summaryrefslogtreecommitdiffstats
Commit message (Collapse)AuthorAgeFilesLines
...
* Revert most of r154844, which was disabled in r155975. Keep around theRichard Smith2012-05-0215-271/+18
| | | | | | | refactorings in that revision, and some of the subsequent bugfixes, which seem to be relevant even without delayed exception specification parsing. llvm-svn: 156031
* Thread safety analysis: additional support for smart pointers in lock ↵DeLesley Hutchins2012-05-022-18/+35
| | | | | | expressions. llvm-svn: 156030
* Teach DAGCombine the same multiply-by-1.0 folding trick when doing FMAs, ↵Owen Anderson2012-05-022-0/+27
| | | | | | just like it now knows for FMULs. llvm-svn: 156029
* For Intel Atom, use ILP scheduling always, instead of ILP for 64 bitPreston Gurd2012-05-021-4/+4
| | | | | | | and Hybrid for 32 bit, since benchmarks show ILP scheduling is better most of the time. llvm-svn: 156028
* Fix some doxycomments using \class instead of \brief.David Blaikie2012-05-022-6/+7
| | | | | | Patches by Csaba Raduly (rcsaba@gmail.com) llvm-svn: 156027
* Patched our local Clang to fix a crash when parsingSean Callanan2012-05-021-0/+26
| | | | | | | classes/structs that have variable-length arrays at the end. llvm-svn: 156026
* Change the Intel Atom detection code to recognizePreston Gurd2012-05-022-3/+5
| | | | | | Lincroft and Medfield. llvm-svn: 156025
* Teach DAG combine that multiplication by 1.0 can always be constant folded.Owen Anderson2012-05-022-0/+12
| | | | llvm-svn: 156023
* [python] Add testcase for annotation cursorAnders Waldenborg2012-05-021-0/+12
| | | | llvm-svn: 156022
* Add tools/lld to .gitignore.Michael J. Spencer2012-05-021-0/+2
| | | | llvm-svn: 156021
* Update the docs to reflect the recent changes to categoriesEnrico Granata2012-05-021-9/+21
| | | | llvm-svn: 156020
* ARM: Add missing two-operand VBIC aliases.Jim Grosbach2012-05-022-0/+7
| | | | llvm-svn: 156019
* Mentioning in the docs that Python commands can have docstringsEnrico Granata2012-05-021-0/+7
| | | | llvm-svn: 156018
* [python] Add Cursor.enum_value wrapping clang_getEnumConstantDeclValueAnders Waldenborg2012-05-022-0/+69
| | | | llvm-svn: 156017
* Walk the decls looking for the last one that has an attribute. We do have to ↵Rafael Espindola2012-05-022-6/+13
| | | | | | | | | | | | | | | | | | walk them, otherwise we cannot produce an error for both struct HIDDEN test4; // canonical struct test4; struct DEFAULT test4; and struct test5; // canonical struct HIDDEN test5; struct DEFAULT test5; llvm-svn: 156016
* [python] Add testcase for enum with specified underlaying typeAnders Waldenborg2012-05-021-0/+8
| | | | llvm-svn: 156009
* [python] Run tests for c++ with std=c++11Anders Waldenborg2012-05-021-2/+3
| | | | llvm-svn: 156008
* [python] Fix warning in c-code inside testcaseAnders Waldenborg2012-05-021-1/+1
| | | | llvm-svn: 156007
* Thread Safety Analysis: fixed attribute handling for lock_returned attribute.DeLesley Hutchins2012-05-022-3/+10
| | | | llvm-svn: 156005
* Modern objective-c translation: Fixing couple of bugsFariborz Jahanian2012-05-022-2/+62
| | | | | | | related to laying out ivar structs and accessing non-fragile-ivar in more compilated cases. // rdar://11323187 llvm-svn: 156004
* Replace the StringSwitch in AttributeList::getKind() with aDouglas Gregor2012-05-022-6/+20
| | | | | | | | TableGen-generated StringMatcher, for a 1.2% speedup in -fparse-only time in <rdar://problem/11004361>. Thanks to Benjamin for pointing me at StringMatcher! llvm-svn: 156003
* Move llvm-tblgen's StringMatcher into the TableGen library so it canDouglas Gregor2012-05-026-4/+4
| | | | | | be used by clang-tblgen. llvm-svn: 156000
* Introduce the notion of 'ignored' attributes, so that all attributesDouglas Gregor2012-05-023-7/+25
| | | | | | we accept are not modeled somehow via Attr.td. llvm-svn: 155998
* [llvm-c] Make a few function declarations proper prototypesAnders Waldenborg2012-05-021-6/+6
| | | | | | | | | This avoids warnings when included in a application that uses -Wstrict-prototypes. e.g: AsmPrinters.def:27:1: warning: function declaration isn't a prototype [-Wstrict-prototypes] llvm-svn: 155997
* This patch continues the work of adding instruction latencies for X86 Atom,Preston Gurd2012-05-024-68/+171
| | | | | | by providing the latencies for the instructions in X86InstrFPStack.td. llvm-svn: 155996
* Introduce the notion of an attribute that has no direct representationDouglas Gregor2012-05-024-23/+89
| | | | | | as an AST node, and fold a number of such attributes into Attr.td. llvm-svn: 155995
* Fix bug in cxa_demangle involving template substitution.Howard Hinnant2012-05-022-6/+10
| | | | llvm-svn: 155994
* Fix same line doxycomments so they comment the adjacent field.David Blaikie2012-05-021-69/+69
| | | | | | | | | | Without the '<' prefix in the doxycomment these comments were incorrectly attached to the proceeding comment on the next line, rather than the preceeding one. Fixes PR12722 llvm-svn: 155993
* Revert r155853Manman Ren2012-05-022-37/+0
| | | | | | | | The commit is intended to fix rdar://10961709. But it is the root cause of PR12720. Revert it for now. llvm-svn: 155992
* Implement the alternate spellings of attributes aligned (as "align")Douglas Gregor2012-05-022-4/+2
| | | | | | | and const (as "__const") using tblgen, rather than explicitly hacking them in. llvm-svn: 155991
* Driver: Turn the default value for -fmath-errno into a proper target hook ↵Benjamin Kramer2012-05-024-4/+23
| | | | | | | | | | | and disable it by default on more platforms. For now -fno-math-errno is the default on BSD-derived platforms (Darwin, DragonFlyBSD, FreeBSD, NetBSD, OpenBSD). If the default is not right for your platform, please yell. I only verified the result with the default compilers on Darwin and FreeBSD. llvm-svn: 155990
* Revert my optimization to AttributeList::getKind() in r155987;Douglas Gregor2012-05-022-33/+20
| | | | | | Benjamin has suggested a better approach. llvm-svn: 155989
* Replace a ConsumeAnyToken() call with ConsumeToken() when we knowDouglas Gregor2012-05-021-1/+1
| | | | | | | we're looking at a normal token. Tiny positive impact -fsyntax-only time for <rdar://problem/11004361>. llvm-svn: 155988
* Replace the StringSwitch in AttributeList::getKind(constDouglas Gregor2012-05-022-20/+33
| | | | | | | IdentifierInfo *) with a static StringMap, improving -fsyntax-only performance by 1% for the example in <rdar://problem/11004361>. llvm-svn: 155987
* [tsan] typo and style (thanks to Nick Lewycky)Kostya Serebryany2012-05-021-9/+9
| | | | llvm-svn: 155986
* The value held in the vector may be RAUW'ed by some of the canonicalizationBill Wendling2012-05-022-2/+53
| | | | | | | methods. Use a weak value handle to keep up with this. PR12245 llvm-svn: 155984
* Disallow YIELD and other allocated nop hints in pre-ARMv6 architectures.Richard Barton2012-05-025-33/+26
| | | | llvm-svn: 155983
* Add support for selecting AVX2 vpshuflw and vpshufhw. Add decoding support ↵Craig Topper2012-05-025-36/+104
| | | | | | for AsmPrinter. llvm-svn: 155982
* Fix forwarding of -fpack-struct from driver to CC1, and add a test.James Molloy2012-05-022-4/+14
| | | | | | | | -fpack-struct's handling has changed in CC1 (one of only two flags that needed changing) because the driver treats "-fpack-struct" as a boolean flag, and CC1 (did) treat it as an option with a separated value. This change causes -fpack-struct=X to be forwarded correctly to -fpack-struct=X instead of erroneously to "-fpack-struct X" llvm-svn: 155981
* Don't try to instantiate std::list at an incomplete type; it's notJohn McCall2012-05-021-1/+1
| | | | | | | allowed, and it'll blow up in unfortunate ways when using a proper C++11 library. llvm-svn: 155980
* Update SmallVector to support move semantics if the host does.John McCall2012-05-022-14/+231
| | | | | | | | | | | | | Note that support for rvalue references does not imply support for the full set of move-related STL operations. I've preserved support for an odd little thing in insert() where we're trying to support inserting a new element from an existing one. If we actually want to support that, there's a lot more we need to do: insert can call either grow or push_back, neither of which is safe against this particular use pattern. llvm-svn: 155979
* Fix unintentional use of operator bool.John McCall2012-05-021-1/+1
| | | | llvm-svn: 155978
* Fix cmake. Remaining two ClangCC1Options deps.Andrew Trick2012-05-022-5/+3
| | | | llvm-svn: 155977
* Fix the implementation of MachOObjectFile::isSectionZeroInit so it follows ↵Eli Friedman2012-05-021-4/+6
| | | | | | the MachO spec. llvm-svn: 155976
* Disable our non-standard delayed parsing of exception specifications. DelayingRichard Smith2012-05-025-36/+15
| | | | | | | the parsing of such things appears to be a conforming extension, but it breaks libstdc++4.7's std::pair. llvm-svn: 155975
* Speculatively fix windows cmake builders.Andrew Trick2012-05-021-1/+1
| | | | | | Really just a wild stab in the dark. llvm-svn: 155974
* If a command takes options and arguments, the help text should warn the user ↵Jim Ingham2012-05-021-3/+15
| | | | | | to use "--" to terminate the options. llvm-svn: 155973
* Refine analyzer diagnostics by adding an expression "cone-of-influence" to ↵Ted Kremenek2012-05-024-45/+378
| | | | | | | | | | | reverse track interesting values through interesting expressions. This allows us to map from interesting values in a caller to interesting values in a caller, thus recovering some precision in diagnostics lost from IPA. Fixes <rdar://problem/11327497> llvm-svn: 155971
* Jeeze... Remove two unneeded #include's of ThreadPlanTestCondition.h, and ↵Jim Ingham2012-05-022-2/+2
| | | | | | | | replace them with includes of ClangUserExpression that were being errantly dragged in through same. llvm-svn: 155970
* Unrevert r155951, reverted in r155962, with two changes:Richard Smith2012-05-023-15/+54
| | | | | | | | | * Work around build failures due to gcc 4.2 bugs. * Remove BodyIndexer::TraverseCXXOperatorCallExpr, which was not being called prior to this change, and whose presence disables a RecursiveASTVisitor stack space optimization after this change. llvm-svn: 155969
OpenPOWER on IntegriCloud