code in preparation for code generation.  The main thing it does
is handle the case when eh.exception calls (and, in a future
patch, eh.selector calls) are far away from landing pads.  Right
now in practice you only find eh.exception calls close to landing
pads: either in a landing pad (the common case) or in a landing
pad successor, due to loop passes shifting them about.  However
future exception handling improvements will result in calls far
from landing pads:
(1) Inlining of rewinds.  Consider the following case:
In function @f:
...
  invoke @g to label %normal unwind label %unwinds
...
unwinds:
  %ex = call i8* @llvm.eh.exception()
...

In function @g:
...
  invoke @something to label %continue unwind label %handler
...
handler:
  %ex = call i8* @llvm.eh.exception()
... perform cleanups ...
  "rethrow exception"

Now inline @g into @f.  Currently this is turned into:
In function @f:
...
  invoke @something to label %continue unwind label %handler
...
handler:
  %ex = call i8* @llvm.eh.exception()
... perform cleanups ...
  invoke "rethrow exception" to label %normal unwind label %unwinds
unwinds:
  %ex = call i8* @llvm.eh.exception()
...

However we would like to simplify invoke of "rethrow exception" into
a branch to the %unwinds label.  Then %unwinds is no longer a landing
pad, and the eh.exception call there is then far away from any landing
pads.

(2) Using the unwind instruction for cleanups.
It would be nice to have codegen handle the following case:
  invoke @something to label %continue unwind label %run_cleanups
...
handler:
... perform cleanups ...
  unwind

This requires turning "unwind" into a library call, which
necessarily takes a pointer to the exception as an argument
(this patch also does this unwind lowering).  But that means
you are using eh.exception again far from a landing pad.

(3) Bugpoint simplifications.  When bugpoint is simplifying
exception handling code it often generates eh.exception calls
far from a landing pad, which then causes codegen to assert.
Bugpoint then latches on to this assertion and loses sight
of the original problem.

Note that it is currently rare for this pass to actually do
anything.  And in fact it normally shouldn't do anything at
all given the code coming out of llvm-gcc!  But it does fire
a few times in the testsuite.  As far as I can see this is
almost always due to the LoopStrengthReduce codegen pass
introducing pointless loop preheader blocks which are landing
pads and only contain a branch to another block.  This other
block contains an eh.exception call.  So probably by tweaking
LoopStrengthReduce a bit this can be avoided.

llvm-svn: 72276

compiled with -fobjc-sender-dependent-dispatch.  This is used in AOP, COP, implementing object 
planes, and a few other things.
Patch by David Chisnall.

llvm-svn: 72275

llvm-svn: 72274

improve codegen in some cases.

llvm-svn: 72273

 - We could just warn about -fno-unit-at-a-time, but in practice people using it
   probably aren't going to get what they want out of clang.

Also, use "clang" specified error for unsupported things instead of driver
unsupported error.

llvm-svn: 72272

for ARM.  This fixes rdar://6908807.

llvm-svn: 72269

 - PR4242.

llvm-svn: 72268

and objects of this class are derived from 'NSProxy'.
Under such conditions, which means that every method possible is 
implemented in the class, we should not issue "Method definition not found" 
warnings.

llvm-svn: 72267

assuming that the use of the value is in a block dominated by the
"normal" destination. LangRef.html and other documentation sources
don't explicitly guarantee this, but it seems to be assumed in
other places in LLVM at least.

This fixes an assertion failure on the included testcase, which
is derived from the Ada testsuite.

FixUsesBeforeDefs is a temporary measure which I'm looking to
replace with a more capable solution.

llvm-svn: 72266

There are no unnecessary action calls period (courtesy of the annotation scheme) and too many 'this means'..

llvm-svn: 72263

Emit debug information for globals (which include automatic variables as well because on PIC16 they are emitted as globals by the frontend).

llvm-svn: 72262

Targets like PIC16 generate Static decls for automatic variables, emit the appropriate debug descriptor as well in that case.

llvm-svn: 72261

Handle correctly a very ugly part of the C++ syntax. We cannot disambiguate between a parenthesized type-id and
a paren expression without considering the context past the parentheses.

Behold:
  (T())x;  - type-id
  (T())*x; - type-id
  (T())/x; - expression
  (T());   - expression

llvm-svn: 72260

Factor the compound literal parsing out from ParseParenExpression and into a new ParseCompoundLiteralExpression.
No functionality change.

llvm-svn: 72259

Now it parses the cast expression unless 'stopIfCastExpr' is true.

llvm-svn: 72258

Embed its functionality into it's only user, ParseCXXCasts.
CXXCasts now get the "actual" expression directly, they no longer always receive a ParenExpr. This is better since the
parentheses are always part of the C++ casts syntax.

llvm-svn: 72257

'ParseExprAfterTypeofSizeofAlignof' method.

llvm-svn: 72256

-Makes typeof consistent with sizeof/alignof
-Fixes a bug when '>' is in a typeof expression, inside a template type param:
  A<typeof(x>1)> a;

llvm-svn: 72255

llvm-svn: 72254

llvm-svn: 72253

llvm-svn: 72252

This attempts to fix PR4239.

llvm-svn: 72251

use in expanding SCEVAddExprs with GEPs. The operands of a
SCEVMulExpr need to be multiplied together, not added.

llvm-svn: 72250

indirect call into direct call, thus the verifier would reject something it
previously accepted.

llvm-svn: 72249

This only rejects mismatches between target specific calling convention
and C/LLVM specific calling convention.
There are too many fastcc/C, coldcc/cc42 mismatches in the testsuite, these are
not reject by the verifier.

llvm-svn: 72248

PR4238.

llvm-svn: 72244

 - This commit has some messy stuff in it to extend string lifetimes, but that
   will go away once we switch to using the enum'd Triple interfaces.

llvm-svn: 72243

DragonFly OS type.

llvm-svn: 72242

llvm-svn: 72241

llvm-svn: 72240

llvm-svn: 72239

llvm-svn: 72238

llvm-svn: 72237

llvm-svn: 72236

llvm-svn: 72235

llvm-svn: 72234

can. Also, delay semantic analysis of initialization for
value-dependent as well as type-dependent expressions, since we can't
always properly type-check a value-dependent expression.

llvm-svn: 72233

operand is the last in a pattern.  There is no
reason this should be true (although apparently
it always is right now).

llvm-svn: 72232

llvm-svn: 72231

llvm-svn: 72229