stored) so that
> the DRI implementati
On Mon, Jan 23, 2017 at 10:21 PM, Ben Widawsky wrote:
> This patch begins introducing how we'll actually handle the potentially
> many modifiers coming in from the API, how we'll store them, and the
> structure in the code to support it.
>
> Prior to this patch, the Y-tiled modifier would be enti
On Mon, Jan 23, 2017 at 10:21 PM, Ben Widawsky wrote:
> Replace the naive, 'save all the modifiers' with a proper query for just
> the modifier that was selected. To accomplish this, two new query tokens
> are added to the extension:
> __DRI_IMAGE_ATTRIB_MODIFIER_UPPER
> __DRI_IMAGE_ATTRIB_MODIFI
On Wed, Jan 25, 2017 at 10:58 AM, Pohjolainen, Topi <
topi.pohjolai...@gmail.com> wrote:
> On Mon, Jan 23, 2017 at 10:21:41PM -0800, Ben Widawsky wrote:
> > This provides a common function or creating miptrees when there is an
> > existing DRIimage to use. That provides an easy way to add CCS
> >
On Mon, Jan 23, 2017 at 10:21 PM, Ben Widawsky wrote:
> This patch shouldn't actually do anything because the libdrm function
> should already do this alignment. However, it preps us for a future
> patch where we add in the CCS AUX size, and in the process it serves as
> a good place to find bise
On Mon, Jan 23, 2017 at 10:21 PM, Ben Widawsky wrote:
> Since the code doesn't support modifiers yet, this patch should do
> nothing other than prepare for more patches.
>
> Signed-off-by: Ben Widawsky
> Acked-by: Daniel Stone
> ---
> src/mesa/drivers/dri/i965/intel_screen.c | 64
> +++
On Wed, Jan 25, 2017 at 10:36 AM, Pohjolainen, Topi <
topi.pohjolai...@gmail.com> wrote:
> On Mon, Jan 23, 2017 at 10:21:46PM -0800, Ben Widawsky wrote:
> > This patch provides the support (and comments) for allocating the BO
> > with space for the CCS buffer just underneath it.
> >
> > This patch
On Mon, Jan 23, 2017 at 10:21 PM, Ben Widawsky wrote:
> There is nothing particularly useful to do currently if the update
> fails, but there is no point carrying on either. As a result, this has a
> behavior change.
>
> v2: Make the return type a bool (Topi)
>
> Signed-off-by: Ben Widawsky
> Ac
On Mon, Jan 23, 2017 at 10:21 PM, Ben Widawsky wrote:
> Allows us to continue utilizing common miptree creation using __DRIimage
> without creating a new DRIimage (for the intel_process_dri2_buffer()
> case).
>
> This is a bit ugly, but I think it's the best one can do.
>
> Signed-off-by: Ben Wid
On Mon, Jan 23, 2017 at 10:21 PM, Ben Widawsky wrote:
> Signed-off-by: Ben Widawsky
> Acked-by: Daniel Stone
> ---
> src/mesa/drivers/dri/i965/intel_screen.c | 18 ++
> 1 file changed, 14 insertions(+), 4 deletions(-)
>
> diff --git a/src/mesa/drivers/dri/i965/intel_screen.c
>
On Mon, Jan 23, 2017 at 10:21 PM, Ben Widawsky wrote:
> v2: Put the commit message as a comment (Topi)
>
> Cc: Topi Pohjolainen
> Cc: Ville Syrjälä
> Cc: Jason Ekstrand
> Signed-off-by: Ben Widawsky
> Acked-by: Daniel Stone
> ---
> src/mesa/drivers/dri/i965/in
On Wed, Jan 25, 2017 at 10:39 AM, Pohjolainen, Topi <
topi.pohjolai...@gmail.com> wrote:
> On Mon, Jan 23, 2017 at 10:21:54PM -0800, Ben Widawsky wrote:
> > On Gen9 hardware, the display engine is able to scanout a compressed
> > framebuffer by providing an offset to auxiliary compression informat
On Mon, Jan 23, 2017 at 10:21 PM, Ben Widawsky wrote:
> FINISHME: Use the kernel's final choice for the fb modifier
>
> bwidawsk@norris2:~/intel-gfx/kmscube (modifiers $)
> ~/scripts/measure_bandwidth.sh ./kmscube none
> Read bandwidth: 603.91 MiB/s
> Write bandwidth: 615.28 MiB/s
> bwidawsk@norr
On Mon, Jan 16, 2017 at 1:13 AM, Topi Pohjolainen <
topi.pohjolai...@gmail.com> wrote:
> This is kept on purpose in i965. It can be moved to ISL if it
> is needed in vulkan.
>
> Pointers to miptrees are given solely for verification purposes.
> These will be dropped in following patches.
>
> Signe
On Tue, Jan 31, 2017 at 4:44 PM, Jason Ekstrand
wrote:
> On Mon, Jan 16, 2017 at 1:13 AM, Topi Pohjolainen <
> topi.pohjolai...@gmail.com> wrote:
>
>> This is kept on purpose in i965. It can be moved to ISL if it
>> is needed in vulkan.
>>
>> Po
On Mon, Jan 16, 2017 at 1:14 AM, Topi Pohjolainen <
topi.pohjolai...@gmail.com> wrote:
> Signed-off-by: Topi Pohjolainen
> ---
> src/mesa/drivers/dri/i965/brw_blorp.c | 25
> -
> src/mesa/drivers/dri/i965/intel_mipmap_tree.c | 6 ++
> 2 files changed, 18 inse
---
src/intel/isl/isl_surface_state.c | 32
1 file changed, 32 insertions(+)
diff --git a/src/intel/isl/isl_surface_state.c
b/src/intel/isl/isl_surface_state.c
index b735478..c7b220b 100644
--- a/src/intel/isl/isl_surface_state.c
+++ b/src/intel/isl/isl_surface_s
Did we actually end up needing this? After you got your gl_DrawId issues
sorted, were you ever able to get element 33 to work? The Sky Lake docs
say it should...
On Tue, Jan 31, 2017 at 7:00 AM, Lionel Landwerlin <
lionel.g.landwer...@intel.com> wrote:
> Signed-off-by: Lionel Landwerlin
> ---
On Tue, Jan 31, 2017 at 7:00 AM, Lionel Landwerlin <
lionel.g.landwer...@intel.com> wrote:
> Signed-off-by: Lionel Landwerlin
> ---
> src/intel/vulkan/genX_cmd_buffer.c | 2 +-
> src/intel/vulkan/genX_pipeline.c | 2 +-
> 2 files changed, 2 insertions(+), 2 deletions(-)
>
> diff --git a/src/in
601 - 700 of 12089 matches
Mail list logo
