This isn't something I expect to occur often. However, we have always
reported that array-ctors with constant arguments are compile-time
constants, and `compareConstant` is supposed to be implemented for all
compile-time constants.
Change-Id: I0bbfe2a3f78c29c14f69c3b6faca71067a0e45c8
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/391116
Commit-Queue: John Stiles <johnstiles@google.com>
Auto-Submit: John Stiles <johnstiles@google.com>
Reviewed-by: Brian Osman <brianosman@google.com>
We don't directly support this today at all. In practice, though, simple
constant arrays are detected as equal in the constant-folding pass
because they hit the `x == x` self-equality check (using
`IsSameExpressionTree`).
This does not work for our inequality tests, though, so those do not
fold.
Change-Id: I6730a9a2d1da9ac613ee58889d651f3ff65b1d2d
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/391057
Commit-Queue: John Stiles <johnstiles@google.com>
Commit-Queue: Brian Osman <brianosman@google.com>
Auto-Submit: John Stiles <johnstiles@google.com>
Reviewed-by: Brian Osman <brianosman@google.com>
As soon as a single VarDeclaration is successfully created, its Variable
is added to the current symbol table. However, if a variable-declaration
line declared several variables in a row, we would stop if ANY of the
declarations contained an error and discard the entire statement, but
would continue processing the rest of the program. This left us in a
position where some Variables existed in the SymbolTable with valid,
reachable names, but their corresponding VarDeclaration statement had
been thrown away as erroneous. Since Variables point back to
VarDeclarations for their initialValues, this gave us a stale pointer.
Any future reference to that variable name which could trigger an
access to its initialValue would read from this dead pointer.
This CL fixes the conversion of VarDeclarations so that we no longer
throw away any VarDeclarations associated with a successfully-parsed
Variable.
Change-Id: If8ec3c160933e48a0e1f36414234b3a849d8978c
Bug: oss-fuzz:32587
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/389636
Commit-Queue: John Stiles <johnstiles@google.com>
Commit-Queue: Brian Osman <brianosman@google.com>
Auto-Submit: John Stiles <johnstiles@google.com>
Reviewed-by: Brian Osman <brianosman@google.com>
This is particularly relevant for "vector / scalar" divides, as these
end up generating an extra OpFDiv to take the reciprocal of the scalar.
This meant we were generating code to calculate "literal 1 / literal N"
at runtime whenever dividing vector V against scalar N. This fix moves
the reciprocal calculation to compile-time. And of course, it's a valid
strength reduction in all cases.
Change-Id: I0986057e516120fe872e70135f1ac365155d5377
Bug: skia:11792
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/389156
Commit-Queue: John Stiles <johnstiles@google.com>
Auto-Submit: John Stiles <johnstiles@google.com>
Reviewed-by: Ethan Nicholas <ethannicholas@google.com>
Due to missing type-checks, we would emit `OpVectorTimesScalar` when
multiplying an ivec with an int, or an `OpFDiv` when dividing an ivec
against an int.
Change-Id: Idc214dbe0ec208cb44f28b22e585584ac2ab7dae
Bug: skia:11267, skia:11788
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/388742
Commit-Queue: John Stiles <johnstiles@google.com>
Auto-Submit: John Stiles <johnstiles@google.com>
Reviewed-by: Ethan Nicholas <ethannicholas@google.com>
We had constant-folding tests for vector-scalar arithmetic, but didn't
have an equivalent SkSL unit test for vector-scalar arithmetic that
actually needs to be computed at runtime.
This exposes two SPIR-V bugs: one was previously known, but the other
is a new discovery.
Change-Id: I28737128f20b445797c6c29872335d05f94cc95c
Bug: skia:11267, skia:11788
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/388739
Auto-Submit: John Stiles <johnstiles@google.com>
Commit-Queue: Ethan Nicholas <ethannicholas@google.com>
Reviewed-by: Ethan Nicholas <ethannicholas@google.com>
Previously, these tests were never actually executed, only read during
code review. They are now properly tested for correctness whenever dm
is run. Non-ES2 compliant statements (do/while/switch) are unfortunately
excluded here, as they are not compatible with Runtime Effects yet.
Change-Id: I965c782baad6f8dd3961a400ae791fb2c1f844d3
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/389296
Auto-Submit: John Stiles <johnstiles@google.com>
Commit-Queue: Ethan Nicholas <ethannicholas@google.com>
Reviewed-by: Ethan Nicholas <ethannicholas@google.com>
If sk_Caps.rewriteMatrixVectorMultiply is false, we generate the same
code as before.
If sk_Caps.rewriteMatrixVectorMultiply is true, and optimization is
enabled, medium-precision multiplication of `matrix * vector` is
rewritten in BinaryExpression as the sum of
(m[0]*v[0] + m[1]*v[1] + ... + m[N]*v[N]).
Change-Id: I9796158fd4203b5cd2278c0d0e9a2e8cda82b136
Bug: skia:11769
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/388460
Reviewed-by: Brian Osman <brianosman@google.com>
Commit-Queue: John Stiles <johnstiles@google.com>
On ARM GPUs, the Vulkan driver does not honor relaxed precision when
evaluating SpvOpMatrixTimesVector. This leads to reduced performance
(compared to GLSL, where mediump matrices and floats do evaluate all
intermediate values at mediump).
This caps bit will be enabled on ARM GPUs and, in a followup CL, will
be used to toggle a workaround where `m*v` is rewritten as the sum of
(m[0]*v[0] + m[1]*v[1] + ... + m[N]*v[N]).
Change-Id: I310fa73639b6498552c9672e76860f2eded15d0a
Bug: skia:11769
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/388459
Auto-Submit: John Stiles <johnstiles@google.com>
Reviewed-by: Brian Osman <brianosman@google.com>
Commit-Queue: John Stiles <johnstiles@google.com>
This reverts commit 7bba1f55e8.
Change-Id: I707a3c215f37376086e22eaa43916afeed6da4c7
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/388456
Reviewed-by: Brian Osman <brianosman@google.com>
Commit-Queue: Ethan Nicholas <ethannicholas@google.com>
Change-Id: I22837a4921238749664217e595d24d196503534d
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/388096
Reviewed-by: Robert Phillips <robertphillips@google.com>
Commit-Queue: Brian Osman <brianosman@google.com>
Removed error tests which no longer test anything:
- UseWithoutInitialize...: we no longer track variable use-before-init.
- InlineDivideByZero: we no longer constant-fold the result of an
inlined function call into its parent statement.
- Unreachable: we no longer report unreachable statements.
And fixed some minor test-case issues:
- StaticSwitchConditionalBreak: we still check this, but the function
was being dead-stripped before this check could run. Renamed to main.
- OssFuzzXxxxx: these cases no longer report errors, but they are still
valuable as regression tests; moved to `shared/`.
Change-Id: Iade3cff821dc998cacfd02f62d3ac4625e48904c
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/387820
Auto-Submit: John Stiles <johnstiles@google.com>
Reviewed-by: Brian Osman <brianosman@google.com>
Commit-Queue: Brian Osman <brianosman@google.com>
SPIR-V doesn't seem to mind overlapping function names, since they're
not load-bearing in any way, but this keeps us consistent with the other
code generators.
Change-Id: Ifdb4cb17795da88eabc0db841af746fb76caf423
Bug: skia:10851
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/387757
Commit-Queue: John Stiles <johnstiles@google.com>
Auto-Submit: John Stiles <johnstiles@google.com>
Reviewed-by: Ethan Nicholas <ethannicholas@google.com>
This will be implemented in Metal and SPIR-V in followup CLs.
Change-Id: I397b4db40b15dd54cf1d8a17f414c3fe184b48d2
Bug: skia:10851
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/387638
Auto-Submit: John Stiles <johnstiles@google.com>
Commit-Queue: John Stiles <johnstiles@google.com>
Reviewed-by: Brian Osman <brianosman@google.com>
Change-Id: If5fb4f99d327bb429f60e8d6c526720dd02b0928
Bug: skia:11342
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/386800
Auto-Submit: John Stiles <johnstiles@google.com>
Reviewed-by: Brian Osman <brianosman@google.com>
Commit-Queue: Brian Osman <brianosman@google.com>
This saves a significant amount of CPU time and, now that the inliner
can handle nested expressions, still inlines almost everything.
Change-Id: I8f198630fa9627bc433ef8fb72f6bcf94595cdaa
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/386917
Commit-Queue: John Stiles <johnstiles@google.com>
Auto-Submit: John Stiles <johnstiles@google.com>
Reviewed-by: Ethan Nicholas <ethannicholas@google.com>
This will allow the inliner to successfully do more work in a single
pass.
Change-Id: I26e8831737c10bdf9a35eebd94ea8b74f6487077
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/386916
Reviewed-by: Ethan Nicholas <ethannicholas@google.com>
Commit-Queue: Ethan Nicholas <ethannicholas@google.com>
Auto-Submit: John Stiles <johnstiles@google.com>
Arguments without side-effects that aren't read from more than once can
be moved directly into the inlined function, and don't need a scratch
variable. This can allow functions like `guarded_divide` to inline
completely in more cases.
Change-Id: I0bfce35635cf9779f4af1bc0790da966ccfe4230
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/386678
Reviewed-by: Ethan Nicholas <ethannicholas@google.com>
Commit-Queue: Ethan Nicholas <ethannicholas@google.com>
Auto-Submit: John Stiles <johnstiles@google.com>
This reverts commit a9c187e5cc.
Change-Id: Icbfb8abdfc67fc2e6428d97a6cdede2726fb56e4
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/385596
Auto-Submit: Ethan Nicholas <ethannicholas@google.com>
Commit-Queue: John Stiles <johnstiles@google.com>
Reviewed-by: John Stiles <johnstiles@google.com>
It is difficult to do this both efficiently and correctly while honoring
GLSL semantics (which require the lvalues to be kept distinct, even when
they point to the same variable). We could make it work by making copies
of every out parameter in each direction (going in for inouts, and
coming out for outs and inouts).
However, this could be self-defeating if it makes it harder for the
driver to track variable lifetimes. Simply opting out of inlining these
functions entirely seems like the best tradeoff; let the driver optimize
them if it can, and we can enjoy reduced complexity in the SkSL inliner.
Change-Id: I62f7b4550cc181cfe789e4f2ff4e408ba1baf9cb
Bug: skia:11326
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/370257
Reviewed-by: Brian Osman <brianosman@google.com>
Reviewed-by: Ethan Nicholas <ethannicholas@google.com>
Commit-Queue: John Stiles <johnstiles@google.com>
It turns out it is not legal to pass the results of OpAccessChain as a
function argument, for... reasons. This CL switches us over to passing
the argument via a temp variable instead.
Bug: skia:11748
Change-Id: Ib5e86c1d000655ebd7bb62ceea6a27b823808645
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/385936
Reviewed-by: John Stiles <johnstiles@google.com>
Commit-Queue: Ethan Nicholas <ethannicholas@google.com>
This allows us to remove 100 LOC from the inliner and is very unlikely
to affect any existing benchmark. We don't have any evidence to support
the idea that a one-iteration `for` loop with `continue`-based exits
will be any faster than a standard function call on any existing GPU.
Our fragment processors are generally written to avoid early returns,
in large part to avoid hitting this path.
This drastically impacts BlendEnum.sksl (which can no longer flatten out
a switch over every blend function in SkSL) but is otherwise a wash.
See: http://go/optimization-in-sksl-inliner suggestion 4(a)
Change-Id: I1f9c27bcd7a8de46cc4e8d0b9768d75957cf1c50
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/385377
Reviewed-by: Ethan Nicholas <ethannicholas@google.com>
Commit-Queue: Ethan Nicholas <ethannicholas@google.com>
Auto-Submit: John Stiles <johnstiles@google.com>
This put the coverage for do-while loops on par with for loops.
Change-Id: I53e0d733edd02a6a139792a8d74c68116453e5ff
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/385500
Auto-Submit: John Stiles <johnstiles@google.com>
Reviewed-by: Brian Osman <brianosman@google.com>
Commit-Queue: John Stiles <johnstiles@google.com>
This can eliminate const variables which have been completely folded
away, unnecessary synthetic variables created during codegen/inlining,
or code that simply didn't need to exist at all.
Change-Id: I37a65e455e6527a6a6c2f4dde918f48d84dc2638
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/383496
Commit-Queue: John Stiles <johnstiles@google.com>
Reviewed-by: Ethan Nicholas <ethannicholas@google.com>
A global variable should be considered "dead" if it's never written and
never read. The previous code checked if it was never written OR never
read, which is not the same.
This would generate GLSL/Metal that didn't compile. In SPIR-V, it would
SkASSERT, then crash, during codegen. The fuzzer was able to detect the
SPIR-V issue, but it was wrong in all three cases.
Change-Id: Id59a2499eb5baa3839b93826bfbc24191bfd490b
Bug: oss-fuzz:32005
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/385280
Commit-Queue: John Stiles <johnstiles@google.com>
Commit-Queue: Brian Osman <brianosman@google.com>
Auto-Submit: John Stiles <johnstiles@google.com>
Reviewed-by: Brian Osman <brianosman@google.com>
This reverts commit e7a8f85e4f.
Reason for revert: must revert dependent CL
Original change's description:
> Only include header once in combined MSL shader.
>
> Bug: skia:11389
> Change-Id: I3e24dcaa2cfeddc7efd7985f9f42a59bfc8175f2
> Reviewed-on: https://skia-review.googlesource.com/c/skia/+/385137
> Reviewed-by: John Stiles <johnstiles@google.com>
> Commit-Queue: Jim Van Verth <jvanverth@google.com>
TBR=jvanverth@google.com,johnstiles@google.com
Change-Id: I7a886b6c57a666e54e65365e41dcb57bd9ab4ba6
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Bug: skia:11389
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/385237
Reviewed-by: Michael Ludwig <michaelludwig@google.com>
Commit-Queue: Michael Ludwig <michaelludwig@google.com>
'in' variables without locations aren't allowed. Use uniforms instead.
Bug: skia:11738
Change-Id: Ic066106deb7409cff154b4be7cfb3e03a7025c7d
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/385000
Commit-Queue: Brian Osman <brianosman@google.com>
Reviewed-by: John Stiles <johnstiles@google.com>
This reverts commit 92748af1a5.
Reason for revert: SkSLCommaSideEffects_GPU crashing on Android
Original change's description:
> Inline functions of the form 'return (expr)' only.
>
> This drastically reduces the number of functions which we allow to be
> inlined. If this change does not hurt our performance, it will allow us
> to trivially remove hundreds of LOC. All current data leads us to
> believe that it may affect the Mali 400 but is highly unlikely to change
> results on any other device in the tree.
>
> More info: http://go/optimization-in-sksl-inliner
>
> Change-Id: Ia6b706742ce5407453e0e697b6c1f9201084c0e8
> Reviewed-on: https://skia-review.googlesource.com/c/skia/+/384858
> Auto-Submit: John Stiles <johnstiles@google.com>
> Commit-Queue: John Stiles <johnstiles@google.com>
> Reviewed-by: Brian Osman <brianosman@google.com>
> Reviewed-by: Ethan Nicholas <ethannicholas@google.com>
TBR=brianosman@google.com,ethannicholas@google.com,johnstiles@google.com
Change-Id: I6a670dacaa58fe3386ff50375ac6d1cac4fd7f2c
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/385161
Reviewed-by: John Stiles <johnstiles@google.com>
Commit-Queue: John Stiles <johnstiles@google.com>
Bug: skia:11389
Change-Id: I3e24dcaa2cfeddc7efd7985f9f42a59bfc8175f2
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/385137
Reviewed-by: John Stiles <johnstiles@google.com>
Commit-Queue: Jim Van Verth <jvanverth@google.com>
This reverts commit 9e476b798f.
Reason for revert: Angry Vulkan bots
Original change's description:
> Refactored SPIR-V RelaxedPrecision handling
>
> The RelaxedPrecision decoration is now handled by nextId(), to make it
> easier to see all spots where a RelaxedPrecision decoration might be
> necessary. The goal of this initial refactor is not to actually fix the
> issues with RelaxedPrecision decorations, but rather to lay the
> groundwork for doing so in followup CLs.
>
> The initial intent of this change was to not affect the SPIR-V at all,
> saving modifications for followups, but there ended up being three kinds
> of changes to the output:
>
> 1. Doing things at nextId() time rather than later means some
> decorations move to an earlier spot in the output. This results in
> diffs, but should not cause any behavioral changes.
> 2. We were incorrectly tagging bools as RelaxedPrecision in some
> situations. By funneling things through fewer code paths, the refactor
> would have caused this to happen in even more situations, and the code
> responsible for the bug was being rewritten in this CL anyway, so it
> seemed worth just fixing the issue as part of this change.
> 3. Funneling things through fewer code paths ended up adding
> (correct) RelaxedPrecision modifiers to binary operations that had
> previously been missing them. It seemed better to just let it happen
> than to try to maintain bug-for-bug compatibility with the previous
> approach.
>
> Change-Id: Ia9654d6b5754e2c797e02226660cb618c9189b36
> Reviewed-on: https://skia-review.googlesource.com/c/skia/+/384318
> Commit-Queue: Ethan Nicholas <ethannicholas@google.com>
> Reviewed-by: Brian Osman <brianosman@google.com>
TBR=brianosman@google.com,ethannicholas@google.com,johnstiles@google.com
Change-Id: I9ada728e5fd5798bc1179640560c2e6045b7efd1
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/385158
Reviewed-by: Ethan Nicholas <ethannicholas@google.com>
Commit-Queue: Ethan Nicholas <ethannicholas@google.com>
The RelaxedPrecision decoration is now handled by nextId(), to make it
easier to see all spots where a RelaxedPrecision decoration might be
necessary. The goal of this initial refactor is not to actually fix the
issues with RelaxedPrecision decorations, but rather to lay the
groundwork for doing so in followup CLs.
The initial intent of this change was to not affect the SPIR-V at all,
saving modifications for followups, but there ended up being three kinds
of changes to the output:
1. Doing things at nextId() time rather than later means some
decorations move to an earlier spot in the output. This results in
diffs, but should not cause any behavioral changes.
2. We were incorrectly tagging bools as RelaxedPrecision in some
situations. By funneling things through fewer code paths, the refactor
would have caused this to happen in even more situations, and the code
responsible for the bug was being rewritten in this CL anyway, so it
seemed worth just fixing the issue as part of this change.
3. Funneling things through fewer code paths ended up adding
(correct) RelaxedPrecision modifiers to binary operations that had
previously been missing them. It seemed better to just let it happen
than to try to maintain bug-for-bug compatibility with the previous
approach.
Change-Id: Ia9654d6b5754e2c797e02226660cb618c9189b36
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/384318
Commit-Queue: Ethan Nicholas <ethannicholas@google.com>
Reviewed-by: Brian Osman <brianosman@google.com>
This drastically reduces the number of functions which we allow to be
inlined. If this change does not hurt our performance, it will allow us
to trivially remove hundreds of LOC. All current data leads us to
believe that it may affect the Mali 400 but is highly unlikely to change
results on any other device in the tree.
More info: http://go/optimization-in-sksl-inliner
Change-Id: Ia6b706742ce5407453e0e697b6c1f9201084c0e8
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/384858
Auto-Submit: John Stiles <johnstiles@google.com>
Commit-Queue: John Stiles <johnstiles@google.com>
Reviewed-by: Brian Osman <brianosman@google.com>
Reviewed-by: Ethan Nicholas <ethannicholas@google.com>
Prevents us from accepting code that can't be correctly transformed to
GLSL, like:
uniform float x;
float y = x;
(Previously, writing code like that in a runtime effect would
effectively produce the exact same code all the way through to GLSL, and
the driver would fail to compile it).
Bug: skia:11336
Change-Id: Iaa797587c4a4a7289ed59ce2736cf0bf0fc5bca3
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/384698
Reviewed-by: Ethan Nicholas <ethannicholas@google.com>
Commit-Queue: Brian Osman <brianosman@google.com>
Bug: skia:11716
Change-Id: Ic09071544b5b5216b01fbc9b478b6269dd96202f
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/382280
Reviewed-by: John Stiles <johnstiles@google.com>
Commit-Queue: Brian Osman <brianosman@google.com>
This only affects fragmentProcessors (children) - and the backend SkSL
we're emitting should not contain those. We've just been silently
ignoring those declarations when converting to GLSL, MSL, etc.
Change-Id: I241f2f4fe4614b49ebccc9c2976fd408e94656d0
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/384316
Commit-Queue: John Stiles <johnstiles@google.com>
Reviewed-by: John Stiles <johnstiles@google.com>
This reverts commit a04692f69e.
Reason for revert: Angry Vulkan bots.
Original change's description:
> Fixed a number of spots where we should have been using RelaxedPrecision
>
> Our SPIR-V output was missing many RelaxedPrecision decorations, which
> was presumably impacting performance.
>
> Change-Id: Iee32d4a42f37af167fe0e45f3db94c2142129695
> Reviewed-on: https://skia-review.googlesource.com/c/skia/+/384178
> Reviewed-by: Brian Osman <brianosman@google.com>
> Commit-Queue: Ethan Nicholas <ethannicholas@google.com>
TBR=egdaniel@google.com,brianosman@google.com,ethannicholas@google.com,johnstiles@google.com
Change-Id: If4fe945cb363c9b61b5a4abfde649a437689d2eb
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/384217
Reviewed-by: Ethan Nicholas <ethannicholas@google.com>
Commit-Queue: Ethan Nicholas <ethannicholas@google.com>
I ran into an issue in an upcoming CL which generated a particularly
ugly switch statement:
switch (x) {
default:
discard;}
So I cleaned this up, and while resolving this issue, managed to improve
a bunch of existing codegen as well. The formatting change has been
split out to a separate CL since it impacts so many golden outputs.
Change-Id: I7a6be29903c47560dcc7f6acd3ef15fd0c5c3c50
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/384179
Commit-Queue: John Stiles <johnstiles@google.com>
Auto-Submit: John Stiles <johnstiles@google.com>
Reviewed-by: Ethan Nicholas <ethannicholas@google.com>
Our SPIR-V output was missing many RelaxedPrecision decorations, which
was presumably impacting performance.
Change-Id: Iee32d4a42f37af167fe0e45f3db94c2142129695
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/384178
Reviewed-by: Brian Osman <brianosman@google.com>
Commit-Queue: Ethan Nicholas <ethannicholas@google.com>
Bug: skia:11738
Change-Id: I1dd5e99830f70d72c292379a45c4e39a55588858
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/383706
Commit-Queue: Brian Osman <brianosman@google.com>
Commit-Queue: Ethan Nicholas <ethannicholas@google.com>
Commit-Queue: Kevin Lubick <kjlubick@google.com>
Auto-Submit: Brian Osman <brianosman@google.com>
Reviewed-by: Ethan Nicholas <ethannicholas@google.com>
Reviewed-by: Kevin Lubick <kjlubick@google.com>
We no longer derive a performance benefit from this pass in practice,
and it is a very expensive compilation step. It is also prone to fuzz-
related errors.
Doc: http://go/optimization-in-sksl
Change-Id: Ief08ffac659a8fe7fe92c92b9a5da14c9f713bc2
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/381261
Reviewed-by: Brian Osman <brianosman@google.com>
Commit-Queue: John Stiles <johnstiles@google.com>
As you might expect, a function tagged with `noinline` will never be
considered as a candidate for inlining.
Change-Id: Ia098f8974e6de251d78bb2a76cd71db8a86bc19c
Bug: skia:11362
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/382337
Reviewed-by: Ethan Nicholas <ethannicholas@google.com>
Commit-Queue: Ethan Nicholas <ethannicholas@google.com>
Auto-Submit: John Stiles <johnstiles@google.com>
Currently, only one of three uses (local variables) does this correctly.
Bug: skia:11716
Change-Id: Iad11e8e5998fcc7caee4d438e0558c5d4e2b1821
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/382277
Reviewed-by: John Stiles <johnstiles@google.com>
Commit-Queue: Brian Osman <brianosman@google.com>
Change-Id: Ic2d1240ab785101365b0fd934562505fb5a3e599
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/381816
Reviewed-by: Brian Osman <brianosman@google.com>
Commit-Queue: Brian Osman <brianosman@google.com>
Auto-Submit: John Stiles <johnstiles@google.com>
Expressions like `value == value` or `color.a != color.a` can be
replaced by `true` or `false` on sight. The GLSL spec makes it clear
that checking for NaN is optional:
4.7.1 Range and Precision
"... NaNs are not required to be generated. Support for signaling NaNs
is not required and exceptions are never raised. Operations and built-in
functions that operate on a NaN are not required to return a NaN as the
result."
Change-Id: I2ad9f2dc505b638ea2904bef41b7a79a2b329551
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/381262
Commit-Queue: John Stiles <johnstiles@google.com>
Auto-Submit: John Stiles <johnstiles@google.com>
Reviewed-by: Brian Osman <brianosman@google.com>
This CL will be used to test for potential performance regressions (or
improvements) that we might cause by disabling this optimization pass.
It will be reverted in ~1 day.
Change-Id: I26b7687c341eb6d81231406381c39869cfccf6d6
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/381259
Auto-Submit: John Stiles <johnstiles@google.com>
Commit-Queue: John Stiles <johnstiles@google.com>
Reviewed-by: Ethan Nicholas <ethannicholas@google.com>
Reviewed-by: Brian Osman <brianosman@google.com>
Change-Id: I7a7874e58bf53978afce8a41b26092406b6490ed
Bug: skia:11342
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/380360
Reviewed-by: Brian Osman <brianosman@google.com>
Commit-Queue: John Stiles <johnstiles@google.com>
Expressions like `x * 1`, `x *= 1`, `x + 0`, `x * 0`, or `0 / x` don't
actually do anything, and can be simplified to just `x` or `0`. (The
zero case must also check that `x` doesn't have side effects, because
`0 * myFunction()` still needs to call `myFunction`.)
`0 - x` is also detected and rewritten as `-x`.
`0 / 0` is left as-is.
This logic works for scalars and vectors; matrices are left as-is.
A similar optimization also occurs in the constant-propagation pass, so
we see almost no diffs in the tests. If control-flow analysis is turned
off, we do see some improvements. (I didn't reuse the existing code at
all, since it was designed around rewriting the CFG tree, but the
concept was identical.)
Change-Id: Ia99cd81f1d4cd3dafaa43ccac6a2261e3257a185
Bug: skia:11343
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/380356
Reviewed-by: Brian Osman <brianosman@google.com>
Commit-Queue: John Stiles <johnstiles@google.com>
Auto-Submit: John Stiles <johnstiles@google.com>
