maxjhandsome
/
llvm-project
forked from OSchip/llvm-project
1
0
Fork 0
Code Issues Pull Requests Packages Releases Wiki Activity
llvm-project/llvm/test/Transforms
History
Alexey Bataev ec95c6cc0a [InstCombine] PR35354: Convert store(bitcast, load bitcast (select (Cond, &V1, &V2)) --> store (, load (select(Cond, load &V1, load &V2))) 
Summary:
If we have the code like this:
```
float a, b;
a = std::max(a ,b);
```
it is converted into something like this:
```
%call = call dereferenceable(4) float* @_ZSt3maxIfERKT_S2_S2_(float* nonnull dereferenceable(4) %a.addr, float* nonnull dereferenceable(4) %b.addr)
%1 = bitcast float* %call to i32*
%2 = load i32, i32* %1, align 4
%3 = bitcast float* %a.addr to i32*
store i32 %2, i32* %3, align 4
```
After inlinning this code is converted to the next:
```
%1 = load float, float* %a.addr
%2 = load float, float* %b.addr
%cmp.i = fcmp fast olt float %1, %2
%__b.__a.i = select i1 %cmp.i, float* %a.addr, float* %b.addr
%3 = bitcast float* %__b.__a.i to i32*
%4 = load i32, i32* %3, align 4
%5 = bitcast float* %arrayidx to i32*
store i32 %4, i32* %5, align 4

```
This pattern is not recognized as minmax pattern.
Patch solves this problem by converting sequence
```
store (bitcast, (load bitcast (select ((cmp V1, V2), &V1, &V2))))
```
to a sequence
```
store (,load (select((cmp V1, V2), &V1, &V2)))
```
After this the code is recognized as minmax pattern.

Reviewers: RKSimon, spatel

Subscribers: llvm-commits

Differential Revision: https://reviews.llvm.org/D40304

llvm-svn: 320157
 		2017-12-08 15:32:10 +00:00
..
ADCE [ADCE][Dominators] Reapply: Teach ADCE to preserve dominators 2017-08-22 16:30:21 +00:00
AddDiscriminators
…
AlignmentFromAssumptions
…
ArgumentPromotion [ArgPromotion] Preserve alignment of byval argument in new alloca 2017-08-04 17:09:11 +00:00
AtomicExpand
…
BDCE [BDCE] Don't check demanded bits on unsized types 2017-08-16 16:09:22 +00:00
BranchFolding
…
CallSiteSplitting [CallSiteSplitting] Remove some indirection (NFC). 2017-11-18 18:14:13 +00:00
CalledValuePropagation Add CalledValuePropagation pass 2017-10-25 13:40:08 +00:00
CodeExtractor - Removed unused lamba (IsReturnBlock) causing build bots to fail for r319398 2017-11-30 03:36:57 +00:00
CodeGenPrepare [BypassSlowDivision] Improve our handling of divisions by constants 2017-12-04 19:21:58 +00:00
ConstProp [X86] Replace 'REQUIRES: x86' in tests with 'REQUIRES: x86-registered-target' which seems to be the correct way to make them run on an x86 build. 2017-06-04 08:21:58 +00:00
ConstantHoisting Fix out-of-order stepping behavior in programs with hoisted constants. 2017-11-09 20:01:31 +00:00
ConstantMerge Canonicalize the representation of empty an expression in DIGlobalVariableExpression 2017-08-30 18:06:51 +00:00
Coroutines [coroutines] Add support for symmetric control transfer (musttail on coro.resumes followed by a suspend) 2017-08-25 02:25:10 +00:00
CorrelatedValuePropagation [CVP] Remove some {s|u}sub.with.overflow checks. 2017-12-05 18:14:24 +00:00
CrossDSOCFI [cfi] Build __cfi_check as Thumb when applicable. 2017-08-29 22:29:15 +00:00
DCE Add an @llvm.sideeffect intrinsic 2017-11-08 21:59:51 +00:00
DeadArgElim Remove the obsolete offset parameter from @llvm.dbg.value 2017-07-28 20:21:02 +00:00
DeadStoreElimination Add an @llvm.sideeffect intrinsic 2017-11-08 21:59:51 +00:00
DivRemPairs [DivRemPairs] split tests per target to account for bots that don't build for all targets 2017-09-09 14:10:59 +00:00
EarlyCSE Add an @llvm.sideeffect intrinsic 2017-11-08 21:59:51 +00:00
EliminateAvailableExternally
…
EntryExitInstrumenter EntryExitInstrumenter: set DebugLocs on the inserted call instructions (PR35412) 2017-11-28 18:44:26 +00:00
ExpandMemCmp/X86 re-land [ExpandMemCmp] Split ExpandMemCmp from CodeGen into its own pass." 2017-11-03 12:12:27 +00:00
Float2Int
…
ForcedFunctionAttrs
…
FunctionAttrs Add an @llvm.sideeffect intrinsic 2017-11-08 21:59:51 +00:00
FunctionImport Fix bot failures by requiring x86 target 2017-08-19 19:15:04 +00:00
GCOVProfiling Canonicalize the representation of empty an expression in DIGlobalVariableExpression 2017-08-30 18:06:51 +00:00
GVN [GVN] Prevent ScalarPRE from hoisting across instructions that don't pass control flow to successors 2017-11-28 07:07:55 +00:00
GVNHoist Add an @llvm.sideeffect intrinsic 2017-11-08 21:59:51 +00:00
GVNSink Add an @llvm.sideeffect intrinsic 2017-11-08 21:59:51 +00:00
GlobalDCE [PM] Teach the PGO instrumentation pasess to run GlobalDCE before 2017-05-25 07:15:09 +00:00
GlobalMerge Canonicalize the representation of empty an expression in DIGlobalVariableExpression 2017-08-30 18:06:51 +00:00
GlobalOpt Add an @llvm.sideeffect intrinsic 2017-11-08 21:59:51 +00:00
GlobalSplit
…
GuardWidening
…
IPConstantProp
…
IRCE [IRCE] Smart range intersection 2017-11-20 06:07:57 +00:00
IndVarSimplify [IndVars] Fix a bug introduced in r317012 2017-12-01 20:57:19 +00:00
InferAddressSpaces InferAddressSpaces: Fix bug about replacing addrspacecast 2017-10-30 21:19:41 +00:00
InferFunctionAttrs Add argmononly attribute to strlen and wcslen, i.e. they only read memory (string) passed to them. 2017-06-18 03:10:26 +00:00
Inline [opt-remarks] If hotness threshold is set, ignore remarks without hotness 2017-12-01 20:41:38 +00:00
InstCombine [InstCombine] PR35354: Convert store(bitcast, load bitcast (select (Cond, &V1, &V2)) --> store (, load (select(Cond, load &V1, load &V2))) 2017-12-08 15:32:10 +00:00
InstMerge
…
InstNamer Add basic test case for -instnamer 2017-05-08 23:18:46 +00:00
InstSimplify [InstSimplify] Add tests for the rL319894 2017-12-07 08:52:24 +00:00
InterleavedAccess [X86][LLVM]Expanding Supports lowerInterleaved{store|load}() in X86InterleavedAccess (VF64 stride 3-4) 2017-10-02 07:35:25 +00:00
Internalize
…
JumpThreading Fix some misc. -enable-var-scope violations 2017-11-13 01:47:52 +00:00
LCSSA
…
LICM [LICM] Fix PR35342 2017-11-17 20:38:25 +00:00
LoadStoreVectorizer Add an @llvm.sideeffect intrinsic 2017-11-08 21:59:51 +00:00
LoopDataPrefetch
…
LoopDeletion [Dominators] Teach LoopDeletion to use the new incremental API 2017-08-02 18:17:52 +00:00
LoopDistribute
…
LoopIdiom Add an @llvm.sideeffect intrinsic 2017-11-08 21:59:51 +00:00
LoopInterchange [LoopInterchange] Fix phi node ordering miscompile. 2017-10-21 13:58:37 +00:00
LoopLoadElim
…
LoopPredication [Loop Predication] Teach LP about reverse loops 2017-12-04 15:11:48 +00:00
LoopReroll Remove the obsolete offset parameter from @llvm.dbg.value 2017-07-28 20:21:02 +00:00
LoopRotate Fix llvm/test/Transforms/LoopRotate/pr35210.ll in rL318237, it uses debug options. 2017-11-15 06:46:58 +00:00
LoopSimplify [SCEV] Teach SCEV to find maxBECount when loop endbound is variant 2017-10-13 14:30:43 +00:00
LoopSimplifyCFG
…
LoopStrengthReduce Revert "[X86] Improvement in CodeGen instruction selection for LEAs." 2017-12-01 22:20:26 +00:00
LoopUnroll loop-unroll: teach remapInstruction to update dbg.value intrinsics. 2017-11-01 23:12:35 +00:00
LoopUnswitch [LoopUnswitch] Fix a simple bug which disables loop unswitch for select statement 2017-08-29 21:45:11 +00:00
LoopVectorize [LV] Interleaved access vectorization: fix computing new alias info 2017-12-06 22:42:24 +00:00
LoopVersioning
…
LoopVersioningLICM
…
LowerAtomic LowerAtomic: Don't skip optnone functions; atomic still need lowering (PR34020) 2017-08-23 15:43:28 +00:00
LowerExpectIntrinsic Fix builin_expect lowering bug 2017-06-07 18:32:24 +00:00
LowerGuardIntrinsic
…
LowerInvoke
…
LowerSwitch
…
LowerTypeTests Current implementation of Value::replaceUsesExceptBlockAddr() uses UseList 2017-11-17 00:30:24 +00:00
Mem2Reg Re-land r313825: "[IR] Add llvm.dbg.addr, a control-dependent version of llvm.dbg.declare" 2017-09-21 19:52:03 +00:00
MemCpyOpt Revert r319482 and r319483 "[memcpyopt] Teach memcpyopt to optimize across basic blocks" 2017-12-06 01:47:55 +00:00
MergeFunc [TailRecursionElimination] Skip debug intrinsics. 2017-11-28 09:32:25 +00:00
MergeICmps Re-land "[MergeICmps] Disable mergeicmps if the target does not want to handle memcmp expansion." 2017-10-10 08:00:45 +00:00
MetaRenamer [MetaRenamer] Leave `@main` alone. 2017-08-01 05:14:45 +00:00
NameAnonGlobals
…
NaryReassociate
…
NewGVN Add an @llvm.sideeffect intrinsic 2017-11-08 21:59:51 +00:00
ObjCARC ObjCARC: do not increment past the end of the BB 2017-10-24 00:09:10 +00:00
PGOProfile [PGO] detect infinite loop and form MST properly 2017-12-07 22:23:28 +00:00
PartiallyInlineLibCalls [PartiallyInlineLibCalls][x86] add TTI hook to allow sqrt inlining to depend on arg rather than result 2017-11-27 21:15:43 +00:00
PhaseOrdering [PassManager, SimplifyCFG] add test for PR34603 / D38566; NFC 2017-11-15 16:37:30 +00:00
PlaceSafepoints All libcalls should be considered to be GC-leaf functions. 2017-07-27 16:49:39 +00:00
PreISelIntrinsicLowering
…
PruneEH
…
Reassociate [Reassociation] regenerate test checks; NFC 2017-11-13 19:46:28 +00:00
Reg2Mem
…
RewriteStatepointsForGC Strip off invariant.start because memory locations arent invariant 2017-11-02 18:24:04 +00:00
SCCP [SCCP] Pick the right lattice value for constants. 2017-11-22 03:04:55 +00:00
SLPVectorizer Revert r319531 "[SLPVectorizer] Failure to beneficially vectorize 'copyable' elements in integer binary ops." 2017-12-01 16:17:24 +00:00
SROA Recommit rL319407: [SROA] enable splitting for non-whole-alloca loads and stores 2017-12-01 06:05:05 +00:00
SafeStack Parse and print DIExpressions inline to ease IR and MIR testing 2017-08-23 20:31:27 +00:00
SampleProfile Include already promoted counts when computing SUM for VP. 2017-11-06 19:52:49 +00:00
Scalarizer Remove the obsolete offset parameter from @llvm.dbg.value 2017-07-28 20:21:02 +00:00
SeparateConstOffsetFromGEP
…
SimpleLoopUnswitch [PM/Unswitch] Teach SimpleLoopUnswitch to do non-trivial unswitching, 2017-11-17 19:58:36 +00:00
SimplifyCFG Bail out of a SimplifyCFG switch table opt at undef values. 2017-12-05 14:14:00 +00:00
Sink Enhance synchscope representation 2017-07-11 22:23:00 +00:00
SpeculateAroundPHIs Add a new pass to speculate around PHI nodes with constant (integer) operands when profitable. 2017-11-28 11:32:31 +00:00
SpeculativeExecution Revert r301950: SpeculativeExecution: Stop using whitelist for costs 2017-05-10 12:30:07 +00:00
StraightLineStrengthReduce
…
StripDeadPrototypes
…
StripSymbols Canonicalize the representation of empty an expression in DIGlobalVariableExpression 2017-08-30 18:06:51 +00:00
StructurizeCFG [Dominators] Include infinite loops in PostDominatorTree 2017-08-15 18:14:57 +00:00
TailCallElim Remove this test 2017-11-28 22:39:38 +00:00
ThinLTOBitcodeWriter ThinLTOBitcodeWriter: Try harder to discard unused references to the merged module. 2017-11-30 23:05:52 +00:00
Util [InstCombine] Add a flag to disable LowerDbgDeclare 2017-09-13 01:43:25 +00:00
WholeProgramDevirt [LTO][ThinLTO] Use the linker resolutions to mark global values as dso_local. 2017-11-04 17:04:39 +00:00