Both loads and atomics had identical code to rewrite destinations,
and all cases had the same two lines to replace instructions.
Signed-off-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
The load/store/atomic cases all duplicated the get_io_offset code, with
a few tiny differences: stores didn't bother checking for per-vertex
inputs, because they can't be stored to, and atomics didn't check at
all, since shared variables aren't per-vertex.
However, it's harmless to check, and allows us to share more code.
Signed-off-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Less typing and word wrapping issues than intrin->variables[0]->var.
Signed-off-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
This is way better than the stupid string approach especially since you
could overflow the string. Again, I thought I had something better at one
point but it obviously got lost.
Signed-off-by: Jason Ekstrand <jason@jlekstrand.net>
Reviewed-by: Iago Toral Quiroga <itoral@igalia.com>
Cc: "12.0" <mesa-stable@lists.freedesktop.org>
It was returning true if the function types have different lengths rather
than false. This was new with the SPIR-V to NIR pass and I thought I'd
fixed it a while ago but it may have gotten lost in rebasing somewhere.
Signed-off-by: Jason Ekstrand <jason@jlekstrand.net>
Reviewed-by: Iago Toral Quiroga <itoral@igalia.com>
Cc: "12.0" <mesa-stable@lists.freedesktop.org>
I have no idea why we were multiplying by 4 before. The offsets we get
from SPIR-V are in bytes and so is nir->num_uniforms so there's no need to
do any adjustment whatsoever.
Signed-off-by: Jason Ekstrand <jason@jlekstrand.net>
Cc: "12.0" <mesa-stable@lists.freedesktop.org>
This was appearing in vc4 VS/CS in mupen64, due to vertex attrib lowering
producing some constants that were getting compared.
total instructions in shared programs: 112276 -> 112198 (-0.07%)
instructions in affected programs: 2239 -> 2161 (-3.48%)
total estimated cycles in shared programs: 283102 -> 283038 (-0.02%)
estimated cycles in affected programs: 2365 -> 2301 (-2.71%)
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
If a shader has an output array, it will get treated as though it were
gl_FragData and rewritten into gl_out_FragData instances. We only want
this to happen on the actual gl_FragData and not everything else.
This is a small part of the problem pointed out by the below bug.
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=96765
Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Cc: "11.2 12.0" <mesa-stable@lists.freedesktop.org>
This just stops counting and assigning a storage location for
these uniforms, the count is only used to create the uniform storage.
These uniform types don't use this storage.
Reviewed-by: Samuel Iglesias Gonsálvez <siglesias@igalia.com>
Some games are sloppy.. perhaps because it is defined behavior for DX or
perhaps because nv blob driver defaults things to zero.
So add driconf param to force uninitialized variables to default to zero.
This issue was observed with rust, from steam store. But has surfaced
elsewhere in the past.
Signed-off-by: Rob Clark <robclark@freedesktop.org>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
The linker deals with atomic counters in terms of uniforms but the
data structure are called after the atomic counters.
Renamed the data structures used in the linker for disambiguation.
Reviewed-by: Timothy Arceri <timothy.arceri@collabora.com>
Signed-off-by: Andres Gomez <agomez@igalia.com>
Currently the linker uses the uniform count for the total number of
atomic counters. However uniforms don't include the innermost array
dimension in their count, but atomic counters are expected to include
them.
Although the spec doesn't directly state this, it's clear how offsets
will be assigned for arrays.
From OpenGL 4.2 (Core Profile), page 98:
" * Arrays of type atomic_uint are stored in memory by element
order, with array element member zero at the lowest offset. The
difference in offsets between each pair of elements in the
array in basic machine units is referred to as the array
stride, and is constant across the entire array. The stride can
be queried by calling GetIntegerv with a pname of
ATOMIC_COUNTER_- ARRAY_STRIDE after a program is linked."
From that it is clear how arrays of atomic counters will interact with
GL_MAX_ATOMIC_COUNTER_BUFFER_SIZE.
For other kinds of uniforms it's also clear that each entry in an
array counts against the relevant limits.
Hence, although inferred, this is the expected behavior.
Fixes GL44-CTS.arrays_of_arrays_gl.AtomicDeclaration
Reviewed-by: Timothy Arceri <timothy.arceri@collabora.com>
Signed-off-by: Andres Gomez <agomez@igalia.com>
There are two distinctly different uses of this struct. The first
is to store GL shader objects. The second is to store information
about a shader stage thats been linked.
The two uses actually share few fields and there is clearly confusion
about their use. For example the linked shaders map one to one with
a program so can simply be destroyed along with the program. However
previously we were calling reference counting on the linked shaders.
We were also creating linked shaders with a name even though it
is always 0 and called the driver version of the _mesa_new_shader()
function unnecessarily for GL shader objects.
Acked-by: Iago Toral Quiroga <itoral@igalia.com>
This will allow us to split gl_shader into two different structs, one for
shader objects and one for linked shaders.
Reviewed-by: Iago Toral Quiroga <itoral@igalia.com>
Rather than passing in gl_shader we now pass in the IR. This will
allow us to later split gl_shader into two structs. One for use
as a linked per stage shader struct and one for use as a GL shader
object.
Reviewed-by: Iago Toral Quiroga <itoral@igalia.com>
The only part of an ir_texture which can be an array is the
offsets array in textureGatherOffsets() calls. We don't want
to lower those, because they're required to remain constants.
Fixes textureGatherOffsets with Gallium drivers such as llvmpipe,
which commit ef78df8d3b regressed.
Cc: mesa-stable@lists.freedesktop.org
Signed-off-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Ilia Mirkin <imirkin@alum.mit.edu>
The intent was to continue down the indirect chain, not to call ourselves
with unchanged input arguments. Found by code inspection, and comparison
to copy_prop_alu_src().
We haven't hit this because callers of NIR's copy prop are doing so in
SSA, before indirect variable dereferences have been lowered to registers.
Reviewed-by: Rob Clark <robdclark@gmail.com>
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
It defaults to true so default behavior doesn't change but it allows you to
do NIR_VALIDATE=false if you don't want validation. Disabling validation
can substantially speed up shader compiles so you frequently want to turn
it off if compiler invariants aren't in question.
Reviewed-by: Matt Turner <mattst88@gmail.com>
Reviewed-by: Matt Turner <mattst88@gmail.com>
Signed-off-by: Rob Clark <robclark@freedesktop.org>
Clean up misrepetitions ('if if', 'the the' etc) found throughout the
comments. This has been done manually, after grepping
case-insensitively for duplicate if, is, the, then, do, for, an,
plus a few other typos corrected in fly-by
v2:
* proper commit message and non-joke title;
* replace two 'as is' followed by 'is' to 'as-is'.
v3:
* 'a integer' => 'an integer' and similar (originally spotted by
Jason Ekstrand, I fixed a few other similar ones while at it)
Signed-off-by: Giuseppe Bilotta <giuseppe.bilotta@gmail.com>
Reviewed-by: Chad Versace <chad.versace@intel.com>
Constant propagation on arrays doesn't make a lot of sense. If the
array is only accessed with constant indexes, then opt_array_splitting
would split it up. Otherwise, we have variable indexing. If there's
multiple accesses, then constant propagation would end up replicating
the data.
The lower_const_arrays_to_uniforms pass creates uniforms for each
ir_constant with array type that it encounters. This means that it
creates redundant uniforms for each copy of the constant, which means
uploading too much data. It can even mean exceeding the maximum number
of uniform components, causing link failures.
We could try and teach the pass to de-duplicate the data by hashing
constants, but it makes more sense to avoid duplicating it in the first
place. We should promote constant arrays to uniforms, then propagate
the uniform access.
Fixes the TressFX shaders from Tomb Raider, which exceeded the maximum
number of uniform components by a huge margin and failed to link.
On Broadwell:
total instructions in shared programs: 9067702 -> 9068202 (0.01%)
instructions in affected programs: 10335 -> 10835 (4.84%)
helped: 10 (Hoard, Shadow of Mordor, Amnesia: The Dark Descent)
HURT: 20 (Natural Selection 2)
loops in affected programs: 4 -> 0
The hurt programs appear to no longer have a constarray uniform, as
all constants were successfully propagated. Apparently before this
patch, we successfully unrolled a loop containing array access, but
only after promoting constant arrays to uniforms. With this patch,
we unroll it first, so all array access is direct, and the array
is split up, and individual constants are propagated. This seems
better.
Cc: mesa-stable@lists.freedesktop.org
Reported-by: Karol Herbst <nouveau@karolherbst.de>
Signed-off-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Timothy Arceri <timothy.arceri@collabora.com>
There's really no point in looking at ir_dereference_array of a
constant. It also misses cases like:
(assign () (var_ref tmp) (constant (array ...) ...))
No changes in shader-db, but keeps it working after the next commit.
Cc: mesa-stable@lists.freedesktop.org
Signed-off-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Timothy Arceri <timothy.arceri@collabora.com>
The new uniform may need precise as well.
Fixes copy propagation of constant array uniforms in Tomb Raider shaders.
Cc: mesa-stable@lists.freedesktop.org
Signed-off-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Timothy Arceri <timothy.arceri@collabora.com>
Previously, we failed to split constant arrays. Code such as
int[2] numbers = int[](1, 2);
would generates a whole-array assignment:
(assign () (var_ref numbers)
(constant (array int 4) (constant int 1) (constant int 2)))
opt_array_splitting generally tried to visit ir_dereference_array nodes,
and avoid recursing into the inner ir_dereference_variable. So if it
ever saw a ir_dereference_variable, it assumed this was a whole-array
read and bailed. However, in the above case, there's no array deref,
and we can totally handle it - we just have to "unroll" the assignment,
creating assignments for each element.
This was mitigated by the fact that we constant propagate whole arrays,
so a dereference of a single component would usually get the desired
single value anyway. However, I plan to stop doing that shortly;
early experiments with disabling constant propagation of arrays
revealed this shortcoming.
This patch causes some arrays in Gl32GSCloth's geometry shaders to be
split, which allows other optimizations to eliminate unused GS inputs.
The VS then doesn't have to write them, which eliminates the entire VS
(5 -> 2 instructions). It still renders correctly.
No other change in shader-db.
v2: Drop !AOA check and improve a comment (feedback from Tim Arceri).
Cc: mesa-stable@lists.freedesktop.org
Signed-off-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Timothy Arceri <timothy.arceri@collabora.com>
opt_constant_propagation.cpp contains constant folding code which can
actually do constant propagation in some cases. It was happily
propagating constants into the left-hand-side of assignments.
For example,
(assign () (var_ref temp) (constant ...))
would brilliantly be turned into:
(assign () (constant ...) (constant ....))
This is a bigger hammer than necessary - it prevents propagation
into the left-hand-side altogether. We could certainly do better
someday. Notably, the constant propagation pass itself already
takes this approach - it's just the constant propagation pass's
built-in constant folding code (which actually propagates, too)
that was broken.
No change in shader-db, but prevents regressions after future commits.
It seems plausible that this could be hit today, but I haven't seen it
happen.
Cc: mesa-stable@lists.freedesktop.org
Signed-off-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Timothy Arceri <timothy.arceri@collabora.com>
We already store these in gl_shader and gl_program here we
remove it from gl_shader_program and just use the values
from gl_shader.
This will allow us to keep the shader cache restore code as
simple as it can be while making it somewhat clearer where these
values originate from.
Reviewed-by: Iago Toral Quiroga <itoral@igalia.com>
We already store this in gl_shader and gl_program here we
remove it from gl_shader_program and just use the values
from gl_shader.
This will allow us to keep the shader cache restore code as
simple as it can be while making it somewhat clearer where these
values originate from.
V2: remove unnecessary NULL check
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Reviewed-by: Iago Toral <itoral@igalia.com>
There's special logic around finding gl_FragData. It latches onto any
array with FRAG_RESULT_DATA0. However gl_SecondaryFragDataEXT[], added
by GL_EXT_blend_func_extended, fits those parameters as well. The real
frag data array should have index 0 though, so we can use that to
distinguish them.
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=96617
Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu>
Cc: "11.1 11.2 12.0" <mesa-stable@lists.freedesktop.org>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
SPIR-V treats it as an input but NIR wants the system value. This
shouldn't have been too much of a surprise given that we have to do the
same conversion in the GLSL IR to NIR pass.
Signed-off-by: Jason Ekstrand <jason@jlekstrand.net>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Cc: "12.0" <mesa-stable@lists.freedesktop.org>
Just setting builder->exact isn't sufficient because that only applies to
instructions that are built with the builder but instructions created
manually and only inserted using the builder are left alone.
Signed-off-by: Jason Ekstrand <jason@jlekstrand.net>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Cc: "12.0" <mesa-stable@lists.freedesktop.org>
This pass is similar to propagate_invariance in the GLSL compiler. The
real "output" of this pass is that any algebraic operations which are
eventually consumed by an invariant variable get marked as "exact".
Signed-off-by: Jason Ekstrand <jason@jlekstrand.net>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Cc: "12.0" <mesa-stable@lists.freedesktop.org>
While mathematically correct, these two optimizations result in an
expression with substantially lower precision than the original. For any
positive finite floating-point value, log2(x) is well-defined and finite.
More precisely, it is in the range [-150, 150] so any sum of logarithms
log2(a) + log2(b) is also well-defined and finite as long as a and b are
both positive and finite. However, if a and b are either very small or
very large, their product may get flushed to infinity or zero causing
log2(a * b) to be nowhere close to log2(a) + log2(b).
This imprecision was causing incorrect rendering in Talos Principal because
part of its HDR rendering process involves doing 8 texture operations,
clamping the result to [0, 65000], taking a dot-product with a constant,
and then taking the log2. This is done 6 or 8 times and summed to produce
the final result which is written to a red texture. In cases where you
have a region of the screen that is very dark, it can end up getting a
result value of -inf which is not what is intended.
Signed-off-by: Jason Ekstrand <jason@jlekstrand.net>
Reviewed-by: Matt Turner <mattst88@gmail.com>
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=96425
Cc: "11.1 11.2 12.0" <mesa-stable@lists.freedesktop.org>
Previously some callers of precision_qualifier_allowed would strip the
arrayness from the type and some would not. As a result, some places
would not notice that float[6], for example, needed a precision
qualifier.
Fixes the new piglit test no-default-float-array-precision.frag.
Signed-off-by: Ian Romanick <ian.d.romanick@intel.com>
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=96358
Cc: "12.0" <mesa-stable@lists.freedesktop.org>
Cc: Gregory Hainaut <gregory.hainaut@gmail.com>
Cc: Ilia Mirkin <imirkin@alum.mit.edu>
Reviewed-by: Timothy Arceri <timothy.arceri@collabora.com>
Instead use the internal gl_shader_stage enum everywhere. This
makes things more consistent and gets rid of unnecessary
conversions.
Ideally it would be nice to remove the Type field from gl_shader
altogether but currently it is used to differentiate between
gl_shader and gl_shader_program in the ShaderObjects hash table.
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
i965 has no special hardware for this, so the best way to implement
this is to pass it in via a uniform.
Signed-off-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Alejandro Piñeiro <apinheiro@igalia.com>
Cc: mesa-stable@lists.freedesktop.org
i965 has no special hardware for this, so we need to pass this value in
as a uniform (unless the TES is linked against a TCS, in which case the
linker can just replace this with a constant).
Signed-off-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Alejandro Piñeiro <apinheiro@igalia.com>
Cc: mesa-stable@lists.freedesktop.org
Built-in variable "MaxCombinedShaderStorageBlocks" was added to GLSL 4.40
revision 9.
Section "1.2.1 Changes since revision 8 of GLSL version 4.40",
page 3 of the PDF states:
"Bug 11734: Add gl_MaxCombinedShaderOutputResources and mark
gl_MaxCombinedImageUnitsAndFragmentOutputs as deprecated."
Fixes: GL44-CTS.shader_image_load_store.basic-glsl-const
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
