DeclarationName can't handle them yet, so right now Parser just errors out on them.

llvm-svn: 90027

llvm-svn: 90026

GetCXXBaseClassOffset to actually pass the test.

llvm-svn: 90025

We always want to create a virtual function pointer entry if the path from the derived return value to the base overridden return value contains a virtual base class.

llvm-svn: 90024

llvm-svn: 90023

to inform GVN about the newly inserted values.  This fixes 
PR5631.

llvm-svn: 90022

llvm-svn: 90021

llvm-svn: 90020

fixes PR5630 and sets the stage for the next phase of goodness (testcase
pending).

llvm-svn: 90019

Add a new PrimaryBaseInfo struct that combines the record decl of a primary base with whether it's virtual or not.

llvm-svn: 90018

llvm-svn: 90017

llvm-svn: 90016

for pointing this out.

llvm-svn: 90015

llvm-svn: 90014

translation of add with immediate.  This allows us
to optimize this function:

void test(int N, double* G) {
  long j;
  G[1] = 1;
    for (j = 1; j < N - 1; j++)
        G[j+1] = G[j] + G[j+1];
}

to only do one load every iteration of the loop.

llvm-svn: 90013

add two simple test cases we now optimize (to one load in the loop each) and one we don't (corresponding to the fixme I added yesterday).

llvm-svn: 90012

llvm-svn: 90011

llvm-svn: 90009

the FIXME I added yesterday to be implemented.

llvm-svn: 90008

array indexes.  The "complex" case of SRoA still handles them, and correctly.

This fixes a weirdness where we'd correctly avoid transforming A[0][42] if
the 42 was too large, but we'd only do it if it was one gep, not two separate
ones.

llvm-svn: 90007

llvm-svn: 90006

the problem only shows for msp430 and pic16 which is why it specifies
them using -march.  But it is wrong to put such tests in CodeGen/Generic,
since not everyone builds these targets.  Put a copy of the test in each
of the target test directories.

llvm-svn: 90005

llvm-svn: 90004

llvm-svn: 90003

llvm-svn: 90002

llvm-svn: 90001

fixes a crash analyzing consumer-lame, which had an "%X = add %X, 1"
in unreachable code.

llvm-svn: 90000

llvm-svn: 89999

llvm-svn: 89998

where it is not available.  It's unclear how to get this inserted 
computation into GVN's scalar availability sets, Owen, help? :)

llvm-svn: 89997

llvm-svn: 89996

llvm-svn: 89995

llvm-svn: 89994

llvm-svn: 89993

translation done by memdep, and reenable gep translation 
again.

llvm-svn: 89992

llvm-svn: 89991

llvm-svn: 89990

llvm-svn: 89989

AggExprEmitter::VisitCastExpr.

llvm-svn: 89988

implement it explicitly or assert that it doesn't make sense for a scalar.
This caught a couple interesting issues: one, CK_BaseToDerivedMemberPointer
casts were getting silently miscompiled, and two, Sema was constructing some
strange implicit casts of type CK_UserDefinedConversion.

The change in SemaExprCXX makes sure the cast kinds are getting set correctly.

llvm-svn: 89987