llvm-svn: 97543

llvm-svn: 97541

stuff now that we don't care about emulating the old broken 
behavior of the old isel.  This eliminates the 
'CheckChainCompatible' check (along with IsChainCompatible) which
did an incorrect and inefficient scan *up* the chain nodes which
happened as the pattern was being formed and does the validation
at the end in HandleMergeInputChains when it forms a structural 
pattern.  This scans "down" the graph, which means that it is
quickly bounded by nodes already selected.  This also handles
token factors that get "trapped" in the dag.

Removing the CheckChainCompatible nodes also shrinks the 
generated tables by about 6K for X86 (down to 83K).

There are two pieces remaining before I can nuke PreprocessRMW:
1. I xfailed a test because we're now producing worse code in a 
   case that has nothing to do with the change: it turns out that
   our use of MorphNodeTo will leave dead nodes in the graph
   which (depending on how the graph is walked) end up causing
   bogus uses of chains and blocking matches.  This is really 
   bad for other reasons, so I'll fix this in a follow-up patch.

2. CheckFoldableChainNode needs to be improved to handle the TF.

llvm-svn: 97539

Based on a patch by Micah Villmow for PR6438.

llvm-svn: 97538

emitted after the increment. Make sure the insert position
reflects this. This fixes PR6453.

llvm-svn: 97537

llvm-svn: 97536

Constructors and operators for anonymous aggregates does not names. Do not force empty AT_name attribute in such cases.

llvm-svn: 97533

respectively.

llvm-svn: 97531

llvm-svn: 97529

a loop and is itself the only dependency).

llvm-svn: 97526

new helper function.

llvm-svn: 97525

predecessors before returning.  Otherwise, if multiple predecessor edges need
splitting, we only get one of them per iteration.  This makes a small but
measurable compile time improvement with -enable-full-load-pre.

llvm-svn: 97521

now that it is gone.

llvm-svn: 97516

llvm-svn: 97512

(the most general) the others are dead.

llvm-svn: 97511

Remove the optimize for code size limitation on r67917. Optimize 64-bit imul by constants into leas + shl regardless if optimizing for code size. The size saving from using imulq isn't worth it. Also, the lea and shl instructions may expose further optimization.

llvm-svn: 97507

problems.  rdar://7697850.

llvm-svn: 97500

Don't accidentally produce unspillable intervals for deeply nested loops.

llvm-svn: 97496

llvm-svn: 97488

llvm-svn: 97485

tantalyzing enough that people keep trying to use it.

llvm-svn: 97483

bit should be set to 0 instead of 1.

llvm-svn: 97481

OPC_SwitchOpcode to use a table lookup instead of having to go
through the interpreter for this.

llvm-svn: 97469

an APInt.

llvm-svn: 97467

by Micah Villmow for PR6335.

llvm-svn: 97461

llvm-svn: 97460

llvm-svn: 97459

llvm-svn: 97458

llvm-svn: 97453

llvm-svn: 97450

llvm-svn: 97449

llvm-svn: 97448

llvm-svn: 97446

llvm-svn: 97443

llvm-svn: 97441

ComplexPattern at the root be generated multiple times, once
for each opcode they are part of.  This encourages factoring
because the opcode checks get treated just like everything
else in the matcher.

llvm-svn: 97439

to a scope where every child starts with a CheckOpcode, but
executes more efficiently.  Enhance DAGISelMatcherOpt to 
form it.

This also fixes a bug in CheckOpcode: apparently the SDNodeInfo
objects are not pointer comparable, we have to compare the
enum name.

llvm-svn: 97438

llvm-svn: 97427

llvm-svn: 97426

(which gets #included into the middle of each 
target's DAGISel class) into a .cpp file where it is
only compiled once.

llvm-svn: 97425