to them.

llvm-svn: 60868

llvm-svn: 60867

llvm-svn: 60866

vec_extract-sse4.ll.

llvm-svn: 60865

llvm-svn: 60864

llvm-svn: 60863

Make sure to promote expressions of the form (floating point + complex integer) correctly, to (complex floating point + complex floating point)

llvm-svn: 60862

llvm-svn: 60861

- Added virtual method 'getSourceLocation()' (no arguments) that gets the location of the next "observable" location (e.g., next character, next token).
PPLexerChange.cpp:
- Implemented FIXME by using PreprocessorLexer::getSourceLocation() to get the location in the file we are returning to after lexing a #included file.  This appears to be slightly faster than having the branch (i.e., 'if(CurLexer)').  It's also not a really hot part of the Preprocessor.

llvm-svn: 60860

llvm-svn: 60859

"else" clause, e.g.,

  if (int X = foo()) {
  } else {
    if (X) { // warning: X is always zero in this context
    }
  }

Fixes rdar://6425550 and lets me think about something other than
DeclContext.

llvm-svn: 60858

them. The DAG combiner expects that nodes that are transformed have one value
result.

llvm-svn: 60857

The "real" fix for <rdar://problem/6424347> clang on xcode: Assertion failed: (0 && "unexpected type"), function mergeTypes,

Commit r60845 was premature.

llvm-svn: 60852

llvm-svn: 60851

llvm-svn: 60850

llvm-svn: 60849

llvm-svn: 60848

Fix <rdar://problem/6424347> clang on xcode: Assertion failed: (0 && "unexpected type"), function mergeTypes,

llvm-svn: 60845

Only perform SETO/SETC to JO/JC conversion if extractvalue is coming from an arithmetic with overflow instruction.

llvm-svn: 60844

Declare PerIDCache as IdentifierInfo** instead of void*.  This is just cleaner.  No performance change.

llvm-svn: 60843

llvm-svn: 60842

llvm-svn: 60841

llvm-svn: 60840

Fix regression caused by fixing <rdar://problem/6429113> clang ObjC rewriter: crash rewriting file with Blocks and properties

llvm-svn: 60839

llvm-svn: 60838

for promoted integer types, eg: i16 on ppc-32, or
i24 on any platform.  Complete support for arbitrary
precision integers would require handling expanded
integer types, eg: i128, but I couldn't be bothered.

llvm-svn: 60834

llvm-svn: 60833

causing a bunch of failures when running
"make ENABLE_EXPENSIVE_CHECKS=1 check".

llvm-svn: 60832

llvm-svn: 60831

Use a scoped object to manage entry/exit from a parser scope rather than explicitly calling EnterScope/ExitScope

llvm-svn: 60830

bit convert that changes the number of elements of a shuffle.

llvm-svn: 60829

Fix MachineCodeEmitter to use uintptr_t instead of intptr_t. This avoids some overflow issues. Patch by Thomas Jablin.

llvm-svn: 60828

llvm-svn: 60827

llvm-svn: 60826

to reflect reality.

llvm-svn: 60825

Fix a bug introduced by r59265. If lazy compilation is disabled, return actual function ptr instead of ptr to stub if function is already compiled.

llvm-svn: 60822

llvm-svn: 60821

parallel, allowing it to decide that P/Q must alias if A/B
must alias in things like:
 P = gep A, 0, i, 1
 Q = gep B, 0, i, 1

This allows GVN to delete 62 more instructions out of 403.gcc.

llvm-svn: 60820

llvm-svn: 60819

llvm-svn: 60818