Commit Graph

977 Commits

Author SHA1 Message Date
Chris Lattner 8a651c7116 Don't warn about unused static functions if they are marked with
attr constructor or destructor.  Patch by Jean-Daniel Dupas!

llvm-svn: 100870
2010-04-09 17:25:05 +00:00
Douglas Gregor f19ac0ede9 Downgrade the "declaration does not declare anything" error to a
warning. It's not harmful to have such pointless declarations, and GCC
does not diagnose this issue consistently.

llvm-svn: 100814
2010-04-08 21:33:23 +00:00
Jeffrey Yasskin 54eba427ed Explain that a template needs arguments to make it into a type, for
variable declarations.

llvm-svn: 100809
2010-04-08 21:04:54 +00:00
Jeffrey Yasskin c76498d409 Make CXXScopeSpec invalid when incomplete, and propagate that into any
Declarator that depends on it.  This fixes several redundant errors and bad
recoveries.

llvm-svn: 100779
2010-04-08 16:38:48 +00:00
John McCall b9c7848ba7 Implement dependent friend function template specializations.
llvm-svn: 100753
2010-04-08 09:05:18 +00:00
Jeffrey Yasskin 4e150f8d35 Fix some redundant errors by changing CXXScopeSpec::isSet calls into
isNotEmpty calls.

llvm-svn: 100722
2010-04-07 23:29:58 +00:00
Fariborz Jahanian 6a0a2e0ccc Patch to not build ivar ASTs when they are ilegally
declared in categories.

llvm-svn: 100577
2010-04-06 22:43:48 +00:00
Daniel Dunbar 229385cd68 Sema/Obj-C: Simplify -- eliminate dead casts & provable asserts, and narrow type.
llvm-svn: 100197
2010-04-02 18:29:09 +00:00
Douglas Gregor f5af3584ca Improve diagnostics when an elaborated-type-specifer containing a
nested-name-specifier (e.g., "class T::foo") fails to find a tag
member in the scope nominated by the
nested-name-specifier. Previously, we gave a bland

  error: 'Nested' does not name a tag member in the specified scope

which didn't actually say where we were looking, which was rather
horrible when the nested-name-specifier was instantiated. Now, we give
something a bit better:

  error: no class named 'Nested' in 'NoDepBase<T>'

llvm-svn: 100060
2010-03-31 23:17:41 +00:00
Douglas Gregor 0208535fda Extend DependentNameType with a keyword enum that specifies whether
this was parsed as a typename-specifier, elaborated-type-specifier
(including the kind), or just a dependent qualified type name.

llvm-svn: 100039
2010-03-31 20:19:30 +00:00
Douglas Gregor a771f46c82 Reinstate my CodeModificationHint -> FixItHint renaming patch, without
the C-only "optimization".

llvm-svn: 100022
2010-03-31 17:46:05 +00:00
Douglas Gregor c1d2d8a014 Rename TypenameType to DependentNameType in anticipation of some
refactoring work in this area.

llvm-svn: 100019
2010-03-31 17:34:00 +00:00
Douglas Gregor 30e631862f Revert r100008, which inexplicably breaks the clang-i686-darwin10 builder
llvm-svn: 100018
2010-03-31 17:25:35 +00:00
Douglas Gregor 3baad0d4f7 Rename CodeModificationHint to FixItHint, since we've been using the
term "fix-it" everywhere and even *I* get tired of long names
sometimes. No functionality change.

llvm-svn: 100008
2010-03-31 15:31:50 +00:00
John McCall cef1582923 Support __attribute__((unused)) on types. This suddenly started firing
a lot for me on selfhosts, I dunno why.

llvm-svn: 99981
2010-03-31 02:47:45 +00:00
John McCall eae5acbbd0 Fix PR6327: restore invariants when there's a parse error in an initializer.
llvm-svn: 99980
2010-03-31 02:13:20 +00:00
Rafael Espindola c50c27cca8 the big refactoring bits of PR3782.
This introduces FunctionType::ExtInfo to hold the calling convention and the
noreturn attribute. The next patch will extend it to include the regparm
attribute and fix the bug.

llvm-svn: 99920
2010-03-30 20:24:48 +00:00
John McCall 2f88d7d72c Implement method friends in class templates and fix a few related problems.
llvm-svn: 99708
2010-03-27 05:57:59 +00:00
John McCall e99d5f3044 When finishing a function definition, leave the function definition *after*
doing all the cleanup tasks and checks.  This gives us the proper context for
checking access to base and member destructors.

llvm-svn: 99559
2010-03-25 22:08:03 +00:00
John McCall e9eaf8e0f8 Handle simple friend-class decls in class templates better by ensuring that
we look for shadow friend decls in the appropriate scope before injecting
a new declaration.

llvm-svn: 99552
2010-03-25 21:28:06 +00:00
Fariborz Jahanian c5b7bfc657 Improve diagnostics when ivar added to class
extension (radar 6812436).

llvm-svn: 99408
2010-03-24 18:08:23 +00:00
Douglas Gregor 8fdd0e8cd6 Silently drop dependent friend function template specializations,
since we have absolutely no way to match them when they are declared
nor do we have a way to represent these parsed-but-not-checked friend
declarations.

llvm-svn: 99407
2010-03-24 17:31:23 +00:00
John McCall 1f0479e3d7 Correct that last fixit: if the user wrote
template <> friend void foo(int);
we need to change it to
  friend void foo<>(int);
or else the user won't get the template specialization they obviously want.

llvm-svn: 99390
2010-03-24 08:27:58 +00:00
John McCall 816d75b701 Support friend function specializations.
llvm-svn: 99389
2010-03-24 07:46:06 +00:00
Douglas Gregor b525ef87fc When recovering from a qualified typedef name, don't clear out the
DeclContext because we don't want a NULL DeclContext. Instead, use the
current context.

llvm-svn: 99281
2010-03-23 15:26:55 +00:00
Ted Kremenek 0b40532b5e Only perform CFG-based warnings on 'static inline' functions that
are called (transitively) by regular functions/blocks within a
translation untion.

llvm-svn: 99233
2010-03-23 00:13:23 +00:00
John McCall df8b37c3f8 -Wshadow should only warn about parameter declarations when we're
entering a function or block definition, not on every single declaration.
Unfortunately we don't have previous-lookup results around when it's time
to make this decision, so we have to redo the lookup.  The alternative is
to use delayed diagnostics.

llvm-svn: 99172
2010-03-22 09:20:08 +00:00
Rafael Espindola 568586ff22 Fix PR6618.
If a struct has an invalid field, mark it as invalid. Also avoid producing
errors about incomplete types that are invalid.

llvm-svn: 99150
2010-03-21 22:56:43 +00:00
Ted Kremenek 918fe8498d Refactor CFG-based warnings in Sema to be run by a worked object called AnalysisBasedWarnings.
This object controls when the warnings are executed, allowing the client code
in Sema to selectively disable warnings as needed.

Centralizing the logic for analysis-based warnings allows us to optimize
when and how they are run.

Along the way, remove the redundant logic for the 'check fall-through' warning
for blocks; now the same logic is used for both blocks and functions.

llvm-svn: 99085
2010-03-20 21:06:02 +00:00
John McCall 2d8c760df7 Implement -Wshadow for parameter declarations as well.
llvm-svn: 99037
2010-03-20 04:12:52 +00:00
John McCall 84f0267301 Redeclaration lookups for parameter names should be flagged as redeclaration lookups
so they don't trigger diagnostics like (say) access control.

llvm-svn: 98806
2010-03-18 06:42:38 +00:00
John McCall 71ba5f27de ActOnTagDefinitionError is supposed to 'unwind' ActOnTagStartDefinition, not
ActOnStartCXXMemberDeclaration.  We haven't started the field collector on this
class yet, so don't stop it.  Fixes a crash in the VS buildbot and a memory error
on all the others.

llvm-svn: 98760
2010-03-17 19:25:57 +00:00
Douglas Gregor 319aa6c4b5 Remove warning about shadowing a built-in; built-ins aren't actually
considered to be a part of the translation unit unless they're named
in a way that brings them into existence.

llvm-svn: 98729
2010-03-17 16:03:44 +00:00
John McCall 2ff380a43a Clean up after ourselves when there's an error parsing the base clause.
Fixes the crash-on-invalid in PR6629.

llvm-svn: 98698
2010-03-17 00:38:33 +00:00
John McCall a2a3f7dc11 Implement -Wshadow. Based on a patch by Mike M.!
llvm-svn: 98684
2010-03-16 21:48:18 +00:00
John McCall a630995f50 Perform access control for the implicit calls to base and member destructors
that occur in constructors (on the unwind path).

llvm-svn: 98681
2010-03-16 21:39:52 +00:00
John McCall 3e11ebebc8 Remember declaration scope qualifiers in the AST. Imposes no memory overhead
on unqualified declarations.

Patch by Enea Zaffanella!  Minimal adjustments:  allocate the ExtInfo nodes
with the ASTContext and delete them during Destroy().  I audited a bunch of
Destroy methods at the same time, to ensure that the correct teardown was
being done.

llvm-svn: 98540
2010-03-15 10:12:16 +00:00
Ted Kremenek f59409623e Use llvm::SmallVector instead of std::vector.
llvm-svn: 98397
2010-03-12 22:22:36 +00:00
John McCall 58f10c3380 Maintain type source information for functions through template
instantiation.  Based on a patch by Enea Zaffanella!  I found a way to
reduce some of the redundancy between TreeTransform's "standard"
FunctionProtoType transformation and TemplateInstantiator's override,
and I killed off the old SubstFunctionType by adding type source info
for the last cases where we were creating FunctionDecls without TSI
(at least that get passed through template instantiation).

llvm-svn: 98252
2010-03-11 09:03:00 +00:00
John McCall e78aac41de Create a new InjectedClassNameType to represent bare-word references to the
injected class name of a class template or class template partial specialization.
This is a non-canonical type;  the canonical type is still a template 
specialization type.  This becomes the TypeForDecl of the pattern declaration,
which cleans up some amount of code (and complicates some other parts, but
whatever).

Fixes PR6326 and probably a few others, primarily by re-establishing a few
invariants about TypeLoc sizes.     

llvm-svn: 98134
2010-03-10 03:28:59 +00:00
Douglas Gregor 589973b097 In C++98/03, an uninitialized variable that has POD class type will be
uninitialized. This seems not to be the case in C++0x, where we still
call the (trivial) default constructor for a POD class
(!). Previously, we had implemented only the C++0x rules; now we
implement both. Fixes PR6536.

llvm-svn: 97928
2010-03-08 02:45:10 +00:00
Eli Friedman b632e30e41 Fix for PR6294: we should only delay recording nested dynamic classes if they
are lexically nested.  Othewise, we never end up recording semantically nested
classes.

llvm-svn: 97900
2010-03-07 05:49:51 +00:00
Douglas Gregor dc70c3abc0 Diagnose the declaration of enum templates. Also, be a bit more
careful about value-dependent enumerators. Fixes PR5786.

llvm-svn: 97570
2010-03-02 17:53:14 +00:00
Douglas Gregor 9a28e84b32 Keep an explicit stack of function and block scopes, each element of
which has the label map, switch statement stack, etc. Previously, we
had a single set of maps in Sema (for the function) along with a stack
of block scopes. However, this lead to funky behavior with nested
functions, e.g., in the member functions of local classes.

The explicit-stack approach is far cleaner, and we retain a 1-element
cache so that we're not malloc/free'ing every time we enter a
function. Fixes PR6382.

Also, tweaked the unused-variable warning suppression logic to look at
errors within a given Scope rather than within a given function. The
prior code wasn't looking at the right number-of-errors count when
dealing with blocks, since the block's count would be deallocated
before we got to ActOnPopScope. This approach works with nested
blocks/functions, and gives tighter error recovery.

llvm-svn: 97518
2010-03-01 23:15:13 +00:00
Chandler Carruth a419dbb5ea Fix the lookup of names used in a friend declaration to not attempt to
re-declare them. This fixes PR6317. Also add the beginnings of an interesting
test case for p1 of [class.friend] which also covers PR6317.

llvm-svn: 97499
2010-03-01 21:17:36 +00:00
Chris Lattner efc83e60d3 fix PR5933: don't warn about unused variables if a function has other errors in it.
llvm-svn: 97498
2010-03-01 21:06:03 +00:00
Chris Lattner 30d0cfda35 Implement jump checking for initialized c++ variables, implementing
a fixme and PR6451.

Only perform jump checking if the containing function has no errors,
and add the infrastructure needed to do this.

On the testcase in the PR, we produce:

t.cc:6:3: error: illegal goto into protected scope
  goto later;
  ^
t.cc:7:5: note: jump bypasses variable initialization
  X x;
    ^

llvm-svn: 97497
2010-03-01 20:59:53 +00:00
Chandler Carruth b306bcc266 Fix a really trivial crasher and begin fleshing out one of the namespace test
cases.

llvm-svn: 97134
2010-02-25 09:32:59 +00:00
Fariborz Jahanian 545643c309 More Sema check for ivars in class continuation.
llvm-svn: 97002
2010-02-23 23:41:11 +00:00
Fariborz Jahanian 4c172c63e5 Early support for declaring ivars in class extensions. wip.
llvm-svn: 96819
2010-02-22 23:04:20 +00:00