summaryrefslogtreecommitdiffstats
path: root/llvm/lib/Analysis/BasicAliasAnalysis.cpp
Commit message (Collapse)AuthorAgeFilesLines
...
* refix PR1143 by making basicaa analyze zexts of indices aggresively,Chris Lattner2010-08-181-11/+30
| | | | | | which I broke with a recent patch. llvm-svn: 111452
* GetLinearExpression is only called when TD is non-null, pass as Chris Lattner2010-08-181-4/+4
| | | | | | a reference instead of pointer. llvm-svn: 111445
* rework GEP decomposition to make a new VariableGEPIndex struct instead ofChris Lattner2010-08-181-19/+39
| | | | | | | using a pair. This tidies up the code a bit. While setting things up, add a (currently unused) field to keep track of how the value is extended. llvm-svn: 111444
* move gep decomposition out of ValueTracking into BasicAA. The form ofChris Lattner2010-08-181-34/+229
| | | | | | | | | decomposition that it is doing is very basicaa specific and is only used by basicaa. Now with less tree breakingness. llvm-svn: 111433
* Revert r111375, "move gep decomposition out of ValueTracking into BasicAA. TheDaniel Dunbar2010-08-181-213/+34
| | | | | | form of", it doesn't pass tests. llvm-svn: 111385
* move gep decomposition out of ValueTracking into BasicAA. The form ofChris Lattner2010-08-181-34/+213
| | | | | | | decomposition that it is doing is very basicaa specific and is only used by basicaa. llvm-svn: 111375
* Reapply r110396, with fixes to appease the Linux buildbot gods.Owen Anderson2010-08-061-7/+7
| | | | llvm-svn: 110460
* Move all the logic for function attributes and call attributes out of theDan Gohman2010-08-061-120/+119
| | | | | | | | | | | | | AliasAnalysis base class and into BasicAliasAnalyais. This avoids confusion about where such logic is happening when there are other AliasAnalysis implementations present. Move the logic for translating two-callsite getModRefInfo queries into other AliasAnalysis queries out of BasicAliasAnalysis and into the AliasAnalysis base class, as it is useful for other AliasAnalysis implementations. llvm-svn: 110421
* Fix botched revert.Owen Anderson2010-08-061-4/+4
| | | | llvm-svn: 110416
* Revert r110396 to fix buildbots.Owen Anderson2010-08-061-3/+3
| | | | llvm-svn: 110410
* Fix 80-column violations.Dan Gohman2010-08-051-6/+6
| | | | llvm-svn: 110401
* Don't use PassInfo* as a type identifier for passes. Instead, use the ↵Owen Anderson2010-08-051-7/+7
| | | | | | | | address of the static ID member as the sole unique type identifier. Clean up APIs related to this change. llvm-svn: 110396
* Implement AccessesArguments checking in the two-callsite formDan Gohman2010-08-051-2/+32
| | | | | | | of BasicAA::getModRefInfo. This allows BasicAA to say that two memset calls to non-aliasing memory locations don't interfere. llvm-svn: 110393
* Fix memdep's code for reasoning about dependences between two calls. A RefDan Gohman2010-08-051-1/+6
| | | | | | | | | | response from getModRefInfo is not useful here. Instead, check for identical calls only in the NoModRef case. Reapply r110270, and strengthen it to compensate for the memdep changes. When both calls are readonly, there is no dependence between them. llvm-svn: 110382
* Revert r110270 for now. It appears to uncover a memdep bug.Dan Gohman2010-08-051-2/+2
| | | | llvm-svn: 110293
* The two-callsite form of AliasAnalysis::getModRefInfo is documentedDan Gohman2010-08-041-2/+2
| | | | | | | | | | to return Ref if the left callsite only reads memory read or written by the right callsite; fix BasicAliasAnalysis to implement this. Add AliasAnalysisEvaluator support for testing the two-callsite form of getModRefInfo. llvm-svn: 110270
* Thread const correctness through a bunch of AliasAnalysis interfaces andDan Gohman2010-08-031-9/+15
| | | | | | | | | | | | eliminate several const_casts. Make CallSite implicitly convertible to ImmutableCallSite. Rename the getModRefBehavior for intrinsic IDs to getIntrinsicModRefBehavior to avoid overload ambiguity with CallSite, which happens to be implicitly convertible to bool. llvm-svn: 110155
* The singular of "indices" is "index".Dan Gohman2010-08-031-3/+3
| | | | llvm-svn: 110135
* Delete an unused function.Dan Gohman2010-08-031-5/+0
| | | | llvm-svn: 110134
* Introduce a symbolic constant for ~0u for use with AliasAnalysis.Dan Gohman2010-08-031-7/+8
| | | | llvm-svn: 110091
* Add INSTANTIATE_AG_PASS, which combines RegisterPass<> with ↵Owen Anderson2010-07-211-10/+6
| | | | | | RegisterAnalysisGroup<> for pass registration. llvm-svn: 109058
* Speculatively revert r108813, in an attempt to get the self-host buildbots ↵Owen Anderson2010-07-201-2/+2
| | | | | | | | working again. I don't see why this patch would cause them to fail the way they are, but none of the other intervening patches seem likely either. llvm-svn: 108818
* Reapply r108794, a fix for the failing test from last time.Owen Anderson2010-07-201-2/+2
| | | | llvm-svn: 108813
* Revert r108794, "Separate PassInfo into two classes: a constructor-freeDaniel Dunbar2010-07-201-2/+2
| | | | | | | superclass (StaticPassInfo) and a constructor-ful subclass (PassInfo).", it is breaking teh everything. llvm-svn: 108805
* Separate PassInfo into two classes: a constructor-free superclass ↵Owen Anderson2010-07-201-2/+2
| | | | | | (StaticPassInfo) and a constructor-ful subclass (PassInfo). llvm-svn: 108794
* Minore code simplification.Dan Gohman2010-07-071-17/+15
| | | | llvm-svn: 107777
* Remove interprocedural-basic-aa and associated code. The AliasAnalysisDan Gohman2010-07-071-77/+14
| | | | | | | | | | | interface needs implementations to be consistent, so any code which wants to support different semantics must use a different interface. It's not currently worthwhile to add a new interface for this new concept. Document that AliasAnalysis doesn't support cross-function queries. llvm-svn: 107776
* Remove context sensitivity concerns from interprocedural-basic-aa, andDan Gohman2010-07-011-35/+55
| | | | | | | make it more aggressive in cases where both pointers are known to live in the same function. llvm-svn: 107420
* Fix whitespace style.Dan Gohman2010-06-291-2/+2
| | | | llvm-svn: 107175
* Use a more obvious way to avoid compiling functions which are only used when ↵Benjamin Kramer2010-06-291-3/+2
| | | | | | XDEBUG is enabled. llvm-svn: 107125
* Jump through some silly hoops to make GCC accept that a function may not alwaysChandler Carruth2010-06-291-0/+3
| | | | | | be called. llvm-svn: 107124
* Add an Intraprocedural form of BasicAliasAnalysis, which aims toDan Gohman2010-06-291-28/+110
| | | | | | | properly handles instructions and arguments defined in different functions, or across recursive function iterations. llvm-svn: 107109
* Fix Value::stripPointerCasts and BasicAA to avoid trouble onDan Gohman2010-06-281-9/+29
| | | | | | | code in unreachable blocks, which have have use-def cycles. This fixes PR7514. llvm-svn: 107071
* use ArgOperand accessorsGabor Greif2010-06-231-11/+11
| | | | llvm-svn: 106697
* Revert 101465, it broke internal OpenGL testing.Eric Christopher2010-04-161-11/+11
| | | | | | | Probably the best way to know that all getOperand() calls have been handled is to replace that API instead of updating. llvm-svn: 101579
* reapply r101434Gabor Greif2010-04-161-11/+11
| | | | | | | | | | | | | with a fix for self-hosting rotate CallInst operands, i.e. move callee to the back of the operand array the motivation for this patch are laid out in my mail to llvm-commits: more efficient access to operands and callee, faster callgraph-construction, smaller compiler binary llvm-svn: 101465
* back out r101423 and r101397, they break llvm-gcc self-host on darwin10Gabor Greif2010-04-161-11/+11
| | | | llvm-svn: 101434
* reapply r101364, which has been backed out in r101368Gabor Greif2010-04-151-11/+11
| | | | | | | | | | | | | with a fix rotate CallInst operands, i.e. move callee to the back of the operand array the motivation for this patch are laid out in my mail to llvm-commits: more efficient access to operands and callee, faster callgraph-construction, smaller compiler binary llvm-svn: 101397
* back out r101364, as it trips the linux nightlybot on some clang C++ testsGabor Greif2010-04-151-11/+11
| | | | llvm-svn: 101368
* rotate CallInst operands, i.e. move callee to the backGabor Greif2010-04-151-11/+11
| | | | | | | | | | of the operand array the motivation for this patch are laid out in my mail to llvm-commits: more efficient access to operands and callee, faster callgraph-construction, smaller compiler binary llvm-svn: 101364
* Pointers to zero-sized objects don't point to overlapping objects.Dan Gohman2010-04-081-0/+5
| | | | llvm-svn: 100789
* There are two ways of checking for a given type, for example isa<PointerType>(T)Duncan Sands2010-02-161-2/+2
| | | | | | | and T->isPointerTy(). Convert most instances of the first form to the second form. Requested by Chris. llvm-svn: 96344
* add some new methods to adjust this pointers. Not used yet.Chris Lattner2010-01-201-0/+20
| | | | llvm-svn: 94013
* move DecomposeGEPExpression out into ValueTracking.cppChris Lattner2009-11-261-158/+1
| | | | llvm-svn: 89956
* teach GetLinearExpression to be a bit more aggressive.Chris Lattner2009-11-261-1/+10
| | | | llvm-svn: 89955
* resolve a fixme. I haven't figured out how to write a testcaseChris Lattner2009-11-261-5/+4
| | | | | | to exercise this though. llvm-svn: 89954
* Teach basicaa that x|c == x+c when the c bits of x are clear. ThisChris Lattner2009-11-261-5/+13
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | allows us to compile the example in readme.txt into: LBB1_1: ## %bb movl 4(%rdx,%rax), %ecx movl %ecx, %esi imull (%rdx,%rax), %esi imull %esi, %ecx movl %esi, 8(%rdx,%rax) imull %ecx, %esi movl %ecx, 12(%rdx,%rax) movl %esi, 16(%rdx,%rax) imull %ecx, %esi movl %esi, 20(%rdx,%rax) addq $16, %rax cmpq $4000, %rax jne LBB1_1 instead of: LBB1_1: movl (%rdx,%rax), %ecx imull 4(%rdx,%rax), %ecx movl %ecx, 8(%rdx,%rax) imull 4(%rdx,%rax), %ecx movl %ecx, 12(%rdx,%rax) imull 8(%rdx,%rax), %ecx movl %ecx, 16(%rdx,%rax) imull 12(%rdx,%rax), %ecx movl %ecx, 20(%rdx,%rax) addq $16, %rax cmpq $4000, %rax jne LBB1_1 GCC (4.2) doesn't seem to be able to eliminate the loads in this testcase either, it generates: L2: movl (%rdx), %eax imull 4(%rdx), %eax movl %eax, 8(%rdx) imull 4(%rdx), %eax movl %eax, 12(%rdx) imull 8(%rdx), %eax movl %eax, 16(%rdx) imull 12(%rdx), %eax movl %eax, 20(%rdx) addl $4, %ecx addq $16, %rdx cmpl $1002, %ecx jne L2 llvm-svn: 89952
* teach basicaa that A[i] != A[i+1].Chris Lattner2009-11-261-0/+32
| | | | llvm-svn: 89951
* Change the other half of aliasGEP (which handles GEP differencing) to use ↵Chris Lattner2009-11-261-466/+108
| | | | | | DecomposeGEPExpression. This dramatically simplifies and shrinks the code by eliminating the horrible CheckGEPInstructions method, fixes a miscompilation (@test3) and makes the code more aggressive. In particular, we now handle the @test4 case, which is reduced from the SmallPtrSet constructor. Missing this caused us to emit a variable length memset instead of a fixed size one. llvm-svn: 89922
* Generalize DecomposeGEPExpression to exactly handle what ↵Chris Lattner2009-11-261-30/+62
| | | | | | Value::getUnderlyingObject does (when TD is around). This allows us to avoid calling DecomposeGEPExpression unless the ultimate alias check we care about passes, speedup up BasicAA a bit. llvm-svn: 89920
OpenPOWER on IntegriCloud