There can be multiple defs for a single virtual register when they are defining
sub-registers.

The missing <dead> flag was stopping the inline spiller from eliminating dead
code after rematerialization.

llvm-svn: 128888

llvm-svn: 128887

This allows us to always keep the smaller slot for an instruction which is what
we want when a register has early clobber defines.

Drop the UsingInstrs set and the UsingBlocks map. They are no longer needed.

llvm-svn: 128886

llvm-svn: 128885

space info. We crash with an assert in this case. This change checks that the
address space of the bitcasted pointer is the same as the gep ptr.

llvm-svn: 128884

llvm-svn: 128882

llvm-svn: 128881

of warnings when passing const arguments into the function.

Fixes PR9426

llvm-svn: 128880

make sense to anyone else.

llvm-svn: 128879

I think this moves the code in the desired direction of the new style
recommendations (and style conventional in Clang), but if anyone prefers
the previous style, or has other suggestions just chime in and I'll
follow up.

llvm-svn: 128878

is a single implementation. No functionality change intended.

llvm-svn: 128877

llvm-svn: 128876

llvm-svn: 128875

inlined path for the common case.

Most basic blocks don't contain a call that may throw, so the last split point
os simply the first terminator.

llvm-svn: 128874

It needed to be moved closer to the setjmp statement, because the code directly
after the setjmp needs to know about values that are on the stack. Also, the
'bitcast' of the function context was causing a dead load. This wouldn't be too
horrible, except that at -O0 it wasn't optimized out, and because it wasn't
using the correct base pointer (if there is a VLA), it would try to access a
value from a garbage address.
<rdar://problem/9130540>

llvm-svn: 128873

docs/UsersManual.html: """ize in a paragraph. A few unpaired quotations might confuse HTML-aware text edtors.

llvm-svn: 128871

llvm-svn: 128870

Fixes rdar://9184526

llvm-svn: 128869

llvm-svn: 128868

llvm-svn: 128867

llvm-svn: 128866

llvm-svn: 128865

rdar://problem/9230801 ARM disassembler discrepancy: erroneously accepting SRS

Plus add invalid-RFEorLDMIA-arm.txt test which should have been checked in with
http://llvm.org/viewvc/llvm-project?view=rev&revision=128859.

llvm-svn: 128864

llvm-svn: 128863

Inst{15-12} should be specified as 0b0000.

rdar://problem/9231168 ARM disassembler discrepancy: erroneously accepting MUL

llvm-svn: 128862

llvm-svn: 128860

rdar://problem/9229922 ARM disassembler discrepancy: erroneously accepting RFE

Also LDC/STC instructions are predicated while LDC2/STC2 instructions are not, fixed while
doing regression testings.

llvm-svn: 128859

Fix PR 9626 (duplicated self-init warnings under -Wuninitialized) with numerous CFG and UninitializedValues analysis changes:

1) Change the CFG to include the DeclStmt for conditional variables, instead of using the condition itself as a faux DeclStmt.
2) Update ExprEngine (the static analyzer) to understand (1), so not to regress.
3) Update UninitializedValues.cpp to initialize all tracked variables to Uninitialized at the start of the function/method.
4) Only use the SelfReferenceChecker (SemaDecl.cpp) on global variables, leaving the dataflow analysis to handle other cases.

The combination of (1) and (3) allows the dataflow-based -Wuninitialized to find self-init problems when the initializer
contained control-flow.

llvm-svn: 128858

llvm-svn: 128857

llvm-svn: 128856

llvm-svn: 128855

llvm-svn: 128854

llvm-svn: 128853

between libFrontend and libDriver.

llvm-svn: 128852

The JITMemory manager references LLVM IR constructs directly, while the
runtime Dyld works at a lower level and can handle objects which may not
originate from LLVM IR. Introduce a new layer for the memory manager to
handle the interface between them. For the MCJIT, this layer will be almost
entirely simply a call-through w/ translation between the IR objects and
symbol names.

llvm-svn: 128851

llvm-svn: 128850

Change test/Analysis/idempotent-operations.c to output the .plist file in the test output directory.

llvm-svn: 128849

Introduce Driver::createInvocationFromArgs used to create a CompilerInvocation from command-line args.

llvm-svn: 128848

llvm-svn: 128847

When a virtual register has a single value that is defined as a copy of a
reserved register, permit that copy to be joined. These virtual register are
usually copies of the stack pointer:

  %vreg75<def> = COPY %ESP; GR32:%vreg75
  MOV32mr %vreg75, 1, %noreg, 0, %noreg, %vreg74<kill>
  MOV32mi %vreg75, 1, %noreg, 8, %noreg, 0
  MOV32mi %vreg75<kill>, 1, %noreg, 4, %noreg, 0
  CALLpcrel32 ...

Coalescing these virtual registers early decreases register pressure.
Previously, they were coalesced by RALinScan::attemptTrivialCoalescing after
register allocation was completed.

The lower register pressure causes the mcinst-lowering-cmp0.ll test case to fail
because it depends on linear scan spilling a particular register.

I am deleting 2008-08-05-SpillerBug.ll because it is counting the number of
instructions emitted, and its revision history shows the 'correct' count being
edited many times.

llvm-svn: 128845