Handle [NSAutoreleasePool addObject:] (an alternative way to add objects to the current autorelease pool).

Added initial code for tracking stack of autorelease pools.

llvm-svn: 57908

llvm-svn: 57907

Found this while fixing another unrelated radar.

llvm-svn: 57904

llvm-svn: 57903

llvm-svn: 57902

Tweak an expected-error to match the output. IMHO it is better to diagnose 'int (void)' because it has the same meaning in C and C++.

llvm-svn: 57901

fix a tricky bug in the JIT global variable emitter, that was triggered when JITing a variable independently of a function. This lead to sharing memory memory between functions and GVs thus changing the value of a GV could change the code in execution. more details on the ML.

llvm-svn: 57900

llvm-svn: 57899

Fix <rdar://problem/6265257> warnings for ambiguous message send swamp other warnings.

Reworked Sema::MatchTwoMethodDeclarations() to optionally match based on method size and alignment (the default in GCC). Changed Sema::LookupInstanceMethodInGlobalPool() to use this feature.

Added -Wno-struct-selector-match to driver, however didn't hook it up yet. Added a FIXME that says this.

llvm-svn: 57898

GRExprEngine::VisitCast() so that other parts of the analyzer can be ignorant.
When we cast "array of type T" to "pointer to T", we get the loc::MemRegionVal
corresponding to the array variable. This is sufficient for BasicStore, but not
for RegionStore. RegionStore should get the element region for the first array
element in the cast. So next comes to the revamping of transfer functions for
different store model.

llvm-svn: 57897

llvm-svn: 57896

llvm-svn: 57895

llvm-svn: 57894

  RValues.h/cpp => SVals.h/cpp

llvm-svn: 57893

initialized to UndefinedVal.

llvm-svn: 57892

llvm-svn: 57891

llvm-svn: 57890

llvm-svn: 57889

llvm-svn: 57888

llvm-svn: 57887

isn't yet prepared for it.

llvm-svn: 57886

in the 32-bit signed offset field of addresses. Even though this
may be intended, some linkers refuse to relocate code where the
relocated address computation overflows.

Also, fix the sign-extension of constant offsets to use the
actual pointer size, rather than the size of the GlobalAddress
node, which may be different, for example on x86-64 where MVT::i32
is used when the address is being fit into the 32-bit displacement
field.

llvm-svn: 57885

llvm-svn: 57878

llvm-svn: 57877

llvm-svn: 57876

Where previously LLVM might emit code like this:

        ucomisd %xmm1, %xmm0
        setne   %al
        setp    %cl
        orb     %al, %cl
        jne     .LBB4_2

it now emits this:

        ucomisd %xmm1, %xmm0
        jne     .LBB4_2
        jp      .LBB4_2

It has fewer instructions and uses fewer registers, but it does
have more branches. And in the case that this code is followed by
a non-fallthrough edge, it may be followed by a jmp instruction,
resulting in three branch instructions in sequence. Some effort
is made to avoid this situation.

To achieve this, X86ISelLowering.cpp now recognizes FCMP_OEQ and
FCMP_UNE in lowered form, and replace them with code that emits
two branches, except in the case where it would require converting
a fall-through edge to an explicit branch.

Also, X86InstrInfo.cpp's branch analysis and transform code now
knows now to handle blocks with multiple conditional branches. It
uses loops instead of having fixed checks for up to two
instructions. It can now analyze and transform code generated
from FCMP_OEQ and FCMP_UNE.

llvm-svn: 57873

the copy instruction from the instruction list before asking the
target to create the new instruction. This gets the old instruction
out of the way so that it doesn't interfere with the target's
rematerialization code. In the case of x86, this helps it find
more cases where EFLAGS is not live.

Also, in the X86InstrInfo.cpp, teach isSafeToClobberEFLAGS to check
to see if it reached the end of the block after scanning each
instruction, instead of just before. This lets it notice when the
end of the block is only two instructions away, without doing any
additional scanning.

These changes allow rematerialization to clobber EFLAGS in more
cases, for example using xor instead of mov to set the return value
to zero in the included testcase.

llvm-svn: 57872

that NaNs are less common.

llvm-svn: 57871

llvm-svn: 57870

llvm-svn: 57869

llvm-svn: 57864

llvm-svn: 57863

for strange asm conditions earlier.  In this case, we have a
double being passed in an integer reg class.  Convert to like
sized integer register so that we allocate the right number 
for the class (two i32's for the f64 in this case).

llvm-svn: 57862

Further improve path-sensitivity with divide-by-zero checking by assuming that a denominator cannot be zero even when the result of an '/' or '%' expression is unknown.

llvm-svn: 57855

Added test case inspired by <rdar://6268365>: recover path-sensitivity after compound assignment when the result of the assignment is not known.

llvm-svn: 57852

Used conjured symbols to recover path-sensitivity when the result of a compound assignment is UnknownVal().

llvm-svn: 57851

llvm-svn: 57850

llvm-svn: 57847

is re-written by the callback to branch directly to the compiled code
in future invocations.

Added back in range-based memory permission functions for the updating of
the stub on Darwin.

llvm-svn: 57846

llvm-svn: 57845