Kristian Høgsberg Kristensen [Fri, 4 Mar 2016 20:56:14 +0000 (12:56 -0800)]
anv: Also cache the struct anv_pipeline_binding maps
This is state the we generate when compiling the shaders and we need it
for mapping resources from descriptor sets to binding table indices.
Kristian Høgsberg Kristensen [Fri, 4 Mar 2016 20:27:31 +0000 (12:27 -0800)]
anv: Don't re-upload shaders when merging
Using anv_pipeline_cache_upload_kernel() will re-upload the kernel and
prog_data when we merge caches. Since the kernel and prog_data is
already in the program_stream, use anv_pipeline_cache_add_entry()
instead to only add the entry to the hash table.
Kristian Høgsberg Kristensen [Fri, 4 Mar 2016 20:25:23 +0000 (12:25 -0800)]
anv: Add anv_pipeline_cache_add_entry()
This function will grow the cache to make room and then add the entry.
Kristian Høgsberg Kristensen [Fri, 4 Mar 2016 20:21:43 +0000 (12:21 -0800)]
anv: Rename anv_pipeline_cache_add_entry() to 'set'
This function is a helper that unconditionally sets a hash table entry
and expects the cache to have enough room. Calling it 'add_entry'
suggests it will grow the cache as needed.
Kristian Høgsberg Kristensen [Fri, 4 Mar 2016 18:59:21 +0000 (10:59 -0800)]
anv: Simplify pipeline cache control flow a bit
No functional change, but the control flow around searching the cache
and falling back to compiling is a bit simpler.
Kristian Høgsberg Kristensen [Fri, 4 Mar 2016 16:15:16 +0000 (08:15 -0800)]
anv: Store prog data in pipeline cache stream
We have to keep it there for the cache to work, so let's not have an
extra copy in struct anv_pipeline too.
Kristian Høgsberg Kristensen [Fri, 4 Mar 2016 00:52:20 +0000 (16:52 -0800)]
anv: Rename 'table' to 'hash_table' in anv_pipeline_cache
A little less ambiguous.
Kristian Høgsberg Kristensen [Fri, 4 Mar 2016 00:48:31 +0000 (16:48 -0800)]
anv: Serialize as much pipeline cache as we can
We can serialize as much as the application asks for and just stop once
we run out of memory. This lets applications use a fixed amount of
space for caching and still get some benefit.
Kristian Høgsberg Kristensen [Fri, 4 Mar 2016 00:39:59 +0000 (16:39 -0800)]
anv: Use 1.0 pipeline cache header
The final version of the pipeline cache header adds a few more fields.
Kristian Høgsberg Kristensen [Fri, 4 Mar 2016 00:21:17 +0000 (16:21 -0800)]
anv: Fix shader key hashing
This was copied from inline code to a helper and wasn't updated to hash
a pointer instead.
Kristian Høgsberg Kristensen [Fri, 4 Mar 2016 00:18:30 +0000 (16:18 -0800)]
anv: Remove excess whitespace
Kristian Høgsberg Kristensen [Fri, 4 Mar 2016 00:10:29 +0000 (16:10 -0800)]
anv: Remove left-over bits of sparse-descriptor code
Jason Ekstrand [Sat, 5 Mar 2016 17:19:01 +0000 (09:19 -0800)]
anv/pipeline: Implement the depth compare EQUAL workaround on gen8+
Jason Ekstrand [Sat, 5 Mar 2016 17:13:44 +0000 (09:13 -0800)]
anv: Don't allow D16_UNORM to be combined with stencil
Among other things, this can cause the depth or stencil test to spurriously
fail when the fragment shader uses discard.
Jason Ekstrand [Sat, 5 Mar 2016 16:45:01 +0000 (08:45 -0800)]
anv/pipeline: Calculate the correct max_source_attr for 3DSTATE_SBE
Jordan Justen [Sat, 5 Mar 2016 08:54:54 +0000 (00:54 -0800)]
anv/hsw: Move query code to genX file for Haswell
This fixes many CTS cases, but will require an update to the kernel
command parser register whitelist. (The CS GPRs and TIMESTAMP
registers need to be whitelisted.)
Signed-off-by: Jordan Justen <jordan.l.justen@intel.com>
Nanley Chery [Sat, 5 Mar 2016 04:41:05 +0000 (20:41 -0800)]
anv: Implement VK_REMAINING_{MIP_LEVELS,ARRAY_LAYERS}
v2: Subtract the baseMipLevel and baseArrayLayer (Jason)
Signed-off-by: Nanley Chery <nanley.g.chery@intel.com>
Reviewed-by: Jason Ekstrand <jason.ekstrand@intel.com>
Jason Ekstrand [Sat, 5 Mar 2016 03:14:48 +0000 (19:14 -0800)]
anv/meta_clear: Set the right number of dynamic states
Jason Ekstrand [Sat, 5 Mar 2016 01:56:12 +0000 (17:56 -0800)]
anv/pipeline: Fix depthBiasEnable on gen7
The first time I tried to fix this, I set the wrong fields.
Jason Ekstrand [Fri, 4 Mar 2016 20:42:03 +0000 (12:42 -0800)]
anv/cmd_buffer: Reset the state streams when resetting the command buffer
Jason Ekstrand [Fri, 4 Mar 2016 23:38:11 +0000 (15:38 -0800)]
anv/cmd_buffer: Include Haswell in set_subpass
Nanley Chery [Thu, 3 Mar 2016 23:49:13 +0000 (15:49 -0800)]
isl: Fix RenderTargetViewExtent for mipmapped 3D surfaces
Match the comment stated above the assignment.
Signed-off-by: Nanley Chery <nanley.g.chery@intel.com>
Reviewed-by: Jason Ekstrand <jason.ekstrand@intel.com>
Nanley Chery [Thu, 3 Mar 2016 23:40:13 +0000 (15:40 -0800)]
isl: Get rid of isl_surf_fill_state_info::level0_extent_px
This field is no longer needed.
Signed-off-by: Nanley Chery <nanley.g.chery@intel.com>
Reviewed-by: Jason Ekstrand <jason.ekstrand@intel.com>
Jason Ekstrand [Fri, 4 Mar 2016 20:23:01 +0000 (12:23 -0800)]
anv/cmd_buffer: Let the pipeline set StencilBufferWriteEnable on gen9
Jason Ekstrand [Fri, 4 Mar 2016 20:22:32 +0000 (12:22 -0800)]
anv/cmd_buffer: Mask stencil reference values
Jason Ekstrand [Fri, 4 Mar 2016 19:07:27 +0000 (11:07 -0800)]
anv/clear: Pull the stencil write mask from the pipeline
The stencil write mask wasn't getting set at all so we were using whatever
write mask happend to be left over by the application.
Jason Ekstrand [Fri, 4 Mar 2016 18:45:24 +0000 (10:45 -0800)]
anv/pipeline: Set StencilBufferWriteEnable from the pipeline
The hardware docs say that StencilBufferWriteEnable should only be set if
StencilTestEnable is set. It seems reasonable to set them together.
Jason Ekstrand [Tue, 1 Mar 2016 21:39:04 +0000 (13:39 -0800)]
anv/pipeline: More competent gen8 clipping
Jason Ekstrand [Fri, 4 Mar 2016 08:02:18 +0000 (00:02 -0800)]
anv/pipeline: Use the right provoking vertex for triangle fans
Jason Ekstrand [Fri, 4 Mar 2016 08:01:25 +0000 (00:01 -0800)]
anv/pipeline: Respect pRasterizationState->depthBiasEnable
Kenneth Graunke [Fri, 4 Mar 2016 00:43:49 +0000 (16:43 -0800)]
anv: Fix backwards shadow comparisons
sample_c is backwards from what GL and Vulkan expect.
See intel_state.c in i965.
v2: Drop unused vk_to_gen_compare_op.
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Kenneth Graunke [Fri, 4 Mar 2016 00:21:09 +0000 (16:21 -0800)]
hack to make dota 2 menus work
Jason Ekstrand [Fri, 4 Mar 2016 00:15:54 +0000 (16:15 -0800)]
isl/surface_state: Set L2 bypass disable for certain BC* formats
Kenneth Graunke [Wed, 24 Feb 2016 23:41:24 +0000 (15:41 -0800)]
anv: Compile shader stages in pipeline order.
Instead of the arbitrary order modules might be specified in.
Acked-by: Jason Ekstrand <jason.ekstrand@intel.com>
Nanley Chery [Mon, 29 Feb 2016 22:37:48 +0000 (14:37 -0800)]
anv/meta: Delete unused functions
Signed-off-by: Nanley Chery <nanley.g.chery@intel.com>
Reviewed-by: Jason Ekstrand <jason.ekstrand@intel.com>
Nanley Chery [Wed, 2 Mar 2016 22:33:51 +0000 (14:33 -0800)]
anv/meta: Use blitter API for state-handling in Buffer Update/Copy
Signed-off-by: Nanley Chery <nanley.g.chery@intel.com>
Reviewed-by: Jason Ekstrand <jason.ekstrand@intel.com>
Nanley Chery [Sun, 28 Feb 2016 01:11:55 +0000 (17:11 -0800)]
anv/meta: Use blitter API in do_buffer_copy()
v2: Keep pitch in units of bytes (Jason)
Signed-off-by: Nanley Chery <nanley.g.chery@intel.com>
Reviewed-by: Jason Ekstrand <jason.ekstrand@intel.com>
Nanley Chery [Mon, 29 Feb 2016 20:19:51 +0000 (12:19 -0800)]
anv/meta: Use blitter API in anv_CmdCopyImage()
Signed-off-by: Nanley Chery <nanley.g.chery@intel.com>
Reviewed-by: Jason Ekstrand <jason.ekstrand@intel.com>
Nanley Chery [Sun, 28 Feb 2016 01:11:55 +0000 (17:11 -0800)]
anv/meta: Use blitter API for copies between Images and Buffers
Signed-off-by: Nanley Chery <nanley.g.chery@intel.com>
Reviewed-by: Jason Ekstrand <jason.ekstrand@intel.com>
Nanley Chery [Mon, 29 Feb 2016 22:28:25 +0000 (14:28 -0800)]
anv/meta: Add function which copies between Buffers and Images
v2: Keep pitch in units of bytes (Jason)
Signed-off-by: Nanley Chery <nanley.g.chery@intel.com>
Reviewed-by: Jason Ekstrand <jason.ekstrand@intel.com>
Nanley Chery [Wed, 2 Mar 2016 07:15:35 +0000 (23:15 -0800)]
anv/meta: Add function to create anv_meta_blit2d_surf from anv_image
v2: Keep pitch in units of bytes (Jason)
Signed-off-by: Nanley Chery <nanley.g.chery@intel.com>
Reviewed-by: Jason Ekstrand <jason.ekstrand@intel.com>
Nanley Chery [Sat, 27 Feb 2016 20:26:21 +0000 (12:26 -0800)]
anv/meta: Implement the blitter API functions
Most of the code in anv_meta_blit2d() is borrowed from do_buffer_copy().
Create an image and image view for each rectangle.
Note: For tiled RGB images, ISL will align the image's row_pitch up to
the nearest tile width.
v2 (Jason):
Keep pitch in units of bytes
Make src_format and dst_format variables
s/dest/dst/ in every usage
v3: Fix dst_image width
Signed-off-by: Nanley Chery <nanley.g.chery@intel.com>
Reviewed-by: Jason Ekstrand <jason.ekstrand@intel.com>
Nanley Chery [Thu, 25 Feb 2016 23:21:12 +0000 (15:21 -0800)]
anv/meta: Modify blitter API fields
Some fields are unnecessary. The variables "pitch" and "bs" are used
for consistency with ISL.
v2: Keep pitch in units of bytes (Jason)
Signed-off-by: Nanley Chery <nanley.g.chery@intel.com>
Reviewed-by: Jason Ekstrand <jason.ekstrand@intel.com>
Jason Ekstrand [Thu, 31 Dec 2015 23:37:54 +0000 (15:37 -0800)]
anv/meta: Add the beginnings of a blitter API
This API is designed to be an abstraction that sits between the VkCmdCopy
commands and the hardware. The idea is that it is simple enough that it
*should* be implementable using the blitter but with enough extra data that
we can implement it with the 3-D pipeline efficiently. One design
objective is to allow the user to supply enough information that we can
handle most blit operations with a single draw call even if they require
copying multiple rectangles.
Nanley Chery [Sat, 27 Feb 2016 21:29:04 +0000 (13:29 -0800)]
anv/meta: Remove redundancies in do_buffer_copy()
Signed-off-by: Nanley Chery <nanley.g.chery@intel.com>
Reviewed-by: Jason Ekstrand <jason.ekstrand@intel.com>
Nanley Chery [Sat, 27 Feb 2016 22:25:00 +0000 (14:25 -0800)]
anv/meta: Replace copy_format w/ block size in do_buffer_copy()
This is a preparatory commit that will simplify the future usage of
this function.
Signed-off-by: Nanley Chery <nanley.g.chery@intel.com>
Reviewed-by: Jason Ekstrand <jason.ekstrand@intel.com>
Nanley Chery [Wed, 2 Mar 2016 22:27:17 +0000 (14:27 -0800)]
anv/meta: Add missing command to exit meta in anv_CmdUpdateBuffer()
Signed-off-by: Nanley Chery <nanley.g.chery@intel.com>
Reviewed-by: Jason Ekstrand <jason.ekstrand@intel.com>
Nanley Chery [Wed, 2 Mar 2016 17:44:48 +0000 (09:44 -0800)]
anv/image: Create a linear image when requested
If a linear image is requested, the only possible result should be a
linearly-tiled surface.
Signed-off-by: Nanley Chery <nanley.g.chery@intel.com>
Reviewed-by: Jason Ekstrand <jason.ekstrand@intel.com>
Nanley Chery [Wed, 2 Mar 2016 01:32:14 +0000 (17:32 -0800)]
isl: Don't filter tiling flags if a specific tiling bit is set
If a specific bit is set, the intention to create a surface with a
specific tiling format should be respected.
Signed-off-by: Nanley Chery <nanley.g.chery@intel.com>
Reviewed-by: Jason Ekstrand <jason.ekstrand@intel.com>
Nanley Chery [Fri, 26 Feb 2016 22:49:51 +0000 (14:49 -0800)]
isl: Add function to get intratile offsets from x/y offsets
Signed-off-by: Nanley Chery <nanley.g.chery@intel.com>
Reviewed-by: Jason Ekstrand <jason.ekstrand@intel.com>
Jason Ekstrand [Thu, 3 Mar 2016 16:17:36 +0000 (08:17 -0800)]
anv/util: Fix vector resizing
It wasn't properly handling the fact that wrap-around in the source may not
translate to wrap-around in the destination. This really needs unit tests.
Jordan Justen [Wed, 2 Mar 2016 09:11:29 +0000 (01:11 -0800)]
anv/gen7: Use predicated rendering for indirect compute
For OpenGL, see commit
9a939ebb47a0d37a6b29e3dbb1b20bdc9538a721.
Fixes:
* dEQP-VK.compute.indirect_dispatch.upload_buffer.empty_command
* dEQP-VK.compute.indirect_dispatch.gen_in_compute.empty_command
Signed-off-by: Jordan Justen <jordan.l.justen@intel.com>
Jordan Justen [Wed, 2 Mar 2016 09:09:16 +0000 (01:09 -0800)]
anv: Save batch to local variable for indirect compute
Signed-off-by: Jordan Justen <jordan.l.justen@intel.com>
Jason Ekstrand [Wed, 2 Mar 2016 19:31:32 +0000 (11:31 -0800)]
anv: Fix make check
Jason Ekstrand [Wed, 2 Mar 2016 19:31:22 +0000 (11:31 -0800)]
isl: Fix make check
Jason Ekstrand [Wed, 2 Mar 2016 18:46:13 +0000 (10:46 -0800)]
gen8/cmd_buffer: Properly return flushed push constant stages
This is required on SKL so that we can properly re-emit binding table
pointers commands.
Jason Ekstrand [Wed, 2 Mar 2016 01:19:43 +0000 (17:19 -0800)]
anv/meta_blit: Use unorm formats for 8 and 16-bit RGB and RGBA values
While Broadwell is very good about UINT formats, HSW is more restrictive.
Neither R8G8B8_UINT nor R16G16B16_UINT really exist on HSW. It should be
safe to just use the unorm formats.
Kenneth Graunke [Wed, 2 Mar 2016 00:26:09 +0000 (16:26 -0800)]
Merge remote-tracking branch 'origin/master' into vulkan
Jason Ekstrand [Tue, 1 Mar 2016 22:43:05 +0000 (14:43 -0800)]
genxml: Make the border color pointer consistent across gens
Jason Ekstrand [Tue, 1 Mar 2016 21:51:50 +0000 (13:51 -0800)]
gen7/pipeline: Add competent blending
This is mostly a copy-and-paste from gen8. Blending still isn't 100% but
it fixes about 1100 CTS blend tests on HSW.
Jason Ekstrand [Tue, 1 Mar 2016 19:38:11 +0000 (11:38 -0800)]
anv: Unify gen7 and gen8 state
Now that we've pulled surface state setup into ISL, there's not much to do
here.
Matt Turner [Sat, 27 Feb 2016 23:30:09 +0000 (15:30 -0800)]
mesa: Remove NV_fragment_program remnants from dlist.c.
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
Acked-by: Brian Paul <brianp@vmware.com>
Matt Turner [Sat, 27 Feb 2016 22:24:58 +0000 (14:24 -0800)]
mesa: Remove NV_fragment_program_option enable bit.
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
Acked-by: Brian Paul <brianp@vmware.com>
Matt Turner [Sat, 27 Feb 2016 22:05:17 +0000 (14:05 -0800)]
program: Remove NV_fragment_program opcode parsing.
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
Acked-by: Brian Paul <brianp@vmware.com>
Matt Turner [Mon, 29 Feb 2016 22:47:43 +0000 (14:47 -0800)]
program: Remove NV_fragment_program scalar suffix parsing.
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
Acked-by: Brian Paul <brianp@vmware.com>
Matt Turner [Sat, 27 Feb 2016 22:05:42 +0000 (14:05 -0800)]
program: Remove NV_fragment_program_option parsing support.
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
Acked-by: Brian Paul <brianp@vmware.com>
Matt Turner [Sat, 27 Feb 2016 22:04:30 +0000 (14:04 -0800)]
program: Remove NV_fragment_program Abs support.
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
Acked-by: Brian Paul <brianp@vmware.com>
Matt Turner [Sat, 27 Feb 2016 21:55:27 +0000 (13:55 -0800)]
program: Remove incorrect comment about OPCODE_TXD.
The table in prog_instruction.h is correct.
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
Acked-by: Brian Paul <brianp@vmware.com>
Matt Turner [Sat, 27 Feb 2016 21:53:50 +0000 (13:53 -0800)]
program: Remove OPCODE_TXP_NV.
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
Acked-by: Brian Paul <brianp@vmware.com>
Matt Turner [Sat, 27 Feb 2016 21:29:32 +0000 (13:29 -0800)]
program: Clean up after previous commit.
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
Acked-by: Brian Paul <brianp@vmware.com>
Matt Turner [Sat, 27 Feb 2016 19:06:25 +0000 (11:06 -0800)]
program: Remove condition-code and precision support.
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
Acked-by: Brian Paul <brianp@vmware.com>
Matt Turner [Sat, 27 Feb 2016 21:19:50 +0000 (13:19 -0800)]
program: Remove OPCODE_KIL_NV.
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
Acked-by: Brian Paul <brianp@vmware.com>
Matt Turner [Sat, 27 Feb 2016 21:44:50 +0000 (13:44 -0800)]
program: Remove RelAddr2 support.
Looks like more never-used crap from the first geometry shader attempt.
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
Acked-by: Brian Paul <brianp@vmware.com>
Matt Turner [Sat, 27 Feb 2016 22:15:26 +0000 (14:15 -0800)]
program: Mark table const.
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
Acked-by: Brian Paul <brianp@vmware.com>
Matt Turner [Sat, 27 Feb 2016 19:23:32 +0000 (11:23 -0800)]
mesa: Remove EmitCondCodes.
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
Acked-by: Brian Paul <brianp@vmware.com>
Matt Turner [Sat, 27 Feb 2016 19:19:41 +0000 (11:19 -0800)]
docs: Remove descriptions of long dead Emit* fields.
Dead since commit
d8a366200 in 2010.
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
Acked-by: Brian Paul <brianp@vmware.com>
Matt Turner [Sun, 28 Feb 2016 00:38:26 +0000 (16:38 -0800)]
glsl: Initialize gl_shader_program::EmptyUniformLocations.
Commit
65dfb30 added exec_list EmptyUniformLocations, but only
initialized the list if ARB_explicit_uniform_location was enabled,
leading to crashes if the extension was not available.
Cc: "11.2" <mesa-stable@lists.freedesktop.org>
Reviewed-by: Tapani Pälli <tapani.palli@intel.com>
Ian Romanick [Sat, 14 Nov 2015 01:57:18 +0000 (17:57 -0800)]
i965/meta: Don't pollute the framebuffer namespace
tl;dr: For many types of GL object, we can *NEVER* use the Gen function.
In OpenGL ES (all versions!) and OpenGL compatibility profile,
applications don't have to call Gen functions. The GL spec is very
clear about how you can mix-and-match generated names and non-generated
names: you can use any name you want for a particular object type until
you call the Gen function for that object type.
Here's the problem scenario:
- Application calls a meta function that generates a name. The first
Gen will probably return 1.
- Application decides to use the same name for an object of the same
type without calling Gen. Many demo programs use names 1, 2, 3,
etc. without calling Gen.
- Application calls the meta function again, and the meta function
replaces the data. The application's data is lost, and the app
fails. Have fun debugging that.
Signed-off-by: Ian Romanick <ian.d.romanick@intel.com>
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=92363
Reviewed-by: Topi Pohjolainen <topi.pohjolainen@intel.com>
Ian Romanick [Sat, 14 Nov 2015 01:35:30 +0000 (17:35 -0800)]
i965/meta: Use _mesa_bind_framebuffers instead of _mesa_BindFramebuffer
Signed-off-by: Ian Romanick <ian.d.romanick@intel.com>
Reviewed-by: Topi Pohjolainen <topi.pohjolainen@intel.com>
Ian Romanick [Sat, 14 Nov 2015 00:51:27 +0000 (16:51 -0800)]
meta: Don't pollute the framebuffer namespace
tl;dr: For many types of GL object, we can *NEVER* use the Gen function.
In OpenGL ES (all versions!) and OpenGL compatibility profile,
applications don't have to call Gen functions. The GL spec is very
clear about how you can mix-and-match generated names and non-generated
names: you can use any name you want for a particular object type until
you call the Gen function for that object type.
Here's the problem scenario:
- Application calls a meta function that generates a name. The first
Gen will probably return 1.
- Application decides to use the same name for an object of the same
type without calling Gen. Many demo programs use names 1, 2, 3,
etc. without calling Gen.
- Application calls the meta function again, and the meta function
replaces the data. The application's data is lost, and the app
fails. Have fun debugging that.
Fixes piglit tests:
- object-namespace-pollution glGetTexImage-compressed framebuffer
- object-namespace-pollution glGenerateMipmap framebuffer
Signed-off-by: Ian Romanick <ian.d.romanick@intel.com>
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=92363
Reviewed-by: Topi Pohjolainen <topi.pohjolainen@intel.com>
Ian Romanick [Sat, 14 Nov 2015 01:21:38 +0000 (17:21 -0800)]
meta/decompress: Track framebuffer using gl_framebuffer instead of GL API object handle
Signed-off-by: Ian Romanick <ian.d.romanick@intel.com>
Reviewed-by: Topi Pohjolainen <topi.pohjolainen@intel.com>
Ian Romanick [Sat, 14 Nov 2015 01:14:00 +0000 (17:14 -0800)]
meta/generate_mipmap: Track framebuffer using gl_framebuffer instead of GL API object handle
Signed-off-by: Ian Romanick <ian.d.romanick@intel.com>
Reviewed-by: Topi Pohjolainen <topi.pohjolainen@intel.com>
Ian Romanick [Fri, 13 Nov 2015 19:12:57 +0000 (11:12 -0800)]
meta: Use _mesa_bind_framebuffers instead of _mesa_BindFramebuffer
Signed-off-by: Ian Romanick <ian.d.romanick@intel.com>
Reviewed-by: Topi Pohjolainen <topi.pohjolainen@intel.com>
Ian Romanick [Fri, 13 Nov 2015 19:06:51 +0000 (11:06 -0800)]
meta: Use _mesa_CreateFramebuffers instead of _mesa_GenFramebuffers
This enables later patches that will stop calling _mesa_GenFramebuffers
or _mesa_CreateFramebuffers which pollute the framebuffer namespace.
For framebuffers, the Bind call is still necessary.
sed -i -e 's/_mesa_GenFramebuffers/_mesa_CreateFramebuffers/' \
src/mesa/drivers/common/*.c
Signed-off-by: Ian Romanick <ian.d.romanick@intel.com>
Reviewed-by: Topi Pohjolainen <topi.pohjolainen@intel.com>
Ian Romanick [Fri, 13 Nov 2015 19:04:25 +0000 (11:04 -0800)]
i965/meta: Use _mesa_CreateFramebuffers instead of _mesa_GenFramebuffers
This enables later patches that will stop calling _mesa_GenFramebuffers
or _mesa_CreateFramebuffers which pollute the framebuffer namespace.
For framebuffers, the Bind call is still necessary.
sed -i -e 's/_mesa_GenFramebuffers/_mesa_CreateFramebuffers/' \
src/mesa/drivers/dri/i965/*.c
Signed-off-by: Ian Romanick <ian.d.romanick@intel.com>
Reviewed-by: Topi Pohjolainen <topi.pohjolainen@intel.com>
Ian Romanick [Fri, 13 Nov 2015 18:51:01 +0000 (10:51 -0800)]
meta: Save and restore the framebuffer using gl_framebuffer instead of GL API object handle
Some meta operations can be called recursively. Future changes (the
"Don't pollute the ... namespace" changes) will cause objects with
invalid names to be used. If a nested meta operation tries to restore
an object named 0xDEADBEEF, it will fail.
This also fixes another latent bug in meta. In a multithreaded,
multicontext application, one thread can delete an object that is bound
in another thread. That object continues to exist until it is unbound
(i.e., its refcount drops to zero). Meta unbinds objects all over the
place. As a result, the rebind in _mesa_meta_end could fail because the
object vanished!
See https://bugs.freedesktop.org/show_bug.cgi?id=92363#c8.
Using _mesa_reference_<object type> to save and restore the objects
prevents the refcount from going to zero.
Signed-off-by: Ian Romanick <ian.d.romanick@intel.com>
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=92363
Reviewed-by: Topi Pohjolainen <topi.pohjolainen@intel.com>
Ian Romanick [Fri, 13 Nov 2015 18:46:40 +0000 (10:46 -0800)]
mesa: Refactor bind_framebuffer to make _mesa_bind_framebuffers
Fixing dd_function_table::BindFramebuffer will come later because that
change is probably not suitable for stable.
v2: Fix whitespace issue noticed by Topi.
Signed-off-by: Ian Romanick <ian.d.romanick@intel.com>
Reviewed-by: Topi Pohjolainen <topi.pohjolainen@intel.com>
Ian Romanick [Fri, 13 Nov 2015 01:18:53 +0000 (17:18 -0800)]
meta: Use _mesa_check_framebuffer_status instead of _mesa_CheckFramebufferStatus
sed -i -e 's/_mesa_CheckFramebufferStatus(GL_DRAW_FRAMEBUFFER/_mesa_check_framebuffer_status(ctx, ctx->DrawBuffer/' \
-e 's/_mesa_CheckFramebufferStatus(GL_FRAMEBUFFER[^)]*/_mesa_check_framebuffer_status(ctx, ctx->DrawBuffer/' \
-e 's/_mesa_CheckFramebufferStatus(GL_READ_FRAMEBUFFER/_mesa_check_framebuffer_status(ctx, ctx->ReadBuffer/' \
$(grep -rl _mesa_CheckFramebufferStatus src/mesa/drivers)
The second expression catches both GL_FRAMEBUFFER and GL_FRAMEBUFFER_EXT.
Signed-off-by: Ian Romanick <ian.d.romanick@intel.com>
Reviewed-by: Topi Pohjolainen <topi.pohjolainen@intel.com>
Ian Romanick [Fri, 13 Nov 2015 01:11:30 +0000 (17:11 -0800)]
meta: Obvious refactor of _mesa_meta_framebuffer_texture_image
Signed-off-by: Ian Romanick <ian.d.romanick@intel.com>
Reviewed-by: Topi Pohjolainen <topi.pohjolainen@intel.com>
Ian Romanick [Fri, 13 Nov 2015 01:05:27 +0000 (17:05 -0800)]
meta: Convert _mesa_meta_bind_fbo_image to take a gl_framebuffer instead of a GL API handle
Also change the name of the function to
_mesa_meta_framebuffer_texture_image. The function is basically a
wrapper around _mesa_framebuffer_texture (which is used to implement
glFramebufferTexture1D and friends), so it makes sense for it's name to
be similar to that.
The next patch will clean _mesa_meta_framebuffer_texture_image up
considerably.
Signed-off-by: Ian Romanick <ian.d.romanick@intel.com>
Reviewed-by: Topi Pohjolainen <topi.pohjolainen@intel.com>
Jason Ekstrand [Tue, 1 Mar 2016 18:59:21 +0000 (10:59 -0800)]
anv/cmd_buffer: Look at both sides for stencil enable
Now it's all consistent with gen9
Jason Ekstrand [Tue, 1 Mar 2016 19:02:12 +0000 (11:02 -0800)]
anv/cmd_buffer: Clean up stencil state setup on gen7
Jason Ekstrand [Tue, 1 Mar 2016 18:56:46 +0000 (10:56 -0800)]
anv/cmd_buffer: Clean up stencil state setup on gen8
Kristian Høgsberg Kristensen [Tue, 1 Mar 2016 17:17:16 +0000 (09:17 -0800)]
anv: Add in image->offset when setting up depth buffer
Fix from Neil Roberts.
https://bugs.freedesktop.org/show_bug.cgi?id=94348
Jason Ekstrand [Tue, 1 Mar 2016 01:27:11 +0000 (17:27 -0800)]
anv/pipeline: Pull 3DSTATE_SBE into a shared helper
Dave Airlie [Tue, 1 Mar 2016 05:48:44 +0000 (15:48 +1000)]
virgl: add support for passing render condition flags to host.
This just passes the extra blit info to fix the render condition
tests.
Cc: "11.2" <mesa-stable@lists.freedesktop.org>
Signed-off-by: Dave Airlie <airlied@redhat.com>
Jason Ekstrand [Tue, 1 Mar 2016 00:47:39 +0000 (16:47 -0800)]
genxml: Break output detail of 3DSTATE_SBE on gen7 into a struct
This makes it work like 3DSTATE_SBE_SWIZ on gen8+ which is much more
convenient.
Kenneth Graunke [Wed, 17 Feb 2016 08:37:04 +0000 (00:37 -0800)]
i965: Push most TES inputs in vec4 mode.
(This is commit
4a1c8a3037cd29938b2a6e2c680c341e9903cfbe for vec4 mode.)
Using the push model for inputs is much more efficient than pulling
inputs - the hardware can simply copy a large chunk into URB registers
at thread creation time, rather than having the thread send messages to
request data from the L3 cache. Unfortunately, it's possible to have
more TES inputs than fit in registers, so we have to fall back to the
pull model in some cases.
However, it turns out that most tessellation evaluation shaders are
fairly simple, and don't use many inputs. An arbitrary cut-off of
24 vec4 slots (12 registers) should suffice. (I chose this instead of
the 32 vec4 slots used in the scalar backend to avoid regressing a few
Piglit tests due to the vec4 register allocator being too stupid to
figure out what to do. We probably ought to fix that, but it's a
separate issue.)
Improves performance in GPUTest's tessmark_x64 microbenchmark by
41.5394% +/- 0.288519% (n = 115) at 1024x768 on my Clevo W740SU
(with Iris Pro 5200).
Improves performance in Synmark's Gl40TerrainFlyTess microbenchmark by
38.3576% +/- 0.759748% (n = 42).
v2: Simplify abs/negate handling, as requested by Matt.
Signed-off-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Chris Forbes <chrisf@ijw.co.nz>
Reviewed-by: Matt Turner <mattst88@gmail.com>
Marek Olšák [Tue, 23 Feb 2016 22:32:44 +0000 (23:32 +0100)]
r600g: remove support for DRM < 2.12.0
Marek Olšák [Tue, 23 Feb 2016 22:32:44 +0000 (23:32 +0100)]
r300g: remove support for DRM < 2.12.0