Brian Paul [Wed, 22 Jul 2015 00:35:38 +0000 (18:35 -0600)]
mesa: add API dispatch for GL_ARB_get_texture_sub_image
This adds the new glGetTextureSubImage() and
glGetCompressedTextureSubImage() functions. Also update the
dispatch sanity test program.
v2: remove stray brace, move xi:include line in gl_API.xml, fix extension
number typo, s/program/texture/ in xml file.
Reviewed-by: Ilia Mirkin <imirkin@alum.mit.edu>
Brian Paul [Wed, 22 Jul 2015 00:35:38 +0000 (18:35 -0600)]
mesa: add new _mesa_Get[Compressed]TextureSubImage() functions
Simple implementations in terms of get_[compressed_]texture_image().
Reviewed-by: Ilia Mirkin <imirkin@alum.mit.edu>
Brian Paul [Wed, 22 Jul 2015 00:35:38 +0000 (18:35 -0600)]
mesa: overhaul the glGetCompressedTexImage code
Same idea as the previous patch.
v2: a few clean-ups spotted by Ilia
Reviewed-by: Ilia Mirkin <imirkin@alum.mit.edu>
Brian Paul [Wed, 22 Jul 2015 00:35:38 +0000 (18:35 -0600)]
mesa: overhaul the glGetTexImage code
1. Reorganize the error checking code.
2. Lay groundwork for getting sub images by passing image offset and
dimensions to the error checking code.
3. Implement _mesa_GetnTexImageARB(), _mesa_GetTexImage() and
_mesa_GetTextureImage() all in terms of get_texture_image().
v2: pass offset/width/height/depth arguments to the error checking
function, avoid using magic width/height/depth values.
v3: remove unused bufSize param to get_texture_image()
Reviewed-by: Ilia Mirkin <imirkin@alum.mit.edu>
Brian Paul [Wed, 22 Jul 2015 00:35:38 +0000 (18:35 -0600)]
mesa: 80-column wrapping in texgetimage.c
Reviewed-by: Ilia Mirkin <imirkin@alum.mit.edu>
Brian Paul [Wed, 22 Jul 2015 00:35:38 +0000 (18:35 -0600)]
mesa: replace Driver.GetCompressedTexImage() w/ GetCompressedTexSubImage()
For now, pass offsets of zero and width/height/depth equal to the
whole image.
Reviewed-by: Ilia Mirkin <imirkin@alum.mit.edu>
Brian Paul [Wed, 22 Jul 2015 00:35:38 +0000 (18:35 -0600)]
mesa: make _mesa_get_[compressed_]texture_image() static
These functions are only called from teximage.c
Reviewed-by: Ilia Mirkin <imirkin@alum.mit.edu>
Brian Paul [Wed, 22 Jul 2015 00:35:38 +0000 (18:35 -0600)]
mesa: plumb offset/size parameters through GetTexSubImage code
Needed for GL_ARB_get_texture_sub_image. But at this point, the
offsets are always zero and the sizes match the whole texture image.
v2: Fixes, suggestions from Laura Ekstrand:
* Fix calls to ctx->Driver.UnmapTextureImage() to pass the correct
slice value.
* Added comments and assertions to check zoffset+depth<=tex->Depth before
the 'img' loops.
* Added a new zoffset==0 assert in get_tex_memcpy().
Reviewed-by: Ilia Mirkin <imirkin@alum.mit.edu>
Brian Paul [Wed, 22 Jul 2015 00:35:38 +0000 (18:35 -0600)]
mesa: replace Driver.GetTexImage with GetTexSubImage()
The new driver hook has x/y/zoffset and width/height/depth parameters
for the new glGetTextureSubImage() function.
The meta code and gallium state tracker are updated to handle the
new parameters.
Callers to Driver.GetTexSubImage() pass in offsets=0 and sizes equal
to the whole texture size.
v2: update i965 driver code, s/GLint/GLsizei/ in GetTexSubImage hook
Reviewed-by: Ilia Mirkin <imirkin@alum.mit.edu>
Brian Paul [Wed, 22 Jul 2015 00:35:38 +0000 (18:35 -0600)]
meta: add offset, width, height parameters to decompress_texture_image()
In preparation for decompressing texture sub images.
Reviewed-by: Ilia Mirkin <imirkin@alum.mit.edu>
Brian Paul [Wed, 22 Jul 2015 00:35:38 +0000 (18:35 -0600)]
meta: handle subimages in _mesa_meta_setup_texture_coords()
v2: fix depth, total_depth mix-up in meta.h, per Laura Ekstrand.
Reviewed-by: Anuj Phogat <anuj.phogat@gmail.com>
Brian Paul [Wed, 22 Jul 2015 00:35:38 +0000 (18:35 -0600)]
mesa: assorted whitespace, formatting fixes in teximage.c
Trivial.
Brian Paul [Wed, 22 Jul 2015 00:35:38 +0000 (18:35 -0600)]
mesa: allow GL_TEXTURE_CUBE_MAP_ARRAY case for glCompressedTexSubImage3D()
Since s3tc works for cube maps and 2D arrays, it should also work for
cube arrays. NVIDIA's driver supports this too. Seems like the spec
should say this.
This is a minor follow-on fix for the commit "mesa: fix up some texture
error checks".
Reviewed-by: Roland Scheidegger <sroland@vmware.com>
Jonathan Gray [Wed, 15 Jul 2015 15:17:37 +0000 (01:17 +1000)]
mesa: include stdarg.h for va_list
Include stdarg.h for va_list. Unbreaks the build on OpenBSD:
In file included from mesa/program/dummy_errors.c:24:
../src/mesa/main/errors.h:85: error: expected declaration specifiers or '...' before 'va_list'
Signed-off-by: Jonathan Gray <jsg@jsg.id.au>
Acked-by: Matt Turner <mattst88@gmail.com>
Reviewed-by: Ilia Mirkin <imirkin@alum.mit.edu>
Ilia Mirkin [Mon, 20 Jul 2015 23:58:43 +0000 (19:58 -0400)]
gallium: replace INLINE with inline
Generated by running:
git grep -l INLINE src/gallium/ | xargs sed -i 's/\bINLINE\b/inline/g'
git grep -l INLINE src/mesa/state_tracker/ | xargs sed -i 's/\bINLINE\b/inline/g'
git checkout src/gallium/state_trackers/clover/Doxyfile
and manual edits to
src/gallium/include/pipe/p_compiler.h
src/gallium/README.portability
to remove mentions of the inline define.
Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu>
Acked-by: Marek Olšák <marek.olsak@amd.com>
Samuel Pitoiset [Tue, 21 Jul 2015 19:58:08 +0000 (21:58 +0200)]
nvc0: force cache flush when binding a new ubo
This fixes the following piglit test:
ext_transform_feedback-immediate-reuse-uniform-buffer
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Ilia Mirkin <imirkin@alum.mit.edu>
Samuel Pitoiset [Tue, 21 Jul 2015 19:35:43 +0000 (21:35 +0200)]
nv50: force cache flush when binding a new ubo
This fixes the following piglit test:
ext_transform_feedback-immediate-reuse-uniform-buffer
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Ilia Mirkin <imirkin@alum.mit.edu>
Vinson Lee [Tue, 21 Jul 2015 04:52:40 +0000 (21:52 -0700)]
st/mesa: Silence GCC unused-variable warning.
Silence a release build warning.
st_glsl_to_tgsi.cpp: In function 'pipe_error st_translate_program(gl_context*, uint, ureg_program*, glsl_to_tgsi_visitor*, const gl_program*, GLuint, const GLuint*, const GLuint*, const ubyte*, const ubyte*, const GLuint*, const GLuint*, GLuint, const GLuint*, const GLuint*, const ubyte*, const ubyte*, boolean, boolean)':
st_glsl_to_tgsi.cpp:5461:36: warning: unused variable 'pscreen' [-Wunused-variable]
struct pipe_screen *pscreen = st->pipe->screen;
^
Signed-off-by: Vinson Lee <vlee@freedesktop.org>
Reviewed-by: Brian Paul <brianp@vmware.com>
Adam Jackson [Tue, 21 Jul 2015 16:08:20 +0000 (12:08 -0400)]
r600/sb: Fix an &/&& mistake
gcc says:
sb/sb_sched.cpp: In member function 'bool r600_sb::alu_group_tracker::try_reserve(r600_sb::alu_node*)':
sb/sb_sched.cpp:492:7: warning: suggest parentheses around operand of '!' or change '&' to '&&' or '!' to '~' [-Wparentheses]
if (!trans & fbs)
It happens to be harmless; if fbs is ever non-zero, it will be VEC_210,
which is 5, so (!trans & 5) == 1 and the branch works as expected. But
logical AND is clearly what was meant.
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Adam Jackson <ajax@redhat.com>
Anuj Phogat [Fri, 17 Jul 2015 21:14:29 +0000 (14:14 -0700)]
Revert "i965/gen9: Plugin the code for selecting YF/YS tiling on skl+"
Commit
c9dbdc0 introduced some dead code which is supposed to be used
once we have Yf/Ys tiling working and performing better. Ken reported
the issue that static analysis tool now shows warnings due to the dead
code. To fix these warnings, this patch reverts the changes made in
commit
c9dbdc0.
It'll be better to add the Yf/Ys tiling selection code later, when we
are ready to use it.
Signed-off-by: Anuj Phogat <anuj.phogat@gmail.com>
Acked-by: Kenneth Graunke <kenneth@whitecape.org>
Francisco Jerez [Mon, 13 Jul 2015 12:52:28 +0000 (15:52 +0300)]
i965: Fix stride field for the result of emit_uniformize().
This is essentially the same problem fixed in an earlier patch for
immediates. Setting the stride to zero will be particularly useful
for my future SIMD lowering pass, because we will be able to just
check whether the stride of a source register is zero and skip
emitting the copies required to unzip it in that case.
Instead of setting stride to zero in every caller of emit_uniformize()
I've changed the function to return the result as its return value
(previously it was being written into a caller-provided destination
register), because this way we can enforce that the result is used with
the correct regioning from the function itself.
The changes to the prototype of its VEC4 counterpart are mainly for
the sake of symmetry, VEC4 registers don't have stride.
Reviewed-by: Samuel Iglesias Gonsálvez <siglesias@igalia.com>
Francisco Jerez [Mon, 13 Jul 2015 12:29:39 +0000 (15:29 +0300)]
i965/fs: Fix stride field for uniforms.
This fixes essentially the same problem as for immediates. Registers
of the UNIFORM file are typically accessed according to the formula:
read_uniform(r, channel_index, array_index) =
read_element(r, channel_index * 0 + array_index * 1)
Which matches the general direct addressing formula for stride=0:
read_direct(r, channel_index, array_index) =
read_element(r, channel_index * stride +
array_index * max{1, stride * width})
In either case if reladdr is present the access will be according to
the composition of two register regions, the first one determining the
per-channel array_index used for the second, like:
read_indirect(r, channel_index, array_index) =
read_direct(r, channel_index,
read(r.reladdr, channel_index, array_index))
where:
read(r, channel_index, array_index) = if r.reladdr == NULL
then read_direct(r, channel_index, array_index)
else read_indirect(r, channel_index, array_index)
In conclusion we can handle uniforms consistently with the other
register files if we set stride to zero. After lowering to a GRF
using VARYING_PULL_CONSTANT_LOAD in demote_pull_constant_loads() the
stride of the source is set to one again because the result of
VARYING_PULL_CONSTANT_LOAD is generally non-uniform.
Reviewed-by: Samuel Iglesias Gonsálvez <siglesias@igalia.com>
Francisco Jerez [Mon, 13 Jul 2015 11:50:24 +0000 (14:50 +0300)]
i965/fs: Fix stride for immediate registers.
When the width field was removed from fs_reg the BROADCAST handling
code in opt_algebraic() started to miss a number of trivial
optimization cases resulting in the ugly indirect-addressing sequence
to be emitted unnecessarily for some variable-indexed texturing and
UBO loads regardless of one of the sources of BROADCAST being
immediate. Apparently the reason was that we were setting the stride
field to one for immediates even though they are typically uniform.
Width used to be set to one too which is why this optimization used to
work previously until the "reg.width == 1" check was removed.
The stride field of vector immediates is intentionally left equal to
one, because they are strictly speaking not uniform. The assertion in
fs_generator makes sure that immediates have the expected stride as
consistency check.
Reviewed-by: Samuel Iglesias Gonsálvez <siglesias@igalia.com>
Iago Toral Quiroga [Mon, 20 Jul 2015 10:58:12 +0000 (12:58 +0200)]
i965/vec4: Fix liveness analysis with BRW_OPCODE_SEL
We only consider a vgrf defined by a given block if the block writes to it
unconditionally. So far we have been checking this by testing that the
instruction is not predicated, however, in the case of BRW_OPCODE_SEL,
the predication is used to select the value to write, not to decide if
the write is actually done. The consequence of this was increased life
spans for affected vgrfs, which could lead to additional register pressure.
Since NIR generates selects for conditional writes this was causing massive
register pressure in a handful of piglit and dEQP tests that had a large
number of select operations with the NIR-vec4 backend.
Fixes the following piglit tests with the NIR-vec4 backend:
spec/glsl-1.50/execution/variable-indexing/vs-output-array-vec4-index-wr-before-gs
spec/glsl-1.50/execution/variable-indexing/gs-input-array-vec4-index-rd
spec/glsl-1.50/execution/variable-indexing/vs-output-array-vec2-index-wr-before-gs
spec/glsl-1.50/execution/variable-indexing/vs-output-array-vec3-index-wr-before-gs
spec/glsl-1.50/execution/variable-indexing/vs-output-array-float-index-wr-before-gs
Fixes 80 dEQP tests with the NIR-vec4 backend in the following category:
dEQP-GLES3.functional.ubo.*
Reviewed-by: Francisco Jerez <currojerez@riseup.net>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Kenneth Graunke [Sat, 18 Jul 2015 08:22:00 +0000 (01:22 -0700)]
mesa: Rename _mesa_lookup_enum_by_nr() to _mesa_enum_to_string().
Generated by sed; no manual changes.
Signed-off-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Brian Paul <brianp@vmware.com>
Samuel Pitoiset [Mon, 20 Jul 2015 19:32:43 +0000 (21:32 +0200)]
nouveau: use bool instead of boolean
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Acked-by: Ilia Mirkin <imirkin@alum.mit.edu>
Tom Stellard [Mon, 20 Jul 2015 15:12:56 +0000 (15:12 +0000)]
gallivm: Initialize LLVM Modules's DataLayout to an empty string.
This fixes crashes in llvmpipe with LLVM 3.8 and also some piglit tests
on radeonsi that use the draw module.
This is just a temporary solution. The correct solution will require
creating a TargetMachine during gallivm initialization and pulling the
DataLayout from there. This will be a somewhat invasive change, and it
will need to be validatated on multiple LLVM versions.
https://llvm.org/bugs/show_bug.cgi?id=24172
Reviewed-by: Roland Scheidegger <sroland@vmware.com>
Samuel Pitoiset [Thu, 16 Jul 2015 21:05:05 +0000 (23:05 +0200)]
nvc0: add a missing parameter to nvc0_set_shader_images()
This fixes a compilation warning introduced in commit
05a12c5
(gallium: add interface for writable shader images).
While we are at it, fix indentation and rename parameters according to
the gallium interface.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Ilia Mirkin <imirkin@alum.mit.edu>
Samuel Pitoiset [Mon, 20 Jul 2015 16:47:17 +0000 (18:47 +0200)]
nouveau: always align buffers to 0x100
Only constbufs must be aligned to 0x100, but since all buffers can be
rebinded as constant buffers they must be also aligned.
This patch prevents this behaviour by aligning everything to 256-byte
increments at buffer creation.
This fixes dmesg fails for the following piglit test:
ext_transform_feedback-immediate-reuse-uniform-buffer -auto -fbo
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Ilia Mirkin <imirkin@alum.mit.edu>
Samuel Pitoiset [Mon, 13 Jul 2015 11:34:31 +0000 (13:34 +0200)]
nv50: limit the maximum number of samplers to 16
NV50_3D_BIND_TSC only allows to bind 16 samplers, and since we don't
want to do anything with NV50_3D_BIND_TSC2, just limit the maximum
number of samplers to 16 like for nvc0.
This fixes dmesg fails with the following piglit test:
max-samplers
But the test still fails.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Ilia Mirkin <imirkin@alum.mit.edu>
Samuel Pitoiset [Mon, 13 Jul 2015 10:52:57 +0000 (12:52 +0200)]
nv50: turn samples counts off during blit
Fixes the following piglit test:
occlusion_query_meta_no_fragments
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Ilia Mirkin <imirkin@alum.mit.edu>
Samuel Pitoiset [Mon, 13 Jul 2015 10:51:21 +0000 (12:51 +0200)]
nv50: add nesting support for occlusion queries
This is loosely based on nvc0.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Ilia Mirkin <imirkin@alum.mit.edu>
Alejandro Piñeiro [Fri, 17 Jul 2015 09:54:34 +0000 (11:54 +0200)]
i965/nir/fs: removed unneeded support for global variables
As functions are inlined, and nir_lower_global_vars_to_local gets
run, all global variables are lowered to local variables.
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Ilia Mirkin [Mon, 20 Jul 2015 04:19:56 +0000 (00:19 -0400)]
nv50: fix max level clamping on G80
It appears that the G80 did not have support for the sampler view
first/last clamping. Put the view's last level in the place of the
texture's so that it doesn't go past what the sampler view allows.
Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu>
Cc: mesa-stable@lists.freedesktop.org
Ilia Mirkin [Sat, 18 Jul 2015 23:02:29 +0000 (19:02 -0400)]
gm107/ir: fix indirect txq emission
Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu>
Cc: mesa-stable@lists.freedesktop.org
Ilia Mirkin [Sat, 18 Jul 2015 22:38:42 +0000 (18:38 -0400)]
nvc0/ir: don't worry about sampler in txq handling
There's no need to deal with samplers for texture size queries. That
code also was accidentally setting an invalid sIndirectSrc position, but
it can now just be removed.
Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu>
Cc: mesa-stable@lists.freedesktop.org
Ilia Mirkin [Sat, 18 Jul 2015 20:43:17 +0000 (16:43 -0400)]
nvc0/ir: fix txq on indirect samplers
Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu>
Cc: mesa-stable@lists.freedesktop.org
Abdiel Janulgue [Wed, 20 May 2015 15:02:44 +0000 (18:02 +0300)]
i965: Disable resource streamer in BLORP
Switch off hardware-generated binding tables and gather push
constants in the blorp. Blorp requires only a minimal set of
simple constants. There is no need for the extra complexity
to program a gather table entry into the pipeline.
Cc: kenneth@whitecape.org
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Signed-off-by: Abdiel Janulgue <abdiel.janulgue@linux.intel.com>
Abdiel Janulgue [Wed, 2 Oct 2013 13:37:20 +0000 (16:37 +0300)]
i965: Upload binding tables in hw-generated binding table format.
When hardware-generated binding tables are enabled, use the hw-generated
binding table format when uploading binding table state.
Normally, the CS will will just consume the binding table pointer commands
as pipelined state. When the RS is enabled however, the RS flushes whatever
edited surface state entries of our on-chip binding table to the binding
table pool before passing the command on to the CS.
Note that the the binding table pointer offset is relative to the binding table
pool base address when resource streamer instead of the surface state base address.
v2: Fix possible buffer overflow when allocating a chunk out of the
hw-binding table pool (Ken).
v3: Remove extra newline and add missing brace around if-statement (Matt).
v4: Fix broken INTEL_DEBUG=shader_time for hw-generated binding tables.
Document PRM WaStateBindingTableOverfetch workaround.
Cc: kenneth@whitecape.org
Cc: mattst88@gmail.com
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Signed-off-by: Abdiel Janulgue <abdiel.janulgue@linux.intel.com>
Abdiel Janulgue [Fri, 17 Jul 2015 09:20:18 +0000 (12:20 +0300)]
i965: Implement interface to edit binding table entries
Unlike normal software binding tables where the driver has to manually
generate and fill a binding table array which are then uploaded to the
hardware, the resource streamer instead presents the driver with an option
to fill out slots for individual binding table indices. The hardware
accumulates the state for these combined edits which it then automatically
flushes to a binding table pool when the binding table pointer state
command is invoked.
v2: Clarify binding table edit bit aligment (Topi).
v3: Make comments and function names more clearer (Ken).
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Signed-off-by: Abdiel Janulgue <abdiel.janulgue@linux.intel.com>
Abdiel Janulgue [Wed, 15 Apr 2015 10:04:45 +0000 (13:04 +0300)]
i965: Enable hardware-generated binding tables on render path.
This patch implements the binding table enable command which is also
used to allocate a binding table pool where where hardware-generated
binding table entries are flushed into. Each binding table offset in
the binding table pool is unique per each shader stage that are
enabled within a batch.
Also insert the required brw_tracked_state objects to enable
hw-generated binding tables in normal render path.
v2: - Use MOCS in binding table pool alloc for GEN8
- Fix spurious offset when allocating binding table pool entry
and start from zero instead.
v3: - Include GEN8 fix for spurious offset above.
v4: - Fixup wrong packet length in enable/disable hw-binding table
for GEN8 (Ville).
- Don't invoke HW-binding table disable command when we dont
have resource streamer (Chris).
v5: - Reorder the state cache invalidate flush so it happens in-between
enabling hw-generated binding tables and the previous sw-binding
table GPU state (Chris).
v6: - Do the same fix in v5 for gen7_disable_hw_binding_tables().
- Adhere to coding guidelines and make comments more informative.
Cc: kenneth@whitecape.org
Cc: syrjala@sci.fi
Cc: chris@chris-wilson.co.uk
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Signed-off-by: Abdiel Janulgue <abdiel.janulgue@linux.intel.com>
Abdiel Janulgue [Tue, 2 Jul 2013 15:48:22 +0000 (11:48 -0400)]
i965: Enable resource streamer for the batchbuffer
Check first if the hardware and kernel supports resource streamer. If this
is allowed, tell the kernel to enable the resource streamer enable bit on
MI_BATCHBUFFER_START by specifying I915_EXEC_RESOURCE_STREAMER
execbuffer flags.
v2: - Use new I915_PARAM_HAS_RESOURCE_STREAMER ioctl to check if kernel
supports RS (Ken).
- Add brw_device_info::has_resource_streamer and toggle it for
Haswell, Broadwell, Cherryview, Skylake, and Broxton (Ken).
v3: - Update I915_PARAM_HAS_RESOURCE_STREAMER to match updated kernel.
v4: - Always inspect the getparam.value (Chris Wilson).
v5: - Fold redundant devinfo->has_resource_streamer check in context create
into init screen.
Cc: kenneth@whitecape.org
Cc: chris@chris-wilson.co.uk
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Signed-off-by: Abdiel Janulgue <abdiel.janulgue@linux.intel.com>
Abdiel Janulgue [Mon, 18 May 2015 11:32:17 +0000 (14:32 +0300)]
i965: Define HW-binding table and resource streamer control opcodes
v2: Use macros for HW binding table edits (Topi)
v3: Add Broadwell support.
v4: Make hardware binding table bit definitions even more clearer (Ken)
Cc: kenneth@whitecape.org
Reviewed-by: Topi Pohjolainen <topi.pohjolainen@intel.com>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Signed-off-by: Abdiel Janulgue <abdiel.janulgue@linux.intel.com>
Eric Anholt [Fri, 17 Jul 2015 17:01:48 +0000 (10:01 -0700)]
vc4: Switch to using a separate ioctl for making shaders.
This gives the kernel a chance to validate and lock down the data,
without having to deal with mmap zapping.
With this, GLBenchmark stops on a texture relocations, because we'd
recycled a shader BO as another shader and failed to revalidate, since we
weren't clearing the cached validation state on mmap faults.
Roland Scheidegger [Fri, 17 Jul 2015 16:01:01 +0000 (18:01 +0200)]
mesa: fix up some texture error checks
In particular, we were incorrectly accepting s3tc (and lots of others)
for CompressedTexSubImage3D (but not CompressedTexImage3D) calls with 3d
targets. At this time, the only allowed formats for these calls are the
bptc ones, since none of the specific extensions allow it (astc hdr would).
Also, fix up a bug in _mesa_target_can_be_compressed - 3d target needs to
be allowed for bptc formats.
Reviewed-by: Brian Paul <brianp@vmware.com>
Eric Anholt [Fri, 17 Jul 2015 18:52:09 +0000 (11:52 -0700)]
vc4: Fix printing of shader-db debug when shader-db isn't turned on.
Eric Anholt [Fri, 17 Jul 2015 18:22:40 +0000 (11:22 -0700)]
vc4: Add debugging on texture relocation validation failures.
Eric Anholt [Thu, 16 Jul 2015 21:30:28 +0000 (14:30 -0700)]
vc4: Also consider uniform 0 in uniform lowering.
The hash table considers key 0 to be the empty key.
Eric Anholt [Fri, 10 Jul 2015 23:30:27 +0000 (16:30 -0700)]
vc4: Use the pure/const attributes on a bunch of our QPU functions.
On a release build, this makes the rest of vc4_qpu_validate.c go away
(the compiler didn't know that our qpu helper function calls had no
side effects).
Eric Anholt [Fri, 10 Jul 2015 23:25:26 +0000 (16:25 -0700)]
mesa: Detect and provide macros for function attributes pure and const.
These are really useful hints to the compiler in the absence of link-time
optimization, and I'm going to use them in VC4.
I've made the const attribute be ATTRIBUTE_CONST unlike other function
attributes, because we have other things in the tree #defining CONST for
their own unrelated purposes.
v2: Alphabetize.
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org> (v1)
Connor Abbott [Tue, 30 Jun 2015 20:42:15 +0000 (13:42 -0700)]
i965/fs: don't make unused payload registers interfere
Before, we were setting payload_last_use_ip for unused payload
registers to 0, which made them interfere with whatever the first
instruction wrote to due to the workaround for SIMD16 uniform arguments.
Just use -1 to mean "unused" instead, and then skip setting any
interferences for unused payload registers.
instructions in affected programs: 0 -> 0
helped: 0
HURT: 0
GAINED: 1
LOST: 0
Reviewed-by: Jordan Justen <jordan.l.justen@intel.com>
Signed-off-by: Connor Abbott <connor.w.abbott@intel.com>
Connor Abbott [Tue, 30 Jun 2015 20:38:20 +0000 (13:38 -0700)]
i965/fs: remove special case in setup_payload_interference()
regs_read() will handle LINTERP for us since the previous commit. In
addition, we were being too conservative, since it will only read 2
registers on SIMD8.
instructions in affected programs: 9061 -> 8893 (-1.85%)
helped: 10
HURT: 0
GAINED: 0
LOST: 0
All of the changes were due to spills being eliminated, mostly in KSP
shaders.
Reviewed-by: Jordan Justen <jordan.l.justen@intel.com>
Signed-off-by: Connor Abbott <connor.w.abbott@intel.com>
Jordan Justen [Mon, 2 Feb 2015 22:23:35 +0000 (14:23 -0800)]
i965/fs: Mark last used ip for all regs read in the payload
If a source register in the push constant registers uses more than one
register, then we wouldn't update payload_last_use_ip for subsequent
registers.
Unlike most uniform data pushed into registers, the CS gl_LocalInvocationID
data varies per execution channel. Therefore for SIMD16 mode, we have vec16
data in the payload. In this case we then need to mark 2 registers in
payload_last_use_ip as last used by the instruction. There's a similar
situation for the z and w coordinates of gl_FragCoord for fragment shaders,
where it had only happened to work before because of some bogus interferences
which the next commit removes.
(Connor: added bit about gl_FragCoord to commit message)
Signed-off-by: Jordan Justen <jordan.l.justen@intel.com>
Reviewed-by: Connor Abbott <connor.w.abbott@intel.com>
Connor Abbott [Wed, 1 Jul 2015 16:58:47 +0000 (09:58 -0700)]
i965/fs: fix regs_read() for LINTERP
The second source always stays within the same SIMD8 register.
Reviewed-by: Jordan Justen <jordan.l.justen@intel.com>
Reviewed-by: Matt Turner <mattst88@gmail.com>
Signed-off-by: Connor Abbott <connor.w.abbott@intel.com>
Connor Abbott [Wed, 15 Jul 2015 19:01:20 +0000 (12:01 -0700)]
nir: add nir_foreach_instr_safe_reverse()
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Jason Ekstrand <jason.ekstrand@intel.com>
Signed-off-by: Connor Abbott <connor.w.abbott@intel.com>
Connor Abbott [Wed, 15 Jul 2015 19:00:47 +0000 (12:00 -0700)]
nir: add nir_instr_is_first() and nir_instr_is_last() helpers
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Jason Ekstrand <jason.ekstrand@intel.com>
Signed-off-by: Connor Abbott <connor.w.abbott@intel.com>
Jordan Justen [Thu, 16 Jul 2015 22:07:05 +0000 (15:07 -0700)]
i965/cs: Use dispatch width of 8 for cs terminate payload setup
This prevents an assertion failure in brw_fs_live_variables.cpp,
fs_live_variables::setup_one_write: Assertion `var < num_vars' failed.
Signed-off-by: Jordan Justen <jordan.l.justen@intel.com>
Reviewed-by: Jason Ekstrand <jason.ekstrand@intel.com>
Jordan Justen [Thu, 16 Jul 2015 22:04:43 +0000 (15:04 -0700)]
i965/cs: Return 1 for regs_read on CS_OPCODE_CS_TERMINATE
This prevents an assertion failure in brw_fs_live_variables.cpp,
fs_live_variables::setup_one_read: Assertion `var < num_vars' failed.
Signed-off-by: Jordan Justen <jordan.l.justen@intel.com>
Reviewed-by: Jason Ekstrand <jason.ekstrand@intel.com>
Kenneth Graunke [Sun, 5 Jul 2015 02:15:16 +0000 (19:15 -0700)]
program: Allow redundant OPTION ARB_fog_* directives.
A fragment program from "Pixel Piracy" contains redundant OPTION
directives:
!!ARBfp1.0
OPTION ARB_precision_hint_fastest;
OPTION ARB_fog_exp2;
OPTION ARB_precision_hint_fastest;
OPTION ARB_fog_exp2;
...
We already allow redundant ARB_precision_hint_fastest directives, but
disallow the redundant (yet consistent) ARB_fog_exp2 directives, failing
to compile the program.
The specification seems to contradict itself - the main text says that
only one fog application option may be specified, but then backpedals,
indicating the intent is to disallow /contradictory/ flags. One of the
issues suggests that specifying contradictory ones is stupid, but
allowed, and only the last one should take effect.
Accepting multiple redundant (but consistent) directives seems harmless,
and like a reasonable interpretation of the specification. It also
fixes a fragment program found in the wild.
Cc: mesa-stable@lists.freedesktop.org
Signed-off-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
Ben Widawsky [Tue, 14 Jul 2015 16:56:09 +0000 (09:56 -0700)]
i965: Push miptree tiling request into flags
With the last few patches a way was provided to influence lower layer miptree
layout and allocation decisions via flags (replacing bools). For simplicity, I
chose not to touch the tiling requests because the change was slightly less
mechanical than replacing the bools.
The goal is to organize the code so we can continue to add new parameters and
tiling types while minimizing risk to the existing code, and not having to
constantly add new function parameters.
v2: Rebased on Anuj's recent Yf/Ys changes
Fix non-msrt MCS allocation (was only happening in gen8 case before)
v3: small fix in assertion requested by Chad
v4: Use parens to get the order right from v3.
Signed-off-by: Ben Widawsky <ben@bwidawsk.net>
Reviewed-by: Jordan Justen <jordan.l.justen@intel.com>
Ben Widawsky [Thu, 16 Jul 2015 23:52:08 +0000 (16:52 -0700)]
Revert "i965: Push miptree tiling request into flags"
This reverts commit
51e8d549e110f86cb7107cf712843aebd956fb9a.
Ben Widawsky [Tue, 14 Jul 2015 16:56:09 +0000 (09:56 -0700)]
i965: Push miptree tiling request into flags
With the last few patches a way was provided to influence lower layer miptree
layout and allocation decisions via flags (replacing bools). For simplicity, I
chose not to touch the tiling requests because the change was slightly less
mechanical than replacing the bools.
The goal is to organize the code so we can continue to add new parameters and
tiling types while minimizing risk to the existing code, and not having to
constantly add new function parameters.
v2: Rebased on Anuj's recent Yf/Ys changes
Fix non-msrt MCS allocation (was only happening in gen8 case before)
v3: small fix in assertion requested by Chad
Signed-off-by: Ben Widawsky <ben@bwidawsk.net>
Reviewed-by: Jordan Justen <jordan.l.justen@intel.com> (v2)
Reviewed-by: Anuj Phogat <anuj.phogat@gmail.com> (v2)
Reviewed-by: Chad Versace <chad.versace@intel.com> (v2)
Francisco Jerez [Tue, 14 Jul 2015 12:43:44 +0000 (15:43 +0300)]
i965/fs: Factor out universally broken calculation of the register component size.
This in principle simple calculation was being open-coded in a number
of places (in a series I haven't yet sent for review there will be a
couple more), all of them were subtly broken in one way or another:
None of them were handling the HW_REG case correctly as pointed out by
Connor, and fs_inst::regs_read() was handling the stride=0 case rather
naively. This patch solves both problems and factors out the
calculation as a new fs_reg method.
Reviewed-by: Jason Ekstrand <jason.ekstrand@intel.com>
Francisco Jerez [Thu, 9 Jul 2015 18:42:28 +0000 (21:42 +0300)]
i965: Implement nir_op_uadd_carry and _usub_borrow without accumulator.
This gets rid of two no16() fall-backs and should allow better
scheduling of the generated IR. There are no uses of usubBorrow() or
uaddCarry() in shader-db so no changes are expected. However the
"arb_gpu_shader5/execution/built-in-functions/fs-usubBorrow" and
"arb_gpu_shader5/execution/built-in-functions/fs-uaddCarry" piglit
tests go from 40 to 28 instructions. The reason is that the plain ADD
instruction can easily be CSE'ed with the original addition, and the
b2i negation can easily be propagated into the source modifier of
another instruction, so effectively both operations are performed with
just one instruction.
v2: Rely on carry_to_arith() and borrow_to_arith() to lower these
(Ilia Mirkin).
Reviewed-by: Matt Turner <mattst88@gmail.com>
Francisco Jerez [Fri, 10 Jul 2015 16:18:39 +0000 (19:18 +0300)]
i965: Implement b2f and b2i using negation.
Booleans are represented as 0/-1 on modern hardware which means we can
just negate them to convert them into a numeric type. Negation has
the benefit that it can be implemented using a source modifier which
can easily be propagated into some other instruction. shader-db
results on HSW:
total instructions in shared programs:
6349082 ->
6346693 (-0.04%)
instructions in affected programs: 40948 -> 38559 (-5.83%)
helped: 123
HURT: 1
GAINED: 1
LOST: 0
Reviewed-by: Matt Turner <mattst88@gmail.com>
Marek Olšák [Sun, 5 Jul 2015 13:00:22 +0000 (15:00 +0200)]
gallium: add interface for writable shader buffers
Reviewed-by: Ilia Mirkin <imirkin@alum.mit.edu>
Marek Olšák [Sun, 5 Jul 2015 12:48:33 +0000 (14:48 +0200)]
gallium: add interface for writable shader images
PIPE_CAPs will be added some other time.
Reviewed-by: Ilia Mirkin <imirkin@alum.mit.edu>
Marek Olšák [Sun, 5 Jul 2015 12:34:13 +0000 (14:34 +0200)]
gallium: add new limits for shader buffers and images
Reviewed-by: Ilia Mirkin <imirkin@alum.mit.edu>
Marek Olšák [Sun, 5 Jul 2015 11:51:16 +0000 (13:51 +0200)]
gallium: add BIND flags for R/W buffers and images
PIPE_CAPs and TGSI support will be added later. The TGSI support should be
straightforward. We only need to split TGSI_FILE_RESOURCE into TGSI_FILE_IMAGE
and TGSI_FILE_BUFFER, though duplicating all opcodes shouldn't be necessary.
The idea is:
* ARB_shader_image_load_store should use set_shader_images.
* ARB_shader_storage_buffer_object should use set_shader_buffers(slots 0..M-1)
if M shader storage buffers are supported.
* ARB_shader_atomic_counters should use set_shader_buffers(slots M..N)
if N-M+1 atomic counter buffers are supported.
PIPE_CAPs can describe various constraints for early DX11 hardware.
Reviewed-by: Ilia Mirkin <imirkin@alum.mit.edu>
Marek Olšák [Fri, 12 Jun 2015 12:24:17 +0000 (14:24 +0200)]
gallium: add PIPE_CAP_MAX_SHADER_PATCH_VARYINGS
Reviewed-by: Ilia Mirkin <imirkin@alum.mit.edu>
Francisco Jerez [Wed, 1 Jul 2015 13:32:24 +0000 (16:32 +0300)]
i965/gen9: Use custom MOCS entries set up by the kernel.
Instead of relying on hardware defaults the i915 kernel driver is
going program custom MOCS tables system-wide on Gen9 hardware. The
"WT" entry previously used for renderbuffers had a number of problems:
It disabled caching on eLLC, it used a reserved L3 cacheability
setting, and it used to override the PTE controls making renderbuffers
always WT on LLC regardless of the kernel's setting. Instead use an
entry from the new MOCS tables with parameters: TC=LLC/eLLC, LeCC=PTE,
L3CC=WB.
The "WB" entry previously used for anything other than renderbuffers
has moved to a different index in the new MOCS tables but it should
have the same caching semantics as the old entry.
Even though the corresponding kernel change ("drm/i915: Added
Programming of the MOCS") is in a way an ABI break it doesn't seem
necessary to check that the kernel is recent enough because the change
should only affect Gen9 which is still unreleased hardware.
v2: Update MOCS values for the new Android-incompatible tables
introduced in v7 of the kernel patch.
Cc: 10.6 <mesa-stable@lists.freedesktop.org>
Reference: http://lists.freedesktop.org/archives/intel-gfx/2015-July/071080.html
Reviewed-by: Ben Widawsky <ben@bwidawsk.net>
EdB [Tue, 7 Jul 2015 15:58:56 +0000 (17:58 +0200)]
clover: little OpenCL status code logging clean
s/build_error/compile_error in order to match the stored OpenCL status code.
Make program::build catch and log every OpenCL error.
Make tgsi error triggering uniform with the llvm one.
Reviewed-by: Francisco Jerez <currojerez@riseup.net>
Renaud Gaubert [Sat, 11 Jul 2015 17:38:10 +0000 (19:38 +0200)]
glsl: avoid compiler's segfault when processing operators with void arguments
This is done by returning an rvalue of type void in the
ast_function_expression::hir function instead of a void expression.
This produces (in the case of the ternary) an hir with a call
to the void returning function and an assignment of a void variable
which will be optimized out (the assignment) during the optimization
pass.
This fix results in having a valid subexpression in the many
different cases where the subexpressions are functions whose
return values are void.
Thus preventing to dereference NULL in the following cases:
* binary operator
* unary operators
* ternary operator
* comparison operators (except equal and nequal operator)
Equal and nequal had to be handled as a special case because
instead of segfaulting on a forbidden syntax it was now accepting
expressions with a void return value on either (or both) side of
the expression.
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=85252
Signed-off-by: Renaud Gaubert <renaud@lse.epita.fr>
Reviewed-by: Gabriel Laskar <gabriel@lse.epita.fr>
Reviewed-by: Samuel Iglesias Gonsalvez <siglesias@igalia.com>
Roland Scheidegger [Thu, 16 Jul 2015 01:55:59 +0000 (03:55 +0200)]
r200: fix some potential big endian issues
The formats chosen (both by texture format choser, fbo storage allocation)
are different for big endian not just for rgba8 but also lower bit width
formats (why I don't actually know). Even the function to test for renderable
formats used different formats, however the actual colorbuffer setup did not.
And the blitter did not take that into account neither.
Untested (what could possibly go wrong...).
Same as for r100.
Acked-by: Marek Olšák <marek.olsak@amd.com>
Roland Scheidegger [Thu, 16 Jul 2015 01:18:20 +0000 (03:18 +0200)]
radeon: fix some potential big endian issues
The formats chosen (both by texture format choser, fbo storage allocation)
are different for big endian not just for rgba8 but also lower bit width
formats (why I don't actually know). Even the function to test for renderable
formats used different formats, however the actual colorbuffer setup did not.
And the blitter did not take that into account neither.
Untested (what could possibly go wrong...).
Acked-by: Marek Olšák <marek.olsak@amd.com>
Roland Scheidegger [Sat, 11 Jul 2015 18:03:27 +0000 (20:03 +0200)]
radeon/r200: mark state atoms as dirty after blits
Blit submits lots of packets which are usually handled by state atoms, so
these must be dirtied.
Not sure if this fixes anything, but it was a concern raised by bug 51658
(with this all issues there seen as actual bugs should be fixed, with the
exception of the patch to upload non-used texenv state atoms which I just
don't understand).
Acked-by: Marek Olšák <marek.olsak@amd.com>
Roland Scheidegger [Thu, 16 Jul 2015 01:06:47 +0000 (03:06 +0200)]
r200: fix fbo rendering by disabling optimized texture format chooser
It is rather unfortunate that we don't know if a texture is going to be used
as a rt later, and we lack the means to do something about a format chosen
which we can't render to directly, so disable this and always chose renderable
format for rgba8 textures.
This addresses an issue raised on (old) bug,
https://bugs.freedesktop.org/show_bug.cgi?id=51658 with gnome-shell, don't
know if that's still applicable but it might fix other things as well.
Acked-by: Marek Olšák <marek.olsak@amd.com>
Anuj Phogat [Tue, 7 Jul 2015 19:23:33 +0000 (12:23 -0700)]
i965: Fix 32 bit build warnings in intel_get_yf_ys_bo_size()
Along with fixing the type of pitch parameter, patch also changes
the types of few local variables and function return type.
Warnings fixed are:
intel_mipmap_tree.c:671:7: warning: passing argument 3 of
'intel_get_yf_ys_bo_size' from incompatible pointer type
intel_mipmap_tree.c:563:1: note: expected 'uint64_t *' but
argument is of type 'long unsigned int *'
Signed-off-by: Anuj Phogat <anuj.phogat@gmail.com>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Matt Turner [Thu, 9 Jul 2015 02:00:48 +0000 (19:00 -0700)]
i965: Optimize batchbuffer macros.
Previously OUT_BATCH was just a macro around an inline function which
does
brw->batch.map[brw->batch.used++] = dword;
When making consecutive calls to intel_batchbuffer_emit_dword() the
compiler isn't able to recognize that we're writing consecutive memory
locations or that it doesn't need to write batch.used back to memory
each time.
We can avoid both of these problems by making a local pointer to the
next location in the batch in BEGIN_BATCH().
Cuts 18k from the .text size.
text data bss dec hex filename
4946956 195152 26192
5168300 4edcac i965_dri.so before
4928956 195152 26192
5150300 4e965c i965_dri.so after
This series (including commit
c0433948) improves performance of Synmark
OglBatch7 by 8.01389% +/- 0.63922% (n=83) on Ivybridge.
Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
Matt Turner [Sat, 11 Jul 2015 21:36:25 +0000 (14:36 -0700)]
i965: Add and use USED_BATCH macro.
The next patch will replace the .used field with an on-demand
calculation of batchbuffer usage.
Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
Matt Turner [Thu, 9 Jul 2015 01:59:51 +0000 (18:59 -0700)]
i965: Split batch emission from relocation functions.
So that everything writing to the batch between BEGIN_BATCH() and
ADVANCE_BATCH() goes through OUT_BATCH.
Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
Matt Turner [Thu, 9 Jul 2015 01:56:52 +0000 (18:56 -0700)]
i965: Move BEGIN_BATCH() into same control flow as ADVANCE_BATCH().
BEGIN_BATCH() and ADVANCE_BATCH() will contain "do {" and "} while (0)"
respectively to allow declaring local variables used by intervening
OUT_BATCH macros. As such, BEGIN_BATCH() and ADVANCE_BATCH() need to be
in the same control flow.
Reviewed-by: Iago Toral Quiroga <itoral@igalia.com>
Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
Brian Paul [Wed, 15 Jul 2015 12:15:06 +0000 (06:15 -0600)]
osmesa: fix OSMesaPixelsStore typo
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=91337
Cc: 10.6 <mesa-stable@lists.freedesktop.org>
Reviewed-by: Emil Velikov <emil.l.velikov@gmail.com>
Eric Anholt [Tue, 14 Jul 2015 19:32:04 +0000 (12:32 -0700)]
vc4: Cache the texture p1 for the sampler.
Cuts another 12% of vc4_uniforms.o, in exchange for computing it at
CSO creation time.
Eric Anholt [Tue, 14 Jul 2015 19:18:40 +0000 (12:18 -0700)]
vc4: Cache texture p0/p1 setup for the sampler view.
In exchange for a bit of space and computation in CSO setup, we cut
vc4_uniform.c (draw time) code size by 4.8%.
Eric Anholt [Tue, 14 Jul 2015 19:21:23 +0000 (12:21 -0700)]
vc4: Move uniforms handling to a separate file.
The rest of vc4_program.c is about compiling, while this is about
uniform emit at draw time.
Eric Anholt [Tue, 14 Jul 2015 18:54:15 +0000 (11:54 -0700)]
vc4: Fix some -Wdouble-promotion warnings.
No code generation changes from this, but it'll be useful to have this
next time I go checking -Wdouble-promotion.
Jordan Justen [Fri, 12 Jun 2015 02:17:03 +0000 (19:17 -0700)]
i965/cs: Initialize GPGPU Thread Count
This field should always be set for gen8. In the bdw PRM, Volume 2d:
Command Reference: Structures under INTERFACE_DESCRIPTOR_DATA, DWORD
6, Bits 9:0, Number of Threads in GPGPU Thread Group:
"This field should not be set to 0 even if the barrier is disabled,
since an accurate value is needed for proper pre-emption."
In the HSW PRM, the it doesn't mention that it must always be set, but
it should not hurt.
Reported-by: Kristian Høgsberg <krh@bitplanet.net>
Signed-off-by: Jordan Justen <jordan.l.justen@intel.com>
Reviewed-by: Ben Widawsky <ben@bwidawsk.net>
Eric Anholt [Sat, 11 Jul 2015 00:00:34 +0000 (17:00 -0700)]
vc4: Fix compiler warnings on release builds.
Eric Anholt [Sat, 11 Jul 2015 00:01:37 +0000 (17:01 -0700)]
vc4: Add better debug for register allocation failure.
Eric Anholt [Fri, 10 Jul 2015 23:11:23 +0000 (16:11 -0700)]
vc4: Drop reloc_count tracking for debug asserts on non-debug builds.
Cuts another 88 bytes of compiled code.
Eric Anholt [Fri, 10 Jul 2015 05:51:06 +0000 (22:51 -0700)]
vc4: Rework cl handling to be friendlier to the compiler.
Drops 680 bytes of code, from avoiding a bunch of extra updates to the
next pointer in the struct.
Eric Anholt [Fri, 10 Jul 2015 21:46:42 +0000 (14:46 -0700)]
vc4: Make a helper function for getting the current offset in the CL.
I needed to rewrite this a bit for safety checking in the next commit.
Despite being a static inline of the same thing that was being done, we
lose 36 bytes of code for some reason.
Eric Anholt [Fri, 10 Jul 2015 05:48:17 +0000 (22:48 -0700)]
vc4: Drop separate cl*_reloc_hindex().
Now that RCL generation is in the kernel, we don't have any other
callers. Oddly, the compiler generates another 8 bytes of code for
this, but the simplification is worth it.
Eric Anholt [Fri, 10 Jul 2015 05:42:22 +0000 (22:42 -0700)]
vc4: Store reloc pointers as pointers, not offsets.
Now that we don't resize the CL as we build (it's set up at the top by
vc4_start_draw()), we can store the pointers instead of offsets from
the base. Saves a bit of math in emitting relocs (about 60 bytes of
code).
Eric Anholt [Tue, 30 Jun 2015 05:32:03 +0000 (22:32 -0700)]
vc4: Add perf debug for when we wait on BOs.
Matt Turner [Mon, 13 Jul 2015 22:40:41 +0000 (15:40 -0700)]
i965: Mark constant static data as const.
Reviewed-by: Topi Pohjolainen <topi.pohjolainen@intel.com>
Samuel Iglesias Gonsalvez [Wed, 8 Jul 2015 15:30:44 +0000 (17:30 +0200)]
glsl: Lower shader storage buffer object loads to GLSL IR instrinsics
Extend the existing lower_ubo_reference pass to also detect SSBO loads
and lower them to __intrinsic_load_ssbo intrinsics.
Signed-off-by: Samuel Iglesias Gonsalvez <siglesias@igalia.com>
Reviewed-by: Jordan Justen <jordan.l.justen@intel.com>
Samuel Iglesias Gonsalvez [Wed, 8 Jul 2015 15:03:06 +0000 (17:03 +0200)]
glsl: Lower shader storage buffer object writes to GLSL IR instrinsics
Extend the existing lower_ubo_reference pass to also detect SSBO writes
and lower them to __intrinsic_store_ssbo intrinsics.
Signed-off-by: Samuel Iglesias Gonsalvez <siglesias@igalia.com>
Reviewed-by: Jordan Justen <jordan.l.justen@intel.com>
Iago Toral Quiroga [Fri, 24 Apr 2015 09:17:15 +0000 (11:17 +0200)]
glsl: Don't do copy propagation on buffer variables
Since the backing storage for these is shared we cannot ensure that
the value won't change by writes from other threads. Normally SSBO
accesses are not guaranteed to be syncronized with other threads,
except when memoryBarrier is used. So, we might be able to optimize
some SSBO accesses, but for now we always take the safe path and emit
the SSBO access.
Reviewed-by: Jordan Justen <jordan.l.justen@intel.com>