llvm-svn: 54968

llvm-svn: 54967

llvm-svn: 54964

llvm-svn: 54958

llvm-svn: 54957

  1) Assign stack slots to new temporaries.
  2) Don't insert an interval into the return vector more than once.

llvm-svn: 54956

Add ExplicitCastExpr to replace the current CastExpr, and have ImplicitCastExpr and ExplicitCastExpr derive from a common base class (CastExpr):

Expr
  -> CastExpr
     -> ExplicitCastExpr
     -> ImplicitCastExpr 

llvm-svn: 54955

llvm-svn: 54954

of pointers to them.  rdar://4304469

llvm-svn: 54953

llvm-svn: 54952

llvm-svn: 54950

  1) Remove an incorrect assertion.
  2) Set the stack slot weight properly.
  3) Resize the VirtRegMap when needed.

llvm-svn: 54949

llvm-svn: 54947

  extension).

llvm-svn: 54946

llvm-svn: 54943

llvm-svn: 54942

no other functionality change.

llvm-svn: 54941

llvm-svn: 54940

llvm-svn: 54939

printing ascii code for hex numbers instead of the hex numbers 
themselves.

llvm-svn: 54936

llvm-svn: 54931

llvm-svn: 54930

llvm-svn: 54929

llvm-svn: 54928

llvm-svn: 54925

Resurrect some ancient code to add spill ranges without attempting folding, remat, or splitting.  This code has been updated to current APIs
in so far as it compiles and, in theory, works, but does not take advantage of recent advancements.  For instance, it could be improved by using
MachineRegisterInfo::use_iterator.

llvm-svn: 54924

llvm-svn: 54923

llvm-svn: 54922

so that lli works out of the box with -enable-eh.

llvm-svn: 54920

llvm-svn: 54914

llvm-svn: 54913

hoisting out some "test" and "cmp" instructions. This was obvious badness.

llvm-svn: 54908

predicate, swap the order of the operands.

llvm-svn: 54907

produce an xor by 127.

llvm-svn: 54906

llvm-svn: 54905

llvm-svn: 54904

Fix a (u)comiss intrinsic lowering bug. It was using anyext which can return junk in higher bits. Patch by Nate Begeman.

llvm-svn: 54903

llvm-svn: 54902

llvm-svn: 54901

llvm-svn: 54900