llvm-svn: 90804

llvm-svn: 90793

llvm-svn: 90792

llvm-svn: 90791

Use FileCheck and set nounwind on calls.

llvm-svn: 90790

number in order to accomodate targets with more than 1024 registers.

llvm-svn: 90789

The semantics of llvm.dbg.value are that starting from where it is executed, an offset into the specified user source variable is specified to get a new value.

An example:
  call void @llvm.dbg.value(metadata !{ i32 7 }, i64 0, metadata !2)
Here the user source variable associated with metadata #2 gets the value "i32 7" at offset 0.
 

llvm-svn: 90788

llvm-svn: 90785

llvm-svn: 90784

llvm-svn: 90782

current form, it is too expensive in compile time.

llvm-svn: 90781

anything.

llvm-svn: 90779

llvm-svn: 90776

llvm-svn: 90775

Use 'forward_values' + 'comma_separated' instead.

llvm-svn: 90774

llvm-svn: 90771

llvm-svn: 90770

llvm-svn: 90764

llvm-svn: 90759

llvm-svn: 90754

sys::cas_flag should be long on this platform, InterlockedAdd() is
defined only for the Itanium architecture (according to MSDN).

Patch by Michael Beck!

llvm-svn: 90748

llvm-svn: 90744

llvm-svn: 90742

Truncate the arguments of llvm.frameaddress / llvm.returnaddress intrinsics from i32 to platform's largest native type

llvm-svn: 90741

llvm-svn: 90740

llvm-svn: 90739

llvm-svn: 90738

llvm-svn: 90737

llvm-svn: 90728

in "bic sp, sp, #15" leads to unpredicatble behaviour in Thumb2 mode.
Emit the following code instead:
mov r4, sp
bic r4, r4, #15
mov sp, r4

llvm-svn: 90724

llvm-svn: 90708

llvm-svn: 90707

because clang lowers nontrivial automatic struct/array inits to memcpy from
a global array.

llvm-svn: 90698

llvm-svn: 90697

llvm-svn: 90696

short x(short *A) {
  memset(A, 1, sizeof(*A)*100);
  return A[42];
}

to 'return 257' instead of doing the load.  

llvm-svn: 90695

shift amount.

llvm-svn: 90694

llvm-svn: 90691

llvm-svn: 90685

Patch by Michael Beck!

llvm-svn: 90683