Marek Olšák [Wed, 7 Aug 2019 02:51:56 +0000 (22:51 -0400)]
radeonsi: clean up si_llvm_context_set_tgsi
Reviewed-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Marek Olšák [Mon, 12 Aug 2019 17:50:00 +0000 (13:50 -0400)]
radeonsi: allocate and resize global_buffers as needed
Reviewed-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Marek Olšák [Tue, 6 Aug 2019 23:49:16 +0000 (19:49 -0400)]
radeonsi/gfx10: don't set PA_SC_TILE_STEERING_OVERRIDE if CLEAR_STATE sets it
Reviewed-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Marek Olšák [Tue, 6 Aug 2019 23:52:24 +0000 (19:52 -0400)]
radeonsi: don't emit PKT3_CONTEXT_CONTROL on amdgpu
Reviewed-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Marek Olšák [Fri, 2 Aug 2019 23:07:58 +0000 (19:07 -0400)]
radeonsi: fix an assertion failure: assert(!res->b.is_shared)
This only appears to happen on Raven2.
Possible way to reproduce:
resource_get_handle(WINSYS_HANDLE_TYPE_KMS) --> sets is_shared = true
resource_get_handle(WINSYS_HANDLE_TYPE_DMABUF) --> fail
Cc: 19.1 19.2 <mesa-stable@lists.freedesktop.org>
Marek Olšák [Fri, 2 Aug 2019 21:27:04 +0000 (17:27 -0400)]
radeonsi: handle the use_ngg_streamout flag in si_update_ngg
Marek Olšák [Fri, 2 Aug 2019 20:26:21 +0000 (16:26 -0400)]
radeonsi: move the tess factor ring size assertion to a place where it matters
Reviewed-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Marek Olšák [Wed, 7 Aug 2019 01:07:21 +0000 (21:07 -0400)]
ac/nir: set image=true when loading FMASK for images
Reviewed-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Christian Gmeiner [Tue, 16 Jul 2019 18:14:00 +0000 (20:14 +0200)]
etnaviv: rs: add support for 64bpp clears
Starting with HALTI2 the RS supports 64bpp clears.
Signed-off-by: Christian Gmeiner <christian.gmeiner@gmail.com>
Reviewed-by: Philipp Zabel <philipp.zabel@gmail.com>
Reviewed-by: Jonathan Marek <jonathan@marek.ca>
Christian GMEINER [Mon, 19 Aug 2019 15:07:29 +0000 (17:07 +0200)]
etnaviv: update headers from rnndb
Update to etna_viv commit
c51353e.
Signed-off-by: Christian GMEINER <christian.GMEINER@bachmann.info>
Reviewed-by: Philipp Zabel <philipp.zabel@gmail.com>
Reviewed-by: Jonathan Marek <jonathan@marek.ca>
Eric Anholt [Wed, 31 Jul 2019 20:30:22 +0000 (13:30 -0700)]
swrast: Make the fetch funcs table sparse.
This shrinks the table, avoids needing to update the table with NULL
entries on every MESA_FORMAT addition, and removes a surprising,
non-unit-tested format number ordering dependency.
Acked-by: Jose Fonseca <jfonseca@vmware.com>
Reviewed-by: Charmaine Lee <charmainel@vmware.com>
Eric Anholt [Tue, 13 Aug 2019 00:11:36 +0000 (17:11 -0700)]
gallium: Remove manual defining of PIPE_FORMAT enum values.
Now that SVGA doesn't have a table that has to be in PIPE_FORMAT
order, we can let the enums have whatever values they naturally would
without worrying about holes.
Acked-by: Jose Fonseca <jfonseca@vmware.com>
Reviewed-by: Charmaine Lee <charmainel@vmware.com>
Eric Anholt [Tue, 13 Aug 2019 00:03:23 +0000 (17:03 -0700)]
svga: Drop unsupported formats from the format table.
Now that we're using the array initializers, we don't need to manually
fill out all these stub entries.
Produced with "sed -i '/.*INVALID.*INVALID.*INVALID/d'
src/gallium/drivers/svga/svga_format.c"
Acked-by: Jose Fonseca <jfonseca@vmware.com>
Reviewed-by: Charmaine Lee <charmainel@vmware.com>
Eric Anholt [Mon, 12 Aug 2019 23:57:52 +0000 (16:57 -0700)]
svga: Remove duplication in the format table.
By using the [ ] = {} array initializer syntax, we no longer need the
entries to be listed in PIPE_FORMAT_* value order. This means that
people adding new gallium formats don't need to cargo-cult changes to
this driver or regress that non-unit-tested requirement.
While I'm here, drop the lines for formats that no longer exist (the
numbered ones in the table).
Acked-by: Jose Fonseca <jfonseca@vmware.com>
Reviewed-by: Charmaine Lee <charmainel@vmware.com>
Eric Anholt [Mon, 12 Aug 2019 23:50:50 +0000 (16:50 -0700)]
svga: Factor out the format conversion table entry lookup.
Seemed like a sensible cleanup, while I was looking at whether I could
make the table sparse.
To make the svga table not require fixups on every new gallium format,
we may want to change how it's populated.
Acked-by: Jose Fonseca <jfonseca@vmware.com>
Reviewed-by: Charmaine Lee <charmainel@vmware.com>
Jason Ekstrand [Mon, 19 Aug 2019 01:53:24 +0000 (20:53 -0500)]
nir: Add more source types to nir_tex_instr_src_type
Reviewed-by: Dave Airlie <airlied@redhat.com>
Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Alyssa Rosenzweig [Thu, 15 Aug 2019 21:53:56 +0000 (14:53 -0700)]
pan/midgard: Compute liveness per-block
Rather than using a regalloc based on live internals, computed hastily
with repeated invocations of a forward-analysis pass, we switch to
compute liveness information on a per-block basis.
Within a given basic block, we compute liveness backwards with a
linear-time algorithm; for common shaders, this may help RA terminate
quicker.
Across blocks, we use a work list (really a work set) and check if we're
making progress. This isn't terribly efficient, but it gets the job
done. Point is, we get the live_in/live_out for each block.
From there, it's simple to rerun the linear-time update algorithm to
compute the interference graph.
The benefit of this technique is the ability to ignore "gaps" in
liveness across intermediate blocks that are never executed. On simple
shaders like the loops in glmark, this results in a minor reduction in
register pressure. The motivation was a complex shader in Krita that
failed register allocation due to an unfortunate interaction between
texture pipeline registers and control flow. This shader now compiles
successfully.
total instructions in shared programs: 3439 -> 3438 (-0.03%)
instructions in affected programs: 22 -> 21 (-4.55%)
helped: 1
HURT: 0
total bundles in shared programs: 2077 -> 2076 (-0.05%)
bundles in affected programs: 12 -> 11 (-8.33%)
helped: 1
HURT: 0
total quadwords in shared programs: 3457 -> 3456 (-0.03%)
quadwords in affected programs: 20 -> 19 (-5.00%)
helped: 1
HURT: 0
total registers in shared programs: 341 -> 338 (-0.88%)
registers in affected programs: 9 -> 6 (-33.33%)
helped: 3
HURT: 0
helped stats (abs) min: 1 max: 1 x̄: 1.00 x̃: 1
helped stats (rel) min: 33.33% max: 33.33% x̄: 33.33% x̃: 33.33%
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Alyssa Rosenzweig [Fri, 16 Aug 2019 14:54:34 +0000 (07:54 -0700)]
pan/midgard: Analyze load/store for swizzle propagation
If there's a nontrivial swizzle fed into an extra (shortened) argument,
we bail on copyprop. No glmark changes (since it doesn't use fancy
texturing/loads).
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Alyssa Rosenzweig [Fri, 16 Aug 2019 14:50:12 +0000 (07:50 -0700)]
pan/midgard: Treat cubemaps "stores" as loads
It's always been ambiguous which they are, but their primary register is
their output, not their input; therefore, they are loads.
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Alyssa Rosenzweig [Fri, 16 Aug 2019 14:41:29 +0000 (07:41 -0700)]
pan/midgard: Clamp cubemap swizzle to XYXX
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Alyssa Rosenzweig [Thu, 15 Aug 2019 23:53:03 +0000 (16:53 -0700)]
pan/midgard: Clamp st_vary swizzle by number of components
Same issue with liveness analysis. If we store out a vec3, we should not
reference the .w component.
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Alyssa Rosenzweig [Thu, 15 Aug 2019 23:41:53 +0000 (16:41 -0700)]
pan/midgard: Use type-appropriate swizzle for texture coordinate
The texture coordinate for a 2D texture could be a vec2 or a vec3,
depending if it's an array texture or not. If it's vec2 (non-array
texture), we should not reference the z component; otherwise, liveness
analysis will get very confused when z is never written.
v2: Fix typo (Ilia).
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Alyssa Rosenzweig [Thu, 15 Aug 2019 23:29:15 +0000 (16:29 -0700)]
pan/midgard: Set mask for lowered read-hazard moves
If we need to lower a move for a read from a vec2 texture coordinate, we
shouldn't write zw, even incidentally.
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Alyssa Rosenzweig [Thu, 15 Aug 2019 21:23:05 +0000 (14:23 -0700)]
pan/midgard: Fix texw lowering with complex control flow
Fixes shaders with control flow like:
out = 0;
if (A) {
if (B)
out = texture(A, ...)
} else {
out = texture(B, ...)
}
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Alyssa Rosenzweig [Thu, 15 Aug 2019 21:22:48 +0000 (14:22 -0700)]
pan/midgard: Add mir_rewrite_index_dst_single helper
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Alyssa Rosenzweig [Thu, 15 Aug 2019 17:49:33 +0000 (10:49 -0700)]
pan/midgard: Print predecessors in MIR
Just as a sanity check.
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Alyssa Rosenzweig [Thu, 15 Aug 2019 17:43:56 +0000 (10:43 -0700)]
pan/midgard: Index blocks for printing
Better than having pointers flying about.
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Alyssa Rosenzweig [Thu, 15 Aug 2019 16:53:25 +0000 (09:53 -0700)]
pan/midgard: Add mir_foreach_src
This is repeated often enough.
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Alyssa Rosenzweig [Thu, 15 Aug 2019 16:51:13 +0000 (09:51 -0700)]
pan/midgard: Add mir_foreach_instr_in_block_rev
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Alyssa Rosenzweig [Thu, 15 Aug 2019 15:48:34 +0000 (08:48 -0700)]
pan/midgard: Add mir_foreach_successor helper
Now we should be able to walk the control-flow graph naturally.
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Alyssa Rosenzweig [Thu, 15 Aug 2019 15:36:41 +0000 (08:36 -0700)]
pan/midgard: Add mir_foreach_predecessor utility
It's ugly, but c'est la vie.
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Alyssa Rosenzweig [Thu, 15 Aug 2019 15:23:48 +0000 (08:23 -0700)]
pan/midgard: Link exit block
The exit block has been 'dangling' in the successors graph, so let's
ensure it's linked in.
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Alyssa Rosenzweig [Thu, 15 Aug 2019 15:16:38 +0000 (08:16 -0700)]
pan/midgard: Add mir_exit_block helper
The exit block is gauranteed to be empty, signaling the end of the
program.
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Alyssa Rosenzweig [Thu, 15 Aug 2019 15:11:10 +0000 (08:11 -0700)]
pan/midgard: Maintain block predecessor set
While we already compute the successors array, for backwards data flow
analysis, it is useful to walk the control flow graph backwards based on
predecessors, so let's compute that information as well.
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Alyssa Rosenzweig [Thu, 15 Aug 2019 15:10:46 +0000 (08:10 -0700)]
pan/midgard: Use ralloc on ctx/blocks
This will allow us to get some level of automatic memory management.
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Alyssa Rosenzweig [Thu, 15 Aug 2019 15:00:04 +0000 (08:00 -0700)]
pan/midgard: Shrink successors[] to 2 length
A block can't have more.
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Roman Stratiienko [Wed, 14 Aug 2019 08:38:43 +0000 (11:38 +0300)]
nir: Add missing dependency in Android.nir.gen.mk
Fixes incremental build with Android
Signed-off-by: Roman Stratiienko <roman.stratiienko@globallogic.com>
Reviewed-by: Tapani Pälli <tapani.palli@intel.com>
Erico Nunes [Fri, 16 Aug 2019 13:19:59 +0000 (15:19 +0200)]
meson: build lima tools as part of 'all' tools
This is primarily so that this build gets tested in CI and we don't
break it again.
Signed-off-by: Erico Nunes <nunes.erico@gmail.com>
Reviewed-by: Christian Gmeiner <christian.gmeiner@gmail.com>
Reviewed-by: Qiang Yu <yuq825@gmail.com>
Connor Abbott [Fri, 16 Aug 2019 10:46:27 +0000 (12:46 +0200)]
ac/nir: Fix store_scratch with a non-full writemask
By adding one more helper to ac_llvm_build, we can also easily keep
vector stores together.
Fixes the
tests/spec/glsl-1.30/execution/fs-large-local-array-vec4.shader_test
piglit test.
Fixes: 74470baebbd ("ac/nir: Lower large indirect variables to scratch")
Reviewed-by: Marek Olšák <marek.olsak@amd.com
Vasily Khoruzhick [Fri, 16 Aug 2019 05:17:26 +0000 (22:17 -0700)]
glsl/standalone: init shader stage in init_gl_program()
Otherwise lima standalone compiler fails when trying to compile fragment
shader with:
lima_compiler: ../src/compiler/nir/nir.c:55: nir_shader_create: Assertion `si->stage == stage' failed
Reviewed-by: Qiang Yu <yuq825@gmail.com>
Reviewed-by: Matt Turner <mattst88@gmail.com>
Signed-off-by: Vasily Khoruzhick <anarsoul@gmail.com>
Jason Ekstrand [Fri, 16 Aug 2019 22:48:53 +0000 (17:48 -0500)]
iris: Only request an input mask if the shader needs it
Fixes: aebca3961b "iris: Fix handling of SIMD32 fragment shaders"
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Xiong, James [Thu, 15 Aug 2019 17:02:52 +0000 (10:02 -0700)]
gallium: add back YVU support
PIPE_FORMAT_YV12 is not handled so switching to PIPE_FORMAT_IYUV and
adding back YVU support.
Signed-off-by: James Xiong <james.xiong@intel.com>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Erico Nunes [Wed, 14 Aug 2019 20:39:33 +0000 (22:39 +0200)]
lima: actually wait for bo in lima_bo_wait
PIPE_TIMEOUT_INFINITE is unsigned and gets assigned to signed fields
where it ends up as -1. When this reaches the kernel as a timeout it
gets translated as no timeout, which cause the waiting functions to
return immediately and not actually wait for a completion.
This seems to cause unstable results with lima where even piglit tests
randomly fail.
Handle this by setting the signed max value in case of infinite timeout.
Signed-off-by: Erico Nunes <nunes.erico@gmail.com>
Reviewed-by: Qiang Yu <yuq825@gmail.com>
Rhys Perry [Thu, 11 Jul 2019 14:31:50 +0000 (15:31 +0100)]
nir/algebraic: add a few masking-before-unpack optimizations
Helps some Dawn of War 3 and F1 2017 shaders with ACO:
Totals from affected shaders:
SGPRS: 2136 -> 2128 (-0.37 %)
VGPRS: 1624 -> 1628 (0.25 %)
Spilled SGPRs: 0 -> 0 (0.00 %)
Spilled VGPRs: 0 -> 0 (0.00 %)
Private memory VGPRs: 0 -> 0 (0.00 %)
Scratch size: 0 -> 0 (0.00 %) dwords per thread
Code Size: 168068 -> 164332 (-2.22 %) bytes
LDS: 44 -> 44 (0.00 %) blocks
Max Waves: 222 -> 221 (-0.45 %)
Wait states: 0 -> 0 (0.00 %)
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
Reviewed-by: Eric Anholt <eric@anholt.net>
Vasily Khoruzhick [Thu, 15 Aug 2019 05:08:50 +0000 (22:08 -0700)]
lima: fix compilation of standalone compiler
Fixes: e0aeee946004("lima: add summary report for shader-db")
Reviewed-by: Qiang Yu <yuq825@gmail.com>
Signed-off-by: Vasily Khoruzhick <anarsoul@gmail.com>
Bas Nieuwenhuizen [Thu, 15 Aug 2019 23:13:00 +0000 (01:13 +0200)]
Revert "radv/gfx10: Enable DCC for storage images."
Quite useless without DCC for LAYOUT_GENERAL.
Fixes: b4dad3afaa0 Revert "radv: Do not decompress on LAYOUT_GENERAL."
Acked-by: Dave Airlie <airlied@redhat.com>
Bas Nieuwenhuizen [Thu, 15 Aug 2019 23:06:55 +0000 (01:06 +0200)]
Revert "radv: Do not decompress on LAYOUT_GENERAL."
Causes issues with a bunch of games with DXVK.
Fixes: 50add1b33ae "radv: Do not decompress on LAYOUT_GENERAL."
Acked-by: Dave Airlie <airlied@redhat.com>
Dave Airlie [Fri, 9 Aug 2019 03:25:56 +0000 (13:25 +1000)]
mesa: add support for CET to x86/x86-64 asm files.
Control-flow enforcement technology is a new instructions on x86
processors to denote where indirect jumps can land. Gcc auto adds
the instruction (which encodes as a NOP on older CPUs) to entrypoints
but assembler files need manual adding. This adds it to all the
entry points in the mesa x86/x86-64 assembler files.
This will only happen if mesa is built with the -fcf-protection flag
to gcc as some distros are wanting to do.
Acked-by: Eric Anholt <eric@anholt.net>
Alyssa Rosenzweig [Thu, 15 Aug 2019 14:03:35 +0000 (07:03 -0700)]
pan/bifrost: Manually constant fold register class
Fixes errors for some people building Mesa:
../src/panfrost/bifrost/bifrost_sched.c:32:31: error: initializer
element is not constant
const unsigned max_vec2_reg = max_primary_reg / 2;
../src/panfrost/bifrost/bifrost_sched.c:33:31: error: initializer
element is not constant
const unsigned max_vec3_reg = max_primary_reg / 4; // XXX: Do we need
to align vec3 to vec4 boundary?
../src/panfrost/bifrost/bifrost_sched.c:34:31: error: initializer
element is not constant
const unsigned max_vec4_reg = max_primary_reg / 4;
../src/panfrost/bifrost/bifrost_sched.c:35:32: error: initializer
element is not constant
const unsigned max_registers = max_primary_reg +
../src/panfrost/bifrost/bifrost_sched.c:40:28: error: initializer
element is not constant
const unsigned vec2_base = primary_base + max_primary_reg;
../src/panfrost/bifrost/bifrost_sched.c:41:28: error: initializer
element is not constant
const unsigned vec3_base = vec2_base + max_vec2_reg;
../src/panfrost/bifrost/bifrost_sched.c:42:28: error: initializer
element is not constant
const unsigned vec4_base = vec3_base + max_vec3_reg;
../src/panfrost/bifrost/bifrost_sched.c:43:27: error: initializer
element is not constant
const unsigned vec4_end = vec4_base + max_vec4_reg;
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Erik Faye-Lund [Tue, 13 Aug 2019 11:02:24 +0000 (13:02 +0200)]
gallium/util: widen type before multiplication
This method returns size_t, but the multiplication multiplies two
integers, leading to overflow rather than type widening.
Noticed by compiling with MSVC, which emits a warning.
Signed-off-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
Reviewed-by: Eric Anholt <eric@anholt.net>
Erik Faye-Lund [Tue, 13 Aug 2019 11:01:42 +0000 (13:01 +0200)]
mesa: avoid warning on Windows
On Windows, p_atomic_inc_return returns an unsigned long long rather
than the type the pointer refers to, so let's make sure we cast the
result to the right type. Otherwise, we'll trigger a warning about
the wrong format-string for the type.
Signed-off-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
Reviewed-by: Eric Anholt <eric@anholt.net>
Erik Faye-Lund [Thu, 8 Aug 2019 12:08:16 +0000 (14:08 +0200)]
win32: unify strcasecmp definitions
There was two incompatible definitions of strcasecmp, which lead to a
compiler warning. Let's clean this up by only leaving one of them, and
using that one all the time.
Signed-off-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
Reviewed-by: Eric Anholt <eric@anholt.net>
Erik Faye-Lund [Thu, 8 Aug 2019 12:01:57 +0000 (14:01 +0200)]
mesa/main: avoid warning when casting offset to pointer
This generates a warning on some 64-bit systems, so let's cast to a
properly sized integer first.
Signed-off-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
Reviewed-by: Eric Anholt <eric@anholt.net>
Erik Faye-Lund [Thu, 8 Aug 2019 11:59:44 +0000 (13:59 +0200)]
nir: avoid warning when casting bogus pointer
This intentionally-bogus pointer generates a warning on some 64-bit
systems, so let's cast to a properly-sized integer first.
Signed-off-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
Reviewed-by: Eric Anholt <eric@anholt.net>
Erik Faye-Lund [Mon, 5 Aug 2019 15:37:15 +0000 (17:37 +0200)]
glsl: fixup u64-warning
Similarly to the unsigned-version, we need to first cast the result to a
suiting integer before negating the number, otherwise we'll trigger a
warning.
Signed-off-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
Reviewed-by: Eric Anholt <eric@anholt.net>
Kenneth Graunke [Wed, 14 Aug 2019 20:54:55 +0000 (13:54 -0700)]
isl: Enable Unorm Path in Color Pipe
Improves performance on my Icelake 8x8 locked to 700Mhz. For example,
some GfxBench5 subtests have the following results:
- [i965] gl_manhattan: ................ 7.01119% +/- 0.180971% (n=5)
- [i965] gl_4 (Car Chase): 4.24351% +/- 0.175622% (n=5)
- [i965] gl_blending: ................ 3.36327% +/- 0.180267% (n=5)
- [i965] gl_5_normal (Aztec Ruins): 1.67962% +/- 0.243534% (n=10)
- [iris] gl_manhattan: ................ 3.92357% +/- 0.073965% (n=25)
- [iris] gl_4 (Car Chase): 2.17746% +/- 0.
0826858% (n=5)
- [iris] gl_blending: ................ 2.79599% +/- 0.803652% (n=15)
- [iris] gl_5_normal (Aztec Ruins): 1.30930% +/- 0.106523% (n=25)
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Rafael Antognolli [Wed, 14 Aug 2019 22:13:55 +0000 (15:13 -0700)]
anv: Properly initialize device->slice_hash.
When subslices_delta == 0 and we take the early return,
device->slice_hash is not initialized on GEN11. It then causes a
segfault when going through anv_DestroyDevice, if compiled with
valgrind.
Fixes: 7bc022b4bbc ("anv/gen11: Emit SLICE_HASH_TABLE when pipes are
unbalanced.)
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Danylo Piliaiev [Tue, 13 Aug 2019 08:25:03 +0000 (11:25 +0300)]
intel/compiler: Fix resource leak in error path
CID:
1452261
Fixes: 04a99515 "intel/compiler: add ability to override shader's assembly"
Signed-off-by: Danylo Piliaiev <danylo.piliaiev@globallogic.com>
Reviewed-by: Tapani Pälli <tapani.palli@intel.com>
Alyssa Rosenzweig [Wed, 14 Aug 2019 15:44:40 +0000 (08:44 -0700)]
panfrost: Implement native RECT textures
We started honouring the normalized_coords flag in the texture
descriptor, but a bisection revealed that broke RECT textures -- since
we were *also* lowering them in the shader. So just remove the
shader-based lowering, use native RECT textures, and enjoy the nominal
reduction in complexity and performance boost.
Fixes: 3e47a1181b7 ("panfrost: Add MALI_SAMP_NORM_COORDS flag")
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Alyssa Rosenzweig [Tue, 13 Aug 2019 17:27:09 +0000 (10:27 -0700)]
panfrost: Add R10G10B10A2_SSCALED vertex format
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Alyssa Rosenzweig [Tue, 13 Aug 2019 18:10:06 +0000 (11:10 -0700)]
pan/midgard: Disassemble UBO index explicitly
It's a bit of a special case but that's fine.
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Alyssa Rosenzweig [Tue, 13 Aug 2019 16:27:43 +0000 (09:27 -0700)]
pan/midgard: Account for unaligned UBOs when promoting uniforms
We only know how to promote aligned accesses, although theoretically we
should be able to promote unaligned to swizzles in the future. Check
this.
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Alyssa Rosenzweig [Tue, 13 Aug 2019 16:27:16 +0000 (09:27 -0700)]
pan/midgard: Add mir_ubo_shift helper
Different UBO reads have different shift requirements.
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Alyssa Rosenzweig [Tue, 13 Aug 2019 16:13:31 +0000 (09:13 -0700)]
pan/midgard: Address emit_ubo_read offset in bytes
We'll want to be smarter about unaligned reads, so let's get this code
all in one place.
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Alyssa Rosenzweig [Tue, 13 Aug 2019 16:11:48 +0000 (09:11 -0700)]
pan/midgard: Wire writemask into UBO reads
Helps the disassembly be clearer and maybe regalloc be smarter.
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Alyssa Rosenzweig [Tue, 13 Aug 2019 15:51:40 +0000 (08:51 -0700)]
pan/midgard: Identify UBO/SSBO op symmetry
It's the same thing, just shifted.
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Alyssa Rosenzweig [Mon, 12 Aug 2019 23:14:03 +0000 (16:14 -0700)]
panfrost: Extend blending to MRT
Our hardware supports independent (per-RT) blending, but we need to
route those settings through from Gallium.
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Alyssa Rosenzweig [Mon, 12 Aug 2019 19:36:46 +0000 (12:36 -0700)]
pan/midgard: Emit store_output branch just-in-time
We'll need multiple branches for MRT, so we can't defer. Also, we need
to track dependencies to ensure r0 is set to the correct value for each
store_output.
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Alyssa Rosenzweig [Mon, 12 Aug 2019 22:29:03 +0000 (15:29 -0700)]
pan/midgard: Add dont_eliminate flag
We need to treat fragment writes specially.
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Alyssa Rosenzweig [Fri, 9 Aug 2019 22:12:30 +0000 (15:12 -0700)]
pan/mfbd: Stuff in RT count
Fixes DATA_INVALID_FAULTs with multiple render targets.
We do always allocate space for 4 cbufs just to keep things sane. This
may not be strictly necessary.
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Alyssa Rosenzweig [Mon, 12 Aug 2019 18:07:00 +0000 (11:07 -0700)]
pan/decode: Dump FBD tagged pointer
Turns out the rt count is stuffed in here..
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Alyssa Rosenzweig [Fri, 9 Aug 2019 23:04:24 +0000 (16:04 -0700)]
pan/decode: Decode invalid access type upon fault
We don't have a good way to confirm this, but it parallels the kernel
definitons for MMU faults nicely.
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Alyssa Rosenzweig [Fri, 9 Aug 2019 21:56:30 +0000 (14:56 -0700)]
pan/decode: Fix duplicate heap_end property
This was supposed to read heap_start. It's the same value but still,
better get this right.
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Alyssa Rosenzweig [Wed, 14 Aug 2019 23:01:38 +0000 (16:01 -0700)]
panfrost: Note "MFBD preload disable" bit
It's a chicken bit, as far as I can tell. Buck buck.
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Alyssa Rosenzweig [Wed, 14 Aug 2019 19:48:04 +0000 (12:48 -0700)]
pan/bifrost: Link in compiler
We enable the standalone compiler, build the new files, and let it
blast.
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Alyssa Rosenzweig [Wed, 14 Aug 2019 19:46:35 +0000 (12:46 -0700)]
pan/bifrost: Check in remainder of the Bifrost compiler
What it says on the tin.
Signed-off-by: Ryan Houdek <Sonicadvance1@gmail.com>
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Alyssa Rosenzweig [Wed, 14 Aug 2019 19:36:30 +0000 (12:36 -0700)]
pan/bifrost: Add bifrost_print.c/h
IR printers.
Signed-off-by: Ryan Houdek <Sonicadvance1@gmail.com>
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Alyssa Rosenzweig [Wed, 14 Aug 2019 19:28:01 +0000 (12:28 -0700)]
pan/bifrost: Style format the disassembler
$ astyle *.c *.h --style=linux -s8
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Alyssa Rosenzweig [Wed, 14 Aug 2019 19:21:23 +0000 (12:21 -0700)]
pan/bifrost: Stub out standalone compiler
We don't actually have a standalone compiler in-tree yet, but let's get
prepared for when we do.
Signed-off-by: Ryan Houdek <Sonicadvance1@gmail.com>
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Alyssa Rosenzweig [Wed, 14 Aug 2019 19:19:01 +0000 (12:19 -0700)]
pan/bifrost: Sync disassembler with Ryan's tree
The disassembler was updated to move common code with the compiler into
a shared header. Additional, some new ops and control registers relating
to rounding were added.
Signed-off-by: Ryan Houdek <Sonicadvance1@gmail.com>
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Alyssa Rosenzweig [Wed, 14 Aug 2019 21:46:21 +0000 (14:46 -0700)]
panfrost: Remove standalone pandecode tool
Now that panwrap has gained the ability to trace directly without
dumping to the filesystem, there's no need to lug around this tool.
I can assure you nobody will miss it.
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Alyssa Rosenzweig [Wed, 14 Aug 2019 21:42:07 +0000 (14:42 -0700)]
pan/midgard: Fix disassembly termination condition
Fixes: 863bdd1f8dc ("pan/midgard: Break, not return, in disassembler")
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Alyssa Rosenzweig [Wed, 14 Aug 2019 21:41:49 +0000 (14:41 -0700)]
panfrost: Ensure we upload at least 1 blend RT
Otherwise we'll get memory junk.
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Alyssa Rosenzweig [Wed, 14 Aug 2019 21:21:02 +0000 (14:21 -0700)]
panfrost: Zero tripipe on initialize
I don't think the hardware cares, but this adds a lot of noise to traces
that we would rather not need to look at.
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Alyssa Rosenzweig [Wed, 14 Aug 2019 21:11:54 +0000 (14:11 -0700)]
pan/midgard: Improve disassembler robustness
Some memory corruption / etc issues let to an accidental "fuzzing" of
the disassembler ;) This uncovered some issues leading to a disassembler
hang, so let's fix that.
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Alyssa Rosenzweig [Wed, 14 Aug 2019 20:52:03 +0000 (13:52 -0700)]
pan/decode: Split public.h out
We want a defined ABI for tracing; this set of functions should be as
small as strictly necessary to minimize panwrap shenanigans.
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Alyssa Rosenzweig [Wed, 14 Aug 2019 20:21:21 +0000 (13:21 -0700)]
pan/decode: Prefer uint64_t to mali_ptr
This removes an unwanted dependency on panfrost-job.h
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Alyssa Rosenzweig [Tue, 13 Aug 2019 22:59:41 +0000 (15:59 -0700)]
pan/midgard: Allocate spill_slot once
Multiple spill moves share a single spill slot. Issue found in Krita.
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Alyssa Rosenzweig [Tue, 13 Aug 2019 22:58:49 +0000 (15:58 -0700)]
pan/midgard: Use hint on midgard_instruction for spill_move
This allows us to have multiple spill moves, whereas otherwise for N
spill moves, the first N-1 would be clobbered. Issue found in Krita.
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Alyssa Rosenzweig [Tue, 13 Aug 2019 20:08:52 +0000 (13:08 -0700)]
panfrost: Remove panfrost_add_dependency asserts
It doesn't... make a ton of sense to need to assert and this routine is
hotter than you might expect. Doesn't matter for release builds, of
course.
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Marek Olšák [Wed, 2 Jan 2019 20:50:13 +0000 (15:50 -0500)]
radeonsi: add support for Renoir
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Eric Engestrom [Fri, 19 Jul 2019 23:42:13 +0000 (00:42 +0100)]
meson: add nir tests to the compiler/nir test suite
Signed-off-by: Eric Engestrom <eric.engestrom@intel.com>
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Eric Anholt <eric@anholt.net>
Eric Engestrom [Thu, 8 Aug 2019 10:13:43 +0000 (11:13 +0100)]
EGL: sync headers with Khronos
Signed-off-by: Eric Engestrom <eric.engestrom@intel.com>
Christian Gmeiner [Wed, 14 Aug 2019 07:37:35 +0000 (09:37 +0200)]
relnotes: Add new ext on etnaviv for 19.2.
Signed-off-by: Christian Gmeiner <christian.gmeiner@gmail.com>
Reviewed-by: Jonathan Marek <jonathan@marek.ca>
Christian Gmeiner [Wed, 14 Aug 2019 13:23:15 +0000 (15:23 +0200)]
etnaviv: fix weird indentation
Fixes: 797a2e4fd03 ("etnaviv: update logic to determine uniform limits")
Signed-off-by: Christian Gmeiner <christian.gmeiner@gmail.com>
Reviewed-by: Jonathan Marek <jonathan@marek.ca>
Ian Romanick [Tue, 6 Aug 2019 20:11:56 +0000 (13:11 -0700)]
nir/algebraic: Reassociate shift-by-constant of shift-by-constant
v2: After some review discussion with Alyssa, the replacements now
correct account for cases where (b+c) >= bitsize.
v3: Use a temporary to simplify the Python code quite a bit. Suggested
by Jason.
Haswell and all Gen8+ platforms had similar results. (Ice Lake shown)
total instructions in shared programs:
16251155 ->
16249576 (<.01%)
instructions in affected programs: 232627 -> 231048 (-0.68%)
helped: 547
HURT: 1
helped stats (abs) min: 1 max: 15 x̄: 2.89 x̃: 3
helped stats (rel) min: 0.04% max: 7.84% x̄: 1.14% x̃: 1.06%
HURT stats (abs) min: 2 max: 2 x̄: 2.00 x̃: 2
HURT stats (rel) min: 0.12% max: 0.12% x̄: 0.12% x̃: 0.12%
95% mean confidence interval for instructions value: -3.12 -2.65
95% mean confidence interval for instructions %-change: -1.20% -1.06%
Instructions are helped.
total cycles in shared programs:
365924392 ->
365372103 (-0.15%)
cycles in affected programs:
59207053 ->
58654764 (-0.93%)
helped: 497
HURT: 34
helped stats (abs) min: 1 max: 29300 x̄: 1118.16 x̃: 16
helped stats (rel) min: <.01% max: 10.59% x̄: 1.82% x̃: 1.82%
HURT stats (abs) min: 2 max: 424 x̄: 101.03 x̃: 63
HURT stats (rel) min: 0.07% max: 46.17% x̄: 4.72% x̃: 2.06%
95% mean confidence interval for cycles value: -1426.41 -653.77
95% mean confidence interval for cycles %-change: -1.66% -1.15%
Cycles are helped.
total spills in shared programs: 8870 -> 8871 (0.01%)
spills in affected programs: 104 -> 105 (0.96%)
helped: 0
HURT: 1
Ivy Bridge and all pre-Gen7 platforms had similar results. (Ivy Bridge shown)
total instructions in shared programs:
11956236 ->
11955635 (<.01%)
instructions in affected programs: 94110 -> 93509 (-0.64%)
helped: 106
HURT: 0
helped stats (abs) min: 1 max: 14 x̄: 5.67 x̃: 4
helped stats (rel) min: 0.12% max: 4.71% x̄: 1.96% x̃: 0.76%
95% mean confidence interval for instructions value: -6.62 -4.72
95% mean confidence interval for instructions %-change: -2.27% -1.64%
Instructions are helped.
total cycles in shared programs:
179296340 ->
178788044 (-0.28%)
cycles in affected programs:
51009603 ->
50501307 (-1.00%)
helped: 82
HURT: 7
helped stats (abs) min: 5 max: 27820 x̄: 6199.00 x̃: 16
helped stats (rel) min: 0.30% max: 8.16% x̄: 2.58% x̃: 3.11%
HURT stats (abs) min: 2 max: 8 x̄: 3.14 x̃: 2
HURT stats (rel) min: 0.02% max: 1.40% x̄: 0.34% x̃: 0.10%
95% mean confidence interval for cycles value: -7649.38 -3773.00
95% mean confidence interval for cycles %-change: -2.71% -1.99%
Cycles are helped.
Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com> [v2]
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Ian Romanick [Wed, 10 Jul 2019 23:28:38 +0000 (16:28 -0700)]
nir/algebraic: Reassociate add-and-shift to be shift-and-add
A common thing in many shaders:
uniform vs { vec4 bones[...]; };
...
x = some_calculation(bones[i + 0]);
y = some_calculation(bones[i + 1]);
z = some_calculation(bones[i + 2]);
This turns into stuff like
vec1 32 ssa_12 = iadd ssa_11, ssa_0
vec1 32 ssa_13 = ishl ssa_12, ssa_3
vec1 32 ssa_14 = intrinsic load_ssbo (ssa_7, ssa_13) (16, 4, 0)
vec1 32 ssa_15 = iadd ssa_11, ssa_1
vec1 32 ssa_16 = ishl ssa_15, ssa_3
vec1 32 ssa_17 = intrinsic load_ssbo (ssa_7, ssa_16) (16, 4, 0)
vec1 32 ssa_18 = iadd ssa_11, ssa_2
vec1 32 ssa_19 = ishl ssa_18, ssa_3
vec1 32 ssa_20 = intrinsic load_ssbo (ssa_7, ssa_19) (16, 4, 0)
By reassociating the shift and the add, we can reduce this to
vec1 32 ssa_12 = ishl ssa_11, ssa_3
vec1 32 ssa_13 = iadd ssa_12, ssa_0
vec1 32 ssa_14 = intrinsic load_ssbo (ssa_7, ssa_13) (16, 4, 0)
vec1 32 ssa_16 = iadd ssa_12, ssa_1
vec1 32 ssa_17 = intrinsic load_ssbo (ssa_7, ssa_16) (16, 4, 0)
vec1 32 ssa_19 = iadd ssa_12, ssa_2
vec1 32 ssa_20 = intrinsic load_ssbo (ssa_7, ssa_19) (16, 4, 0)
v2: Add some commentary from Rhys Perry's nearly identical patch.
All Intel platforms had similar results. (Ice Lake shown)
total instructions in shared programs:
16277758 ->
16250704 (-0.17%)
instructions in affected programs:
1440284 ->
1413230 (-1.88%)
helped: 4920
HURT: 6
helped stats (abs) min: 1 max: 69 x̄: 5.50 x̃: 4
helped stats (rel) min: 0.10% max: 18.33% x̄: 2.21% x̃: 1.79%
HURT stats (abs) min: 1 max: 12 x̄: 4.50 x̃: 3
HURT stats (rel) min: 0.18% max: 3.23% x̄: 1.91% x̃: 2.55%
95% mean confidence interval for instructions value: -5.67 -5.31
95% mean confidence interval for instructions %-change: -2.26% -2.16%
Instructions are helped.
total cycles in shared programs:
367118526 ->
365895358 (-0.33%)
cycles in affected programs:
93504145 ->
92280977 (-1.31%)
helped: 2754
HURT: 1269
helped stats (abs) min: 1 max: 47039 x̄: 460.66 x̃: 16
helped stats (rel) min: <.01% max: 34.93% x̄: 3.77% x̃: 1.12%
HURT stats (abs) min: 1 max: 1500 x̄: 35.85 x̃: 9
HURT stats (rel) min: 0.01% max: 17.35% x̄: 2.18% x̃: 0.75%
95% mean confidence interval for cycles value: -387.31 -220.78
95% mean confidence interval for cycles %-change: -2.11% -1.68%
Cycles are helped.
LOST: 1
GAINED: 1
Reviewed-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Andrii Simiklit [Wed, 7 Aug 2019 10:56:38 +0000 (13:56 +0300)]
nir/find_array_copies: Reject copies with mismatched lengths
copy_deref for wildcard dereferences requires the same
arrays lengths otherwise it leads to a crash in optimizations
like 'nir_opt_copy_prop_vars' because these optimizations expect
'copy_deref' just for arrays with the same lengths.
v2: check was moved to 'try_match_deref' to fix aoa cases
(Jason Ekstrand <jason@jlekstrand.net>)
v3: -fixed comment
-the condition merged with other one
(Jason Ekstrand <jason@jlekstrand.net>)
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=111286
Signed-off-by: Andrii Simiklit <andrii.simiklit@globallogic.com>
Alyssa Rosenzweig [Wed, 14 Aug 2019 16:19:54 +0000 (09:19 -0700)]
pan/midgard: Prefix blobber-db output for grepping
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Alyssa Rosenzweig [Wed, 14 Aug 2019 16:11:55 +0000 (09:11 -0700)]
pan/midgard: Implement blobber-db
We wire through some shader-db-style stats on the current shader in the
disassemble so we can get a quick estimate of shader complexity from a
trace.
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Suggested-by: Rob Clark <robdclark@chromium.org>