llvm-svn: 24237

llvm-svn: 24236

llvm-svn: 24167

llvm-svn: 24166

Get the UIUC copyright to actually print.

llvm-svn: 24165

llvm-svn: 24160

other, traditional UNIX greps).

llvm-svn: 24156

The configure script no longer tells us that we're configuring for SparcV9
specifically.
2004-06-17-UnorderedCompares may work on SparcV8, but it's experiental
anyway.
2005-02-20-AggregateSAVEEXPR should fail on any Solaris machine, as Solaris
doesn't provide complex number support.

llvm-svn: 24155

llvm-svn: 24138

llvm-svn: 24137

of having to hack the nightly tester script itself.

as an example, I use the following for my machine:

$HOME/llvm/utils/NightlyTest.pl -parallel -release -enable-llcbeta \
     -spec2000path /Volumes/ProjectsDisk/cvs/benchmarks/speccpu2000-llvm/benchspec/ \
     -povraypath /Volumes/ProjectsDisk/cvs/benchmarks/povray31 \
     -namdpath /Volumes/ProjectsDisk/cvs/benchmarks/namd

llvm-svn: 24136

need to send chris, jim and sampo a box of fish each

llvm-svn: 24135

one sometimes needs to pass FP args in both FP *and* integer registers.

llvm-svn: 24134

clever little tablegen!

llvm-svn: 24133

llvm-svn: 24132

llvm-svn: 24131

llvm-svn: 24130

not compiling a whole program at a time :)

llvm-svn: 24129

llvm-svn: 24126

llvm-svn: 24125

llvm-svn: 24124

bad cases.  This fixes Markus's second testcase in PR639, and should
seal it for good.

llvm-svn: 24123

llvm-svn: 24122

llvm-svn: 24121

llvm-svn: 24120

1. Embed and not inherit vector for NodeGroup.

2. Iterate operands and not uses (performance.)

3. Some long pending comment changes.

llvm-svn: 24119

llvm-svn: 24118

llvm-svn: 24117

a special case hack for X86, make the hack more general: if an incoming argument
register is not used in any block other than the entry block, don't copy it to
a vreg.  This helps us compile code like this:

%struct.foo = type { int, int, [0 x ubyte] }
int %test(%struct.foo* %X) {
        %tmp1 = getelementptr %struct.foo* %X, int 0, uint 2, int 100
        %tmp = load ubyte* %tmp1                ; <ubyte> [#uses=1]
        %tmp2 = cast ubyte %tmp to int          ; <int> [#uses=1]
        ret int %tmp2
}

to:

_test:
        lbz r3, 108(r3)
        blr

instead of:

_test:
        lbz r2, 108(r3)
        or r3, r2, r2
        blr

The (dead) copy emitted to copy r3 into a vreg for extra-block uses was
increasing the live range of r3 past the load, preventing the coallescing.

This implements CodeGen/PowerPC/reg-coallesce-simple.ll

llvm-svn: 24115

llvm-svn: 24112

generating results in vregs that will need them.  In the case of something
like this:  CopyToReg((add X, Y), reg1024), we no longer emit code like
this:

   reg1025 = add X, Y
   reg1024 = reg 1025

Instead, we emit:

   reg1024 = add X, Y

Whoa! :)

llvm-svn: 24111

llvm-svn: 24110

llvm-svn: 24109

llvm-svn: 24108

llvm-svn: 24107

This implements test/Regression/CodeGen/PowerPC/mul-neg-power-2.ll,
producing:

_foo:
        slwi r2, r3, 1
        subfic r3, r2, 63
        blr

instead of:

_foo:
        mulli r2, r3, -2
        addi r3, r2, 63
        blr

llvm-svn: 24106

llvm-svn: 24103

When inserting code for an addrec expression with a non-unit stride, be
more careful where we insert the multiply.  In particular, insert the multiply
in the outermost loop we can, instead of the requested insertion point.

This allows LSR to notice the mul in the right loop, reducing it when it gets
to it.  This allows it to reduce the multiply, where before it missed it.

This happens quite a bit in the test suite, for example, eliminating 2
multiplies in art, 3 in ammp, 4 in apsi, reducing from 1050 multiplies to
910 muls in galgel (!), from 877 to 859 in applu, and 36 to 30 in bzip2.

This speeds up galgel from 16.45s to 16.01s, applu from 14.21 to 13.94s and
fourinarow from 66.67s to 63.48s.

This implements Transforms/LoopStrengthReduce/nested-reduce.ll

llvm-svn: 24102

llvm-svn: 24099

llvm-svn: 24098