summaryrefslogtreecommitdiffstats
Commit message (Collapse)AuthorAgeFilesLines
...
* Update Xcode project.Ted Kremenek2008-11-181-0/+6
| | | | llvm-svn: 59478
* Added conditional guard 'if (CurLexer)' when using SetCommentRetentionState().Ted Kremenek2008-11-181-6/+6
| | | | | | | | | | | | This is because the PTHLexer will not support this method. Performance testing on preprocessing Cocoa.h shows that this results in a negligible performance difference (less than 1%). I tried making Lexer::SetCommentRetentionState() an out-of-line function (a precursor to making it a virtual function in PreprocessorLexer) and noticed a 1% decrease in speed (it is called in a hot part of the Preprocessor). llvm-svn: 59477
* Give SIToFPInst preference over UIToFPInst because it is faster on platforms ↵Devang Patel2008-11-182-3/+55
| | | | | | that are widely used. llvm-svn: 59476
* Avoid using a loop in ReleasePred and ReleaseSucc methods to compute theDan Gohman2008-11-183-55/+64
| | | | | | | | | | | | | new CycleBound value. Instead, just update CycleBound on each call. Also, make ReleasePred and ReleaseSucc methods more consistent accross the various schedulers. This also happens to make ScheduleDAGRRList's CycleBound computation somewhat more interesting, though it still doesn't have any noticeable effect, because no current targets that use the register-pressure reduction scheduler provide pipeline models. llvm-svn: 59475
* Change a bunch of uses of 'CurLexer->' to 'CurPPLexer->', which should be theTed Kremenek2008-11-181-40/+40
| | | | | | | | alias for the current PreprocessorLexer. No functionality change. Performance testing shows this results in no performance degradation when preprocessing Cocoa.h. llvm-svn: 59474
* - Add 'CurPPLexer' to Preprocessor to keep track of the currentTed Kremenek2008-11-182-9/+31
| | | | | | | | | PreprocessorLexer, which will either be a 'Lexer' or 'PTHLexer'. - Added stub field 'CurPTHLexer' to keep track of the current PTHLexer. - Modified IncludeStackInfo to track both the current PTHLexer and current PreprocessorLexer. llvm-svn: 59472
* While handling floating point IVs lift restrictions on initial value and ↵Devang Patel2008-11-172-26/+91
| | | | | | increment value. llvm-svn: 59471
* Implement effects of 'mutable', and a few comments from Chris on its parsing.Sebastian Redl2008-11-176-6/+32
| | | | llvm-svn: 59470
* Eliminate all of the placeholder identifiers used for constructors,Douglas Gregor2008-11-1718-167/+160
| | | | | | | | | destructors, and conversion functions. The placeholders were used to work around the fact that the parser and some of Sema really wanted declarators to have simple identifiers; now, the code that deals with declarators will use DeclarationNames. llvm-svn: 59469
* Fold assertion into second valid else branch. This removes a compiler warningTed Kremenek2008-11-171-3/+2
| | | | | | | | where the control reaches the end of a non-void function and also allows the compiler to generate better code. When this assertion is false we can easily add more else cases. llvm-svn: 59468
* Fix <rdar://problem/6333904> [sema] message lookup on super is incorrectSteve Naroff2008-11-173-1/+43
| | | | | | Missing special lookup rule in Sema::ActOnInstanceMessage(). llvm-svn: 59467
* Handle floating point ivs during doInitialization().Devang Patel2008-11-171-144/+130
| | | | llvm-svn: 59466
* Eliminate some trivial differences between the ScheduleNodeTopDownDan Gohman2008-11-172-6/+8
| | | | | | functions in these two schedulers. llvm-svn: 59465
* Add soft float support for a bunch more operations. OriginalDuncan Sands2008-11-1713-4/+356
| | | | | | | patch by Richard Osborne, tweaked and extended by your humble servant. llvm-svn: 59464
* Only test test/CodeGenObjC/synchronized.m as a i686-apple-darwin8 targetDouglas Gregor2008-11-171-1/+2
| | | | llvm-svn: 59463
* Updated IdentifierResolver to deal with DeclarationNames. The names ofDouglas Gregor2008-11-1716-70/+223
| | | | | | | | | | | | | | | | | | | | | C++ constructors, destructors, and conversion functions now have a FETokenInfo field that IdentifierResolver can access, so that these special names are handled just like ordinary identifiers. A few other Sema routines now use DeclarationNames instead of IdentifierInfo*'s. To validate this design, this code also implements parsing and semantic analysis for id-expressions that name conversion functions, e.g., return operator bool(); The new parser action ActOnConversionFunctionExpr takes the result of parsing "operator type-id" and turning it into an expression, using the IdentifierResolver with the DeclarationName of the conversion function. ActOnDeclarator pushes those conversion function names into scope so that the IdentifierResolver can find them, of course. llvm-svn: 59462
* Don't bother doing latency calculations in the "fast" scheduler.Dan Gohman2008-11-171-15/+5
| | | | llvm-svn: 59461
* Implement rdar://6319320: give a good diagnostic for cases where peopleChris Lattner2008-11-176-2/+22
| | | | | | | are trying to use the old GCC "casts as lvalue" extension. We don't and will hopefully never support this. llvm-svn: 59460
* Fix <rdar://problem/6316324> [sema] spurious warning on comparison of ↵Steve Naroff2008-11-172-2/+10
| | | | | | qualified id. llvm-svn: 59459
* Use SUnit's CycleBound field instead of duplicating it inDan Gohman2008-11-171-8/+9
| | | | | | a side-car datastructure llvm-svn: 59458
* Updated checker build.Ted Kremenek2008-11-171-1/+1
| | | | llvm-svn: 59457
* Remove these, which test for optimizations thatDale Johannesen2008-11-172-45/+0
| | | | | | are not currently done (cf PowerPC/README.txt). llvm-svn: 59456
* Move some former testcases (low-probability codegenDale Johannesen2008-11-171-0/+45
| | | | | | optimizations) into this wishlist. llvm-svn: 59455
* Let AnalyzeAlloca() remove debug intrinsics.Devang Patel2008-11-171-20/+11
| | | | llvm-svn: 59454
* Test case for objective-c's @synchronized statement.Fariborz Jahanian2008-11-171-0/+11
| | | | llvm-svn: 59451
* Don't produce ADDC/ADDE when expanding SHL unless they are legalRichard Osborne2008-11-172-1/+8
| | | | | | for the target. This fixes PR3080. llvm-svn: 59450
* Support dependencies between plugins by priority-sorting.Mikhail Glushenkov2008-11-174-5/+38
| | | | llvm-svn: 59449
* Filter ToolPropertiesList to exclude all Tools not mentioned in the ↵Mikhail Glushenkov2008-11-171-49/+78
| | | | | | compilation graph. llvm-svn: 59448
* Add a layer of indirection to make plugins more flexible.Mikhail Glushenkov2008-11-177-109/+111
| | | | | | | Use strings instead of TableGen defs in the compilation graph definition. Makes it easier for the plugins to modify an existing graph. llvm-svn: 59447
* Add TODO item for adding more source-range information to declaratorsDouglas Gregor2008-11-171-0/+21
| | | | llvm-svn: 59446
* Don't use the isPending flag to mean what the isAvailable flag means.Dan Gohman2008-11-172-2/+2
| | | | llvm-svn: 59445
* Fix <rdar://problem/6320086> parser rejects block capturing ivar.Steve Naroff2008-11-172-1/+23
| | | | llvm-svn: 59444
* Some cleanups for C++ operator overloadingDouglas Gregor2008-11-176-89/+68
| | | | llvm-svn: 59443
* Simplify error messages for two-parameter overloaded increment/decrement ↵Douglas Gregor2008-11-173-17/+6
| | | | | | operators llvm-svn: 59442
* Introduction the DeclarationName class, as a single, general method ofDouglas Gregor2008-11-1729-202/+890
| | | | | | | | representing the names of declarations in the C family of languages. DeclarationName is used in NamedDecl to store the name of the declaration (naturally), and ObjCMethodDecl is now a NamedDecl. llvm-svn: 59441
* pad header out to 80 colsChris Lattner2008-11-171-1/+1
| | | | llvm-svn: 59438
* add OperatorKinds.* to the project.Chris Lattner2008-11-171-0/+4
| | | | llvm-svn: 59437
* line up some columns, no functionality change.Chris Lattner2008-11-171-4/+4
| | | | llvm-svn: 59436
* A few corrections to the expr constant work. Not enabled at the Eli Friedman2008-11-171-2/+3
| | | | | | moment. llvm-svn: 59435
* move partial template specialization to FoldingSet.h.Zhongxing Xu2008-11-172-5/+6
| | | | llvm-svn: 59434
* More expr constant work. (Off by default).Anders Carlsson2008-11-171-8/+25
| | | | llvm-svn: 59433
* Fixed legalization of CONVERT_RNDSAT for integers.Mon P Wang2008-11-172-24/+16
| | | | llvm-svn: 59432
* Removed 2008-10-17-SpillerBug.ll as it does not provide an accurate test of ↵Lang Hames2008-11-161-130/+0
| | | | | | PR2898. llvm-svn: 59431
* Address some comments Eli had.Anders Carlsson2008-11-161-11/+6
| | | | llvm-svn: 59430
* use HandleConversionToBool() to check if a given cond is foldable (per Eli's ↵Nuno Lopes2008-11-161-3/+3
| | | | | | comment) llvm-svn: 59429
* More complex float evaluator support.Anders Carlsson2008-11-161-0/+45
| | | | llvm-svn: 59428
* Describe how constant folding and i-c-e diagnosing should work. Chris Lattner2008-11-161-0/+115
| | | | | | | Unfortunately, we're not here yet, but eventually Expr::isConstantExpr and friends should go away. llvm-svn: 59427
* rename Expr::tryEvaluate to Expr::Evaluate.Chris Lattner2008-11-167-18/+18
| | | | llvm-svn: 59426
* Add very limited support for evaluating complex floats.Anders Carlsson2008-11-161-1/+46
| | | | llvm-svn: 59425
* Trivial tidyingChris Lattner2008-11-161-2/+0
| | | | llvm-svn: 59424
OpenPOWER on IntegriCloud