Jason Ekstrand [Thu, 15 Oct 2015 20:45:07 +0000 (13:45 -0700)]
anv/cmd_buffer: Add an alloc_surface_state helper
Jason Ekstrand [Thu, 15 Oct 2015 02:00:17 +0000 (19:00 -0700)]
anv: Get rid of the descriptor_set_binding struct
We no longer need it as we have a better way to deal with dynamic offsets.
Jason Ekstrand [Thu, 15 Oct 2015 00:02:55 +0000 (17:02 -0700)]
anv: Get rid of backend compiler hacks for descriptor sets
Now that we have anv_nir_apply_pipeline_layout, we can hand the backend
compiler intrinsics and texture instructions that use a flat buffer index
just like it wants. There's no longer any reason for any of these hacks.
Jason Ekstrand [Thu, 15 Oct 2015 00:46:24 +0000 (17:46 -0700)]
anv/nir: Rewrite apply_dynamic_offsets to handle the new vk intrinsics
Jason Ekstrand [Wed, 14 Oct 2015 23:39:58 +0000 (16:39 -0700)]
anv/nir: Add a pass for applying a applying a pipeline layout to a shader
This new pass lowers the _vk intrinsics which take a (set, binding, index)
tripple to the single-index non-vk intrinsics based on the pipeline layout.
Jason Ekstrand [Wed, 14 Oct 2015 23:49:34 +0000 (16:49 -0700)]
nir/spirv: Use the Vulkan ubo intrinsics
Jason Ekstrand [Wed, 14 Oct 2015 23:37:59 +0000 (16:37 -0700)]
nir/intrinsics: Add new Vulkan load/store intrinsics
Jason Ekstrand [Wed, 14 Oct 2015 23:37:35 +0000 (16:37 -0700)]
nir/intrinsic: Allow up to four indices
Jason Ekstrand [Wed, 14 Oct 2015 22:18:49 +0000 (15:18 -0700)]
anv: Completely rework descriptor set layouts
This patch reworks a bunch of stuff in the way we do descriptor set
layouts. Our previous approach had a couple of problems. First, it was
based on a misunderstanding of arrays in descriptor sets. Second, it
didn't properly handle descriptor sets where some bindings were missing
stages. The new apporach should be correct and also makes some operations,
particularly those on the hot-path, a bit easier.
We use the descriptor set layout for four things:
1) To determine the map from bindings to the actual flattened descriptor
set in vkUpdateDescriptorSets().
2) To determine the descriptor <-> binding table entry mapping to use in
anv_cmd_buffer_flush_descriptor_sets().
3) To determine the mappings of dynamic indices.
4) To determine the (set, binding, array index) -> binding table entry
mapping inside of shaders.
The new approach is directly taylored towards these operations.
Chad Versace [Wed, 14 Oct 2015 20:56:48 +0000 (13:56 -0700)]
vk: Add README
Requested by developers outside Intel.
During the driver's pre-release development, let's make the README easy
to find for external experimenters. Keep it at the top of the source
tree.
Jason Ekstrand [Wed, 14 Oct 2015 03:00:05 +0000 (20:00 -0700)]
nir/spirv: Fix a bug in indirect OpAccessChain handling
Jason Ekstrand [Tue, 13 Oct 2015 22:50:02 +0000 (15:50 -0700)]
anv/image: Add a basic implementation of GetImageSubresourceLayout
Jason Ekstrand [Tue, 13 Oct 2015 22:20:28 +0000 (15:20 -0700)]
anv/formats: Use the surface_format_info struct from brw_surface_formats.h
The surface_format_info struct changed in mesa but the copied-and-pasted
version didn't get updated on the last mesa master merge. This both fixes
the bug and should prevent this in the future.
Jason Ekstrand [Tue, 13 Oct 2015 22:19:58 +0000 (15:19 -0700)]
i965/surface_formats: Pull the surface_format_info struct into a header
Jason Ekstrand [Tue, 13 Oct 2015 22:00:35 +0000 (15:00 -0700)]
anv/x11: Do something sensible if get_geometry fails in GetSurfaceProperties
Jason Ekstrand [Tue, 13 Oct 2015 18:45:58 +0000 (11:45 -0700)]
anv/wsi: Add the GetSurfacePresentModesKHR stub
Support has existed in the X11 and Wayland backends for a while but,
somehow, the entrypoint got missed in the API shuffle.
Jason Ekstrand [Tue, 13 Oct 2015 01:25:19 +0000 (18:25 -0700)]
anv: Declare/validate the correct API version
Jason Ekstrand [Sat, 10 Oct 2015 16:31:03 +0000 (09:31 -0700)]
anv/device: Return VK_SUCCESS after setting pCount in QueueFamilyProperties
Kristian Høgsberg Kristensen [Thu, 8 Oct 2015 19:25:34 +0000 (12:25 -0700)]
Merge ../mesa into vulkan
Jason Ekstrand [Thu, 8 Oct 2015 19:15:58 +0000 (12:15 -0700)]
anv/queue: Get rid of the serial
This was a remnant of the object tagging implementation we had at one
point. We haven't used it for a long time so there's no good reason to
keep it around.
Kristian Høgsberg Kristensen [Wed, 7 Oct 2015 12:09:48 +0000 (05:09 -0700)]
i965: Link compiler unit tests to libi965_compiler.la
We can now link the unit tests against just libi965_compiler.la. This
lets us drop a lot of DRI driver dependencies, but we still pull in all
of libmesa and more.
This also provides a few standalone users of libi965_compiler.la, which
will help us accidentally using i965_dri.so functions from the compiler.
Reviewed-by: Jason Ekstrand <jason.ekstrand@intel.com>
Signed-off-by: Kristian Høgsberg Kristensen <krh@bitplanet.net>
Kristian Høgsberg Kristensen [Tue, 6 Oct 2015 23:54:52 +0000 (16:54 -0700)]
i965: Break out backend compiler to its own library
This introduces a new libtool helper library, libi965_compiler.la. This
library is moderately self-contained, but still needs to link to all of
libmesa.la among other things.
Reviewed-by: Jason Ekstrand <jason.ekstrand@intel.com>
Signed-off-by: Kristian Høgsberg Kristensen <krh@bitplanet.net>
Kristian Høgsberg Kristensen [Wed, 7 Oct 2015 12:13:50 +0000 (05:13 -0700)]
i965/cs: Get max_cs_threads from brw_compiler devinfo
Reviewed-by: Jason Ekstrand <jason.ekstrand@intel.com>
Signed-off-by: Kristian Høgsberg Kristensen <krh@bitplanet.net>
Kristian Høgsberg Kristensen [Wed, 7 Oct 2015 12:06:30 +0000 (05:06 -0700)]
i965: Move brw_get_shader_time_index() call out of emit functions
brw_get_shader_time_index() is all tangled up in brw_context state and
we can't call it from the compiler. Thanks the Jasons recent
refactoring, we can just get the index and pass to the emit functions
instead.
Reviewed-by: Jason Ekstrand <jason.ekstrand@intel.com>
Signed-off-by: Kristian Høgsberg Kristensen <krh@bitplanet.net>
Kristian Høgsberg Kristensen [Wed, 7 Oct 2015 11:19:39 +0000 (04:19 -0700)]
i965: Move brw_select_clip_planes() to brw_shader.cpp
We call this from the compiler so move it to brw_shader.cpp.
Reviewed-by: Topi Pohjolainen <topi.pohjolainen@intel.com>
Signed-off-by: Kristian Høgsberg Kristensen <krh@bitplanet.net>
Kristian Høgsberg Kristensen [Mon, 5 Oct 2015 21:02:56 +0000 (14:02 -0700)]
i965: Use util_next_power_of_two() for brw_get_scratch_size()
This function computes the next power of two, but at least 1024. We can
do that by bitwise or'ing in 1023 and calling util_next_power_of_two().
We use brw_get_scratch_size() from the compiler so we need it out of
brw_program.c. We could move it to brw_shader.cpp, but let's make it a
small inline function instead.
Reviewed-by: Topi Pohjolainen <topi.pohjolainen@intel.com>
Signed-off-by: Kristian Høgsberg Kristensen <krh@bitplanet.net>
Kristian Høgsberg Kristensen [Tue, 6 Oct 2015 23:11:08 +0000 (16:11 -0700)]
i965: Move brw_mark_surface_used() to brw_shader.cpp
brw_program.c won't be part of the compiler library, but we need
brw_mark_surface_used() in the compiler. Move to brw_shader.cpp.
Reviewed-by: Topi Pohjolainen <topi.pohjolainen@intel.com>
Signed-off-by: Kristian Høgsberg Kristensen <krh@bitplanet.net>
Kristian Høgsberg Kristensen [Tue, 6 Oct 2015 05:07:58 +0000 (22:07 -0700)]
i965/cs: Split out helper for building local id payload
The initial motivation for this patch was to avoid calling
brw_cs_prog_local_id_payload_dwords() in gen7_cs_state.c from the
compiler. This commit ends up refactoring things a bit more so as to
split out the logic to build the local id payload to brw_fs.cpp. This
moves the payload building closer to the compiler code that uses the
payload layout and makes it available to other users of the compiler.
Reviewed-by: Topi Pohjolainen <topi.pohjolainen@intel.com>
Signed-off-by: Kristian Høgsberg Kristensen <krh@bitplanet.net>
Kristian Høgsberg Kristensen [Mon, 5 Oct 2015 21:22:23 +0000 (14:22 -0700)]
i965: Move brw_link_shader() and friends to new file brw_link.cpp
We want to use the rest of brw_shader.cpp with the rest of the compiler
without pulling in the GLSL linking code.
Reviewed-by: Topi Pohjolainen <topi.pohjolainen@intel.com>
Signed-off-by: Kristian Høgsberg Kristensen <krh@bitplanet.net>
Kristian Høgsberg Kristensen [Tue, 6 Oct 2015 23:19:04 +0000 (16:19 -0700)]
i965: Configure bufmgr debug options from intel_screen.c
We need the debug flag parsing and INTEL_DEBUG in the compiler, but we
don't want the dependency on bufmgr (libdrm_intel) in there. Move to
intel_screen.c.
There are now only two lines left in brw_process_intel_debug_variable(),
but we keep it in intel_debug.h to avoid having to expose
'debug_control' as a global variable.
Reviewed-by: Topi Pohjolainen <topi.pohjolainen@intel.com>
Reviewed-by: Iago Toral Quiroga <itoral@igalia.com>
Signed-off-by: Kristian Høgsberg Kristensen <krh@bitplanet.net>
Kristian Høgsberg Kristensen [Mon, 5 Oct 2015 21:13:29 +0000 (14:13 -0700)]
util: Move DRI parse_debug_string() to util
We want to use intel_debug.c in code that doesn't link to dri common.
v2: Remove unnecessary stddef.h include (Topi), use util/debug.h
in all DRI driver and remove driParseDebugString() (Iago).
Reviewed-by: Topi Pohjolainen <topi.pohjolainen@intel.com>
Signed-off-by: Kristian Høgsberg Kristensen <krh@bitplanet.net>
Kristian Høgsberg Kristensen [Mon, 5 Oct 2015 20:58:05 +0000 (13:58 -0700)]
i965: Move brw_dump_ir() out of brw_*_emit() functions
We move these calls one level up into the codegen functions.
Reviewed-by: Topi Pohjolainen <topi.pohjolainen@intel.com>
Signed-off-by: Kristian Høgsberg Kristensen <krh@bitplanet.net>
Jason Ekstrand [Wed, 7 Oct 2015 21:13:55 +0000 (14:13 -0700)]
vk/0.170.2: Update to the new VK_EXT_KHR_swapchain extensions
Emil Velikov [Thu, 8 Oct 2015 14:50:54 +0000 (15:50 +0100)]
gallium/ddebug: add missing dd_util.h to sources list
Signed-off-by: Emil Velikov <emil.velikov@collabora.com>
Emil Velikov [Thu, 8 Oct 2015 14:50:12 +0000 (15:50 +0100)]
gallium/ddebug: automake: sort sources alphabetically
Signed-off-by: Emil Velikov <emil.velikov@collabora.com>
Jason Ekstrand [Tue, 6 Oct 2015 00:16:02 +0000 (17:16 -0700)]
nir/sweep: Reparent the shader name
Previously the name of the nir shader was being freed prematurely during
nir_sweep. Since
756613ed35d the name was later being used to generate
filenames for the optimiser debug output and these would end up with
garbage from the dangling pointer.
Co-authored-by: Neil Roberts <neil@linux.intel.com>
Reviewed-by: Matt Turner <mattst88@gmail.com>
Jan Vesely [Sun, 4 Oct 2015 00:19:13 +0000 (19:19 -0500)]
c11/threads: initialize timeout structure
Signed-off-by: Jan Vesely <jano.vesely@gmail.com>
Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
Boyan Ding [Thu, 8 Oct 2015 07:38:15 +0000 (15:38 +0800)]
docs/relnotes: document EGL_KHR_create_context on llvmpipe and softpipe
Signed-off-by: Boyan Ding <boyan.j.ding@gmail.com>
Iago Toral Quiroga [Wed, 23 Sep 2015 06:52:07 +0000 (08:52 +0200)]
i965/gs/gen6: Maximum allowed size of SEND messages is 15 (4 bits)
Comit
d48ac9306619 addressed this for VS, but we forgot to do the same for
URB writes generated by the gen6 GS.
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Iago Toral Quiroga [Tue, 22 Sep 2015 11:14:52 +0000 (13:14 +0200)]
i965: Define FIRST_SPILL_MRF and FIRST_PULL_LOAD_MRF only once and in one place
That should make tracking where we do spills and pull loads a bit easier.
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Iago Toral Quiroga [Tue, 22 Sep 2015 11:01:18 +0000 (13:01 +0200)]
i965: make pull constant loads in gen6 start at MRFs 16/17
So they do not conflict with our (un)spills (MRF 21..23) or our
URB writes (MRF 1..15)
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Iago Toral Quiroga [Tue, 22 Sep 2015 10:53:08 +0000 (12:53 +0200)]
i965: Fix remove_duplicate_mrf_writes so it can handle 24 MRFs in gen6
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Tapani Pälli [Thu, 8 Oct 2015 06:25:16 +0000 (09:25 +0300)]
mesa: include bad type in error string of _mesa_pack_depth_span
Signed-off-by: Tapani Pälli <tapani.palli@intel.com>
Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
Tapani Pälli [Wed, 7 Oct 2015 07:04:06 +0000 (10:04 +0300)]
glsl: add varyings to resource list only with SSO
Varyings can be considered inputs or outputs of a program only when
SSO is in use. With multi-stage programs, inputs contain only inputs
for first stage and outputs contains outputs of the final shader stage.
I've tested that fix works for Assault Android Cactus (demo version)
and does not cause Piglit or CTS regressions in glGetProgramiv tests.
Following ES 3.1 CTS separate shader tests that do query properties
of varyings in SSO shader programs pass:
ES31-CTS.program_interface_query.separate-programs-vertex
ES31-CTS.program_interface_query.separate-programs-fragment
Signed-off-by: Tapani Pälli <tapani.palli@intel.com>
Tested-by: Dieter Nützel <Dieter@nuetzel-hh.de>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=92122
Jason Ekstrand [Wed, 7 Oct 2015 22:52:09 +0000 (15:52 -0700)]
mesa: Correctly handle GL_BGRA_EXT in ES3 format_and_type checks
The EXT_texture_format_BGRA8888 extension (which mesa supports
unconditionally) adds a new format and internal format called GL_BGRA_EXT.
Previously, this was not really handled at all in
_mesa_ex3_error_check_format_and_type. When the checks were tightened in
commit
f15a7f3c, we accidentally tightened things too far and GL_BGRA_EXT
would always cause an error to be thrown.
There were two primary issues here. First, is that
_mesa_es3_effective_internal_format_for_format_and_type didn't handle the
GL_BGRA_EXT format. Second is that it blindly uses _mesa_base_tex_format
which returns GL_RGBA for GL_BGRA_EXT. This commit fixes both of these
issues as well as adds explicit checks that GL_BGRA_EXT is only ever used
with GL_BGRA_EXT and GL_UNSIGNED_BYTE.
Signed-off-by: Jason Ekstrand <jason.ekstrand@intel.com>
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=92265
Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
Cc: "11.0" <mesa-stable@lists.freedesktop.org>
Emil Velikov [Wed, 7 Oct 2015 20:23:16 +0000 (21:23 +0100)]
Revert "mesa: enable KHR_debug for ES contexts"
This reverts commit
b69cfbdf18fa64606a76761b20bc268f4ac731e5.
This isn't quite baked yet. Seems that despite building the ES piglits,
none of them got executed.
Chad Versace [Wed, 7 Oct 2015 18:48:29 +0000 (11:48 -0700)]
Merge branch 'vulkan-0.170.2' into vulkan
This updates the API from 0.138.2 to 0.170.2,
and updates SPIR-V to v32.
Matt Turner [Wed, 7 Oct 2015 18:43:58 +0000 (11:43 -0700)]
egl/dri2: Properly dereference array.
Fixes a regression that broke EGL since
commit
858f2f2ae6d72f338fdd6d544b0c733814e22724
Author: Emil Velikov <emil.l.velikov@gmail.com>
Date: Sun Sep 13 12:25:27 2015 +0100
egl/dri2: ease srgb __DRIconfig conditionals
Chad Versace [Wed, 7 Oct 2015 18:45:48 +0000 (11:45 -0700)]
vk: Remove temporary vulkan headers
Remove vulkan-0.138.2.h and vulkan-0.170.2.h. Their purpose was to aid
the header update to 0.170.2.
Chad Versace [Wed, 7 Oct 2015 18:44:44 +0000 (11:44 -0700)]
vk/0.170.2: Bump header version
The header is now fully updated.
Chad Versace [Wed, 7 Oct 2015 18:43:13 +0000 (11:43 -0700)]
vk/0.170.2: Update C++ errata and typedefs
Chad Versace [Wed, 7 Oct 2015 18:39:49 +0000 (11:39 -0700)]
vk/0.170.2: Update remaining enums
Chad Versace [Wed, 7 Oct 2015 18:36:51 +0000 (11:36 -0700)]
vk/0.170.2: Update VkResult
Version 0.170.2 removes most of the error enums. In many cases, I had to
replace an error with a less accurate (or even incorrect) one.
In other cases, the error path is replaced with an assertion.
Chad Versace [Wed, 7 Oct 2015 17:58:55 +0000 (10:58 -0700)]
vk/0.170: Update VkDescriptorInfo
Ignore the new bufferInfo field with a anv_finishme.
Chad Versace [Wed, 7 Oct 2015 17:45:41 +0000 (10:45 -0700)]
vk/0.170.2: Update vkCreateDescriptorPool
Nothing to do. In Mesa the pool is a stub.
Chad Versace [Wed, 7 Oct 2015 17:44:40 +0000 (10:44 -0700)]
vk/0.170.2: Update VkAttachmentDescription
Chad Versace [Wed, 7 Oct 2015 17:43:44 +0000 (10:43 -0700)]
vk/0.170.2: Update VkImageViewCreateInfo
Chad Versace [Wed, 7 Oct 2015 17:42:59 +0000 (10:42 -0700)]
vk/0.170.2: Update VkImageCreateInfo
Chad Versace [Wed, 7 Oct 2015 17:36:46 +0000 (10:36 -0700)]
vk/0.170.2: Update VkPhysicalDeviceProperties
Chad Versace [Wed, 7 Oct 2015 17:28:30 +0000 (10:28 -0700)]
vk/0.170.2: Update VkImageFormatProperties
Marek Olšák [Fri, 2 Oct 2015 17:21:54 +0000 (19:21 +0200)]
radeonsi: fix a GS hang on VI
Broken by one of the cleanups:
0d46c3bc9d09b376d74f7399e1a2d1b0a923640b
Not applicable to stable.
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Marek Olšák [Sun, 6 Sep 2015 13:43:23 +0000 (15:43 +0200)]
radeonsi: remove TC L2 cache flush for index buffers on VI
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Chad Versace [Wed, 7 Oct 2015 17:15:59 +0000 (10:15 -0700)]
vk/0.170.2: Update VkFormatProperties
Chad Versace [Wed, 7 Oct 2015 17:05:02 +0000 (10:05 -0700)]
vk/0.170.2: Update VkPhysicalDeviceFeatures
Chad Versace [Wed, 7 Oct 2015 16:57:51 +0000 (09:57 -0700)]
vk/0.170.2: Update VkPhysicalDeviceLimits
Jason Ekstrand [Wed, 7 Oct 2015 16:45:47 +0000 (09:45 -0700)]
anv/device: Remove some #ifdef'd out code
This was a left-over from the dynamic state update.
Jason Ekstrand [Wed, 7 Oct 2015 16:44:51 +0000 (09:44 -0700)]
vk/0.170.2: Make vkUpdateDescriptorSets return void
Jason Ekstrand [Wed, 7 Oct 2015 16:28:21 +0000 (09:28 -0700)]
anv/pipeline: Add support for dynamic state in pipelines
Jason Ekstrand [Wed, 7 Oct 2015 00:21:44 +0000 (17:21 -0700)]
vk/0.170.2: Switch to the new dynamic state model
Jason Ekstrand [Wed, 7 Oct 2015 16:31:53 +0000 (09:31 -0700)]
anv: Add a dynamic state data structure and basic helpers
Jason Ekstrand [Wed, 7 Oct 2015 16:27:31 +0000 (09:27 -0700)]
anv/private: Add a typed_memcpy macro
This is amazingly helpful when copying arrays of things around.
Chad Versace [Wed, 7 Oct 2015 16:09:37 +0000 (09:09 -0700)]
vk/meta: Fix -Wstrict-prototypes
In C, functions with no arguments require a void argument.
build_nir_clear_fragment_shader() lacked that.
Fixes:
anv_meta.c:70:1: warning: function declaration isn't a prototype
[-Wstrict-prototypes]
Chad Versace [Wed, 7 Oct 2015 14:30:52 +0000 (07:30 -0700)]
vk/0.170.2: Merge VkAttachmentView into VkImageView
Chad Versace [Wed, 7 Oct 2015 16:03:47 +0000 (09:03 -0700)]
vk/image: Fix retrieval of anv_surface for depthstencil aspect
If anv_image_get_surface_for_aspect_mask() is given a combined
depthstencil aspect mask, and the image has a stencil surface but no
depth surface, then return the stencil surface.
Hacks on hacks.
Brian Paul [Tue, 6 Oct 2015 22:55:39 +0000 (16:55 -0600)]
svga: whitespace fixes in svga_sampler_view.c
Brian Paul [Tue, 6 Oct 2015 22:55:39 +0000 (16:55 -0600)]
svga: whitespace fixes in svga_resource_buffer.c
Stefan Dösinger [Tue, 6 Oct 2015 22:55:39 +0000 (16:55 -0600)]
mesa: Remove GL_ARB_sampler_object depth compare error checking.
Version 3: Simplify the code comment, word wrap commit description.
Version 2: Return GL_FALSE if ARB_shadow is unsupported instead of
pretending to store the value as suggested by Brian Paul.
This fixes a GL error warning on r200 in Wine.
The GL_ARB_sampler_objects extension does not specify a dependency on
GL_ARB_shadow or GL_ARB_depth_texture for setting the depth texture
compare mode and function. Silently ignore attempts to change these
settings. They won't matter without a depth texture being assigned
anyway.
Reviewed-by: Brian Paul <brianp@vmware.com>
Brian Paul [Tue, 6 Oct 2015 22:55:39 +0000 (16:55 -0600)]
svga: round UBO constant buffer size up/down to multiple of 16 bytes
The svga3d device requires constant buffers to be a multiple of 16 bytes
in size. OpenGL UBOs may not fit that restriction. As a work-around,
round the size up if possible, else round down.
Note that this patch only effects UBO constant buffers (index 1 or higher),
not the 0th/default constant buffer.
Fixes the game Grim Fandango Remastered. VMware bug
1510130.
Reviewed-by: Charmaine Lee <charmainel@vmware.com>
Reviewed-by: José Fonseca <jfonseca@vmware.com>
Chad Versace [Wed, 7 Oct 2015 14:28:46 +0000 (07:28 -0700)]
vk: Drop -Wextra
Eliminates lots of warnings due to anv_meta.c's inclusion of nir.h.
I like the extra warnings, and they should probably get fixed. However,
git-grep reveals that no other Mesa directory uses -Wextra. Building
Vulkan produces a lot of compiler warnings from core Mesa headers that
no other Mesa developer sees, and hence no other Mesa developer will
fix.
Emil Velikov [Sun, 13 Sep 2015 11:36:54 +0000 (12:36 +0100)]
egl/dri2: enable EGL_KHR_gl_colorspace for swrast
No driver changes needed for softpipe/llvmpipe - things just work.
v2: Whitespace fixes.
Signed-off-by: Emil Velikov <emil.l.velikov@gmail.com>
Reviewed-by: Boyan Ding <boyan.j.ding@gmail.com>
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Emil Velikov [Sun, 13 Sep 2015 11:25:27 +0000 (12:25 +0100)]
egl/dri2: ease srgb __DRIconfig conditionals
One can simplify the if-else chain, by declaring the driconfigs as a
two sized array, whist using srgb as a index to the correct entry.
Signed-off-by: Emil Velikov <emil.l.velikov@gmail.com>
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Emil Velikov [Mon, 14 Sep 2015 17:35:48 +0000 (18:35 +0100)]
mesa: enable KHR_debug for ES contexts
Signed-off-by: Emil Velikov <emil.l.velikov@gmail.com>
Matthew Waters [Wed, 16 Sep 2015 15:38:27 +0000 (16:38 +0100)]
main/get: make KHR_debug enums available everywhere
Move all the enums but CONTEXT_FLAGS. The spec seems quite explicit
about the latter (wrt OpenGL ES)
"In OpenGL ES versions prior to and including ES 3.1 there is no
CONTEXT_FLAGS state and therefore the CONTEXT_FLAG_DEBUG_BIT cannot
be queried."
v2 [Emil Velikov] Rebase.
v3 [Emil Veliokv] Drop the CONTEXT_FLAGS hunk - not applicable for GLES
Signed-off-by: Matthew Waters <ystreet00@gmail.com>
Signed-off-by: Emil Velikov <emil.l.velikov@gmail.com>
Matthew Waters [Mon, 14 Sep 2015 17:35:46 +0000 (18:35 +0100)]
glapi: add function pointers for KHR_debug for gles
v2 [Emil Velikov]
- Rebase.
- Correct version in gles11 dispatch_sanity.
- Move the extension enable to a separate patch.
Signed-off-by: Matthew Waters <ystreet00@gmail.com>
Signed-off-by: Emil Velikov <emil.l.velikov@gmail.com>
Varad Gautam [Wed, 7 Oct 2015 04:18:15 +0000 (09:48 +0530)]
egl: move memcpy to bring conf->base operations together
Signed-off-by: Varad Gautam <varadgautam@gmail.com>
Suggested-by: Emil Velikov <emil.l.velikov@gmail.com>
Reviewed-by: Emil Velikov <emil.l.velikov@gmail.com>
Varad Gautam [Wed, 7 Oct 2015 04:18:14 +0000 (09:48 +0530)]
egl: restore surface type before linking config to its display
commit
c2c2e9a (egl: implement EGL_KHR_gl_colorspace (v2)) leaves
_EGLConfig->SurfaceType set incorrectly before calling _eglLinkConfig(),
and the bad value is passed around to platform_android. set it to zero
as earlier.
v2: Set SurfaceType to 0, rather than surface_type (Suggested by Emil)
Cc: mesa-stable@lists.freedesktop.org
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=91596
Signed-off-by: Varad Gautam <varadgautam@gmail.com>
Reviewed-by: Emil Velikov <emil.l.velikov@gmail.com>
Ilia Mirkin [Mon, 5 Oct 2015 19:19:05 +0000 (15:19 -0400)]
nouveau: make sure there's always room to emit a fence
I started seeing a lot of situations on nv30 where fence emission
wouldn't fit into the previous buffer (causing assertions). This ensures
that whenever checking for space, we always leave a bit of extra room
for the fence emission commands. Adjusts the nv30 and nvc0 fence
emission logic to bypass the space checking as well.
Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu>
Cc: mesa-stable@lists.freedesktop.org
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Chad Versace [Wed, 7 Oct 2015 02:11:58 +0000 (19:11 -0700)]
vk: Embed two surface states in anv_image_view
This prepares for merging VkAttachmentView into VkImageView. The two
surface states are:
anv_image_view::color_rt_surface_state:
RENDER_SURFACE_STATE when using image as a color render target.
anv_image_view::nonrt_surface_state;
RENDER_SURFACE_STATE when using image as a non render target.
No Crucible regressions.
Chad Versace [Wed, 7 Oct 2015 01:55:16 +0000 (18:55 -0700)]
vk/pipeline: Emit MSAA finishme only if samples > 1
If samples == 1, then there's nothing for Mesa to do, and the finishme
message is only noise.
Chad Versace [Wed, 7 Oct 2015 01:53:29 +0000 (18:53 -0700)]
vk: Remove stale finishme for stencil image views
They don't work completely. But they work well enough to satisfy
Crucible.
Chad Versace [Wed, 7 Oct 2015 01:17:09 +0000 (18:17 -0700)]
vk: Add anv_image::usage
It's a copy of VkImageCreateInfo::usage. Will be used for the
VkAttachmentView/VkImageView merge.
Chad Versace [Wed, 7 Oct 2015 01:34:37 +0000 (18:34 -0700)]
vk/meta: Fix usage flags for image-wrapped-buffers
In make_image_for_buffer(), use VK_IMAGE_USAGE_SAMPLED_BIT when
transferring from the buffer and use VK_IMAGE_USAGE_COLOR_ATTACHMENT_BIT
when transferring to the buffer.
Chad Versace [Wed, 7 Oct 2015 01:18:47 +0000 (18:18 -0700)]
vk/image: Remove stale anv_asserts for depthstencil attachments
We don't fully handle mipmapped, array depthstencil attachments. But we
handle the well enough for Crucible's miptree tests.
Boyan Ding [Sun, 27 Sep 2015 09:16:57 +0000 (17:16 +0800)]
vc4: use nir two-sided-color lowering
Similar to
9ffc1049ca (freedreno/ir3: use nir two-sided-color lowering).
No piglit regression.
Signed-off-by: Boyan Ding <boyan.j.ding@gmail.com>
Reviewed-by: Eric Anholt <eric@anholt.net>
Eric Anholt [Tue, 6 Oct 2015 23:32:03 +0000 (16:32 -0700)]
vc4: Fix a leak of the last color read/write surface on context destroy.
Eric Anholt [Tue, 6 Oct 2015 23:12:50 +0000 (16:12 -0700)]
vc4: Fix a memory leak in the simulator case.
We validate per draw call, and need to free the shader per draw call, too.
Kristian Høgsberg Kristensen [Fri, 2 Oct 2015 21:38:10 +0000 (14:38 -0700)]
i965: Delete brw_cs.cpp which was deleted in master
Jason Ekstrand [Tue, 6 Oct 2015 22:10:45 +0000 (15:10 -0700)]
nir/spirv: Better texture handling
Jason Ekstrand [Tue, 6 Oct 2015 21:44:38 +0000 (14:44 -0700)]
nir/spirv: Update to SPIR-V Rev. 32
Jason Ekstrand [Tue, 6 Oct 2015 17:34:59 +0000 (10:34 -0700)]
vk/0.170.2: Fairly trivial enum shuffling