Chris Lattner
61f60a05b2
c89 does not perform array -> pointer promotion unless the array is an lvalue. This
...
is different than C99. This fixes the rest of rdar://6095180.
llvm-svn: 54064
2008-07-25 21:33:13 +00:00
Daniel Dunbar
b5586dad97
Fix long double test case
...
- Had wrong sizeof(long double) encoded into test case
- Results now agree with llvm-gcc
llvm-svn: 54059
2008-07-25 20:54:30 +00:00
Chris Lattner
f6e1e309c7
In c99 mode, comma does do function/array promotion even though
...
it does not do unary promotions (like short->int).
llvm-svn: 54058
2008-07-25 20:54:07 +00:00
Ted Kremenek
e17c037f4f
Remove '.' in expected warning
...
llvm-svn: 54051
2008-07-25 20:30:34 +00:00
Chris Lattner
e174acc910
Comma does not perform unary promotions, rdar://6095180
...
llvm-svn: 54045
2008-07-25 20:02:29 +00:00
Chris Lattner
e08e47ab2f
gcc requires a semicolon at the end of an interface, unlike its extension for structs.
...
rdar://6095245.
llvm-svn: 54044
2008-07-25 19:54:55 +00:00
Steve Naroff
e2177fbfdc
Cleaunup Sema::ActOnClassMessage(). This commit:
...
(a) removes a bogus warning.
(b) removes an undesirable usage of the ObjCMessageExpr constructor that takes an IdentifierInfo * (which I will abolish).
llvm-svn: 54042
2008-07-25 19:39:00 +00:00
Chris Lattner
f9c49e5d27
Reject typedef redefinitions when the underlying types are not identical,
...
even if in a system header etc. rdar://6079937
llvm-svn: 54038
2008-07-25 18:44:27 +00:00
Chris Lattner
31614aeb0a
update this testcase.
...
llvm-svn: 54037
2008-07-25 18:42:46 +00:00
Daniel Dunbar
c481f72857
Add test case for hex floating point constants in < C99 mode
...
- For: rdar://6096838
llvm-svn: 54036
2008-07-25 18:41:57 +00:00
Chris Lattner
ec8996d771
Add support for __extension__ as an lvalue. rdar://6097308
...
llvm-svn: 54033
2008-07-25 18:07:19 +00:00
Ted Kremenek
092ec76923
Don't emit 'dead initialization' warnings for variables marked 'unused'.
...
This fixes PR 2573: http://llvm.org/bugs/show_bug.cgi?id=2573
llvm-svn: 54009
2008-07-25 04:47:34 +00:00
Daniel Dunbar
0e5845c13a
Rework codegen of case ranges
...
- Fix multiple issues with the way case ranges were emitted, see test
cases details about the specific issues.
- The root issue was not being careful about how basic blocks were
emitted which led to them being chained together incorrectly,
resulting in improper control flow.
- Fixes <rdar://problem/6098585>
llvm-svn: 54006
2008-07-25 01:11:38 +00:00
Daniel Dunbar
dd49a04d18
Allow CLANG env variable to override "clang" command in TestRunner.sh
...
llvm-svn: 54005
2008-07-25 01:05:44 +00:00
Ted Kremenek
9e9afb552c
Correctly handle NSAssertionHandle -handleFailureInMethod:object:file:lineNumber:description:
...
This fixes: http://llvm.org/bugs/show_bug.cgi?id=2593
llvm-svn: 53993
2008-07-24 18:47:16 +00:00
Ted Kremenek
0e60b7588a
Don't issue a missing +dealloc warning for classes that just contain SEL ivars.
...
This fixes PR 2592: http://llvm.org/bugs/show_bug.cgi?id=2592
llvm-svn: 53987
2008-07-24 17:45:56 +00:00
Ted Kremenek
53b29025f1
Rename test case file.
...
llvm-svn: 53984
2008-07-24 17:02:18 +00:00
Ted Kremenek
87b16f4258
Issue dead store warnings for preincrements involved in a subexpression.
...
llvm-svn: 53983
2008-07-24 17:01:17 +00:00
Ted Kremenek
0b5eb1ac87
Added dead stores test case.
...
llvm-svn: 53966
2008-07-23 23:18:43 +00:00
Ted Kremenek
ecc851bb6e
Further refine dead store checking to distinguish between dead stores and dead increments.
...
llvm-svn: 53960
2008-07-23 21:16:38 +00:00
Ted Kremenek
46abc7db6b
Properly skip IBOutlets when checking for unused ivars.
...
Refine the error message of unused ivars.
Added test case.
llvm-svn: 53957
2008-07-23 18:21:36 +00:00
Chris Lattner
070968957e
Fix rdar://6094010 various asserts happening with wide strings in inline asm.
...
llvm-svn: 53947
2008-07-23 06:46:56 +00:00
Chris Lattner
d258325c2b
Fix a codegen crash on:
...
int foo(void) {
float x[2];
return x;
}
rdar://6093986
llvm-svn: 53946
2008-07-23 06:31:27 +00:00
Ted Kremenek
98f6e582f2
Added path-sensitive checking for null pointer values passed to function arguments marked nonnull.
...
This implements <rdar://problem/6069935>
llvm-svn: 53891
2008-07-22 00:46:16 +00:00
Daniel Dunbar
327acd7a6c
Implement bzero, memset, memmove builtins.
...
This subsumes Nico Weber's patch and implements bzero in terms
of llvm.memset to match llvm-gcc.
llvm-svn: 53888
2008-07-22 00:26:45 +00:00
Daniel Dunbar
b725726c29
Implement nans, prefetch, and trap builtins.
...
This closes <rdar://problem/6080720>, support for __builtin_constant_p
has been filed separately.
llvm-svn: 53885
2008-07-21 22:59:13 +00:00
Ted Kremenek
5224e6a81d
Add test case for nonnull attribute.
...
Fix indexing bug.
llvm-svn: 53882
2008-07-21 22:09:15 +00:00
Daniel Dunbar
c2f67966f4
Add __builtin_powi[fl] support
...
llvm-svn: 53866
2008-07-21 18:44:41 +00:00
Daniel Dunbar
5b288788b3
Fix a test RUN line to not generate '=-' in test directory
...
llvm-svn: 53865
2008-07-21 18:42:27 +00:00
Daniel Dunbar
d93abc3bb0
Implement ffs, parity, and popcount builtins.
...
+ test case
llvm-svn: 53851
2008-07-21 17:19:41 +00:00
Chris Lattner
ca1e848588
Switch initialization of the protocol list for an interface decl to use
...
the standard "set these as the list of protocols" interface instead of a
strange "set this as the size and then set each one to the value" interface.
The problem with the later is that it a) is completely different from
everything else, b) is awkward, and c) doesn't handle the case when a
referenced protocol is invalid: it set it to null.
This meant that all clients downstream would have to handle null protocols
in the protocol list, and empirically they didn't. Fix this by not setting
invalid protocols in the referenced protocol list, fixing the crash on
test/Sema/objc-interface-1.m
While I'm at it, clean up some locations so that we produce:
t.m:1:25: error: cannot find interface declaration for 'NSObject', superclass of 'NSWhatever'
@interface NSWhatever : NSObject <NSCopying>
~~~~~~~~~~~~~~~~~~~~~ ^
instead of:
t.m:1:1: error: cannot find interface declaration for 'NSObject', superclass of 'NSWhatever'
@interface NSWhatever : NSObject <NSCopying>
^
llvm-svn: 53846
2008-07-21 07:06:49 +00:00
Chris Lattner
87899be4cc
when in the context of an @implementation, look for private methods in the
...
@implementation to resolve nullary selector references.
llvm-svn: 53845
2008-07-21 06:44:27 +00:00
Chris Lattner
bd2d634424
continue cleaning up code, and disable sending a message directly to an
...
interface. This fixes a bug where we used to accept:
void test2(NSNumber x) {
[x METH];
}
which doesn't make sense and GCC rejects.
llvm-svn: 53841
2008-07-21 06:12:56 +00:00
Nuno Lopes
8a48c2f7d9
fix test
...
llvm-svn: 53802
2008-07-19 22:11:43 +00:00
Ted Kremenek
f487548c17
Patch by Kovarththanan Rajaratnam:
...
"When dumping the tokens (-dumptokens output type), the column numbers are not
correctly shown. This patch fixes that issue."
llvm-svn: 53797
2008-07-19 19:10:40 +00:00
Ted Kremenek
8857703b57
Added test case.
...
llvm-svn: 53770
2008-07-18 20:48:10 +00:00
Daniel Dunbar
155ed24359
Prevent clang from emitting output when input has errors
...
+ test case
<rdar://problem/6080040>
http://llvm.org/bugs/show_bug.cgi?id=2280
llvm-svn: 53756
2008-07-18 16:38:05 +00:00
Argyrios Kyrtzidis
bd25998a45
When in C++, make EnumConstant names hide tag names in the same scope, instead of colliding with them.
...
llvm-svn: 53702
2008-07-16 21:01:53 +00:00
Steve Naroff
ce2dca186b
RewriteObjC::RewriteObjCTryStmt():Don't synthesize a catch begin if there are 0 catch clauses.
...
This fixes <rdar://problem/5987211> clang ObjC rewriter: @try / @finally block produces unbalanced output.
llvm-svn: 53679
2008-07-16 15:31:30 +00:00
Argyrios Kyrtzidis
7da34d050b
When checking for name collision between a tag and a previously defined namespace, the collision occured even when the tag was in a different nested scope.
...
Fix it by taking into account the scope when checking for namespace-tag name collisions.
llvm-svn: 53667
2008-07-16 07:45:46 +00:00
Ted Kremenek
ed1f72b67c
For the MissingDealloc check, don't treat IBOutlet ivars as being needed to be released
...
llvm-svn: 53647
2008-07-15 23:04:27 +00:00
Ted Kremenek
81bfc074c9
Distinguish between dead stores and dead initializations.
...
llvm-svn: 53628
2008-07-15 18:06:32 +00:00
Ted Kremenek
fd7efdf20d
Provide an "Analysis Scope" for Analyses so checks can either be run on code declarations (bodies) or Objective-C @implementation blocks.
...
llvm-svn: 53584
2008-07-15 00:46:02 +00:00
Nate Begeman
191a6b1f1b
http://llvm.org/bugs/show_bug.cgi?id=2523
...
Add some code to handle vector comparisons, which is the language side
of the llvm vicmp/vfcmp instructions. Also make the vector-vector and
vector-scalar asign checks a bit more sane under the presence of lax vector
conversions.
llvm-svn: 53565
2008-07-14 18:02:46 +00:00
Ted Kremenek
3bfb314c25
Add new check: -check-objc-methodsigs. This check scans methods in
...
ObjCImplementationDecls and sees if a ancestor class defines a method with the
same selector but with a different type signature. Right now it just compares
return types, and mainly looks at differences in primitive values. The checking
will be expanded in the future.
llvm-svn: 53482
2008-07-11 22:40:47 +00:00
Ted Kremenek
1aaa6d8c19
Add test case.
...
llvm-svn: 53335
2008-07-09 18:11:43 +00:00
Nuno Lopes
7bfa180056
fix CheckForConstantInitializer() for Compound Literals
...
also fix the correspondent test (it was expecting more errors than it should. please confirm my fix is correct (at least gcc agrees with me)
llvm-svn: 53174
2008-07-07 16:46:50 +00:00
Argyrios Kyrtzidis
5708a4661e
'&&' commands together so that the test status reflects the results of all the commands, otherwise the test status will be the result of only the last command.
...
llvm-svn: 53135
2008-07-04 10:33:02 +00:00
Ted Kremenek
4449ec09d6
'&&' clang commands together so that the test status reflects the results of all three clang executions.
...
llvm-svn: 53132
2008-07-04 04:38:48 +00:00
Ted Kremenek
8044046efb
Fix a bug in the dead stores checker reported in the following email:
...
http://lists.cs.uiuc.edu/pipermail/cfe-dev/2008-July/002157.html
Essentially the observer mechanism in LiveVariables was observing block-level
expressions multiple times, leading to a case where the dead store checker could
see a value as dead when it was really live.
llvm-svn: 53115
2008-07-03 22:25:27 +00:00