Dave Airlie [Tue, 10 May 2016 04:40:54 +0000 (14:40 +1000)]
st/glsl_to_tgsi: brown paper bag for the input offsets fix.
Oops, thanks compiler.
Signed-off-by: Dave Airlie <airlied@redhat.com>
Dave Airlie [Tue, 10 May 2016 01:56:31 +0000 (11:56 +1000)]
glsl: check geometry output vertices limits.
This fixes:
GL45-CTS.geometry_shader.limits.max_output_vertices
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Signed-off-by: Dave Airlie <airlied@redhat.com>
Dave Airlie [Tue, 10 May 2016 01:06:47 +0000 (11:06 +1000)]
mesa/vbo: fix check for zero aliases with 2/10/10/10
This fixes:
GL33-CTS.gtf33.GL3Tests.vertex_type_2_10_10_10_rev.vertex_type_2_10_10_10_rev_attrib
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Cc: mesa-stable@lists.freedesktop.org
Signed-off-by: Dave Airlie <airlied@redhat.com>
Eduardo Lima Mitev [Thu, 5 May 2016 11:52:38 +0000 (13:52 +0200)]
nir/print: Print memory qualifiers in a variable declaration
Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
Eduardo Lima Mitev [Thu, 5 May 2016 11:52:37 +0000 (13:52 +0200)]
glsl: Apply memory qualifiers to vars inside named block interfaces
This is missing and memory qualifiers are currently being ignored for SSBOs.
Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
Dave Airlie [Tue, 10 May 2016 01:41:02 +0000 (11:41 +1000)]
st/glsl_to_tgsi: handle offsets from inputs
This fixes:
GL45-CTS.gpu_shader5.texture_gather_offset_color_repeat
Reviewed-by: Nicolai Hähnle <nicolai.haehnle@amd.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
Rob Clark [Mon, 9 May 2016 23:27:59 +0000 (19:27 -0400)]
scripts: bump git_reviewer.pl --git-min-percent default
Bump up default percentage of commits required to be auto-picked for CC.
Seems from a bit of trial-and-error to come up with a more reasonable
list of CC's this way.
Signed-off-by: Rob Clark <robclark@freedesktop.org>
Kenneth Graunke [Mon, 9 May 2016 04:49:34 +0000 (21:49 -0700)]
Revert "Revert "i965: Switch to scalar TCS by default.""
This reverts commit
bd326c229c528a214c9fda705e7a961cfa49ac9e.
Now that we've fixed the GPU hangs, let's turn it back on.
Reviewed-by: Kristian Høgsberg <krh@bitplanet.net>
Kenneth Graunke [Mon, 9 May 2016 04:15:04 +0000 (21:15 -0700)]
i965: Actually assign binding table offsets for the TCS.
As far as I can tell, this was just entirely missing...honestly, I'm
not sure how anything worked at all.
Caught by noticing GPU hangs in image load store tests with scalar TCS,
but probably has broader implications.
Cc: mesa-stable@lists.freedesktop.org
Signed-off-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Kristian Høgsberg <krh@bitplanet.net>
Kenneth Graunke [Sun, 8 May 2016 00:30:02 +0000 (17:30 -0700)]
i965: Clamp "Maximum VP Index" to 1 when gl_ViewportIndex isn't written.
fs_visitor::emit_urb_writes skips writing the VUE header for shaders
that don't write gl_PointSize, gl_Layer, or gl_ViewportIndex. This
leaves their values uninitialized. Kristian's nearby comment says:
"But often none of the special varyings that live there are written
and in that case we can skip writing to the vue header, provided the
corresponding state properly clamps the values further down the
pipeline."
However, we were clamping gl_ViewportIndex to [0, 15], so we would end
up using a random viewport. To fix this, detect when the shader doesn't
write gl_ViewportIndex, and clamp it to [0, 0].
The vec4 backend always writes zeros to the VUE header, so it doesn't
suffer from this problem. With vec4-style HWord writes, we can write
the header and position together in a single message. In the FS world,
we would need 4 extra MOVs of 0 and a longer message, or a separate
OWord write. It's likely cheaper to just clamp the value.
Fixes DiRT Showdown and Bioshock Infinite, which only rendered half of
the screen - the lower left of two triangles.
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=93054
Cc: mesa-stable@lists.freedesktop.org
Signed-off-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Kristian Høgsberg <krh@bitplanet.net>
Reviewed-by: Matt Turner <mattst88@gmail.com>
Jordan Justen [Mon, 9 May 2016 21:46:25 +0000 (14:46 -0700)]
i965/hsw: Fix brw_store_data_imm*
For Gen6 through Haswell dword 1 is MBZ. In gen 8 it becomes part of
the 64-bit address.
Signed-off-by: Jordan Justen <jordan.l.justen@intel.com>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Kenneth Graunke [Thu, 5 May 2016 06:44:25 +0000 (23:44 -0700)]
i965: Reimplement ARB_transform_feedback2 on Haswell and later.
My old implementation accumulated <start, end> pairs in a buffer,
and eventually processed that data on the CPU. This meant flushing
the batchbuffer and waiting for it to completely execute before we
could map it, resulting in really long stalls. We could also run out
of space in the buffer, and have to do this early.
Instead, we can use Haswell's MI_MATH command to do the (end - start)
subtraction, as well as the multiplication by 2 or 3 to convert from
the number of primitives written to the number of vertices written.
We still need to CS stall to read the counters, but otherwise everything
is completely pipelined - there's no CPU<->GPU synchronization required.
It also uses only 80 bytes in the buffer, no matter what.
Improves performance in Manhattan on Skylake GT3e at 800x600 by
6.1086% +/- 0.954166% (n=9). At 1920x1080, improves performance
by 2.82103% +/- 0.148596% (n=84).
v2: Fix number of primitives -> number of vertices calculation for
GL_TRIANGLES (I was multiplying by 4 instead of 3.) Caught by
Jordan Justen.
Signed-off-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Jordan Justen <jordan.l.justen@intel.com>
Kenneth Graunke [Thu, 5 May 2016 05:52:52 +0000 (22:52 -0700)]
i965: Add a brw_load_register_reg64 helper.
It appears that we can't do this in a single command (like we do for
MI_LOAD_REGISTER_IMM) - the Skylake simulator gets rather grumpy about
the command length if I try to combine them. No matter.
Signed-off-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Jordan Justen <jordan.l.justen@intel.com>
Kenneth Graunke [Thu, 5 May 2016 08:57:30 +0000 (01:57 -0700)]
i965: Only enable ARB_query_buffer_object for newer kernels on Haswell.
On Haswell, we need version 6 of the kernel command parser in order to
write the math registers. Our implementation of ARB_query_buffer_object
heavily relies on MI_MATH, so we should only advertise it when MI_MATH
is available. We also need MI_LOAD_REGISTER_REG, which requires version
7 of the command parser.
To make these checks easier, introduce a screen->has_mi_math_and_lrr
flag that will be set when both commands are supported.
Signed-off-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Jordan Justen <jordan.l.justen@intel.com>
Dave Airlie [Mon, 9 May 2016 06:13:45 +0000 (16:13 +1000)]
mesa/objectlabel: don't return info on genned but never bound textures.
This fixes some cases in the CTS KHR debug tests where it uses
glIsTexture to find an invalid ID and then call GetObjectLabel.
Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Signed-off-by: Dave Airlie <airlied@redhat.com>
Dave Airlie [Mon, 9 May 2016 06:09:13 +0000 (16:09 +1000)]
mesa: don't use genned but unnamed xfb objects.
If we try to draw or query an XFB object that hasn't been bound,
we shouldn't return any information.
This fixes a couple if cases in:
GL33-CTS.transform_feedback.api_errors_test
The ObjectLabel test is inspired by another test.
Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Signed-off-by: Dave Airlie <airlied@redhat.com>
Samuel Pitoiset [Sun, 8 May 2016 22:13:41 +0000 (00:13 +0200)]
nv50/ir: silence unsupported TGSI_PROPERTY_CS_FIXED_BLOCK_*
We don't need them for compute shaders.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Ilia Mirkin <imirkin@alum.mit.edu>
Jordan Justen [Thu, 5 May 2016 14:43:02 +0000 (07:43 -0700)]
mesa/compute: Fix indirect dispatch buffer size check on 32-bit systems
2655265fcba9017e793026c76e490e04db088c8f, but for compute.
Signed-off-by: Jordan Justen <jordan.l.justen@intel.com>
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Rob Clark [Mon, 9 May 2016 17:51:18 +0000 (13:51 -0400)]
freedreno/ir3: fix fallout from new block iterators
Since this is potentially modifying the block structure of the shader,
it needs the _safe() version of the iterator.
Signed-off-by: Rob Clark <robclark@freedesktop.org>
Nicolai Hähnle [Fri, 6 May 2016 16:52:17 +0000 (11:52 -0500)]
radeonsi: workaround for tesselation on SI
We request more than 32KB of LDS here, which SI doesn't have. Since LLVM
recently started checking the size of declared LDS allocations, all shaders
involved in tesselation fail to compile on SI.
Note that the entire calculation here seems wrong, given how we calculate
indices for generic attributes, so the number ends up wrong on CI+ as well.
A proper solution is clearly needed, but this patch should serve as a band-aid
for SI in the meantime.
Also note that the real size of the LDS allocation in hardware is independent
from what we tell LLVM, so this is really more of a "cosmetic" change.
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=95198
Cc: "11.2" <mesa-stable@lists.freedesktop.org>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Nicolai Hähnle [Thu, 5 May 2016 00:05:14 +0000 (19:05 -0500)]
radeonsi: always allocate export memory for pixel shaders
Experiments with framebuffer-no-attachments type draw calls have shown that
NULL exports stall terribly unless we ensure that export memory is allocated
by the SPI.
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Nicolai Hähnle [Wed, 4 May 2016 18:53:45 +0000 (13:53 -0500)]
radeonsi: expose performance counters as 64 bit
This is useful for shader-related counters, since they tend to quickly
exceed 32 bits.
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Rob Clark [Mon, 9 May 2016 16:46:13 +0000 (12:46 -0400)]
nir/search: fix typo
Signed-off-by: Rob Clark <robclark@freedesktop.org>
Tim Rowley [Mon, 28 Mar 2016 19:29:30 +0000 (14:29 -0500)]
gallium: enable intel jitevents profiling
LLVM when configured with "intel jitevents" enabled can inform
VTune about dynamic code, so individual shaders are attributed
profiling data and the resulting assembly can be examined.
Acked-by: Roland Scheidegger <sroland@vmware.com>
Bruce Cherniak [Mon, 9 May 2016 14:00:04 +0000 (09:00 -0500)]
swr: Add missing break in query switch statement.
Missed a switch break in query stat collection when refactoring queries.
Reviewed-by: George Kyriazis <george.kyriazis@intel.com>
Rob Clark [Fri, 6 May 2016 17:21:08 +0000 (13:21 -0400)]
freedreno/ir3: allow for additional VS sysval inputs
There are a total of four possible currently, rather than 2. So we need
to be prepared for the input array to grow by 16 components. We could
get away with less if we could pack sysval inputs.. and the way this is
handled currently isn't really the nicest thing. But it's a tactical
fix for an issue hit in:
GL31-CTS.gtf30.GL3Tests.transform_feedback.transform_feedback_vertex_id
Signed-off-by: Rob Clark <robclark@freedesktop.org>
Emil Velikov [Mon, 9 May 2016 13:28:20 +0000 (14:28 +0100)]
docs: add news item and link release notes for 11.1.4/11.2.2
Signed-off-by: Emil Velikov <emil.velikov@collabora.com>
Emil Velikov [Mon, 9 May 2016 13:21:01 +0000 (14:21 +0100)]
docs: add sha256 checksums for 11.2.2
Signed-off-by: Emil Velikov <emil.velikov@collabora.com>
Emil Velikov [Mon, 9 May 2016 12:54:59 +0000 (13:54 +0100)]
docs: add release notes for 11.2.2
Signed-off-by: Emil Velikov <emil.velikov@collabora.com>
Emil Velikov [Mon, 9 May 2016 12:49:18 +0000 (13:49 +0100)]
docs: add sha256 checksums for 11.1.4
Signed-off-by: Emil Velikov <emil.velikov@collabora.com>
Emil Velikov [Mon, 9 May 2016 12:25:35 +0000 (13:25 +0100)]
docs: add release notes for 11.1.4
Signed-off-by: Emil Velikov <emil.velikov@collabora.com>
Jose Fonseca [Fri, 6 May 2016 13:03:05 +0000 (14:03 +0100)]
scons: Improve Python module dependency discovery.
Several NIR scripts were using `from ... import ...` syntax, which wasn't
supported.
Using Python standard libary's modulefinder solves the problem with less
effort and hacks.
Reviewed-by: Brian Paul <brianp@vmware.com>
Marek Olšák [Sun, 24 Apr 2016 11:23:52 +0000 (13:23 +0200)]
r300g: add support for PIPE_FORMAT_x8R8G8B8_*
And set endian swap for packed formats the way it should be done
in theory.
This allows big endian to work again, but it can still be buggy.
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=71789
Cc: 11.1 11.2 <mesa-stable@lists.freedesktop.org>
Reviewed-by: Michel Dänzer <michel.daenzer@amd.com>
Daniel Stone [Mon, 2 May 2016 14:34:40 +0000 (15:34 +0100)]
Revert "i965: Always use Y-tiled buffers on SKL+"
This commit broke Weston, Mutter, and xf86-video-modesetting, on KMS.
In order to use Y-tiled buffers, the kernel requires the tiling mode to
be explicitly named through the I915_FORMAT_MOD_Y_TILED AddFB2 modifier;
it disallows any attempt to infer the buffer's tiling mode.
As the GBM API does not have a way to extract modifiers for a buffer,
this commit broke all users of GBM on SKL+. Revert it for now, until we
get a way to extract modifier information from GBM, and also let GBM
users inform the implementation that it intends to use the modifiers.
This reverts commit
6a0d036483caf87d43ebe2edd1905873446c9589.
Signed-off-by: Daniel Stone <daniels@collabora.com>
Acked-by: Ben Widawsky <ben@bwidawsk.net>
Tested-by: Hans de Goede <hdegoede@redhat.com>
Dave Airlie [Fri, 6 May 2016 00:21:14 +0000 (10:21 +1000)]
mesa/shader_query: add missing subroutines cases
ARRAY_SIZE and LOCATION should accept the SUBROUTINE_UNIFORM types.
Fixes:
GL43-CTS.program_interface_query.subroutines-vertex
GL43-CTS.program_interface_query.subroutines-tess-control
GL43-CTS.program_interface_query.subroutines-tess-eval
GL43-CTS.program_interface_query.subroutines-geometry
GL43-CTS.program_interface_query.subroutines-fragment
GL43-CTS.program_interface_query.subroutines-compute
Reviewed-by: Antia Puentes <apuentes@igalia.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
Kenneth Graunke [Sun, 13 Mar 2016 08:08:56 +0000 (00:08 -0800)]
spirv: Fix structure splitting with per-vertex interface arrays.
We want to use interface_type, not vtn_var->type. They're normally
equivalent, but for geometry/tessellation per-vertex interface arrays,
we need to unwrap a level.
Otherwise, we tried to iterate a structure members but instead used
an array length. If the array length was longer than the number of
fields in the structure, we'd crash.
Fixes the CreatePipelineGeometryInputBlockPositive layer validation
test.
v2: Just use glsl_without_array() on the vtn_var type
(requested by Jason Ekstrand).
Signed-off-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Chris Forbes <chrisforbes@google.com>
Kenneth Graunke [Tue, 15 Mar 2016 05:01:29 +0000 (22:01 -0700)]
compiler: Add a C wrapper for glsl_type::without_array().
Signed-off-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Chris Forbes <chrisforbes@google.com>
Nicolai Hähnle [Sat, 30 Apr 2016 04:41:59 +0000 (23:41 -0500)]
radeonsi: fix undefined behavior (memcpy arguments must be non-NULL)
Reviewed-by: Michel Dänzer <michel.daenzer@amd.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Nicolai Hähnle [Sat, 30 Apr 2016 03:29:22 +0000 (22:29 -0500)]
radeonsi: fix some reported undefined left-shifts
One of these is an unsigned bitfield, which I suspect is a false positive, but
gcc 5.3.1 complains about it with -fsanitize=undefined.
Reviewed-by: Michel Dänzer <michel.daenzer@amd.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Nicolai Hähnle [Sat, 30 Apr 2016 03:15:48 +0000 (22:15 -0500)]
gallium/radeon: clean left-shift undefined behavior
Shifting into the sign bit of a signed int is undefined behavior.
Unfortunately, there are potentially many places where this happens using
the register macros.
This commit is the result of running
sed -ie "s/(((\(\w\+\)) & 0x\(\w\+\)) << \(\w\+\))/(((unsigned)(\1) \& 0x\2) << \3)/g"
on all header files in gallium/{r600,radeon,radeonsi}.
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Nicolai Hähnle [Sat, 30 Apr 2016 02:56:25 +0000 (21:56 -0500)]
gallium: fix various undefined left shifts into sign bit
Funnily enough, some of these were turned into a compile-time error by gcc
with -fsanitize=undefined ("initializer is not a constant").
Reviewed-by: Brian Paul <brianp@vmware.com>
Reviewed-by: Michel Dänzer <michel.daenzer@amd.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Nicolai Hähnle [Sat, 30 Apr 2016 06:36:59 +0000 (01:36 -0500)]
compiler/glsl: do not downcast list sentinel
This crashes gcc's undefined behaviour sanitizer.
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Nicolai Hähnle [Sat, 30 Apr 2016 04:38:09 +0000 (23:38 -0500)]
mesa/main: fix another undefined left shift
Reviewed-by: Brian Paul <brianp@vmware.com>
Reviewed-by: Michel Dänzer <michel.daenzer@amd.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Nicolai Hähnle [Sat, 30 Apr 2016 03:28:00 +0000 (22:28 -0500)]
mesa/main: define _NEW_xxx flags as unsigned shifts
Since 1 << 31 complains about undefined behaviour; the others are changed
only for consistency.
Reviewed-by: Brian Paul <brianp@vmware.com>
Reviewed-by: Michel Dänzer <michel.daenzer@amd.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Bas Nieuwenhuizen [Thu, 5 May 2016 20:25:01 +0000 (22:25 +0200)]
radeonsi: Compute correct LDS size for fragment shaders.
No sure where the 36 came from, but we clearly need at least
48 bytes per attribute per primitive.
Signed-off-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Eric Anholt [Fri, 6 May 2016 01:04:14 +0000 (18:04 -0700)]
vc4: Add support for loading immediate values in QIR.
This will be used for resetting the uniform stream in the presence of
branching, but may also be useful as an optimization to reduce how many
uniforms we have to copy out per draw call (in exchange for increasing
icache pressure).
Eric Anholt [Tue, 3 May 2016 18:16:54 +0000 (11:16 -0700)]
vc4: Make vc4_qpu_validate() produce more verbose failures.
Seeing the expansion of a QPU_GET_FIELD in an assert isn't very
informative, and it's hard find what's going wrong without getting a dump
of the instruction that failed.
Eric Anholt [Mon, 2 May 2016 19:21:45 +0000 (12:21 -0700)]
vc4: Add a small QIR validate pass.
This has caught a couple of bugs during loop development so far, and I
should probably have written it long ago.
Eric Anholt [Fri, 6 May 2016 02:19:39 +0000 (19:19 -0700)]
vc4: Fix the src count on exp2/log2.
Found by the upcoming QIR validate pass.
Eric Anholt [Tue, 15 Mar 2016 21:15:51 +0000 (14:15 -0700)]
vc4: Reuse QPU disasm's cond flags in QIR.
In the process, this made me flatten out the "%s%s%s%s" fprintf arguments.
Eric Anholt [Mon, 2 May 2016 19:36:56 +0000 (12:36 -0700)]
vc4: When emitting an instruction to an existing temp, mark it non-SSA.
Prevents a bug in the later control-flow support series.
Eric Anholt [Mon, 2 May 2016 21:45:45 +0000 (14:45 -0700)]
vc4: Make sure that we don't overwrite the signal for PROG_END.
We should have already emitted a NOP due to the last instruction being a
TLB or VPM write. However, if you disable dead code elimination then you
might get dead code at the end, and that dead code might have the signal
bits set to something non-default, at which point you die in assertion
failure.
Samuel Pitoiset [Fri, 6 May 2016 14:25:45 +0000 (16:25 +0200)]
nvc0: unreference images when the context is destroyed
Like other resources, we need to unreference all images.
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Ilia Mirkin <imirkin@alum.mit.edu>
Jose Fonseca [Fri, 6 May 2016 11:00:08 +0000 (12:00 +0100)]
nir: Remove spurious return from void function.
Left over from
450c0613627d5a472fcf1122c15b66988abfb372.
Trivial. Built locally with clang and gcc.
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=95296
Marek Olšák [Tue, 3 May 2016 10:29:16 +0000 (12:29 +0200)]
radeonsi: set DECOMPRESS_Z_ON_FLUSH if nr_samples >= 4
Vulkan always sets this. It only affects in-place Z decompression.
This is recommended for performance, but what app uses MSAA depth
texturing?
Reviewed-by: Nicolai Hähnle <nicolai.haehnle@amd.com>
Marek Olšák [Sat, 30 Apr 2016 19:44:41 +0000 (21:44 +0200)]
r600g: use the hw MSAA resolving if formats are compatible
This allows resolving RGBA into RGBX.
This should improve HL2 Lost Coast performance.
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Kenneth Graunke [Fri, 6 May 2016 01:03:23 +0000 (18:03 -0700)]
Revert "i965: Switch to scalar TCS by default."
This reverts commit
b593737ed8349b280fa29242c35f565b59ab3025.
Apparently it causes GPU hangs on some image load store tests.
Let's turn it back off until we figure out why.
Leo Liu [Mon, 2 May 2016 17:18:52 +0000 (13:18 -0400)]
st/omx/enc: fix incorrect reference picture order for B frames
Stacking frames is for driver that's capable to do dual instances
encoding. Such feature is not enabled for B frames currently.
Signed-off-by: Leo Liu <leo.liu@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Cc: "11.1 11.2" <mesa-stable@lists.freedesktop.org>
Jason Ekstrand [Tue, 3 May 2016 19:24:51 +0000 (12:24 -0700)]
i965/fs: Move handling of samples_identical into the switch statement
This is where we handle texop_texture_samples so it makes things more
consistent.
Jason Ekstrand [Tue, 3 May 2016 18:57:30 +0000 (11:57 -0700)]
i965/fs: Simplify texture destination fixups
There are a few different fixups that we have to do for texture
destinations that re-arrange channels, fix hardware vs. API mismatches, or
just shrink the result to fit in the NIR destination. These were all being
done in a somewhat haphazard manner. This commit replaces all of the
shuffling with a single LOAD_PAYLOAD operation at the end and makes it much
easier to insert fixups between the texture instruction itself and the
LOAD_PAYLOAD.
Shader-db results on Haswell:
total instructions in shared programs:
6227035 ->
6226669 (-0.01%)
instructions in affected programs: 19119 -> 18753 (-1.91%)
helped: 85
HURT: 0
total cycles in shared programs:
56491626 ->
56476126 (-0.03%)
cycles in affected programs: 672420 -> 656920 (-2.31%)
helped: 92
HURT: 42
Jason Ekstrand [Tue, 3 May 2016 18:46:31 +0000 (11:46 -0700)]
i965/fs: stop inclinding glsl/ir.h in brw_fs.h
We are no longer using anything from GLSL IR in the FS backend.
Jason Ekstrand [Tue, 3 May 2016 17:41:38 +0000 (10:41 -0700)]
i965/fs: Merge nir_emit_texture and emit_texture
The fs_visitor::emit_texture helper originated when we still had both NIR
and IR visitors for the FS backend. Since the old visitor was removed,
emit_texture serves no real purpose beyond arbitrarily splitting
heavily-linked code across two functions.
Connor Abbott [Wed, 13 Apr 2016 21:11:53 +0000 (17:11 -0400)]
nir: remove now-unused nir_foreach_block*_call()
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Reviewed-by: Connor Abbott <cwabbott0@gmail.com>
Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
Connor Abbott [Wed, 13 Apr 2016 03:59:06 +0000 (23:59 -0400)]
vc4: fixup for new nir_foreach_block()
Reviewed-by: Eric Anholt <eric@anholt.net>
Connor Abbott [Wed, 13 Apr 2016 03:42:42 +0000 (23:42 -0400)]
ir3: fixup for new nir_foreach_block()
Jason Ekstrand [Thu, 28 Apr 2016 17:34:52 +0000 (10:34 -0700)]
nir/lower_double_ops: fixup for new nir_foreach_block()
Signed-off-by: Jason Ekstrand <jason@jlekstrand.net>
Reviewed-by: Connor Abbott <cwabbott0@gmail.com>
Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
Jason Ekstrand [Wed, 13 Apr 2016 23:26:57 +0000 (16:26 -0700)]
nir/lower_double_pack: fixup for new nir_foreach_block()
Signed-off-by: Jason Ekstrand <jason@jlekstrand.net>
Reviewed-by: Connor Abbott <cwabbott0@gmail.com>
Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
Jason Ekstrand [Wed, 13 Apr 2016 23:26:39 +0000 (16:26 -0700)]
nir/gather_info: fixup for new foreach_block()
Signed-off-by: Jason Ekstrand <jason@jlekstrand.net>
Reviewed-by: Connor Abbott <cwabbott0@gmail.com>
Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
Connor Abbott [Fri, 8 Apr 2016 21:20:53 +0000 (17:20 -0400)]
nir/lower_two_sided_color: fixup for new foreach_block()
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
Connor Abbott [Fri, 8 Apr 2016 20:30:23 +0000 (16:30 -0400)]
nir/lower_tex: fixup for new foreach_block()
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
Connor Abbott [Fri, 8 Apr 2016 20:22:13 +0000 (16:22 -0400)]
nir/lower_outputs_to_temporaries: fixup for new foreach_block()
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
Kenneth Graunke [Fri, 22 Apr 2016 04:42:08 +0000 (21:42 -0700)]
i965: Switch to scalar TCS by default.
Normally, we expect SIMD8 shaders to be more instructions than SIMD4x2
shaders, as it takes four instructions to operate on a vec4, rather than
a single instruction. However, the benefit is that it can process 8
objects per shader thread instead of 2.
Surprisingly, the shader-db statistics show an improvement in both
instruction and cycle counts:
Synmark: -31.25% instructions, -29.27% cycles, 0 hurt.
Tessmark: -36.92% instructions, -37.81% cycles, 0 hurt.
Unigine Heaven: -3.42% instructions, -17.95% cycles, 0 hurt.
Shadow of Mordor:
+13.24% instructions (26 with fewer instructions, 45 with more),
-5.23% cycles (44 with fewer cycles, 27 with more cycles).
Presumably, this is because the SIMD8 URB messages are a much more
natural fit than the SIMD4x2 URB messages - there's a ton less header
setup.
I benchmarked Shadow of Mordor and Unigine Heaven on my Skylake GT3e,
and the performance seems to be the same or increase ever so slightly
(< 1 FPS difference). So I believe it's strictly superior.
There's also a lot more optimization potential we can do in scalar mode.
This will also help us finish fp64 support, as scalar support is going
to land much sooner than vec4-mode support.
Signed-off-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Matt Turner <mattst88@gmail.com>
Kenneth Graunke [Wed, 20 Apr 2016 09:38:57 +0000 (02:38 -0700)]
nir: Optimize out stores of undefs.
There are a couple of cycle count changes in shader-db, but it's
basically a wash.
However, with the Broadwell scalar TCS backend enabled, many
Shadow of Mordor shaders benefit from this patch. Because we don't
batch up output writes for TCS, vec4 outputs might not have all
components defined. Many output writes have a value of undef,
which is useless.
With scalar TCS, stats for tessellation shaders on Broadwell:
total instructions in shared programs:
1283000 ->
1280444 (-0.20%)
instructions in affected programs: 34302 -> 31746 (-7.45%)
helped: 71
HURT: 0
total cycles in shared programs:
10798768 ->
10780682 (-0.17%)
cycles in affected programs: 158004 -> 139918 (-11.45%)
helped: 71
HURT: 0
Signed-off-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Matt Turner <mattst88@gmail.com>
Kenneth Graunke [Wed, 20 Apr 2016 08:42:25 +0000 (01:42 -0700)]
nir: Replace vecN(undef, undef, ...) with a single undef.
shader-db statistics on Broadwell:
total instructions in shared programs:
8963409 ->
8962455 (-0.01%)
instructions in affected programs: 60858 -> 59904 (-1.57%)
helped: 318
HURT: 0
total cycles in shared programs:
71408022 ->
71406276 (-0.00%)
cycles in affected programs: 398416 -> 396670 (-0.44%)
helped: 199
HURT: 51
GAINED: 1
The only shaders affected were in Dota 2 Reborn.
It also sets up for the next optimization.
Signed-off-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Matt Turner <mattst88@gmail.com>
Kenneth Graunke [Wed, 20 Apr 2016 08:15:53 +0000 (01:15 -0700)]
nir: Rename opt_undef_alu to opt_undef_csel; update comments.
This better reflects what it does. I plan to add other ALU
optimizations as well, so the old name would be confusing.
In preparation for that, also move the file comments about csels
above the opt_undef_csel function, and delete the ones about there
not being other optimizations.
Signed-off-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Matt Turner <mattst88@gmail.com>
Kenneth Graunke [Tue, 3 May 2016 20:14:31 +0000 (13:14 -0700)]
i965: Rework passthrough TCS checks.
According to Timothy, using program_string_id == 0 to identify the
passthrough TCS is going to be problematic for his shader cache work.
So, change it to strcmp() the name at visitor creation time.
Signed-off-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Matt Turner <mattst88@gmail.com>
Tim Rowley [Mon, 2 May 2016 20:09:39 +0000 (14:09 -0600)]
swr: [rasterizer core] Faster modulo operator in ProcessVerts
Avoid % operator, since we know that curVertex is always incrementing.
Reviewed-by: Bruce Cherniak <bruce.cherniak@intel.com>
Tim Rowley [Mon, 2 May 2016 19:35:25 +0000 (13:35 -0600)]
swr: [rasterizer] Small warning cleanup
Reviewed-by: Bruce Cherniak <bruce.cherniak@intel.com>
Tim Rowley [Mon, 2 May 2016 17:37:37 +0000 (11:37 -0600)]
swr: [rasterizer] Add SWR_ASSUME / SWR_ASSUME_ASSERT macros
Fix static code analysis errors found by coverity on Linux
Reviewed-by: Bruce Cherniak <bruce.cherniak@intel.com>
Tim Rowley [Sat, 30 Apr 2016 20:07:20 +0000 (14:07 -0600)]
swr: [rasterizer] Miscellaneous backend changes
Reviewed-by: Bruce Cherniak <bruce.cherniak@intel.com>
Tim Rowley [Fri, 29 Apr 2016 18:15:34 +0000 (12:15 -0600)]
swr: [rasterizer] Add support for X24_TYPELESS_G8_UINT format
Reviewed-by: Bruce Cherniak <bruce.cherniak@intel.com>
Tim Rowley [Fri, 29 Apr 2016 06:25:45 +0000 (00:25 -0600)]
swr: [rasterizer jitter] Fix printing bugs for tracing.
Reviewed-by: Bruce Cherniak <bruce.cherniak@intel.com>
Tim Rowley [Thu, 28 Apr 2016 06:58:26 +0000 (00:58 -0600)]
swr: [rasterizer memory] Add missing store tiles function
Storing color hot tile to 8bit w-major stencil format.
Reviewed-by: Bruce Cherniak <bruce.cherniak@intel.com>
Tim Rowley [Wed, 27 Apr 2016 23:16:26 +0000 (17:16 -0600)]
swr: [rasterizer jitter] Add asserts for supported formats in fetch shader
Reviewed-by: Bruce Cherniak <bruce.cherniak@intel.com>
Tim Rowley [Wed, 27 Apr 2016 22:45:13 +0000 (16:45 -0600)]
swr: [rasterizer core] Fix thread allocation
Fix windows in 32-bit mode when hyperthreading is disabled on Xeons.
Some support for asymmetric processor topologies.
Reviewed-by: Bruce Cherniak <bruce.cherniak@intel.com>
Tim Rowley [Wed, 27 Apr 2016 19:02:27 +0000 (13:02 -0600)]
swr: [rasterizer core] Fix threadviz support in buckets
Need to do lazy eval of the threadviz knob since order of globals
is undefined.
Reviewed-by: Bruce Cherniak <bruce.cherniak@intel.com>
Tim Rowley [Mon, 25 Apr 2016 20:19:58 +0000 (14:19 -0600)]
swr: [rasterizer] Whitespace cleanup and misc changes
Reviewed-by: Bruce Cherniak <bruce.cherniak@intel.com>
Nicolai Hähnle [Mon, 2 May 2016 21:22:03 +0000 (16:22 -0500)]
radeonsi: mark descriptor loads as using dynamically uniform indices
This tells LLVM to always use SMEM loads for descriptors. It fixes a
regression in piglit's arb_shader_storage_buffer_object/execution/indirect.shader_test
that was caused by LLVM r268259 (but the proper fix is really here in Mesa).
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Matt Turner [Tue, 3 May 2016 06:32:13 +0000 (23:32 -0700)]
i965/fs: Don't follow pow with an instruction with two dest regs.
Beginning with commit
7b208a73, Unigine Valley began hanging the GPU on
Gen >= 8 platforms.
Evidently that commit allowed the scheduler to make different choices
that somehow finally ran afoul of a hardware bug in which POW and FDIV
instructions may not be followed by an instruction with two destination
registers (including compressed instructions). I presume the conditions
are more complex than that, but the internal hardware bug report (BDWGFX
bug_de
1696294) does not contain much more information.
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=94924
Reviewed-by: Topi Pohjolainen <topi.pohjolainen@intel.com> [v1]
Tested-by: Mark Janes <mark.a.janes@intel.com> [v1]
Reviewed-by: Francisco Jerez <currojerez@riseup.net>
Bruce Cherniak [Thu, 28 Apr 2016 17:13:15 +0000 (12:13 -0500)]
swr: Remove stall waiting for core query counters.
When gathering query results, swr_gather_stats was
unnecessarily stalling the entire pipeline. Results are now
collected asynchronously, with a fence marking completion.
Reviewed-By: George Kyriazis <george.kyriazis@intel.com>
Dave Airlie [Thu, 5 May 2016 03:11:11 +0000 (13:11 +1000)]
mesa/ubo: add missing compute cases for ubo/atomic buffers
This fixes: GL43-CTS.compute_shader.resource-ubo
Reviewed-by: Anuj Phogat <anuj.phogat@gmail.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
Dave Airlie [Thu, 5 May 2016 00:41:29 +0000 (10:41 +1000)]
mesa/compute: drop pointless casts.
We already are a GLintptr, casting won't help.
Reviewed-by: Alejandro Piñeiro <apinheiro@igalia.com>
Reviewed-by: Anuj Phogat <anuj.phogat@gmail.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
Thomas Hindoe Paaboel Andersen [Wed, 4 May 2016 04:15:37 +0000 (06:15 +0200)]
mesa: remove null check before free
Reviewed-by: Eduardo Lima Mitev <elima@igalia.com>
Thomas Hindoe Paaboel Andersen [Wed, 4 May 2016 04:15:36 +0000 (06:15 +0200)]
freedreno: remove null check before free
Reviewed-by: Eduardo Lima Mitev <elima@igalia.com>
Thomas Hindoe Paaboel Andersen [Wed, 4 May 2016 03:48:39 +0000 (05:48 +0200)]
nir: fix assert for wildcard pairs
The assert was null checking dest_arr_parent twice. The intention
seems to be to check both dest_ and src_.
Added in
d3636da9
Reviewed-by: Eduardo Lima Mitev <elima@igalia.com>
Brian Paul [Wed, 4 May 2016 20:02:30 +0000 (14:02 -0600)]
glapi: fix parameter type for GetSamplerParameterIuivEXT() in es_EXT.xml
The function returns GLuint, not GLfloat values.
v2: also fix the OES function
Cc: "11.2" <mesa-stable@lists.freedesktop.org>
Reviewed-by: Charmaine Lee <charmainel@vmware.com>
Reviewed-by: Ilia Mirkin <imirkin@alum.mit.edu>
Brian Paul [Wed, 4 May 2016 20:01:29 +0000 (14:01 -0600)]
mesa: include texture format in glGenerateMipmap error message
Reviewed-by: Charmaine Lee <charmainel@vmware.com>
Brian Paul [Tue, 3 May 2016 01:10:06 +0000 (19:10 -0600)]
main: uses casts to silence some _mesa_debug() format warnings
Silences warnings with 32-bit Linux gcc builds and MinGW which doesn't
recognize the ‘t’ conversion character.
Reviewed-by: Sinclair Yeh <syeh@vmware.com>
Jordan Justen [Sun, 17 Apr 2016 21:07:47 +0000 (14:07 -0700)]
docs: Mark GL_ARB_query_buffer_object as done for i965/hsw+
Signed-off-by: Jordan Justen <jordan.l.justen@intel.com>
Jordan Justen [Tue, 19 Apr 2016 16:42:13 +0000 (09:42 -0700)]
i965: Implement ARB_query_buffer_object for HSW+
v2:
* Declare loop index variable at loop site (idr)
* Make arrays of MI_MATH instructions 'static const' (idr)
* Remove commented debug code (idr)
* Updated comment in set_query_availability (Ken)
* Replace switch with if/else in hsw_result_to_gpr0 (Ken)
* Only divide GL_FRAGMENT_SHADER_INVOCATIONS_ARB by 4 on
hsw and gen8 (Ken)
Signed-off-by: Jordan Justen <jordan.l.justen@intel.com>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>