The optimizations are only valid for 32-bit integers. They were
mistakenly firing for 64-bit integers as well.
Cc: mesa-stable@lists.freedesktop.org
Reviewed-by: Matt Turner <mattst88@gmail.com>
Two of the ARB_shader_ballot piglit tests hit the find_lsb case,
removing some of the noise allowed me to better debug the test when it
was failing.
Reviewed-by: Connor Abbott <cwabbott0@gmail.com>
This shuffles constants down in the reverse of what the previous
patch does and applies some simpilifications that may be made
possible from doing so.
Shader-db results BDW:
total instructions in shared programs: 12980814 -> 12977822 (-0.02%)
instructions in affected programs: 281889 -> 278897 (-1.06%)
helped: 1231
HURT: 128
total cycles in shared programs: 246562852 -> 246567288 (0.00%)
cycles in affected programs: 11271524 -> 11275960 (0.04%)
helped: 1630
HURT: 1378
V2: mark float opts as inexact
Reviewed-by: Elie Tournier <elie.tournier@collabora.com>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
V2: mark float opts as inexact
If one of the inputs to an mul/add is the result of another
mul/add there is a chance that we can reuse the result of that
mul/add in other calls if we do the multiplication in the right
order.
Also by attempting to move all constants to the top we increase
the chance of constant folding.
For example it is a fairly common pattern for shaders to do something
similar to this:
const float a = 0.5;
in vec4 b;
in float c;
...
b.x = b.x * c;
b.y = b.y * c;
...
b.x = b.x * a + a;
b.y = b.y * a + a;
So by simply detecting that constant a is part of the multiplication
in ffma and switching it with previous fmul that updates b we end up
with:
...
c = a * c;
...
b.x = b.x * c + a;
b.y = b.y * c + a;
Shader-db results BDW:
total instructions in shared programs: 13011050 -> 12967888 (-0.33%)
instructions in affected programs: 4118366 -> 4075204 (-1.05%)
helped: 17739
HURT: 1343
total cycles in shared programs: 246717952 -> 246410716 (-0.12%)
cycles in affected programs: 166870802 -> 166563566 (-0.18%)
helped: 18493
HURT: 7965
total spills in shared programs: 14937 -> 14560 (-2.52%)
spills in affected programs: 9331 -> 8954 (-4.04%)
helped: 284
HURT: 33
total fills in shared programs: 20211 -> 19671 (-2.67%)
fills in affected programs: 12586 -> 12046 (-4.29%)
helped: 286
HURT: 33
LOST: 39
GAINED: 33
Some of the hurt will go away when we shuffle things back down to the
bottom in the following patch. It's also noteworthy that almost all of the
spill changes are in Deus Ex both hurt and helped.
Reviewed-by: Elie Tournier <elie.tournier@collabora.com>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Didn't turn out as useful as I'd hoped, but it will help alot more on
i965 by reducing regressions when we drop brw_do_channel_expressions()
and brw_do_vector_splitting().
I'm not sure how much sense 'is_not_used_by_conditional' makes on
platforms other than i965 but since this is a new opt it at least
won't do any harm.
shader-db BDW:
total instructions in shared programs: 13029581 -> 13029415 (-0.00%)
instructions in affected programs: 15268 -> 15102 (-1.09%)
helped: 86
HURT: 0
total cycles in shared programs: 247038346 -> 247036198 (-0.00%)
cycles in affected programs: 692634 -> 690486 (-0.31%)
helped: 183
HURT: 27
Reviewed-by: Elie Tournier <elie.tournier@collabora.com>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
The NIR story on conversion opcodes is a mess. We've had way too many
of them, naming is inconsistent, and which ones have explicit sizes was
sort-of random. This commit re-organizes things and makes them all
consistent:
- All non-bool conversion opcodes now have the explicit size in the
destination and are named <src_type>2<dst_type><size>.
- Integer <-> integer conversion opcodes now only come in i2i and u2u
forms (i2u and u2i have been removed) since the only difference
between the different integer conversions is whether or not they
sign-extend when up-converting.
- Boolean conversion opcodes all have the explicit size on the bool and
are named <src_type>2<dst_type>.
Making things consistent also allows nir_type_conversion_op to be moved
to nir_opcodes.c and auto-generated using mako. This will make adding
int8, int16, and float16 versions much easier when the time comes.
Reviewed-by: Eric Anholt <eric@anholt.net>
NIR is a typeless IR and the two opcodes, when considered bitwise, do
exactly the same thing. There's no reason to have two versions.
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Previously both sources were unsized. This caused problems when the
thing being shifted was 64-bit but the shift count was 32-bit. The
expectation in NIR is that all unsized sources (and destination) will
ultimately have the same size.
The changes in nir_opt_algebraic.py are to prevent errors like:
Failed to parse transformation:
03:12:25 (('extract_i8', 'a', 'b'), ('ishr', ('ishl', 'a', ('imul', ('isub', 3, 'b'), 8)), 24), 'options->lower_extract_byte')
03:12:25 Traceback (most recent call last):
03:12:25 File "/home/jenkins/workspace/Leeroy_2/repos/mesa/src/compiler/nir/nir_algebraic.py", line 610, in __init__
03:12:25 xform = SearchAndReplace(xform)
03:12:25 File "/home/jenkins/workspace/Leeroy_2/repos/mesa/src/compiler/nir/nir_algebraic.py", line 495, in __init__
03:12:25 BitSizeValidator(varset).validate(self.search, self.replace)
03:12:25 File "/home/jenkins/workspace/Leeroy_2/repos/mesa/src/compiler/nir/nir_algebraic.py", line 311, in validate
03:12:25 validate_dst_class = self._validate_bit_class_up(replace)
03:12:25 File "/home/jenkins/workspace/Leeroy_2/repos/mesa/src/compiler/nir/nir_algebraic.py", line 414, in _validate_bit_class_up
03:12:25 src_class = self._validate_bit_class_up(val.sources[i])
03:12:25 File "/home/jenkins/workspace/Leeroy_2/repos/mesa/src/compiler/nir/nir_algebraic.py", line 420, in _validate_bit_class_up
03:12:25 assert src_class == src_type_bits
03:12:25 AssertionError
Signed-off-by: Ian Romanick <ian.d.romanick@intel.com>
Suggested-by: Connor Abbott <cwabbott0@gmail.com>
Reviewed-by: Connor Abbott <cwabbott0@gmail.com>
Cc: Jason Ekstrand <jason@jlekstrand.net>
Otherwise we will end up with an extra instruction to compare the
result of the inot.
On BDW:
total instructions in shared programs: 13060620 -> 13060481 (-0.00%)
instructions in affected programs: 103379 -> 103240 (-0.13%)
helped: 127
HURT: 0
total cycles in shared programs: 256590950 -> 256587408 (-0.00%)
cycles in affected programs: 11324730 -> 11321188 (-0.03%)
helped: 114
HURT: 21
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
We turn these from bcsel into inot/b2f combos in order for other
optimisation passes to get further. Once we have finished turn
the ones that remain and are used in more than a single expression
back into a bcsel.
On BDW:
total instructions in shared programs: 13060965 -> 13060297 (-0.01%)
instructions in affected programs: 835701 -> 835033 (-0.08%)
helped: 670
HURT: 2
total cycles in shared programs: 256599536 -> 256598006 (-0.00%)
cycles in affected programs: 114655488 -> 114653958 (-0.00%)
helped: 419
HURT: 240
LOST: 0
GAINED: 1
The 2 HURT is because inserting bcsel creates the only use of
const 1.0 in two shaders from tri-of-friendship-and-madness.
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
This sequence shows up The Talos Principal, at least under Vulkan,
and prevents loop analysis from properly computing trip counts in a
few loops.
Reviewed-by: Ian Romanick <ian.d.romanick@intel.com>
The previous power-of-two rules didn't catch idiv (because i965 doesn't
set lower_idiv) and imod cases. The udiv and umod cases should have
been caught, but I included them for orthogonality.
This fixes silly code observed from compute shaders with local_size_[xy]
= 1.
Signed-off-by: Ian Romanick <ian.d.romanick@intel.com>
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=98299
Reviewed-by: Jordan Justen <jordan.l.justen@intel.com>
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
According to Connor, it's safe to assume that the first operand of
bcsel, as well as the operand of b2f and b2i, must be well formed
booleans.
https://lists.freedesktop.org/archives/mesa-dev/2016-August/125658.html
With the previous improvements to a@bool handling, this now has no
change in shader-db instruction counts on Broadwell.
Signed-off-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Reviewed-by: Matt Turner <mattst88@gmail.com>
Some shaders include code that looks like:
uniform int i;
uniform vec4 bones[...];
foo(bones[i * 3], bones[i * 3 + 1], bones[i * 3 + 2]);
CSE would do some work on this:
x = i * 3
foo(bones[x], bones[x + 1], bones[x + 2]);
The compiler may then add '<< 4 + base' to the index calculations.
This results in expressions like
x = i * 3
foo(bones[x << 4], bones[(x + 1) << 4], bones[(x + 2) << 4]);
Just rearranging the math to produce (i * 48) + 16 saves an
instruction, and it allows CSE to do more work.
x = i * 48;
foo(bones[x], bones[x + 16], bones[x + 32]);
So, ~6 instructions becomes ~3.
Some individual shader-db results look pretty bad. However, I have a
really, really hard time believing the change in estimated cycles in,
for example, 3dmmes-taiji/51.shader_test after looking that change in
the generated code.
G45
total instructions in shared programs: 4020840 -> 4010070 (-0.27%)
instructions in affected programs: 177460 -> 166690 (-6.07%)
helped: 894
HURT: 0
total cycles in shared programs: 98829000 -> 98784990 (-0.04%)
cycles in affected programs: 3936648 -> 3892638 (-1.12%)
helped: 894
HURT: 0
Ironlake
total instructions in shared programs: 6418887 -> 6408117 (-0.17%)
instructions in affected programs: 177460 -> 166690 (-6.07%)
helped: 894
HURT: 0
total cycles in shared programs: 143504542 -> 143460532 (-0.03%)
cycles in affected programs: 3936648 -> 3892638 (-1.12%)
helped: 894
HURT: 0
Sandy Bridge
total instructions in shared programs: 8357887 -> 8339251 (-0.22%)
instructions in affected programs: 432715 -> 414079 (-4.31%)
helped: 2795
HURT: 0
total cycles in shared programs: 118284184 -> 118207412 (-0.06%)
cycles in affected programs: 6114626 -> 6037854 (-1.26%)
helped: 2478
HURT: 317
Ivy Bridge
total instructions in shared programs: 7669390 -> 7653822 (-0.20%)
instructions in affected programs: 388234 -> 372666 (-4.01%)
helped: 2795
HURT: 0
total cycles in shared programs: 68381982 -> 68263684 (-0.17%)
cycles in affected programs: 1972658 -> 1854360 (-6.00%)
helped: 2458
HURT: 307
Haswell
total instructions in shared programs: 7082636 -> 7067068 (-0.22%)
instructions in affected programs: 388234 -> 372666 (-4.01%)
helped: 2795
HURT: 0
total cycles in shared programs: 68282020 -> 68164158 (-0.17%)
cycles in affected programs: 1891820 -> 1773958 (-6.23%)
helped: 2459
HURT: 261
Broadwell
total instructions in shared programs: 9002466 -> 8985875 (-0.18%)
instructions in affected programs: 658784 -> 642193 (-2.52%)
helped: 2795
HURT: 5
total cycles in shared programs: 78503092 -> 78450404 (-0.07%)
cycles in affected programs: 2873304 -> 2820616 (-1.83%)
helped: 2275
HURT: 415
Skylake
total instructions in shared programs: 9156978 -> 9140387 (-0.18%)
instructions in affected programs: 682625 -> 666034 (-2.43%)
helped: 2795
HURT: 5
total cycles in shared programs: 75591392 -> 75550574 (-0.05%)
cycles in affected programs: 3192120 -> 3151302 (-1.28%)
helped: 2271
HURT: 425
Signed-off-by: Ian Romanick <ian.d.romanick@intel.com>
Reviewed-by: Thomas Helland <thomashelland90@gmail.com>
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
I noticed this when I tried to do frexp(float(some_unsigned)) in the
ir_unop_find_lsb lowering pass. The code generated for frexp() uses
fabs, and this resulted in an extra instruction. Ultimately I ended up
not using frexp.
Signed-off-by: Ian Romanick <ian.d.romanick@intel.com>
Reviewed-by: Matt Turner <mattst88@gmail.com>
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>
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>
Some optimizations, like converting integer multiply/divide into left/
right shifts, have additional constraints on the search expression.
Like requiring that a variable is a constant power of two. Support
these cases by allowing a fxn name to be appended to the search var
expression (ie. "a#32(is_power_of_two)").
Signed-off-by: Rob Clark <robclark@freedesktop.org>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
The i965 driver has its own pass for fusing mul+add combinations that's
much smarter than what nir_opt_algebraic can do so we don't want to get the
nir_opt_algebraic one just because we didn't set lower_ffma.
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Split 32-bit and 64-bit fmod lowering as the drivers might need to
lower them separately inside NIR depending on the HW support.
Signed-off-by: Samuel Iglesias Gonsálvez <siglesias@igalia.com>
Reviewed-by: Jordan Justen <jordan.l.justen@intel.com>
There is no sense in having the double version of ldexp take a 64-bit
integer. Instead, let's just take a 32-bit int all the time. This also
matches what GLSL does where both variants of ldexp take a regular integer
for the exponent argument.
Reviewed-by: Samuel Iglesias Gonsálvez <siglesias@igalia.com>
Some hardware (i965 on Broadwell generation, for example) does not support
natively the execution of lrp instruction with double arguments.
Add 'lower_flrp64' flag to lower this instruction in that case.
v2:
- Rename lower_flrp_double to lower_flrp64 (Jason)
- Fix typo (Jason)
- Adapt the code to define bit_size information in the opcodes.
Signed-off-by: Samuel Iglesias Gonsálvez <siglesias@igalia.com>
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
A later patch will add lower_flrp64 option to NIR.
Signed-off-by: Samuel Iglesias Gonsálvez <siglesias@igalia.com>
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
The algorithm used is different from both the naive suggestion from the
GLSL spec and the one used in GLSL IR today. Unfortunately, the GLSL IR
implementation that we have today doesn't handle denormals (for those that
care) or the case where the float source is +-inf.
Reviewed-by: Matt Turner <mattst88@gmail.com>
Dolphin uses them a lot. Range tracking would be better in the long term,
but this two lines works fine for now.
Signed-off-by: Markus Wick <markus@selfnet.de>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Many shaders contain expression trees of the form:
const_1 * (value * const_2)
Reorganizing these to
(const_1 * const_2) * value
will allow constant folding to combine the constants. Sometimes, these
constants are 2 and 0.5, so we can remove a multiply altogether. Other
times, it can create more immediate constants, which can actually hurt.
Finding a good balance here is tricky. While much more could be done,
this simple patch seems to have a lot of positive benefit while having
a low downside.
shader-db results on Broadwell:
total instructions in shared programs: 8963768 -> 8961369 (-0.03%)
instructions in affected programs: 438318 -> 435919 (-0.55%)
helped: 1502
HURT: 245
total cycles in shared programs: 71527354 -> 71421516 (-0.15%)
cycles in affected programs: 11541788 -> 11435950 (-0.92%)
helped: 3445
HURT: 1224
Signed-off-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Eduardo Lima Mitev <elima@igalia.com>
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Reviewed-by: Matt Turner <mattst88@gmail.com>
Oddly, this did not affect the shader where I first noticed the pattern.
That particular shader doesn't get its if-statement converted to a bcsel
because there are two assignments in the else-statement. This led to me
submitting https://bugs.freedesktop.org/show_bug.cgi?id=94747.
shader-db results:
Sandy Bridge
total instructions in shared programs: 8467384 -> 8467069 (-0.00%)
instructions in affected programs: 36594 -> 36279 (-0.86%)
helped: 46
HURT: 0
total cycles in shared programs: 117573448 -> 117568518 (-0.00%)
cycles in affected programs: 339114 -> 334184 (-1.45%)
helped: 46
HURT: 0
Ivy Bridge / Haswell / Broadwell / Skylake:
total instructions in shared programs: 7774258 -> 7773999 (-0.00%)
instructions in affected programs: 30874 -> 30615 (-0.84%)
helped: 46
HURT: 0
total cycles in shared programs: 65739190 -> 65734530 (-0.01%)
cycles in affected programs: 180380 -> 175720 (-2.58%)
helped: 45
HURT: 1
No change on G45 or Ironlake.
I also tried these expressions, but none of them affected any shaders in
shader-db:
(('bcsel', a, 'a@bool', 'b@bool'), ('ior', a, b)),
(('bcsel', a, 'b@bool', False), ('iand', a, b)),
(('bcsel', a, 'b@bool', 'a@bool'), ('iand', a, b)),
Signed-off-by: Ian Romanick <ian.d.romanick@intel.com>
Reviewed-by: Iago Toral Quiroga <itoral@igalia.com>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Many of our optimizations, while great for cutting shaders down to size,
aren't really precision-safe. This commit tries to flag all of the
inexact floating-point optimizations so they don't get run on values that
are flagged "exact". It's a bit conservative and maybe flags some safe
optimizations as unsafe but that's better than missing one.
Reviewed-by: Francisco Jerez <currojerez@riseup.net>
The previous transformation got the arguments to fmin backwards. When NaNs
are involved, the GLSL min/max aren't commutative so it matters.
Reviewed-by: Francisco Jerez <currojerez@riseup.net>
The fxor opcode is required to return 1.0f or 0.0f but the input variable
may not be 1.0f or 0.0f.
Reviewed-by: Francisco Jerez <currojerez@riseup.net>
No shader-db changes, but this is symmetric with the previous commit.
Signed-off-by: Ian Romanick <ian.d.romanick@intel.com>
Reviewed-by: Matt Turner <mattst88@gmail.com>