Re-apply 66024 with fixes: 1. Fixed indirect call to immediate address assembly. 2. Fixed JIT encoding by making the address pc-relative.

llvm-svn: 66803

llvm-svn: 66801

llvm-svn: 66800

llvm-svn: 66799

llvm-svn: 66798

llvm-svn: 66797

width of bitfields. 

I'll be burning this down and replacing it with a properly-dispatched
implementation like the one used for types.

llvm-svn: 66796

llvm-svn: 66795

llvm-svn: 66794

llvm-svn: 66793

llvm-svn: 66792

llvm-svn: 66791

llvm-svn: 66790

 - Notably, clang now exits with an error if it can't find a
   file. This flushed out a bug in the CGColorSpace.c test case. :)

llvm-svn: 66789

llvm-svn: 66788

llvm-svn: 66787

llvm-svn: 66786

llvm-svn: 66785

llvm-svn: 66784

llvm-svn: 66783

llvm-svn: 66782

llvm-svn: 66781

llvm-svn: 66780

related transformations out of target-specific dag combine into the
ARM backend.  These were added by Evan in r37685 with no testcases
and only seems to help ARM (e.g. test/CodeGen/ARM/select_xform.ll).

Add some simple X86-specific (for now) DAG combines that turn things
like cond ? 8 : 0  -> (zext(cond) << 3).  This happens frequently
with the recently added cp constant select optimization, but is a
very general xform.  For example, we now compile the second example
in const-select.ll to:

_test:
        movsd   LCPI2_0, %xmm0
        ucomisd 8(%esp), %xmm0
        seta    %al
        movzbl  %al, %eax
        movl    4(%esp), %ecx
        movsbl  (%ecx,%eax,4), %eax
        ret

instead of:

_test:
        movl    4(%esp), %eax
        leal    4(%eax), %ecx
        movsd   LCPI2_0, %xmm0
        ucomisd 8(%esp), %xmm0
        cmovbe  %eax, %ecx
        movsbl  (%ecx), %eax
        ret

This passes multisource and dejagnu.

llvm-svn: 66779

llvm-svn: 66778

Enable Chris' value propagation change. It make available known sign, zero, one bits information for values that are live out of basic blocks. The goal is to eliminate unnecessary sext, zext, truncate of values that are live-in to blocks. This does not handle PHI nodes yet.

llvm-svn: 66777

On x86, if the only use of a i64 load is a i64 store, generate a pair of double load and store instead.

llvm-svn: 66776

llvm-svn: 66775

llvm-svn: 66774

llvm-svn: 66773

llvm-svn: 66772

llvm-svn: 66770

understands.

llvm-svn: 66769

llvm-svn: 66768

 - Use OPT_ prefix for ids.

 - Reference groups and aliases by shortend id (on the theory that
   this is more readable).

 - Rename the special option ids to more protected names.

llvm-svn: 66767

llvm-svn: 66766

Remove the explicit if OS = Darwin test around the setting of -m32/-m64.

llvm-svn: 66765

-fblocks, and there were some duplicate options scattered in.

llvm-svn: 66764

llvm-svn: 66763

 - Mike, please verify.

llvm-svn: 66762