Commit Graph

10185 Commits

Author SHA1 Message Date
Serguei Katkov 66182d6c38 [SimplifyCFG] Re-apply Relax restriction for folding unconditional branches
The commit rL308422 introduces a restriction for folding unconditional
branches. Specifically if empty block with unconditional branch leads to
header of the loop then elimination of this basic block is prohibited.
However it seems this condition is redundantly strict.
If elimination of this basic block does not introduce more back edges
then we can eliminate this block.

The patch implements this relax of restriction.

The test profile/Linux/counter_promo_nest.c in compiler-rt project
is updated to meet this change.

Reviewers: efriedma, mcrosier, pacxx, hsung, davidxl	
Reviewed By: pacxx
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D42691

llvm-svn: 324572
2018-02-08 07:16:29 +00:00
Mircea Trofin 06ac8cfbd1 Verify profile data confirms large loop trip counts.
Summary:
Loops with inequality comparers, such as:

   // unsigned bound
   for (unsigned i = 1; i < bound; ++i) {...}

have getSmallConstantMaxTripCount report a large maximum static
trip count - in this case, 0xffff fffe. However, profiling info
may show that the trip count is much smaller, and thus
counter-recommend vectorization.

This change:
- flips loop-vectorize-with-block-frequency on by default.
- validates profiled loop frequency data supports vectorization,
  when static info appears to not counter-recommend it. Absence
  of profile data means we rely on static data, just as we've
  done so far.

Reviewers: twoh, mkuper, davidxl, tejohnson, Ayal

Reviewed By: davidxl

Subscribers: bkramer, llvm-commits

Differential Revision: https://reviews.llvm.org/D42946

llvm-svn: 324543
2018-02-07 23:29:52 +00:00
Alexey Bataev cd8d6de381 [SLP] Add a tests for PR36280, NFC.
llvm-svn: 324510
2018-02-07 20:11:37 +00:00
Max Kazantsev b299ade2c5 Re-enable "[SCEV] Make isLoopEntryGuardedByCond a bit smarter"
The failures happened because of assert which was overconfident about
SCEV's proving capabilities and is generally not valid.

Differential Revision: https://reviews.llvm.org/D42835

llvm-svn: 324473
2018-02-07 11:16:29 +00:00
Serguei Katkov 69246ca787 Revert [SCEV] Make isLoopEntryGuardedByCond a bit smarter
Revert rL324453 commit which causes buildbot failures.

Differential Revision: https://reviews.llvm.org/D42835

llvm-svn: 324462
2018-02-07 09:10:08 +00:00
Max Kazantsev dd5ee6f5d9 [SCEV] Make isLoopEntryGuardedByCond a bit smarter
Sometimes `isLoopEntryGuardedByCond` cannot prove predicate `a > b` directly.
But it is a common situation when `a >= b` is known from ranges and `a != b` is
known from a dominating condition. Thia patch teaches SCEV to sum these facts
together and prove strict comparison via non-strict one.

Differential Revision: https://reviews.llvm.org/D42835

llvm-svn: 324453
2018-02-07 07:56:26 +00:00
Michael Zolotukhin cae66ba5f8 The xfailed test from r324448 passed on one of the bots: remove it entirely for now.
llvm-svn: 324451
2018-02-07 06:54:11 +00:00
Michael Zolotukhin 1713dd5b8d Xfail the test added in r324445 until the underlying issue in LoopSink is fixed.
llvm-svn: 324448
2018-02-07 06:11:50 +00:00
Michael Zolotukhin e82e83fcce Follow-up for r324429: "[LCSSAVerification] Run verification only when asserts are enabled."
Before r324429 we essentially didn't have a verification of LCSSA, so
no wonder that it has been broken: currently loop-sink breaks it (the
attached test illustrates the failure).

It was detected during a stage2 RA build, so to unbreak it I'm disabling
the check for now.

llvm-svn: 324445
2018-02-07 04:24:44 +00:00
Alexey Bataev 1e593fe73e [SLP] Update test checks, NFC.
llvm-svn: 324387
2018-02-06 20:00:05 +00:00
Simon Pilgrim 9f2ae7e2d1 [InstCombine][ValueTracking] Match non-uniform constant power-of-two vectors
Generalize existing constant matching to work with non-uniform constant vectors as well.

Differential Revision: https://reviews.llvm.org/D42818

llvm-svn: 324369
2018-02-06 18:39:23 +00:00
Simon Pilgrim e11c64162c Regenerate vector-urem test. NFCI.
llvm-svn: 324357
2018-02-06 16:10:12 +00:00
Clement Courbet a7a1746865 [MergeICmps] Handle chains with several complex BCE basic blocks.
- Fix condition for detecting that a complex basic block was the first in
   the chain.
 - Add tests.

This was caught by buildbots when submitting rL324319.

llvm-svn: 324341
2018-02-06 12:25:33 +00:00
Hiroshi Inoue ba3585eaf2 [ThinLTO] fix test failure without x86 backend
This patch moves ThinLTOBitcodeWriter/module-asm.ll test case into x86 directory to avoid a test failure when x86 backend is not enabled.

llvm-svn: 324316
2018-02-06 07:03:09 +00:00
Peter Collingbourne 29c6f4833c ThinLTOBitcodeWriter: Do not include module-level inline asm in the merged module.
If the inline asm provides the definition of a symbol, this can result
in duplicate symbol errors.

Differential Revision: https://reviews.llvm.org/D42944

llvm-svn: 324313
2018-02-06 03:29:18 +00:00
Teresa Johnson 791c98e4c8 [ThinLTO] Remove dead and dropped symbol declarations when possible
Summary:
Removing the dropped symbols will prevent indirect call promotion in the
ThinLTO Backend from adding a new reference to a symbol, which can
result in linker unsats. This can happen when we compile with a sample
profile collected from one binary by used for another, which may have
profiled targets that aren't used in the new binary.

Note that until dropDeadSymbols handles variables and aliases (in
progress), we may not be able to remove the declaration and can still
have an issue.

Reviewers: grimar, davidxl

Subscribers: mehdi_amini, inglorion, llvm-commits, eraman

Differential Revision: https://reviews.llvm.org/D42816

llvm-svn: 324299
2018-02-06 00:43:39 +00:00
Sanjay Patel d7c702b451 [LoopStrengthReduce, x86] don't add cost for a cmp that will be macro-fused (PR35681)
In the motivating case from PR35681 and represented by the macro-fuse-cmp test:
https://bugs.llvm.org/show_bug.cgi?id=35681
...there's a 37 -> 31 byte size win for the loop because we eliminate the big base 
address offsets.

SPEC2017 on Ryzen shows no significant perf difference.

Differential Revision: https://reviews.llvm.org/D42607

llvm-svn: 324289
2018-02-05 23:43:05 +00:00
Sanjay Patel 49aafec2e6 [InstCombine] don't try to evaluate instructions with >1 use (revert r324014)
This example causes a compile-time explosion:

define i16 @foo(i16 %in) {
  %x = zext i16 %in to i32
  %a1 = mul i32 %x, %x
  %a2 = mul i32 %a1, %a1
  %a3 = mul i32 %a2, %a2
  %a4 = mul i32 %a3, %a3
  %a5 = mul i32 %a4, %a4
  %a6 = mul i32 %a5, %a5
  %a7 = mul i32 %a6, %a6
  %a8 = mul i32 %a7, %a7
  %a9 = mul i32 %a8, %a8
  %a10 = mul i32 %a9, %a9
  %a11 = mul i32 %a10, %a10
  %a12 = mul i32 %a11, %a11
  %a13 = mul i32 %a12, %a12
  %a14 = mul i32 %a13, %a13
  %a15 = mul i32 %a14, %a14
  %a16 = mul i32 %a15, %a15
  %a17 = mul i32 %a16, %a16
  %a18 = mul i32 %a17, %a17
  %a19 = mul i32 %a18, %a18
  %a20 = mul i32 %a19, %a19
  %a21 = mul i32 %a20, %a20
  %a22 = mul i32 %a21, %a21
  %a23 = mul i32 %a22, %a22
  %a24 = mul i32 %a23, %a23
  %T = trunc i32 %a24 to i16
  ret i16 %T
}

 

llvm-svn: 324276
2018-02-05 21:50:32 +00:00
Sanjay Patel 1c84dd9a8f [InstCombine] add test corresponding to r324252 (PR36225); NFC
As PR36225 shows, we definitely don't want to enable the
canEvaluate* logic with phis. 

There's still a question of whether we should just revert 
r324014 completely because it exposes a compile-time sinkhole
(although that problem might exist independently).

llvm-svn: 324266
2018-02-05 19:59:52 +00:00
Sanjay Patel e9a153f414 [InstCombine] add unsigned saturation subtraction canonicalizations
This is the instcombine part of unsigned saturation canonicalization.
Backend patches already commited: 
https://reviews.llvm.org/D37510
https://reviews.llvm.org/D37534

It converts unsigned saturated subtraction patterns to forms recognized 
by the backend:
(a > b) ? a - b : 0 -> ((a > b) ? a : b) - b)
(b < a) ? a - b : 0 -> ((a > b) ? a : b) - b)
(b > a) ? 0 : a - b -> ((a > b) ? a : b) - b)
(a < b) ? 0 : a - b -> ((a > b) ? a : b) - b)
((a > b) ? b - a : 0) -> - ((a > b) ? a : b) - b)
((b < a) ? b - a : 0) -> - ((a > b) ? a : b) - b)
((b > a) ? 0 : b - a) -> - ((a > b) ? a : b) - b)
((a < b) ? 0 : b - a) -> - ((a > b) ? a : b) - b)

Patch by Yulia Koval!

Differential Revision: https://reviews.llvm.org/D41480

llvm-svn: 324255
2018-02-05 17:53:29 +00:00
Hans Wennborg 22db17cf43 Revert r323472 "[Debug] Add dbg.value intrinsics for PHIs created during LCSSA."
This broke the Chromium build; see PR36238.

> This patch is an enhancement to propagate dbg.value information when
> Phis are created on behalf of LCSSA.  I noticed a case where a value
> carried across a loop was reported as <optimized out>.
>
> Specifically this case:
>
>   int bar(int x, int y) {
>     return x + y;
>   }
>
>   int foo(int size) {
>     int val = 0;
>     for (int i = 0; i < size; ++i) {
>       val = bar(val, i);  // Both val and i are correct
>     }
>     return val; // <optimized out>
>   }
>
> In the above case, after all of the interesting computation completes
> our value is reported as "optimized out." This change will add a
> dbg.value to correct this.
>
> This patch also moves the dbg.value insertion routine from
> LoopRotation.cpp into Local.cpp, so that we can share it in both places
> (LoopRotation and LCSSA).
>
> Patch by Matt Davis!
>
> Differential Revision: https://reviews.llvm.org/D42551

llvm-svn: 324247
2018-02-05 16:10:42 +00:00
Serguei Katkov 276b32bb14 Revert [SimplifyCFG] Relax restriction for folding unconditional branches
The patch causes the failure of the test
compiler-rt/test/profile/Linux/counter_promo_nest.c

To unblock buildbot, revert the patch while investigation is in progress.

Differential Revision: https://reviews.llvm.org/D42691

llvm-svn: 324214
2018-02-05 09:05:43 +00:00
Max Kazantsev f7667483c1 [NFC] Add tests for PR35743
llvm-svn: 324209
2018-02-05 08:09:49 +00:00
Serguei Katkov 6e93980e82 [SimplifyCFG] Relax restriction for folding unconditional branches
The commit rL308422 introduces a restriction for folding unconditional
branches. Specifically if empty block with unconditional branch leads to
header of the loop then elimination of this basic block is prohibited.
However it seems this condition is redundantly strict.
If elimination of this basic block does not introduce more back edges
then we can eliminate this block.

The patch implements this relax of restriction.

Reviewers: efriedma, mcrosier, pacxx, hsung, davidxl	
Reviewed By: pacxx
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D42691

llvm-svn: 324208
2018-02-05 07:56:43 +00:00
Serguei Katkov ec7029c286 Re-apply [SCEV] Fix isLoopEntryGuardedByCond usage
ScalarEvolution::isKnownPredicate invokes isLoopEntryGuardedByCond without check
that SCEV is available at entry point of the loop. It is incorrect and fixed by patch.

To bugs additionally fixed:
assert is moved after the check whether loop is not a nullptr.
Usage of isLoopEntryGuardedByCond in ScalarEvolution::isImpliedCondOperandsViaNoOverflow
is guarded by isAvailableAtLoopEntry.

Reviewers: sanjoy, mkazantsev, anna, dorit, reames
Reviewed By: mkazantsev
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D42417

llvm-svn: 324204
2018-02-05 05:49:47 +00:00
Florian Hahn 642637aab4 [PartialInliner] Update test (NFC).
llvm-svn: 324199
2018-02-04 18:40:24 +00:00
Florian Hahn 8f804fc07d [InlineFunction] Set arg attrs even if there only are VarArg attrs.
When using the partial inliner, we might have attributes for forwarded
varargs, but the CodeExtractor does not create an empty argument
attribute set for regular arguments in that case, because it does not know
of the additional arguments. So in case we have attributes for VarArgs, we
also have to make sure we create (empty) attributes for all regular arguments.

This fixes PR36210.

llvm-svn: 324197
2018-02-04 18:27:47 +00:00
Chad Rosier a097bc69df [LV] Use Demanded Bits and ValueTracking for reduction type-shrinking
The type-shrinking logic in reduction detection, although narrow in scope, is
also rather ad-hoc, which has led to bugs (e.g., PR35734). This patch modifies
the approach to rely on the demanded bits and value tracking analyses, if
available. We currently perform type-shrinking separately for reductions and
other instructions in the loop. Long-term, we should probably think about
computing minimal bit widths in a more complete way for the loops we want to
vectorize.

PR35734
Differential Revision: https://reviews.llvm.org/D42309

llvm-svn: 324195
2018-02-04 15:42:24 +00:00
David Green 9688ed61fe Remove unneeded -debug argument from new test
llvm-svn: 324176
2018-02-03 17:33:50 +00:00
David Green 7174023f57 [InstCombine] Allow common type conversions to i8/i16/i32
This, in instcombine, allows conversions to i8/i16/i32 (very
common cases) even if the resulting type is not legal according
to the data layout. This can often open up extra combine
opportunities.

Differential Revision: https://reviews.llvm.org/D42424

llvm-svn: 324174
2018-02-03 16:51:03 +00:00
Sanjay Patel a767ee5af0 [InstCombine] make sure tests are providing coverage for the stated pattern; NFC
Without extra instructions and uses, swapMayExposeCSEOpportunities() would change
the icmp (as seen in the check lines), so we were not actually testing patterns 
that should be handled by D41480.

llvm-svn: 324143
2018-02-02 21:40:54 +00:00
Sanjay Patel 5b8cb26bcc [InstCombine] add baseline tests for unsigned saturated sub (D41480); NFC
llvm-svn: 324109
2018-02-02 17:43:16 +00:00
Yaxun Liu 2a22c5deff [AMDGPU] Switch to the new addr space mapping by default
This requires corresponding clang change.

Differential Revision: https://reviews.llvm.org/D40955

llvm-svn: 324101
2018-02-02 16:07:16 +00:00
Sanjay Patel 3343fcef86 [InstCombine] allow multi-use values in canEvaluate* if all uses are in 1 inst
This is the enhancement suggested in D42536 to fix a shortcoming in 
regular InstCombine's canEvaluate* functionality.
When we have multiple uses of a value, but they're all in one instruction, we can 
allow that expression to be narrowed or widened for the same cost as a single-use 
value.

AFAICT, this can only matter for multiply: sub/and/or/xor/select would be simplified 
away if the operands are the same value; add becomes shl; shifts with a variable shift 
amount aren't handled.

Differential Revision: https://reviews.llvm.org/D42739

llvm-svn: 324014
2018-02-01 21:55:53 +00:00
David Green 184df0c35d Revert commit rL323951
Looks like it's causing timeouts out on at least ppc64le
buildbots.

llvm-svn: 323959
2018-02-01 13:05:25 +00:00
David Green e11f0545db [InstCombine] Allow common type conversions to i8/i16/i32
This, in instcombine, allows conversions to i8/i16/i32 (very
common cases) even if the resulting type is not legal according
to the data layout. This can often open up extra combine
opportunities.

Differential Revision: https://reviews.llvm.org/D42424

llvm-svn: 323951
2018-02-01 11:06:18 +00:00
Mikael Holmen 6d06976e74 [LSR] Don't force bases of foldable formulae to the final type.
Summary:
Before emitting code for scaled registers, we prevent
SCEVExpander from hoisting any scaled addressing mode
by emitting all the bases first. However, these bases
are being forced to the final type, resulting in some
odd code.

For example, if the type of the base is an integer and
the final type is a pointer, we will emit an inttoptr
for the base, a ptrtoint for the scale, and then a
'reverse' GEP where the GEP pointer is actually the base
integer and the index is the pointer. It's more intuitive
to use the pointer as a pointer and the integer as index.

Patch by: Bevin Hansson

Reviewers: atrick, qcolombet, sanjoy

Reviewed By: qcolombet

Subscribers: llvm-commits

Differential Revision: https://reviews.llvm.org/D42103

llvm-svn: 323946
2018-02-01 06:38:34 +00:00
Amjad Aboud b86b771c02 [AggressiveInstCombine] Fixed TruncCombine class to handle TruncInst leaf node correctly.
This covers the case where TruncInst leaf node is a constant expression.
See PR36121 for more details.

Differential Revision: https://reviews.llvm.org/D42622

llvm-svn: 323926
2018-01-31 22:39:05 +00:00
Puyan Lotfi 43e94b15ea Followup on Proposal to move MIR physical register namespace to '$' sigil.
Discussed here:

http://lists.llvm.org/pipermail/llvm-dev/2018-January/120320.html

In preparation for adding support for named vregs we are changing the sigil for
physical registers in MIR to '$' from '%'. This will prevent name clashes of
named physical register with named vregs.

llvm-svn: 323922
2018-01-31 22:04:26 +00:00
Marek Olsak 8f2df9d26c [SeparateConstOffsetFromGEP] Fix up addrspace in the AMDGPU test
llvm-svn: 323913
2018-01-31 20:49:19 +00:00
Marek Olsak 13e4741275 AMDGPU: Add intrinsics llvm.amdgcn.cvt.{pknorm.i16, pknorm.u16, pk.i16, pk.u16}
Reviewers: arsenm, nhaehnle

Subscribers: kzhuravl, wdng, yaxunl, dstuttard, tpr, t-tye

Differential Revision: https://reviews.llvm.org/D41663

llvm-svn: 323908
2018-01-31 20:18:04 +00:00
Marek Olsak 8e7d149a31 [SeparateConstOffsetFromGEP] Preserve metadata when splitting GEPs
Summary:
!amdgpu.uniform needs to be preserved for AMDGPU, otherwise bad things
happen.

Reviewers: arsenm, nhaehnle, jingyue, broune, majnemer, bjarke.roune, dblaikie

Subscribers: wdng, tpr, llvm-commits

Differential Revision: https://reviews.llvm.org/D42744

llvm-svn: 323907
2018-01-31 20:17:52 +00:00
Daniel Neilson be58a220e9 [CodeGenPrepare] Improve source and dest alignments of memory intrinsics independently
Summary:
  This change is part of step five in the series of changes to remove alignment argument from
memcpy/memmove/memset in favour of alignment attributes. In particular, this changes the
CodeGenPrepare pass to be more aggressive in improving the source and destination alignments
of memcpy/memmove/memset by exploiting our new ability to record independent alignments
for each argument.

Steps:
Step 1) Remove alignment parameter and create alignment parameter attributes for
memcpy/memmove/memset. ( rL322965, rC322964, rL322963 )
Step 2) Expand the IRBuilder API to allow creation of memcpy/memmove with differing
source and dest alignments. ( rL323597 )
Step 3) Update Clang to use the new IRBuilder API. ( rC323617 )
Step 4) Update Polly to use the new IRBuilder API. ( rL323618 )
Step 5) Update LLVM passes that create memcpy/memmove calls to use the new IRBuilder API,
and those that use use MemIntrinsicInst::[get|set]Alignment() to use [get|set]DestAlignment()
and [get|set]SourceAlignment() instead. ( rL323886 )
Step 6) Remove the single-alignment IRBuilder API for memcpy/memmove, and the
MemIntrinsicInst::[get|set]Alignment() methods.

Reference
   http://lists.llvm.org/pipermail/llvm-dev/2015-August/089384.html
   http://lists.llvm.org/pipermail/llvm-commits/Week-of-Mon-20151109/312083.html

llvm-svn: 323891
2018-01-31 17:24:53 +00:00
Sanjay Patel fd58ade81c [InstCombine] move related tests into the same file; NFC
llvm-svn: 323882
2018-01-31 15:47:59 +00:00
Sanjay Patel 8c74a9a155 [InstCombine] add tests to show limit of canEvaluate* ; NFC
llvm-svn: 323881
2018-01-31 15:28:39 +00:00
Amjad Aboud d895bff5f2 [AggressiveInstCombine] Make TruncCombine class ignore unreachable basic blocks.
Because dead code may contain non-standard IR that causes infinite looping or crashes in underlying analysis.
See PR36134 for more details.

Differential Revision: https://reviews.llvm.org/D42683

llvm-svn: 323862
2018-01-31 10:41:31 +00:00
Alexey Bataev 1c8f53f47d [SLP] Add extra test for extractelement shuffle, NFC.
llvm-svn: 323815
2018-01-30 21:06:06 +00:00
Sanjay Patel ffb37a29d1 [LoopStrengthReduce] add test to show potential macro-fusion-based diff (PR35681); NFC
This is the baseline output for the test proposed with D42607.

llvm-svn: 323806
2018-01-30 19:17:38 +00:00
Simon Pilgrim 073f089c6e [X86][XOP] Update isVectorShiftByScalarCheap with cases covered by XOP
Similar to D42437, XOP supports variable shift for v16i8/v8i16/v4i32/v2i64 types.

Differential Revision: https://reviews.llvm.org/D42526

llvm-svn: 323797
2018-01-30 18:10:21 +00:00
Petar Jovanovic 9208e8fbf6 [DeadArgumentElimination] Preserve llvm.dbg.values's first argument
When removing return value Dead Argument Elimination pass clobbers first
llvm.dbg.value’s argument for live arguments of that function by replacing
it with nullptr. In the next pass it will be deleted, so debug location
about those arguments are lost. This change fixes it.

Patch by Djordje Todorovic.

Differential Revision: https://reviews.llvm.org/D42541

llvm-svn: 323784
2018-01-30 16:42:04 +00:00