llvm-svn: 24884

Only run lower-allocations and lower-select for the simple isel

llvm-svn: 24881

For example, instead of emitting this:

test:
        save -40112, %o6, %o6   ;; imm too large
        add %i6, -40016, %o0    ;; imm too large
        call caller
        nop
        restore %g0, %g0, %g0
        retl
        nop

emit this:

test:
        sethi 4194264, %g1
        or %g1, 848, %g1
        save %o6, %g1, %o6
        sethi 4194264, %g1
        add %g1, %i6, %g1
        add %i1, 944, %o0
        call caller
        nop
        restore %g0, %g0, %g0
        retl
        nop

which doesn't cause the assembler to barf.

llvm-svn: 24880

llvm-svn: 24879

llvm-svn: 24877

last night, breaking crafty and twolf.  Make sure that the newly found
legal nodes are themselves not re-legalized until the next iteration.

Also, since this functionality exists now, we can reduce number of legalizer
iterations by depending on this behavior instead of having to misuse 'do
another iteration' to get the same effect.

llvm-svn: 24875

llvm-svn: 24874

us to load and store vectors directly at a pointer (offset of zero) by
using r0 as the base register.  This also requires some asm printer work
to satisfy the darwin assembler.

For
void %foo(<4 x float> * %a) {
entry:
  %tmp1 = load <4 x float> * %a;
  %tmp2 = add <4 x float> %tmp1, %tmp1
  store <4 x float> %tmp2, <4 x float> *%a
  ret void
}

We now produce:
_foo:
        lvx v0, 0, r3
        vaddfp v0, v0, v0
        stvx v0, 0, r3
        blr

Instead of:
_foo:
        li r2, 0
        lvx v0, r2, r3
        vaddfp v0, v0, v0
        stvx v0, r2, r3
        blr

llvm-svn: 24872

llvm-svn: 24871

llvm-svn: 24870

llvm-svn: 24869

llvm-svn: 24867

creating graphs that cannot be scheduled.

llvm-svn: 24866

Make the dsa-alloc-list and dsa-free-list options hidden.

llvm-svn: 24864

llvm-svn: 24863

functions that allocate memory.

llvm-svn: 24862

llvm-svn: 24861

latency period between the two.

llvm-svn: 24860

we were storing into [FP+88] instead of [FP+92].
Improve codegen by emitting [FP+92], instead of emitting a copy of FP into
another GPR which wouldn't be coallesced because FP isn't register allocated.

llvm-svn: 24859

llvm-svn: 24857

llvm-svn: 24856

llvm-svn: 24855

llvm-svn: 24854

llvm-svn: 24853

llvm-svn: 24852

llvm-svn: 24850

llvm-svn: 24849

llvm-svn: 24848

llvm-svn: 24847

llvm-svn: 24846

llvm-svn: 24845

llvm-svn: 24844

llvm-svn: 24843

llvm-svn: 24842

llvm-svn: 24841

llvm-svn: 24840

llvm-svn: 24839

llvm-svn: 24838

llvm-svn: 24837

llvm-svn: 24836