- bb9caa9 Switch CodeMetrics itself over to use TTI to determine if an instruction by Chandler Carruth · 13 years ago
- 9fb823b Move all of the header files which are involved in modelling the LLVM IR by Chandler Carruth · 13 years ago
- 698e84f Remove the Function::getFnAttributes method in favor of using the AttributeSet by Bill Wendling · 13 years ago
- 4f6fb95 Add a new attribute, 'noduplicate'. If a function contains a noduplicate call, the call cannot be duplicated - Jump threading, loop unrolling, loop unswitching, and loop rotation are inhibited if they would duplicate the call. by James Molloy · 13 years ago
- 3d7b0b8 Rename the 'Attributes' class to 'Attribute'. It's going to represent a single attribute in the future. by Bill Wendling · 13 years ago
- ed0881b Use the new script to sort the includes of every file under lib. by Chandler Carruth · 13 years ago
- 8f46e91 Simplify code. No functionality change. by Jakub Staszak · 13 years ago
- c9b22d7 Create enums for the different attributes. by Bill Wendling · 13 years ago
- 863bab6 Remove the `hasFnAttr' method from Function. by Bill Wendling · 13 years ago
- 9a03c73 Do not pass an invalid domtree to SimplifyInstruction from by Peter Collingbourne · 13 years ago
- bf4b9af Second attempt at PR12573: by Bill Wendling · 14 years ago
- 712d85a Remove hack from r154987. The problem persists even with it, so it's not even a good hack. by Bill Wendling · 14 years ago
- 9f97595 Put this expensive check below the less expensive ones. by Bill Wendling · 14 years ago
- 4d4d025 Use a heavy hammer to fix PR12573. by Bill Wendling · 14 years ago
- 4442bfe Fix 12513: Loop unrolling breaks with indirect branches. by Andrew Trick · 14 years ago
- 4104ed9 whitespace by Andrew Trick · 14 years ago
- f62ffec Fast fix for PR12343: http://llvm.org/bugs/show_bug.cgi?id=12343 by Stepan Dyatkovskiy · 14 years ago
- 3904590 This pass didn't want the inline cost per-se, it just wants generic code by Chandler Carruth · 14 years ago
- 97b02fc llvm::SwitchInst by Stepan Dyatkovskiy · 14 years ago
- 5b648af Taken into account Duncan's comments for r149481 dated by 2nd Feb 2012: by Stepan Dyatkovskiy · 14 years ago
- 513aaa5 SwitchInst refactoring. by Stepan Dyatkovskiy · 14 years ago
- 2931a59 Fixed comment in loop-unswitch. by Stepan Dyatkovskiy · 14 years ago
- 7ec12e4 Cosmetic patch for r148215. by Stepan Dyatkovskiy · 14 years ago
- cb2adbac Fixup for r148132. Type replacement for LoopsProperties: from DenseMap to std::map, since we need to keep a valid pointer to properties of current loop. by Stepan Dyatkovskiy · 14 years ago
- d476fdc Speculatively revert r148132+r148133 to try and fix a buildbot failure. by Eli Friedman · 14 years ago
- 0a920fa Cosmetic patch for r148132. by Stepan Dyatkovskiy · 14 years ago
- cbcbdb2 LoopUnswitch: All helper data that is collected during loop-unswitch iterations was moved to separated class (LUAnalysisCache). by Stepan Dyatkovskiy · 14 years ago
- 8216569 Improved compile time: by Stepan Dyatkovskiy · 14 years ago
- d9d1de4 Fix typo "infinte". by Nick Lewycky · 14 years ago
- 3ba90a1 Add the actual code for r147175. by Chad Rosier · 14 years ago
- 1b7e2ba Speculatively revert r146578 to determine if it is the cause of a number of by Chad Rosier · 14 years ago
- d7b2bb3 Fix for bug #11429: Wrong behaviour for switches. Small improvement for code size heuristics. by Stepan Dyatkovskiy · 14 years ago
- f5b32e5 SplitBlockPredecessors uses ArrayRef instead of Data and Size. by Jakub Staszak · 14 years ago
- c24b86f Propagate TargetLibraryInfo throughout ConstantFolding.cpp and by Chad Rosier · 14 years ago
- 31798ef Potential bug in RewriteLoopBodyWithConditionConstant: use iterator should not be changed inside the uses enumeration loop. by Stepan Dyatkovskiy · 14 years ago
- 547b6c5 Stop emitting instructions with the name "tmp" they eat up memory and have to be uniqued, without any benefit. by Benjamin Kramer · 14 years ago
- 90f90da Split the landing pad basic block with the correct function. Also merge the by Bill Wendling · 14 years ago
- 61ea0e4 Reinstate r133513 (reverted in r133700) with an additional fix for a by Jay Foad · 14 years ago
- 9651312 Revert r133513: by Eric Christopher · 14 years ago
- a97a2c9 Reinstate r133435 and r133449 (reverted in r133499) now that the clang by Jay Foad · 14 years ago
- 184f3b3 Revert r133435 and r133449 to appease buildbots. by Chad Rosier · 14 years ago
- e03c05c Change how PHINodes store their operands. by Jay Foad · 14 years ago
- 6115824 Bail on unswitching a switch statement for a case with a critical edge. We name by Nick Lewycky · 14 years ago
- 9605a69 Simplify r132022 based on Cameron's feedback. by Evan Cheng · 14 years ago
- 73e6c09 Forgot dyn_cast check. by Evan Cheng · 14 years ago
- 1b55f56 Fix LoopUnswitch bug. RewriteLoopBodyWithConditionConstant can delete a dead by Evan Cheng · 14 years ago
- 99de19b Make LoopUnswitch preserve ScalarEvolution by just forgetting everything about by Cameron Zwarich · 15 years ago
- 43f8d16 Revamp the ValueMapper interfaces in a couple ways: by Chris Lattner · 15 years ago
- bb2cd02 Constant folding here is pointless, because InstructionSimplify by Duncan Sands · 15 years ago
- aef146b Factor code for testing whether replacing one value with another by Duncan Sands · 15 years ago
- b99f39b If dom tree information is available, make it possible to pass by Duncan Sands · 15 years ago
- a4fefc1 Passes do not need to recursively initialize passes that they preserve, if by Owen Anderson · 15 years ago
- 6c18d1a Get rid of static constructors for pass registration. Instead, every pass exposes an initializeMyPassFunction(), which by Owen Anderson · 15 years ago
- 229e38f Be more consistent in using ValueToValueMapTy. by Rafael Espindola · 15 years ago
- 8ac477f Begin adding static dependence information to passes, which will allow us to by Owen Anderson · 15 years ago
- df7a4f2 Now with fewer extraneous semicolons! by Owen Anderson · 15 years ago
- 0ad7d9c Fix loop unswitching's assumption that a code path which either by Dan Gohman · 15 years ago
- bc1a65a apparently unswitch had the same "Feature". Stop its by Chris Lattner · 15 years ago
- a7aed18 Reapply r110396, with fixes to appease the Linux buildbot gods. by Owen Anderson · 15 years ago
- bda59bd Revert r110396 to fix buildbots. by Owen Anderson · 15 years ago
- 755aceb Don't use PassInfo* as a type identifier for passes. Instead, use the address of the static by Owen Anderson · 15 years ago
- a57b97e Fix batch of converting RegisterPass<> to INTIALIZE_PASS(). by Owen Anderson · 15 years ago
- 0dc3c2d Use ValueMap instead of DenseMap. by Devang Patel · 15 years ago
- 9b21208 Pulled CodeMetrics out of InlineCost.h and made it a bit more general, so it can be reused from PartialSpecializationCost by Kenneth Uildriks · 15 years ago
- 66e809a remove a bunch of ad-hoc code to simplify instructions from by Chris Lattner · 16 years ago
- 5814d9d9 RewriteLoopBodyWithConditionConstant can end up rewriting the by Chris Lattner · 16 years ago
- c832c1b some code cleanups, use SwitchInst::findCaseValue, reduce indentation by Chris Lattner · 16 years ago
- ba93044 Code clean up. by Evan Cheng · 16 years ago
- ed66db3 Code refactoring. by Evan Cheng · 16 years ago
- 2734ebd Add a DominatorTree argument to isLCSSA so that it doesn't have to by Dan Gohman · 16 years ago
- 19d0b47 There are two ways of checking for a given type, for example isa<PointerType>(T) by Duncan Sands · 16 years ago
- 9dff9be Uniformize the names of type predicates: rather than having isFloatTy and by Duncan Sands · 16 years ago
- 302240d fix a crash in loop unswitch on a loop invariant vector condition. by Chris Lattner · 16 years ago
- a81a6df Convert a ton of simple integer type equality tests to the new predicate. by Benjamin Kramer · 16 years ago
- d9c355d Change errs() to dbgs(). by David Greene · 16 years ago
- 18fa568 Add Loop contains utility methods for testing whether a loop by Dan Gohman · 16 years ago
- 72c367f Dereference loopHeader after checking for null rather than before. by Dan Gohman · 16 years ago
- 2b8fed1 Reverting patch in revision 89758, initial attempt at fixing PR5373 has proven to be bogus. by Edward O'Callaghan · 16 years ago
- 5fd452d Fix for PR5373, Credit to Jakub Staszak. by Edward O'Callaghan · 16 years ago
- 46b5c64 remove a bunch of extraneous LLVMContext arguments by Chris Lattner · 16 years ago
- a83ac2d Update various Loop optimization passes to cope with the possibility that by Dan Gohman · 16 years ago
- 7a8db3a if CostMetrics says to never duplicate some code, don't unswitch a loop. by Chris Lattner · 16 years ago
- b6b8ec7 Restore LoopUnswitch's block-oriented threshold. LoopUnswitch now checks both by Dan Gohman · 16 years ago
- 92f8619 Use isVoidTy() by Devang Patel · 16 years ago
- a677136 Check void type before using RAUWd. by Devang Patel · 16 years ago
- 115741b Do not check use_empty() before replaceAllUsesWith(). This gives ValueHandles a chance to get properly updated. by Devang Patel · 16 years ago
- 2dc6f8d Use the new CodeMetrics class to compute code size instead of by Dan Gohman · 16 years ago
- 71ca652 Make LoopUnswitch's cost estimation count Instructions, rather than by Dan Gohman · 16 years ago
- deaf572 Reflow comment. by Mike Stump · 16 years ago
- c56af25 Fix an 80-column violation. by Dan Gohman · 16 years ago
- 3ddbc24 Re-apply r80926, with fixes: keep the domtree informed of new blocks by Dan Gohman · 16 years ago
- 9041995 Revert r80926. It causes loop unswitch assertion and slow down some JIT tests significantly. by Evan Cheng · 16 years ago
- 4c1bdcf Add a verifyAnalysis to LoopInfo, LoopSimplify, and LCSSA form that verify by Dan Gohman · 16 years ago
- 2dd09db eliminate VISIBILITY_HIDDEN from Transforms/Scalar. PR4861 by Chris Lattner · 16 years ago
- b25de3f eliminate the "Value" printing methods that print to a std::ostream. by Chris Lattner · 16 years ago
- 55f1c09 Push LLVMContexts through the IntegerType APIs. by Owen Anderson · 16 years ago
- 23a204d Move getTrue() and getFalse() to 2.5-like APIs. by Owen Anderson · 16 years ago
- b292b8c Move more code back to 2.5 APIs. by Owen Anderson · 16 years ago
- 0dd5e1e More migration to raw_ostream, the water has dried up around the iostream hole. by Daniel Dunbar · 16 years ago
- edb4a70 Revert the ConstantInt constructors back to their 2.5 forms where possible, thanks to contexts-on-types. More to come. by Owen Anderson · 16 years ago