aboutsummaryrefslogtreecommitdiff
path: root/lib/Transforms
AgeCommit message (Collapse)Author
2012-12-26Remove mid-optimizer warning. This situation should be handled differently,Nick Lewycky
such as by a compiler warning, a check in clang -fsanitizer=undefined, being optimized to unreachable, or a combination of the above. PR14722. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@171119 91177308-0d34-0410-b5e6-96231b3b80d8
2012-12-26LoopVectorizer: Optimize the vectorization of consecutive memory access when ↵Nadav Rotem
the iteration step is -1 git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@171114 91177308-0d34-0410-b5e6-96231b3b80d8
2012-12-26[msan] Raise alignment of origin stores/loads when possible.Evgeniy Stepanov
Origin alignment is as high as the alignment of the corresponding application location, but never less than 4. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@171110 91177308-0d34-0410-b5e6-96231b3b80d8
2012-12-26[msan] Expand the file comment with track-origins info.Evgeniy Stepanov
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@171109 91177308-0d34-0410-b5e6-96231b3b80d8
2012-12-26BBVectorize: Use VTTI to compute costs for intrinsics vectorizationHal Finkel
For the time being this includes only some dummy test cases. Once the generic implementation of the intrinsics cost function does something other than assuming scalarization in all cases, or some target specializes the interface, some real test cases can be added. Also, for consistency, I changed the type of IID from unsigned to Intrinsic::ID in a few other places. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@171079 91177308-0d34-0410-b5e6-96231b3b80d8
2012-12-25LoopVectorize: Enable vectorization of the fmuladd intrinsicHal Finkel
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@171076 91177308-0d34-0410-b5e6-96231b3b80d8
2012-12-25BBVectorize: Enable vectorization of the fmuladd intrinsicHal Finkel
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@171075 91177308-0d34-0410-b5e6-96231b3b80d8
2012-12-25[msan] Fix handling of vectors of pointers.Evgeniy Stepanov
VectorType::getInteger() can not be used with them, because pointer size depends on the target. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@171070 91177308-0d34-0410-b5e6-96231b3b80d8
2012-12-25[msan] Fix handling of select with vector condition.Evgeniy Stepanov
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@171069 91177308-0d34-0410-b5e6-96231b3b80d8
2012-12-25ASan: initialize callbacks from ASan module pass in a separate function for ↵Alexey Samsonov
consistency git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@171061 91177308-0d34-0410-b5e6-96231b3b80d8
2012-12-25ASan: move stack poisoning logic into FunctionStackPoisoner structAlexey Samsonov
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@171060 91177308-0d34-0410-b5e6-96231b3b80d8
2012-12-24Add LLVMContext::emitWarning methods and use them. <rdar://problem/12867368>Bob Wilson
When the backend is used from clang, it should produce proper diagnostics instead of just printing messages to errs(). Other clients may also want to register their own error handlers with the LLVMContext, and the same handler should work for warnings in the same way as the existing emitError methods. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@171041 91177308-0d34-0410-b5e6-96231b3b80d8
2012-12-24LoopVectorizer: When checking for vectorizable types, also checkNadav Rotem
the StoreInst operands. PR14705. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@171023 91177308-0d34-0410-b5e6-96231b3b80d8
2012-12-24Fix typo in commentsAlexey Samsonov
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@171021 91177308-0d34-0410-b5e6-96231b3b80d8
2012-12-24LoopVectorizer: Fix an endless loop in the code that looks for reductions.Nadav Rotem
The bug was in the code that detects PHIs in if-then-else block sequence. PR14701. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@171008 91177308-0d34-0410-b5e6-96231b3b80d8
2012-12-23LoopVectorize: Fix accidentaly inverted condition.Benjamin Kramer
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@171001 91177308-0d34-0410-b5e6-96231b3b80d8
2012-12-23LoopVectorize: For scalars and void types there is no need to compute vector ↵Benjamin Kramer
insert/extract costs. Fixes an assert during the build of oggenc in the test suite. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@171000 91177308-0d34-0410-b5e6-96231b3b80d8
2012-12-23Loop Vectorizer: Update the cost model of scatter/gather operations and makeNadav Rotem
them more expensive. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@170995 91177308-0d34-0410-b5e6-96231b3b80d8
2012-12-22Remove trailing whitespaceCraig Topper
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@170990 91177308-0d34-0410-b5e6-96231b3b80d8
2012-12-22Change 'AttrVal' to 'AttrKind' to better reflect that it's a kind of ↵Bill Wendling
attribute instead of the value of the attribute. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@170972 91177308-0d34-0410-b5e6-96231b3b80d8
2012-12-21Remove duplicate includes.Roman Divacky
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@170902 91177308-0d34-0410-b5e6-96231b3b80d8
2012-12-21[msan] Remove unreachable blocks before instrumenting a function.Evgeniy Stepanov
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@170883 91177308-0d34-0410-b5e6-96231b3b80d8
2012-12-21Enable if-conversion.Nadav Rotem
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@170841 91177308-0d34-0410-b5e6-96231b3b80d8
2012-12-21Every pass deserves a name, even codegenprep.Evan Cheng
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@170831 91177308-0d34-0410-b5e6-96231b3b80d8
2012-12-21BB-Vectorizer: Check the cost of the store pointer typeNadav Rotem
and not the return type, which is void. A number of test cases fail after adding the assertion in TTImpl. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@170828 91177308-0d34-0410-b5e6-96231b3b80d8
2012-12-21Fix a bug in the code that checks if we can vectorize loops while using dynamicNadav Rotem
memory bound checks. Before the fix we were able to vectorize this loop from the Livermore Loops benchmark: for ( k=1 ; k<n ; k++ ) x[k] = x[k-1] + y[k]; git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@170811 91177308-0d34-0410-b5e6-96231b3b80d8
2012-12-20LoopVectorize: Fix a bug in the scalarization of instructions.Nadav Rotem
Before if-conversion we could check if a value is loop invariant if it was declared inside the basic block. Now that loops have multiple blocks this check is incorrect. This fixes External/SPEC/CINT95/099_go/099_go git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@170756 91177308-0d34-0410-b5e6-96231b3b80d8
2012-12-20Loop Vectorizer: turn-off if-conversion.Nadav Rotem
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@170708 91177308-0d34-0410-b5e6-96231b3b80d8
2012-12-20Add a new attribute, 'noduplicate'. If a function contains a noduplicate ↵James Molloy
call, the call cannot be duplicated - Jump threading, loop unrolling, loop unswitching, and loop rotation are inhibited if they would duplicate the call. Similarly inlining of the function is inhibited, if that would duplicate the call (in particular inlining is still allowed when there is only one callsite and the function has internal linkage). git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@170704 91177308-0d34-0410-b5e6-96231b3b80d8
2012-12-20Formatting fixes. Remove some unnecessary 'else' after 'return'. No ↵Craig Topper
functional change. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@170676 91177308-0d34-0410-b5e6-96231b3b80d8
2012-12-20Removing trailing whitespaceCraig Topper
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@170675 91177308-0d34-0410-b5e6-96231b3b80d8
2012-12-20Loop Vectorizer: Enable if-conversion.Nadav Rotem
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@170632 91177308-0d34-0410-b5e6-96231b3b80d8
2012-12-20whitespaceNadav Rotem
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@170626 91177308-0d34-0410-b5e6-96231b3b80d8
2012-12-19Transform (x&C)>V into (x&C)!=0 where possiblePaul Redmond
When the least bit of C is greater than V, (x&C) must be greater than V if it is not zero, so the comparison can be simplified. Although this was suggested in Target/X86/README.txt, it benefits any architecture with a directly testable form of AND. Patch by Kevin Schoedel git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@170576 91177308-0d34-0410-b5e6-96231b3b80d8
2012-12-19[msan] Add track-origins argument to the pass constructor.Evgeniy Stepanov
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@170544 91177308-0d34-0410-b5e6-96231b3b80d8
2012-12-19[msan] Heuristically instrument unknown intrinsics.Evgeniy Stepanov
This changes adds shadow and origin propagation for unknown intrinsics by examining the arguments and ModRef behaviour. For now, only 3 classes of intrinsics are handled: - those that look like simple SIMD store - those that look like simple SIMD load - those that don't have memory effects and look like arithmetic/logic/whatever operation on simple types. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@170530 91177308-0d34-0410-b5e6-96231b3b80d8
2012-12-19LoopVectorize: Make iteration over induction variables not depend on pointer ↵Benjamin Kramer
values. MapVector is a bit heavyweight, but I don't see a simpler way. Also the InductionList is unlikely to be large. This should help 3-stage selfhost compares (PR14647). git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@170528 91177308-0d34-0410-b5e6-96231b3b80d8
2012-12-19Inline the 'hasIncompatibleWithVarArgsAttrs' method into its only uses. And ↵Bill Wendling
some minor comment reformatting. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@170516 91177308-0d34-0410-b5e6-96231b3b80d8
2012-12-19Rename the 'Attributes' class to 'Attribute'. It's going to represent a ↵Bill Wendling
single attribute in the future. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@170502 91177308-0d34-0410-b5e6-96231b3b80d8
2012-12-19Make sure the buffer, which containas an instance of APFloat, has proper ↵Shuxin Yang
alignment. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@170486 91177308-0d34-0410-b5e6-96231b3b80d8
2012-12-18rdar://12801297 Shuxin Yang
InstCombine for unsafe floating-point add/sub. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@170471 91177308-0d34-0410-b5e6-96231b3b80d8
2012-12-18Enable the loop vectorizer in clang and not in the pass manager, so that we ↵Nadav Rotem
can disable it in clang. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@170470 91177308-0d34-0410-b5e6-96231b3b80d8
2012-12-18LoopVectorize: Emit reductions as log2(vectorsize) shuffles + vector ops ↵Benjamin Kramer
instead of scalar operations. For example on x86 with SSE4.2 a <8 x i8> add reduction becomes movdqa %xmm0, %xmm1 movhlps %xmm1, %xmm1 ## xmm1 = xmm1[1,1] paddw %xmm0, %xmm1 pshufd $1, %xmm1, %xmm0 ## xmm0 = xmm1[1,0,0,0] paddw %xmm1, %xmm0 phaddw %xmm0, %xmm0 pextrb $0, %xmm0, %edx instead of pextrb $2, %xmm0, %esi pextrb $0, %xmm0, %edx addb %sil, %dl pextrb $4, %xmm0, %esi addb %dl, %sil pextrb $6, %xmm0, %edx addb %sil, %dl pextrb $8, %xmm0, %esi addb %dl, %sil pextrb $10, %xmm0, %edi pextrb $14, %xmm0, %edx addb %sil, %dil pextrb $12, %xmm0, %esi addb %dil, %sil addb %sil, %dl git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@170439 91177308-0d34-0410-b5e6-96231b3b80d8
2012-12-18Enable the loop vectorizer.Nadav Rotem
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@170416 91177308-0d34-0410-b5e6-96231b3b80d8
2012-12-18SROA: Replace calls to getScalarSizeInBits to DataLayout's API becauseNadav Rotem
getScalarSizeInBits could not handle vectors of pointers. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@170412 91177308-0d34-0410-b5e6-96231b3b80d8
2012-12-18Initialize NoRedZone and remove unused default values.Rafael Espindola
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@170404 91177308-0d34-0410-b5e6-96231b3b80d8
2012-12-17Fix another SROA crasher, PR14601.Chandler Carruth
This was a silly oversight, we weren't pruning allocas which were used by variable-length memory intrinsics from the set that could be widened and promoted as integers. Fix that. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@170353 91177308-0d34-0410-b5e6-96231b3b80d8
2012-12-17[msan] Fix lint warning.Evgeniy Stepanov
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@170347 91177308-0d34-0410-b5e6-96231b3b80d8
2012-12-17Teach the rewriting of memcpy calls to support subvector copies.Chandler Carruth
This also cleans up a bit of the memcpy call rewriting by sinking some irrelevant code further down and making the call-emitting code a bit more concrete. Previously, memcpy of a subvector would actually miscompile (!!!) the copy into a single vector element copy. I have no idea how this ever worked. =/ This is the memcpy half of PR14478 which we probably weren't noticing previously because it didn't actually assert. The rewrite relies on the newly refactored insert- and extractVector functions to do the heavy lifting, and those are the same as used for loads and stores which makes the test coverage a bit more meaningful here. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@170338 91177308-0d34-0410-b5e6-96231b3b80d8
2012-12-17Optimize tree walking in markAliveBlocks.Evgeniy Stepanov
Check whether a BB is known as reachable before adding it to the worklist. This way BB's with multiple predecessors are added to the list no more than once. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@170335 91177308-0d34-0410-b5e6-96231b3b80d8
generated by cgit v1.2.3-18-g5258 (git 2.32.0) at 2025-05-02 09:32:25 +0000