Commit Graph

3342 Commits

Author SHA1 Message Date
Roman Lebedev ffec58bce6 [InstCombine][NFC] Add tests for one more masked merge pattern.
This pattern came up in D46494.
I'm pretty sure we want to canonicalize it from
	(x | ~m) & (y &  m)
to
	(x &  m) | (y & ~m)

https://rise4fun.com/Alive/TEM

llvm-svn: 331625
2018-05-07 09:42:45 +00:00
Sanjay Patel e7b6654711 [InstCombine] refine select-of-constants to bitwise ops
Add logic for the special case when a cmp+select can clearly be
reduced to just a bitwise logic instruction, and remove an 
over-reaching chunk of general purpose bit magic. The primary goal 
is to remove cases where we are not improving the IR instruction 
count when doing these select transforms, and in all cases here that 
is true.

In the motivating 3-way compare tests, there are further improvements
because we can combine/propagate select values (not sure if that
belongs in instcombine, but it's there for now).

DAGCombiner has folds to turn some of these selects into bit magic,
so there should be no difference in the end result in those cases.
Not all constant combinations are handled there yet, however, so it
is possible that some targets will see more cmov/csel codegen with
this change in IR canonicalization. 

Ideally, we'll go further to *not* turn selects into multiple 
logic/math ops in instcombine, and we'll canonicalize to selects.
But we should make sure that this step does not result in regressions
first (and if it does, we should fix those in the backend).

The general direction for this change was discussed here:
http://lists.llvm.org/pipermail/llvm-dev/2016-September/105373.html
http://lists.llvm.org/pipermail/llvm-dev/2017-July/114885.html

Alive proofs for the new bit magic:
https://rise4fun.com/Alive/XG7

Differential Revision: https://reviews.llvm.org/D46086

llvm-svn: 331486
2018-05-03 21:58:44 +00:00
Omer Paparo Bivas 54596e1c1a [InstCombine] new testcases for OverflowingBinaryOperators and PossiblyExactOperators transformations; NFC
instcombine should transform the relevant cases if the OverflowingBinaryOperator/PossiblyExactOperator can be proven to be safe.

Change-Id: I7aec62a31a894e465e00eb06aed80c3ea0c9dd45
llvm-svn: 331265
2018-05-01 14:27:10 +00:00
Omer Paparo Bivas 82ef8e19ef [InstCombine] Adjusting bswap pattern matching to hold for And/Shift mixed case
Differential Revision: https://reviews.llvm.org/D45731

Change-Id: I85d4226504e954933c41598327c91b2d08192a9d
llvm-svn: 331257
2018-05-01 12:25:46 +00:00
Sanjay Patel 1934cb1c49 [InstCombine] fix test to restore intent
This test had values that differed in only in capitalization,
and that causes problems for the auto-generating check line
script. So I changed that in rL331226, but I accidentally
forgot to change a subsequent use of a param.

llvm-svn: 331228
2018-04-30 21:28:18 +00:00
Sanjay Patel dab8515370 [InstCombine] add tests, update checks; NFC
llvm-svn: 331226
2018-04-30 21:03:36 +00:00
Roman Lebedev aa4faec114 [InstCombine] Unfold masked merge with constant mask
Summary:
As discussed in D45733, we want to do this in InstCombine.

https://rise4fun.com/Alive/LGk

Reviewers: spatel, craig.topper

Reviewed By: spatel

Subscribers: chandlerc, xbolva00, llvm-commits

Differential Revision: https://reviews.llvm.org/D45867

llvm-svn: 331205
2018-04-30 17:59:33 +00:00
Roman Lebedev b7ae4ef82b [InstCombine][NFC] Add tests for unfolding masked merge with constant mask
Summary: As discussed in D45733, we want to do this in InstCombine.

Differential Revision: https://reviews.llvm.org/D45866

llvm-svn: 331204
2018-04-30 17:59:26 +00:00
Roman Lebedev 136867931a [InstCombine] Canonicalize variable mask in masked merge
Summary:
Masked merge has a pattern of: `((x ^ y) & M) ^ y`.
But, there is no difference between `((x ^ y) & M) ^ y` and `((x ^ y) & ~M) ^ x`,
We should canonicalize the pattern to non-inverted mask.

https://rise4fun.com/Alive/Yol

Reviewers: spatel, craig.topper

Reviewed By: spatel

Subscribers: llvm-commits

Differential Revision: https://reviews.llvm.org/D45664

llvm-svn: 331112
2018-04-28 15:45:07 +00:00
Roman Lebedev 6b1e66b188 [InstCombine][NFC] Add tests for variable mask canonicalization in masked merge
Summary:
Masked merge has a pattern of: `((x ^ y) & M) ^ y`.
But, there is no difference between `((x ^ y) & M) ^ y` and `((x ^ y) & ~M) ^ x`,
We should canonicalize the pattern to non-inverted mask.

Differential Revision: https://reviews.llvm.org/D45663

llvm-svn: 331111
2018-04-28 15:45:00 +00:00
Max Kazantsev 303572f3df [NFC] Add some tests that demonstrate unrecognized three-way comparison patterns
llvm-svn: 331100
2018-04-28 04:38:21 +00:00
Roman Lebedev 6959b8e76f [PatternMatch] Stabilize the matching order of commutative matchers
Summary:
Currently, we
1. match `LHS` matcher to the `first` operand of binary operator,
2. and then match `RHS` matcher to the `second` operand of binary operator.
If that does not match, we swap the `LHS` and `RHS` matchers:
1. match `RHS` matcher to the `first` operand of binary operator,
2. and then match `LHS` matcher to the `second` operand of binary operator.

This works ok.
But it complicates writing of commutative matchers, where one would like to match
(`m_Value()`) the value on one side, and use (`m_Specific()`) it on the other side.

This is additionally complicated by the fact that `m_Specific()` stores the `Value *`,
not `Value **`, so it won't work at all out of the box.

The last problem is trivially solved by adding a new `m_c_Specific()` that stores the
`Value **`, not `Value *`. I'm choosing to add a new matcher, not change the existing
one because i guess all the current users are ok with existing behavior,
and this additional pointer indirection may have performance drawbacks.
Also, i'm storing pointer, not reference, because for some mysterious-to-me reason
it did not work with the reference.

The first one appears trivial, too.
Currently, we
1. match `LHS` matcher to the `first` operand of binary operator,
2. and then match `RHS` matcher to the `second` operand of binary operator.
If that does not match, we swap the ~~`LHS` and `RHS` matchers~~ **operands**:
1. match ~~`RHS`~~ **`LHS`** matcher to the ~~`first`~~ **`second`** operand of binary operator,
2. and then match ~~`LHS`~~ **`RHS`** matcher to the ~~`second`~ **`first`** operand of binary operator.

Surprisingly, `$ ninja check-llvm` still passes with this.
But i expect the bots will disagree..

The motivational unittest is included.
I'd like to use this in D45664.

Reviewers: spatel, craig.topper, arsenm, RKSimon

Reviewed By: craig.topper

Subscribers: xbolva00, wdng, llvm-commits

Differential Revision: https://reviews.llvm.org/D45828

llvm-svn: 331085
2018-04-27 21:23:20 +00:00
Matt Morehouse 1ae1febfde Revert "[SimplifyLibcalls] Replace locked IO with unlocked IO"
This reverts r331002 due to sanitizer bot breakage.

llvm-svn: 331011
2018-04-27 01:48:09 +00:00
David Bolvansky 2c9cc9c731 [SimplifyLibcalls] Replace locked IO with unlocked IO
Summary: If file stream arg is not captured and source is fopen, we could replace IO calls by unlocked IO ("_unlocked" function variants) to gain better speed,

Reviewers: efriedma, RKSimon, spatel, sanjoy, hfinkel, majnemer

Subscribers: lebedev.ri, llvm-commits

Differential Revision: https://reviews.llvm.org/D45736

llvm-svn: 331002
2018-04-26 22:31:43 +00:00
Roman Lebedev 33095e3610 [InstCombine][NFC] Regenerate checks in or-xor.ll
llvm-svn: 330996
2018-04-26 21:41:56 +00:00
Roman Lebedev cabaeac29c [InstCombine][NFC] Regenerate checks in and-or-not.ll
llvm-svn: 330994
2018-04-26 21:13:09 +00:00
Sanjoy Das 6f1937b10f [InstCombine] Simplify Add with remainder expressions as operands.
Summary:
Simplify integer add expression X % C0 + (( X / C0 ) % C1) * C0 to
X % (C0 * C1).  This is a common pattern seen in code generated by the XLA
GPU backend.

Add test cases for this new optimization.

Patch by Bixia Zheng!

Reviewers: sanjoy

Reviewed By: sanjoy

Subscribers: efriedma, craig.topper, lebedev.ri, llvm-commits, jlebar

Differential Revision: https://reviews.llvm.org/D45976

llvm-svn: 330992
2018-04-26 20:52:28 +00:00
Sanjoy Das 0e643db48f Add test cases to prepare for the optimization that simplifies Add with
remainder expressions as operands.

Summary:
Add test cases to prepare for the new optimization that Simplifies integer add
expression X % C0 + (( X / C0 ) % C1) * C0 to X % (C0 * C1).

Patch by Bixia Zheng!

Reviewers: sanjoy

Reviewed By: sanjoy

Subscribers: jlebar, llvm-commits

Differential Revision: https://reviews.llvm.org/D46017

llvm-svn: 330991
2018-04-26 20:52:27 +00:00
Roman Lebedev 7cc56f1599 [InstCombine][NFC] add2.ll: add a few commutative checks.
Fixes some missing test coverage in InstCombineAddSub.cpp, visitAdd()

llvm-svn: 330986
2018-04-26 20:07:17 +00:00
Roman Lebedev 1efe879641 [InstCombine][NFC] Autogenerate checks in add2.ll
llvm-svn: 330985
2018-04-26 20:07:12 +00:00
Roman Lebedev 3d7b22621c [NFC][InstCombine] rem.ll: add a few commutative tests.
This closes a gap in missing test coverage in
isKnownToBeAPowerOfTwo() from ValueTracking.cpp

llvm-svn: 330975
2018-04-26 18:44:37 +00:00
Roman Lebedev e117e1a440 [NFC][InstCombine] Regenerate rem.ll test
llvm-svn: 330974
2018-04-26 18:44:32 +00:00
David Bolvansky cb8ca5f37c [SimplifyLibcalls] Atoi, strtol replacements
Reviewers: spatel, lebedev.ri, xbolva00, efriedma

Reviewed By: xbolva00, efriedma

Subscribers: efriedma, llvm-commits

Differential Revision: https://reviews.llvm.org/D45418

llvm-svn: 330860
2018-04-25 18:58:53 +00:00
Sanjay Patel 0387ceb67a [InstCombine] add tests for select to logic folds; NFC
As discussed in D45862, we want these folds sometimes
because they're good improvements.
But as we can see here, the current logic doesn't
check uses and doesn't produce optimal code in all
cases. 

llvm-svn: 330837
2018-04-25 15:59:23 +00:00
Sanjay Patel 54795bb16b [InstCombine] move tests for select with bit-test of condition; NFC
These are all but 1 of the select-of-constant tests that appear 
to be transformed within foldSelectICmpAnd() and the block above 
it predicated by decomposeBitTestICmp().

As discussed in D45862 (and can be seen in several tests here),
we probably want to stop doing those transforms because they
can increase the instruction count without benefitting other
passes or codegen.

The 1 test not included here is a urem test where the bit hackery
allows us to remove a urem. To preserve killing that urem, we 
should do some stronger known-bits analysis or pattern matching of 
'urem x, (select-of-pow2-constants)'.

llvm-svn: 330768
2018-04-24 21:06:06 +00:00
Sanjay Patel 510af48e5d [InstCombine] regenerate checks; NFC
The first step in fixing problems raised in D45862
is to make the problems visible. Now we can more easily
see/update cases where selects have been turned into 
multiple instructions with no apparent improvement in 
analysis or benefits for other passes (vectorization).

llvm-svn: 330731
2018-04-24 16:08:03 +00:00
Sanjay Patel f03ec65517 [InstCombine] regenerate checks; NFC
The current version of the script uses regex for params.
This could mask a bug (param values got wrongly swapped),
but it seems unlikely in practice, so let's just update
the whole file to reduce diffs when there is a meaningful
change here.

llvm-svn: 330729
2018-04-24 15:42:30 +00:00
Sanjay Patel 30be665e82 [PatternMatch] allow undef elements when matching a vector zero
This is the last step in getting constant pattern matchers to allow
undef elements in constant vectors.

I'm adding a dedicated m_ZeroInt() function and building m_Zero() from
that. In most cases, calling code can be updated to use m_ZeroInt()
directly when there's no need to match pointers, but I'm leaving that
efficiency optimization as a follow-up step because it's not always
clear when that's ok.

There are just enough icmp folds in InstSimplify that can be used for 
integer or pointer types, that we probably still want a generic m_Zero()
for those cases. Otherwise, we could eliminate it (and possibly add a
m_NullPtr() as an alias for isa<ConstantPointerNull>()).

We're conservatively returning a full zero vector (zeroinitializer) in
InstSimplify/InstCombine on some of these folds (see diffs in InstSimplify),
but I'm not sure if that's actually necessary in all cases. We may be 
able to propagate an undef lane instead. One test where this happens is 
marked with 'TODO'.
 

llvm-svn: 330550
2018-04-22 17:07:44 +00:00
Sanjay Patel c1265ab99e [InstCombine] add vector test with undef elts; NFC
llvm-svn: 330547
2018-04-22 15:59:14 +00:00
Sanjay Patel e187cd3273 [InstSimplify, InstCombine] add vector tests with undef elts; NFC
llvm-svn: 330543
2018-04-22 14:19:37 +00:00
Sanjay Patel 5f845732ed [InstSimplify] move tests for shifts; NFC
llvm-svn: 330516
2018-04-21 16:58:00 +00:00
Sanjay Patel d0b27a1156 [InstSimplify] move/add/regenerate checks for tests; NFC
llvm-svn: 330515
2018-04-21 16:23:47 +00:00
Roman Lebedev f6934d725b [NFC][InstCombine] Regenerate two tests that are affected by folding masked merge
llvm-svn: 330415
2018-04-20 10:49:19 +00:00
Roman Lebedev d536de1e7b [NFC][InstCombine] A few more tests for masked merge add/xor -> or with constant mask
llvm-svn: 330325
2018-04-19 13:02:17 +00:00
Sanjay Patel b2ab3f28d5 [SimplifyLibcalls] Realloc(null, N) -> Malloc(N)
Patch by Dávid Bolvanský!

Differential Revision: https://reviews.llvm.org/D45413

llvm-svn: 330259
2018-04-18 14:21:31 +00:00
Bjorn Pettersson bc4f19b6bd [DebugInfo] Sink related dbg users when sinking in InstCombine
Summary:
When sinking an instruction in InstCombine we now also sink
the DbgInfoIntrinsics that are using the sunken value.

Example)

When sinking the load in this input

bb.X:
  %0 = load i64, i64* %start, align 4, !dbg !31
  tail call void @llvm.dbg.value(metadata i64 %0, ...)
  br i1 %cond, label %for.end, label %for.body.lr.ph
for.body.lr.ph:
  br label %for.body

we now also move the dbg.value, like this

bb.X:
  br i1 %cond, label %for.end, label %for.body.lr.ph
for.body.lr.ph:
  %0 = load i64, i64* %start, align 4, !dbg !31
  tail call void @llvm.dbg.value(metadata i64 %0, ...)
  br label %for.body

In the past we haven't moved the dbg.value so we got

bb.X:
  tail call void @llvm.dbg.value(metadata i64 %0, ...)
  br i1 %cond, label %for.end, label %for.body.lr.ph
for.body.lr.ph:
  %0 = load i64, i64* %start, align 4, !dbg !31
  br label %for.body


So in the past we got a debug-use before the def of %0.
And that dbg.value was also on the path jumping to %for.end, for
which %0 never was defined.

CodeGenPrepare normally comes to rescue later (when not moving
the dbg.value), since it moves dbg.value instrinsics quite
brutally, without really analysing if it is correct to move
the intrinsic (see PR31878).
So at the moment this patch isn't expected to have much impact,
besides that it is moving the dbg.value already in opt, making
the IR look more sane directly.

This can be seen as a preparation to (hopefully) make it possible
to turn off CodeGenPrepare::placeDbgValues later as a solution
to PR31878.

I also adjusted test/DebugInfo/X86/sdagsplit-1.ll to make the
IR in the test case up-to-date with this behavior in InstCombine.

Reviewers: rnk, vsk, aprantl

Reviewed By: vsk, aprantl

Subscribers: mattd, JDevlieghere, llvm-commits

Tags: #debug-info

Differential Revision: https://reviews.llvm.org/D45425

llvm-svn: 330243
2018-04-18 08:08:04 +00:00
Sanjay Patel aea15131db [InstCombine] peek through bitcasted vector/array pointer GEP operand
The bitcast may be interfering with other combines or vectorization 
as shown in PR16739:
https://bugs.llvm.org/show_bug.cgi?id=16739

Most pointer-related optimizations are probably able to look through 
this bitcast, but removing the bitcast shrinks the IR, so it's at
least a size savings.

Differential Revision: https://reviews.llvm.org/D44833

llvm-svn: 330237
2018-04-18 00:36:40 +00:00
Sanjay Patel 1170daa277 [InstCombine] simplify fneg+fadd folds; NFC
Two cleanups:
1. As noted in D45453, we had tests that don't need FMF that were misplaced in the 'fast-math.ll' test file.
2. This removes the final uses of dyn_castFNegVal, so that can be deleted. We use 'match' now.

llvm-svn: 330126
2018-04-16 14:13:57 +00:00
Roman Lebedev f84bfb2147 [InstCombine] Simplify 'xor' to 'or' if no common bits are set.
Summary:
In order to get the whole fold as specified in [[ https://bugs.llvm.org/show_bug.cgi?id=6773 | PR6773 ]],
let's first handle the simple straight-forward things.
Let's start with the `and` -> `or` simplification.

The one obvious thing missing here: the constant mask is not handled.
I have an idea how to handle it, but it will require some thinking,
and is not strictly required here, so i've left that for later.

https://rise4fun.com/Alive/Pkmg

Reviewers: spatel, craig.topper, eli.friedman, jingyue

Reviewed By: spatel

Subscribers: llvm-commits

Was reviewed as part of https://reviews.llvm.org/D45631

llvm-svn: 330103
2018-04-15 18:59:44 +00:00
Roman Lebedev 620b3da38f [InstCombine] Simplify 'add' to 'or' if no common bits are set.
Summary:
In order to get the whole fold as specified in [[ https://bugs.llvm.org/show_bug.cgi?id=6773 | PR6773 ]],
let's first handle the simple straight-forward things.
Let's start with the `and` -> `or` simplification.

The one obvious thing missing here: the constant mask is not handled.
I have an idea how to handle it, but it will require some thinking,
and is not strictly required here, so i've left that for later.

https://rise4fun.com/Alive/Pkmg

Reviewers: spatel, craig.topper, eli.friedman, jingyue

Reviewed By: spatel

Subscribers: llvm-commits

Differential Revision: https://reviews.llvm.org/D45631

llvm-svn: 330101
2018-04-15 18:59:33 +00:00
Warren Ristow 8b2f27ce3a [InstCombine] Enable Add/Sub simplifications with only 'reassoc' FMF
These simplifications were previously enabled only with isFast(), but that
is more restrictive than required. Since r317488, FMF has 'reassoc' to
control these cases at a finer level.

llvm-svn: 330089
2018-04-14 19:18:28 +00:00
Sanjay Patel 713f09d014 [InstCombine] add shift+logic tests (PR37098); NFC
It debateable whether instcombine should be in the business of
reassociation, but it is currently. 

These tests and PR37098 demonstrate a missing ability to do a 
simple reassociation that allows eliminating shifts. 

If we decide that functionality belongs somewhere else, then we 
should still have some tests to show that we've intentionally 
limited instcombine to *not* include this ability.

llvm-svn: 330086
2018-04-14 13:39:02 +00:00
Roman Lebedev 8db3e115e7 [InstCombine][NFC] masked-merge: add 'and' tests, too.
(and plain 'or', for completeness sake.)

After submitting D45631, i have realized that it will *already*
affect 'and' pattern, and it was obvious that there were no
good test patterns to show that.

Since the masked-merge.ll is getting kinda big,
unify naming schemes a bit, and split into 'xor'/'and'/'or'
testfiles, with the only difference being the last operation.

llvm-svn: 330072
2018-04-13 21:57:01 +00:00
Roman Lebedev fe6a0b9a65 [InstCombine][NFC] masked-merge: commutativity tests: ensure the ordering.
This was intended since initially, but i did not really think
about it, and did not know how to force that. Now that the
xor->or fold is working (patch upcoming), this came up
to improve the test coverage.

A followup for rL330003, rL330007
https://bugs.llvm.org/show_bug.cgi?id=6773

llvm-svn: 330039
2018-04-13 17:15:55 +00:00
Roman Lebedev 4899a9cc89 [InstCombine][NFC] Regenerate logical-select.ll test
llvm-svn: 330017
2018-04-13 14:07:29 +00:00
Roman Lebedev 53e423ed1e [InstCombine][NFC] Add last few tests with constant mask for masked merge folding.
A followup for rL330003
https://bugs.llvm.org/show_bug.cgi?id=6773

llvm-svn: 330007
2018-04-13 12:00:00 +00:00
Roman Lebedev 038d996c80 [InstCombine][NFC] Add tests for masked merge folding.
https://bugs.llvm.org/show_bug.cgi?id=6773

As discussed there, some backends may want to undo this fold
(x86+bmi for scalars, x86+sse for vectors, ...)
https://bugs.llvm.org/show_bug.cgi?id=37104

https://rise4fun.com/Alive/JXt

llvm-svn: 330003
2018-04-13 10:56:35 +00:00
Roman Lebedev c00659328a [InstCombine]: foldSelectICmpAndAnd(): and is commutative
Summary:
The fold added in D45108 did not account for the fact that
the and instruction is commutative, and if the mask is a variable,
the mask variable and the fold variable may be swapped.

I have noticed this by accident when looking into [[ https://bugs.llvm.org/show_bug.cgi?id=6773 | PR6773 ]]

This extends/generalizes that fold, so it is handled too.

Reviewers: spatel, craig.topper

Reviewed By: spatel

Subscribers: llvm-commits

Differential Revision: https://reviews.llvm.org/D45539

llvm-svn: 330001
2018-04-13 09:57:57 +00:00
Craig Topper 254ed028a4 [X86] Remove the pmuldq/pmuldq intrinsics and replace with native IR.
This completes the work started in r329604 and r329605 when we changed clang to no longer use the intrinsics.

We lost some InstCombine SimplifyDemandedBit optimizations through this change as we aren't able to fold 'and', bitcast, shuffle very well.

llvm-svn: 329990
2018-04-13 06:07:18 +00:00
Roman Lebedev 53271ba1d2 [InstCombine][NFC]: Add tests: foldSelectICmpAndAnd(): and is commutative
Summary:
The fold added in D45108 did not account for the fact that
the and instruction is commutative, and if the mask is a variable,
the mask variable and the fold variable may be swapped.

I have noticed this by accident when looking into [[ https://bugs.llvm.org/show_bug.cgi?id=6773 | PR6773 ]]

Reviewers: spatel, craig.topper

Reviewed By: spatel

Subscribers: llvm-commits

Differential Revision: https://reviews.llvm.org/D45538

llvm-svn: 329901
2018-04-12 12:04:57 +00:00