c0f15487c6
Fix analyzer breakage introduced by r49213: http://llvm.org/viewvc/llvm-project?rev=49213&view=rev
...
The problem is that some clients of gcc lookout at the stdout output of gcc;
this requires that all the analyzer diagnostics go to stderr to avoid polluting
stdout.
llvm-svn: 49355
2008-04-07 23:27:54 +00:00
a518bb79b1
Make getDirnameSep a static method (not part of Path's interface).
...
llvm-svn: 49354
2008-04-07 22:01:32 +00:00
e0680b7e7a
Pass the ASTContext object around when deserializing Decl and Stmt objects, so
...
they can be created using the same allocator as in the "from source code" case.
llvm-svn: 49353
2008-04-07 21:55:54 +00:00
c042bf1db5
Added method Path::getDirname().
...
llvm-svn: 49352
2008-04-07 21:53:57 +00:00
3a13067557
Added support for Create() calls that take an argument besides the
...
deserializer.
llvm-svn: 49350
2008-04-07 21:36:46 +00:00
fba3942438
Have Parser::FuzzyParseMicrosoftAsmStatement() return the null statement (';').
...
llvm-svn: 49349
2008-04-07 21:06:54 +00:00
3bc3ddd638
Rename MemOperand to MachineMemOperand. This was suggested by
...
review feedback from Chris quite a while ago. No functionality
change.
llvm-svn: 49348
2008-04-07 19:35:22 +00:00
18948355e1
Remove llvm-upgrade.
...
llvm-svn: 49347
2008-04-07 18:32:47 +00:00
c7972310bb
Added note to getting started guide to use llvm-gcc4.2.
...
Removed llvm-upgrade from documentations.
llvm-svn: 49346
2008-04-07 18:32:25 +00:00
ed92b41a39
Add operator= implementations to SparseBitVector, allowing it to be used in GVN. This results
...
in both time and memory savings for GVN. For example, one testcase went from 10.5s to 6s with
this patch.
llvm-svn: 49345
2008-04-07 17:38:23 +00:00
d6481955db
Testcase for pr2169.
...
llvm-svn: 49344
2008-04-07 17:03:16 +00:00
7b93268305
Fix test.
...
llvm-svn: 49343
2008-04-07 17:02:18 +00:00
b646062ad5
Fix PR 2169.
...
llvm-svn: 49339
2008-04-07 14:20:50 +00:00
813384951e
Use Intrinsic::getDeclaration in more places.
...
llvm-svn: 49338
2008-04-07 13:45:04 +00:00
1416ebf1fe
The "stacksave is not nounwind problem" no longer
...
needs to be fixed here - a previous commit made sure
that intrinsics always get the right attributes.
So remove no-longer needed code, and while there use
Intrinsic::getDeclaration rather than getOrInsertFunction.
llvm-svn: 49337
2008-04-07 13:43:58 +00:00
fbc6adcc59
Use Intrinsic::getDeclaration to get hold of
...
intrinsics. Fix up the argument type (should
be i8*, was an array*).
llvm-svn: 49336
2008-04-07 13:41:19 +00:00
a8ff6ca2e3
Make sure that intrinsics automagically get the
...
right parameter attributes no matter how they
are obtained.
llvm-svn: 49335
2008-04-07 13:39:11 +00:00
51f532f92d
Re-commit of the r48822, where the infinite looping problem discovered
...
by Dan Gohman is fixed.
llvm-svn: 49330
2008-04-07 10:06:32 +00:00
0c1e634cbb
Make GVN more memory efficient, particularly on code that contains a large number of
...
allocations, which GVN can't optimize anyways.
llvm-svn: 49329
2008-04-07 09:59:07 +00:00
4ba0cef93b
random whitespace fixes.
...
llvm-svn: 49328
2008-04-07 07:01:58 +00:00
fa00732562
fix a bug I introduced in my previous checkin.
...
llvm-svn: 49327
2008-04-07 06:58:21 +00:00
e6b27c6ecd
simplify array compatibility testing.
...
llvm-svn: 49326
2008-04-07 06:56:55 +00:00
7460fd297d
simplify reference handling.
...
llvm-svn: 49325
2008-04-07 06:52:53 +00:00
49af6a4beb
move some code around, no other change.
...
llvm-svn: 49324
2008-04-07 06:51:04 +00:00
ec646834e7
simplify compatibility testing for tag types.
...
llvm-svn: 49323
2008-04-07 06:49:41 +00:00
3c7961597f
merge compatibility testing of qualified/unqualified interfaces together
...
and fix a bug.
llvm-svn: 49322
2008-04-07 06:38:24 +00:00
64bf6ba308
Fix a really bad bug where type uniquing would merge a<x> with b<x> as the same
...
type, because it did not include a/b in the hash.
llvm-svn: 49321
2008-04-07 06:37:47 +00:00
c9062d01d7
MyOtherClass<MyProtocol>* is compatible with MyClass*
...
if MyClass is a superclass of MyOtherClass, there is no need for
an exact interface match.
llvm-svn: 49320
2008-04-07 06:06:56 +00:00
7bbd3d75ef
Remove a dead check for compatible builtin types
...
llvm-svn: 49319
2008-04-07 05:55:38 +00:00
90bef90ad4
futher simplify compatibility testing of objc interface types.
...
llvm-svn: 49318
2008-04-07 05:53:18 +00:00
856d8f5b58
Remove tabs. Patch by Mike Stump!
...
llvm-svn: 49317
2008-04-07 05:47:52 +00:00
9555466f7e
ocuvector and vector should be compatible. Fix ASQual compatibility.
...
llvm-svn: 49316
2008-04-07 05:43:21 +00:00
aee96c3a90
Fix comment typo, do reference eval at the correct type.
...
llvm-svn: 49315
2008-04-07 05:37:56 +00:00
f8a6b4d4fc
simplify vector type compatibility testing.
...
llvm-svn: 49314
2008-04-07 05:36:14 +00:00
2a3569b5d9
move ObjCQualifiedIdTypesAreCompatible out of ASTContext into Sema.
...
While it is similar to the other compatibility predicates in ASTContext,
it is not used by them and is different.
In addition, greatly simplify ObjCQualifiedIdTypesAreCompatible and
fix some canonical type bugs. Also, simplify my Type::getAsObjC* methods.
llvm-svn: 49313
2008-04-07 05:30:13 +00:00
ae947fe3f9
Replace an O(n^2) algorithm in areCompatObjCQualInterfaces with
...
an O(n) algorithm by taking advantage of the fact that the
protocol qualifier list is already guaranteed sorted.
llvm-svn: 49312
2008-04-07 05:05:41 +00:00
fde0761735
improve comments.
...
llvm-svn: 49311
2008-04-07 04:58:07 +00:00
e0ea37ace7
move sorting of qualifying protocols from the parser into
...
sema. This allows clients of the parser to have the unmolested
list if desired, and guarantees that noone can create an
ObjCQualifiedInterfaceType with an unsorted list.
llvm-svn: 49310
2008-04-07 04:56:42 +00:00
513494b32b
add an example.
...
llvm-svn: 49309
2008-04-07 04:47:29 +00:00
bd7981a978
eliminate getReferencedProtocols from
...
ObjCQualifiedIdType/ObjCQualifiedInterfaceType, adding an interator
interface instead.
llvm-svn: 49308
2008-04-07 04:44:08 +00:00
f96084ee73
make QualifiedInterfaceTypesAreCompatible a static function
...
and start simplifying it.
llvm-svn: 49307
2008-04-07 04:17:40 +00:00
ae1b6078e3
This predicate is just a generic "issuperclass" predicate, move it to the
...
ObjCInterfaceType as a method.
llvm-svn: 49306
2008-04-07 04:13:03 +00:00
ddfdaf9a69
Simplify some objc compatibility testing, make interfaceTypesAreCompatible
...
a static function named isCompatibleInterfaceAssign.
llvm-svn: 49305
2008-04-07 04:07:56 +00:00
f8c3e82897
simplify the logic in ASTContext::objcTypesAreCompatible
...
llvm-svn: 49302
2008-04-07 01:30:37 +00:00
ba5862e74c
clean up some logic in objc type handling. Specifically, make it so that
...
there are QualType::getAsObjc* type methods, and make isa<ObjCInterfaceType>
return true for ObjCQualifiedInterfaceType's.
llvm-svn: 49300
2008-04-07 00:27:04 +00:00
87e484f08b
Mark calls to llvm.stacksave, llvm.stackrestore as
...
nounwind. When such calls are inlined into something
else that is invoked, they were getting changed to invokes,
which is badness.
llvm-svn: 49299
2008-04-07 00:08:48 +00:00
b9dfb03ddb
trivial changes to getFloatingTypeOfSizeWithinDomain, nothing significant.
...
llvm-svn: 49298
2008-04-06 23:58:54 +00:00
d4bacd68bf
simplify max type computation by making it return an integer (like
...
getFloatingTypeOrder) instead of a type. Fix a fixme.
llvm-svn: 49297
2008-04-06 23:55:33 +00:00
b90739d556
minor simplifications/cleanups to type comparisons.
...
llvm-svn: 49296
2008-04-06 23:38:49 +00:00
fc4379f916
remove the Decl::getCanonicalType() method.
...
llvm-svn: 49295
2008-04-06 23:10:54 +00:00
Ted Kremenek