Kenneth Graunke [Fri, 21 Apr 2017 08:28:13 +0000 (01:28 -0700)]
i965/vec4: Avoid reswizzling MACH instructions in opt_register_coalesce().
opt_register_coalesce() was optimizing sequences such as:
mul(8) acc0:D, attr18.xyyy:D, attr19.xyyy:D
mach(8) vgrf5.xy:D, attr18.xyyy:D, attr19.xyyy:D
mov(8) m4.zw:F, vgrf5.xxxy:F
into:
mul(8) acc0:D, attr18.xyyy:D, attr19.xyyy:D
mach(8) m4.zw:D, attr18.xxxy:D, attr19.xxxy:D
This doesn't work - if we're going to reswizzle MACH, we'd need to
reswizzle the MUL as well. Here, the MUL fills the accumulator's .zw
components with attr18.yy * attr19.yy. But the MACH instruction expects
.z to contain attr18.x * attr19.x. Bogus results ensue.
No change in shader-db on Haswell. Prevents regressions in Timothy's
patches to use enhanced layouts for varying packing (which rearrange
code just enough to trigger this pre-existing bug, but were fine
themselves).
Acked-by: Timothy Arceri <tarceri@itsqueeze.com>
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Timothy Arceri [Fri, 21 Apr 2017 07:04:10 +0000 (17:04 +1000)]
mesa: validate sampler type across the whole program
Currently we were only making sure types were the same within a
single stage. This looks to have regressed with
953a0af8e3f73.
Fixes:
953a0af8e3f73 ("mesa: validate sampler uniforms during gluniform calls")
Reviewed-by: Nicolai Hähnle <nicolai.haehnle@amd.com>
Reviewed-by: Tapani Pälli <tapani.palli@intel.com>
https://bugs.freedesktop.org/show_bug.cgi?id=97524
Timothy Arceri [Fri, 7 Apr 2017 01:40:40 +0000 (11:40 +1000)]
mesa: don't lock hashtables that are not shared across contexts
From Chapter 5 'Shared Objects and Multiple Contexts' of
the OpenGL 4.5 spec:
"Objects which contain references to other objects include
framebuffer, program pipeline, query, transform feedback,
and vertex array objects. Such objects are called container
objects and are not shared"
For we leave locking in place for framebuffer objects because
the EXT fbo extension allowed sharing.
We could maybe just replace the hash with an ordinary hash table
but for now this should remove most of the unnecessary locking.
Reviewed-by: Nicolai Hähnle <nicolai.haehnle@amd.com>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Matt Turner [Wed, 5 Apr 2017 04:29:00 +0000 (14:29 +1000)]
mesa: Remove deleteFlag pattern from container objects.
This pattern was only useful when we used mutex locks, which the previous
commit removed.
Reviewed-by: Nicolai Hähnle <nicolai.haehnle@amd.com>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Matt Turner [Fri, 21 Apr 2017 03:48:38 +0000 (13:48 +1000)]
mesa: Remove unnecessary locking from container objects.
From Chapter 5 'Shared Objects and Multiple Contexts' of
the OpenGL 4.5 spec:
"Objects which contain references to other objects include
framebuffer, program pipeline, query, transform feedback,
and vertex array objects. Such objects are called container
objects and are not shared"
For we leave locking in place for framebuffer objects because
the EXT fbo extension allowed sharing.
V2: (Timothy Arceri)
- rebased and dropped changes to framebuffer objects
Reviewed-by: Nicolai Hähnle <nicolai.haehnle@amd.com>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Timothy Arceri [Fri, 21 Apr 2017 03:29:46 +0000 (13:29 +1000)]
mesa: remove fallback RefCount == 0 pattern
We should never get here if this is 0 unless there is a
bug. Replace the check with an assert.
Reviewed-by: Nicolai Hähnle <nicolai.haehnle@amd.com>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Elie TOURNIER [Fri, 21 Apr 2017 16:20:05 +0000 (17:20 +0100)]
egl: add gitignore
Since commit
ce562f9e3fa, two new files are generated.
We don't want to track them.
Signed-off-by: Elie Tournier <elie.tournier@collabora.com>
Reviewed-by: Plamena Manolova <plamena.manolova@intel.com>
Reviewed-by: Emil Velikov <emil.velikov@collabora.com>
Samuel Pitoiset [Fri, 21 Apr 2017 09:18:50 +0000 (11:18 +0200)]
glsl: make use of glsl_type::is_float()
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Samuel Iglesias Gonsálvez <siglesias@igalia.com>
Reviewed-by: Edward O'Callaghan <funfunctor@folklore1984.net>
Samuel Pitoiset [Fri, 21 Apr 2017 08:53:32 +0000 (10:53 +0200)]
glsl: make use of glsl_type::is_double()
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Samuel Iglesias Gonsálvez <siglesias@igalia.com>
Reviewed-by: Edward O'Callaghan <funfunctor@folklore1984.net>
Samuel Pitoiset [Fri, 21 Apr 2017 08:46:24 +0000 (10:46 +0200)]
glsl: make use of glsl_type::is_integer_64()
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Samuel Iglesias Gonsálvez <siglesias@igalia.com>
Reviewed-by: Edward O'Callaghan <funfunctor@folklore1984.net>
Samuel Pitoiset [Fri, 21 Apr 2017 08:42:12 +0000 (10:42 +0200)]
glsl: simplify glsl_type::is_integer_32_64()
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Samuel Iglesias Gonsálvez <siglesias@igalia.com>
Reviewed-by: Edward O'Callaghan <funfunctor@folklore1984.net>
Samuel Pitoiset [Fri, 21 Apr 2017 08:41:13 +0000 (10:41 +0200)]
glsl: add glsl_type::is_integer_64()
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Samuel Iglesias Gonsálvez <siglesias@igalia.com>
Reviewed-by: Edward O'Callaghan <funfunctor@folklore1984.net>
Samuel Pitoiset [Fri, 21 Apr 2017 08:36:05 +0000 (10:36 +0200)]
glsl: make use of glsl_type::is_boolean()
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Samuel Iglesias Gonsálvez <siglesias@igalia.com>
Reviewed-by: Edward O'Callaghan <funfunctor@folklore1984.net>
Samuel Pitoiset [Fri, 21 Apr 2017 08:32:39 +0000 (10:32 +0200)]
glsl: make use of glsl_type::is_record()
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Samuel Iglesias Gonsálvez <siglesias@igalia.com>
Reviewed-by: Edward O'Callaghan <funfunctor@folklore1984.net>
Samuel Pitoiset [Fri, 21 Apr 2017 08:28:58 +0000 (10:28 +0200)]
glsl: make use of glsl_type::is_interface()
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Samuel Iglesias Gonsálvez <siglesias@igalia.com>
Reviewed-by: Edward O'Callaghan <funfunctor@folklore1984.net>
Samuel Pitoiset [Fri, 21 Apr 2017 08:25:42 +0000 (10:25 +0200)]
glsl: make use of glsl_type::is_array()
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Samuel Iglesias Gonsálvez <siglesias@igalia.com>
Reviewed-by: Edward O'Callaghan <funfunctor@folklore1984.net>
Samuel Pitoiset [Fri, 21 Apr 2017 08:16:58 +0000 (10:16 +0200)]
glsl: make use glsl_type::is_atomic_uint()
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Samuel Iglesias Gonsálvez <siglesias@igalia.com>
Reviewed-by: Edward O'Callaghan <funfunctor@folklore1984.net>
Samuel Pitoiset [Fri, 21 Apr 2017 08:13:49 +0000 (10:13 +0200)]
glsl: add glsl_type::is_atomic_uint() helper
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Samuel Iglesias Gonsálvez <siglesias@igalia.com>
Reviewed-by: Edward O'Callaghan <funfunctor@folklore1984.net>
Emil Velikov [Thu, 20 Apr 2017 15:24:08 +0000 (16:24 +0100)]
mesa/glthread: correctly compare thread handles
As mentioned in the manual - comparing pthread_t handles via the C
comparison operator is incorrect and pthread_equal() should be used
instead.
Cc: Timothy Arceri <tarceri@itsqueeze.com>
Fixes:
d8d81fbc316 ("mesa: Add infrastructure for a worker thread to process GL commands.")
Signed-off-by: Emil Velikov <emil.velikov@collabora.com>
Reviewed-by: Plamena Manolova <plamena.manolova@intel.com>
Reviewed-by: Brian Paul <brianp@vmware.com>
Reviewed-by: Eric Anholt <eric@anholt.net>
Emil Velikov [Wed, 19 Apr 2017 10:35:10 +0000 (11:35 +0100)]
st/clover: add space between < and ::
As pointed out by compiler
./llvm/codegen.hpp:52:22: error: ‘<::’ cannot begin a template-argument list [-fpermissive]
./llvm/codegen.hpp:52:22: note: ‘<:’ is an alternate spelling for ‘[’. Insert whitespace between ‘<’ and ‘::’
Cc: Francisco Jerez <currojerez@riseup.net>
Signed-off-by: Emil Velikov <emil.velikov@collabora.com>
Reviewed-by: Nicolai Hähnle <nicolai.haehnle@amd.com>
Reviewed-by: Vedran Miletić <vedran@miletic.net>
Samuel Pitoiset [Thu, 20 Apr 2017 17:02:28 +0000 (19:02 +0200)]
glsl: get rid of values_for_type()
This function is actually a wrapper for component_slots()
and it always returns 1 (or N) for samplers. Since
component_slots() now return 1 for samplers, it can go.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Timothy Arceri <tarceri@itsqueeze.com>
Reviewed-by: Edward O'Callaghan <funfunctor@folklore1984.net>
Reviewed-by: Nicolai Hähnle <nicolai.haehnle@amd.com>
Samuel Pitoiset [Thu, 20 Apr 2017 17:02:27 +0000 (19:02 +0200)]
glsl: make component_slots() returns 1 for sampler types
It looks inconsistent to return 1 for image types and 0 for
sampler types. Especially because component_slots() is mostly
used by values_for_type() which always returns 1 for samplers.
For bindless, this value will be bumped to 2 because the
ARB_bindless_texture states that bindless samplers/images
should consume two components.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Timothy Arceri <tarceri@itsqueeze.com>
Reviewed-by: Edward O'Callaghan <funfunctor@folklore1984.net>
Reviewed-by: Nicolai Hähnle <nicolai.haehnle@amd.com>
Kai Wasserbäch [Thu, 20 Apr 2017 10:21:02 +0000 (12:21 +0200)]
docs/features: mark KHR_no_error as started
The OpenGL extension KHR_no_error is exposed since commit
d42d150ad26e29d9e894ba9f9e28f8134e2e5393 by Timothy Arceri. Therefore it
should be marked as "started" in the features.txt
Signed-off-by: Kai Wasserbäch <kai@dev.carbon-project.org>
Reviewed-by: Nicolai Hähnle <nicolai.haehnle@amd.com>
Tapani Pälli [Fri, 21 Apr 2017 07:00:21 +0000 (10:00 +0300)]
Revert "android: fix segfault within swap_buffers"
This reverts commit
4d4558411db166d2d66f8cec9cb581149dbe1597.
This was a wrong call, while it fixed issue with 3DMark it
actually introduced regression elsewhere.
Signed-off-by: Tapani Pälli <tapani.palli@intel.com>
Ilia Mirkin [Wed, 19 Apr 2017 22:38:02 +0000 (18:38 -0400)]
nvc0: Add support for setting viewport index/layer from VS/TES
This enables support on GM200+ for:
- GL_AMD_vertex_shader_layer
- GL_AMD_vertex_shader_layer_viewport_index
- GL_ARB_shader_viewport_layer_array
Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu>
[lyude: add relnotes/TES cap]
Signed-off-by: Lyude <lyude@redhat.com>
[imirkin: move relnotes to right place, add features.txt]
Reviewed-by: Ilia Mirkin <imirkin@alum.mit.edu>
Lyude [Wed, 19 Apr 2017 22:38:01 +0000 (18:38 -0400)]
nvc0/ir: Only store viewport in scratch register for GP
EMIT only applies to geometry shaders. For everything else, we want to
export the viewport normally.
Signed-off-by: Lyude <lyude@redhat.com>
Reviewed-by: Boyan Ding <boyan.j.ding@gmail.com>
Reviewed-by: Ilia Mirkin <imirkin@alum.mit.edu>
Bas Nieuwenhuizen [Thu, 20 Apr 2017 20:03:15 +0000 (22:03 +0200)]
radv: Prefetch compute shader too.
For consistency, doesn't really impact performance.
Signed-off-by: Bas Nieuwenhuizen <basni@google.com>
Reviewed-by: Dave Airlie <airlied@redhat.com>
Jason Ekstrand [Thu, 20 Apr 2017 00:16:49 +0000 (17:16 -0700)]
anv/query: Use genxml for MI_MATH
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed by: Iago Toral Quiroga <itoral@igalia.com>
Jason Ekstrand [Wed, 19 Apr 2017 23:21:08 +0000 (16:21 -0700)]
genxml: Add better support for MI_MATH
This breaks the guts of MI_MATH (the instruction part) out into its own
structure with proper named values.
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed by: Iago Toral Quiroga <itoral@igalia.com>
Jason Ekstrand [Wed, 19 Apr 2017 23:20:44 +0000 (16:20 -0700)]
genxml/pack: Allow hex values in the XML
Acked-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed by: Iago Toral Quiroga <itoral@igalia.com>
Reviewed-by: Dylan Baker <dylan@pnwbakers.com>
Dave Airlie [Thu, 20 Apr 2017 19:53:37 +0000 (20:53 +0100)]
radv/ac: use tex_lz if we can.
Looking at some Talos shaders vs radeonsi, I noticed they use
tex_lz in a few places, so we should be able to as well.
Reviewed-by: Bas Nieuwenhuizen <basni@google.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
Marek Olšák [Sun, 9 Apr 2017 20:35:36 +0000 (22:35 +0200)]
st/mesa: use one big translation table in st_pipe_vertex_format
for lower overhead.
Reviewed-by: Ilia Mirkin <imirkin@alum.mit.edu>
Reviewed-by: Brian Paul <brianp@vmware.com>
Marek Olšák [Sun, 9 Apr 2017 14:48:40 +0000 (16:48 +0200)]
st/mesa: check in advance in st_draw_vbo whether the bitmap cache is empty
Reviewed-by: Ilia Mirkin <imirkin@alum.mit.edu>
Reviewed-by: Brian Paul <brianp@vmware.com>
Marek Olšák [Sun, 9 Apr 2017 14:45:00 +0000 (16:45 +0200)]
st/mesa: put the bitmap_cache structure inside st_context
This is nicer on caches, and the next commit will need to access
the structure from a different place.
Reviewed-by: Ilia Mirkin <imirkin@alum.mit.edu>
Reviewed-by: Brian Paul <brianp@vmware.com>
Marek Olšák [Sun, 9 Apr 2017 14:39:39 +0000 (16:39 +0200)]
st/mesa: inline and optimize st_invalidate_readpix_cache
Reviewed-by: Ilia Mirkin <imirkin@alum.mit.edu>
Reviewed-by: Brian Paul <brianp@vmware.com>
Marek Olšák [Sun, 9 Apr 2017 14:03:59 +0000 (16:03 +0200)]
st/mesa: invalidate the readpix cache in st_indirect_draw_vbo
Cc: <mesa-stable@lists.freedesktop.org>
Reviewed-by: Ilia Mirkin <imirkin@alum.mit.edu>
Reviewed-by: Brian Paul <brianp@vmware.com>
Marek Olšák [Sun, 2 Apr 2017 14:33:07 +0000 (16:33 +0200)]
gallium/util: remove util_draw_range_elements helper
min/max_index are typically hints for the u_vbuf module, not the driver.
Reviewed-by: Ilia Mirkin <imirkin@alum.mit.edu>
Reviewed-by: Brian Paul <brianp@vmware.com>
Marek Olšák [Sun, 9 Apr 2017 13:57:45 +0000 (15:57 +0200)]
gallium: fold u_trim_pipe_prim call from st/mesa to drivers
Most drivers don't need it and shouldn't need it because it can't be used
in some cases (indirect draws, primitive restart, count from streamout).
Reviewed-by: Brian Paul <brianp@vmware.com>
Samuel Iglesias Gonsálvez [Thu, 20 Apr 2017 12:02:44 +0000 (14:02 +0200)]
docs/envvars: sort INTEL_DEBUG envvar options by name
It helps to find the envvar option you are looking for.
Signed-off-by: Samuel Iglesias Gonsálvez <siglesias@igalia.com>
Reviewed-by: Iago Toral Quiroga <itoral@igalia.com>
Christoph Haag [Thu, 20 Apr 2017 08:34:18 +0000 (10:34 +0200)]
ac: fix build after LLVM 5.0 SVN r300718
v2: previously getWithDereferenceableBytes() exists, but addAttr() doesn't take that type
Signed-off-by: Christoph Haag <haagch+mesadev@frickel.club>
Reviewed-by: Nicolai Hähnle <nicolai.haehnle@amd.com>
Tested-and-reviewed-by: Mike Lothian <mike@fireburn.co.uk>
Juan A. Suarez Romero [Wed, 5 Apr 2017 18:18:42 +0000 (20:18 +0200)]
bin/get-{extra,fixes}-pick-list.sh: improve output
Show the commit hash and the title in a way that it is easier to copy
and paste in the bin/.cherry-ignore-extra file if we want to ignore
those commits for the future.
v2:
- Use printf instead echo (Eric Engestrom)
Reviewed-by: Eric Engestrom <eric.engestrom@imgtec.com>
Reviewed-by: Emil Velikov <emil.velikov@collabora.com>
Juan A. Suarez Romero [Wed, 5 Apr 2017 18:14:22 +0000 (20:14 +0200)]
bin/get-{extra,fixes}-pick-list.sh: add support for ignore list
Both scripts does not use a file with the commits to ignore. So if we
have handled one of the suggested commits and decided we won't pick it,
the scripts will continue suggesting them.
v2:
- Mark the candidates in bin/get-extra-pick-list.sh (Juan A. Suarez)
- Use bin/.cherry-ignore to store rejected patches (Emil)
Reviewed-by: Emil Velikov <emil.velikov@collabora.com>
Brian Paul [Fri, 14 Apr 2017 18:42:40 +0000 (12:42 -0600)]
mesa: print target string in glBindTexture() error message
Reviewed-by: Edward O'Callaghan <funfunctor@folklore1984.net>
Reviewed-by: Timothy Arceri <tarceri@itsqueeze.com>
Brian Paul [Wed, 19 Apr 2017 18:13:36 +0000 (12:13 -0600)]
mesa: fix Windows build error related to getuid()
getuid() and geteuid() are not present on Windows.
Reviewed-by: Timothy Arceri <tarceri@itsqueeze.com>
Tim Rowley [Tue, 18 Apr 2017 22:02:56 +0000 (17:02 -0500)]
swr: simd16 vs work
Build VS with alternating output for the current simd16 fe double-pump
of a simd8 shader.
Reviewed-by: Bruce Cherniak <bruce.cherniak@intel.com>
Bas Nieuwenhuizen [Wed, 19 Apr 2017 22:59:31 +0000 (00:59 +0200)]
radv: Set variant code_size when created from the cache.
Signed-off-by: Bas Nieeuwenhuizen <basni@google.com>
Reviewed-by: Dave Airlie <airlied@redhat.com>
Bas Nieuwenhuizen [Wed, 19 Apr 2017 20:32:16 +0000 (22:32 +0200)]
radv: Add shader prefetch.
Gives me approximately a 2% perf increase in bot dota2 & talos.
Having descriptors (both sets and vertex buffers) prefetched
didn't help so I didn't include that.
Signed-off-by: Bas Nieuwenhuizen <basni@google.com>
Reviewed-by: Dave Airlie <airlied@redhat.com>
Bas Nieuwenhuizen [Mon, 17 Apr 2017 10:22:02 +0000 (12:22 +0200)]
radv: Remove binding buffer count.
In cases where it is used it is always 1.
Signed-off-by: Bas Nieuwenhuizen <basni@google.com>
Reviewed-by: Bas Nieuwenhuizen <basni@google.com>
Bas Nieuwenhuizen [Tue, 18 Apr 2017 23:08:06 +0000 (01:08 +0200)]
radv: Don't try to find gaps for non-freeable descriptors.
With this we don't have any operations on a pool with non-freeable
descriptors left that have O(#descriptors) complexity.
Signed-off-by: Bas Nieuwenhuizen <basni@google.com>
Reviewed-by: Bas Nieuwenhuizen <basni@google.com>
Bas Nieuwenhuizen [Mon, 17 Apr 2017 00:14:06 +0000 (02:14 +0200)]
radv: Use host memory pool for non-freeable descriptors.
v2: Handle out of pool memory error.
v3: Actually use VK_ERROR_OUT_OF_POOL_MEMORY_KHR for the error condition.
Signed-off-by: Bas Nieuwenhuizen <basni@google.com>
Reviewed-by: Bas Nieuwenhuizen <basni@google.com>
Bas Nieuwenhuizen [Mon, 17 Apr 2017 09:30:43 +0000 (11:30 +0200)]
radv: Don't allocate dynamic descriptors separately.
Signed-off-by: Bas Nieuwenhuizen <basni@google.com>
Reviewed-by: Bas Nieuwenhuizen <basni@google.com>
Emil Velikov [Sun, 16 Apr 2017 14:46:28 +0000 (15:46 +0100)]
st/mesa: automake: honour the vdpau header install location
If VDPAU is installed in the non-default location, we'll fail to find
the headers and error at build time.
../../src/gallium/include/state_tracker/vdpau_dmabuf.h:37:25: fatal error: vdpau/vdpau.h: No such file or directory
#include <vdpau/vdpau.h>
^
Fixes:
faba96bc60b ("st/vdpau: add new interop interface")
Cc: Christian König <christian.koenig@amd.com>
Signed-off-by: Emil Velikov <emil.velikov@collabora.com>
Reviewed-by: Nicolai Hähnle <nicolai.haehnle@amd.com>
Emil Velikov [Sun, 16 Apr 2017 13:39:03 +0000 (14:39 +0100)]
winsys/sw/dri: don't use GNU void pointer arithmetic
Resolves build issues like the following:
src/gallium/winsys/sw/dri/dri_sw_winsys.c:203:31: error: pointer of type ‘void *’ used in arithmetic [-Werror=pointer-arith]
data = dri_sw_dt->data + (dri_sw_dt->stride * box->y) + box->x * blsize;
^
src/gallium/winsys/sw/dri/dri_sw_winsys.c:203:62: error: pointer of type ‘void *’ used in arithmetic [-Werror=pointer-arith]
data = dri_sw_dt->data + (dri_sw_dt->stride * box->y) + box->x * blsize;
^
Cc: <mesa-stable@lists.freedesktop.org>
Signed-off-by: Emil Velikov <emil.velikov@collabora.com>
Reviewed-by: Nicolai Hähnle <nicolai.haehnle@amd.com>
Emil Velikov [Sun, 16 Apr 2017 00:46:59 +0000 (01:46 +0100)]
configure.ac: check require_basic_egl only if egl enabled
Fixes:
1ac40173c2a ("configure.ac: simplify EGL requirements for drivers dependent on EGL")
Signed-off-by: Emil Velikov <emil.velikov@collabora.com>
Reviewed-by: Nicolai Hähnle <nicolai.haehnle@amd.com>
Emil Velikov [Tue, 18 Apr 2017 10:33:42 +0000 (11:33 +0100)]
configure.ac: manually expand PKG_CHECK_VAR
The macro is introduced with pkgconfig v0.28 which isn't universally
available. Thus it will error at configure stage.
Reported-by: Brian Paul <brianp@vmware.com>
Tested-by: Brian Paul <brianp@vmware.com>
Fixes:
ce562f9e3fa ("EGL: Implement the libglvnd interface for EGL (v3)")
Signed-off-by: Emil Velikov <emil.velikov@collabora.com>
Timothy Arceri [Tue, 4 Apr 2017 05:45:06 +0000 (15:45 +1000)]
mesa: add KHR_no_error support to glVertexAttribDivisor()
Reviewed-by: Nicolai Hähnle <nicolai.haehnle@amd.com>
Timothy Arceri [Mon, 3 Apr 2017 05:11:29 +0000 (15:11 +1000)]
mesa/vbo: add KHR_no_error support to DrawElements*() functions
V2: move MESA_VERBOSE checks back into the common code path.
Reviewed-by: Nicolai Hähnle <nicolai.haehnle@amd.com>
Timothy Arceri [Fri, 31 Mar 2017 04:38:48 +0000 (15:38 +1100)]
mesa/vbo: add KHR_no_error support to vbo_exec_DrawArrays*()
V2: add missing FLUSH_CURRENT() to no_error path
Reviewed-by: Nicolai Hähnle <nicolai.haehnle@amd.com>
Timothy Arceri [Mon, 3 Apr 2017 06:38:18 +0000 (16:38 +1000)]
mesa/vbo: move some Draw checks out of validation
These checks do not generate any errors. Move them so we can add
KHR_no_error support and still make sure we do these checks.
Reviewed-by: Nicolai Hähnle <nicolai.haehnle@amd.com>
Timothy Arceri [Mon, 3 Apr 2017 04:37:25 +0000 (14:37 +1000)]
mesa/varray: add KHR_no_error support to *Pointer() functions
Reviewed-by: Nicolai Hähnle <nicolai.haehnle@amd.com>
Timothy Arceri [Fri, 31 Mar 2017 04:26:34 +0000 (15:26 +1100)]
mesa/varray: add KHR_no_error support to some callers of validate_array_format()
The only caller we don't update is update_arrays(), we leave that to the
following commit.
Reviewed-by: Nicolai Hähnle <nicolai.haehnle@amd.com>
Timothy Arceri [Fri, 31 Mar 2017 04:02:40 +0000 (15:02 +1100)]
mesa/varray: rename update_array_format() -> validate_array_format()
We also move _mesa_update_array_format() into the caller.
This gets these functions ready for KHR_no_error support.
V2: Updated function comment as suggested by Brian.
Reviewed-by: Nicolai Hähnle <nicolai.haehnle@amd.com>
Timothy Arceri [Fri, 31 Mar 2017 03:44:59 +0000 (14:44 +1100)]
mesa/varray: create get_array_format() helper
This will help us split array validation from array update.
V2: add const to ctx param
Reviewed-by: Nicolai Hähnle <nicolai.haehnle@amd.com>
Timothy Arceri [Fri, 31 Mar 2017 03:18:35 +0000 (14:18 +1100)]
mesa/varray: split update_array() into validate_array() and update_array()
This will be used for adding KHR_no_error support.
Reviewed-by: Nicolai Hähnle <nicolai.haehnle@amd.com>
Timothy Arceri [Thu, 30 Mar 2017 12:22:46 +0000 (23:22 +1100)]
mesa: add KHR_no_error support to glUniform*() functions
V2: restore lost comment, add static to validate_uniform(),
simplify array offset logic.
Reviewed-by: Nicolai Hähnle <nicolai.haehnle@amd.com>
Timothy Arceri [Tue, 28 Mar 2017 03:43:23 +0000 (14:43 +1100)]
mesa: always return GL_OUT_OF_MEMORY or GL_NO_ERROR when KHR_no_error enabled
Reviewed-by: Nicolai Hähnle <nicolai.haehnle@amd.com>
Timothy Arceri [Thu, 30 Mar 2017 12:11:34 +0000 (23:11 +1100)]
mesa: add _mesa_is_no_error_enabled() helper
Reviewed-by: Nicolai Hähnle <nicolai.haehnle@amd.com>
Timothy Arceri [Wed, 12 Apr 2017 05:08:01 +0000 (15:08 +1000)]
mesa: add env var to force enable the KHR_no_error ctx flag
V2: typo know -> known
V3: add security check (Suggested by Nicolai)
Reviewed-by: Nicolai Hähnle <nicolai.haehnle@amd.com>
Timothy Arceri [Tue, 28 Mar 2017 03:47:58 +0000 (14:47 +1100)]
mesa: expose KHR_no_error
Reviewed-by: Nicolai Hähnle <nicolai.haehnle@amd.com>
Constantine Kharlamov [Thu, 13 Apr 2017 20:56:28 +0000 (23:56 +0300)]
r600g: update dirty_level_mask after the 1-st draw after FB change
Ported from radeonsi. Testing with Kane&Lynch2 shows ≈1k skipped updates per
frame on average.
No piglit changes with tests/gpu.py, gbm mode.
Signed-off-by: Constantine Kharlamov <Hi-Angel@yandex.ru>
Tested-by: Dieter Nützel <Dieter@nuetzel-hh.de>
Reviewed-by: Nicolai Hähnle <nicolai.haehnle@amd.com>
Nicolai Hähnle [Fri, 7 Apr 2017 16:24:44 +0000 (18:24 +0200)]
vbo: fix gl_DrawID handling in glMultiDrawArrays
Fixes a bug in
KHR-GL45.shader_draw_parameters_tests.ShaderMultiDrawArraysParameters.
Cc: mesa-stable@lists.freedesktop.org
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Nicolai Hähnle [Fri, 7 Apr 2017 16:20:34 +0000 (18:20 +0200)]
mesa: move glMultiDrawArrays to vbo and fix error handling
When any count[i] is negative, we must skip all draws.
Moving to vbo makes the subsequent change easier.
v2:
- provide the function in all contexts, including GLES
- adjust validation accordingly to include the xfb check
v3:
- fix mix-up of pre- and post-xfb prim count (Nils Wallménius)
Cc: mesa-stable@lists.freedesktop.org
Reviewed-by: Timothy Arceri <tarceri@itsqueeze.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Nicolai Hähnle [Thu, 13 Apr 2017 19:06:11 +0000 (21:06 +0200)]
mesa: extract need_xfb_remaining_prims_check
The same logic needs to be applied to glMultiDrawArrays.
Cc: mesa-stable@lists.freedesktop.org
Reviewed-by: Timothy Arceri <tarceri@itsqueeze.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Nicolai Hähnle [Thu, 13 Apr 2017 18:53:17 +0000 (20:53 +0200)]
mesa: fix remaining xfb prims check for GLES with multiple instances
Found by inspection.
Cc: mesa-stable@lists.freedesktop.org
Reviewed-by: Timothy Arceri <tarceri@itsqueeze.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Mike Lothian [Wed, 19 Apr 2017 02:09:05 +0000 (03:09 +0100)]
radv/meta: Fix nir_builder.h include
This fixes the build after:
commit
399ebd2a84a133bd2ca3da388a059fd3bafe33f5
Author: Dave Airlie <airlied@redhat.com>
Date: Wed Apr 19 06:18:23 2017 +1000
radv/meta: add common shader vertex generation function
Signed-off-by: Mike Lothian <mike@fireburn.co.uk>
Signed-off-by: Dave Airlie <airlied@redhat.com>
Mike Lothian [Wed, 19 Apr 2017 02:09:04 +0000 (03:09 +0100)]
radv/ac: Fix nir.h include
This fixes the build after:
commit
224cf2906a8f38ce47411afc93a223ac0e41795f
Author: Dave Airlie <airlied@redhat.com>
Date: Mon Apr 17 13:01:52 2017 +1000
radv/ac: add initial pre-pass for shader info gathering
Signed-off-by: Mike Lothian <mike@fireburn.co.uk>
Signed-off-by: Dave Airlie <airlied@redhat.com>
Dave Airlie [Tue, 18 Apr 2017 23:13:06 +0000 (09:13 +1000)]
radv/meta: refactor out some common shaders.
The vs vertex generate and fs noop shaders are used in a few places,
so refactor them out.
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Signed-off-by: Dave Airlie <airlied@redhat.com>
Dave Airlie [Tue, 18 Apr 2017 20:40:29 +0000 (06:40 +1000)]
radv/meta: generate position for blit shaders.
This generates the position info using the vertex shader.
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Signed-off-by: Dave Airlie <airlied@redhat.com>
Dave Airlie [Tue, 18 Apr 2017 20:36:08 +0000 (06:36 +1000)]
radv/meta: reduce vertex buffer in blit2d.
Generate the position vertices.
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Signed-off-by: Dave Airlie <airlied@redhat.com>
Dave Airlie [Tue, 18 Apr 2017 05:50:10 +0000 (15:50 +1000)]
radv/meta: reduce vertex buffer usage in clear shaders
For depth clears we have to pass the depth in the 2nd
component, we can use push constants for some of this
later to drop the vertex buffer completely
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Signed-off-by: Dave Airlie <airlied@redhat.com>
Dave Airlie [Tue, 18 Apr 2017 05:38:09 +0000 (15:38 +1000)]
radv/meta: avoid using vertex buffer for resolve shader.
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Signed-off-by: Dave Airlie <airlied@redhat.com>
Dave Airlie [Tue, 18 Apr 2017 05:31:55 +0000 (15:31 +1000)]
radv/meta: move depth decompress to using inline vertex data
This removes the vertex buffer, and just generates the values
in the shader.
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Signed-off-by: Dave Airlie <airlied@redhat.com>
Dave Airlie [Tue, 18 Apr 2017 05:29:21 +0000 (15:29 +1000)]
radv/meta: move fast clear to generate vertices in shader.
Avoids having to setup vertex buffers.
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Signed-off-by: Dave Airlie <airlied@redhat.com>
Dave Airlie [Tue, 18 Apr 2017 20:18:23 +0000 (06:18 +1000)]
radv/meta: add common shader vertex generation function
Instead of passing in the same 1.0, -1.0 combinations via
vertex buffers, we can just use vertex id to have the vertex
shader build them. This function introduces the generator
code needed, later patches will use this.
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Signed-off-by: Dave Airlie <airlied@redhat.com>
Dave Airlie [Tue, 18 Apr 2017 05:28:04 +0000 (15:28 +1000)]
radv/meta: add support for save/restore meta without vertex data.
Some of the shaders could just generate the vertex data in the
shader, so add helpers to allow us to move to doing that.
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Signed-off-by: Dave Airlie <airlied@redhat.com>
Dave Airlie [Tue, 18 Apr 2017 23:30:26 +0000 (09:30 +1000)]
radv: drop debugging leftovers code in descriptor set patches.
Signed-off-by: Dave Airlie <airlied@redhat.com>
Dave Airlie [Tue, 18 Apr 2017 03:22:32 +0000 (13:22 +1000)]
radv: add support for 32 descriptor sets.
This bumps the limit to the number of sets to 32, now that
we have proper support for it. It also uses 1u in a few places
to make things a bit safer.
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Signed-off-by: Dave Airlie <airlied@redhat.com>
Dave Airlie [Tue, 18 Apr 2017 00:21:59 +0000 (10:21 +1000)]
radv/ac: add support for indirect access of descriptor sets.
We want to expose more descriptor sets to the applications,
but currently we have a 1:1 mapping between shader descriptor
sets and 2 user sgprs, limiting us to 4 per stage. This commit
check if we don't have enough user sgprs for the number of
bound sets for this shader, we can ask for them to be indirected.
Two sgprs are then used to point to a buffer or 64-bit pointers
to the number of allocated descriptor sets. All shaders point
to the same buffer.
We can use some user sgprs to inline one or two descriptor sets
in future, but until we have a workload that needs this I don't
think we should spend too much time on it.
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Signed-off-by: Dave Airlie <airlied@redhat.com>
Dave Airlie [Mon, 17 Apr 2017 20:18:01 +0000 (06:18 +1000)]
radv: start allocating user sgprs
This adds an initial implementation to allocate the user
sgprs and make sure we don't run out if we try to bind
a bunch of descriptor sets.
This can be enhanced further in the future if we add
support for inlining push constants.
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Signed-off-by: Dave Airlie <airlied@redhat.com>
Dave Airlie [Mon, 17 Apr 2017 19:44:46 +0000 (05:44 +1000)]
radv/ac: mark used descriptor sets in shader info.
This pre calculates the used descriptor sets.
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Signed-off-by: Dave Airlie <airlied@redhat.com>
Dave Airlie [Mon, 17 Apr 2017 19:35:05 +0000 (05:35 +1000)]
radv/ac: frag shader only needs ring offsets if sample positions enabled
mostly documenting things, since with modern llvm we always have the
spill enabled.
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Signed-off-by: Dave Airlie <airlied@redhat.com>
Dave Airlie [Mon, 17 Apr 2017 19:29:34 +0000 (05:29 +1000)]
radv/ac: move needs_push_constants to shader info.
First step to optimising push constants.
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Signed-off-by: Dave Airlie <airlied@redhat.com>
Dave Airlie [Mon, 17 Apr 2017 19:23:47 +0000 (05:23 +1000)]
radv: optimise compute shader grid size emission.
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Signed-off-by: Dave Airlie <airlied@redhat.com>
Dave Airlie [Mon, 17 Apr 2017 19:15:46 +0000 (05:15 +1000)]
radv: start conditionalising vertex inputs. (v2)
In practice this will probably just drop draw id in a few places.
v2: just do draw_id for now. (Bas)
it might be possible to do something more if we need it in the
future. (nha)
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Signed-off-by: Dave Airlie <airlied@redhat.com>
Dave Airlie [Mon, 17 Apr 2017 03:01:52 +0000 (13:01 +1000)]
radv/ac: add initial pre-pass for shader info gathering
There is some radv specific info we need to gather from shaders
before we get into converting nir->llvm, so we can make
better decisions especially around user sgpr allocation.
This is just an initial placeholder to gather if sample positions
are required in the frag shader.
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Signed-off-by: Dave Airlie <airlied@redhat.com>
Rob Clark [Sat, 15 Apr 2017 14:04:55 +0000 (10:04 -0400)]
freedreno: refactor dirty state handling
In particular, move per-shader-stage info out to a seperate array of
enum's indexed by shader stage. This will make it easier to add more
shader stages as well as new per-stage state (like SSBOs).
Signed-off-by: Rob Clark <robdclark@gmail.com>
Rob Clark [Sun, 16 Apr 2017 18:47:35 +0000 (14:47 -0400)]
freedreno: move clear path dirty state hack to a2xx backend
a3xx/a4xx use the generic u_blitter path, which will make state dirty
bits be set appropriately thanks to the automagic of generic code
setting generic state in the driver. And a5xx has a blit/dma engine
(actually, two) so it doesn't need these extra dirty bits set.
Signed-off-by: Rob Clark <robdclark@gmail.com>
Rob Clark [Sun, 16 Apr 2017 16:39:59 +0000 (12:39 -0400)]
freedreno/ir3: split out per-stage emit_consts fxns
This makes it easier to deal with adding additional stages which have
their own driver-params. The duplicated code this introduces can be
refactored out after a later patch moves to per-shader-stage dirty
flags.
Signed-off-by: Rob Clark <robdclark@gmail.com>
Rob Clark [Sun, 16 Apr 2017 18:52:16 +0000 (14:52 -0400)]
freedreno: add helper to mark all state clean
Note that this involves juggling around a bit when we emit and clear
texture state. So split out from the patch that adds the helper to set
all state dirty.
Signed-off-by: Rob Clark <robdclark@gmail.com>
Rob Clark [Sun, 16 Apr 2017 16:21:49 +0000 (12:21 -0400)]
freedreno: add helper to mark all state dirty
This will simplify things when we break out per-shader-stage dirty bits.
Signed-off-by: Rob Clark <robdclark@gmail.com>