| Commit message (Collapse) | Author | Age | Files | Lines | |
|---|---|---|---|---|---|
| * | Use zero-based shadow by default on Android. | Evgeniy Stepanov | 2012-05-23 | 1 | -8/+13 |
| | | | | | llvm-svn: 157318 | ||||
| * | Use zero-based shadow by default on Android. | Evgeniy Stepanov | 2012-05-23 | 1 | -2/+7 |
| | | | | | llvm-svn: 157317 | ||||
| * | PR1255(case ranges) related changes in Local Transformations. | Stepan Dyatkovskiy | 2012-05-23 | 1 | -10/+14 |
| | | | | | llvm-svn: 157315 | ||||
| * | asan: fix typo in comment | Dmitry Vyukov | 2012-05-23 | 1 | -1/+1 |
| | | | | | llvm-svn: 157314 | ||||
| * | Tidy up spacing. | Craig Topper | 2012-05-23 | 1 | -2/+2 |
| | | | | | llvm-svn: 157313 | ||||
| * | small refinement to r157218 to save a tiny amount of table size in the common | Chris Lattner | 2012-05-23 | 2 | -5/+13 |
| | | | | | | | case. llvm-svn: 157312 | ||||
| * | Correct the starting location for instantiations of field declarations which | Richard Smith | 2012-05-23 | 1 | -1/+1 |
| | | | | | | | | | | | | | | start with a cv-qualifier. DeclaratorDecl::getTypeSpecStartLoc() does not produce the location of the first type-specifier (the cv-qualifier) in this case, because we don't track source locations for cv-qualifiers. No test here: I've not found a way to test this with a lit-style test, and introducing a gtest test for this seems unwarranted. Suggestions welcome! Patch by Daniel Jasper! llvm-svn: 157311 | ||||
| * | If the first argument of __builtin_object_size can be folded to a constant | Richard Smith | 2012-05-23 | 3 | -5/+21 |
| | | | | | | | | | pointer, but such folding encounters side-effects, ignore the side-effects rather than performing them at runtime: CodeGen generates wrong code for __builtin_object_size in that case. llvm-svn: 157310 | ||||
| * | Fix indentation of wrapped line for readability. No functional change. | Craig Topper | 2012-05-23 | 1 | -1/+1 |
| | | | | | llvm-svn: 157309 | ||||
| * | Clang 3.1 is out: turn the relevant C++11 status boxes green. | Richard Smith | 2012-05-23 | 1 | -13/+13 |
| | | | | | llvm-svn: 157308 | ||||
| * | Move file to be more representative. | Eric Christopher | 2012-05-23 | 1 | -0/+0 |
| | | | | | llvm-svn: 157306 | ||||
| * | Check that we have an unknown upper bound range for our flexible array members. | Eric Christopher | 2012-05-23 | 1 | -1/+1 |
| | | | | | llvm-svn: 157305 | ||||
| * | Emit C++11 enum class information if it exists. | Eric Christopher | 2012-05-23 | 2 | -1/+16 |
| | | | | | | | Part of rdar://11496790 llvm-svn: 157304 | ||||
| * | Add support for C++11 enum classes in llvm. | Eric Christopher | 2012-05-23 | 4 | -4/+55 |
| | | | | | | | Part of rdar://11496790 llvm-svn: 157303 | ||||
| * | Add more convenience registers to x86_64 and a simple test scenario: | Johnny Chen | 2012-05-22 | 2 | -10/+28 |
| | | | | | | | | self.expect("expr -- $ax == (($ah << 8) | $al)", substrs = ['true']) llvm-svn: 157302 | ||||
| * | address some of John Criswell's comments | Nuno Lopes | 2012-05-22 | 1 | -31/+84 |
| | | | | | | | teach computeAllocSize about realloc, reallocf, and valloc llvm-svn: 157298 | ||||
| * | ARMDisassembler.cpp: Fix utf8 char in comments. | NAKAMURA Takumi | 2012-05-22 | 1 | -3/+3 |
| | | | | | llvm-svn: 157292 | ||||
| * | Integrated a fix for an ARM disassembler crash | Sean Callanan | 2012-05-22 | 1 | -3/+29 |
| | | | | | | | (0xd 0xd 0xa0 0xf4, or "vld2hs.8 {d0[], d1[]}, [r0]!") llvm-svn: 157289 | ||||
| * | Recognize the MS inheritance attributes and turn them into attributes | John McCall | 2012-05-22 | 10 | -20/+123 |
| | | | | | | | | | | on the RecordDecl. Persist the MS portability type attributes and ignore them in Sema rather than the parser. Patch by João Matos! llvm-svn: 157288 | ||||
| * | Fix line endings. | John McCall | 2012-05-22 | 1 | -23/+18 |
| | | | | | llvm-svn: 157287 | ||||
| * | Fix a typo. | Johnny Chen | 2012-05-22 | 1 | -1/+1 |
| | | | | | llvm-svn: 157278 | ||||
| * | Add a test case to check that eax's content equals the lower half of rax. | Johnny Chen | 2012-05-22 | 1 | -4/+36 |
| | | | | | | | Plus fix the test class name as well as wrong directory path. llvm-svn: 157277 | ||||
| * | The RegisterInfo descriptors for the convenience registers can specify an ↵ | Johnny Chen | 2012-05-22 | 1 | -1/+1 |
| | | | | | | | | | | | offset to be added to the offset as derived from the value_regs field, which is useful for future expansion purposes. As of now, we have: calculated_offset_of_eax = offset_of_rax + (offset_of_eax_from_the_descriptor which is 0) llvm-svn: 157275 | ||||
| * | Untabify and 80-col. | Eric Christopher | 2012-05-22 | 1 | -17/+17 |
| | | | | | llvm-svn: 157274 | ||||
| * | Formatting consistency. | Eric Christopher | 2012-05-22 | 1 | -3/+4 |
| | | | | | llvm-svn: 157273 | ||||
| * | Fix wrong offset of eax and friends pointed out by Greg. | Johnny Chen | 2012-05-22 | 1 | -1/+1 |
| | | | | | | | rdar://problem/11487457 llvm-svn: 157272 | ||||
| * | Also compute TopoSigs in synthetic register classes. | Jakob Stoklund Olesen | 2012-05-22 | 2 | -3/+8 |
| | | | | | | | | CodeGenRegisterClass has two constructors. Both need to compute the TopoSigs BitVector. llvm-svn: 157271 | ||||
| * | tsan: add shadow memory flush + fix few bugs | Dmitry Vyukov | 2012-05-22 | 11 | -59/+60 |
| | | | | | llvm-svn: 157270 | ||||
| * | Test that we emit a subrange type for vlas. | Eric Christopher | 2012-05-22 | 1 | -0/+9 |
| | | | | | | | Part of rdar://11457152 llvm-svn: 157269 | ||||
| * | tsan: fix sizeof sigset_t | Dmitry Vyukov | 2012-05-22 | 1 | -1/+1 |
| | | | | | llvm-svn: 157268 | ||||
| * | Fixes the autoconf build. | Manuel Klimek | 2012-05-22 | 1 | -2/+2 |
| | | | | | llvm-svn: 157266 | ||||
| * | hopefully fix the CMake build. sorry for breakage | Nuno Lopes | 2012-05-22 | 1 | -0/+1 |
| | | | | | llvm-svn: 157264 | ||||
| * | LSR fix: add a missing phi check during IV hoisting. | Andrew Trick | 2012-05-22 | 2 | -1/+59 |
| | | | | | | | Fixes PR12898: SCEVExpander crash. llvm-svn: 157263 | ||||
| * | wire -fbounds-checking to the new LLVM bounds checking pass | Nuno Lopes | 2012-05-22 | 7 | -20/+35 |
| | | | | | llvm-svn: 157262 | ||||
| * | add a new pass to instrument loads and stores for run-time bounds checking | Nuno Lopes | 2012-05-22 | 9 | -62/+357 |
| | | | | | | | | | move EmitGEPOffset from InstCombine to Transforms/Utils/Local.h (a draft of this) patch reviewed by Andrew, thanks. llvm-svn: 157261 | ||||
| * | Adds a method overwriteChangedFiles to the Rewriter. This is implemented by | Manuel Klimek | 2012-05-22 | 5 | -2/+235 |
| | | | | | | | | | | first writing the changed files to a temporary location and then overwriting the original files atomically. Also adds a RewriterTestContext to aid unit testing rewrting logic in general. llvm-svn: 157260 | ||||
| * | tsan: fix makefile | Dmitry Vyukov | 2012-05-22 | 1 | -0/+1 |
| | | | | | | | allow to build custom configurations llvm-svn: 157259 | ||||
| * | revert the usage of the objectsize intrinsic with 3 parameters (to match ↵ | Nuno Lopes | 2012-05-22 | 4 | -12/+8 |
| | | | | | | | LLVM r157255) llvm-svn: 157256 | ||||
| * | revert my previous patches that introduced an additional parameter to the ↵ | Nuno Lopes | 2012-05-22 | 15 | -276/+105 |
| | | | | | | | | | objectsize intrinsic. After a lot of discussion, we realized it's not the best option for run-time bounds checking llvm-svn: 157255 | ||||
| * | Only erase virtregs with no uses left. | Jakob Stoklund Olesen | 2012-05-22 | 2 | -4/+133 |
| | | | | | | | | | | Also make sure registers aren't erased twice if the dead def mentions the register twice. This fixes PR12911. llvm-svn: 157254 | ||||
| * | Remove some compilation warnings. | Filipe Cabecinhas | 2012-05-22 | 2 | -2/+2 |
| | | | | | llvm-svn: 157253 | ||||
| * | tsan: reduce per-thread memory usage | Dmitry Vyukov | 2012-05-22 | 9 | -21/+19 |
| | | | | | llvm-svn: 157252 | ||||
| * | Fix PR12858, a crash due to GVN's PRE not fully removing an instruction from the | Duncan Sands | 2012-05-22 | 2 | -6/+45 |
| | | | | | | | | | | | | | leader table. That's because it wasn't expecting instructions to turn up as leader for a value number that is not its own, but equality propagation could create this situation. One solution is to have the leader table use a WeakVH but this slows down GVN by about 5%. Instead just have equality propagation not add instructions to the leader table, only constants and arguments. In theory this might cause GVN to run more (each time it changes something it runs again) but it doesn't seem to occur enough to cause a slow down. llvm-svn: 157251 | ||||
| * | [asan] nuke some old unused code | Kostya Serebryany | 2012-05-22 | 1 | -51/+0 |
| | | | | | llvm-svn: 157250 | ||||
| * | [asan] increase the stack size limit to 256M (yes, that happens); also CHECK ↵ | Kostya Serebryany | 2012-05-22 | 2 | -8/+6 |
| | | | | | | | that the stack size is less than that on a non-main thread llvm-svn: 157249 | ||||
| * | tsan: simple memory profiler | Dmitry Vyukov | 2012-05-22 | 9 | -4/+110 |
| | | | | | llvm-svn: 157248 | ||||
| * | Replace inline asm constraint "=a" by the more general constraint "=r". | Simon Atanasyan | 2012-05-22 | 2 | -2/+2 |
| | | | | | | | That extend a range of platforms support this test case. llvm-svn: 157247 | ||||
| * | ScheduleOptimizer: Simplify some code | Tobias Grosser | 2012-05-22 | 1 | -22/+10 |
| | | | | | | | We now use isl_map_equate, which makes the code a lot simpler. llvm-svn: 157246 | ||||
| * | ScopInfo: Add parameter bounds to context | Tobias Grosser | 2012-05-22 | 4 | -2/+39 |
| | | | | | | | | | Derive the maximal and minimal values of a parameter from the type it has. Add this information to the scop context. This information is needed, to derive optimal types during code generation. llvm-svn: 157245 | ||||
| * | Replace some asserts with llvm_unreachable | Tobias Grosser | 2012-05-22 | 1 | -4/+4 |
| | | | | | llvm-svn: 157244 | ||||

