For some reason, we occasionally write the flag register with a MOV.NZ
instruction:
add(8) g25<1>F -g6<0,1,0>F g15<8,8,1>F
cmp.l.f0(8) g26<1>D g25<8,8,1>F 0F
mov.nz.f0(8) null g26<8,8,1>D
A MOV.NZ instruction on the result of a CMP is like comparing for
equality with true in C. It's useless. Removing it allows us to
generate:
add.l.f0(8) null -g6<0,1,0>F g15<8,8,1>F
total instructions in shared programs: 5955701 -> 5951657 (-0.07%)
instructions in affected programs: 302910 -> 298866 (-1.34%)
GAINED: 1
LOST: 0
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
This allows us to apply the optimization in cases where the CMP's
argument is negated, by flipping the conditional mod. For example, it
allows us to optimize this:
add(8) temp a b
cmp.l.f0(8) null -temp 0.0
into
add.g.f0(8) temp a b
total instructions in shared programs: 5958360 -> 5955701 (-0.04%)
instructions in affected programs: 466880 -> 464221 (-0.57%)
GAINED: 0
LOST: 1
Reviewed-by: Jason Ekstrand <jason.ekstrand@intel.com>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Otherwise we'll apply the conditional mod to only one of SIMD8
instructions and trigger an assertion.
NoDDClr/NoDDChk have the same problem but we never apply those to these
instructions, so I'm leaving them for a later time.
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
3-src instructions can only have GRF/MRF destinations. It's really
difficult to deal with that restriction in dead code elimination (that
wants to give instructions null destinations to show that their result
isn't used) while allowing 3-src instructions to have conditional mod,
so don't, and just give then a destination before register allocation.
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Now that we properly track accumulator dependencies, the scheduler is
able to schedule instructions between the mach and mov in the common
the integer multiplication pattern:
mul acc0, x, y
mach null, x, y
mov dest, acc0
Since a null destination implies no dependency on the destination, we
can also safely schedule instructions (that don't write the accumulator)
between the mul and mach.
GAINED: 103
LOST: 43
Causes one program to spill (643 -> 1076 instructions).
I committed this patch last year (commit 42a26cb5) but reverted it
(commit 0d3f83f4) after inexplicable artifacts in Kerbal Space Program
(bug 78648). Tapani reapplied this patch and could not reproduce the bug
with current Mesa.
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Jason Ekstrand <jason.ekstrand@intel.com>
If try_replace_with_sel is able to replace the flow control with a SEL
instruction, then there is no flow control... failing SIMD16 because
of nonexistent flow control is wrong.
No piglit regressions on any i965 platform in Jenkins.
total instructions in shared programs: 4382707 -> 4382707 (0.00%)
instructions in affected programs: 0 -> 0
helped: 0
HURT: 0
GAINED: 2089
LOST: 0
No other platforms affected in shader-db.
Signed-off-by: Ian Romanick <ian.d.romanick@intel.com>
Reviewed-by: Matt Turner <mattst88@gmail.com>
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
It's nice to have this present in your default cases so you can see what
instruction is triggering an abort.
v2: Just pass a NULL state, now that it won't crash when you do.
Reviewed-by: Jason Ekstrand <jason.ekstrand@intel.com>
This is the equivalent of brw_fs_channel_expressions.cpp, which I wanted
for vc4.
v2: Use the nir_src_for_ssa() helper, and another instance of
nir_alu_src_copy().
v3: Drop the non-SSA support. All intended callers will have SSA-only ALU
ops.
v4: Use insert_before, drop stale bcsel/fcsel comment, drop now-unused
unsupported() function, drop lower_context struct.
v5: Completely rename the pass to nir_lower_alu_to_scalar(), add an assert
about weird input_sizes[].
Reviewed-by: Jason Ekstrand <jason.ekstrand@iastate.edu>
There aren't many users yet, but I wanted to do this from my scalarizing
pass.
v2: Constify the src arguments.
Reviewed-by: Jason Ekstrand <jason.ekstrand@intel.com>
Most of these exist in the GLSL IR algebraic pass already. However,
SSA allows us to find more instances of the patterns.
total NIR instructions in shared programs: 2015593 -> 2011430 (-0.21%)
NIR instructions in affected programs: 124189 -> 120026 (-3.35%)
helped: 604
total i965 instructions in shared programs: 6025505 -> 6018717 (-0.11%)
i965 instructions in affected programs: 261295 -> 254507 (-2.60%)
helped: 1295
HURT: 3
Signed-off-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Jason Ekstrand <jason.ekstrand@intel.com>
Reviewed-by: Matt Turner <mattst88@gmail.com>
The first batch removes bonus fnot/inot operations, possibly allowing
other optimizations to better recognize patterns.
The next batch replaces a fadd and constant 0.0 with an fneg - negation
is usually free on GPUs, while addition is not.
total NIR instructions in shared programs: 2020814 -> 2015593 (-0.26%)
NIR instructions in affected programs: 411143 -> 405922 (-1.27%)
helped: 2233
HURT: 214
A few shaders are hurt by a few instructions due to moving neg such
that it has a constant operand, which is then folded, resulting in two
distinct load_consts for x and -x. We can always clean that up later.
total i965 instructions in shared programs: 6035392 -> 6025505 (-0.16%)
i965 instructions in affected programs: 784980 -> 775093 (-1.26%)
helped: 4508
HURT: 2
Signed-off-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Jason Ekstrand <jason.ekstrand@intel.com>
Reviewed-by: Matt Turner <mattst88@gmail.com>
The GLSL IR optimization pass contained these; we may as well include
them too.
v2: Fix a >> 0 and a << 0 optimizations (caught by Matt).
No change in the number of NIR instructions on a shader-db run.
total i965 instructions in shared programs: 6035397 -> 6035392 (-0.00%)
i965 instructions in affected programs: 542 -> 537 (-0.92%)
helped: 2 (in glamor)
Signed-off-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Jason Ekstrand <jason.ekstrand@intel.com>
Reviewed-by: Matt Turner <mattst88@gmail.com>
Matt and I noticed a bunch of "val <- ior a a" operations in a shader,
so we decided to add an algebraic optimization for that. While there,
I decided to add a bunch more of them.
v2: Delete bogus fand/for optimizations (caught by Jason).
total NIR instructions in shared programs: 2023511 -> 2020814 (-0.13%)
NIR instructions in affected programs: 149634 -> 146937 (-1.80%)
helped: 1032
total i965 instructions in shared programs: 6035392 -> 6035397 (0.00%)
i965 instructions in affected programs: 537 -> 542 (0.93%)
HURT: 2
Signed-off-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Jason Ekstrand <jason.ekstrand@intel.com>
Reviewed-by: Matt Turner <mattst88@gmail.com>
Matt and I noticed that one of the shaders hurt by INTEL_USE_NIR=1 had
load_input and load_uniform intrinsics repeated several times, with the
same parameters, but each one generating a distinct SSA value. This
made ALU operations on those values appear distinct as well.
Generating distinct SSA values is silly - these are read only variables.
CSE'ing them makes everything use a single SSA value, which then allows
other operations to be CSE'd away as well.
Generalizing a bit, it seems like we should be able to safely CSE any
intrinsics that can be eliminated and reordered. I didn't implement
support for variables for the time being.
v2: Assert that info->num_variables == 0 (requested by Jason).
total NIR instructions in shared programs: 2435936 -> 2023511 (-16.93%)
NIR instructions in affected programs: 2413496 -> 2001071 (-17.09%)
helped: 16872
total i965 instructions in shared programs: 6028987 -> 6008427 (-0.34%)
i965 instructions in affected programs: 640654 -> 620094 (-3.21%)
helped: 2071
HURT: 585
GAINED: 14
LOST: 25
Signed-off-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Jason Ekstrand <jason.ekstrand@intel.com>
This should not be a change in behavior, as all current cases that
potentially answer "yes" require SSA.
The next patch will introduce another case that requires SSA.
Signed-off-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Jason Ekstrand <jason.ekstrand@intel.com>
Reviewed-by: Matt Turner <mattst88@gmail.com>
This allows us to count NIR instructions via shader-db.
Use "run" as normal. The results file will contain both NIR and
assembly.
Then, to generate a NIR report:
./report.py <(grep NIR results/foo) <(grep NIR results/bar)
Or, to generate an i965 report:
./report.py <(grep -v NIR results/foo) <(grep -v NIR results/bar)
Signed-off-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Jason Ekstrand <jason.ekstrand@intel.com>
Reviewed-by: Matt Turner <mattst88@gmail.com>
This is useful for debugging and looking for optimization opportunities.
It will need to be expanded when we add support for other scalar stages.
Signed-off-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Matt Turner <mattst88@gmail.com>
We want to run CSE and algebraic optimizations again after lowering IO.
Some of the passes in the optimization loop don't handle saturates and
other modifiers, so run it before lowering to source modifiers.
total instructions in shared programs: 6046190 -> 6045768 (-0.01%)
instructions in affected programs: 22406 -> 21984 (-1.88%)
helped: 47
HURT: 0
GAINED: 0
LOST: 0
Signed-off-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Jason Ekstrand <jason.ekstrand@intel.com>
Apparently $(top_srcdir) is not expanded in a source list when using
subdir-objects, so remove that. It's not clear to me why we were going
to such lengths to prefix each source file anyway.
According to BSpec the correct number for min_vs_entries is 34 for CHV.
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Change max_wm_threads to match the spec on CHV. The max number of
threads in 3DSTATE_PS is always programmed to 64 and the hardware
internally scales that depending on the GT SKU. So this doesn't
change the max number of threads actually used, but it does affect
the scratch space calculation.
On CHV the old value was too small, so the amount of scratch space
allocated wasn't sufficient to satisfy the actual max number of
threads used.
Cc: mesa-stable@lists.freedesktop.org
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
When emitting texturing from indirect texture units, we need to be able to
scratch around in the header message. Since we only do this for >= HSW,
this is ok since there are no MRFs.
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Anuj phogat <anuj.phogat@gmail.com>
