summaryrefslogtreecommitdiffstats
path: root/clang
Commit message (Collapse)AuthorAgeFilesLines
* Fix diagnostic pragmas.Argyrios Kyrtzidis2010-12-1523-117/+454
| | | | | | | | | | | | Diagnostic pragmas are broken because we don't keep track of the diagnostic state changes and we only check the current/latest state. Problems manifest if a diagnostic is emitted for a source line that has different diagnostic state than the current state; this can affect a lot of places, like C++ inline methods, template instantiations, the lexer, etc. Fix the issue by having the Diagnostic object keep track of the source location of the pragmas so that it is able to know what is the diagnostic state at any given source location. Fixes rdar://8365684. llvm-svn: 121873
* Move the work-in-progress implementation of variadic templates to its own ↵Douglas Gregor2010-12-153-29/+47
| | | | | | file in Sema. No functionality change. llvm-svn: 121869
* Reorganize LookupMemberExpr for clarity and to make the obvious fast pathsJohn McCall2010-12-151-267/+321
| | | | | | come first. llvm-svn: 121866
* __attribute__((nonnull)) can apply to reference-to-pointerDouglas Gregor2010-12-152-2/+13
| | | | | | parameters. Fixes <rdar://problem/8769025>. llvm-svn: 121864
* Sema: have BuildExpressionFromIntegralTemplateArgument produce well-formed ↵Peter Collingbourne2010-12-152-1/+16
| | | | | | | | | | | | | IntegerLiterals BuildExpressionFromIntegralTemplateArgument can produce malformed IntegerLiterals with an EnumType if the template parameter type is an EnumType. This breaks the AST printer which expects all IntegerLiterals to have a plain integer type. Instead, give the IntegerLiteral the enum's promotion type and wrap in an implicit cast to the EnumType. llvm-svn: 121862
* Silence GCC warning about control reaching the end of the function and ↵Chandler Carruth2010-12-151-0/+2
| | | | | | explicitly mark that all cases are handled. llvm-svn: 121855
* Fix gcc warning: 'clang::ASTStmtReader' is already a friend of ↵Nico Weber2010-12-151-2/+0
| | | | | | 'clang::OverloadExpr' llvm-svn: 121852
* Sundry missing lvalue-to-rvalue conversions. Also leave a TODO for the vitalJohn McCall2010-12-153-7/+19
| | | | | | | future task of performing contextual conversion to size_t in a VLA size expression. :) llvm-svn: 121836
* Set the "implicitly inline" bit on a method as soon as we see a definitionJohn McCall2010-12-154-24/+60
| | | | | | | | within the class. Teach IR gen to look for function definitions in record lexical contexts when deciding whether to emit a function whose address was taken. Fixes PR8789. llvm-svn: 121833
* Variadic templates: extend the Expr class with a bit that specifiesDouglas Gregor2010-12-1517-308/+572
| | | | | | | | | | | | | | | | | | whether the expression contains an unexpanded parameter pack, in the same vein as the changes to the Type hierarchy. Compute this bit within all of the Expr subclasses. This change required a bunch of reshuffling of dependency calculations, mainly to consolidate them inside the constructors and to fuse multiple loops that iterate over arguments to determine type dependence, value dependence, and (now) containment of unexpanded parameter packs. Again, testing is painfully sparse, because all of the diagnostics will change and it is more important to test the to-be-written visitor that collects unexpanded parameter packs. llvm-svn: 121831
* Function types are compatible (in the C sense) if their regparms are identical.John McCall2010-12-151-16/+20
| | | | llvm-svn: 121821
* Added missing IgnoreParens().Abramo Bagnara2010-12-1414-32/+47
| | | | llvm-svn: 121795
* Rewrite ComplexExprEvaluator::VisitCastExpr to use cast kinds, and fixJohn McCall2010-12-142-82/+127
| | | | | | | the basic casting logic to insert intermediate casts and preserve the exact complex-cast design. Fixes a crash in the test suite. llvm-svn: 121776
* Minor fix for clang-completion-mode, from Dve AbrahamsDouglas Gregor2010-12-141-2/+2
| | | | llvm-svn: 121771
* Improve some comments, shrink FunctionType::ExtInfo, and fix a bug foundJohn McCall2010-12-142-7/+16
| | | | | | | by valgrind where we were doing the wrong thing in the presence of invalid exception specs. llvm-svn: 121770
* Fix documentation typo, from nobled.Douglas Gregor2010-12-141-1/+1
| | | | llvm-svn: 121768
* Restore r121752 without modification.John McCall2010-12-1419-367/+347
| | | | llvm-svn: 121763
* Pull out r121752 in case it's causing the selfhost breakage.John McCall2010-12-1419-347/+367
| | | | llvm-svn: 121759
* Factor out most of the extra state in a FunctionProtoType into a separateJohn McCall2010-12-1419-367/+347
| | | | | | | class to be passed around. The line between argument and return types and everything else is kindof vague, but I think it's justifiable. llvm-svn: 121752
* Generalize this test to work without instruction names.Dan Gohman2010-12-141-2/+2
| | | | llvm-svn: 121742
* Implement CodeGen support for the may_alias attribute.Dan Gohman2010-12-132-0/+42
| | | | llvm-svn: 121734
* Variadic templates: extend Type, NestedNameSpecifier, TemplateName,Douglas Gregor2010-12-1320-65/+392
| | | | | | | | | | | | | | | | | | | | | | | | and TemplateArgument with an operation that determines whether there are any unexpanded parameter packs within that construct. Use this information to diagnose the appearance of the names of parameter packs that have not been expanded (C++ [temp.variadic]p5). Since this property is checked often (every declaration, ever expression statement, etc.), we extend Type and Expr with a bit storing the result of this computation, rather than walking the AST each time to determine whether any unexpanded parameter packs occur. This commit is deficient in several ways, which will be remedied with future commits: - Expr has a bit to store the presence of an unexpanded parameter pack, but it is never set. - The error messages don't point out where the unexpanded parameter packs were named in the type/expression, but they should. - We don't check for unexpanded parameter packs in all of the places where we should. - Testing is sparse, pending the resolution of the above three issues. llvm-svn: 121724
* Skip ParenType on function instantiations.Abramo Bagnara2010-12-133-3/+13
| | | | llvm-svn: 121720
* Remove a type that got reduced away from this test case but not actually ↵Chandler Carruth2010-12-131-1/+0
| | | | | | deleted. llvm-svn: 121694
* Fix PR8774 by restricting when hasInit returns true. Previously, itChandler Carruth2010-12-132-1/+20
| | | | | | | | | | | | | | would return true if the initializer pointer union had *any* non-null pointer in it, even if the pointer wasn't one that would actually be returned via getInit(). This makes it more accurately model the logic of 'getInit() != NULL'. This still isn't completely satisfying. From a principled stance, I suspect we should make hasInit() and getInit() *always* return false and NULL (resp.) for ParmVarDecl. We shouldn't at the API level treat initializers and default arguments as the same thing. llvm-svn: 121692
* Reduce the number of builtin operator overload candidates added in certainChandler Carruth2010-12-131-27/+89
| | | | | | | | | | | | cases. First, omit all builtin overloads when no non-record type is in the set of candidate types. Second, avoid arithmetic type overloads for non-arithmetic or enumeral types (counting vector types as arithmetic due to Clang extensions). When heavily using constructs such as STL's '<<' based stream logging, this can have a significant impact. One logging-heavy test case's compile time dropped by 10% with this. Self-host shows 1-2% improvement in compile time, but that's likely in the noise. llvm-svn: 121665
* Basic, Sema: add support for CUDA launch_bounds attributePeter Collingbourne2010-12-124-0/+55
| | | | llvm-svn: 121654
* Sema: diagnose kernel functions with non-void return typePeter Collingbourne2010-12-123-1/+22
| | | | llvm-svn: 121653
* Move the functionality to mark all vtables of key functions as used withinChandler Carruth2010-12-122-13/+13
| | | | | | | | | | | | | | a translation unit to the ActOnEndOfTranslationUnit function instead of doing it at the start of DefineUsedVTables. The latter is now called *recursively* during template instantiation, which causes an absolutely insane number of walks of every record decl in the translation unit. After this patch, an extremely template instantiation heavy test case's compile time drops by 10x, and we see between 15% and 20% improvement in average compile times across a project. This is just recovering a regression, it doesn't make anything faster than it was several weeks ago. llvm-svn: 121644
* Change MBlaze target to have the same description string as its LLVM equivalent.Wesley Peck2010-12-121-2/+1
| | | | llvm-svn: 121640
* Finish cleaning up the static utility code for adding builtin operator overloadChandler Carruth2010-12-121-62/+65
| | | | | | | candidates. They're now wrapped in nice APIs which hide the tables, etc. Also removes some repetitive code from clients. llvm-svn: 121634
* Clean up the helpers used to compute the usual arithmetic conversions' resultChandler Carruth2010-12-121-67/+56
| | | | | | | | | | | | type. Localize all of the logic within a single function rather than spreading it throughout the class. Also fixes a buglet where we failed to check for a RHS arithmetic type wider than the LHS and return its canonical type. I've yet to produce a test case that breaks because of this, but it was spotted by inspection by folks on the IRC channel and is obviously correct now. llvm-svn: 121633
* Add a comment to a helper function.Chandler Carruth2010-12-121-0/+2
| | | | llvm-svn: 121632
* Sink the logic to suppress builtin operator overloads in the presence ofChandler Carruth2010-12-121-52/+52
| | | | | | | | | | | user-defined operator overloads on the same enumeral types to the one place where it is used. In theory this removes wasted computation from several paths through this code, but I'm not aware of a case where it actually matters. This is mostly for cleanliness. llvm-svn: 121630
* Reorder the cases in the switch to be more logically grouped (to my mind). IfChandler Carruth2010-12-121-34/+34
| | | | | | | others have another ordering they would prefer, I'm all ears, but this one made it much easier for me to find the group of operators I'm interested in. llvm-svn: 121629
* Remove the final goto from this switch making it explict which overload set isChandler Carruth2010-12-121-2/+3
| | | | | | added for binary operator&. llvm-svn: 121628
* Fold away completely identical code with simple fallthrough.Chandler Carruth2010-12-121-7/+2
| | | | llvm-svn: 121627
* Move and copy function calls around to remove the indirection through gotosChandler Carruth2010-12-121-14/+10
| | | | | | from the switch statement. llvm-svn: 121626
* Simplify the flow through the switch by explicitly listing the added overloadsChandler Carruth2010-12-121-13/+7
| | | | | | for a few cases. llvm-svn: 121625
* Fix 80-column violations and reflowing some code to facilitate those fixes.Chandler Carruth2010-12-121-48/+59
| | | | llvm-svn: 121624
* Begin the refactoring of how builtin operators are added to the overloadChandler Carruth2010-12-121-736/+842
| | | | | | | | | | | | | | candidate set. This breaks apart a huge switch + goto system into distinct methods on a class. It also places the current mess of tables and other static state used in the process within that class. This is still a work in progress. I did a few simplifications that jumped out at me as I went, but I plan to iterate on this a bit before it's truly clean. However, this is easily the most invasive chunk. I benchmarked it on all-std-headers.cpp and an internal testcase that has a major hotspot in overload resolution and saw no real performance impact. llvm-svn: 121623
* Enhance my implementation of //rdar ://8747333 in r121597 to allowFariborz Jahanian2010-12-112-20/+44
| | | | | | | | for declaration of property setter/getter in forward class extensions and also skip over propeties which are @dynamic. llvm-svn: 121617
* Improved complex constants evaluation.Abramo Bagnara2010-12-112-2/+112
| | | | llvm-svn: 121616
* Any property declared in a class extension might have userFariborz Jahanian2010-12-103-3/+51
| | | | | | | | | | | declared setter or getter in current class extension or one of the other class extensions. Mark them as synthesized as property will be synthesized when property with same name is seen in the @implementation. This prevents bogus warning about unimplemented methods to be issued for these methods. Fixes // rdar://8747333 llvm-svn: 121597
* Add missing switch case for the quad-register version of the Neon vmul builtin.Bob Wilson2010-12-101-0/+1
| | | | llvm-svn: 121595
* Fix clang crashes on Neon vld[234]_dup intrinsics with 64-bit element types.Bob Wilson2010-12-101-0/+21
| | | | | | The 64-bit element vectors need to be handled as a special case. llvm-svn: 121592
* Introduce ObjCMessageExpr::getReceiverRange() to get the source range of the ↵Argyrios Kyrtzidis2010-12-102-1/+20
| | | | | | receiver. llvm-svn: 121517
* Keep the source location of the selector in ObjCMessageExpr.Argyrios Kyrtzidis2010-12-108-47/+78
| | | | llvm-svn: 121516
* Make RecursiveASTVisitor traverse a ObjCMethodDecl.Argyrios Kyrtzidis2010-12-101-1/+3
| | | | llvm-svn: 121515
* Do not assert on shifts of Neon polynomial types.Bob Wilson2010-12-101-4/+2
| | | | | | | | Most Neon shift intrinsics do not have variants for polynomial types, but vsri_n and vsli_n do support them, and we need to properly range-check the shift immediates for them. llvm-svn: 121509
OpenPOWER on IntegriCloud