Anders Carlsson [Fri, 24 Apr 2009 05:23:13 +0000 (05:23 +0000)]
Add a VarDecl parameter to the CXXTemporaryObjectExpr constructor. It's unused for now, so no functionality change yet. Also, create CXXTempVarDecls to pass to the CXXTemporaryObjectExpr ctor.
Anders Carlsson [Fri, 24 Apr 2009 04:57:02 +0000 (04:57 +0000)]
Move the CXXConstructExpr before the CXXTemporaryObjectExpr so that the temporary object expr can inherit from the construct expr. No functionality change.
Douglas Gregor [Fri, 24 Apr 2009 02:57:34 +0000 (02:57 +0000)]
Eliminate Sema::ObjCAliasDecls. This is based on Steve's fix, but also
updates name lookup so that we see through @compatibility_alias
declarations to their underlying interfaces.
Chris Lattner [Fri, 24 Apr 2009 00:30:45 +0000 (00:30 +0000)]
Fix rdar://6821047 - clang crashes on subscript of interface in 64-bit mode
Several changes here:
1. We change Type::isIncompleteType to realize that forward declared
interfaces are incomplete. This eliminate special case code for this
from the sizeof path, and starts us rejecting P[4] when P is a pointer
to an incomplete interface.
2. Explicitly reject P[4] when P points to an interface in non-fragile ABI
mode.
3. Switch the sizeof(interface) diagnostic back to an error instead of a
warning in non-fragile abi mode.
Douglas Gregor [Fri, 24 Apr 2009 00:11:27 +0000 (00:11 +0000)]
Eliminate Sema::ObjCImplementations, relying instead on name lookup. What's good for uniformity is good for PCH (or is it the other way around?).
As part of this, make ObjCImplDecl inherit from NamedDecl (since
ObjCImplementationDecls now need to have names so that they can be
found). This brings ObjCImplDecl very, very close to
ObjCContainerDecl; we may be able to merge them soon.
Douglas Gregor [Thu, 23 Apr 2009 23:18:26 +0000 (23:18 +0000)]
Eliminate Sema::ObjCProtocols. Instead, we place ObjCProtocolDecls in
their own namespace (IDNS_Protocol) and use the normal name-lookup
routines to find them. Aside from the simplification this provides
(one less DenseMap!), it means that protocols will be lazily
deserialized from PCH files.
Make the code size of the selector table block match the code size of
the type and decl blocks.
Douglas Gregor [Thu, 23 Apr 2009 22:29:11 +0000 (22:29 +0000)]
PCH support for all of the predefined Objective-C types, such as id,
SEL, Class, Protocol, CFConstantString, and
__objcFastEnumerationState. With this, we can now run the Objective-C
methods and properties PCH tests.
retain/release checker: Don't call isTrackedObject() with the canonical type.
This was preventing the checker from tracking return objects referenced by 'id'.
Per discussions with Ken Ferry and Paul Marks (<rdar://problem/6815234>) greatly
extend the number of objects tracked by the retain/release checker by assuming
that all class and instance methods should follow Cocoa object "getter" and
"alloc/new" conventions.
Handle corner case where clang-cc is invoked directly to compile preprocessed source file without -main-file-name. In this case, CDDebugInfo is not able identify correct main source file becase SM.isFromMainFile() returns true for locations from header files as well as locations from main source file.
This patch takes conservative approach by not emitting more then one compile unit with isMain bit set.
Steve Naroff [Thu, 23 Apr 2009 15:15:40 +0000 (15:15 +0000)]
Sema::ActOnStartClassInterface(): Use PushOnScopeChains().
This enables class recognition to work with PCH. I believe this means we can remove Sema::ObjCInterfaceDecls and it's usage within Sema::LookupName(). Will investigate.
Daniel Dunbar [Thu, 23 Apr 2009 08:03:21 +0000 (08:03 +0000)]
Mark IMAGE_INFO as constant on x86_64-darwin.
- This shouldn't change anything, we never actually access it, but
this is consistent with llvm-gcc (and 32-bit)
Daniel Dunbar [Thu, 23 Apr 2009 07:00:09 +0000 (07:00 +0000)]
Add implicit definition of objc_msgSend.
- As with malloc and friends, this is important where the return type
on a 64-bit platform would otherwise end up discarding the upper
32-bits.
Chris Lattner [Thu, 23 Apr 2009 05:30:27 +0000 (05:30 +0000)]
the logic for computing __func__ and friends is really broken:
the type assigned by sema (and is visible with sizeof(__func__) for
example) has nothing to do with what codegen ends up producing.
We should eventually add a method on PredefinedExpr to handle this.
In the meantime, just set up some framework and add some fixme's.
Douglas Gregor [Thu, 23 Apr 2009 03:23:08 +0000 (03:23 +0000)]
The ivars in an ObjCImplementationDecl are now stored in the
DeclContext rather than in a separate list. This makes PCH
(de-)serialization trivial, so that ivars can be loaded lazily.
Anders Carlsson [Thu, 23 Apr 2009 02:32:43 +0000 (02:32 +0000)]
Add a CXXConstructExpr that represents an implicit call to a C++ constructor. I think CXXTemporaryObjectExpr is going to become a subclass of CXXConstructExpr, since CXXTemporaryObjectExpr represents a syntactic temporary, for example T()
Douglas Gregor [Thu, 23 Apr 2009 01:02:12 +0000 (01:02 +0000)]
Eliminate the three SmallVectors in ObjCImplDecl (for instance
methods, class methods, and property implementations) and instead
place all of these entities into the DeclContext.
This eliminates more linear walks when looking for class or instance
methods and should make PCH (de-)serialization of ObjCDecls trivial
(and lazy).
Add PathDiagnosticRange to PathDiagnostics. These simply wrap SourceRange and
indicate whether or not the range represents an absolute range or should be
extended by lexing to the end of the token.
Douglas Gregor [Wed, 22 Apr 2009 20:56:09 +0000 (20:56 +0000)]
Eliminate Sema::KnownFunctionIDs, so that Sema doesn't end up pulling
in a bunch of declarations from the PCH file. We're down to loading
very few declarations in Carbon-prefixed "Hello, World!":
Eli Friedman [Wed, 22 Apr 2009 19:23:09 +0000 (19:23 +0000)]
Add handling for complex->int, int->complex float, and float->complex
int. Note that constant int->complex float and float->complex int casts
were being miscompiled.
Douglas Gregor [Wed, 22 Apr 2009 19:09:20 +0000 (19:09 +0000)]
Minimize the number and kind of "external definitions" that the PCH
file needs to store. CodeGen needs to see these definitions (via
HandleTopLevelDecl), otherwise it won't be able to generate code for
them.
This patch notifies the consumer (e.g., CodeGen) about function
definitions and variable definitions when the corresponding
declarations are deserialized. Hence, we don't eagerly deserialize the
declarations for every variable or function that has a definition in
the PCH file. This gives another 5% speedup for the Carbon-prefixed
"Hello, World!", and brings our PCH statistics down to something far
more reasonable:
Douglas Gregor [Wed, 22 Apr 2009 18:49:13 +0000 (18:49 +0000)]
Lazy loading of builtins for precompiled headers.
PCH files now contain complete information about builtins, including
any declarations that have been synthesized as part of building the
PCH file. When using a PCH file, we do not initialize builtins at all;
when needed, they'll be found in the PCH file.
This optimization translations into a 9% speedup for "Hello, World!"
with Carbon.h as a prefix header and roughly a 5% speedup for 403.gcc
with its prefix header. We're also reading less of the PCH file for
"Hello, World!":
BugReporter (extensive diagnostics):
- Remove stale assertion that was breaking the test suite.
- When popping location contexts, only add a control-flow piece for fileID
locations.
PathDiagnosticLocation: ranges for terminators now only include the first
character instead of the entire range for the IfStmt, ForStmt, etc. We may
gradually refine these ranges later, but basically terminator ranges just refer
to the first keyword.
Daniel Dunbar [Wed, 22 Apr 2009 17:43:55 +0000 (17:43 +0000)]
Reapply r69771, with updates & fixes:
Rework the shadow struct that is layed out for Objective-C classes.
- Superclasses are now always laid out in their shadow structure at
the first field.
- Prior to this, the entire class heirarchy was flattened into a
single structure which meant that alignment, padding, and bitfields
were incorrect (the ASTRecordLayout was correct however, which
meant our debug info didn't coincide with ivar offsets, for
example).
- This is still very suboptimal (for example, ivar are looked up
recursively, but I believe the ivar layout itself is now at least
close to correct.
Daniel Dunbar [Wed, 22 Apr 2009 10:28:39 +0000 (10:28 +0000)]
Don't convert interface types (to structs) as part of CodeGenTypes.
- This has pros and cons, but for now the pros seem to significantly
outway the con.
The con is that we will always need to cast in the runtime
implementation to a struct type, if we wish to access an interface
directly.
The pros are:
- Avoid the cost of generating types which are used. Most
manipulation of Objective-C objects is done through messages, and
only the implementation of a class will directly access
memory. Previously, we would convert the type even if it only
appear as a function parameter, for example.
- We don't need to worry about incomplete types, and
UpdateCompletedType for interfaces is gone.
- It becomes easier to narrow the interface to the shadow struct for
Objective-C interfaces (so it can be eliminated).
Currently the runtimes still use the CodeGenTypes machinery to
generate the LLVM structure they need via ConvertTagDecl, but this can
eventually be replaced.
Daniel Dunbar [Wed, 22 Apr 2009 08:53:01 +0000 (08:53 +0000)]
Add another workaround for -include.
- If we don't find a file looking relative to the current working
directory, fall back to header search. This is closer to what would
happen if the lookup was starting from right directory in the first
place (except it will find files in the directory of the main
source file, which I *think* should not be found).
Daniel Dunbar [Wed, 22 Apr 2009 07:32:20 +0000 (07:32 +0000)]
Merge ivar access amongst the three runtimes.
- For now, this means we are always doing the address computations by
hand instead of constructing a proper GEP. Right now, however, this
is less important than having fewer entry points to dealing with
Objective-C interface layout.
projects / clang / log