On Monday, January 23, 2017 11:23:13 AM PST Ian Romanick wrote:
> From: Ian Romanick
>
> OpenGL ES implementations are not allowed to ship ARB extensions, and
> OpenGL implementations are not allowed to ship OES extensions.
>
> The functionality is also included in GL_ARB_ES2_compatibility. Eve
On 24/01/17 05:44 AM, Samuel Pitoiset wrote:
> Useful when debugging applications which map too much VRAM.
Is the number of mapped buffers really useful, as opposed to the total
size of buffer mappings? Even if it was the latter though, it doesn't
show which mappings are for BOs in VRAM vs GTT, do
On Tue, 2017-01-24 at 07:54 +0200, Tapani Pälli wrote:
> Hi Timothy;
>
> On 01/24/2017 01:12 AM, Timothy Arceri wrote:
> > From: Timothy Arceri
> >
>
> 8<
>
> > +
> > +static void
> > +write_uniforms(struct blob *metadata, struct gl_shader_program
> > *prog)
> > +{
> > + blob_write_uint32(me
We no longer allocate a miptree for the mcs_buf, so this is not a useful
assertion.
While here, move the CCS disabling so that we only conditionally shut it
off. This helps us dtrt later when CCS is used in more places.
Recommended-by: Topi Pohjolainen
Signed-off-by: Ben Widawsky
---
src/mesa/
v2: Try to keep the assert as recommended by Topi. This requires
modifying the num_samples check to be <= 1 because internally created
buffers set num_samples = 0.
v3: Buffers are proactively marked as scanout, often, and so checking
is_scanout in whether or not the buffer supports non-msrt fast c
Upper layers of the code will have the need to specify full or partial
resolves (more on this in the next patch). This code simply adds the new
enums and plumbs it in as minimally as necessary.
Signed-off-by: Ben Widawsky
Acked-by: Daniel Stone
Reviewed-by: Topi Pohjolainen
---
src/mesa/driver
In the foreseeable future it doesn't seem to make sense to have multiple
resolve flags. What does make sense is to have the caller give an
indication to the lower layers what it things should be done for
resolve. The enum change distinguishes this binary selection.
v2: Make setting the hint more c
Signed-off-by: Ben Widawsky
Acked-by: Daniel Stone
Reviewed-by: Topi Pohjolainen
---
src/mesa/drivers/dri/i965/brw_blorp.c | 3 ++-
src/mesa/drivers/dri/i965/brw_blorp.h | 3 ++-
src/mesa/drivers/dri/i965/intel_mipmap_tree.c | 4 ++--
3 files changed, 6 insertions(+), 4 deletion
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@norris2:~/intel-gfx/kmscube (modifiers $)
~/scripts/measure_bandwidth.s
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
Acked-by: Daniel Stone
Reviewed-by: Topi Pohjolainen
---
src/mesa/dri
On Gen9 hardware, the display engine is able to scanout a compressed
framebuffer by providing an offset to auxiliary compression information.
Unfortunately, the hardware is incapable of doing the same thing for the
fast clear color.
To mitigate this, the hardware introduced a new resolve type call
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 bisectable issues if libdrm or kernel does
something incorrectly.
S
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/intel_screen.c | 5 -
1 file changed, 4 insertions(+), 1 deletion(-)
diff --git a/src/mesa/drivers/dr
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 Widawsky
Acked-by: Daniel Stone
---
src/mesa/drivers/dri/i965/brw_co
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
b/src/mesa/drivers/dri/i965/intel_screen.c
index 971013f2dd..85070bb54
Make the code only disable CCS when it has to, unlike before where it
disabled CCS and enabled it when it could. This is much more inline with
how it should work in a few patches, where we have fewer restrictions as
to when we disable CCS.
Signed-off-by: Ben Widawsky
---
src/mesa/drivers/dri/i96
This provides a common function or creating miptrees when there is an
existing DRIimage to use. That provides an easy way to add CCS
allocation.
Signed-off-by: Ben Widawsky
Acked-by: Daniel Stone
---
src/mesa/drivers/dri/i965/intel_fbo.c | 17 -
src/mesa/drivers/dri/i965
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
1 file changed, 49 insertions(+), 15 deletions(-
Previously our aux buffers (MCS, and HiZ) never had an offset because
they were in their own buffer object. When using the CCS lossless
compression feature, it's desirable to store the data at an offset from
the main framebuffer, ie. share a buffer object. This patch just makes
having an aux offset
This patch provides the support (and comments) for allocating the BO
with space for the CCS buffer just underneath it.
This patch was originally titled:
"i965: Create correctly sized mcs for an image"
In order to make things more bisectable, reviewable, and to have the
CCS_MODIFIER token saved fo
This patch adds support for handling X tiled modifier. This isn't
particularly useful but it makes our code complete.
Signed-off-by: Ben Widawsky
Acked-by: Daniel Stone
---
src/mesa/drivers/dri/i965/intel_screen.c | 11 +++
1 file changed, 7 insertions(+), 4 deletions(-)
diff --git a/s
This code will disable actually creating these buffers for the scanout,
but it puts the allocation in place.
Primarily this patch is split out for review, it can be squashed in
later if preferred.
v2:
assert(mt->offset == 0) in ccs creation (as requested by Topi)
Remove bogus is_scanout check in
This doesn't really "do" anything because the default tiling for the
winsys buffer is X tiled. We do however want the X tiled modifier to
work correctly from the API perspective, which would imply that if you
set this modifier, and later do a get_modifier, you get back at least X
tiled.
Cc: Kristi
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_MODIFIER_LOWER
The query extension only supported 32b queries, and modi
This will expose all the modifier based APIs used by GBM.
This patch was split out for v3 of the series.
Signed-off-by: Ben Widawsky
---
src/mesa/drivers/dri/i965/intel_screen.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/src/mesa/drivers/dri/i965/intel_screen.c
b/src/m
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 implementation can create an image with the appropriate
attributes.
A gette
At image creation create a path for dealing with the linear modifier.
This works exactly like the old usage flags where __DRI_IMAGE_USE_LINEAR
was specified.
During development of this patch series, it was decided that a lack of
modifier was an insufficient way to express the required modifiers. A
Modifiers will be obtains or guessed by the client and passed in during
image creation/import.
This requires bumping the DRIimage version.
As of this patch, the modifiers aren't plumbed all the way down, this
patch simply makes sure the interface level stuff is correct.
v2: Don't allow usage + m
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 entirely ignored. It
shouldn't actually be used until this point beca
I intend to need to get to the devinfo structure, and storing the screen
is an easy way to do that.
It seems to be the consensus that you cannot share an image between
multiple screens.
Scape-goat: Rob Clark
Signed-off-by: Ben Widawsky
Reviewed-by: Eric Engestrom
Acked-by: Daniel Stone
---
s
v2: Other way round... to make consistent, make both return type have
the fixed width - uint32_t.
Cc: Daniel Stone
Signed-off-by: Ben Widawsky
Reviewed-by: Eric Engestrom
Acked-by: Daniel Stone
---
src/gbm/main/gbm.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/src/
This will be used by clients that need to know the number of planes
allocated for them on behalf of the GL or other API. The best current
example of this is when an extra "plane" is allocated to store
compression data for the primary plane.
v2: Return 1 for cases where there is no image, ie. dumb
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 when getting offsets we consider dumb images (Daniel)
Signed-off-by:
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 count preventing bad index succeeding (Daniel)
Signed-off-by: Ben Widawsky
Reviewed-b
v2: Make the error return be -1 instead of 0 because I think 0 is
actually valid.
v3: Set errno to EINVAL when the specified plane is above the total
planes. (Jason Ekstrand)
Return the bo's handle if there is no image ie. for dumb images like cursor
(Daniel)
Signed-off-by: Ben Widawsky
Acked-b
This will be used so we can query information per plane.
Signed-off-by: Ben Widawsky
Reviewed-by: Eric Engestrom
Acked-by: Daniel Stone
---
src/gbm/backends/dri/gbm_dri.c | 7 +++
src/gbm/main/gbm.c | 2 +-
src/gbm/main/gbmint.h | 1 +
3 files changed, 9 insertions(+),
Other things are out of order, but I need to add a getter so I'm just
fixing those.
This helps people adding to GBM know where the right place to put things
is.
Signed-off-by: Ben Widawsky
Reviewed-by: Eric Engestrom
Acked-by: Daniel Stone
---
src/gbm/main/gbm.c | 22 +++---
1
I've merged the first few patches from v2. I'd like to merge the GBM stuff at
least this time around. Most of it has review but enough has changed that it
probably needs eyes on it all again.
Here is the branch:
https://cgit.freedesktop.org/~bwidawsk/mesa/log/?h=modifiers
Since v2 I've gotten GET
Third-party can put their backend to a directory configured with
'--with-gbm-backenddir' and create a /etc/gbm.conf.d/*.conf file
which contains the backend so file name to overwrite the default
builtin DRI backend.
The /etc/gbm.conf.d/*.conf will be sorted and the backends added
will be tried one
Hi Timothy;
On 01/24/2017 01:12 AM, Timothy Arceri wrote:
From: Timothy Arceri
8<
+
+static void
+write_uniforms(struct blob *metadata, struct gl_shader_program *prog)
+{
+ blob_write_uint32(metadata, prog->SamplersValidated);
+ blob_write_uint32(metadata, prog->data->NumUniformStorage
All the other calls to retrieve the attachment have been covered except
this one - return the proper error for attachment points that are valid
enums but out of bound for the driver.
Fixes GL45-CTS.geometry_shader.layered_fbo.fb_texture_invalid_attachment
Signed-off-by: Ilia Mirkin
---
src/mesa
When a texture is immutable, we can't tack on extra levels
after-the-fact like we could with glTexImage. So check against that
level limit and return an error if it's surpassed.
The spec is a little unclear in that it says to check if "level is not a
supported texture level", however that is never
*** This is a test, please discard ***
Hi,
I've been trying to send a patch lately and it seems that it does not
get through the mailing list filters properly, since it does not have
the [Mesa-dev] prefix added to its subject nor appear in patchwork.
After talking to some people working for
*** This is a test, please discard ***
Hi,
I've been trying to send a patch lately and it seems that it does not
get through the mailing list filters properly, since it does not have
the [Mesa-dev] prefix added to its subject nor appear in patchwork.
After talking to some people working for
*** This is a test, please discard ***
Hi,
I've been trying to send a patch lately and it seems that it does not
get through the mailing list filters properly, since it does not have
the [Mesa-dev] prefix added to its subject nor appear in patchwork.
After talking to some people working for
On Mon, 2017-01-23 at 12:43 -0500, Romain Failliot wrote:
> 2017-01-23 0:06 GMT-05:00 Timothy Arceri :
> > I can confirm a similar problem. I have the same card and also got
> > a 4K
> > monitor recently. For me I was running a game in windowed mode (F1
> > 2015
> > I think) for a very short amount
On Mon, Jan 23, 2017 at 10:51 PM, Ilia Mirkin wrote:
> There are variants of the instructions that don't have legacy versions
> that we don't prevent from being selected. As such, I think this
> approach is doomed.
>
> Not for inclusion upstream.
> ---
>
> On top of everything, totally untested. A
There are variants of the instructions that don't have legacy versions
that we don't prevent from being selected. As such, I think this
approach is doomed.
Not for inclusion upstream.
---
On top of everything, totally untested. And also unsure I got the instruction
patching logic right - just goi
From: Timothy Arceri
Previously the constant array would not get copy propagated until the backend
did its GLSL IR opt loop. I plan on removing that from i965 shortly which
caused huge regressions in Deus-ex and Tomb Raider which have large
constant arrays. Moving lowering before the opt loop in
This matches the behavior of most other drivers, including nouveau.
Signed-off-by: Ilia Mirkin
---
Untested. This also leaves RCP and RSQ with the clamped variants. I suspect
this is wrong, but seems unrelated to multiplication per se.
src/gallium/drivers/r600/r600_shader.c | 36 ++
Signed-off-by: Ilia Mirkin
---
Untested. Can be verified with Xnine. It should pass before 1/2 of this series,
start failing with it, and pass again with 2/2 in place.
src/gallium/drivers/r600/r600_pipe.c | 2 +-
src/gallium/drivers/r600/r600_shader.c | 20 +---
2 files chang
https://bugs.freedesktop.org/show_bug.cgi?id=97102
--- Comment #13 from Bruce Cherniak ---
I pushed the change that corrects the crash you were seeing. This is the same
as Tim's suggestion to prune empty nodes.
I don't know all the other topologies well enough to suggest a change that
heuristic
Mesa 12.0.6 is now available.
Note: This is an extra release for the 12.0 stable branch, as per developers'
feedback. There are no plans to have further 12.0.x releases. Users are
encouraged to migrate to the 13.0 series in order to obtain future fixes.
In this release we have:
Patch for the CSO
From: Dave Airlie
bf8e1f9e7 radv: generate entrypoints from vk.xml
ripped out the ability to add extra header files (and writing
extra xml files is a bit messier). For some non-public
development or even developing future EXT/MESA extensions,
it would be nice to allow this option before things la
On Mon, Jan 23, 2017 at 05:42:44PM -0800, Jason Ekstrand wrote:
> On Mon, Jan 23, 2017 at 5:40 PM, Nanley Chery wrote:
>
> > On Mon, Jan 23, 2017 at 05:30:22PM -0800, Jason Ekstrand wrote:
> > > On Mon, Jan 23, 2017 at 4:55 PM, Nanley Chery
> > wrote:
> > >
> > > > Compressing a render target an
On Mon, Jan 23, 2017 at 5:40 PM, Nanley Chery wrote:
> On Mon, Jan 23, 2017 at 05:30:22PM -0800, Jason Ekstrand wrote:
> > On Mon, Jan 23, 2017 at 4:55 PM, Nanley Chery
> wrote:
> >
> > > Compressing a render target and decompressing it in the same
> > > single-subpass render pass may waste band
On Mon, Jan 23, 2017 at 05:30:22PM -0800, Jason Ekstrand wrote:
> On Mon, Jan 23, 2017 at 4:55 PM, Nanley Chery wrote:
>
> > Compressing a render target and decompressing it in the same
> > single-subpass render pass may waste bandwidth. While this may be
> > beneficial in some circumstances, it
On Mon, Jan 23, 2017 at 4:55 PM, Nanley Chery wrote:
> Compressing a render target and decompressing it in the same
> single-subpass render pass may waste bandwidth. While this may be
> beneficial in some circumstances, it does not help in all.
>
> Cc: "13.0 17.0"
>
This doesn't really fix a bu
From: Dave Airlie
bf8e1f9e7 radv: generate entrypoints from vk.xml
ripped out the ability to add extra header files (and writing
extra xml files is a bit messier). For some non-public
development or even developing future EXT/MESA extensions,
it would be nice to allow this option before things la
Signed-off-by: Lionel Landwerlin
---
src/intel/vulkan/anv_device.c | 593 +
src/intel/vulkan/anv_formats.c | 42 +++
2 files changed, 398 insertions(+), 237 deletions(-)
diff --git a/src/intel/vulkan/anv_device.c b/src/intel/vulkan/anv_device.c
index f80
Hi
- Original Message -
> Fix the leak of some ctx fields in error path.
>
> Signed-off-by: Li Qiang
> ---
> src/gallium/auxiliary/tgsi/tgsi_sanity.c | 6 +-
> 1 file changed, 5 insertions(+), 1 deletion(-)
>
> diff --git a/src/gallium/auxiliary/tgsi/tgsi_sanity.c
> b/src/gallium/a
When parsing texture instruction, it doesn't stop if the
'cur' is ',', the loop variable 'i' will also be increased
and be used to index the 'inst.TexOffsets' array. This can lead
an oob access issue. This patch avoid this.
Signed-off-by: Li Qiang
---
src/gallium/auxiliary/tgsi/tgsi_text.c | 2 +
Hi
- Original Message -
> When parsing texture instruction, it doesn't stop if the
> 'cur' is ',', the loop variable 'i' will also be increased
> and be used to index the 'inst.TexOffsets' array. This can lead
> an oob access issue. This patch avoid this.
>
> Signed-off-by: Li Qiang
> --
2017-01-23 16:17 GMT+08:00 Marc-André Lureau :
> Hi
>
> - Original Message -
> > When parsing texture instruction, it doesn't stop if the
> > 'cur' is ',', the loop variable 'i' will also be increased
> > and be used to index the 'inst.TexOffsets' array. This can lead
> > an oob access iss
Fix the leak of some ctx fields in error path.
Signed-off-by: Li Qiang
---
src/gallium/auxiliary/tgsi/tgsi_sanity.c | 6 +-
1 file changed, 5 insertions(+), 1 deletion(-)
diff --git a/src/gallium/auxiliary/tgsi/tgsi_sanity.c
b/src/gallium/auxiliary/tgsi/tgsi_sanity.c
index f867925..c3d5094
This series increases Dota 2 frame rates by about 2.61% on a SKL GT4 at
1080p. The first patch contributes to most of the gain.
In the long-term, I think we'll want to rework our CCS implementation to
perform resolves upon layout transitions. This will reduce the bandwidth
consumed by resolves. I'
Cc: "13.0 17.0"
Signed-off-by: Nanley Chery
---
src/intel/vulkan/anv_blorp.c | 10 --
1 file changed, 8 insertions(+), 2 deletions(-)
diff --git a/src/intel/vulkan/anv_blorp.c b/src/intel/vulkan/anv_blorp.c
index 2edd0712c9..4a4b479116 100644
--- a/src/intel/vulkan/anv_blorp.c
+++ b/src
On Mon, Jan 23, 2017 at 4:53 PM, Chad Versace
wrote:
> On Mon 23 Jan 2017, Jason Ekstrand wrote:
> > In brw_blorp_copyteximage, we use the format from the render buffer.
> > This could be a combined depth/stencil format. In this case, we handle
> > stencil properly but we give blorp the wrong IS
Compressing a render target and decompressing it in the same
single-subpass render pass may waste bandwidth. While this may be
beneficial in some circumstances, it does not help in all.
Cc: "13.0 17.0"
Signed-off-by: Nanley Chery
---
src/intel/vulkan/genX_cmd_buffer.c | 6 --
1 file changed
These files belong to the vulkan loader.
---
src/intel/vulkan/Makefile.am | 8 +---
1 file changed, 5 insertions(+), 3 deletions(-)
diff --git a/src/intel/vulkan/Makefile.am b/src/intel/vulkan/Makefile.am
index fba3d66..95f276e 100644
--- a/src/intel/vulkan/Makefile.am
+++ b/src/intel/vulkan/
On Mon 23 Jan 2017, Jason Ekstrand wrote:
> In brw_blorp_copyteximage, we use the format from the render buffer.
> This could be a combined depth/stencil format. In this case, we handle
> stencil properly but we give blorp the wrong ISL format. Specifically,
> we would give blorp ISL_FORMAT_R32G3
From: Dave Airlie
bf8e1f9e7 radv: generate entrypoints from vk.xml
ripped out the ability to add extra header files (and writing
extra xml files is a bit messier). For some non-public
development or even developing future EXT/MESA extensions,
it would be nice to allow this option before things la
On Mon, Jan 23, 2017 at 4:25 PM, Chad Versace
wrote:
> On Mon 23 Jan 2017, Jason Ekstrand wrote:
> > On Mon, Jan 23, 2017 at 3:31 PM, Chad Versace
> wrote:
> >
> > On Mon 23 Jan 2017, Jason Ekstrand wrote:
> > > On Mon, Jan 23, 2017 at 2:28 PM, Chad Versace <
> chadvers...@chromium.org>
On Mon 23 Jan 2017, Jason Ekstrand wrote:
> On Mon, Jan 23, 2017 at 3:31 PM, Chad Versace
> wrote:
>
> On Mon 23 Jan 2017, Jason Ekstrand wrote:
> > On Mon, Jan 23, 2017 at 2:28 PM, Chad Versace
> wrote:
> >
> > Implement each vkFoo2KHR() by trivially passing it through
Apologies if you are wondering where this is used. This patch really
should be before patch 24.
___
mesa-dev mailing list
mesa-dev@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/mesa-dev
On Mon, Jan 23, 2017 at 3:31 PM, Chad Versace
wrote:
> On Mon 23 Jan 2017, Jason Ekstrand wrote:
> > On Mon, Jan 23, 2017 at 2:28 PM, Chad Versace
> wrote:
> >
> > Implement each vkFoo2KHR() by trivially passing it through to the
> > original vkFoo().
> >
> >
> > As I mentioned to Lionel
Bump. Requesting more review please.
On Fri 13 Jan 2017, Chad Versace wrote:
> This series depends on fence fd support in I915_GEM_EXECBUFFER2, which
> isn't upstream in libdrm nor the kernel yet. I tested this with kmscube
> on Skylake, and everything looked good to me.
>
> I pushed tags for th
On Fri 20 Jan 2017, Rafael Antognolli wrote:
> I have tested this series with the branches that you mentioned, and with
> piglit with the patches from my own branch:
>
> https://github.com/rantogno/piglit/tree/review/fences-v02
>
> Everything seems to work fine. You can add:
>
> Tested-by: Rafae
On Mon 23 Jan 2017, Jason Ekstrand wrote:
> On Mon, Jan 23, 2017 at 2:28 PM, Chad Versace
> wrote:
>
> Implement each vkFoo2KHR() by trivially passing it through to the
> original vkFoo().
>
>
> As I mentioned to Lionel when he wrote basically this exact same patch, I
> think
> that m
From: Timothy Arceri
Since we know how big the list will be we can allocate the storage
upfront.
---
src/compiler/glsl/shader_cache.cpp | 1 +
1 file changed, 1 insertion(+)
diff --git a/src/compiler/glsl/shader_cache.cpp
b/src/compiler/glsl/shader_cache.cpp
index 91b79a5..0e7deb1 100644
--- a
From: Timothy Arceri
If the shader cache falls back to doing a compile and link we need the
original FragData bindings as they could have changed after the program
was first linked.
---
src/compiler/glsl/shader_cache.cpp | 8
src/mesa/main/mtypes.h | 2 ++
src/mesa/main/sh
From: Timothy Arceri
Pointers will have different lengths so we simply create a different
sha1 for each platform.
In theory we should be able to share cached shaders as we cache all
pointer as uint64_t however if a pointer is ever added to one of the
structs we write to file with blob_write_byte
From: Timothy Arceri
This will allow us to make use of the hash_table_call_foreach()
function from outside the object.
---
src/util/string_to_uint_map.h | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/src/util/string_to_uint_map.h b/src/util/string_to_uint_map.h
index e05
From: Timothy Arceri
Also add an assert to catch buffer overflows.
---
src/compiler/glsl/shader_cache.cpp | 13 +++--
1 file changed, 11 insertions(+), 2 deletions(-)
diff --git a/src/compiler/glsl/shader_cache.cpp
b/src/compiler/glsl/shader_cache.cpp
index bb63da5..91b79a5 100644
---
From: Timothy Arceri
If the shader cache falls back to doing a compile and link we need the
original attribute bindings as they could have changed after the program
was first linked.
---
src/compiler/glsl/shader_cache.cpp | 15 +++
src/mesa/main/mtypes.h | 7 +++
src
From: Timothy Arceri
---
src/util/disk_cache.c | 4
1 file changed, 4 insertions(+)
diff --git a/src/util/disk_cache.c b/src/util/disk_cache.c
index da01be7..01fe4d3 100644
--- a/src/util/disk_cache.c
+++ b/src/util/disk_cache.c
@@ -142,6 +142,10 @@ disk_cache_create(void)
struct stat
Am 23.01.2017 um 20:39 schrieb Tobias Droste:
> Am Montag, 23. Januar 2017, 11:53:18 CET schrieb Jose Fonseca:
>> On 20/01/17 02:48, Emil Velikov wrote:
>>> On 19 January 2017 at 19:26, Tobias Droste wrote:
Am Mittwoch, 18. Januar 2017, 18:45:04 CET schrieb Emil Velikov:
> On 18 January 2
From: Timothy Arceri
We already pull these values from the metadata cache so no need to
recreate them.
---
src/compiler/glsl/linker.cpp | 20
src/mesa/main/shaderobj.c| 8 +---
2 files changed, 17 insertions(+), 11 deletions(-)
diff --git a/src/compiler/glsl/linker
From: Timothy Arceri
---
src/compiler/glsl/linker.cpp | 62 +++-
src/mesa/main/shaderobj.c| 16 +++-
2 files changed, 42 insertions(+), 36 deletions(-)
diff --git a/src/compiler/glsl/linker.cpp b/src/compiler/glsl/linker.cpp
index 3271a19..705
From: Timothy Arceri
This will allow us to skip certain things when falling back to
a full recompile on a cache miss such as avoiding reinitialising
uniforms.
In this chage we use it to avoid reading the program metadata
from the cache and skipping linking during a fallback.
---
src/compiler/gl
From: Timothy Arceri
A number of things can happen that change the shader source after it is
compiled or linked.
For example:
- Source changed after it is first compiled
- Source changed after linking
- Shader detached after linking
In order to be able to fallback to a full rebuild on a cache m
From: Timothy Arceri
Here we skip the recreation of uniform storage if we are relinking
after a cache miss. This is improtant because uniform values may
have already been set by the application and we don't want to reset
them.
---
src/compiler/glsl/link_uniforms.cpp | 31 +++-
From: Timothy Arceri
Also remove cache item and fallback to full recompile if current Mesa
version differs.
V2: don't leak buffer
---
src/compiler/glsl/shader_cache.cpp | 13 +
1 file changed, 13 insertions(+)
diff --git a/src/compiler/glsl/shader_cache.cpp
b/src/compiler/glsl/sha
From: Timothy Arceri
These may be lowered constant arrays or uniform values that we set before
linking
so we need to cache the actual uniform values.
---
src/compiler/glsl/shader_cache.cpp | 33 +
1 file changed, 33 insertions(+)
diff --git a/src/compiler/glsl/s
From: Timothy Arceri
---
src/compiler/glsl/link_uniforms.cpp | 2 +-
src/compiler/glsl/linker.h | 3 +++
2 files changed, 4 insertions(+), 1 deletion(-)
diff --git a/src/compiler/glsl/link_uniforms.cpp
b/src/compiler/glsl/link_uniforms.cpp
index 8930d26..41fd79a 100644
--- a/src/compi
From: Timothy Arceri
We already have a reference.
---
src/compiler/glsl/linker.cpp | 3 ++-
src/mesa/main/shaderobj.c| 3 ++-
2 files changed, 4 insertions(+), 2 deletions(-)
diff --git a/src/compiler/glsl/linker.cpp b/src/compiler/glsl/linker.cpp
index fa9e154..d34ec97 100644
--- a/src/com
From: Timothy Arceri
---
src/compiler/glsl/shader_cache.cpp | 57 ++
1 file changed, 51 insertions(+), 6 deletions(-)
diff --git a/src/compiler/glsl/shader_cache.cpp
b/src/compiler/glsl/shader_cache.cpp
index b5468a9..9ed1c4e 100644
--- a/src/compiler/glsl/s
From: Timothy Arceri
The scenario is a program has been linked for the first time and we
cache the program metadata, then glLinkProgram() is called for a second
time. Since we will now retrieve the program metadata from cache we need
to skip linking.
---
src/compiler/glsl/shader_cache.cpp | 1 +
From: Timothy Arceri
---
src/compiler/glsl/shader_cache.cpp | 89 ++
1 file changed, 89 insertions(+)
diff --git a/src/compiler/glsl/shader_cache.cpp
b/src/compiler/glsl/shader_cache.cpp
index 29fc70a..9db4f25 100644
--- a/src/compiler/glsl/shader_cache.cpp
From: Timothy Arceri
---
src/compiler/glsl/shader_cache.cpp | 10 ++
1 file changed, 10 insertions(+)
diff --git a/src/compiler/glsl/shader_cache.cpp
b/src/compiler/glsl/shader_cache.cpp
index 9db4f25..358eb4f 100644
--- a/src/compiler/glsl/shader_cache.cpp
+++ b/src/compiler/glsl/shad
1 - 100 of 186 matches
Mail list logo
