Sebastian Redl [Sat, 24 Sep 2011 17:48:00 +0000 (17:48 +0000)]
Give InitListChecker a verification-only mode, where it neither emits diagnostics nor
builds a semantic (structured) initializer list, just reports on whether it can match
the given list to the target type.
Use this mode for doing init list checking in the initial step of initialization, which
will eventually allow us to do overload resolution based on the outcome.
Sebastian Redl [Sat, 24 Sep 2011 17:47:52 +0000 (17:47 +0000)]
In Initialization, add step kind SK_ListConstructorCall (list-initialization
resolves to a constructor call in C++11) and failure kind
FK_ListInitializationFailed (early InitListChecker run failed).
[microsoft] In Microsoft mode, if we are inside a template class member function and we can't resolve an identifier then assume the identifier is type dependent. The goal is to postpone name lookup to instantiation time to be able to search into type dependent base classes.
This fixes a few errors when parsing MFC code with clang.
BTW clang trunk is now about 5 patches away to be able the parse the default wizard-generated MFC project.
objc - fixes a crash when undefined typed property
followed by it implementation crashes when attempt
is made to access the synthesized ivar.
// rdar://10177744
The token stream was not getting properly reset when leaving
ParseLexedMethodDef in some error cases. In the testcase, that caused later
accesses to the token stream to touch memory which had been freed as we
finished parsing the class definition. Major hat-tip to AddressSanitizer for
helping pinpoint the use-after-free, including the allocation and deallocation
points:
Daniel Dunbar [Fri, 23 Sep 2011 20:33:41 +0000 (20:33 +0000)]
Driver: Add a --working-directory option which can be used to cause the compiler
to operate "as if" in a certain working directory.
- For now, we just implement this by changing the actual working directory, but
eventually we would want to handle this transparently. This is useful to
avoid an extra exec() pair in some situations, and will be something we would
want to support for more flexibility in using the Clang libraries.
Douglas Gregor [Fri, 23 Sep 2011 20:23:42 +0000 (20:23 +0000)]
Don't propagate the 'availability' attribute through declaration
merging for overrides. One might want to make a method's availability
in a superclass different from that of its subclass. Fixes
<rdar://problem/10166223>.
Richard Trieu [Fri, 23 Sep 2011 20:10:00 +0000 (20:10 +0000)]
Add a new warning to -Wliteral-conversion to catch cases where a string literal
is cast to a boolean. An exception has been made for string literals in
logical expressions to allow the common case of use in assert statements.
bool x;
x = "hi"; // Warn here
void foo(bool x);
foo("hi"); // Warn here
assert(0 && "error");
assert("error); // Warn here
Douglas Gregor [Fri, 23 Sep 2011 19:19:41 +0000 (19:19 +0000)]
Clean up parsing the category names in interfaces slightly, using
MatchRHSPunctuation appropriately and giving a useful source location
for the complaint about attributes being added to a category.
Douglas Gregor [Fri, 23 Sep 2011 19:04:03 +0000 (19:04 +0000)]
When checking for weak vtables, check whether the actual definition of
the key function is inline, rather than the original
declaration. Perhaps FunctionDecl::isInlined() is poorly named. Fixes
<rdar://problem/9979458>.
[microsoft] Fix a bug in -fdelayed-template-parsing mode where we were not reentering the delayed function context correctly. The problem was that all template params were reintroduced inside the same scope. So if we had a situation where we had 2 template params with the same name at different scope then clang would generate an error about ambiguous name.
The solution is to create a new ParseScope(Scope::TemplateParamScope) for each template scope that we want to reenter. (from the outmost to the innermost scope)
This fixes some errors when parsing MFC code with clang.
Benjamin Kramer [Thu, 22 Sep 2011 21:41:16 +0000 (21:41 +0000)]
Add support for CPATH and friends.
This moves the existing code for CPATH into the driver and adds the environment lookup and path splitting there.
The paths are then passed down to cc1 with -I options (CPATH), added after the normal user-specified include dirs.
Language specific paths are passed via -LANG-isystem and the actual filtering is performed in the frontend.
I tried to match GCC's behavior as close as possible
Douglas Gregor [Thu, 22 Sep 2011 20:32:43 +0000 (20:32 +0000)]
Only trigger the initialize-an-array-via-elementwise-copy/move code
generation when we're dealing with an implicitly-defined copy or move
constructor. And, actually set the implicitly-defined bit for
implicitly-defined constructors and destructors. Should fix self-host.
Anna Zaks [Thu, 22 Sep 2011 18:10:41 +0000 (18:10 +0000)]
ST->scanReachableSymbols() is creating a SubRegionMap (SRM) on every call since one SRM is created in each ScanReachableSymbols instance. Creating the object just once and calling only scan inside the loop gives ~ 14% speed up of the StaticAnalyzer run (Release+Asserts).
Pull out the declaration of the ScanReachableSymbols so that it can be used directly. Document ProgramState::scanReachableSymbols() methods.
In OpenCL, conversions between different vector types are disallowed
OpenCL 6.2.1 says: "Implicit conversions between built-in vector data types are
disallowed." OpenCL 6.2.2 says: "Explicit casts between vector types are not
legal." For example:
uint4 u = (uint4)(1);
int4 i = u; // invalid implicit conversion
int4 e = (int4)u; // invalid explicit conversion
Fixes PR10967. Submitted by: Anton Lokhmotov <Anton.lokhmotov@gmail.com>
In the OpenCL mode, the AltiVec mode must be off and checks must be strict
OpenCL is different from AltiVec in the way it supports vector literals. OpenCL
is strict with regards to semantic checks. For example, implicit conversions
and explicit casts between vectors of different types are disallowed.
Fixes PR10975. Submitted by: Anton Lokhmotov <Anton.lokhmotov@gmail.com>
John McCall [Wed, 21 Sep 2011 08:08:30 +0000 (08:08 +0000)]
ANSI C requires that a call to an unprototyped function type succeed
if the definition has a non-variadic prototype with compatible
parameters. Therefore, the default rule for such calls must be to
use a non-variadic convention. Achieve this by casting the callee to
the function type with which it is required to be compatible, unless
the target specifically opts out and insists that unprototyped calls
should use the variadic rules. The only case of that I'm aware of is
the x86-64 convention, which passes arguments the same way in both
cases but also sets a small amount of extra information; here we seek
to maintain compatibility with GCC, which does set this when calling
an unprototyped function.
Anna Zaks [Wed, 21 Sep 2011 00:35:58 +0000 (00:35 +0000)]
[analyzer] Fix a bug where PathDiagnosticLocation did not generate a valid range and add asserts to check validity of locations early on. Ignore invalid ranges in PathDiagnosticPiece (they could be added by checker writers).
Addresses radar://10124836 and radar://radar10102244.
Anna Zaks [Tue, 20 Sep 2011 23:27:32 +0000 (23:27 +0000)]
[analyzer] Refactor PathDiagnosticLocation: Remove SourceRange member from PathDiagnosticLocation - FullSourceLoc Loc and PathDiagnosticRange Range are sufficient.
Anna Zaks [Tue, 20 Sep 2011 22:30:48 +0000 (22:30 +0000)]
[analyzer] Remove dead code. (This code is trying to implement the idea that PathDiagnosticClient could implement DiagnosticClient and has been dead for a while).
Bob Wilson [Tue, 20 Sep 2011 22:00:38 +0000 (22:00 +0000)]
For i386 kext fallback to llvm-gcc, search paths for several Darwin versions.
This replaces the hack to read UNAME_RELEASE from the environment when
identifying the OS version on Darwin, and it's more flexible. It's also
horribly ugly, but at least this consolidates the ugliness to touch less of
the code so that it will be easier to rip out later.
Anna Zaks [Tue, 20 Sep 2011 21:38:35 +0000 (21:38 +0000)]
[analyzer] Refactor PathDiagnosticLocation: Make PathDiagnosticLocation(SourceLocation...) private. Most of the effort here goes to making BugReport refer to a PathDiagnosticLocation instead of FullSourceLocation.
(Another step closer to the goal of having Diagnostics which can recover from invalid SourceLocations.)
Richard Trieu [Tue, 20 Sep 2011 20:03:50 +0000 (20:03 +0000)]
Fix a problem in digraph handling where "[:" might be treated as "<::" and
erronously trigger the digraph correction fix-it. Include a new test to catch
this in the future.
Anna Zaks [Tue, 20 Sep 2011 18:23:52 +0000 (18:23 +0000)]
[analyzer] Refactor PathDiagnosticLocation: Use PointerUnion of LocationContext and AnalysisContext to support creation of PathDiagnosticLocations for checkers which no context sensitivity.
Douglas Gregor [Tue, 20 Sep 2011 18:13:03 +0000 (18:13 +0000)]
Introduce an egregious hack for modules to cope with headers that come
from unfriendly (== not at all modularized) directories. This is
temporary, and it only affects module construction until I'll figured
out how to deal with system headers.
Ivan Krasin [Tue, 20 Sep 2011 14:56:54 +0000 (14:56 +0000)]
Clarify PNaCl target characteristics: set LongDoubleWidth, PtrDiffType, IntPtrType,
change __builtin_va_list to from a structure to int[4] (same alignment
and size, but with a simpler representation). Patch by David Meyer!
John McCall [Tue, 20 Sep 2011 06:21:28 +0000 (06:21 +0000)]
Fix a crash-on-invalid with bad CV-qualification on 'this' in the
presence of an implicit move assignment operator. I think the implicit
copy assignment operator case was also wrong, but just in a "displaying
the wrong diagnostic" way.
Fix a pretty nasty bug in noreturn destructors that cascaded into lots
of false positive warnings that depend on noreturn destructors pruning
the CFGs, but only in C++0x mode!
This was really surprising as the debugger quickly reveals that the
attributes are parsed correctly (and using the same code) in both modes.
The warning fires in the same way in both modes. But between parsing and
building the destructor declaration with the noreturn attribute and the
warning, it magically disappears. The key? The 'noexcept' appears!
When we were rebuilding the destructor type with the computed implicit
noexcept we completely dropped the old type on the floor. This almost
makes sense (as the arguments and return type to a destructor aren't
exactly unpredictable), but lost any function type attributes as well.
The fix is simple, we build the new type off of the old one rather than
starting fresh.
Testing this is a bit awkward. I've done it by running the
noreturn-sensitive tests in both modes, which previous failed and now
passes, but if anyone has ideas about how to more specifically and
thoroughly test that the extended info on a destructor is preserved when
adding noexcept, I'm all ears.
Anna Zaks [Tue, 20 Sep 2011 01:51:40 +0000 (01:51 +0000)]
[analyzer] Refactor PathDiagnosticLocation: Pre-compute Range and Location with gen methods on object creation instead of computing on demand. This would allow to remove dependency on the other members which help with construction and might not even be valid at later stages (to be removed later on).
projects / clang / log