summaryrefslogtreecommitdiffstats
path: root/clang
Commit message (Collapse)AuthorAgeFilesLines
* Fix quoting of #pragma comment for MS compat, clang part.Michael Kuperstein2015-02-162-4/+12
| | | | | | | | | For #pragma comment(linker, ...) MSVC expects the comment string to be quoted, but for #pragma comment(lib, ...) the compiler itself quotes the library name. Since this distinction disappears by the time the directive reaches the backend, move quoting for the "lib" version to the frontend. Differential Revision: http://reviews.llvm.org/D7653 llvm-svn: 229376
* ASTMatchers: Replace enable_if with static_assert.Benjamin Kramer2015-02-161-7/+3
| | | | | | | This is nicer in general and gives a better error message, but it also might bring MSVC 2013 back to life. llvm-svn: 229374
* ASTMatchers: Make VariadicOperatorMatcher a real variadic template.Benjamin Kramer2015-02-161-119/+15
| | | | llvm-svn: 229370
* DR1467: If aggregate initialization encounters an initializer list for whichRichard Smith2015-02-162-1/+9
| | | | | | | | subobject initialization is not possible, be sure to note the overall initialization as having failed so that overload resolution knows that the relevant candidate is not viable. llvm-svn: 229353
* [X86] Teach clang to lower __builtin_ia32_psrldqi256 and ↵Craig Topper2015-02-162-2/+56
| | | | | | __builtin_ia32_pslldqi256 to vector shuffles the backend recognizes. This is a step towards removing the corresponding intrinsics from the backend. llvm-svn: 229348
* Removing LLVM_DELETED_FUNCTION, as MSVC 2012 was the last reason for ↵Aaron Ballman2015-02-1570-166/+165
| | | | | | requiring the macro. NFC; Clang edition. llvm-svn: 229339
* Partial revert of r229336; this wasn't intended to go in.Aaron Ballman2015-02-157-214/+62
| | | | llvm-svn: 229338
* Removing LLVM_EXPLICIT, as MSVC 2012 was the last reason for requiring the ↵Aaron Ballman2015-02-1527-83/+235
| | | | | | macro. NFC; Clang edition. llvm-svn: 229336
* Unbreak the build.Benjamin Kramer2015-02-151-1/+1
| | | | llvm-svn: 229329
* Make pushCleanup a variadic template. NFC.Benjamin Kramer2015-02-151-46/+5
| | | | llvm-svn: 229327
* Format: Make FormatToken's isOneOf a variadic templateBenjamin Kramer2015-02-151-20/+3
| | | | llvm-svn: 229326
* Analyzer: Replace a set of particularly ugly faux variadics.Benjamin Kramer2015-02-153-106/+14
| | | | llvm-svn: 229325
* Fix a typo in r229291 causing buildbot failure.Larisse Voufo2015-02-151-1/+1
| | | | llvm-svn: 229295
* Add more tests for crashes that happend to be fixed by r229288.Larisse Voufo2015-02-151-0/+214
| | | | llvm-svn: 229294
* Don't crash on `struct ::, struct ::` (and the same for enums).Larisse Voufo2015-02-151-214/+0
| | | | | | | | | | | | | | | | | | | The first part of that line doesn't parse correctly and ParseClassSpecifier() for some reason skips to tok::comma to recover, and then ParseDeclarationSpecifiers() sees the next struct and calls ParseClassSpecifier() again with the same DeclSpec object. However, the first call already called ActOnCXXGlobalScopeSpecifier() on the DeclSpec's CXXScopeSpec, and sema gets confused when this gets called again. As a fix, let ParseClassSpecifier() (and ParseEnumSpecifier()) call ParseOptionalCXXScopeSpec() with a temporary CXXScopeSpec object, and only copy it into the DeclSpec if things work out. (This is also how all the other functions that set the DeclSpec's TypeSpecScope set it.) Found by SLi's bot. llvm-svn: 229293
* More on fixing sized deallocation implementation logic: Fix PR21754.Larisse Voufo2015-02-152-5/+21
| | | | llvm-svn: 229291
* Add more tests for crashes that happend to be fixed by r229288.Nico Weber2015-02-151-0/+2
| | | | llvm-svn: 229289
* Don't crash on `struct ::, struct ::` (and the same for enums).Nico Weber2015-02-153-6/+25
| | | | | | | | | | | | | | | | | | | The first part of that line doesn't parse correctly and ParseClassSpecifier() for some reason skips to tok::comma to recover, and then ParseDeclarationSpecifiers() sees the next struct and calls ParseClassSpecifier() again with the same DeclSpec object. However, the first call already called ActOnCXXGlobalScopeSpecifier() on the DeclSpec's CXXScopeSpec, and sema gets confused when this gets called again. As a fix, let ParseClassSpecifier() (and ParseEnumSpecifier()) call ParseOptionalCXXScopeSpec() with a temporary CXXScopeSpec object, and only copy it into the DeclSpec if things work out. (This is also how all the other functions that set the DeclSpec's TypeSpecScope set it.) Found by SLi's bot. llvm-svn: 229288
* Wrap to 80 columns. No behavior change.Nico Weber2015-02-152-2/+4
| | | | llvm-svn: 229282
* Remove trailing whitespace to make test compatible with typeless pointer ↵David Blaikie2015-02-151-1/+1
| | | | | | migration llvm-svn: 229274
* Remove trailing whitespace to make test compatible with typeless pointer ↵David Blaikie2015-02-151-2/+2
| | | | | | migration llvm-svn: 229273
* Remove trailing whitespace to make test compatible with typeless pointer ↵David Blaikie2015-02-151-1/+1
| | | | | | migration llvm-svn: 229272
* Remove trailing whitespace to help make test compatible with typeless ↵David Blaikie2015-02-151-1/+1
| | | | | | pointer migration llvm-svn: 229271
* Update test to be ready for typeless pointer changesDavid Blaikie2015-02-151-1/+1
| | | | llvm-svn: 229270
* Update test to make it robust to migration to typeless pointersDavid Blaikie2015-02-151-3/+3
| | | | llvm-svn: 229269
* Replace snprintf with a Twine. NFC.Benjamin Kramer2015-02-141-5/+3
| | | | llvm-svn: 229264
* ItaniumMangle: Correctly mangle <base-unresolved-name>David Majnemer2015-02-142-33/+95
| | | | | | | | | | We had two bugs: - We were missing the "on" prefix for unresolved operators. - We didn't handle the mangling of destructors at all. This fixes PR22584. llvm-svn: 229255
* Make a reference const. No behavior change.Nico Weber2015-02-141-1/+1
| | | | llvm-svn: 229242
* Revise the implementation logic of sized deallocation: Do not automatically ↵Larisse Voufo2015-02-148-17/+38
| | | | | | | | generate weak definitions of the sized operator delete (in terms of unsized operator delete). Instead, provide the funcitonality via a new compiler flag, -fdef-sized-delete. The current implementation causes link-time ODR violations when the delete symbols are exported into the dynamic table. llvm-svn: 229241
* [modules] Accept //-style comments in module maps on purpose rather than byRichard Smith2015-02-142-1/+7
| | | | | | accident, and accept them even when they begin '//*'. llvm-svn: 229240
* Update cxx_dr_status to cover dr1748 and add test.Richard Smith2015-02-142-1/+36
| | | | llvm-svn: 229229
* Try to unbreak the Hexagon botDavid Majnemer2015-02-142-2/+2
| | | | llvm-svn: 229219
* DR1748: the reserved placement allocation functions have undefined behavior ifRichard Smith2015-02-145-16/+50
| | | | | | | they're given a null pointer as an argument, so we do not need to emit null checks on their results. llvm-svn: 229213
* CodeGen: _Atomic(_Complex) shouldn't crashDavid Majnemer2015-02-143-2/+16
| | | | | | | We could be a little kinder if we did a compare-exchange loop instead of an atomic-load/store pair. llvm-svn: 229212
* Revert "Revert r229082 for a bit, it caused PR22577."David Majnemer2015-02-1414-26/+181
| | | | | | | This reverts commit r229123. It was a red herring, the bug was present without r229082. llvm-svn: 229205
* CodeGen: Correctly convert atomic bool from i8 to i1David Majnemer2015-02-142-1/+11
| | | | | | | | | Bools are a little tricky, they are i8 in memory and must be coerced back to i1 before further operations can be performed on them. This fixes PR22577. llvm-svn: 229204
* Update test case to be compatible with auto-migration to new getelementptr ↵David Blaikie2015-02-141-1/+1
| | | | | | | | | | | syntax coming in the near future The first change won't touch GEPOperators such as these, but the update script only identifies them by the leading '(' after getelementptr or 'getelementptr inbounds', so update this test to at least have those features to allow auto-migrating. llvm-svn: 229198
* Remove trailing whitespace getting in the way of near-future getelementptr ↵David Blaikie2015-02-141-1/+1
| | | | | | change updates llvm-svn: 229197
* Adjust test case to be compatible with future changes to explicitly pass the ↵David Blaikie2015-02-141-2/+2
| | | | | | type to getelementptr llvm-svn: 229196
* [modules] If we have a choice between including a file textually and importingRichard Smith2015-02-136-2/+36
| | | | | | a prebuilt form from a module, prefer the modular form, all else being equal. llvm-svn: 229188
* [modules] Guard against recursively re-entering diagnosing ODR violations whileRichard Smith2015-02-131-0/+11
| | | | | | | in the middle of emitting an ODR violation diagnostic. I don't yet have a reduced testcase for this issue, working on it... llvm-svn: 229167
* [modules] Don't produce duplicate lookup results if the same declaration isRichard Smith2015-02-131-5/+12
| | | | | | visible through multiple imported modules. No functionality change. llvm-svn: 229147
* PR17938: This has already been fixed, add regression test.Richard Smith2015-02-131-0/+10
| | | | llvm-svn: 229146
* [modules] Improve llvm-bcanalyzer output on AST files a little. No ↵Richard Smith2015-02-134-7/+8
| | | | | | functionality change. llvm-svn: 229145
* Revert r229082 for a bit, it caused PR22577.Nico Weber2015-02-1314-181/+26
| | | | llvm-svn: 229123
* Remove a stray blank line that my sort_include.py script found.Chandler Carruth2015-02-131-1/+0
| | | | llvm-svn: 229093
* [PM] Fix #include order here that I forgot to update before changing it.Chandler Carruth2015-02-131-1/+1
| | | | llvm-svn: 229092
* [PM] Move away from the old llvm/PassManager.h header and its usingChandler Carruth2015-02-131-22/+22
| | | | | | | | | | | | declarations and just use the legacy namespace qualifier in this file and the permanent header name. The old wrapper header is going away to make LLVM's build more modular, and without updating Clang I can't easily start to add usage of the new pass manager to Clang. This should make it more clear in the code which set of types is doing what. llvm-svn: 229090
* MS ABI: Implement /volatile:msDavid Majnemer2015-02-1314-26/+181
| | | | | | | | | | | | The /volatile:ms semantics turn volatile loads and stores into atomic acquire and release operations. This distinction is important because volatile memory operations do not form a happens-before relationship with non-atomic memory. This means that a volatile store is not sufficient for implementing a mutex unlock routine. Differential Revision: http://reviews.llvm.org/D7580 llvm-svn: 229082
* [X86] Fix test cases that I foolishly copied and modified from another file ↵Craig Topper2015-02-131-4/+4
| | | | | | that had optimizations on. This caused the check patterns to not quite match. llvm-svn: 229073
OpenPOWER on IntegriCloud