Tested on gen9.
v2: Rename lower_txd_3d_surafaces flag to lower_txd_3d (Jason Ekstrand)
Signed-off-by: Sagar Ghuge <sagar.ghuge@intel.com>
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Note that we have an official GL extension number, pick the appropriate
section of the GLX spec to modify, and add changelog.
Reviewed-by: Emil Velikov <emil.velikov@collabora.com>
Signed-off-by: Adam Jackson <ajax@redhat.com>
This has not even had an attempt at implementation. If you asked for
renderer 0 - which, the spec implies, should always work - then
dri2_convert_glx_attribs would fail, we'd silently fall back to creating
an indirect context, and xserver would also not recognize the attribute
and would throw BadValue at you.
The API would be difficult to use in any case, since there's no way to
enumerate how many renderers the screen has. I'd be tempted to add that
by defining:
glXQueryRendererIntegerMESA(dpy, screen,
/* renderer = */ -1,
0, &value);
to return the number of renderers, but a new entrypoint might be
cleaner. Still, better to not specify it at all than to lie about it.
Reviewed-by: Emil Velikov <emil.velikov@collabora.com>
Signed-off-by: Adam Jackson <ajax@redhat.com>
In one place we say, if GLES isn't supported then the profile version
will be 0.0. Then later we say, if the GLES profile extension isn't
supported then GLX_RENDERER_OPENGL_ES_PROFILE_VERSION_MESA is not
mentioned in the spec. A strict reading of the latter would mean that
GLX_RENDERER_OPENGL_ES_PROFILE_VERSION_MESA is not a recognized token,
and the query should instead return False.
The implementation does not check for the GLES profile extensions, and
the additional complexity doesn't seem worth it. Removing the
interaction text makes the spec match the implementation.
Reviewed-by: Emil Velikov <emil.velikov@collabora.com>
Signed-off-by: Adam Jackson <ajax@redhat.com>
emit_intrinsic_store_image() is always using 4 components when
collecting registers for the value. When image has less than
4 components (e.g, r32f, rg32i, etc) this results in extra mov
instructions.
This patch uses the actual number of components from the image format.
For example, in a shader like:
layout (r32f, binding=0) writeonly uniform imageBuffer u_image;
...
void main(void) {
...
imageStore (u_image, some_offset, vec4(1.0));
...
}
instruction count is reduced in at least 3 instructions (note image
format is r32f, 1 component only).
This obviously reduces register pressure as well.
v2: - Added support for image formats from NV_image_format extension
(Ilia Mirkin).
- Return 4 components by default instead of asserting. (Rob Clark).
v3: Added more missing formats (Ilia Mirkin).
v4: Added a debug message for unknown image formats (Rob Clark).
Reviewed-by: Ilia Mirkin <imirkin@alum.mit.edu>
Reviewed-by: Rob Clark <robdclark@gmail.com>
Instead of going all the way back to the variable, just look at the
deref. The modes are guaranteed to be the same by nir_validate whenever
the variable can be found. This fixes clear_unused_for_modes for
derefs that don't have an accessible variable.
Reviewed-by: Timothy Arceri <tarceri@itsqueeze.com>
Instead of going all the way back to the variable, just look at the
deref. The modes are guaranteed to be the same by nir_validate whenever
the variable can be found. This fixes apply_barrier_for_modes for
derefs that don't have an accessible variable.
Reviewed-by: Timothy Arceri <tarceri@itsqueeze.com>
This is instead of looking all the way back to the variable which may
not exist for all derefs. This makes this code properly ignore casts
with modes other than the mode[s] we care about (where casts aren't
allowed).
Reviewed-by: Timothy Arceri <tarceri@itsqueeze.com>
This is instead of looking all the way back to the variable which may
not exist for all derefs. This makes this code properly ignore casts
with modes other than the mode[s] we care about (where casts aren't
allowed).
Reviewed-by: Timothy Arceri <tarceri@itsqueeze.com>
This is instead of looking all the way back to the variable which may
not exist for all derefs. This makes this code properly ignore casts
with modes other than the mode[s] we care about (where casts aren't
allowed).
Reviewed-by: Timothy Arceri <tarceri@itsqueeze.com>
This is instead of looking all the way back to the variable which may
not exist for all derefs. This makes this code properly ignore casts
with modes other than the mode[s] we care about (where casts aren't
allowed).
Reviewed-by: Timothy Arceri <tarceri@itsqueeze.com>
If we can't find the variable from the deref, just assume it isn't
invariant and continue on. This can happen if, for instance, we're
writing to a deref that points into an SSBO.
Reviewed-by: Timothy Arceri <tarceri@itsqueeze.com>
"There's no point in walking the program if we're never going to
actually lower anything."
Except we might lower compacted local arrays. In that case, modes will
be 0, but there is still lowering to be done.
This reverts commit 7f75cf2a94.
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=109081
Suggested-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Tested-by: Clayton Craft <clayton.a.craft@intel.com>
Cc: Kenneth Graunke <kenneth@whitecape.org>
Each time I have to touch the buffer import/export functions in the dri
state tracker I get lost in the maze of functions converting between
DRI_IMAGE_FOURCC, DRI_IMAGE_FORMAT, DRI_IMAGE_COMPONENTS and pipe format.
Rip it out and replace by a single table, which defines the correspondence
between the different representations.
Also this now stores all the known representations in the __DRIimageRec,
to avoid the loss of information we currently have when importing a buffer
with a fourcc, which doesn't have a corresponding dri format.
Signed-off-by: Lucas Stach <l.stach@pengutronix.de>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Currently all the EGL APIs are missing a way to specify how an imported
dma-buf is intended to be used. Demanding the format to be both usable
for sampling and rendering artificially restricts the list of formats a
driver is able to import.
Looking at how the Intel driver implements those DRI2 image APIs it
doesn't distinguish between render or sampler compatible formats. So
this patch aligns behavior between Intel and Gallium based drivers.
Signed-off-by: Lucas Stach <l.stach@pengutronix.de>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
There is no need to do the detour over the resource behind the
surface to get the format. Use the surface format directly.
Signed-off-by: Lucas Stach <l.stach@pengutronix.de>
Reviewed-by: Philipp Zabel <p.zabel@pengutronix.de>
GNU Hurd needs to turn off glx-direct, rather than special case it,
we'll just add a toggle.
CC: 18.3 <mesa-stable@lists.freedesktop.org>
Reviewed-by: Eric Engestrom <eric.engestrom@intel.com>
Old versions of meson returned ppc64le as the cpu_family for little
endian power8 cpus, versions >=0.48 don't do this, so the check wouldn't
work in that case. This generalizes the check to work for both old and
new versions of meson.
Fixes: 34bbb24ce7
("meson: Add support for ppc assembly/optimizations")
Reviewed-by: Eric Engestrom <eric.engestrom@intel.com>
Gen9-10 have fewer than 4 subslices per slice, so they need this to be
rounded up. Gen11 isn't documented as needing this hack, and it can
also have more than 4 subslices, so the hack actually can break things.
Reviewed-by: Anuj Phogat <anuj.phogat@gmail.com>
On some GPUs, especially older Intel GPUs, some math instructions are
very expensive. On those architectures, don't reduce flow control to a
csel if one of the branches contains one of these expensive math
instructions.
This prevents a bunch of cycle count regressions on pre-Gen6 platforms
with a later patch (intel/compiler: More peephole select for pre-Gen6).
v2: Remove stray #if block. Noticed by Thomas.
Signed-off-by: Ian Romanick <ian.d.romanick@intel.com>
Reviewed-by: Thomas Helland <thomashelland90@gmail.com>
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
That flow control may be trying to avoid invalid loads. On at least
some platforms, those loads can also be expensive.
No shader-db changes on any Intel platform (even with the later patch
"intel/compiler: More peephole select").
v2: Add a 'indirect_load_ok' flag to nir_opt_peephole_select. Suggested
by Rob. See also the big comment in src/intel/compiler/brw_nir.c.
v3: Use nir_deref_instr_has_indirect instead of deref_has_indirect (from
nir_lower_io_arrays_to_elements.c).
v4: Fix inverted condition in brw_nir.c. Noticed by Lionel.
Signed-off-by: Ian Romanick <ian.d.romanick@intel.com>
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
If there is a CMP.NZ that compares a single component (via a .zzzz
swizzle, for example) with 0, it can propagate its conditional modifier
back to a previous CMP that writes only that component. The specific
case that I saw was:
cmp.l.f0(8) g42<1>.xF g61<4>.xF (abs)g18<4>.zF
...
cmp.nz.f0(8) null<1>D g42<4>.xD 0D
In this case we can just delete the second CMP.
No changes on Broadwell or Skylake because they do not use the vec4
backend. Also no changes on GM45 or Iron Lake.
Sandy Bridge, Ivy Bridge, and Haswell had similar results. (Sandy Bridge shown)
total instructions in shared programs: 10856676 -> 10852569 (-0.04%)
instructions in affected programs: 228322 -> 224215 (-1.80%)
helped: 1331
HURT: 0
helped stats (abs) min: 1 max: 7 x̄: 3.09 x̃: 4
helped stats (rel) min: 0.11% max: 6.67% x̄: 1.88% x̃: 1.83%
95% mean confidence interval for instructions value: -3.19 -2.99
95% mean confidence interval for instructions %-change: -1.93% -1.83%
Instructions are helped.
total cycles in shared programs: 154788865 -> 154732047 (-0.04%)
cycles in affected programs: 2485892 -> 2429074 (-2.29%)
helped: 1097
HURT: 59
helped stats (abs) min: 2 max: 168 x̄: 51.96 x̃: 64
helped stats (rel) min: 0.12% max: 12.70% x̄: 3.44% x̃: 2.22%
HURT stats (abs) min: 2 max: 16 x̄: 3.02 x̃: 2
HURT stats (rel) min: 0.18% max: 0.83% x̄: 0.64% x̃: 0.71%
95% mean confidence interval for cycles value: -51.04 -47.26
95% mean confidence interval for cycles %-change: -3.40% -3.07%
Cycles are helped.
Signed-off-by: Ian Romanick <ian.d.romanick@intel.com>
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
The (-abs(x) >= 0) => (x == 0) optimization is removed from the vec4 and
scalar parts. In the VS part, adding the new pattern was not
helpful. The pattern that is removed is really old, and it has been
handled by NIR for ages.
All Gen7+ platforms had similar results. (Broadwell shown)
total instructions in shared programs: 14715715 -> 14715709 (<.01%)
instructions in affected programs: 474 -> 468 (-1.27%)
helped: 6
HURT: 0
helped stats (abs) min: 1 max: 1 x̄: 1.00 x̃: 1
helped stats (rel) min: 1.12% max: 1.35% x̄: 1.28% x̃: 1.35%
95% mean confidence interval for instructions value: -1.00 -1.00
95% mean confidence interval for instructions %-change: -1.40% -1.15%
Instructions are helped.
total cycles in shared programs: 559569911 -> 559569809 (<.01%)
cycles in affected programs: 5963 -> 5861 (-1.71%)
helped: 6
HURT: 0
helped stats (abs) min: 16 max: 18 x̄: 17.00 x̃: 17
helped stats (rel) min: 1.45% max: 1.88% x̄: 1.73% x̃: 1.85%
95% mean confidence interval for cycles value: -18.15 -15.85
95% mean confidence interval for cycles %-change: -1.95% -1.51%
Cycles are helped.
Iron Lake and Sandy Bridge had similar results. (Iron Lake shown)
total instructions in shared programs: 7780915 -> 7780913 (<.01%)
instructions in affected programs: 246 -> 244 (-0.81%)
helped: 2
HURT: 0
total cycles in shared programs: 177876108 -> 177876106 (<.01%)
cycles in affected programs: 3636 -> 3634 (-0.06%)
helped: 1
HURT: 0
GM45
total instructions in shared programs: 4799152 -> 4799151 (<.01%)
instructions in affected programs: 126 -> 125 (-0.79%)
helped: 1
HURT: 0
total cycles in shared programs: 122052654 -> 122052652 (<.01%)
cycles in affected programs: 3640 -> 3638 (-0.05%)
helped: 1
HURT: 0
Signed-off-by: Ian Romanick <ian.d.romanick@intel.com>
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
In an instruction sequence like
cmp(8).ge.f0.0 vgrf17:D, vgrf2.xxxx:D, vgrf9.xxxx:D
(+f0.0) sel(8) vgrf1:UD, vgrf8.xyzw:UD, vgrf1.xyzw:UD
The other fields of vgrf17 may be unused, but the CMP still needs to
generate the other flag bits.
To my surprise, nothing in shader-db or any test suite appears to hit
this. However, I have a change to brw_vec4_cmod_propagation that
creates cases where this can happen. This fix prevents a couple dozen
regressions in that patch.
Signed-off-by: Ian Romanick <ian.d.romanick@intel.com>
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Fixes: 5df88c20 ("i965/vec4: Rewrite dead code elimination to use live in/out.")
We were not using the view mask for depth clears, causing only the
first view to be cleared.
Fixes: 2e86f6b259 "radv: Add multiview clears."
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
The switch directly after the check has a default case that returns
NULL too, so the effective return value is not changed. Also this
check is wrong once we start dealing with formats introduced by an
extension (e.g. YUV formats).
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
I botched some copy-and-paste and clamped to signed int max instead of
uint max. Fixes KHR-GL46.shader_image_load_store.multiple-uniforms on
skl.
Fixes: d3e046e76c ("nir: Pull some of intel's image load/store format
conversion to nir_format.h")
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
I thought the value was correctly propagated, but actually not.
Fixes: 2ac6d55f38 ("radv: bump reported version to 1.1.90")
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Now at version 2 with the fixed header.
Reviewed-by: Samuel Iglesias Gonsálvez <siglesias@igalia.com>
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Fixes: 191a1dce92 ('nir: Add 1-bit Boolean opcodes')
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Fixes: f8d5b377c8 ('radv: set cb base tile swizzles for MRT speedups (v4)')
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=108116
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Reviewed-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
nir_sweep already marks all metadata invalid, so it is safe to release
the memory here too.
mean soft fp64 using uint64: 1,342,759,331 => 1,010,670,475
gfxbench5 aztec ruins high 11: 63,555,571 => 61,889,811
deus ex mankind divided 148: 62,845,304 => 62,829,640
deus ex mankind divided 2890: 71,922,686 => 71,922,686
dirt showdown 676: 69,238,607 => 69,238,607
dolphin ubershaders 210: 77,822,072 => 77,822,072
Signed-off-by: Ian Romanick <ian.d.romanick@intel.com>
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Found using pahole.
Changes in peak memory usage according to Valgrind massif:
mean soft fp64 using uint64: 1,343,991,403 => 1,342,759,331
gfxbench5 aztec ruins high 11: 63,619,971 => 63,555,571
deus ex mankind divided 148: 62,887,728 => 62,845,304
deus ex mankind divided 2890: 72,399,750 => 71,922,686
dirt showdown 676: 69,464,023 => 69,238,607
dolphin ubershaders 210: 78,359,728 => 77,822,072
Signed-off-by: Ian Romanick <ian.d.romanick@intel.com>
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Replace the old array in each value with a hash table in each value.
Changes in peak memory usage according to Valgrind massif:
mean soft fp64 using uint64: 5,499,875,082 => 1,343,991,403
gfxbench5 aztec ruins high 11: 63,619,971 => 63,619,971
deus ex mankind divided 148: 62,887,728 => 62,887,728
deus ex mankind divided 2890: 72,402,222 => 72,399,750
dirt showdown 676: 74,466,431 => 69,464,023
dolphin ubershaders 210: 109,630,376 => 78,359,728
Run-time change for a full run on shader-db on my Haswell desktop (with
-march=native) is 1.22245% +/- 0.463879% (n=11). This is about +2.9
seconds on a 237 second run. The first time I sent this version of this
patch out, the run-time data was quite different. I had misconfigured
the script that ran the test, and none of the tests from higher GLSL
versions were run. These are generally more complex shaders, and they
are more affected by this change.
The previous version of this patch used a single hash table for the
whole phi builder. The mapping was from [value, block] -> def, so a
separate allocation was needed for each [value, block] tuple. There was
quite a bit of per-allocation overhead (due to ralloc), so the patch was
followed by a patch that added the use of the slab allocator. The
results of those two patches was not quite as good:
mean soft fp64 using uint64: 5,499,875,082 => 1,343,991,403
gfxbench5 aztec ruins high 11: 63,619,971 => 63,619,971
deus ex mankind divided 148: 62,887,728 => 62,887,728
deus ex mankind divided 2890: 72,402,222 => 72,402,222 *
dirt showdown 676: 74,466,431 => 72,443,591 *
dolphin ubershaders 210: 109,630,376 => 81,034,320 *
The * denote tests that are better now. In the tests that are the same
in both patches, the "after" peak memory usage was at a different
location. I did not check the local peaks.
Signed-off-by: Ian Romanick <ian.d.romanick@intel.com>
Suggested-by: Jason Ekstrand <jason@jlekstrand.net>
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
D3D Booleans use a 32-bit 0/-1 representation. Because this previously
matched NIR exactly, we didn't have to really optimize for it. Now that
we have 1-bit Booleans, we need some specific optimizations to chew
through the D3D12-style Booleans.
Shader-db results on Kaby Lake:
total instructions in shared programs: 15136811 -> 14967944 (-1.12%)
instructions in affected programs: 2457021 -> 2288154 (-6.87%)
helped: 8318
HURT: 10
total cycles in shared programs: 373544524 -> 359701825 (-3.71%)
cycles in affected programs: 151029683 -> 137186984 (-9.17%)
helped: 7749
HURT: 682
total loops in shared programs: 4431 -> 4399 (-0.72%)
loops in affected programs: 32 -> 0
helped: 21
HURT: 0
total spills in shared programs: 10290 -> 10051 (-2.32%)
spills in affected programs: 2532 -> 2293 (-9.44%)
helped: 18
HURT: 18
total fills in shared programs: 22203 -> 21732 (-2.12%)
fills in affected programs: 3319 -> 2848 (-14.19%)
helped: 18
HURT: 18
Note that a large chunk of the improvement fixing regressions caused by
switching to 1-bit Booleans. Previously, our ability to optimize D3D
booleans was improved by using the D3D representation directly in NIR.
Now that NIR does 1-bit bools, we need a few more optimizations.
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Reviewed-by: Eric Anholt <eric@anholt.net>
Tested-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Reviewed-by: Eric Anholt <eric@anholt.net>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Tested-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
This is a squash of a few distinct changes:
glsl,spirv: Generate 1-bit Booleans
Revert "Use 32-bit opcodes in the NIR producers and optimizations"
Revert "nir/builder: Generate 32-bit bool opcodes transparently"
nir/builder: Generate 1-bit Booleans in nir_build_imm_bool
Reviewed-by: Eric Anholt <eric@anholt.net>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Tested-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
