- 'for i in $(find . -type f); do sed -e 's#\(RUN:.*[^ ]\) *&& *$#\1#g' $i | FileUpdate $i; done', for the curious.

llvm-svn: 86430

llvm-svn: 86429

llvm-svn: 86428

llvm-svn: 86427

Use aligned load/store instructions for spilling Q registers when we know the stack slot is 128 bit aligned

llvm-svn: 86425

llvm-svn: 86423

the 'build_config' value in at runtime using the new lit runtime user parameter
feature.

This simplifies things and drops a dependency on 'sed', FWIW.

llvm-svn: 86421

llvm-svn: 86420

llvm-svn: 86419

llvm-svn: 86418

llvm-svn: 86417

llvm-svn: 86416

1. rename the movhp patfrag to movlhps, since thats what it actually matches
2. eliminate the bogus movhps load and store patterns, they were incorrect.  The load transforms are already handled (correctly) by shufps/unpack.
3. revert a recent test change to its correct form.

llvm-svn: 86415

llvm-svn: 86414

llvm-svn: 86413

llvm-svn: 86412

llvm-svn: 86411

llvm-svn: 86410

llvm-svn: 86408

llvm-svn: 86407

llvm-svn: 86404

llvm-svn: 86403

than an error about incompatible types. Patch by Sean Hunt.

llvm-svn: 86402

llvm-svn: 86400

Eric, please investigate.

llvm-svn: 86399

(eliminating some extends) if the new type of the
computation is legal or if both the source and dest
are illegal.  This prevents instcombine from changing big
chains of computation into i64 on 32-bit targets for 
example.

llvm-svn: 86398

Please verify.

llvm-svn: 86397

llvm-svn: 86396

llvm-svn: 86395

make it optional doesn't work out.  If you don't want to specify this, don't
specify a TD string at all.

llvm-svn: 86394

llvm-svn: 86393

llvm-svn: 86392

llvm-gcc-i386-darwin9 build bot.

llvm-svn: 86391

llvm-svn: 86390

llvm-svn: 86389

llvm-svn: 86387

llvm-svn: 86386

llvm-svn: 86385

llvm-svn: 86384

llvm-svn: 86383