Stuart Hastings 65d0bc94b4 Avoid fusing bitcasts with dynamic allocas if the amount-to-allocate ...

might overflow.  Re-typing the alloca to a larger type (e.g. double)
hoists a shift into the alloca, potentially exposing overflow in the
expression.  rdar://problem/9265821

llvm-svn: 132926

2011-06-13 18:48:49 +00:00

..

ADCE

…

ArgumentPromotion

…

BlockPlacement

…

BranchFolding

…

CodeExtractor

…

CodeGenPrepare

…

ConstantMerge

…

ConstProp

implement PR9315, constant folding exp2 in terms of pow (since hosts without

2011-05-22 22:22:35 +00:00

CorrelatedValuePropagation

…

DeadArgElim

Fix grammar in test.

2011-05-22 01:16:00 +00:00

DeadStoreElimination

fix a really nasty basicaa mod/ref calculation bug that was causing miscompilation of

2011-05-23 05:15:43 +00:00

EarlyCSE

…

FunctionAttrs

…

GlobalDCE

…

GlobalOpt

Add testcase for r132290, to check for the crasher caught by the buildbots

2011-05-29 19:41:14 +00:00

GVN

Add a testcase to demonstrate the problem where phi translation is

2011-06-04 07:05:05 +00:00

IndVarSimplify

Test case pasto (failed when run with IR verifier).

2011-06-02 23:57:27 +00:00

Inline

SplitCriticalEdge can sometimes split the edge from an invoke to a landing

2011-06-09 20:06:24 +00:00

InstCombine

Avoid fusing bitcasts with dynamic allocas if the amount-to-allocate

2011-06-13 18:48:49 +00:00

InstSimplify

The comparision "max(x,y)==x" is equivalent to "x>=y". Since the max is

2011-05-07 16:56:49 +00:00

Internalize

…

IPConstantProp

…

JumpThreading

…

LCSSA

…

LICM

…

LoopDeletion

…

LoopIdiom

Fix PR9815: I was trying to get out of "generating code and then

2011-05-22 17:39:56 +00:00

LoopRotate

…

LoopSimplify

…

LoopStrengthReduce

When forming an ICmpZero LSRUse, normalize the non-IV operand

2011-05-18 21:02:18 +00:00

LoopUnroll

…

LoopUnswitch

Bail on unswitching a switch statement for a case with a critical edge. We name

2011-06-03 06:27:15 +00:00

LowerAtomic

…

LowerInvoke

…

LowerSetJmp

…

LowerSwitch

…

Mem2Reg

…

MemCpyOpt

PR10067: Add missing safety check to call return transformation in MemCpyOpt::processStore. If something accesses the dest of the "copy" between the call and the copy, the performCallSlotOptzn transformation is not valid.

2011-06-02 21:24:42 +00:00

MergeFunc

…

PhaseOrdering

This directory was missing the dg.exp to cause the tests to run. Some time since

2011-06-06 20:23:00 +00:00

PruneEH

…

Reassociate

…

ScalarRepl

Fix PR10104 by adding a bounds check on a vector element access check. It was

2011-06-09 01:45:33 +00:00

SCCP

…

SimplifyCFG

If the block that we're threading through is jumped to by an indirect branch,

2011-06-04 09:42:04 +00:00

SimplifyLibCalls

…

Sink

…

StripSymbols

…

TailCallElim

Don't do tail calls in a function that call setjmp. The stack might be

2011-05-16 03:05:33 +00:00

TailDup

Move few target-dependant tests to appropriate directories.

2011-05-06 18:24:46 +00:00