Re: [Mesa-dev] [PATCH 8/8] radeonsi: make const and stream uploaders allocate read-only memory

For the series: Tested-by: Dieter Nützel on RX580 with UH, UV, Blender 2.79, glmark2 Shouldn't we set R600_DEBUG=sisched for UH, UV, Blender at least? Maybe general (with LLVM 5.0/6.0)? I'm seeing ~7-10 (> 10%) more fps with UH,UV and Blender (even with tess).

[Mesa-dev] [PATCH kmscube] Init and clean up VT settings

This puts VT input into raw (unbuffered) mode so that we can detect single key strokes. Also uses KD_GRAPHICS mode so that fbcon gets restored properly on exit and inhibits VT switching since we don't properly get/set drm master. Finally, handle signals and clean up if we catch one. --- common.c

Re: [Mesa-dev] [PATCH] docs/release-calendar: update and extend

On Thu, 2017-11-30 at 17:36 +0200, Andres Gomez wrote: > On Wed, 2017-11-29 at 18:20 +, Emil Velikov wrote: > > From: Emil Velikov > > > > Cc: Juan A. Suárez > > Cc: Andres Gomez > > Signed-off-by: Emil Velikov

Re: [Mesa-dev] [PATCH v4 21/44] i965/fs: Use byte scattered read for 16-bit load_ssbo

Reviewed-by: Jason Ekstrand On Wed, Nov 29, 2017 at 6:50 PM, Jose Maria Casanova Crespo < jmcasan...@igalia.com> wrote: > Used to enable 16-bit reads at do_untyped_vector_read, that is used on > the following intrinsics: > >* nir_intrinsic_load_shared >*

Re: [Mesa-dev] [PATCH 2/2] r600/atomic: add cayman version of atomic save/restore from GDS

On 01.12.2017 06:06, Dave Airlie wrote: From: Dave Airlie On Cayman we don't use the append/consume counters (fglrx doesn't) and they don't seem to work well with compute shaders. This just uses GDS instead to do the atomic operations. Interesting. This is kind of what

Re: [Mesa-dev] [PATCH v4 28/44] i965/fs: Use untyped_surface_read for 16-bit load_ssbo

On Wed, Nov 29, 2017 at 6:57 PM, Jose Maria Casanova Crespo < jmcasan...@igalia.com> wrote: > SSBO loads were using byte_scattered read messages as they allow > reading 16-bit size components. byte_scattered messages can only > operate one component at a time so we needed to emit as many messages

Re: [Mesa-dev] [PATCH v4 19/44] i965/fs: Use byte_scattered_write on 16-bit store_ssbo

I've left some comments below that I think clean things up and make this better, but I believe it is correct as-is. Reviewed-by: Jason Ekstrand On Wed, Nov 29, 2017 at 6:42 PM, Jose Maria Casanova Crespo < jmcasan...@igalia.com> wrote: > From: Alejandro Piñeiro

Re: [Mesa-dev] [PATCH v4 22/44] i965/fs: Helpers for un/shuffle 16-bit pairs in 32-bit components

On Thu, Nov 30, 2017 at 3:41 PM, Chema Casanova wrote: > > > On 30/11/17 23:21, Jason Ekstrand wrote: > > On Wed, Nov 29, 2017 at 6:50 PM, Jose Maria Casanova Crespo > > > wrote: > > > > This helpers are used to

Re: [Mesa-dev] [PATCH v4 27/44] i965/fs: Predicate byte scattered writes if needed

Please put this one somewhere before we turn the extension on in anv as it's needed for correctness. Reviewed-by: Jason Ekstrand wrote: > From: Alejandro Piñeiro > >

Re: [Mesa-dev] [PATCH 2/2] Implement WaClearTDRRegBeforeEOTForNonPS.

On Thu, Nov 30, 2017 at 04:50:19PM -0800, Rafael Antognolli wrote: > It looks like I forgot to prefix the subject with "intel/compiler:". > Fixed locally. > > On Thu, Nov 30, 2017 at 04:42:48PM -0800, Rafael Antognolli wrote: > > The bspec describes: > > > >"WA: Clear tdr register before

Re: [Mesa-dev] [PATCH v4 19/44] i965/fs: Use byte_scattered_write on 16-bit store_ssbo

On Fri, Dec 1, 2017 at 2:46 AM, Chema Casanova wrote: > On 01/12/17 11:12, Jason Ekstrand wrote: > > I've left some comments below that I think clean things up and make this > > better, but I believe it is correct as-is. > > > > Reviewed-by: Jason Ekstrand

Re: [Mesa-dev] [PATCH mesa] u_debug: do not compile asserts when they are disabled

On Thu, 30 Nov 2017 12:16:06 + Eric Engestrom wrote: > Commit f0ba7d897d1c22202531a added this code to expose asserts to the > compiler in an attempt to hide 'unused variable' warnings, incorrectly > claiming it was a no-op. This has two bad effects: > - any assert

Re: [Mesa-dev] [PATCH mesa] u_debug: do not compile asserts when they are disabled

On 2017-11-30 01:16 PM, Eric Engestrom wrote: > Commit f0ba7d897d1c22202531a added this code to expose asserts to the > compiler in an attempt to hide 'unused variable' warnings, incorrectly > claiming it was a no-op. This has two bad effects: > - any assert with side-effects are executed when

Re: [Mesa-dev] [PATCH v4 15/44] i965/fs: Define new shader opcode to set rounding modes

On Thu, Nov 30, 2017 at 03:07:59AM +0100, Jose Maria Casanova Crespo wrote: > From: Alejandro Piñeiro > > Although it is possible to emit them directly as AND/OR on brw_fs_nir, > having a specific opcode makes it easier to remove duplicate settings > later. > > v2: (Curro)

[Mesa-dev] [Bug 103699] Latest mesa breaks firefox on kde plasma with compositing on

https://bugs.freedesktop.org/show_bug.cgi?id=103699 --- Comment #26 from Tapani Pälli --- (In reply to Mario Kleiner from comment #25) > Another problem caused by the new Mesa srgb visuals -- and assigning srgb to > the special 32 bit composite overlay -- seems to be that the

Re: [Mesa-dev] [PATCH v4 19/44] i965/fs: Use byte_scattered_write on 16-bit store_ssbo

On 01/12/17 11:12, Jason Ekstrand wrote: > I've left some comments below that I think clean things up and make this > better, but I believe it is correct as-is. > > Reviewed-by: Jason Ekstrand > > > On Wed, Nov 29, 2017 at 6:42 PM, Jose Maria

Re: [Mesa-dev] [PATCH v4 18/44] i965/fs: Add byte scattered write message and fs support

On Thu, Nov 30, 2017 at 03:08:02AM +0100, Jose Maria Casanova Crespo wrote: > v2: (Jason Ekstrand) > - Enable bit_size parameter to scattered messages to enable different > bitsizes byte/word/dword. > - Remove use of brw_send_indirect_scattered_message in favor of >

Re: [Mesa-dev] [PATCH] radeon/vcn: determine idr by pic type

Am 30.11.2017 um 22:18 schrieb Leo Liu: On 11/30/2017 04:12 PM, boyuan.zh...@amd.com wrote: From: Boyuan Zhang Vaapi encode interface provides idr frame flags, where omx interface doesn't. Therefore, change to use picture type to determine idr frame, which will work

Re: [Mesa-dev] [PATCH] meta: Fix ClearTexture with GL_DEPTH_COMPONENT.

On 30.11.2017 19:53, Kenneth Graunke wrote: We only handled unpacking for GL_DEPTH_STENCIL formats. Cemu was hitting _mesa_problem() for an unsupported format in _mesa_unpack_float_32_uint_24_8_depth_stencil_row(), because the format was depth-only, rather than depth-stencil. Bugzilla:

Re: [Mesa-dev] [PATCH v4 26/44] i965/fs: Optimize 16-bit SSBO stores by packing two into a 32-bit reg

I left some cleanup comments below and one request for an additional comment. Reviewed-by: Jason Ekstrand On Wed, Nov 29, 2017 at 6:57 PM, Jose Maria Casanova Crespo < jmcasan...@igalia.com> wrote: > From: Eduardo Lima Mitev > > Currently, we use

Re: [Mesa-dev] [PATCH v4 11/44] i965: Support for 16-bit base types in helper functions

On Thu, Nov 30, 2017 at 03:07:55AM +0100, Jose Maria Casanova Crespo wrote: > v2: Fixed calculation of scalar size for 16-bit types. (Jason Ekstrand) > > Signed-off-by: Jose Maria Casanova Crespo > Signed-off-by: Eduardo Lima > Reviewed-by: Jason

Re: [Mesa-dev] [PATCH] st/va: Enable vaExportSurfaceHandle()

Am 01.12.2017 um 01:31 schrieb Mark Thompson: It will be present from libva 2.1 (VAAPI 1.1.0 or higher). Signed-off-by: Mark Thompson Acked-by: Christian König --- See:

Re: [Mesa-dev] [PATCH 0/5] i965: ASTC5x5 workaround

On Fri, Dec 1, 2017 at 10:06 AM, Rogovin, Kevin wrote: > Hi, > > Yes ANV will need something like this as well. If the GPU samples from both > an ASTC5x5 texture and one with an aux buffer without a texture cache > invalidate between such accesses, then the GPU hangs,

Re: [Mesa-dev] GBM and the Device Memory Allocator Proposals

On Fri, Dec 1, 2017 at 12:09 PM, Nicolai Hähnle wrote: > On 01.12.2017 16:06, Rob Clark wrote: >> >> On Thu, Nov 30, 2017 at 5:43 PM, Nicolai Hähnle >> wrote: >>> >>> Hi, >>> >>> I've had a chance to look a bit more closely at the allocator prototype >>>

[Mesa-dev] [PATCH 4/8] gallium/radeon: remove RADEON_HEAP_VRAM_GTT

From: Marek Olšák Only winsyses can set VRAM|GTT. Drivers shouldn't if they want to use winsys allocators. --- src/gallium/drivers/radeon/radeon_winsys.h | 10 ++ 1 file changed, 2 insertions(+), 8 deletions(-) diff --git

[Mesa-dev] [PATCH 5/8] winsys/amdgpu: add RADEON_FLAG_READ_ONLY

From: Marek Olšák --- src/gallium/drivers/radeon/radeon_winsys.h | 47 ++ src/gallium/winsys/amdgpu/drm/amdgpu_bo.c | 9 +- 2 files changed, 49 insertions(+), 7 deletions(-) diff --git a/src/gallium/drivers/radeon/radeon_winsys.h

[Mesa-dev] [PATCH 2/8] radeonsi: allow DMABUF exports for local buffers

From: Marek Olšák Cc: 17.3 --- src/gallium/drivers/radeon/r600_texture.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/src/gallium/drivers/radeon/r600_texture.c b/src/gallium/drivers/radeon/r600_texture.c index

[Mesa-dev] [PATCH 3/8] gallium/radeon: move setting VRAM|GTT into winsyses

From: Marek Olšák The combined VRAM|GTT heap will be removed. --- src/gallium/drivers/r600/r600_buffer_common.c | 14 -- src/gallium/drivers/radeon/r600_buffer_common.c | 14 -- src/gallium/winsys/amdgpu/drm/amdgpu_bo.c | 14 +-

[Mesa-dev] [PATCH 7/8] gallium/u_upload_mgr: allow drivers to specify pipe_resource::flags

From: Marek Olšák --- src/gallium/auxiliary/util/u_upload_mgr.c| 9 ++--- src/gallium/auxiliary/util/u_upload_mgr.h| 2 +- src/gallium/drivers/freedreno/a3xx/fd3_context.c | 2 +- src/gallium/drivers/freedreno/a4xx/fd4_context.c | 2 +-

[Mesa-dev] [PATCH 8/8] radeonsi: make const and stream uploaders allocate read-only memory

From: Marek Olšák and anything that clones these uploaders, like u_threaded_context. --- src/gallium/drivers/radeon/r600_pipe_common.c | 6 -- 1 file changed, 4 insertions(+), 2 deletions(-) diff --git a/src/gallium/drivers/radeon/r600_pipe_common.c

[Mesa-dev] [PATCH 6/8] radeonsi: make IBs and shader binaries read-only for the GPU

From: Marek Olšák Now Mesa can't corrupt them from the GPU. --- src/gallium/drivers/radeon/r600_buffer_common.c | 3 +++ src/gallium/drivers/radeon/r600_pipe_common.h | 1 + src/gallium/drivers/radeonsi/si_shader.c| 8 +---

[Mesa-dev] [PATCH 1/8] radeonsi: flush the context after resource_copy_region for buffer exports

From: Marek Olšák Cc: 17.2 17.3 --- src/gallium/drivers/radeon/r600_texture.c | 14 -- 1 file changed, 12 insertions(+), 2 deletions(-) diff --git a/src/gallium/drivers/radeon/r600_texture.c

Re: [Mesa-dev] GBM and the Device Memory Allocator Proposals

On 01.12.2017 18:09, Nicolai Hähnle wrote: [snip] As for the actual transition API, I accept that some metadata may be required, and the metadata probably needs to depend on the memory layout, which is often vendor-specific. But even linear layouts need some transitions for caches. We probably

Re: [Mesa-dev] [PATCH 3/3] i965/gen6-7/sol: Bump primitive counter BO size.

Eero Tamminen writes: > Hi, > > Tested-By: Eero Tamminen > > On 18.11.2017 00:28, Francisco Jerez wrote: >> Improves performance of SynMark2 OglGSCloth by a further 9.65%±0.59% >> due to the reduction in overwraps of the primitive count

Re: [Mesa-dev] [PATCH 0/5] i965: ASTC5x5 workaround

Hi, For ANV I do not know as I have not really poked into its code. For i965, this patch series handles the situation as to what to do if a draw of dispatch compute accesses both an ASTC5x5 texture and a texture with an auxiliary buffer. It does this by checking if there are both such

Re: [Mesa-dev] [PATCH v2 14/32] vulkan/wsi: Do image creation in common code

On Tue 28 Nov 2017, Jason Ekstrand wrote: > This uses the mock extension created in a previous commit to tell the > driver that the image it's just been asked to create is, in fact, a > window system image with whatever assumptions that implies. There was a > lot of redundant code between the two

Re: [Mesa-dev] [PATCH] docs/release-calendar: update and extend

On 30 November 2017 at 15:36, Andres Gomez wrote: > On Wed, 2017-11-29 at 18:20 +, Emil Velikov wrote: >> From: Emil Velikov >> >> Cc: Juan A. Suárez >> Cc: Andres Gomez >> Signed-off-by: Emil Velikov

Re: [Mesa-dev] [PATCH] egl/android: Partially handle HAL_PIXEL_FORMAT_IMPLEMENTATION_DEFINED

On Fri, Dec 1, 2017 at 8:44 AM, Tomasz Figa wrote: > On Fri, Dec 1, 2017 at 11:20 PM, Rob Herring wrote: >> On Fri, Dec 1, 2017 at 7:30 AM, Robert Foss >> wrote: >>> On Thu, 2017-11-30 at 11:14 -0600, Rob Herring wrote: On

[Mesa-dev] [Bug 103814] incorrect dust rendering in hl2 without sisched

https://bugs.freedesktop.org/show_bug.cgi?id=103814 --- Comment #5 from Hleb Valoshka <375...@gmail.com> --- I have checked mesa snapshot build against llvm6 and no more able to reproduce this error. Debian's 17.1.5 is linked against llvm4, while 17.2.5 is linked against llvm5, so it looks like

Re: [Mesa-dev] V2 Initial GS NIR support for radeonsi

Commit "radeonsi: enable gs support for nir backend" also modifies st/mesa. If there are no piglit regressions for TGSI, all ac and radeonsi patches are: Reviewed-by: Marek Olšák Marek On Thu, Nov 30, 2017 at 5:47 AM, Timothy Arceri wrote: > On

Re: [Mesa-dev] GBM and the Device Memory Allocator Proposals

On 01.12.2017 16:06, Rob Clark wrote: On Thu, Nov 30, 2017 at 5:43 PM, Nicolai Hähnle wrote: Hi, I've had a chance to look a bit more closely at the allocator prototype repository now. There's a whole bunch of low-level API design feedback, but for now let's focus on the

[Mesa-dev] [PATCH 1/5] i965: define astc5x5 workaround infrastructure

From: Kevin Rogovin Some GEN's have a bug in the sample where if the sampler accesses a texture with an auxialry surface and an ASTC5x5 texture without having the texture cache invalidated between such accesses, then the GPU will hang. This patch defines the

[Mesa-dev] [PATCH 4/5] i965: use ASTC5x5 workaround in brw_draw

From: Kevin Rogovin Perform the ASTC5x5 workaround tasks for drawing; note that the function does not do anything and immediately returns if the bug is not present on the hardware. Signed-off-by: Kevin Rogovin ---

[Mesa-dev] [PATCH 0/5] i965: ASTC5x5 workaround

From: Kevin Rogovin This patch series implements a needed workaround for Gen9 for ASTC5x5 sampler reads. The crux of the work around is to make sure that the sampler does not read an ASTC5x5 texture and a surface with an auxilary buffer without having a texture cache

[Mesa-dev] [PATCH 2/5] i965: ASTC5x5 workaround logic for blorp

From: Kevin Rogovin Blorp will only read from an ASTC5x5 texture if it copies from such a surface, that can only if an application is fetching such pixels. Because an ASTC5x3 texture can never be a render target, we do not need to worry about blorp reading such surfaces

[Mesa-dev] [PATCH 3/5] i965: set ASTC5x5 workaround texture type tracking on texture validate

From: Kevin Rogovin One of the presteps in each draw (and compute) call is to validate the textures. This is the perfect place (since all texture units are looped through) to see if ASTC5x5 and/or textures with an auxilary surface are accessed by the GPU. Signed-off-by:

[Mesa-dev] [PATCH 5/5] i965: use ASTC5x5 workaround in brw_compute

From: Kevin Rogovin Perform the ASTC5x5 workaround tasks for compute; note that the function does not do anything and immediately returns if the bug is not present on the hardware. Signed-off-by: Kevin Rogovin ---

Re: [Mesa-dev] [PATCH 2/2] Implement WaClearTDRRegBeforeEOTForNonPS.

On Fri, Dec 01, 2017 at 10:33:26AM +0200, Pohjolainen, Topi wrote: > On Thu, Nov 30, 2017 at 04:50:19PM -0800, Rafael Antognolli wrote: > > It looks like I forgot to prefix the subject with "intel/compiler:". > > Fixed locally. > > > > On Thu, Nov 30, 2017 at 04:42:48PM -0800, Rafael Antognolli

Re: [Mesa-dev] [PATCH 0/5] i965: ASTC5x5 workaround

On Fri, Dec 1, 2017 at 12:19 PM, wrote: > From: Kevin Rogovin > > This patch series implements a needed workaround for Gen9 for ASTC5x5 > sampler reads. The crux of the work around is to make sure that the > sampler does not read an ASTC5x5

Re: [Mesa-dev] [PATCH 14/29] anv/cmd_buffer: Apply subpass flushes before set_subpass

On Fri, Dec 1, 2017 at 5:47 AM, Pohjolainen, Topi < topi.pohjolai...@gmail.com> wrote: > On Mon, Nov 27, 2017 at 07:06:04PM -0800, Jason Ekstrand wrote: > > This seems slightly more correct because it means that the flushes > > happen before any clears or resolves implied by the subpass

Re: [Mesa-dev] [PATCH v4 07/44] spirv/nir: Handle 16-bit types

On 30/11/17 21:24, Jason Ekstrand wrote: > I sprinkled a few mostly trivial comments below.  With those fixed, > > Reviewed-by: Jason Ekstrand > > > On Wed, Nov 29, 2017 at 6:07 PM, Jose Maria Casanova Crespo >

Re: [Mesa-dev] [PATCH 5/6] glx: Prepare driFetchDrawable for no-config contexts

On Tue, 2017-11-14 at 14:03 -0800, Ian Romanick wrote: > On 11/14/2017 12:13 PM, Adam Jackson wrote: > > diff --git a/src/glx/glx_pbuffer.c b/src/glx/glx_pbuffer.c > > index 933b5d9ecd..42e7996e37 100644 > > --- a/src/glx/glx_pbuffer.c > > +++ b/src/glx/glx_pbuffer.c > > @@ -272,7 +272,7 @@

Re: [Mesa-dev] [PATCH 08/22] i965/screen: Honor 'allow_rgb10_configs' option. (v2)

Not sure of the end results of discussion on this drirc option but personally I think it is good to have as temporary workaround. Patches 7,8 Reviewed-by: Tapani Pälli On 29.11.2017 06:20, Mario Kleiner wrote: Allows to prevent exposing RGB10 configs and visuals to

Re: [Mesa-dev] [PATCH 7/7] meson: fix deps and underlinkage of libGL

On 30/11/2017 16:46, Emil Velikov wrote: On 30 November 2017 at 15:13, Jon Turney wrote: On 29/11/2017 17:34, Dylan Baker wrote: [...] Maybe it's me that's missing something... There are references to functions provided by these libraries (xcb_glx, xcb, x11_xcb) in common code. Having a

Re: [Mesa-dev] [PATCH 20/22] nv50, nvc0: Support BGRX1010102 format for visuals.

On Fri, Dec 1, 2017 at 12:54 AM, Mario Kleiner wrote: > On 11/29/2017 04:38 PM, Ilia Mirkin wrote: >> >> Why is this required? Can't you just use the BGR10_A2 format directly? >> > > If i don't define this PIPE_FORMAT_B10G10R10X2_UNORM as "TD" = displayable, > then it

Re: [Mesa-dev] [PATCH 2/2] glsl: don't run intrastage array validation when the interface type is not an array

On 01.12.2017 08:20, Samuel Iglesias Gonsálvez wrote: On Thu, 2017-11-30 at 15:47 +0100, Nicolai Hähnle wrote: Can you add an explanation / spec quote for this? Yes. "We validate that the interface block array type's definition matches. However, the function could be previously called if an

Re: [Mesa-dev] [PATCH v4 28/44] i965/fs: Use untyped_surface_read for 16-bit load_ssbo

On 01/12/17 11:49, Jason Ekstrand wrote: > On Wed, Nov 29, 2017 at 6:57 PM, Jose Maria Casanova Crespo > > wrote: > > SSBO loads were using byte_scattered read messages as they allow > reading 16-bit size components. byte_scattered

Re: [Mesa-dev] [PATCH] st/va: Enable vaExportSurfaceHandle()

On Friday, 2017-12-01 00:31:48 +, Mark Thompson wrote: > It will be present from libva 2.1 (VAAPI 1.1.0 or higher). > > Signed-off-by: Mark Thompson > --- > See: > >

Re: [Mesa-dev] [PATCH 1/2] meson: install dri internal headers.

On Thursday, 2017-11-30 10:40:11 -0800, Dylan Baker wrote: > Reported-by: Marc Dietrich > Signed-off-by: Dylan Baker Both patches are: Reviewed-by: Eric Engestrom > --- > include/meson.build | 4 > 1 file changed, 4

[Mesa-dev] [PATCH] radv: fix a crash in radv_can_dump_shader()

module can be NULL, oops. Signed-off-by: Samuel Pitoiset --- src/amd/vulkan/radv_shader.h | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/src/amd/vulkan/radv_shader.h b/src/amd/vulkan/radv_shader.h index 91f2e7f2a1..f404c49ad5 100644 ---

Re: [Mesa-dev] [PATCH 06/22] i965/screen: Add XRGB2101010 and ARGB2101010 support for DRI3.

Reviewed-by: Tapani Pälli On 29.11.2017 06:20, Mario Kleiner wrote: Allow DRI3/Present buffer sharing for 10 bpc buffers. Otherwise composited desktops under DRI3 will only display black client areas for redirected windows. Signed-off-by: Mario Kleiner

Re: [Mesa-dev] [PATCH 02/22] i965: Support xrgb/argb2101010 formats for glx_texture_from_pixmap.

Reviewed-by: Tapani Pälli On 29.11.2017 06:20, Mario Kleiner wrote: Makes compositing under X11/GLX work. Signed-off-by: Mario Kleiner --- src/mesa/drivers/dri/i965/intel_tex_image.c | 12 ++-- 1 file changed, 10 insertions(+), 2

Re: [Mesa-dev] [PATCH 03/22] i965: Support accelerated blit for depth 30 formats. (v2)

Reviewed-by: Tapani Pälli On 29.11.2017 06:20, Mario Kleiner wrote: Extend intel_miptree_blit() to handle at least ARGB2101010 -> XRGB2101010, ARGB2101010 -> ARGB2101010, and XRGB2101010 -> XRGB2101010 via the BLT engine, but not XRGB2101010 -> ARGB2101010 yet. This

[Mesa-dev] [PATCH] i965/nir: do int64 lowering before optimization

Otherwise loop unrolling will fail to see the actual cost of the unrolling operations when the loop body contains 64-bit integer instructions, and very specially when the divmod64 lowering applies, since its lowering is quite expensive. Without this change, some in-development CTS tests for int64

Re: [Mesa-dev] [PATCH 1/2] gallium/util: add u_transfer_helper

On Thu, Nov 30, 2017 at 10:33 AM, Rob Clark wrote: > On Thu, Nov 30, 2017 at 10:04 AM, Nicolai Hähnle wrote: >> On 29.11.2017 14:48, Rob Clark wrote: >>> >>> Add a new helper that drivers can use to emulate various things that >>> need special handling in

Re: [Mesa-dev] [PATCH 01/22] i965/screen: Add basic support for rendering 10 bpc/depth 30 framebuffers. (v3)

IMO this patch should be moved (when committing) as patch .. 6? So that we would have the dri enablers in place. Otherwise when if/bisecting these configs will fail to initialize. On 29.11.2017 06:20, Mario Kleiner wrote: Expose formats which are supported at least back to Gen 5 Ironlake,

Re: [Mesa-dev] [PATCH] egl/android: Partially handle HAL_PIXEL_FORMAT_IMPLEMENTATION_DEFINED

On Thu, 2017-11-30 at 11:14 -0600, Rob Herring wrote: > On Thu, Nov 30, 2017 at 12:11 AM, Tapani Pälli m> wrote: > > > > > > On 11/30/2017 06:13 AM, Tomasz Figa wrote: > > > > > > On Thu, Nov 30, 2017 at 3:43 AM, Robert Foss > > ra.com> > > >

[Mesa-dev] [PATCH] glx/dri3: Remove unused deviceName variable

From: Vadym Shovkoplias deviceName string is declared, assigned and freed but actually never used in dri3_create_screen() function. Fixes: 2d94601582e ("Add DRI3+Present loader") Signed-off-by: Vadym Shovkoplias ---

[Mesa-dev] [PATCH v2 5/6] st/omx/tizonia: Add H.264 encoder

v2: Refactor out screen functions to st/omx Example Gstreamer pipeline : gst-launch-1.0 filesrc location=movie.mp4 ! qtdemux ! h264parse ! avdec_h264 ! videoconvert ! omxh264enc ! h264parse ! avdec_h264 ! videoconvert ! ximagesink --- src/gallium/state_trackers/omx/Makefile.sources| 4 +-

[Mesa-dev] [PATCH v2 2/6] st/omx/tizonia: Add --enable-omx-tizonia flag and build files

Allow only bellagio or tizonia to be used at the same time. Detect tizonia package config file Generate libomx_mesa.so and install it to libtizcore.pc::pluginsdir Only compile empty source (target.c) for now. GSoC Project link: https://summerofcode.withgoogle.com/projects/#4737166321123328 ---

[Mesa-dev] [PATCH v2 3/6] st/omx/tizonia: Add entrypoint

Adds base files for adding components --- .../state_trackers/omx/tizonia/Makefile.sources| 4 ++- .../state_trackers/omx/tizonia/entrypoint.c| 37 ++ .../state_trackers/omx/tizonia/entrypoint.h| 35 3 files changed, 75 insertions(+), 1

[Mesa-dev] [PATCH v2 6/6] st/omx/tizonia/h264d: Add EGLImage support

Example Gstreamer pipeline : MESA_ENABLE_OMX_EGLIMAGE=1 GST_GL_API=gles2 GST_GL_PLATFORM=egl gst-launch-1.0 filesrc location=movie.mp4 ! qtdemux ! h264parse ! omxh264dec ! glimagesink --- src/gallium/state_trackers/omx/Makefile.am | 1 +

Re: [Mesa-dev] [PATCH] glx/dri3: Remove unused deviceName variable

On Friday, 2017-12-01 13:23:02 +0200, vadim.shovkopl...@gmail.com wrote: > From: Vadym Shovkoplias > > deviceName string is declared, assigned and freed but actually > never used in dri3_create_screen() function. > > Fixes: 2d94601582e ("Add DRI3+Present

Re: [Mesa-dev] [PATCH] egl/android: Partially handle HAL_PIXEL_FORMAT_IMPLEMENTATION_DEFINED

On Fri, Dec 1, 2017 at 7:30 AM, Robert Foss wrote: > On Thu, 2017-11-30 at 11:14 -0600, Rob Herring wrote: >> On Thu, Nov 30, 2017 at 12:11 AM, Tapani Pälli > m> wrote: >> > >> > >> > On 11/30/2017 06:13 AM, Tomasz Figa wrote: >> > > >> > > On

Re: [Mesa-dev] [PATCH 12/22] egl/x11: Handle depth 30 drawables for EGL_KHR_image_pixmap.

Patches 9,10,11,12 LGTM Reviewed-by: Tapani Pälli On 29.11.2017 06:20, Mario Kleiner wrote: Enables eglCreateImageKHR() with target set to EGL_NATIVE_PIXMAP_KHR to handle color depth 30 X11 drawables. Note that in theory the drawable depth 32 case in the current

Re: [Mesa-dev] [PATCH 14/29] anv/cmd_buffer: Apply subpass flushes before set_subpass

On Mon, Nov 27, 2017 at 07:06:04PM -0800, Jason Ekstrand wrote: > This seems slightly more correct because it means that the flushes > happen before any clears or resolves implied by the subpass transition. After reading the next patch this patch seems incomplete both before and after. Next patch

Re: [Mesa-dev] [PATCH 7/7] meson: fix deps and underlinkage of libGL

On 1 December 2017 at 12:50, Jon Turney wrote: > On 30/11/2017 16:46, Emil Velikov wrote: >> >> On 30 November 2017 at 15:13, Jon Turney wrote: >>> >>> On 29/11/2017 17:34, Dylan Baker wrote: > > [...] >>> >>> >>> Maybe it's me that's missing something... >>> >>>

Re: [Mesa-dev] GBM and the Device Memory Allocator Proposals

On Thu, Nov 30, 2017 at 5:43 PM, Nicolai Hähnle wrote: > Hi, > > I've had a chance to look a bit more closely at the allocator prototype > repository now. There's a whole bunch of low-level API design feedback, but > for now let's focus on the high-level stuff first. > > Going

Re: [Mesa-dev] [PATCH] egl/x11: Remove unneeded free() on always null string

On Friday, 2017-12-01 17:08:53 +0200, vadim.shovkopl...@gmail.com wrote: > From: Vadym Shovkoplias > > In this condition dri2_dpy->driver_name string always equals > NULL, so call to free() is useless > > Signed-off-by: Vadym Shovkoplias

Re: [Mesa-dev] [PATCH] egl/x11: Remove unneeded free() on always null string

Hi Eric, Mostly by a static analysis tool. It found at least 7 issues with useless free() calls and other problems that probably should be fixed. Suggest please should I create one cumulative commit for this or it should be a separate commits ? On Fri, Dec 1, 2017 at 5:41 PM, Eric Engestrom

Re: [Mesa-dev] [PATCH] egl/android: Partially handle HAL_PIXEL_FORMAT_IMPLEMENTATION_DEFINED

On Fri, Dec 1, 2017 at 11:20 PM, Rob Herring wrote: > On Fri, Dec 1, 2017 at 7:30 AM, Robert Foss wrote: >> On Thu, 2017-11-30 at 11:14 -0600, Rob Herring wrote: >>> On Thu, Nov 30, 2017 at 12:11 AM, Tapani Pälli >> m> wrote:

[Mesa-dev] [PATCH] egl/x11: Remove unneeded free() on always null string

From: Vadym Shovkoplias In this condition dri2_dpy->driver_name string always equals NULL, so call to free() is useless Signed-off-by: Vadym Shovkoplias --- src/egl/drivers/dri2/platform_x11.c | 1 - 1 file changed, 1

[Mesa-dev] [PATCH] radv: do not print ASM to stderr when dumping shaders

Signed-off-by: Samuel Pitoiset --- src/amd/vulkan/radv_debug.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/src/amd/vulkan/radv_debug.c b/src/amd/vulkan/radv_debug.c index cdf8e7a114..c8794d06ea 100644 --- a/src/amd/vulkan/radv_debug.c +++

Re: [Mesa-dev] [PATCH 4/5] xlib: remove dummy GLX_MESA_set_3dfx_mode implementation

On 11/30/2017 12:21 PM, Emil Velikov wrote: On 30 November 2017 at 19:09, Ian Romanick wrote: Is xmesa.h something that apps could see? Removing stuff could, hypothetically, cause compilation problems... but also, app developers, fix your old crap. :) Some digging

Re: [Mesa-dev] [PATCH 0/5] i965: ASTC5x5 workaround

Hi, Yes ANV will need something like this as well. If the GPU samples from both an ASTC5x5 texture and one with an aux buffer without a texture cache invalidate between such accesses, then the GPU hangs, which in turn makes the system unresponsive for a few seconds until the kernel resets the

[Mesa-dev] [PATCH] i965: Disable regular fast-clears (CCS_D) on gen9+

This partially reverts commit 3e57e9494c2279580ad6a83ab8c065d01e7e634e which caused a bunch of GPU hangs on several Source titles. To date, we have no clue why these hangs are actually happening. This undoes the final effect of 3e57e9494c227 and gets us back to not hanging. Tested with Team

[Mesa-dev] [PATCH] meson: Install dri.pc file when building gallium dri drivers

Currently this pkg-config file is only installed if a classic dri driver is built. This is wrong, it should be installed if any dri driver is installed, which includes the gallium dri target. Reported-by: Marc Dietrich Signed-off-by: Dylan Baker ---

[Mesa-dev] [PATCH] i965/cnl: Avoid fast-clearing sRGB render buffers

Gen10 doesn't automatically decode the clear color of sRGB buffers. To get correct rendering, avoid fast-clearing such buffers for now. The driver now passes the following piglit tests: * spec@arb_framebuffer_srgb@msaa-fast-clear * spec@ext_texture_srgb@multisample-fast-clear gl_ext_texture_srgb

Re: [Mesa-dev] [PATCH v4 18/44] i965/fs: Add byte scattered write message and fs support

On 30/11/17 21:42, Jason Ekstrand wrote: > On Wed, Nov 29, 2017 at 6:08 PM, Jose Maria Casanova Crespo > > wrote: > > v2: (Jason Ekstrand) >     - Enable bit_size parameter to scattered messages to enable > different >      

[Mesa-dev] [Bug 104024] xinit hangs on black screen with cursor, kernel oopses , bisected

https://bugs.freedesktop.org/show_bug.cgi?id=104024 Bug ID: 104024 Summary: xinit hangs on black screen with cursor, kernel oopses , bisected Product: Mesa Version: git Hardware: Other OS: All

Re: [Mesa-dev] GBM and the Device Memory Allocator Proposals

On Fri, 1 Dec 2017 13:38:41 -0500 Rob Clark wrote: > On Fri, Dec 1, 2017 at 12:09 PM, Nicolai Hähnle wrote: > > On 01.12.2017 16:06, Rob Clark wrote: > >> > >> On Thu, Nov 30, 2017 at 5:43 PM, Nicolai Hähnle > >> wrote: > >>> >

[Mesa-dev] [PATCH 1/3] meson: Reformat meson code to match more common style

Generally in our meson build large arrays are formated in the form: [ ..., ..., ..., $ ..., ] So use that form Signed-off-by: Dylan Baker --- src/glx/meson.build | 13 - 1 file changed, 8 insertions(+), 5 deletions(-) diff --git a/src/glx/meson.build

[Mesa-dev] [PATCH 2/3] meson: fix underlinkage without dri3

There are some case where the dri3 loader is covering for underlinkage for GLX and EGL, provide the linkage that they actually need. Signed-off-by: Dylan Baker --- src/egl/meson.build | 2 +- src/glx/meson.build | 3 ++- 2 files changed, 3 insertions(+), 2 deletions(-)

[Mesa-dev] [PATCH 3/3] meson: Fix overlinkage of dri3 loader

This was covering for underinkage elsewhere. With that fixed these can be removed. Signed-off-by: Dylan Baker --- src/loader/meson.build | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) diff --git a/src/loader/meson.build b/src/loader/meson.build index

[Mesa-dev] [PATCH 0/3] Fix linkage for libEGL and libGLX without dri3

As we discussed elsewhere, this should fix the linkage of the dri3 loader and glx and egl. Dylan Baker (3): meson: Reformat meson code to match more common style meson: fix underlinkage without dri3 meson: Fix overlinkage of dri3 loader src/egl/meson.build| 2 +- src/glx/meson.build

[Mesa-dev] [PATCH] util: Add backtrace support based on gallium u_debug_stack

Only the libunwind support is retained from u_debug_stack. Signed-off-by: Jordan Justen --- src/util/Makefile.am | 4 +- src/util/Makefile.sources | 2 + src/util/backtrace.c | 223 ++ src/util/backtrace.h

[Mesa-dev] [PATCH] i965: Serialize nir later in the linking process

Fixes MESA_GLSL=cache_fb with piglit tests/spec/glsl-1.50/execution/geometry/clip-distance-vs-gs-out.shader_test Fixes: 0610a624a12 i965/link: Serialize program to nir after linking for shader cache Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=103988 Signed-off-by: Jordan Justen

Re: [Mesa-dev] [PATCH v4 18/44] i965/fs: Add byte scattered write message and fs support

Assuming the changes you describe below, this patch would get a Reviewed-by: Jason Ekstrand I think that's enough for you to land the UBO/SSBO portion of 16bit storage. Feel free to send v2 versions of any patches you want me to look at again before you push but I think

Re: [Mesa-dev] [PATCH] i965/cnl: Avoid fast-clearing sRGB render buffers

On Fri, Dec 01, 2017 at 02:44:42PM -0800, Nanley Chery wrote: > Gen10 doesn't automatically decode the clear color of sRGB buffers. To > get correct rendering, avoid fast-clearing such buffers for now. > > The driver now passes the following piglit tests: > *

Re: [Mesa-dev] [PATCH] i965: Serialize nir later in the linking process

Based on what you pasted on IRC the other day, I'm guessing the problem was the missing unify_interfaces(). In any case, this looks much better. The closer serialization happens to back-end shader compilation, the better. Reviewed-by: Jason Ekstrand On Fri, Dec 1, 2017

  1   2   >