Re: [Mesa-dev] [PATCH v5 1/2] anv/cmd_buffer: consider multiview masks for tracking pending clear aspects

On Tue, 2018-03-27 at 09:06 -0700, Jason Ekstrand wrote: > I'm sorry I've been so incredibly out-to-lunch on reviewing this. :-( > > On Tue, Mar 27, 2018 at 12:53 AM, Iago Toral Quiroga om> wrote: > > When multiview is active a subpass clear may only clear a subset of > > the > > > > attachment

Re: [Mesa-dev] [PATCH 42/61] nir: Add a concept of per-member structs and a lowering pass

So I've been thinking about structs and I'm pretty sure we should be able to write some passes to completely lower them away. vertex shader inputs, buffer block and shader interface blocks cannot contain structs so it seems to me the only blocker is the way we assign uniform and varying locati

[Mesa-dev] [PATCH] st: Don't try to finalize the texture in st_render_texture().

We can't necessarily finalize the texture at this point if we're rendering to a texture image whose format is different from the baselevel's format. This was introduced as a fix for fbo-incomplete-texture-03 in de414f491526610bb260c73805c81ba413388e20, but the later fix for vmware on that testcase

Re: [Mesa-dev] [PATCH mesa] meson/configure: detect endian.h instead of trying to guess when it's available

On Tue, Mar 27, 2018 at 07:36:17PM +0100, Emil Velikov wrote: > On 25 March 2018 at 09:06, Jonathan Gray wrote: > > On Wed, Mar 21, 2018 at 05:09:17PM +, Eric Engestrom wrote: > >> Cc: Maxin B. John > >> Cc: Khem Raj > >> Suggested-by: Jon Turney > >> Signed-off-by: Eric Engestrom > >> ---

[Mesa-dev] [Bug 105738] commit f7ffa504a065dc2631fd38cc5fe885b277f4e7e7 causes artifacting in radv

https://bugs.freedesktop.org/show_bug.cgi?id=105738 --- Comment #5 from gloriouseggr...@gmail.com --- heya sorry for the late reply, just got the time to check this, and yes, it does fix the issue! On Tue, Mar 27, 2018 at 5:28 PM, wrote: > *Comment # 4

[Mesa-dev] [PATCH] radeonsi/nir: fix explicit component packing for geom/tess doubles

--- src/gallium/drivers/radeonsi/si_shader.c | 19 +++ 1 file changed, 11 insertions(+), 8 deletions(-) diff --git a/src/gallium/drivers/radeonsi/si_shader.c b/src/gallium/drivers/radeonsi/si_shader.c index 8ae742c93f6..d5607a99d32 100644 --- a/src/gallium/drivers/radeonsi/si_sha

[Mesa-dev] [Bug 105775] F1 2017 crashes on GCN 1.0 cards

https://bugs.freedesktop.org/show_bug.cgi?id=105775 Bug ID: 105775 Summary: F1 2017 crashes on GCN 1.0 cards Product: Mesa Version: 17.3 Hardware: x86-64 (AMD64) OS: Linux (All) Status: NEW Severity: block

[Mesa-dev] [Bug 105755] Mesa freezes when the GLSL shader contains a `for` loop with an uninitialized `i` index/counter variable

https://bugs.freedesktop.org/show_bug.cgi?id=105755 --- Comment #15 from Roland Scheidegger --- (In reply to Ilia Mirkin from comment #14) > (In reply to Swyter from comment #12) > > detecting when the draw call/batch/job is idle spinning/stuck by having some > > kind of reasonable timeout and ki

[Mesa-dev] [PATCH] vc4: Fix out-of-tree build

Signed-off-by: Aaron Watry Cc: Eric Anholt --- I still have to build at least one drm driver with pci support, but I've at least managed an out-of-tree build with these changes. src/broadcom/Makefile.am | 2 ++ 1 file changed, 2 insertions(+) diff --git a/src/broadcom/Makefile.am b/src/broadco

[Mesa-dev] [Bug 105755] Mesa freezes when the GLSL shader contains a `for` loop with an uninitialized `i` index/counter variable

https://bugs.freedesktop.org/show_bug.cgi?id=105755 --- Comment #14 from Ilia Mirkin --- (In reply to Swyter from comment #12) > detecting when the draw call/batch/job is idle spinning/stuck by having some > kind of reasonable timeout and killing it with a descriptive message when > things go out

[Mesa-dev] [Bug 104553] mat4: m[i][j] incorrect result with row_major UBO

https://bugs.freedesktop.org/show_bug.cgi?id=104553 --- Comment #10 from Timothy Arceri --- (In reply to Alejandro Piñeiro (freenode IRC: apinheiro) from comment #9) > Found this bug when I was about to report the same problem. Some comments > below: > > (In reply to florian.will from comment #0

Re: [Mesa-dev] [PATCH] mesa: Advertise support for GL_MESA_program_binary_formats

On 2018-03-27 16:29:34, Ian Romanick wrote: > Do *all* drivers set binary formats count >= 1? Like... i915? I don't see the extension as implying that a driver supports >= 1 binary formats. Rather, the extension just defines what enum Mesa can use with the *_get_program_binary extensions. I real

Re: [Mesa-dev] [PATCH] intel/fs: Don't emit a des copy for image ops with has_dest == false

On 03/27/2018 04:35 PM, Ian Romanick wrote: > On 03/27/2018 04:29 PM, Jason Ekstrand wrote: >> This was causing us to walk dest_components times over a thing with no >> destination. This happened to work because all of the image intrinsics >> without a destination also happened to have dest_compon

[Mesa-dev] [Bug 105670] [regression][hang] Trine1EE hangs GPU after loading screen on Mesa3D-17.3 and later

https://bugs.freedesktop.org/show_bug.cgi?id=105670 --- Comment #19 from Timothy Arceri --- Ok here is a piglit test for the original GLSL IR bug: https://patchwork.freedesktop.org/patch/213332/ The NIR bug is a little harder to test reliably for since it depends on the order in which optimisat

[Mesa-dev] [Bug 105670] [regression][hang] Trine1EE hangs GPU after loading screen on Mesa3D-17.3 and later

https://bugs.freedesktop.org/show_bug.cgi?id=105670 i...@yahoo.com changed: What|Removed |Added Status|NEW |RESOLVED Resolution|---

Re: [Mesa-dev] [PATCH] intel/fs: Don't emit a des copy for image ops with has_dest == false

On Tue, Mar 27, 2018 at 7:35 PM, Ian Romanick wrote: > On 03/27/2018 04:29 PM, Jason Ekstrand wrote: >> This was causing us to walk dest_components times over a thing with no >> destination. This happened to work because all of the image intrinsics >> without a destination also happened to have d

Re: [Mesa-dev] [PATCH] intel/fs: Don't emit a des copy for image ops with has_dest == false

On Tue, Mar 27, 2018 at 4:33 PM, Matt Turner wrote: > On Tue, Mar 27, 2018 at 4:29 PM, Jason Ekstrand > wrote: > > This was causing us to walk dest_components times over a thing with no > > destination. This happened to work because all of the image intrinsics > > without a destination also hap

Re: [Mesa-dev] [PATCH] intel/fs: Don't emit a des copy for image ops with has_dest == false

On 03/27/2018 04:29 PM, Jason Ekstrand wrote: > This was causing us to walk dest_components times over a thing with no > destination. This happened to work because all of the image intrinsics > without a destination also happened to have dest_components == 0. We > shouldn't be reading dest_compon

Re: [Mesa-dev] [PATCH] intel/fs: Don't emit a des copy for image ops with has_dest == false

On Tue, Mar 27, 2018 at 4:29 PM, Jason Ekstrand wrote: > This was causing us to walk dest_components times over a thing with no > destination. This happened to work because all of the image intrinsics > without a destination also happened to have dest_components == 0. We > shouldn't be reading d

[Mesa-dev] [PATCH] intel/fs: Don't emit a des copy for image ops with has_dest == false

This was causing us to walk dest_components times over a thing with no destination. This happened to work because all of the image intrinsics without a destination also happened to have dest_components == 0. We shouldn't be reading dest_components if has_dest == false. --- src/intel/compiler/brw

Re: [Mesa-dev] [PATCH] mesa: Advertise support for GL_MESA_program_binary_formats

Do *all* drivers set binary formats count >= 1? Like... i915? On 03/27/2018 03:22 PM, Jordan Justen wrote: > This extension defines an enum that can be used with > GL_ARB_get_program_binary and GL_OES_get_program_binary. > > There is no reason for an application to check for this extension, > si

Re: [Mesa-dev] [PATCH] nir/intrinsics: Don't report negative dest_components

On Tue, Mar 27, 2018 at 7:17 PM, Jason Ekstrand wrote: > I have no idea why but having dest_components == -1 was causing a memory > leak somewhere. Without this, you can't get through a full shader-db > run without running out of memory. I guess something somewhere looking at dest_components but

[Mesa-dev] [PATCH] nir/intrinsics: Don't report negative dest_components

I have no idea why but having dest_components == -1 was causing a memory leak somewhere. Without this, you can't get through a full shader-db run without running out of memory. --- src/compiler/nir/nir_intrinsics_c.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/src/compile

Re: [Mesa-dev] [PATCH 3/8] vbo: Remove vbo_indirect_draw_func.

On Sun, Mar 25, 2018 at 2:41 PM, wrote: > From: Mathias Fröhlich > > Remove the vbo_indirect_draw_func vbo callback and make the default > implementation use the drivers main draw callback function directly. > This will be needed with the next changes when drivers without own main > drivers Draw

Re: [Mesa-dev] [PATCH 2/2 v2] nir: mako all the intrinsics

Quoting Ian Romanick (2018-03-27 12:21:36) > I haven't investigated the root cause yet, but something about this > patch *explodes* memory usage. My shader-db runs on an 8-thread, 16GB > RAM + 16GB swap system get OOM killed. Previous to this patch, RAM > usage would never exceed 3.1% (based on w

Re: [Mesa-dev] [PATCH] nvc0/ir: fix INTERP_* with indirect inputs

Reviewed-by: Karol Herbst On Sat, Mar 24, 2018 at 8:19 PM, Ilia Mirkin wrote: > There were two problems, both of which are fixed now: > - The indirect address was not being shifted by 4 > - The indirect address was being placed as an argument in the offset case > > This fixes some of the new i

Re: [Mesa-dev] [PATCH 2/2] st/glsl_to_nir: correctly handle arrays packed across multiple vars

For the series: Reviewed-by: Marek Olšák Marek On Thu, Mar 22, 2018 at 9:49 PM, Timothy Arceri wrote: > Fixes piglit test: > tests/spec/arb_enhanced_layouts/execution/component- > layout/vs-fs-array-interleave-range.shader_test > --- > src/mesa/state_tracker/st_glsl_to_nir.cpp | 24 +

[Mesa-dev] [PATCH] nv50/ir: add more advanced slct constant folding code

From: Karol Herbst shader-db changes: total instructions in shared programs : 5894114 -> 5887031 (-0.12%) total gprs used in shared programs: 666558 -> 666514 (-0.01%) total shared used in shared programs : 520416 -> 520416 (0.00%) total local used in shared programs : 53524 -> 53572 (0.09

Re: [Mesa-dev] [PATCH] nv50/ir: optimise slct(t, f, set) to mov(set) or not(set)

just noticed I sent out the wrong version of that patch... On Tue, Mar 27, 2018 at 10:50 PM, Karol Herbst wrote: > From: Karol Herbst > > helps mainly Feral-ported games > > changes in shader-db: > total instructions in shared programs : 3940749 -> 3935015 (-0.15%) > total gprs used in shared pr

Re: [Mesa-dev] [PATCH] nv50/ir: optimise slct(t, f, set) to mov(set) or not(set)

On Tue, Mar 27, 2018 at 11:04 PM, Ilia Mirkin wrote: > On Tue, Mar 27, 2018 at 4:50 PM, Karol Herbst wrote: >> From: Karol Herbst >> >> helps mainly Feral-ported games >> >> changes in shader-db: >> total instructions in shared programs : 3940749 -> 3935015 (-0.15%) >> total gprs used in shared

Re: [Mesa-dev] [PATCH] nv50/ir: optimise slct(t, f, set) to mov(set) or not(set)

On Tue, Mar 27, 2018 at 11:19 PM, Ian Romanick wrote: > It will be interesting to see if this still occurs after nouveau > finishes switching to NIR. There's a pattern in nir_opt_algebraic for this. > well there is no plan to switch to NIR for everything where we can use TGSI. > On 03/27/2018 0

Re: [Mesa-dev] [PATCH 3/3] ac/nir_to_llvm: fix component packing for double outputs

For the series: Reviewed-by: Marek Olšák Marek On Tue, Mar 20, 2018 at 11:50 PM, Timothy Arceri wrote: > We need to wait until after the writemask is widened before we > adjust it for component packing. > > Together with the previous patch this fixes a number of > arb_enhanced_layouts compone

[Mesa-dev] [PATCH] mesa: Advertise support for GL_MESA_program_binary_formats

This extension defines an enum that can be used with GL_ARB_get_program_binary and GL_OES_get_program_binary. There is no reason for an application to check for this extension, since the *_get_program_binary extensions will return whatever enum value is appropriate for that implementation. Neverth

Re: [Mesa-dev] [PATCH v4 13/18] i965/surface_state: Emit the clear color address instead of value.

On Tue, Mar 27, 2018 at 11:19:43AM -0700, Jason Ekstrand wrote: > On Tue, Mar 27, 2018 at 4:37 AM, Pohjolainen, Topi > > wrote: > > On Thu, Mar 08, 2018 at 08:49:06AM -0800, Rafael Antognolli wrote: > > On Gen10, when emitting the surface state, use the value stored in the > > clear

[Mesa-dev] [Bug 105738] commit f7ffa504a065dc2631fd38cc5fe885b277f4e7e7 causes artifacting in radv

https://bugs.freedesktop.org/show_bug.cgi?id=105738 --- Comment #4 from Samuel Pitoiset --- Can you try https://patchwork.freedesktop.org/patch/213115/ and confirm it fixes the issue? -- You are receiving this mail because: You are the QA Contact for the bug. You are the assignee for the bug.__

Re: [Mesa-dev] [PATCH] nv50/ir: optimise slct(t, f, set) to mov(set) or not(set)

It will be interesting to see if this still occurs after nouveau finishes switching to NIR. There's a pattern in nir_opt_algebraic for this. On 03/27/2018 01:50 PM, Karol Herbst wrote: > From: Karol Herbst > > helps mainly Feral-ported games > > changes in shader-db: > total instructions in sh

Re: [Mesa-dev] [PATCH] nv50/ir: optimise slct(t, f, set) to mov(set) or not(set)

On Tue, Mar 27, 2018 at 4:50 PM, Karol Herbst wrote: > From: Karol Herbst > > helps mainly Feral-ported games > > changes in shader-db: > total instructions in shared programs : 3940749 -> 3935015 (-0.15%) > total gprs used in shared programs: 481460 -> 481433 (-0.01%) > total local used in s

[Mesa-dev] [PATCH] nv50/ir: optimise slct(t, f, set) to mov(set) or not(set)

From: Karol Herbst helps mainly Feral-ported games changes in shader-db: total instructions in shared programs : 3940749 -> 3935015 (-0.15%) total gprs used in shared programs: 481460 -> 481433 (-0.01%) total local used in shared programs : 27481 -> 27513 (0.12%) total bytes used in shared

Re: [Mesa-dev] [PATCH v4 12/18] i965/blorp: Update the fast clear color address.

On Tue, Mar 27, 2018 at 11:16:37AM -0700, Jason Ekstrand wrote: > On Thu, Mar 8, 2018 at 8:49 AM, Rafael Antognolli > > wrote: > > On Gen10, whenever we do a fast clear, blorp will update the clear color > state buffer for us, as long as we set the clear color address > correctly. >

[Mesa-dev] [PATCH] android: Use new nir intrinsics python scripts

Fixes: 76dfed8ae2d5 ("nir: mako all the intrinsics") Signed-off-by: Stefan Schake Acked-by: Rob Clark --- src/compiler/Android.nir.gen.mk | 9 + 1 file changed, 9 insertions(+) diff --git a/src/compiler/Android.nir.gen.mk b/src/compiler/Android.nir.gen.mk index aaa2712..fa0707e 100644 -

Re: [Mesa-dev] [Mesa-stable] [PATCH 1/4] Revert "mesa: simplify _mesa_is_image_unit_valid for buffers"

On Tue, Mar 27, 2018 at 8:09 AM, Juan A. Suarez Romero wrote: > On Mon, 2018-03-19 at 20:55 -0400, Marek Olšák wrote: > > Reviewed-by: Marek Olšák > > > > I understand the R-b is only for first patch, not for the full series. > Yes. Marek ___ mesa-de

Re: [Mesa-dev] [PATCH v4 00/18] Use clear color address in surface state.

I finally got around to reading through this series again. Everything which I didn't comment on is Reviewed-by: Jason Ekstrand On Thu, Mar 8, 2018 at 8:48 AM, Rafael Antognolli < rafael.antogno...@intel.com> wrote: > New revision of this series. > > Major differences are: > - I tried to incor

Re: [Mesa-dev] [PATCH 2/2 v2] nir: mako all the intrinsics

I haven't investigated the root cause yet, but something about this patch *explodes* memory usage. My shader-db runs on an 8-thread, 16GB RAM + 16GB swap system get OOM killed. Previous to this patch, RAM usage would never exceed 3.1% (based on watching top during the run). So... > 60x memory usa

Re: [Mesa-dev] [PATCH] nir: fix generated nir_intrinsics.c for MSVC

Reviewed-by: Dylan Baker Quoting Rob Clark (2018-03-27 11:59:37) > Apparently it is not happy about things like: .foo = {} > > So skip over initializers for empty lists. > > Fixes: 76dfed8ae2d5c6c509eb2661389be3c6a25077df > Reported-by: Roland Scheidegger > Signed-off-by: Rob Clark > --- > s

Re: [Mesa-dev] [PATCH] nir: fix generated nir_intrinsics.c for MSVC

This looks alright to me, though I'm not sure why msvc seemingly didn't like the empty braces in the first place. Acked-by: Roland Scheidegger Am 27.03.2018 um 20:59 schrieb Rob Clark: > Apparently it is not happy about things like: .foo = {} > > So skip over initializers for empty lists. > >

Re: [Mesa-dev] [PATCH] glsl: remove GCC construct from VECN macro

Am 27.03.2018 um 20:40 schrieb Emil Velikov: > On 27 March 2018 at 00:08, Roland Scheidegger wrote: >> Am 27.03.2018 um 00:44 schrieb Rob Clark: >>> On Mon, Mar 26, 2018 at 6:38 PM, Rob Clark wrote: On Mon, Mar 26, 2018 at 6:07 PM, Roland Scheidegger wrote: > Am 26.03.2018 um 23:0

[Mesa-dev] [PATCH] nir: fix generated nir_intrinsics.c for MSVC

Apparently it is not happy about things like: .foo = {} So skip over initializers for empty lists. Fixes: 76dfed8ae2d5c6c509eb2661389be3c6a25077df Reported-by: Roland Scheidegger Signed-off-by: Rob Clark --- src/compiler/nir/nir_intrinsics_c.py | 4 1 file changed, 4 insertions(+) diff -

Re: [Mesa-dev] [PATCH] glsl: remove GCC construct from VECN macro

On Tue, Mar 27, 2018 at 2:40 PM, Emil Velikov wrote: > On 27 March 2018 at 00:08, Roland Scheidegger wrote: >> Am 27.03.2018 um 00:44 schrieb Rob Clark: >>> On Mon, Mar 26, 2018 at 6:38 PM, Rob Clark wrote: On Mon, Mar 26, 2018 at 6:07 PM, Roland Scheidegger wrote: > Am 26.03.201

[Mesa-dev] [ANNOUNCE] mesa 18.0.0

Mesa 18.0.0 is now available. This release consists of nearly 2700 commits from approximately 110 developers. To put it in perspective - that's slight decrease from Mesa 17.3.0. The top highlights include: - Cannonlake support on i965 and anv - GL 4.3/GLES 3.1 support on r600/evergreen with hw

Re: [Mesa-dev] [Mesa-stable] [PATCH 1/4] Revert "mesa: simplify _mesa_is_image_unit_valid for buffers"

On Tue, Mar 27, 2018 at 02:09:16PM +0200, Juan A. Suarez Romero wrote: > On Mon, 2018-03-19 at 20:55 -0400, Marek Olšák wrote: > > Reviewed-by: Marek Olšák > > > > I understand the R-b is only for first patch, not for the full series. > > Any chance for this series to be reviewed? This was nomi

Re: [Mesa-dev] [PATCH] glsl: remove GCC construct from VECN macro

On 27 March 2018 at 00:08, Roland Scheidegger wrote: > Am 27.03.2018 um 00:44 schrieb Rob Clark: >> On Mon, Mar 26, 2018 at 6:38 PM, Rob Clark wrote: >>> On Mon, Mar 26, 2018 at 6:07 PM, Roland Scheidegger >>> wrote: Am 26.03.2018 um 23:01 schrieb Rob Clark: > correct me if I'm wrong,

Re: [Mesa-dev] [PATCH mesa] meson/configure: detect endian.h instead of trying to guess when it's available

On 25 March 2018 at 09:06, Jonathan Gray wrote: > On Wed, Mar 21, 2018 at 05:09:17PM +, Eric Engestrom wrote: >> Cc: Maxin B. John >> Cc: Khem Raj >> Suggested-by: Jon Turney >> Signed-off-by: Eric Engestrom >> --- >> configure.ac| 1 + >> meson.build | 2 +- >> src/util/u

Re: [Mesa-dev] [PATCH v4 13/18] i965/surface_state: Emit the clear color address instead of value.

On Tue, Mar 27, 2018 at 4:37 AM, Pohjolainen, Topi < topi.pohjolai...@gmail.com> wrote: > On Thu, Mar 08, 2018 at 08:49:06AM -0800, Rafael Antognolli wrote: > > On Gen10, when emitting the surface state, use the value stored in the > > clear color entry buffer by using a clear color address in the

Re: [Mesa-dev] [PATCH v4 12/18] i965/blorp: Update the fast clear color address.

On Thu, Mar 8, 2018 at 8:49 AM, Rafael Antognolli < rafael.antogno...@intel.com> wrote: > On Gen10, whenever we do a fast clear, blorp will update the clear color > state buffer for us, as long as we set the clear color address > correctly. > > However, on a hiz clear, if the surface is already on

Re: [Mesa-dev] [PATCH v4 05/18] intel/isl: Add support to emit clear value address.

On Tue, Mar 27, 2018 at 4:31 AM, Pohjolainen, Topi < topi.pohjolai...@gmail.com> wrote: > On Thu, Mar 08, 2018 at 08:48:58AM -0800, Rafael Antognolli wrote: > > gen10 can emit the clear color by setting it on a buffer somewhere, and > > then adding only the address to the surface state. > > > > Th

Re: [Mesa-dev] [PATCH v2 3/5] mesa: simplify MESA_GL_VERSION_OVERRIDE behavior of API override

On 19 March 2018 at 00:41, Andres Gomez wrote: > From: Marek Olšák > > v2: > - Provide a correct explanation on the envvars documentation (Ian). > - Provide a more correct explanation on the function comments (Andres). > > Fixes: 2599b92eb97 ("mesa: allow forcing >=3.1 compatibility contexts >

Re: [Mesa-dev] [PATCH 3/3] egl/x11: Handle both depth 30 formats for eglCreateImage().

Hi Mario, On 12 March 2018 at 20:45, Mario Kleiner wrote: > We need to distinguish if a backing pixmap of a window is > XRGB2101010 or XBGR2101010, as different gpu hw supports > different formats. NVidia hw prefers XBGR, whereas AMD and > Intel are happy with XRGB. > > We use the red channel mas

[Mesa-dev] [AppVeyor] mesa mesa-18.0.0 #7304 completed

Build mesa 7304 completed Commit dceb1ce807 by Emil Velikov on 3/27/2018 4:19 PM: docs: Update 18.0.0 release notes\n\nNote: the file was originally 17.4.0, yet git stuggles to detect the\nmove :-\\n\nSigned-off-by: Emil Velikov Configure your notifi

Re: [Mesa-dev] [PATCH v4 02/18] genxml: Preserve fields that share dword space with addresses.

On Thu, Mar 8, 2018 at 8:48 AM, Rafael Antognolli < rafael.antogno...@intel.com> wrote: > Some instructions contain fields that are either an address or a value > of some type based on the content of other fields, such as clear color > values vs address. That works fine if these fields are in the

Re: [Mesa-dev] [PATCH 1/3] wayland-drm: Expose server-side xbgr2101010 and abgr2101010 formats.

Hi Ilia, On 14 March 2018 at 19:02, Ilia Mirkin wrote: > On Tue, Mar 13, 2018 at 5:30 AM, Daniel Stone wrote: >> On 12 March 2018 at 20:45, Mario Kleiner wrote: >>> This way the wayland server can signal support for these formats >>> to wayland EGL clients. This is currently used by nouveau for

[Mesa-dev] [AppVeyor] mesa master #7303 failed

Build mesa 7303 failed Commit 76dfed8ae2 by Rob Clark on 3/15/2018 10:42 PM: nir: mako all the intrinsics\n\nI threatened to do this a long time ago.. I probably *should* have done\nit a long time ago when there where many fewer intrinsics. But the\nsystem of

[Mesa-dev] [PATCH] nv50/ra: prefer def == src2 for mad/sad with immediates on nvc0

From: Karol Herbst This helps with the PostRALoadPropagation pass moving long immediates into FMA/MAD instructions. changes in shader-db: total instructions in shared programs : 5894114 -> 5886074 (-0.14%) total gprs used in shared programs: 666558 -> 666563 (0.00%) total shared used in shar

Re: [Mesa-dev] [Mesa-stable] [PATCH] i965/miptree: Fail gracefully when make_surface returns NULL

On 03/27/2018 04:49 AM, Juan A. Suarez Romero wrote: > On Thu, 2018-01-04 at 11:38 -0800, Ian Romanick wrote: >> On 01/03/2018 01:27 PM, Jason Ekstrand wrote: >>> This patch looks good in isolation and you can have my rb. However, >>> this all a bit concerning and it makes me wonder if we're paper

Re: [Mesa-dev] [PATCH v5 1/2] anv/cmd_buffer: consider multiview masks for tracking pending clear aspects

I'm sorry I've been so incredibly out-to-lunch on reviewing this. :-( On Tue, Mar 27, 2018 at 12:53 AM, Iago Toral Quiroga wrote: > When multiview is active a subpass clear may only clear a subset of the > attachment layers. Other subpasses in the same render pass may also > clear too and we wan

Re: [Mesa-dev] [PATCH 3/4] radeonsi: don't use the SPI barrier management bug workaround

On Tue, Mar 27, 2018 at 4:11 AM, Samuel Pitoiset wrote: > > > On 03/27/2018 02:39 AM, Marek Olšák wrote: > >> From: Marek Olšák >> >> --- >> src/gallium/drivers/radeonsi/si_shader.c | 5 + >> 1 file changed, 5 insertions(+) >> >> diff --git a/src/gallium/drivers/radeonsi/si_shader.c >> b/

Re: [Mesa-dev] [PATCH v3 4/8] intel: Handle firstvertex in an identical way to BaseVertex

Thanks for looking at this. On 27/03/18 03:06, Ian Romanick wrote: It looks like there are a couple places in brw_fs_nir.cpp (emit_system_values_block and fs_visitor::nir_emit_vs_intrinsic) that handle nir_intrinsic_load_base_instance but not nir_intrinsic_load_first_vertex. Should those cases

Re: [Mesa-dev] [PATCH] mesa: Inherit texture view multi-sample information from the original texture images.

LGTM. I guess we probably don't have much piglit coverage for texture_view + MSAA. Reviewed-by: Brian Paul On 03/27/2018 07:52 AM, Henri Verbeet wrote: Found running "The Witness" in Wine. Without this patch, texture views created on multi-sample textures would have a GL_TEXTURE_SAMPLES of 0

Re: [Mesa-dev] [Mesa-stable] [PATCH] egl/wayland: Make swrast display_sync the correct queue

On Thu, 2018-03-22 at 16:45 +, Eric Engestrom wrote: > On Thursday, 2018-03-22 15:28:49 +, Daniel Stone wrote: > > On 22 March 2018 at 15:20, Derek Foreman wrote: > > > commit 03dd9a88b0be17ff0ce91e92f6902a9a85ba584a introduced per surface > > > queues, but the display_sync for swrast_comm

[Mesa-dev] [PATCH] mesa: Inherit texture view multi-sample information from the original texture images.

Found running "The Witness" in Wine. Without this patch, texture views created on multi-sample textures would have a GL_TEXTURE_SAMPLES of 0. All things considered such views actually work surprisingly well, but when combined with (plain) multi-sample textures in a framebuffer object, the resulting

Re: [Mesa-dev] [PATCH] nvc0/ir: fix emiting NOTs with predicates

On Tue, Mar 27, 2018 at 9:24 AM, Karol Herbst wrote: > From: Karol Herbst > > Signed-off-by: Karol Herbst Reviewed-by: Ilia Mirkin > --- > src/gallium/drivers/nouveau/codegen/nv50_ir_emit_nvc0.cpp | 2 ++ > 1 file changed, 2 insertions(+) > > diff --git a/src/gallium/drivers/nouveau/codegen/

[Mesa-dev] [PATCH] nvc0/ir: fix emiting NOTs with predicates

From: Karol Herbst Signed-off-by: Karol Herbst --- src/gallium/drivers/nouveau/codegen/nv50_ir_emit_nvc0.cpp | 2 ++ 1 file changed, 2 insertions(+) diff --git a/src/gallium/drivers/nouveau/codegen/nv50_ir_emit_nvc0.cpp b/src/gallium/drivers/nouveau/codegen/nv50_ir_emit_nvc0.cpp index e2c41a0

[Mesa-dev] [AppVeyor] mesa master #7301 completed

Build mesa 7301 completed Commit 1e0a06000b by Rob Clark on 3/26/2018 10:45 PM: glsl_types: fix build break with intel/msvc compiler\n\nThe VECN() macro was taking advantage of a GCC specific feature that is\nnot available on lesser compilers, mostly for the pu

[Mesa-dev] [Bug 105755] Mesa freezes when the GLSL shader contains a `for` loop with an uninitialized `i` index/counter variable

https://bugs.freedesktop.org/show_bug.cgi?id=105755 --- Comment #13 from Swyter --- I have opened a glslang issue here, I thought that cross-referencing both reports might be interesting, for completeness: https://github.com/KhronosGroup/glslang/issues/1315 -- You are receiving this mail becau

Re: [Mesa-dev] [Mesa-stable] [PATCH 1/4] Revert "mesa: simplify _mesa_is_image_unit_valid for buffers"

On Mon, 2018-03-19 at 20:55 -0400, Marek Olšák wrote: > Reviewed-by: Marek Olšák > I understand the R-b is only for first patch, not for the full series. Any chance for this series to be reviewed? This was nominated for stable, but wasn't reviewed yet, so I can't pick it for 17.3 J.A.

Re: [Mesa-dev] [PATCH] radv: Unset ZRANGE_PRECISION when depth was zeroed

On Thu, 2018-03-22 at 12:31 +, James Legg wrote: > On Thu, 2018-03-22 at 02:36 +0100, Bas Nieuwenhuizen wrote: > > On Thu, Mar 8, 2018 at 12:59 PM, James Legg > > wrote: > > > This avoids bug 105396 somehow. I suspect it is a VI and GFX9 hardware > > > bug which PAL calls WaTcCompatZRange, bu

Re: [Mesa-dev] [Mesa-stable] [PATCH] i965/miptree: Fail gracefully when make_surface returns NULL

On Thu, 2018-01-04 at 11:38 -0800, Ian Romanick wrote: > On 01/03/2018 01:27 PM, Jason Ekstrand wrote: > > This patch looks good in isolation and you can have my rb. However, > > this all a bit concerning and it makes me wonder if we're papering over > > the real bug. > > It may be a bit of both.

Re: [Mesa-dev] [PATCH v4 06/18] intel/blorp: Add support for fast clear address.

On Thu, Mar 08, 2018 at 08:48:59AM -0800, Rafael Antognolli wrote: > On gen10+, if surface->clear_color_addr is present, use it directly > intead of copying it to the surface state. > > v4: Remove redundant #if clause for GEN <= 10 (Jason) > > Signed-off-by: Rafael Antognolli > Reviewed-by: Jaso

[Mesa-dev] [Bug 105755] Mesa freezes when the GLSL shader contains a `for` loop with an uninitialized `i` index/counter variable

https://bugs.freedesktop.org/show_bug.cgi?id=105755 --- Comment #12 from Swyter --- Yeah. That was it, I only use a few shaders and they are all compiled at startup. Good to know that caching is implemented Mesa-wide now. > *EE: [! OGL LOG ] type: 0, local: 0, gpr: 28, inst: 134, bytes: 1432 > *

Re: [Mesa-dev] [PATCH v4 13/18] i965/surface_state: Emit the clear color address instead of value.

On Thu, Mar 08, 2018 at 08:49:06AM -0800, Rafael Antognolli wrote: > On Gen10, when emitting the surface state, use the value stored in the > clear color entry buffer by using a clear color address in the surface > state. > > v4: Use the clear color offset from the clear_color_bo, when available.

Re: [Mesa-dev] [PATCH v4 05/18] intel/isl: Add support to emit clear value address.

On Thu, Mar 08, 2018 at 08:48:58AM -0800, Rafael Antognolli wrote: > gen10 can emit the clear color by setting it on a buffer somewhere, and > then adding only the address to the surface state. > > This commit add support for that on isl_surf_fill_state, and if that is > requested, skip setting th

[Mesa-dev] [REBASED PATCH v5 16/21] clover/spirv: Add functions for validating SPIR-V binaries

Signed-off-by: Pierre Moreau --- This patch was not modified, but no longer applied cleanly after modifying “clover/llvm: Allow translating from SPIR-V to LLVM IR” in the v6. src/gallium/state_trackers/clover/Makefile.am | 15 ++- src/gallium/state_trackers/clover/Makefile.sources | 4 +

[Mesa-dev] [PATCH v6 14/21] clover/llvm: Allow translating from SPIR-V to LLVM IR

Signed-off-by: Pierre Moreau --- v6: Update Makefile.am to add to libclllvm its new llvm-spirv dependency. (Karol Herbst) src/gallium/state_trackers/clover/Makefile.am | 6 - .../state_trackers/clover/llvm/invocation.cpp | 29 ++ .../state_trackers/clover/l

[Mesa-dev] [REBASED PATCH v5 09/21] clover: Disallow creating libraries from other libraries

If creating a library, do not allow non-compiled object in it, as executables are not allowed, and libraries would make it really hard to enforce the "-enable-link-options" flag. Reviewed-by: Francisco Jerez Reviewed-by: Aaron Watry Signed-off-by: Pierre Moreau --- This patch was not modified,

[Mesa-dev] [PATCH v6 07/21] clover/api: Rework the validation of devices for building

Signed-off-by: Pierre Moreau --- v6: * Drop the `valid_devs` argument to `validate_build_common()` (Francisco Jerez) * Change `clLinkProgram()` to initialise `prog`’s devices prior to calling `validate_build_common()`. src/gallium/state_trackers/clover/api/program.cpp | 18 +-

Re: [Mesa-dev] [MOSTLY RESEND PATCH 0/8] Emit SPIR-V pre-pre-work

On 27/03/18 04:00, Ian Romanick wrote: > Most of these patches have been on the list at least once. Major > changes are: > > - Patches 1, 6, and 7 had a bunch of changes due to rebasing. > > - Patches 2 had some minor changes due to rebasing. > > - Patches 3 and 8 are new. > > Only patches 4 and 5

Re: [Mesa-dev] [RFC] Mesa 17.3.x release problems and process improvements

On Thu, 2018-03-22 at 16:24 -0700, Dylan Baker wrote: > Quoting Ilia Mirkin (2018-03-22 15:16:18) > > On Thu, Mar 22, 2018 at 6:00 PM, Dylan Baker wrote: > > > Quoting Ilia Mirkin (2018-03-21 17:39:14) > > > > Just one bit of feedback, for the rest I either agree or have no > > > > opinion: > > >

Re: [Mesa-dev] [RFC] Mesa 17.3.x release problems and process improvements

On Thu, 2018-03-22 at 09:46 +0100, Juan A. Suarez Romero wrote: On Thu, 2018-03-22 at 00:28 +, Emil Velikov wrote: [...] > > > In detail: > > * make the patch queue, release date and blockers accessible at any > >point in time: [...] > > > > * patches with trivial conflicts can b

Re: [Mesa-dev] [PATCH 7/8] spirv: Move SPIR-V building to Makefile.spirv.am and spirv/meson.build

The am part looks good to me, although probably you want a second opinion: Reviewed-by: Alejandro Piñeiro On 27/03/18 04:00, Ian Romanick wrote: > From: Ian Romanick > > Future changes will add generated files used only from > src/compiler/glsl. These can't be built from Makefile.nir.am, and

Re: [Mesa-dev] [PATCH v3 0/8] freedreno: a2xx improvements

On Tue, Mar 27, 2018 at 5:53 AM, Wladimir J. van der Laan wrote: > >> Patchset looks reasonable to me, from a quick look. Testing a22x >> isn't so easy without resurrecting an ancient downstream kernel for >> old snadragon devices, so I think it is ok to ignore that. If the > > Do you mean: > >

Re: [Mesa-dev] [PATCH 3/8] util: Use util_is_power_of_two_nonzero in u_vector

Reviewed-by: Alejandro Piñeiro On 27/03/18 04:00, Ian Romanick wrote: > From: Ian Romanick > > Previously size=0, element_size=0 would have been allowed. That > combination can only lead to despair. > > Signed-off-by: Ian Romanick > --- > src/util/u_vector.c | 4 ++-- > 1 file changed, 2 ins

Re: [Mesa-dev] [PATCH 1/8] util: Move util_is_power_of_two to bitscan.h and rename to util_is_power_of_two_or_zero

Reviewed-by: Alejandro Piñeiro On 27/03/18 04:00, Ian Romanick wrote: > From: Ian Romanick > > The new name make the zero-input behavior more obvious. The next > patch adds a new function with different zero-input behavior. > > Signed-off-by: Ian Romanick > Suggested-by: Matt Turner > --- >

Re: [Mesa-dev] [PATCH 8/8] i965: Silence unused parameter warning

Reviewed-by: Alejandro Piñeiro On 27/03/18 04:00, Ian Romanick wrote: > From: Ian Romanick > > src/mesa/drivers/dri/i965/brw_draw_upload.c: In function ‘double_types’: > src/mesa/drivers/dri/i965/brw_draw_upload.c:225:34: warning: unused parameter > ‘brw’ [-Wunused-parameter] > double_types(s

Re: [Mesa-dev] [PATCH v3 0/8] freedreno: a2xx improvements

> Patchset looks reasonable to me, from a quick look. Testing a22x > isn't so easy without resurrecting an ancient downstream kernel for > old snadragon devices, so I think it is ok to ignore that. If the Do you mean: - try to keep supporting a22x on best-effort basis by providing fallback b

Re: [Mesa-dev] [RFC] Mesa release improvements - Feature and Stable releases

On Wed, 2018-03-21 at 20:43 +, Emil Velikov wrote: > > All in all the idea sounds sane. As a summary/overall: > - Maintainer: open meta bug, send usual release plan + mention that > feature should be added to the tracker > - Developers: add bugs to tracker, or > - Developers: list via othe

Re: [Mesa-dev] [PATCH] i965: Extend the negative 32-bit deltas to 64-bits

Hello Chris and Emil. Could we make a final decision? Because I'm bit confused what you are expecting exactly. I would prefer to left initial patch as is (if its possible). On Tue, Mar 27, 2018 at 9:20 AM, Sergii Romantsov < sergii.romant...@gmail.com> wrote: > Hello Stuart, > Nice to hear. > An

Re: [Mesa-dev] [PATCH 3/4] radeonsi: don't use the SPI barrier management bug workaround

On 03/27/2018 02:39 AM, Marek Olšák wrote: From: Marek Olšák --- src/gallium/drivers/radeonsi/si_shader.c | 5 + 1 file changed, 5 insertions(+) diff --git a/src/gallium/drivers/radeonsi/si_shader.c b/src/gallium/drivers/radeonsi/si_shader.c index 8ae742c93f6..00ebbb9b0f2 100644 ---

Re: [Mesa-dev] [PATCH 1/4] ac/surface: set AddrSurfInfoIn.format = ADDR_FMT_8, add assertions

Tested-by: Samuel Pitoiset On 03/27/2018 02:39 AM, Marek Olšák wrote: From: Marek Olšák Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=105738 --- src/amd/common/ac_surface.c | 8 1 file changed, 8 insertions(+) diff --git a/src/amd/common/ac_surface.c b/src/amd/common/ac_

[Mesa-dev] [PATCH v5 1/2] anv/cmd_buffer: consider multiview masks for tracking pending clear aspects

When multiview is active a subpass clear may only clear a subset of the attachment layers. Other subpasses in the same render pass may also clear too and we want to honor those clears as well, however, we need to ensure that we only clear a layer once, on the first subpass that uses a particular la

[Mesa-dev] [PATCH v5 2/2] anv/cmd_buffer: honor pending clear views for depth/stencil attachments

v2: rebased on top of subpass rework. v3: rebased v4: - rebased - reset pending clear views in one go rather one bit at a time (Caio) --- src/intel/vulkan/genX_cmd_buffer.c | 22 +- 1 file changed, 21 insertions(+), 1 deletion(-) diff --git a/src/intel/vulkan/genX_cmd_buff

[Mesa-dev] [PATCH v3] i965: initialize SPIR-V capabilities

Needed for ARB_gl_spirv. Those are not the same that the Intel vulkan driver. From the ARB_spirv_extensions spec: "3. If a new GL extension is added that includes SPIR-V support via a new SPIR-V extension does it's SPIR-V extension also get enumerated by the SPIR_V_EXTENSIONS_ARB query?.

  1   2   >