llvm-svn: 11088

llvm-svn: 11087

TwoAddressInstructionPass can correctly update use/def information.

llvm-svn: 11086

outside of loops = 0.

llvm-svn: 11085

llvm-svn: 11084

llvm-svn: 11083

llvm-svn: 11082

llvm-svn: 11081

llvm-svn: 11080

(I just love typing haX0r.   haX0r haX0r haX0r.)

llvm-svn: 11079

llvm-svn: 11078

llvm-svn: 11077

and addPassesToJITCompile() slightly.

llvm-svn: 11076

big-endian, 32-bit module, or if __ppc__, __POWERPC__, or __APPLE__ are
defined.

llvm-svn: 11075

llvm-svn: 11074

PowerPC-specific passes here.

llvm-svn: 11073

llvm-svn: 11072

llvm-svn: 11070

llvm-svn: 11069

llvm-svn: 11068

llvm-svn: 11067

much better when I get rid of the reserved registers.

llvm-svn: 11066

llvm-svn: 11065

Basically we store floating point values as their integral components, instead of relying
on the semantics of floating point < to differentiate between values.  This is likely to
make the map search be faster anyway.

llvm-svn: 11064

llvm-svn: 11063

llvm-svn: 11062

registers (not as the max number of registers).

Change toSpill from a std::set into a std::vector<bool>.

Use the reverse iterator adapter to do a reverse scan of allocatable
registers.

llvm-svn: 11061

where appropriate.

llvm-svn: 11060

llvm-svn: 11059

llvm-svn: 11058

llvm-svn: 11057

llvm-svn: 11055

llvm-svn: 11054

This tremendously improves the code generated by the LLVM optimizer, primarily
by making the inliner more aggressive.  For example, it improves the stepanov
benchmark from 55.56 mega-additions/sec to 98.04 Ma/s.  It also improves the
oopack/iterator benchmark from 338.3MFLOPS/s to 1103.4MFLOPS/s.  Less noteworthy,
it improves oopack/matrix from 573 -> 641 MFLOPS/s.

llvm-svn: 11053

llvm-svn: 11052

llvm-svn: 11050

llvm-svn: 11049

llvm-svn: 11048

llvm-svn: 11047

term) working on bytecode size stuff.

llvm-svn: 11046