llvm-svn: 157060

It is no longer necessary to separate VirtCopies, PhysCopies, and
ImpDefCopies. Implicitly defined copies are extremely rare after we
added the ProcessImplicitDefs pass, and physical register copies are not
joined any longer.

llvm-svn: 157059

Patch by Jack Carter.

llvm-svn: 157057

This has been disabled for a while, and it is not a feature we want to
support. Copies between physical and virtual registers are eliminated by
good hinting support in the register allocator. Joining virtual and
physical registers is really a form of register allocation, and the
coalescer is not properly equipped to do that. In particular, it cannot
backtrack coalescing decisions, and sometimes that would cause it to
create programs that were impossible to register allocate, by exhausting
a small register class.

It was also very difficult to keep track of the live ranges of aliasing
registers when extending the live range of a physreg. By disabling
physreg joining, we can let fixed physreg live ranges remain constant
throughout the register allocator super-pass.

One type of physreg joining remains: A virtual register that has a
single value which is a copy of a reserved register can be merged into
the reserved physreg. This always lowers register pressure, and since we
don't compute live ranges for reserved registers, there are no problems
with aliases.

llvm-svn: 157055

SelectionDAGBuilder::Clusterify : main functinality was replaced with CRSBuilder::optimize, so big part of Clusterify's code was reduced.

llvm-svn: 157046

llvm-svn: 157044

llvm-svn: 157043

non-profitable commute using outdated info. The test case would still fail
because of poor pre-RA schedule. That will be fixed by MI scheduler.

rdar://11472010

llvm-svn: 157038

llvm-svn: 157033

this fixes the missed optimization I was seeing in the CorrelatedValuePropagation pass

llvm-svn: 157032

llvm-svn: 157024

llvm-svn: 157020

the 0b10 mask encoding bits.  Make MSR APSR writes without a _<bits> qualifier
an alias for MSR APSR_nzcvq even though ARM as deprecated it use.  Also add
support for suffixes (_nzcvq, _g, _nzcvqg) for APSR versions.  Some FIXMEs in
the code for better error checking when versions shouldn't be used.
rdar://11457025

llvm-svn: 157019

llvm-svn: 157011

llvm-svn: 157007

Introduce the basic strategy for register pressure scheduling.

1) Respect target limits at all times.

2) Indentify critical register classes (pressure sets).
   Track pressure within the scheduled region.
   Avoid increasing scheduled pressure for critical registers.

3) Avoid exceeding the max pressure of the region prior to scheduling.

Added logic for picking between the top and bottom ready Q's based on
regpressure heuristics.

Status: functional but needs to be asjusted to achieve good results.
llvm-svn: 157006

llvm-svn: 157005

llvm-svn: 157004

llvm-svn: 157003

llvm-svn: 157002

RegisterCoalescer set <undef> flags on all operands of copy instructions
that are scheduled to be removed. This is so they won't affect
shrinkToUses() by introducing false register reads.

Make sure those <undef> flags are never cleared, or shrinkToUses() could
cause live intervals to end at instructions about to be deleted.

This would be a lot simpler if RegisterCoalescer could just erase joined
copies immediately instead of keeping all the to-be-deleted instructions
around.

This fixes PR12862. Unfortunately, bugpoint can't create a sane test
case for this. Like many other coalescer problems, this failure depends
of a very fragile series of events.

<rdar://problem/11474428>

llvm-svn: 157001

Make sure useless (def-only) intervals also get verified.

llvm-svn: 157000

bugpoint-ing, it may turn into something else.

llvm-svn: 156998

separate side table, using the handy SequenceToOffsetTable class.  This encodes all
these weird things into another 256 bytes, allowing all intrinsics to be encoded this way.

llvm-svn: 156995

Patch by Meador Inge.

llvm-svn: 156989

llvm-svn: 156986

SelectionDAGBuilder: CaseBlock, CaseRanges and CaseCmp changed representation of Low and High from signed to unsigned. Since unsigned ints usually simpler, faster and allows to reduce some extra signed bit checks needed before <,>,<=,>= comparisons.

llvm-svn: 156985

for future expansion, no functionality change yet though.

llvm-svn: 156979

are only rejected because they can't be encoded into a 32-bit unit, not because
they contain an unencodable feature.

llvm-svn: 156978

intrinsics that use passed-in arguments.

llvm-svn: 156977

llc to recognize MIPS16 as a MIPS ASE extension. -mips16 will mean the
mips16 ASE for mips32 by default.

As part of fixing of adding this we discovered some small changes that
need to be made to MipsInstrInfo::storeRegToStackSLot and
MipsInstrInfo::loadRegFromStackSlot. We were using some "==" equality tests
where in fact we should have been using Mips::<regclas>.hasSubClassEQ instead,
per suggestion of Jakob Stoklund Olesen.

Patch by Reed Kotler.

llvm-svn: 156958

When widening an existing <def,reads-undef> operand to a super-register,
it may be necessary to clear the <undef> flag because the wider register
is now read-modify-write through the instruction.

Conversely, it may be necessary to add an <undef> flag when the
coalescer turns a full-register def into a sub-register def, but the
larger register wasn't live before the instruction.

This happens in test/CodeGen/ARM/coalesce-subregs.ll, but the test
is too small for the <undef> flags to affect the generated code.

llvm-svn: 156951

Added LLIMCJITMemoryManager to the lli. This manager will be used for MCJIT instead of DefaultJIMMemoryManager.
It's more flexible for MCJIT tasks, in addition it's provides a invalidation instruction cache for code sections which will be used before JIT code will be executed.

llvm-svn: 156933

llvm-svn: 156917

llvm-svn: 156909

generated code (for Intrinsic::getType) into a table.  This handles common cases right now,
but I plan to extend it to handle all cases and merge in type verification logic as well
in follow-on patches.

llvm-svn: 156905

llvm-svn: 156896

It is now possible to coalesce weird skewed sub-register copies by
picking a super-register class larger than both original registers. The
included test case produces code like this:

  vld2.32 {d16, d17, d18, d19}, [r0]!
  vst2.32 {d18, d19, d20, d21}, [r0]

We still perform interference checking as if it were a normal full copy
join, so this is still quite conservative. In particular, the f1 and f2
functions in the included test case still have remaining copies because
of false interference.

llvm-svn: 156878

It is possible to coalesce two overlapping registers to a common
super-register that it larger than both of the original registers.

The important difference is that it may be necessary to rewrite DstReg
operands as well as SrcReg operands because the sub-register index has
changed.

This behavior is still disabled by CoalescerPair.

llvm-svn: 156869

This can happen when widening a virtual register to a super-register
class.

llvm-svn: 156867