Douglas Gregor [Wed, 20 Oct 2010 01:32:02 +0000 (01:32 +0000)]
Introduce a simple cache for unqualified typo corrections, so that we
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.
The type-to-delete may not be a pointer if it's a dependent type.
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.
Ted Kremenek [Tue, 19 Oct 2010 22:15:20 +0000 (22:15 +0000)]
Really^2 fix <rdar://problem/8361834>, this time without crashing.
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.
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.
Bill Wendling [Tue, 19 Oct 2010 20:08:12 +0000 (20:08 +0000)]
Now that mm_malloc.h was rewritten to *not* include errno.h (see
http://llvm.org/viewvc/llvm-project?rev=116771&view=rev) we can get rid of these
hacks.
Douglas Gregor [Tue, 19 Oct 2010 19:39:10 +0000 (19:39 +0000)]
Improve the performance of typo correction, by using a simple
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.
Ted Kremenek [Tue, 19 Oct 2010 18:16:54 +0000 (18:16 +0000)]
Simplify lifetime management of MacroInfo objects in Preprocessor by having the Preprocessor maintain them in a linked
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).
Douglas Gregor [Tue, 19 Oct 2010 17:17:35 +0000 (17:17 +0000)]
When marking declarations referenced within an expression (e.g.,
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.
Chandler Carruth [Tue, 19 Oct 2010 08:47:51 +0000 (08:47 +0000)]
Use CLANG_RESOURCE_DIR define if one is provided, otherwise use the default of
'../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.
John McCall [Tue, 19 Oct 2010 01:40:49 +0000 (01:40 +0000)]
Redirect templated friend class decls to a new Sema callback and
construct an unsupported friend when there's a friend with a templated
scope specifier. Fixes a consistency crash, rdar://problem/8540527
Douglas Gregor [Tue, 19 Oct 2010 00:03:23 +0000 (00:03 +0000)]
Tweak code-completion result priorities, so that exact and similar
type matches have a bigger impact. The impetus for this change was
that, when initializing an enumeration value, we want enumerators of
that enumeration type to have a higher priority than, e.g., unrelated
local variables.
Ted Kremenek [Mon, 18 Oct 2010 23:36:05 +0000 (23:36 +0000)]
"Fix" bogus idempotent operations warning due to loop unrolling not unrolling enough loops to show that an invariant
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.
Daniel Dunbar [Mon, 18 Oct 2010 22:36:15 +0000 (22:36 +0000)]
Driver/IA: Accept and ignore -force_cpusubtype_ALL, as in 'clang -c
-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.
Douglas Gregor [Mon, 18 Oct 2010 22:01:46 +0000 (22:01 +0000)]
Fix the translation of the PCC_ForInit code-completion context for
C++/C99/Objective-C, so that we properly include types. This fix
affects global caching of code-completion results; without caching,
the behavior was already correct.
Douglas Gregor [Mon, 18 Oct 2010 21:05:04 +0000 (21:05 +0000)]
Introduce code completion results for Objective-C methods, both when
declaring methods and when sending messages to them, by bringing all
of the selector into TypedCheck chunks in the completion result. This
way, we can improve the sorting of these results to account for the
full selector name rather than just the first chunk.
Douglas Gregor [Mon, 18 Oct 2010 18:21:28 +0000 (18:21 +0000)]
When providing code completions of Objective-C method declarations
(after - or +), always traverse superclasses and all categories. The
programmer may want to complete a method from *anywhere*.
Francois Pichet [Mon, 18 Oct 2010 15:01:13 +0000 (15:01 +0000)]
Microsoft enum extensions. 2 things will change on -fms-extensions:
1. enum underlying type is int by default.
2. Error "enumerator value is not representable in the underlying type"is a ExtWarning
Bill Wendling [Sun, 17 Oct 2010 07:38:01 +0000 (07:38 +0000)]
The "gcc.dg/compat/vector-1 -m32" test was broken after the MMX rewrite. The
function parameters weren't converted to use the correct type (x86_mmx). Add a
check, similar to the one in llvm-gcc, to see if we need the x86_mmx type for
that function parameter. If so, it coerces the type to be that.
John McCall [Sat, 16 Oct 2010 06:59:13 +0000 (06:59 +0000)]
White-listing templated-scope friend decls is a good idea, but doing it
by marking the decl invalid isn't. Make some steps towards supporting these
and then hastily shut them down at the last second by marking them as
unsupported.
Daniel Dunbar [Sat, 16 Oct 2010 05:04:10 +0000 (05:04 +0000)]
Revert r116656, "IRgen/Obj-C/NeXT: Fix the IR signature for
objc_exception_rethrow, so we don't...", since something is actually trying to
call this with the wrong signature (!). Unfortunately I don't understand the new
EH infrastructure well enough to fix it immediately.
Douglas Gregor [Fri, 15 Oct 2010 23:53:28 +0000 (23:53 +0000)]
Allow list-initialization of a local variable of class type with a
flexible array member, so long as the flexibility array member is
either not initialized or is initialized with an empty initializer
list. Fixes <rdar://problem/8540437>.
Ted Kremenek [Fri, 15 Oct 2010 22:50:23 +0000 (22:50 +0000)]
Tweak retain/release checker diagnostics to specify a leak occurs because an object is not referenced later in the path,
not that it isn't referenced later in the code. Fixes <rdar://problem/8527839>.
Check for ivar being a C++ object before attempting to
find a copy constructor/assignment operator used
in getter/setter synthesis. This removes an unintended
diagnostics and makes objc++ consistant with objective-c.
// rdar: //8550657.
projects / clang / log