llvm-svn: 22963

float %test2(float* %P) {
        %Q = load float* %P
        %R = add float %Q, %Q
        ret float %R
}

By returning the right result.

llvm-svn: 22961

Make sure expressions only have one use before emitting them into a place that is conditionally executed

llvm-svn: 22960

llvm-svn: 22959

llvm-svn: 22953

simplifies BRTWOWAY into BR if one of the results is a fall-through.
Unless I'm missing something, there is no reason to duplicate this
in the target-specific code.

llvm-svn: 22952

llvm-svn: 22951

it in the block.  This codegens:

int %test(bool %c) {
        %retval = select bool %c, int 17, int 1
        ret int %retval
}

as:

_test:
        rlwinm r2, r3, 0, 31, 31
        li r2, 17
        cmpwi cr0, r3, 0
        bne .LBB_test_2 ;
.LBB_test_1:    ;
        li r2, 1
.LBB_test_2:    ;
        or r3, r2, r2
        blr

instead of:

_test:
        rlwinm r2, r3, 0, 31, 31
        li r2, 17
        li r4, 1
        cmpwi cr0, r3, 0
        bne .LBB_test_2 ;
.LBB_test_1:    ;
        or r2, r4, r4
.LBB_test_2:    ;
        or r3, r2, r2
        blr

... which is one fewer instruction.  The savings are more significant for
global address and constantfp nodes.

llvm-svn: 22946

llvm-svn: 22944

llvm-svn: 22942

This REALLY should be lowered by the legalizer!

llvm-svn: 22941

llvm-svn: 22935

llvm-svn: 22929

llvm-svn: 22927

llvm-svn: 22926

llvm-svn: 22925

llvm-svn: 22924

namespace instead of the reg class namespace.  Update getRegClassForType()
to use modified names due to tblgen change.

llvm-svn: 22923

llvm-svn: 22922

llvm-svn: 22921

llvm-svn: 22917

llvm-svn: 22916

llvm-svn: 22915

llvm-svn: 22914

llvm-svn: 22913

llvm-svn: 22909

(which died about a week ago) so we're back to load-(2^n-1)-then-AND
sequences. slow, but things should now be Almost Completely Working,
modulo those pesky alignment/ABI issues.

llvm-svn: 22904

instruction.

llvm-svn: 22901

cleanest backend in this metric :)

llvm-svn: 22893

I'm not convinced this is all of them,  but I can't do much testing, because
IA64 LLC crashes on big programs :(

llvm-svn: 22892

llvm-svn: 22891

Give a whole bunch of other stuff variable operands, particularly FP.  The
FP stackifier is playing fast and loose with operands here, so we have to
mark them all as variable.  This will have to be fixed before we can dag->dag
the X86 backend.  The solution is for the pre-stackifier and post-stackifier
instructions to all be disjoint.

llvm-svn: 22890

llvm-svn: 22889

llvm-svn: 22888

only take one operand.  The other comes implicitly in through CL.

llvm-svn: 22887

passed.

llvm-svn: 22886

llvm-svn: 22884

the operand lists of a few other nodes.

llvm-svn: 22883

llvm-svn: 22882

llvm-svn: 22881