functions in these two schedulers.

llvm-svn: 59465

patch by Richard Osborne, tweaked and extended by your humble
servant.

llvm-svn: 59464

llvm-svn: 59461

a side-car datastructure

llvm-svn: 59458

are not currently done (cf PowerPC/README.txt).

llvm-svn: 59456

optimizations) into this wishlist.

llvm-svn: 59455

llvm-svn: 59454

for the target. This fixes PR3080.

llvm-svn: 59450

llvm-svn: 59449

llvm-svn: 59448

Use strings instead of TableGen defs in the compilation graph
definition. Makes it easier for the plugins to modify an existing graph.

llvm-svn: 59447

llvm-svn: 59445

llvm-svn: 59434

llvm-svn: 59432

llvm-svn: 59431

llvm-svn: 59417

This fixed PR3078.

llvm-svn: 59416

2008-10-17-SpillerBug.ll is currently failing, but this doesn't reflect an actual regression of PR2898. This test should probably be removed. I've XFAILed it for now to keep buildbot quiet while this is considered.

llvm-svn: 59415

llvm-svn: 59414

llvm-svn: 59409

llvm-svn: 59406

llvm-svn: 59403

llvm-svn: 59402

llvm-svn: 59401

llvm-svn: 59400

can extract using different indexes for two vectors. Added a few tests
for vector shuffles.

llvm-svn: 59399

llvm-svn: 59398

a bit.  Next step is to factor out into their own helper functions.

llvm-svn: 59397

llvm-svn: 59395

llvm-svn: 59394

llvm-svn: 59393

llvm-svn: 59392

llvm-svn: 59391

llvm-svn: 59390

folded into the select.

llvm-svn: 59389

Eliminate the cases checking for cond?0:-1, since that is already
handled by commutative checking.

llvm-svn: 59388

llvm-svn: 59387

llvm-svn: 59386

llvm-svn: 59385

If this patch causes a performance regression for anyone, please let me know,
and it can be fixed in a different way with much more effort.

llvm-svn: 59384