llvm-svn: 60769

llvm-svn: 60767

llvm-svn: 60766

llvm-svn: 60765

llvm-svn: 60764

llvm-svn: 60755

llvm-svn: 60754

llvm-svn: 60753

the first block of a query specially.  This makes the "complete query
caching" subsystem more effective, avoiding predecessor queries.  This
speeds up GVN another 4%.

llvm-svn: 60752

and split its inner loop out into a new GetNonLocalInfoForBlock
function.  No functionality change.

llvm-svn: 60751

llvm-svn: 60750

one of its operand.

llvm-svn: 60749

llvm-svn: 60748

This speeds up the new GVN by another 3%

llvm-svn: 60747

position instead of using a full sort. This speeds up GVN by ~4% with the
new memdep stuff.

llvm-svn: 60746

pred iterator.

llvm-svn: 60745

llvm-svn: 60744

queries.  This speeds up GVN using the new queries (not yet
checked in) by just over 10%.

llvm-svn: 60743

really simple cache class for these queries.  Hopefully this can
be removed if pred_iterator speeds back up.

llvm-svn: 60742

- Fix call.ll and call_indirect.ll expected results, now that it's using a
  different pre-register allocation scheduler.

llvm-svn: 60741

Fix the shift amount when unrolling a vector shift into scalar shifts.
Fix problem in getShuffleScalarElt where it assumes that the input of
a bit convert must be a vector.

llvm-svn: 60740

on test/CodeGen/Generic/2007-06-06-CriticalEdgeLandingPad.

llvm-svn: 60739

- Change default scheduling preference to list-burr, which produces somewhat
  better code than the default. Could also use list-tdrr, but need to ask
  dev list about the appropriate handy mnemonic before commiting.

llvm-svn: 60738

complete. For instance, it lowers the common case into this less-than-optimal
code:

        addl    %ecx, %eax
        seto    %cl
        testb   %cl, %cl
        jne     LBB1_2  ## overflow

instead of:

        addl    %ecx, %eax
        jo      LBB1_2  ## overflow

That will come in a future commit.

llvm-svn: 60737

llvm-svn: 60736

llvm-svn: 60735

pipeline model.

llvm-svn: 60733

for spotting these!

llvm-svn: 60728

llvm-svn: 60727

jump threading has been shown to only expose problems not
have bugs itself.  I'm sure it's completely bug free! ;-)

llvm-svn: 60725

llvm-svn: 60711

llvm-svn: 60707

avoids GCC's warning even on 64-bit hosts.

llvm-svn: 60706

llvm-svn: 60705

llvm-svn: 60703

Thanks Duncan!

llvm-svn: 60702

llvm-svn: 60701

llvm-svn: 60698

llvm-svn: 60697

and use it in x86 address mode folding. Also, make
getRegForValue return 0 for illegal types even if it has a
ValueMap for them, because Argument values are put in the
ValueMap. This fixes PR3181.

llvm-svn: 60696