llvm-svn: 123406

llvm-svn: 123405

the symbolic immediate names used for these instructions, fixing their pretty-printers, and
adding proper encoding information for them.

With this, we can properly pretty-print and encode assembly like:
	mrc p15, #0, r3, c13, c0, #3

Fixes <rdar://problem/8857858>.

llvm-svn: 123404

Relax an assertion. On archs like ARM, an immediate field may be scattered. So it's possible for some bits of every 8 bits to be encoded already, and the rest still needs to be fixed up.

llvm-svn: 123403

for efficiancy (still part of //rdar://8761767).
Per John's comment.

llvm-svn: 123401

llvm-svn: 123400

llvm-svn: 123399

variables.

llvm-svn: 123398

llvm-svn: 123397

llvm-svn: 123396

llvm-svn: 123395

dead stores within nested assignments.  I have
never seen an actual bug found by this specific
warning, and it can lead to many false positives.

llvm-svn: 123394

directional local labels like 1f and 2b.

llvm-svn: 123393

llvm-svn: 123392

assigns. // rdar://8761767 

llvm-svn: 123391

Patch by Bob Wilson.

llvm-svn: 123390

llvm-svn: 123389

Patch by arrowdodger!

llvm-svn: 123388

set up the source operands. The original instr has an immediate operand that
should be replaced with the frame reg operand rather than just adding the
reg operand. Previously, the instruction ended up with too many operands
causing an assert() when adding the default predicate. rdar://8825456

llvm-svn: 123387

process, perform a number of refactorings:

- Move MiscNameMangler member functions to MangleContext
- Remove GlobalDecl dependency from MangleContext
- Make MangleContext abstract and move Itanium/Microsoft functionality
  to their own classes/files
- Implement ASTContext::createMangleContext and have CodeGen use it

No (intended) functionality change.

llvm-svn: 123386

when we're actually matching a template template argument to a
template template parameter. Otherwise, use strict matching.

Fixes <rdar://problem/8859985> clang++: variadics and out-of-line definitions.

llvm-svn: 123385

It will still return an iterator that points to the first terminator or end(),
but there may be DBG_VALUE instructions following the first terminator.

llvm-svn: 123384

llvm-svn: 123383

llvm-svn: 123382

This is a minor extension of SROA to handle a special case that is
important for some ARM NEON operations.  Some of the NEON intrinsics
return multiple values, which are handled as struct types containing
multiple elements of the same vector type.  The corresponding return
types declared in the arm_neon.h header have equivalent arrays.  We
need SROA to recognize that it can split up those arrays and structs
into separate vectors, even though they are not always accessed with
the same type.  SROA already handles loads and stores of an entire
alloca by using insertvalue/extractvalue to access the individual
pieces, and that code works the same regardless of whether the type
is a struct or an array.  So, all that needs to be done is to check
for compatible arrays and homogeneous structs.

llvm-svn: 123381

SROA only split up structs and arrays one level at a time, so padding can
only cause trouble if it is located in between the struct or array elements.

llvm-svn: 123380

llvm-svn: 123379

llvm-svn: 123378

llvm-svn: 123377

Patch by arrowdodger!

llvm-svn: 123376

getting the base region. This makes the RemoveDeadBindings() correct.

llvm-svn: 123375

since the bindings are purged after they are set up. Need to investigate
RemoveDeadBindings algorithm.

llvm-svn: 123374

the comment I added): an extern weak global may have a null address.

llvm-svn: 123373

is "X != 0 -> X" when X is a boolean.  This occurs a lot because of the way
llvm-gcc converts gcc's conditional expressions.  Add this, and a few other
similar transforms for completeness.

llvm-svn: 123372

by LLDB.  Instead of being materialized into the input structure
passed to the expression, variables are left in place and pointers
to them are materialzied into the structure.  Variables not resident
in memory (notably, registers) get temporary memory regions allocated
for them.

Persistent variables are the most complex part of this, because they
are made in various ways and there are different expectations about
their lifetime.  Persistent variables now have flags indicating their
status and what the expectations for longevity are.  They can be
marked as residing in target memory permanently -- this is the
default for result variables from expressions entered on the command
line and for explicitly declared persistent variables (but more on
that below).  Other result variables have their memory freed.

Some major improvements resulting from this include being able to
properly take the address of variables, better and cleaner support
for functions that return references, and cleaner C++ support in
general.  One problem that remains is the problem of explicitly
declared persistent variables; I have not yet implemented the code
that makes references to them into indirect references, so currently
materialization and dematerialization of these variables is broken.

llvm-svn: 123371

llvm-svn: 123370

in the right direction. It eliminated some hacks and will unblock codegen
work. But it's far from being done. It doesn't reject illegal expressions,
e.g. (FOO - :lower16:BAR). It also doesn't work in Thumb2 mode at all.

llvm-svn: 123369

a struct value to a symbolic index into array.
RegionStore can't actually reason about this,
so we were getting bogus warnings about loading
uninitialized values from the array.  The solution
is invalidate the entire array when we cannot
represent the binding explicitly.

Fixes <rdar://problem/8848957>

llvm-svn: 123368

16 bytes for PR8969. Update all testcases accordingly.

llvm-svn: 123367

llvm-svn: 123364