independently of whether they're definitions, then teach IR generation to
ignore non-explicit visibility when emitting declarations. Use this to
make sure that RTTI, vtables, and VTTs get the right visibility.
More of rdar://problem/8613093
llvm-svn: 117781
whether it's a declaration or not, then ignores that information for
declarations unless it was explicitly given. It's not totally clear
how that should be mapped into a sane system, but make an effort.
llvm-svn: 117780
load identifiers without loading their corresponding macro
definitions. This is likely to improve PCH performance slightly, and
reduces deserialization stack depth considerably when using
preprocessor metaprogramming.
llvm-svn: 117750
in asm's. PR 8501, 8602988.
I don't like including Type.h where it is; the idea was
to get references to X86_MMXTy out of the common code.
Maybe there's a better way?
llvm-svn: 117736
for namespace-scope variable declarations.
Apply visibility in IR gen to variables that are merely declared
and never defined. We were previously emitting these with default
visibility unless they were declared with private_extern.
Ignore global visibility settings when computing visibility for
a declaration's context, and key several conditions on whether a
visibility attribute exists anywhere in the hierarchy as opposed
to whether it exists at the current level.
llvm-svn: 117729
and never defined. We were previously emitting these with default
visibility unless they were declared with private_extern.
Ignore global visibility settings when computing visibility for
a declaration's context, and key several conditions on whether a
visibility attribute exists anywhere in the hierarchy as opposed
to whether it exists at the current level.
llvm-svn: 117644
timers to be dumped whenever the ASTUnit is destroyed. Instead, just
print the time elapsed for each operation after we perform the
operation.
llvm-svn: 117550
in the scope checker. With that done, turn an indirect goto into a
protected scope into a hard error; otherwise IR generation has to start
worrying about declarations not dominating their scopes, as exemplified
in PR8473.
If this really affects anyone, I can probably adjust this to only hard-error
on possible indirect gotos into VLA scopes rather than arbitrary scopes.
But we'll see how people cope with the aggressive change on the marginal
feature.
llvm-svn: 117539
is that we need more information to decide the exact conditions for whether
one ObjCObjectPointer is an acceptable return/parameter override for another,
so we're going to disable that entire class of warning for now. The
"forward developement" warning category, -Wmethod-signatures, can receive
unrestricted feature work, and when we're happy with how it acts, we'll
turn it on by default.
This is a pretty conservative change, and nobody's totally content with it.
llvm-svn: 117524
containing a DoStmt, and the LHS doesn't create a new block, then we should
return RBlock. Otherwise we'll incorrectly return NULL.
Also relax an assertion in VisitWhileStmt(). Reset 'Block' when it is finished.
llvm-svn: 117436
typo. This can happen with context-sensitive keywords like "super",
when typo correction didn't know that "super" wasn't permitted in this
context.
llvm-svn: 117372
There's probably still significant padding waste on x86-64 UNIXen, but
the difference in 32-bit compiles should be significant.
There are a lot of Expr nodes left that could lose a word this way.
llvm-svn: 117359
until after we've checked/promoted the argument. Hopefully fixes the
Emacs regression due to my recent change that expanded type-checking
in the presence of K&R function definitions.
llvm-svn: 117353
- tags with C linkage should ignore visibility=hidden
- functions and variables with explicit visibility attributes should
ignore the linkage of their types
Either of these should be sufficient to fix PR8457.
Also, FileCheck-ize a test case.
llvm-svn: 117351
covariant/contravariant overrides and implementations, but do so under
control of a new flag (-Wno-objc-covariant-overrides, which yes does cover
contravariance too).
*At least* the covariance cases will probably be enabled by default shortly,
but that's not totally uncontroversial.
llvm-svn: 117346
getCanonicalType() to make sure that the type we got back is actually
canonical. This is the case for most types, which always build a
canonical type when given canonical components. However, some types that
involve expressions in their canonicalization (e.g., array types with
dependent sizes) don't always build canonical types from canonical
components, because there is no such thing as a "canonical"
expression. Therefore, we do this extra mapping to ensure that the
canonical types we store are actually canonical.
llvm-svn: 117344
globals memory space gets assigned a symbolic value, but that value was not being used for lazy symbolication
of fields of globals. This could result in cases where bogus null dereferences were being reported.
Fixes PR 8440.
llvm-svn: 117336
function definition, we should still use a prototype to type-check and
convert the function arguments, if such a prototype exists. Fixes
PR8314.
llvm-svn: 117305
A common idiom in Objective-C is to provide a definition of a method in a subclass that returns a more-specified version of an object than the superclass. This does not violate the principle of substitutability, because you can always use the object returned by the subclass anywhere that you could use the type returned by the superclass. It was, however, generating warnings with clang, leading people to believe that semantically correct code was incorrect and requiring less accurate type specification and explicit down-casts (neither of which is a good thing to encourage).
This change ensures that any method definition has parameter and return types that make it accept anything that something conforming to the declaration may pass and return something that the caller will expect, but allows stricter definitions.
llvm-svn: 117271
This adds them where missing, and traces them through PCH. We fix at least one
bug in the extents found by the Index library, and make a lot of refactoring
tools which care about the exact formulation of a constructor call easier to
write. Also some minor cleanups to more consistently follow the friend pattern
instead of the setter pattern when rebuilding a serialized AST.
Patch originally by Samuel Benzaquen.
llvm-svn: 117254
In that case a chained PCH will record the updates to the DefinitionData pointer of forward references.
If a forward reference mutated into a definition re-write it into the chained PCH, this is too big of a change.
llvm-svn: 117239
- Pass around RecordDataImpl instead of the concrete RecordData so that any SmallVector can be used.
- Move ASTDeclWriter::WriteCXXDefinitionData to ASTWriter::AddCXXDefinitionData.
llvm-svn: 117236
its initial creation/deserialization and store the changes in a chained PCH.
The idea is that the AST entities call methods on the ASTMutationListener to give notifications
of changes; the PCHWriter implements the ASTMutationListener interface and stores the incremental changes
of the updated entity. WIP
llvm-svn: 117235
members in class subobjects of different types. So long as the
underlying declaration sets are the same, and the declaration sets
involve non-instance members, this is not an ambiguity.
llvm-svn: 117163
more closely parallel the computation of linkage. This gets us to a state
much closer to what gcc emits, modulo bugs, which will undoubtedly arise in
abundance.
llvm-svn: 117147
declaration have the 'readwrite' attribute. This is a common case, and we can issue a more lucid diagnostic.
Fixes <rdar://problem/7629420>.
llvm-svn: 117045
themselves have no template parameters. This is actually a restriction
due to the grammar of template template parameters, but we choose to
diagnose it in Sema to provide better recovery.
llvm-svn: 117032
redeclarations of main appropriately rather than allowing it to be
overloaded. Also, disallowing declaring main as a template.
Fixes GCC DejaGNU g++.old-deja/g++.other/main1.C.
llvm-svn: 117029
nil. Otherwise we can get false paths where a second @synchronized using the mutex
can have a bogus warning. Fixes <rdar://problem/8578650>.
llvm-svn: 117016
This adds an option to set the _MSC_VER macro without
recompiling. This is very useful when testing compatibility
with the Windows SDK and c++stdlib headers.
-fmsc-version=<version> (defaults to VS2003 (1300))
llvm-svn: 116999
inclusion directives, keeping track of every #include, #import,
etc. in the translation unit. We keep track of the source location and
kind of the inclusion, how the file name was spelled, and the
underlying file to which the inclusion resolved.
llvm-svn: 116952
As far as I can see, gcc is right to think this! The following change
will cause a nice segfault rather than undefined behaviour if this case
occurs. Someone who understands what this code is supposed to do should
probably take a proper look.
llvm-svn: 116917
don't repeatedly loop through identifiers, correcting the same typo'd
identifier over and over again.
We still bail out after 20 typo corrections, but this should help
improve performance in the common case where we're typo-correcting
because the user forgot to include a header.
llvm-svn: 116901
Here's example code:
---
template<class T> class MyClass {
struct S { };
S* NewS() { return new S; }
void DeleteS() { delete NewS(); }
};
---
CXXDeleteExpr::getDestroyedType() on the 'delete NewS()' expression
would crash before this change. Now it returns a dependent type
object. Solution suggested by dgregor.
llvm-svn: 116891
Now MICache is a linked list (per the FIXME), where we tradeoff between MacroInfo objects being in MICache
and MIChainHead. MacroInfo objects in the MICache chain are already "Destroy()'ed", so they can be reused. When
inserting into MICache, we need to remove them from the regular linked list so that they aren't destroyed more than
once.
llvm-svn: 116869
The problem was not the management of MacroInfo objects, but that when we recycle them
via the MICache the memory of the underlying SmallVector (within MacroInfo) was not getting
released. This is because objects stashed into MICache simply are reused with a placement
new, and never have their destructor called.
llvm-svn: 116862
computation to compute the lower bound of the edit distance, so that
we can avoid computing the edit distance for names that will clearly
be rejected later. Since edit distance is such an expensive algorithm
(M x N), this leads to a 7.5x speedup when correcting NSstring ->
NSString in the presence of a Cocoa PCH.
llvm-svn: 116849
list of allocated MacroInfos. This requires only 1 extra pointer per MacroInfo object, and allows us to blow them
away in one place. This fixes an elusive memory leak with MacroInfos (whose exact location I couldn't still figure
out despite substantial digging).
Fixes <rdar://problem/8361834>.
llvm-svn: 116842
within a default argument), recurse into default arguments. Fixes
PR8401, a regression I introduced in r113700 while refactoring our
handling of "used" declarations in default arguments.
llvm-svn: 116817
'../lib/clang/<version>'. Actually use '..' rather than removing the trailing
component to correctly handle paths containing '.' or symlinks in the presence
of -no-canonical-prefixes, etc. This shouldn't change any existing behavior.
llvm-svn: 116803
construct an unsupported friend when there's a friend with a templated
scope specifier. Fixes a consistency crash, rdar://problem/8540527
llvm-svn: 116786
doesn't hold. This fix is to increase the loop unrolling count to 4, which experiments show doesn't typically impact
analysis time. The real fix is to modify the IdempotentOperationsChecker to suppress warnings where an analysis point
could be preceded by a point where we gave up due to loop unrolling.
llvm-svn: 116769
-Wa,-force_cpusubtype_ALL t.c'.
- Tweaks -Wa, and -Xassembler handling to only accept an explicit short list of
arguments and give an obvious unsupported error on others.
llvm-svn: 116759
Anders Carlsson