Re: [Mesa-dev] [PATCH 5/7] util/disk_cache: hash pointer size and gpu name into cache keys

On Thu, Mar 16, 2017 at 3:14 PM, Timothy Arceri wrote: > > > On 17/03/17 00:00, Grazvydas Ignotas wrote: >> >> On Thu, Mar 16, 2017 at 2:49 AM, Timothy Arceri >> wrote: >>> >>> >>> >>> On 16/03/17 10:09, Grazvydas Ignotas wrote: This

Re: [Mesa-dev] [PATCH] i965: remove uneeded asm/unistd.h include

On 16 March 2017 at 02:48, Jonathan Gray wrote: > Fix the build on OpenBSD by removing an uneeded include for asm/unistd.h. > > Signed-off-by: Jonathan Gray Reviewed-by: Emil Velikov and push both of your patches. Thank you Jonathan,

Re: [Mesa-dev] [PATCH 1/3] dri: Add an image creation with modifiers

Hi Ben, On 13 March 2017 at 21:47, Ben Widawsky wrote: > On 17-03-10 10:28:42, Emil Velikov wrote: >> >> Hi Ben, >> >> Mostly pointing out a few things that look strange, pardon if some >> seem too pedantic. >> >> On 10 March 2017 at 01:48, Ben Widawsky

Re: [Mesa-dev] [PATCH] android: i965: generate code for OA counter queries

On 13 March 2017 at 22:36, Mauro Rossi wrote: > 2017-03-13 14:00 GMT+01:00 Emil Velikov : >> >> On 12 March 2017 at 23:01, Mauro Rossi wrote: >> > Automake generation rules are replicated for android. >> > $* macro was

[Mesa-dev] [PATCH 4/6] radeon_drm_bo: explicitly check return value of drmCommandWriteRead

CID 1313492 Signed-off-by: Julien Isorce --- src/gallium/winsys/radeon/drm/radeon_drm_bo.c | 9 +++-- 1 file changed, 7 insertions(+), 2 deletions(-) diff --git a/src/gallium/winsys/radeon/drm/radeon_drm_bo.c b/src/gallium/winsys/radeon/drm/radeon_drm_bo.c index

Re: [Mesa-dev] [PATCH 3/3] docs/releasing: document how to squash/announce queued patches

On 15 March 2017 at 19:05, Eric Engestrom wrote: > On Wednesday, 2017-03-15 17:34:26 +, Emil Velikov wrote: >> From: Emil Velikov >> >> In the odd case where a patch needs to be fixed, squash the appropriate >> fix and document how. Add

[Mesa-dev] [PATCH 3/6] si_pipe: do not dereference sscreen before checking if it is NULL.

CID 1354483 Signed-off-by: Julien Isorce --- src/gallium/drivers/radeonsi/si_pipe.c | 12 ++-- 1 file changed, 6 insertions(+), 6 deletions(-) diff --git a/src/gallium/drivers/radeonsi/si_pipe.c b/src/gallium/drivers/radeonsi/si_pipe.c index a9a7c8f..6cbaa72 100644

[Mesa-dev] [PATCH 6/6] si_descriptor: move velems nullity check before dereference

CID 1399479: Dereference before null check (REVERSE_INULL) check_after_deref: Null-checking velems suggests that it may be null, but it has already been dereferenced on all paths leading to the check. Signed-off-by: Julien Isorce ---

[Mesa-dev] [PATCH 5/6] si_blit: check zstex nullity

Few lines before there is a if (zstex && zstex->htile_buffer ...) which tells that there are some situation where it can be NULL. So let's check for its nullity in other places too. Spotted by coverity: CID 1237295 Signed-off-by: Julien Isorce ---

[Mesa-dev] [PATCH 1/6] radeon: initialize hole variable before calling container_of

Like in a few other places in that radeon_drm_bo.c file. CID 715739. Signed-off-by: Julien Isorce --- src/gallium/winsys/radeon/drm/radeon_drm_bo.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/src/gallium/winsys/radeon/drm/radeon_drm_bo.c

[Mesa-dev] [PATCH 0/6] Fixes a few radeon issues found by Coverity

CID 1399479: medium CID 1237295: medium CID 1313492: medium CID 1354483: medium CID 1250588: high CID 715739: high It is all more cosmetic than anything but worth to add some more checks. To Emil: thx for your reply about "question about container_of". Here is a cosmetic patch for it.

[Mesa-dev] [PATCH 2/6] r600_shader.c: check validity of the value returned by eg_get_interpolator_index

Like done in another place in that same file. Not sure if this should be an assert though. CID 1250588 Signed-off-by: Julien Isorce --- src/gallium/drivers/r600/r600_shader.c | 6 -- 1 file changed, 4 insertions(+), 2 deletions(-) diff --git

Re: [Mesa-dev] [PATCH] i965: avoid using a GNU make pattern rule

On 16 March 2017 at 02:49, Jonathan Gray wrote: > % pattern rules are a GNU extension. As there is only one file here > avoid patterns and globbing entirely to fix the build on non-GNU make. > > Signed-off-by: Jonathan Gray > --- >

[Mesa-dev] [PATCH] util/disk_cache: pass predicate functions file stats directly (v3)

Since switching to LRU eviction the only user of these predicate functions now resolves directory entry stats itself so pass them directly saving calling fstat and strlen twice (and the expensive strlen is skipped entirely if access time is newer). v2: Update for empty cache dir detection changes

Re: [Mesa-dev] MESA and KOTOR

Patch for implementing WGL_ARB_make_current_read attached. I can’t test it at the moment since I’m not near my Windows development environment. Let me know what you find. -Brian 0001-st-wgl-add-support-for-WGL_ARB_make_current_read.patch Description: Binary data > On Mar 15, 2017, at

Re: [Mesa-dev] [PATCH 04/12] genxml: Make MI_STORE_DATA_IMM have a single 64-bit data field

On Thu, Mar 16, 2017 at 3:24 AM, Lionel Landwerlin < lionel.g.landwer...@intel.com> wrote: > I might be wrong, but are you sure it's a good idea? > > The documentation says that dword0 is the lower 32bits of a qword and > dword1 the upper bits. > With this change we now make it look like it's a

Re: [Mesa-dev] [PATCH kmscube] cube-tex: Handle SMOOTH switch case

On 16 March 2017 at 00:27, Eric Engestrom wrote: > On Wednesday, 2017-03-15 18:56:53 -0300, Fabio Estevam wrote: >> In kmscube.c there is the following logic: >> >> if (mode == SMOOTH) { >> egl = init_cube_smooth(gbm); >> } else { >> egl

[Mesa-dev] [Bug 100236] Undefined symbols for architecture x86_64: "typeinfo for llvm::RTDyldMemoryManager"

https://bugs.freedesktop.org/show_bug.cgi?id=100236 Bug ID: 100236 Summary: Undefined symbols for architecture x86_64: "typeinfo for llvm::RTDyldMemoryManager" Product: Mesa Version: git Hardware: x86-64 (AMD64)

[Mesa-dev] [PATCH] radv: Fix using more than 4 bound descriptor sets

Avoid a buffer overflow in ac_nir_to_llvm.c's create_function when using more than 4 descriptor sets. radv claims support for 8. Cc: 17.0 --- src/amd/common/ac_nir_to_llvm.h | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git

Re: [Mesa-dev] [PATCH] i965/gen8+: Do full stall when switching pipeline

Thanks for tracking this down! Reviewed-by: Jason Ekstrand I'm not sure how we missed this when CS was brought up. Oh well. On Thu, Mar 16, 2017 at 1:24 AM, Topi Pohjolainen < topi.pohjolai...@gmail.com> wrote: > just as earlier gens do. > > CC: Jason Ekstrand

[Mesa-dev] [PATCH] mesa: more unused linux/version.h include

From: Emil Velikov The header provides the LINUX_VERSION_CODE and KERNEL_VERSION macros. With neither of which being used by any part of mesa. Signed-off-by: Emil Velikov --- src/mesa/x86/common_x86.c | 3 --- 1 file changed, 3

Re: [Mesa-dev] [PATCH] util/build-id: check dlpi_name before strstr call

On Wed, Mar 15, 2017 at 11:10 PM, Tapani Pälli wrote: > According to dl_iterate_phdr man page first object visited is the > main program where dlpi_name is an empty string. This fixes segfault > on Android when using build-id as identifier. If Android's libc returns

[Mesa-dev] [PATCH] radv/ac: Fix shared memory offset calculation

The index passed to get_shared_memory_ptr is an attribute slot index, i.e. the index of a vec4 within LDS. Therefore this must be scaled by sizeof(vec4) to give the LDS byte offset. Signed-off-by: Alex Smith --- src/amd/common/ac_nir_to_llvm.c | 2 +- 1 file

Re: [Mesa-dev] [PATCH] radv/ac: Fix shared memory offset calculation

On 16 March 2017 at 16:40, Alex Smith wrote: > The index passed to get_shared_memory_ptr is an attribute slot index, > i.e. the index of a vec4 within LDS. Therefore this must be scaled by > sizeof(vec4) to give the LDS byte offset. > Whoever is reviewing/merging this

Re: [Mesa-dev] [PATCH v3 07/24] i965/fs: generalize the legalization d2x pass

On 16/03/17 04:23, Francisco Jerez wrote: > Samuel Iglesias Gonsálvez writes: > >> On 04/03/17 00:19, Francisco Jerez wrote: >>> Samuel Iglesias Gonsálvez writes: >>> Add support to SEL instruction and add an assert to detect unsupported

Re: [Mesa-dev] [PATCH v3 07/24] i965/fs: generalize the legalization d2x pass

On 16/03/17 04:23, Francisco Jerez wrote: > Samuel Iglesias Gonsálvez writes: > >> On 04/03/17 00:19, Francisco Jerez wrote: >>> Samuel Iglesias Gonsálvez writes: >>> Add support to SEL instruction and add an assert to detect unsupported

[Mesa-dev] [PATCH] mesa: disable glthread when glNewList() is called

glNewList() swaps dispatch tables, and we don't have anything in place to handle that in glthread. --- src/mapi/glapi/gen/gl_API.xml | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/src/mapi/glapi/gen/gl_API.xml b/src/mapi/glapi/gen/gl_API.xml index 15d7e4f..c1f0f8f 100644 ---

[Mesa-dev] [PATCH] util/build-id: check dlpi_name before strstr call

According to dl_iterate_phdr man page first object visited is the main program where dlpi_name is an empty string. This fixes segfault on Android when using build-id as identifier. Signed-off-by: Tapani Pälli --- src/util/build_id.c | 6 ++ 1 file changed, 6

[Mesa-dev] nir isinf vs NAN handling

radv has failed the SPIR-V isinf tests for a good while, I've talked to Matt and the llvm backend matches on an ordered feq to infinity, where as we map the nir feq to unordered. 06:00 < arsenm> airlied: unordered eq for nan input would say true. isinf for NAN says false So it

Re: [Mesa-dev] [PATCH] mesa: disable glthread when glNewList() is called

On 16/03/17 03:01 PM, Timothy Arceri wrote: > glNewList() swaps dispatch tables, and we don't have anything in > place to handle that in glthread. This prevents three piglit tests from crashing with mesa_glthread=true. (Many more left though, as discussed on IRC) Tested-by: Michel Dänzer

Re: [Mesa-dev] [PATCH v3] anv: handle failures when growing reloc lists

On Wed, 2017-03-15 at 16:56 +0200, Pohjolainen, Topi wrote: > On Wed, Mar 15, 2017 at 01:03:37PM +0100, Iago Toral Quiroga wrote: > > > > Growing the reloc list happens through calling anv_reloc_list_add() > > or > > anv_reloc_list_append(). Make sure that we call these through > > helpers > >

Re: [Mesa-dev] [PATCH] genxml: Add XML version tags

Reviewed-by: Lionel Landwerlin On 16/03/17 05:56, Jason Ekstrand wrote: There's not much point to having them or not having them but this reduces some pointless diff from the version we can auto-generate --- src/intel/genxml/gen4.xml | 1 +

Re: [Mesa-dev] [PATCH 12/12] anv: implement pipeline statistics queries

On Wed, 2017-03-15 at 17:03 -0700, Jason Ekstrand wrote: > From: Ilia Mirkin > > In the end, pipeline statistics queries look a lot like occlusion > queries only with between 1 and 11 begin/end pairs being generated > instead of just the one. > --- >  

Re: [Mesa-dev] [PATCH 1/4] aubinator: Fix indentation.

Thanks for the cleanups! This series is : Reviewed-by: Lionel Landwerlin On 16/03/17 03:01, Kenneth Graunke wrote: Three space, not four. --- src/intel/tools/aubinator.c | 13 ++--- 1 file changed, 6 insertions(+), 7 deletions(-) diff --git

[Mesa-dev] [PATCH v2] android: fix segfault within swap_buffers

Function droid_swap_buffers may get called without dri2_surf->buffer set, in these cases we don't have a back buffer set either. Patch fixes segfault seen with 3DMark that uses android.opengl.GLSurfaceView for rendering it's UI. backtrace: #00 pc 00013f88 /system/lib/egl/libGLES_mesa.so

[Mesa-dev] [Bug 100223] marshal_generated.c:38:10: fatal error: 'X11/Xlib-xcb.h' file not found

https://bugs.freedesktop.org/show_bug.cgi?id=100223 --- Comment #2 from Tapani Pälli --- FYI we are getting this error also on Android where 'X11/Xlib-xcb.h' does not exist -- You are receiving this mail because: You are the assignee for the bug. You are the QA Contact for

[Mesa-dev] [PATCH] i965/gen8+: Do full stall when switching pipeline

just as earlier gens do. CC: Jason Ekstrand Cc: "17.0 13.0" Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=96743 Signed-off-by: Topi Pohjolainen --- src/mesa/drivers/dri/i965/brw_misc_state.c | 3

Re: [Mesa-dev] [PATCH v3] anv: handle failures when growing reloc lists

On Thu, Mar 16, 2017 at 08:56:30AM +0100, Iago Toral wrote: > On Wed, 2017-03-15 at 16:56 +0200, Pohjolainen, Topi wrote: > > On Wed, Mar 15, 2017 at 01:03:37PM +0100, Iago Toral Quiroga wrote: > > > > > > Growing the reloc list happens through calling anv_reloc_list_add() > > > or > > >

[Mesa-dev] [Bug 100223] marshal_generated.c:38:10: fatal error: 'X11/Xlib-xcb.h' file not found

https://bugs.freedesktop.org/show_bug.cgi?id=100223 Vinson Lee changed: What|Removed |Added CC|

[Mesa-dev] [PATCH] android: add marshal_generated c and h files to generated sources

Fixes: efd63e2 ("mesa: Connect the generated GL command marshalling code to the build.") Signed-off-by: Tapani Pälli --- src/mesa/Android.gen.mk | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/src/mesa/Android.gen.mk b/src/mesa/Android.gen.mk index

[Mesa-dev] [Bug 100223] marshal_generated.c:38:10: fatal error: 'X11/Xlib-xcb.h' file not found

https://bugs.freedesktop.org/show_bug.cgi?id=100223 --- Comment #3 from Tapani Pälli --- IMO there should be a check if X11 exists or maybe removal of include if this code does not utilize Xlib/Xcb? -- You are receiving this mail because: You are the assignee for the bug. You

Re: [Mesa-dev] [PATCH 08/11] st/va: clear the video surface on allocation

Christian König wrote: From: Christian König This makes debugging of decoding problems quite a bit easier. This breaks gstreamer encode for me. ffmpeg is OK, but then IIRC ffmpeg only uses one of something that gstreamer uses two of, not wishing to get too

[Mesa-dev] [Bug 92366] [BSW SKL] Regression: glx@glx-swap-event_async failed

https://bugs.freedesktop.org/show_bug.cgi?id=92366 cprigent changed: What|Removed |Added Assignee|ramix.ben.hass...@intel.com

Re: [Mesa-dev] [PATCH 7/7] util/disk_cache: also write key_blob to cache entry

On Thu, Mar 16, 2017 at 3:28 AM, Timothy Arceri wrote: > On 16/03/17 10:09, Grazvydas Ignotas wrote: >> >> This can be used to deal with key hash collisions from different >> version (should we find that to actually happen) and to find >> which mesa version produced it the

[Mesa-dev] [Bug 100075] st/va raises segmentation fault while using videotestsrc with gstreamer

https://bugs.freedesktop.org/show_bug.cgi?id=100075 --- Comment #1 from Julien Isorce --- Thx Gurkirpal for the report. The other day I was going further into https://bugzilla.gnome.org/show_bug.cgi?id=779642 and I wonder if it is not the same issue in the end, see its

Re: [Mesa-dev] [PATCH 6/7] util/disk_cache: add a write helper

On Thu, Mar 16, 2017 at 3:00 AM, Timothy Arceri wrote: > > > On 16/03/17 10:09, Grazvydas Ignotas wrote: >> >> Simplifies the write code a bit and handles EINTR. >> >> Signed-off-by: Grazvydas Ignotas >> --- >> src/util/disk_cache.c | 40

[Mesa-dev] [PATCH] st/mesa: call glthread_destroy() before _vbo_DestroyContext()

Otherwise we have a race condition between vbo calls in the glthread and the _vbo_DestroyContext() call. This fixes a bunch of piglit crashes. --- src/mesa/main/context.c | 2 -- src/mesa/state_tracker/st_context.c | 2 ++ 2 files changed, 2 insertions(+), 2 deletions(-) diff --git

Re: [Mesa-dev] [PATCH] util/build-id: check dlpi_name before strstr call

On 16 March 2017 at 06:10, Tapani Pälli wrote: > According to dl_iterate_phdr man page first object visited is the > main program where dlpi_name is an empty string. This fixes segfault > on Android when using build-id as identifier. > > Signed-off-by: Tapani Pälli

Re: [Mesa-dev] [PATCH v2] android: fix segfault within swap_buffers

On 16 March 2017 at 08:55, Tapani Pälli wrote: > Function droid_swap_buffers may get called without dri2_surf->buffer set, > in these cases we don't have a back buffer set either. Patch fixes segfault > seen with 3DMark that uses android.opengl.GLSurfaceView for rendering

Re: [Mesa-dev] [PATCH] android: add marshal_generated c and h files to generated sources

On 16 March 2017 at 09:24, Tapani Pälli wrote: > Fixes: efd63e2 ("mesa: Connect the generated GL command marshalling code to > the build.") > Signed-off-by: Tapani Pälli > --- > src/mesa/Android.gen.mk | 4 +++- > 1 file changed, 3 insertions(+),

Re: [Mesa-dev] [PATCH] mapi: remove Xlib/xcb include in gl_marshal.py

On 16/03/17 22:09, Emil Velikov wrote: From: Emil Velikov The only use of it was to provide the _X_INLINE macro. We already require (and provide where needed) 'inline', not to mention that it's used in the file already. I see, thanks. Please push this.

[Mesa-dev] [Bug 92361] [BSW SKL] Regression: glx@glx-copy-sub-buffer failed

https://bugs.freedesktop.org/show_bug.cgi?id=92361 cprigent changed: What|Removed |Added Assignee|christophe.prig...@intel.co

[Mesa-dev] [PATCH 06/32] swr: [rasterizer archrast] Add additional API events

--- .../drivers/swr/rasterizer/archrast/events.proto | 36 ++ src/gallium/drivers/swr/rasterizer/core/api.cpp| 12 2 files changed, 48 insertions(+) diff --git a/src/gallium/drivers/swr/rasterizer/archrast/events.proto

[Mesa-dev] [PATCH 25/32] swr: [rasterizer scripts] Put codegen scripts into a separate directory

--- src/gallium/drivers/swr/.gitignore | 4 +- src/gallium/drivers/swr/Makefile.am| 126 ++--- src/gallium/drivers/swr/SConscript | 38 +++ .../{scripts => codegen}/gen_archrast.py | 0 .../{scripts =>

[Mesa-dev] [PATCH 23/32] swr: [rasterizer core] SIMD16 Frontend WIP - fix tesselation crashes

--- .../drivers/swr/rasterizer/core/frontend.cpp | 19 +- src/gallium/drivers/swr/rasterizer/core/pa.h | 7 ++-- src/gallium/drivers/swr/rasterizer/core/pa_avx.cpp | 40 -- 3 files changed, 35 insertions(+), 31 deletions(-) diff --git

[Mesa-dev] [PATCH 20/32] swr: [rasterizer core] Fix typo in SIMD16 code path

--- src/gallium/drivers/swr/rasterizer/core/pa.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/src/gallium/drivers/swr/rasterizer/core/pa.h b/src/gallium/drivers/swr/rasterizer/core/pa.h index 2e159da..afeed65 100644 --- a/src/gallium/drivers/swr/rasterizer/core/pa.h +++

[Mesa-dev] [PATCH 01/32] swr: [rasterizer core] Support sparse numa id values on all OSes

--- .../drivers/swr/rasterizer/core/threads.cpp| 80 ++ 1 file changed, 53 insertions(+), 27 deletions(-) diff --git a/src/gallium/drivers/swr/rasterizer/core/threads.cpp b/src/gallium/drivers/swr/rasterizer/core/threads.cpp index f1c3030..e11291b 100644 ---

[Mesa-dev] [PATCH 18/32] swr: [rasterizer core] Allow no arguments to SWR_INVALID macro

Turns out this is somewhat tricky with gcc/g++. --- src/gallium/drivers/swr/rasterizer/common/swr_assert.h | 14 +- 1 file changed, 13 insertions(+), 1 deletion(-) diff --git a/src/gallium/drivers/swr/rasterizer/common/swr_assert.h

[Mesa-dev] [PATCH 26/32] swr: [rasterizer codegen] Quiet gen_backends.py execution

--- src/gallium/drivers/swr/rasterizer/codegen/gen_backends.py | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/src/gallium/drivers/swr/rasterizer/codegen/gen_backends.py b/src/gallium/drivers/swr/rasterizer/codegen/gen_backends.py index cbbc378..d6367ca 100644 ---

Re: [Mesa-dev] [Mesa-stable] [PATCH] i965/gen8+: Do full stall when switching pipeline

Jason Ekstrand writes: > Thanks for tracking this down! > > Reviewed-by: Jason Ekstrand > > I'm not sure how we missed this when CS was brought up. Oh well. > I don't think we did. The hardware docs are somewhat contradictory about whether this is

[Mesa-dev] [Bug 100236] Undefined symbols for architecture x86_64: "typeinfo for llvm::RTDyldMemoryManager"

https://bugs.freedesktop.org/show_bug.cgi?id=100236 --- Comment #1 from Emil Velikov --- Interesting - this might be related to my pthread-stubs patches. Vinson, please provide a complete configure line alongside the summary and more information about the

Re: [Mesa-dev] [PATCH 00/28] update swr rasterizer

On Mar 15, 2017, at 9:59 PM, Emil Velikov > wrote: Hi Tim, On 16 March 2017 at 00:12, Tim Rowley > wrote: Hightlights include: lots of simd16 work, assert rework, and

[Mesa-dev] [PATCH 13/32] swr: [rasterizer archrast/core/scripts] Fix archrast multithreading issue

Per pixel stats are cached but were not always being flushed as threads moved from one draw context to the next. Added an explicit flush to allow all archrast objects to flush any cached events. --- .../drivers/swr/rasterizer/archrast/archrast.cpp | 52 +++---

[Mesa-dev] [PATCH 27/32] swr: [rasterizer codegen] Rewrite gen_llvm_ir_macros.py to use mako

Don't create/use cpp files, header only now. --- src/gallium/drivers/swr/.gitignore | 6 +- src/gallium/drivers/swr/Makefile.am| 61 +-- src/gallium/drivers/swr/SConscript | 24 +- .../swr/rasterizer/codegen/gen_llvm_ir_macros.py | 411

[Mesa-dev] [PATCH 10/32] swr: [rasterizer core] Fix RECT_LIST primitive assembly

The bug would make the 3rd component of attributes on the second triangle of a RECT be invalid. --- src/gallium/drivers/swr/rasterizer/core/pa_avx.cpp | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/src/gallium/drivers/swr/rasterizer/core/pa_avx.cpp

[Mesa-dev] [PATCH 21/32] swr: [rasterizer] Convert more SWR_ASSERT(false, ...) to SWR_INVALID(...)

--- .../drivers/swr/rasterizer/common/simdintrin.h | 2 +- src/gallium/drivers/swr/rasterizer/core/api.cpp | 2 +- src/gallium/drivers/swr/rasterizer/core/backend.cpp | 2 +- src/gallium/drivers/swr/rasterizer/core/binner.cpp | 2 +-

[Mesa-dev] [PATCH 02/32] swr: [rasterizer core] Finish SIMD16 PA OPT except tesselation

--- src/gallium/drivers/swr/rasterizer/core/pa.h | 84 - src/gallium/drivers/swr/rasterizer/core/pa_avx.cpp | 1625 +--- 2 files changed, 1420 insertions(+), 289 deletions(-) diff --git a/src/gallium/drivers/swr/rasterizer/core/pa.h

[Mesa-dev] [PATCH 03/32] swr: [rasterizer core] Finish SIMD16 PA OPT including tesselation

--- src/gallium/drivers/swr/rasterizer/core/pa_avx.cpp | 268 +++-- 1 file changed, 247 insertions(+), 21 deletions(-) diff --git a/src/gallium/drivers/swr/rasterizer/core/pa_avx.cpp b/src/gallium/drivers/swr/rasterizer/core/pa_avx.cpp index 22643bd..6fb37e5 100644 ---

[Mesa-dev] [PATCH 00/32] update swr rasterizer

Hightlights include: lots of simd16 work, assert rework, and autogen changes (scripts centralized, one file added, two removed). v2: * adjust scons build system along with automake * separate backend template comment style change * only one declaration of gBackendPixelRateTable * prefix template

[Mesa-dev] [PATCH 24/32] swr: [rasterizer core] Fix trifan regression from 886a196cf2

Fixes piglit triangle-rasterization-overdraw. SIMD16 path not working. --- src/gallium/drivers/swr/rasterizer/core/pa.h | 2 ++ src/gallium/drivers/swr/rasterizer/core/pa_avx.cpp | 14 +- 2 files changed, 11 insertions(+), 5 deletions(-) diff --git

[Mesa-dev] [PATCH 05/32] swr: [rasterizer core/scripts] Autogen backend initialization function(s)

Autogen functions that instantiates different BackendPixelRate templates. Functions get split into separate files after reaching a user defined threshold (currently 512 per file) to speed up compilation. This change will enable the addition of more template flags in the pixel back end. ---

[Mesa-dev] [PATCH 14/32] swr: [rasterizer core] Implement double pumped SIMD16 TESS

--- .../drivers/swr/rasterizer/core/frontend.cpp | 256 ++--- 1 file changed, 177 insertions(+), 79 deletions(-) diff --git a/src/gallium/drivers/swr/rasterizer/core/frontend.cpp b/src/gallium/drivers/swr/rasterizer/core/frontend.cpp index 761e044..b2f4643 100644 ---

[Mesa-dev] [PATCH 08/32] swr: [rasterizer archrast] Fix performance issue with archrast stats

Performance is now 50x faster with archrast now that we're properly filtering out all of the rdtsc begin/end. --- .../drivers/swr/rasterizer/archrast/archrast.cpp | 30 +++--- 1 file changed, 15 insertions(+), 15 deletions(-) diff --git

[Mesa-dev] [PATCH 07/32] swr: [rasterizer core] Implement SIMD16 GS and STREAMOUT

--- .../drivers/swr/rasterizer/core/frontend.cpp | 302 + 1 file changed, 251 insertions(+), 51 deletions(-) diff --git a/src/gallium/drivers/swr/rasterizer/core/frontend.cpp b/src/gallium/drivers/swr/rasterizer/core/frontend.cpp index 18728e5..761e044 100644 ---

[Mesa-dev] [PATCH 16/32] swr: [rasterizer] Backend code adjustments

--- .../drivers/swr/rasterizer/common/simdintrin.h | 9 ++ src/gallium/drivers/swr/rasterizer/core/api.cpp| 5 +- .../drivers/swr/rasterizer/core/backend.cpp| 1 + src/gallium/drivers/swr/rasterizer/core/backend.h | 96 --

[Mesa-dev] [PATCH 11/32] swr: [rasterizer archrast/scripts] Further archrast cleanups

Removed redundant data being written out to file --- .../drivers/swr/rasterizer/archrast/archrast.cpp | 264 - .../scripts/templates/ar_eventhandler_h.template | 2 +- .../templates/ar_eventhandlerfile_h.template | 2 +- 3 files changed, 104 insertions(+), 164

[Mesa-dev] [PATCH 04/32] swr: [rasterizer core] backend.h declares gBackendPixelRateTable

--- src/gallium/drivers/swr/rasterizer/core/api.cpp | 1 - src/gallium/drivers/swr/rasterizer/core/backend.h | 8 2 files changed, 8 insertions(+), 1 deletion(-) diff --git a/src/gallium/drivers/swr/rasterizer/core/api.cpp b/src/gallium/drivers/swr/rasterizer/core/api.cpp index

[Mesa-dev] [PATCH 12/32] swr: [rasterizer archrast] Remove redundant data from archrast files

If count can be derived from other counts then this can be done in post processing scripts. --- .../drivers/swr/rasterizer/archrast/archrast.cpp | 42 ++--- .../drivers/swr/rasterizer/archrast/events.proto | 198 ++--- 2 files changed, 103 insertions(+), 137 deletions(-)

[Mesa-dev] [PATCH 15/32] swr: [rasterizer archrast] Fix the early and late depthstencil events

The coverage and stencil mask arguments were reversed. --- src/gallium/drivers/swr/rasterizer/core/backend.cpp | 10 +- 1 file changed, 5 insertions(+), 5 deletions(-) diff --git a/src/gallium/drivers/swr/rasterizer/core/backend.cpp b/src/gallium/drivers/swr/rasterizer/core/backend.cpp

[Mesa-dev] [Bug 100236] Undefined symbols for architecture x86_64: "typeinfo for llvm::RTDyldMemoryManager"

https://bugs.freedesktop.org/show_bug.cgi?id=100236 Vinson Lee changed: What|Removed |Added CC|

[Mesa-dev] [PATCH 32/32] swr: [rasterizer] Cleanup naming of codegen files

All template files and generated files are prefixed with gen_. --- src/gallium/drivers/swr/.gitignore | 12 ++--- src/gallium/drivers/swr/Makefile.am| 62 +++--- src/gallium/drivers/swr/SConscript | 18 +++

[Mesa-dev] [PATCH 17/32] swr: [rasterizer] Slight assert refactoring

Make asserts more robust. Add SWR_INVALID(...) as a replacement for SWR_ASSERT(0, ...) --- .../drivers/swr/rasterizer/archrast/archrast.cpp | 2 +- .../drivers/swr/rasterizer/archrast/eventmanager.h | 2 +- .../drivers/swr/rasterizer/common/swr_assert.cpp | 331 ++---

[Mesa-dev] [PATCH 30/32] swr: [rasterizer codegen] Rewrite gen_llvm_types.py to use mako

--- src/gallium/drivers/swr/Makefile.am| 9 +- src/gallium/drivers/swr/SConscript | 2 +- .../swr/rasterizer/codegen/gen_llvm_types.py | 206 ++--- .../swr/rasterizer/codegen/templates/gen_llvm.hpp | 72 +++ 4 files changed, 171

[Mesa-dev] [PATCH 19/32] swr: [rasterizer core/common] Fix the native AVX512 build under ICC

--- .../drivers/swr/rasterizer/common/simd16intrin.h | 26 +--- src/gallium/drivers/swr/rasterizer/core/pa.h | 46 -- src/gallium/drivers/swr/rasterizer/core/pa_avx.cpp | 2 +- 3 files changed, 47 insertions(+), 27 deletions(-) diff --git

[Mesa-dev] [PATCH 31/32] swr: [rasterizer codegen] Remove BOM from knob_defs.py

--- src/gallium/drivers/swr/rasterizer/codegen/knob_defs.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/src/gallium/drivers/swr/rasterizer/codegen/knob_defs.py b/src/gallium/drivers/swr/rasterizer/codegen/knob_defs.py index 3c6940d..0c39a77 100644 ---

[Mesa-dev] [PATCH 29/32] swr: [rasterizer codegen] Fix generation of knobs

--- src/gallium/drivers/swr/Makefile.am| 10 +- src/gallium/drivers/swr/SConscript | 4 ++-- src/gallium/drivers/swr/rasterizer/codegen/gen_backends.py | 2 ++ src/gallium/drivers/swr/rasterizer/codegen/gen_knobs.py| 2

[Mesa-dev] [PATCH 09/32] swr: [rasterizer common] Add InterpolateComponentFlat utility

--- src/gallium/drivers/swr/rasterizer/common/simdintrin.h | 13 + 1 file changed, 13 insertions(+) diff --git a/src/gallium/drivers/swr/rasterizer/common/simdintrin.h b/src/gallium/drivers/swr/rasterizer/common/simdintrin.h index 8926e66..ea79902 100644 ---

[Mesa-dev] [PATCH 22/32] swr: [rasterizer jitter] Fix LogicOp blend jit after assert changes

--- .../drivers/swr/rasterizer/jitter/blend_jit.cpp| 35 +++--- 1 file changed, 25 insertions(+), 10 deletions(-) diff --git a/src/gallium/drivers/swr/rasterizer/jitter/blend_jit.cpp b/src/gallium/drivers/swr/rasterizer/jitter/blend_jit.cpp index cbcc343..55961d1 100644 ---

[Mesa-dev] [PATCH 28/32] swr: [rasterizer codegen] Change backend template comment style

--- .../codegen/templates/backend_template.cpp | 58 +++--- 1 file changed, 29 insertions(+), 29 deletions(-) diff --git a/src/gallium/drivers/swr/rasterizer/codegen/templates/backend_template.cpp

Re: [Mesa-dev] [PATCH 2/2] radv: add external memory support.

On Wed, Mar 15, 2017 at 1:25 AM, Dave Airlie wrote: > From: Dave Airlie > > This adds support for exporting 2D images, to an > opaque fd. > > This implements the: > VK_KHX_external_memory_capabilities > VK_KHX_external_memory > VK_KHX_external_memory_fd > >

[Mesa-dev] [PATCH 1/4] i965: Fall back to GL 4.2 on Haswell if the kernel isn't new enough.

In commit d2590eb65ff28a9cbd592353d15d7e6cbd2c6fc6 I enabled GL 4.5 on Haswell...but failed to check if we could do indirect compute shader dispatch...and query buffer objects. Indirect compute shader dispatch requires command parser version 5 (kernel commit

Re: [Mesa-dev] [PATCH 1/4] i965: Fall back to GL 4.2 on Haswell if the kernel isn't new enough.

On Thu, Mar 16, 2017 at 6:12 PM, Kenneth Graunke wrote: > In commit d2590eb65ff28a9cbd592353d15d7e6cbd2c6fc6 I enabled GL 4.5 > on Haswell...but failed to check if we could do indirect compute > shader dispatch...and query buffer objects. > > Indirect compute shader

Re: [Mesa-dev] [PATCH v3 1/1] clover: use pipe_resource references

Jan Vesely writes: > v2: buffers are created with one reference. > v3: add pipe_resource reference to mapping object > Mapping objects are supposed to be short-lived, they're logically part of the parent resource object so they shouldn't ever out-live it. What is this

[Mesa-dev] [PATCH] configure.ac: Use POSIX word boundary regex.

Fixes: fe56c745b8cb ("Convert sed(1) syntax to be compatible with FreeBSD and OpenBSD") Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=100236 Signed-off-by: Vinson Lee --- configure.ac | 10 +- 1 file changed, 5 insertions(+), 5 deletions(-) diff --git

[Mesa-dev] [PATCH 2/4] i965: Document the sad story of the kernel command parser.

This should help us figure out the complexities of which kernel versions we need to get various features on various platforms. Signed-off-by: Kenneth Graunke --- src/mesa/drivers/dri/i965/intel_screen.c | 97 1 file changed, 97

[Mesa-dev] [PATCH 4/4] i965: Skip register write detection when possible.

Detecting register write support by trial and error introduces a stall at screen creation time, which it would be nice to avoid. Certain command parser versions guarantee this will work (see the giant comment in intelInitScreen2 below, or a few commits ago): - Ivybridge: version >= 1 (kernel

[Mesa-dev] [PATCH 3/4] i965: Set screen->cmd_parser_version to 0 if we can't write registers.

If we can't write registers, then the effective command parser version is 0 - it may exist, but it's not usefully enabling anything. See kernel commit 1ca3712ca3429a617ed6c5f87718e4f6fe4ae0c6 (in v4.8) where the kernel starts doing this for us. This makes us do more or less the same thing on

Re: [Mesa-dev] [PATCH] i965: Select pipeline and emit state base address in Gen8+ HiZ ops.

On Wed, Mar 08, 2017 at 10:27:20AM -0800, Nanley Chery wrote: > On Wed, Mar 08, 2017 at 10:07:12AM -0800, Nanley Chery wrote: > > On Wed, Mar 08, 2017 at 02:17:59AM -0800, Kenneth Graunke wrote: > > > On Thursday, March 2, 2017 4:36:08 PM PST Nanley Chery wrote: > > > > On Mon, Feb 06, 2017 at

Re: [Mesa-dev] [RFC libdrm 0/2] Replace the build system with meson

Is there a way not to use ninja with meson, because ninja redirects all stderr output from gcc to stdout, which breaks many development environments that expect errors in stderr? I'm basically saying that if ninja can't keep gcc errors in stderr, I wouldn't like any project that I might be

Re: [Mesa-dev] [PATCH v3 1/1] clover: use pipe_resource references

On Thu, 2017-03-16 at 15:24 -0700, Francisco Jerez wrote: > Jan Vesely writes: > > > v2: buffers are created with one reference. > > v3: add pipe_resource reference to mapping object > > > > Mapping objects are supposed to be short-lived, they're logically part > of the

  1   2   3   >