This is not yet completely enabled.

llvm-svn: 123274

llvm-svn: 123270

llvm-svn: 123253

llvm-svn: 123248

llvm-svn: 123247

llvm-svn: 123246

llvm-svn: 123243

llvm-svn: 123242

McARM: Sketch some logic for determining when to add carry set and predication code operands based on the "canonical mnemonic".

llvm-svn: 123239

carry setting flag from the mnemonic.

Note that this currently involves me disabling a number of working cases in
arm_instructions.s, this is a hopefully short term evil which will be rapidly
fixed (and greatly surpassed), assuming my current approach flies.

llvm-svn: 123238

"this" pointer for any subclass of User, you could static_cast it to
User* and then reinterpret_cast that to Use* to get the end of the
operand list. This isn't a safe assumption in general, because the
static_cast might adjust the "this" pointer. Fixed by having these
OperandTraits classes take an extra template parameter, which is the
subclass of User. This is groundwork for PR889.

llvm-svn: 123235

Factor the actual simplification out of SimplifyIndirectBrOnSelect and into a new helper function so it can be reused in e.g. an upcoming SimplifySwitchOnSelect.
No functional change.

llvm-svn: 123234

llvm-svn: 123229

Filling no-ops is done just before emitting of assembly,
when the instruction stream is final. No-ops are inserted
to align the instructions so the dual-issue of the pipeline
is utilized. This speeds up generated code with a minimum of 
1% on a select set of algorithms.

This pass may be redundant if the instruction scheduler and 
all subsequent passes that modify the instruction stream 
(prolog+epilog inserter, register scavenger, are there others?)
are made aware of the instruction alignments.

llvm-svn: 123226

to get a testcase.

llvm-svn: 123225

actually reached in the testcase in PR8954, but it's safe and good
practice.

llvm-svn: 123224

is floating around in the ether.

llvm-svn: 123223

phi nodes.  It is called from MergeBlockIntoPredecessor which is 
called from GVN, which claims to preserve these.

I'm skeptical that this is the actual problem behind PR8954, but
this is a stab in the right direction.

llvm-svn: 123222

llvm-svn: 123221

neccesarily an uncond branch to the header.  This fixes 
PR8955 (the assertion tripping).

llvm-svn: 123219

llvm-svn: 123218

Revert r123207: "Turn on memdep's verifyRemoved() in an attempt to smoke out the cause of our gcc bootstrap miscompare."

It didn't.

llvm-svn: 123215

llvm-svn: 123209

Turn on memdep's verifyRemoved() in an attempt to smoke out the cause of our gcc bootstrap miscompare.

llvm-svn: 123207

point values to their integer representation through the SSE intrinsic
calls. This is the last part of a README.txt entry for which I have real
world examples.

llvm-svn: 123206

Fix a random missed optimization by making InstCombine more aggressive when determining which bits are demanded by
a comparison against a constant.

llvm-svn: 123203

llvm-svn: 123202

restore the stack pointer from the frame pointer on thumbv6.

Fixes rdar://8819685

llvm-svn: 123196

llvm-svn: 123193

There's an inherent tension in DAGCombine between assuming
that things will be put in canonical form, and the Depth
mechanism that disables transformations when recursion gets
too deep.  It would not surprise me if there's a lot of little
bugs like this one waiting to be discovered.  The mechanism
seems fragile and I'd suggest looking at it from a design viewpoint.

llvm-svn: 123191

operation in some cases.

llvm-svn: 123190

llvm-svn: 123189

instructions but are restricted pseudo forms.

llvm-svn: 123177

instructions but are restricted pseudo forms.

llvm-svn: 123176

llvm-svn: 123175

llvm-svn: 123172

llvm-svn: 123171

Rename TargetFrameInfo into TargetFrameLowering. Also, put couple of FIXMEs and fixes here and there.

llvm-svn: 123170

llvm-svn: 123169

IDs when available rather than using a mixture of IDs and textual name
comparisons.

llvm-svn: 123165