llvm-svn: 96178

with multiplication by constants distributed through, occasionally
those subexpressions can include both x and -x. For now, if this
condition is discovered within LSR, just prune such cases away,
as they won't be profitable. This fixes a "zero allocated in a
base register" assertion failure.

llvm-svn: 96177

1. restore these across direct/indirect calls.
2. restore pagesel for any macros with gotos.

llvm-svn: 96175

llvm-svn: 96174

llvm-svn: 96172

Otherwise AT&T asm printer is used with non-compatible MCAsmInfo and
there is no way to override this behaviour.

llvm-svn: 96165

as suggested by Bob Wilson.

llvm-svn: 96153

the no-TLI case. But it should still default to declining the
transformation.

llvm-svn: 96152

current insertion point, advance the current insertion point.
This avoids a use-before-def situation in a testcase extracted
from clang which is difficult to reduce to a reasonable-sized
regression test.

llvm-svn: 96151

to override anything from the parent class.

llvm-svn: 96150

because profitability can't be sufficiently approximated.

llvm-svn: 96148

from opt.

llvm-svn: 96135

odd offsets since the bitcasted pointer size and the offset pointer
size are going to be different types for the GEP vs base object.

llvm-svn: 96134

encoding them into nothing.

llvm-svn: 96110

llvm-svn: 96109

--
ddunbar@ozzy:tmp$ clang -m32 -integrated-as hello.c && ./a.out
hello world!
--

llvm-svn: 96096

llvm-svn: 96095

llvm-svn: 96094

llvm-svn: 96093

MCAssembler: Switch MCFillFragment to only taking constant values. Symbolic expressions can always be emitted as data + fixups.

llvm-svn: 96092

llvm-svn: 96091

llvm-svn: 96090

llvm-svn: 96089

llvm-svn: 96088

and add a doxygen comment.

Cache the phi entry to avoid doing tons of 
PHINode::getBasicBlockIndex calls in the common case.

On my insane testcase from re2c, this speeds up CGP from
617.4s to 7.9s (78x).

llvm-svn: 96083

llvm-svn: 96081

to a PHI, avoid it in the common case where the BB occurs
in the same index for multiple phis.  This speeds up CGP on
an insane testcase from 8.35 to 3.58s.

llvm-svn: 96080

using pred_begin/end.  It is much faster.

llvm-svn: 96079

instead of with pred_begin/end.

llvm-svn: 96078

llvm-svn: 96076

llvm-svn: 96075

tables.

llvm-svn: 96073

When coalescing with a physreg, remember to add imp-def and imp-kill when
dealing with sub-registers.

Also fix a related bug in VirtRegRewriter where substitutePhysReg may
reallocate the operand list on an instruction and invalidate the reg_iterator.
This can happen when a register is mentioned twice on the same instruction.

llvm-svn: 96072

deterministically sorted.

llvm-svn: 96071

created. This ensures it's updated at all time. It means targets which perform
dynamic stack alignment would know whether it is required and whether frame
pointer register cannot be made available register allocation.
This is a fix for rdar://7625239. Sorry, I can't create a reasonably sized test
case.

llvm-svn: 96069

Functions explicitly marked inline will get an inlining threshold slightly
more aggressive than the default for -O3. This means than -O3 builds are
mostly unaffected while -Os builds will be a bit bigger and faster.

The difference depends entirely on how many 'inline's are sprinkled on the
source.

In the CINT2006 suite, only these tests are significantly affected under -Os:

               Size   Time
471.omnetpp   +1.63% -1.85%
473.astar     +4.01% -6.02%
483.xalancbmk +4.60%  0.00%

Note that 483.xalancbmk runs too quickly to give useful timing results.

llvm-svn: 96066

llvm-svn: 96065

MC/AsmParser: Attempt to constant fold expressions up-front. This ensures we avoid fixups for obvious cases like '-(16)'.

llvm-svn: 96064

llvm-svn: 96063

We still have the templated X86 JIT emitter, *and* the
almost-copy in X86InstrInfo for getting instruction sizes.

llvm-svn: 96059