summaryrefslogtreecommitdiffstats
Commit message (Collapse)AuthorAgeFilesLines
...
* Revert a failing test from r215261Justin Bogner2014-08-091-71/+0
| | | | | | | The (new) CoverageMapping/macroparams.c test is failing on a number of buildbots. Reverting it until Alex can investigate and fix the test. llvm-svn: 215274
* Implement some common file operations on Windows.Zachary Turner2014-08-094-62/+114
| | | | llvm-svn: 215273
* [modules] Weaken an out-of-date assertion: an #undef can have no location if weRichard Smith2014-08-096-1/+16
| | | | | | | imported it from a module when performing finalization before writing out an AST file. llvm-svn: 215272
* Move some X86 subtarget configuration onto the subtarget that's beingEric Christopher2014-08-092-21/+22
| | | | | | created. llvm-svn: 215271
* R600/SI: Custom lower CONCAT_VECTORSTom Stellard2014-08-092-3/+4
| | | | | | | This will lower them using register copies rather than loads and stores to the stack. llvm-svn: 215270
* R600/SI: Update concat_vectors.ll to check for scratch usageTom Stellard2014-08-091-0/+36
| | | | | | | | | | | | These tests were using SI-NOT: MOVREL to make sure concat vectors weren't being lowered to stack loads and stores, but we are using scratch buffers for the stack now instead of registers, so we need to add an additional SI-NOT check for scratch buffers. With this change I was able to uncover one broken test which will be fixed in a future commit. llvm-svn: 215269
* Refactor the module map file used for uniquing a module name out ofBen Langmuir2014-08-099-54/+81
| | | | | | | | | class Module. It's almost always going to be the same as getContainingModule() for top-level modules, so just add a map to cover the remaining cases. This lets us do less bookkeeping to keep the ModuleMap fields up to date. llvm-svn: 215268
* Fix expected windows result.Rafael Espindola2014-08-091-1/+1
| | | | llvm-svn: 215267
* Fix typo.Eric Christopher2014-08-092-2/+2
| | | | llvm-svn: 215266
* Fix a bug where SetFilePermissions was calling MakeDirectory.Zachary Turner2014-08-091-1/+1
| | | | | | This appears to have been a copy / paste error. llvm-svn: 215265
* Fixup Android tests build rulesAlexey Samsonov2014-08-092-2/+2
| | | | llvm-svn: 215264
* Remove some transient raw pointer ownership in ClangAttrEmitter::createArgumentDavid Blaikie2014-08-081-23/+28
| | | | | | | | This function might be a bit easier if it were split in two with a lot of early returns - and that setOptional bit in the outer function, but anyway. llvm-svn: 215263
* Fix bug causing FileSpec::GetPath() to crash with a null dest.Zachary Turner2014-08-081-0/+3
| | | | llvm-svn: 215262
* Add tests for coverage mapping generation.Alex Lorenz2014-08-0828-0/+1233
| | | | | | | | | | This patch adds the tests for the coverage mapping generation. Most of the tests check the mapping regions produced by the generator, and one checks the llvm IR. Differential Revision: http://reviews.llvm.org/D4793 llvm-svn: 215261
* Add predefined macros to identify x86_64h architectures.Bob Wilson2014-08-082-0/+11
| | | | | | Patch by Jim Grosbach. llvm-svn: 215260
* Predefine IB_DESIGNABLE and IBInspectable macros. <rdar://problem/17441860>Bob Wilson2014-08-082-4/+4
| | | | | | | | | These macros are used as markers for Interface Builder and need to be defined to empty strings since they have no impact on the code. Patch by Ted Kremenek. llvm-svn: 215259
* Add a cc1 "dump-coverage-mapping" for testing coverage mapping.Alex Lorenz2014-08-087-5/+71
| | | | | | Differential Revision: http://reviews.llvm.org/D4799 llvm-svn: 215258
* Simplify ownership of ExplodedGraph in the CoreEngine by removing unique_ptr ↵David Blaikie2014-08-082-37/+30
| | | | | | | | | | | | indirection. Summary: I was going to fix the use of raw pointer ownership in "takeGraph" when I realized that function was unused and the whole ExplodedGraph could just be owned by value without the std::unique_ptr indirection at all. Reviewers: jordan_rose Differential Revision: http://reviews.llvm.org/D4833 llvm-svn: 215257
* Fix some python shutdown / ordering issues.Zachary Turner2014-08-084-15/+21
| | | | | | | | Differential Revision: http://reviews.llvm.org/D4826 Reviewed by: Enrico Granata llvm-svn: 215256
* [MCJIT] Simplify immediate decoding code in the RuntimeDyldMachO hierarchy.Lang Hames2014-08-086-44/+49
| | | | | | | | | | | Cleanup only: no functional change. This patch makes RuntimeDyldMachO targets directly responsible for decoding immediates, rather than letting them implement catch a callback from generic code. Since this is a very target specific operation, it makes sense to let the target-specific code drive it. llvm-svn: 215255
* [FastISel][X86] Silence -Wenum-compare warningRui Ueyama2014-08-081-2/+6
| | | | llvm-svn: 215253
* Add -link-cxx-sanitizer driver flag.Alexey Samsonov2014-08-085-10/+22
| | | | | | | | | | | | | | | | | | | | | | Summary: This flag can be used to force linking of CXX-specific parts of sanitizer runtimes into the final executable. It gives more precise control than --driver-mode=g++ and comes handy when user links several object files with sanitized C++ code into an executable, but wants to provide libstdc++ himself, instead of relying on Clang dirver's behavior. Test Plan: clang regression test suite Reviewers: chandlerc, rsmith Reviewed By: rsmith Subscribers: cfe-commits Differential Revision: http://reviews.llvm.org/D4824 llvm-svn: 215252
* Extend tautological pointer compare and pointer to bool conversion warnings toRichard Trieu2014-08-085-1/+140
| | | | | | | | | | | | | | | macro arguments. Previously, these warnings skipped any code in a macro expansion. Preform an additional check and warn when the expression and context locations are both in the macro argument. The most obvious case not caught is passing a pointer directly to a macro, i.e 'assert(&array)' but 'assert(&array && "valid array")' is caught. This is because macro arguments are not typed and the conversion happens inside the macro. llvm-svn: 215251
* Objective-C [qoi]. Issue warning and fixit if property-dot syntaxFariborz Jahanian2014-08-085-5/+53
| | | | | | | use mis-cased property name (which is currently accepted silently due to the way property setters are named). rdar://17911746 llvm-svn: 215250
* Fix the windows build.Rafael Espindola2014-08-081-1/+1
| | | | | | Sorry for the noise. llvm-svn: 215249
* Reword comment slightly.Eric Christopher2014-08-082-8/+6
| | | | llvm-svn: 215248
* [CMake] Determine which compiler-rt libraries are supported onAlexey Samsonov2014-08-085-112/+185
| | | | | | | | a given platform in a top-level CMakeLists.txt to use it both in lib/ and in test/ subdirectories. Move architecture/platform checks to config-ix. llvm-svn: 215247
* Sure-up MemoryBuffer ownership in JSONCompilationDatabase's ctor.David Blaikie2014-08-082-4/+5
| | | | llvm-svn: 215246
* Change __ENVIRONMENT_MAC_OS_X_VERSION_MIN_REQUIRED__ for versions > 10.9.Bob Wilson2014-08-082-6/+19
| | | | | | | | The previous encoding only allowed a single digit for the minor version number. This changes it to use 2 digits for both the minor version and the revision number. llvm-svn: 215245
* Add an option to suppress the persistent result variable when running ↵Jim Ingham2014-08-083-0/+28
| | | | | | | | | | | | EvaluateExpression from Python. If you don't need to refer to the result in another expression, there's no need to bloat the persistent variable table with them since you already have the result SBValue to work with. <rdar://problem/17963645> llvm-svn: 215244
* Remove dead code. Fixes pr20544.Rafael Espindola2014-08-084-32/+6
| | | | llvm-svn: 215243
* Use llvm::sys::path::native instead of llvm::sys::fs::normalize_separators.Rafael Espindola2014-08-081-2/+2
| | | | | | | llvm::sys::path::native has a superset of the functionality and this was the only use of llvm::sys::fs::normalize_separators. llvm-svn: 215242
* Convert from Windows to Unix paths in sys::path::native.Rafael Espindola2014-08-081-1/+11
| | | | | | Part of pr20544. Test to follow in a second. llvm-svn: 215241
* Allow large immediates for branch instructions in 32bit mode.Joerg Sonnenberger2014-08-082-3/+23
| | | | llvm-svn: 215240
* Fix a thinko in the memory allocator that was causing us to strand an ↵Jim Ingham2014-08-081-7/+9
| | | | | | | | | | | | | | allocation range in all but one of the AllocatedBlocks that matched the requested permissions. Over time this would make the performance of expressions slow down considerably. Also added a little bit of logging that was helpful in resolving the issue. <rdar://problem/17954438> llvm-svn: 215239
* Provide an implementation of getNoopForMachoTarget for PPC, otherwiseJoerg Sonnenberger2014-08-083-2/+10
| | | | | | empty functions will assert in the MC object writer. llvm-svn: 215238
* [FastISel][X86] Fix INC/DEC optimization (r215230)Juergen Ributzka2014-08-082-1/+48
| | | | | | | | I accidentally also used INC/DEC for unsigned arithmetic which doesn't work, because INC/DEC don't set the required flag which is used for the overflow check. llvm-svn: 215237
* Partially revert r215204 - [mips] Add -mabicalls/-mno-abicalls to the driverDaniel Sanders2014-08-082-53/+40
| | | | | | | | It wasn't actually a bug that -mabicalls/-mno-abicalls wasn't being passed to GAS. The only reason we pass it to the integrated assembler is because it shares the same framework with CodeGen. llvm-svn: 215236
* Objective-C. Minor refactoring of my previous patch.Fariborz Jahanian2014-08-081-36/+29
| | | | | | rdar://17554063 llvm-svn: 215235
* Restore some reverted fixes for spelling, newlines and __linux__ macro.Deepak Panickal2014-08-0820-29/+37
| | | | llvm-svn: 215234
* AArch64: avoid deleting the current iterator in a loop.Tim Northover2014-08-081-3/+4
| | | | | | | | std::map invalidates the iterator to any element that gets deleted, which means we can't increment it correctly afterwards. This was causing Darwin test failures. llvm-svn: 215233
* Objective-C ARC. Use of non-retain/autorelease APIFariborz Jahanian2014-08-089-21/+69
| | | | | | | for building Objective-C array literals in ARC mode. rdar://17554063 llvm-svn: 215232
* [FastISel][AArch64] Attach MachineMemOperands to load and store instructions.Juergen Ributzka2014-08-081-17/+29
| | | | llvm-svn: 215231
* [FastISel][X86] Use INC/DEC when possible for {sadd|ssub}.with.overflow ↵Juergen Ributzka2014-08-082-31/+83
| | | | | | | | | | intrinsics. This is a small peephole optimization to emit INC/DEC when possible. Fixes <rdar://problem/17952308>. llvm-svn: 215230
* Fix a bug when scan-build is used in a cross-compilation environment withSylvestre Ledru2014-08-081-2/+13
| | | | | | | | | | | | | the --use-cc option. Instead, we will search in the PATH For example: scan-build --use-cc=arm-none-eabi-gcc -o out make -e Initially reported as a Debian Bug: https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=748777 llvm-svn: 215229
* DebugInfo: Recommit (reverted in r215217, originally committed in r215157) ↵David Blaikie2014-08-081-0/+1
| | | | | | | | | the assertion that no argument variable is overwritten by subsequent argument variables. This turned up a bug in clang where arguments were emitted with duplicate argument numbers (see r215227). llvm-svn: 215228
* DebugInfo: Blocks: Do not depend on LLVM argument numbering when choosing ↵David Blaikie2014-08-084-3/+30
| | | | | | | | | | | | the debug info argument numbering. Due to the possible presence of return-by-out parameters, using the LLVM argument number count when numbering debug info arguments can end up off-by-one. This could produce two arguments with the same number, which would in turn cause LLVM to emit only one of those arguments (whichever it found last) or assert (r215157). llvm-svn: 215227
* AArch64A57FPLoadBalancing.cpp: Define ColorNames in !NDEBUG.NAKAMURA Takumi2014-08-081-0/+2
| | | | llvm-svn: 215226
* DataTypes.h.cmake: Define PRIx32 &c for !HAVE_INTTYPES_H hosts.NAKAMURA Takumi2014-08-081-0/+7
| | | | | | I supposed PRIx32 might be unused in the tree. llvm-svn: 215225
* Delete dead code. NFC.Rafael Espindola2014-08-086-155/+0
| | | | llvm-svn: 215224
OpenPOWER on IntegriCloud