| Commit message (Collapse) | Author | Age | Files | Lines | |
|---|---|---|---|---|---|
| * | Driver: Resolve inconsistency in matching options against options which are | Daniel Dunbar | 2009-11-19 | 2 | -7/+6 |
| | | | | | | | aliases -- just treat this case as an (unchecked) client error. llvm-svn: 89306 | ||||
| * | Use Option::matches instead of direct ID comparison. | Daniel Dunbar | 2009-11-19 | 3 | -13/+13 |
| | | | | | llvm-svn: 89305 | ||||
| * | Remove extra ';', found by clang++. | Daniel Dunbar | 2009-11-19 | 1 | -2/+2 |
| | | | | | llvm-svn: 89304 | ||||
| * | The sub-statement of a case statement is not an unevaluated context! | Eli Friedman | 2009-11-19 | 2 | -10/+34 |
| | | | | | llvm-svn: 89303 | ||||
| * | autoconf config.* claims to not know about auroraux triple. | Edward O'Callaghan | 2009-11-19 | 2 | -1/+8 |
| | | | | | llvm-svn: 89301 | ||||
| * | Teach IVUsers to keep things simpler and track loop-invariant strides only | Jim Grosbach | 2009-11-19 | 1 | -0/+10 |
| | | | | | | | | for uses inside the loop. This works better with LSR. Disabled behind -simplify-iv-users while benchmarking. llvm-svn: 89299 | ||||
| * | Eliminate duplicate phi nodes in loops. Loop rotation, for example, can ↵ | Jim Grosbach | 2009-11-19 | 1 | -0/+6 |
| | | | | | | | introduce these, and it's beneficial to later passes to clean them up. llvm-svn: 89298 | ||||
| * | Make EliminateDuplicatePHINodes() available as a utility function | Jim Grosbach | 2009-11-19 | 2 | -2/+9 |
| | | | | | llvm-svn: 89297 | ||||
| * | Test from Dhrystone to make sure that we're not emitting an aligned load for a | Bill Wendling | 2009-11-19 | 1 | -0/+28 |
| | | | | | | | string that's aligned at 8-bytes instead of 16-bytes. llvm-svn: 89295 | ||||
| * | Improve code completion for Objective-C message sends, so that we | Douglas Gregor | 2009-11-19 | 6 | -41/+155 |
| | | | | | | | | | | | | provide completion results before each keyword argument, e.g., [foo Method:arg WithArg1:arg1 WithArg2:arg2] We now complete before "WithArg1" and before "WithArg2", in addition to completing before "Method". llvm-svn: 89290 | ||||
| * | Improve instantiation control for rtti data and allow key functions to | Mike Stump | 2009-11-19 | 4 | -31/+96 |
| | | | | | | | instantiate a class. WIP. llvm-svn: 89289 | ||||
| * | Driver: Switch to using TableGen'erated Options.inc instead of Options.def file. | Daniel Dunbar | 2009-11-19 | 10 | -667/+745 |
| | | | | | llvm-svn: 89288 | ||||
| * | Refine linkage on thunks. WIP. | Mike Stump | 2009-11-19 | 1 | -2/+1 |
| | | | | | llvm-svn: 89287 | ||||
| * | Use TOOLALIAS so that clang++ gets 'built'. | Daniel Dunbar | 2009-11-19 | 1 | -0/+1 |
| | | | | | llvm-svn: 89286 | ||||
| * | Driver: Split OptTable out into OptTable.{h,cpp} | Daniel Dunbar | 2009-11-19 | 8 | -158/+193 |
| | | | | | llvm-svn: 89283 | ||||
| * | Add TOOLALIAS makefile variable; this defines an alternate name for a program | Daniel Dunbar | 2009-11-19 | 1 | -4/+45 |
| | | | | | | | | | | | which the makefiles will create by symlinking the actual tool to. - For use by clang, where we want to make 'clang++' and alias for clang (which enables C++ support in the driver) - Not sure this is the best approach, alternative suggestions welcome! llvm-svn: 89282 | ||||
| * | Improve the accuracy of code completion for the attributes of an @property ↵ | Douglas Gregor | 2009-11-19 | 2 | -15/+39 |
| | | | | | | | declaration, so that we don't produce completions for attributes that would conflict with already-specified attributes llvm-svn: 89281 | ||||
| * | The "ReadOnlyWithRel" enum seems to apply more to what Darwin does with the EH | Bill Wendling | 2009-11-19 | 2 | -5/+6 |
| | | | | | | | exception table than DataRel. llvm-svn: 89279 | ||||
| * | Twine: Stores kinds as uchar instead of bitfield to be friendlier to the | Daniel Dunbar | 2009-11-19 | 1 | -4/+4 |
| | | | | | | | optimizer. llvm-svn: 89278 | ||||
| * | Improve code-completion results for the flags in an @property | Douglas Gregor | 2009-11-19 | 5 | -38/+147 |
| | | | | | | | | | | declaration by providing patterns for "getter = <method>" and "setter = <method>". As part of this, invented a new "pattern" result kind that is merely a semantic string. The "pattern" result kind should help with other kinds of code templates. llvm-svn: 89277 | ||||
| * | There should be no need to keep renumbering blocks during tail duplication. | Bob Wilson | 2009-11-18 | 1 | -3/+0 |
| | | | | | llvm-svn: 89275 | ||||
| * | Fix buildbots. | Bob Wilson | 2009-11-18 | 1 | -1/+1 |
| | | | | | llvm-svn: 89274 | ||||
| * | Add XCore support for indirectbr / blockaddress. | Richard Osborne | 2009-11-18 | 5 | -1/+68 |
| | | | | | llvm-svn: 89273 | ||||
| * | De-bork CMake build | Douglas Gregor | 2009-11-18 | 1 | -0/+1 |
| | | | | | llvm-svn: 89272 | ||||
| * | Attempt #2: | Bill Wendling | 2009-11-18 | 2 | -7/+29 |
| | | | | | | | Place the EH table in the __TEXT section on MachO. It saves space. llvm-svn: 89270 | ||||
| * | Do not enter forward class 'Protocol' in decl context. | Fariborz Jahanian | 2009-11-18 | 2 | -1/+7 |
| | | | | | | | Will do it later. Fixes pr5552. llvm-svn: 89269 | ||||
| * | Do overload resolution for compound assignment even if only the RHS is ↵ | Sebastian Redl | 2009-11-18 | 2 | -5/+35 |
| | | | | | | | overloadable. Compound assignment may be overloaded as a non-member, and anyway the overload resolution is necessary because it triggers implicit (used-defined) conversions. Fixes PR5512, but not really the deeper issues lurking. Those are standard defects. llvm-svn: 89268 | ||||
| * | Rename CodeCompleteObjCProperty to something more specific. No functionality ↵ | Douglas Gregor | 2009-11-18 | 4 | -4/+4 |
| | | | | | | | change. llvm-svn: 89267 | ||||
| * | Resolve this FIXME: unelaborated lookups are ambiguous, too. | John McCall | 2009-11-18 | 1 | -5/+3 |
| | | | | | llvm-svn: 89266 | ||||
| * | Code completion after @dynamic | Douglas Gregor | 2009-11-18 | 5 | -15/+32 |
| | | | | | llvm-svn: 89265 | ||||
| * | Tail duplication still needs to iterate. Duplicating new instructions onto | Bob Wilson | 2009-11-18 | 2 | -3/+71 |
| | | | | | | | the tail of a block may make that block a new candidate for duplication. llvm-svn: 89264 | ||||
| * | Overhaul previous-declaration and overload checking to work on lookup results | John McCall | 2009-11-18 | 9 | -317/+498 |
| | | | | | | | | rather than NamedDecl*. This is a major step towards eliminating OverloadedFunctionDecl. llvm-svn: 89263 | ||||
| * | Test case for parsing-recovery change to @synthesized | Douglas Gregor | 2009-11-18 | 1 | -0/+15 |
| | | | | | llvm-svn: 89260 | ||||
| * | Code completion for Objective-C @synthesized. | Douglas Gregor | 2009-11-18 | 5 | -14/+150 |
| | | | | | llvm-svn: 89259 | ||||
| * | ignore parens surounding the type when diagnosing | Fariborz Jahanian | 2009-11-18 | 2 | -1/+3 |
| | | | | | | | pointer-to-member cast types used in expressions. llvm-svn: 89255 | ||||
| * | Add another statistic to measure code size due to tail duplication. | Bob Wilson | 2009-11-18 | 1 | -0/+3 |
| | | | | | llvm-svn: 89254 | ||||
| * | Remove spurious @verbatim. Patch by Timo Juhani Lindfors! | Jeffrey Yasskin | 2009-11-18 | 1 | -1/+1 |
| | | | | | llvm-svn: 89252 | ||||
| * | This patch fixes a bug in misdiagnosing correct | Fariborz Jahanian | 2009-11-18 | 5 | -16/+17 |
| | | | | | | | use of pointer to data member. llvm-svn: 89251 | ||||
| * | Not all ASM has # for comments. | Bill Wendling | 2009-11-18 | 1 | -1/+1 |
| | | | | | llvm-svn: 89250 | ||||
| * | Track overriding methods when instantiating a template class. Fixes PR5550. | Sebastian Redl | 2009-11-18 | 4 | -18/+42 |
| | | | | | llvm-svn: 89248 | ||||
| * | Make MSVC happy. | Daniel Dunbar | 2009-11-18 | 1 | -12/+12 |
| | | | | | llvm-svn: 89247 | ||||
| * | Fix PR5300. | Jakob Stoklund Olesen | 2009-11-18 | 2 | -12/+34 |
| | | | | | | | | | When TwoAddressInstructionPass deletes a dead instruction, make sure that all register kills are accounted for. The 2-addr register does not get special treatment. llvm-svn: 89246 | ||||
| * | TableGen: Add initial backend for clang Driver's option parsing. | Daniel Dunbar | 2009-11-18 | 3 | -10/+248 |
| | | | | | llvm-svn: 89245 | ||||
| * | Pretend destructors are const and volatile. This allows calling them with ↵ | Sebastian Redl | 2009-11-18 | 2 | -3/+13 |
| | | | | | | | const and/or volatile objects. Fixes PR5548. llvm-svn: 89244 | ||||
| * | Don't generate superfluous and ambiguous built-in candidates for multi-level ↵ | Sebastian Redl | 2009-11-18 | 2 | -0/+22 |
| | | | | | | | array subscript and arithmetic. Fixes PR5546. llvm-svn: 89242 | ||||
| * | Allow the machine verifier to be run outside the PassManager. | Jakob Stoklund Olesen | 2009-11-18 | 2 | -18/+161 |
| | | | | | | | Verify LiveVariables information when present. llvm-svn: 89241 | ||||
| * | Remove the -early-coalescing option | Jakob Stoklund Olesen | 2009-11-18 | 2 | -147/+0 |
| | | | | | llvm-svn: 89240 | ||||
| * | Construct INTPTR_MIN, INTPTR_MAX, and UINTPTR_MAX from the exact-width limit | Ken Dyck | 2009-11-18 | 1 | -9/+4 |
| | | | | | | | macros corresponding to __INTPTR_WIDTH__. llvm-svn: 89239 | ||||
| * | Fixed the in-place spiller and trivial rewriter, which had been broken by ↵ | Lang Hames | 2009-11-18 | 3 | -221/+56 |
| | | | | | | | the recent SlotIndexes work. llvm-svn: 89238 | ||||
| * | Define intptr_t and uintptr_t in terms of their equivalent exact-width types. | Ken Dyck | 2009-11-18 | 2 | -26/+27 |
| | | | | | llvm-svn: 89237 | ||||

