Fold c2 in (x << c1) | c2 where (c2 < c1)
e.g.
int test(int x) {
  return (x << 3) + 7;
}

This can be codegen'd as:
leal 7(,%eax,8), %eax

llvm-svn: 28550

llvm-svn: 28549

llvm-svn: 28548

tests. The output in a few cases still doesn't compile, however.

llvm-svn: 28547

llvm-svn: 28546

uses DSA to make find targets of calls.  It provides a very convinient
interface to DSA results to do things with indirect calls, such as
write a devirtualizer (which I have and may commit one of these days).

llvm-svn: 28545

llvm-svn: 28544

llvm-svn: 28543

llvm-svn: 28542

1. Get rid of old AsmWriter cruft that's not needed.
2. Implement several instructions. Enough to get by globalvars.ll and
   alignment.ll in the Feature test suite.
3. Handle constants properly (don't repeat definitions).
4. Make the output compatible with llvm-dis for diff purposes.

llvm-svn: 28541

llvm-svn: 28540

llvm-svn: 28539

llvm-svn: 28538

llvm-svn: 28537

1. Return the module from the MakeModule function so it can be verified.
2. Make sure types get generated with their names

llvm-svn: 28536

llvm-svn: 28535

llvm-svn: 28534

llvm-svn: 28533

llvm-svn: 28532

llvm-svn: 28531

llvm-svn: 28530

is now theoretically feature-complete.  It has not, however, been thoroughly
test, and is still considered experimental.

llvm-svn: 28529

This is a safekeeping commit. The program is not finished. It currently
handles modules, types, global variables and function declarations. Blocks
and instructions remain to be done.

llvm-svn: 28528

other calculations on each individually, rather than trying to delay it and do
them all at the end.

llvm-svn: 28527

"libdir" for some reason. Changing to llvmlibsdir instead fixes it.

llvm-svn: 28526

committed). This infrastructure is only activated when RUNLLVM2CPP=1 is
specified on the make command line. Currently it is only supported in the
Feature test suite.

llvm-svn: 28525

the iterated Dominance Frontier of the loop-closure Phi's.  This is the
second phase of the LCSSA pass.  The third phase (coming soon) will be to
update all uses of loop variables to use the loop-closure Phi's instead.

llvm-svn: 28524

ldecod, lencod, and SPASS.

llvm-svn: 28523

llvm-svn: 28522

makes it so that it constant folds instructions on the fly.  This is good
for several reasons:

0. Many instructions are constant foldable after inlining, particularly if
   inlining a call with constant arguments.
1. Without this, the inliner has to allocate memory for all of the instructions
   that can be constant folded, then a subsequent pass has to delete them.  This
   gets the job done without this extra work.
2. This makes the inliner *pass* a bit more aggressive: in particular, it
   partially solves a phase order issue where the inliner would inline lots
   of code that folds away to nothing, but think that the resultant function
   is big because of this code that will be gone.  Now the code never exists.

This is the first part of a 2-step process.  The second part will be smart
enough to see when this implicit constant folding propagates a constant into
a branch or switch instruction, making CFG edges dead.

This implements Transforms/Inline/inline_constprop.ll

llvm-svn: 28521

llvm-svn: 28520

llvm-svn: 28519

llvm-svn: 28518

Refactor some code to expose an interface to constant fold and instruction given it's opcode, typeand operands.

llvm-svn: 28517

and operands.

llvm-svn: 28516

fly if it can.

llvm-svn: 28515

SimplifySelectOps would eliminate a Select, delete it, then return true.

The clients would see that it did something and return null.

The top level would see a null return, and decide that nothing happened,
proceeding to process the node in other ways: boom.

The fix is simple: clients of SimplifySelectOps should return the select
node itself.

In order to catch really obnoxious boogs like this in the future, add an
assert that nodes are not deleted.  We do this by checking for a sentry node
type that the SDNode dtor sets when a node is destroyed.

llvm-svn: 28514

use-after-deleted errors.

llvm-svn: 28513

llvm-svn: 28512

instead of MVT::i1. Either is fine except MVT::i32 is probably a legal type
for most (if not all) platforms while MVT::i1 is not.

llvm-svn: 28511