llvm-svn: 26096

aggrssive inlining. This reduces Select_store frame size from 24k to 10k.

llvm-svn: 26095

problem where it inline the map insertion call too aggressively. Before this
change it was producing a frame size of 24k for Select_store(), now it's down
to 10k (by calling this method rather than calling the map insertion operator).

llvm-svn: 26094

llvm-svn: 26093

llvm-svn: 26091

llvm-svn: 26090

uses of loop values outside the loop.  We need loop-closed SSA form to do
this right, or to use SSA rewriting if we really care.

llvm-svn: 26089

llvm-svn: 26088

1. Teach it new tricks: in particular how to propagate through signed shr and sexts.
2. Teach it to return a bitset of known-1 and known-0 bits, instead of just zero.
3. Teach instcombine (AND X, C) to fold when we know all C bits of X.

This implements Regression/Transforms/InstCombine/bittest.ll, and allows
future things to be simplified.

llvm-svn: 26087

llvm-svn: 26086

llvm-svn: 26085

llvm-svn: 26084

Move all getTargetNode() out of SelectionDAG.h into SelectionDAG.cpp. This
prevents them from being inlined.
Change getTargetNode() so they return SDNode * instead of SDOperand to prevent
copying. It should also help compilation speed.

llvm-svn: 26083

llvm-svn: 26082

llvm-svn: 26080

llvm-svn: 26079

value/alignment pair for each constant, keep a value/offset pair.

llvm-svn: 26078

offset of each entry from the start of the constant pool.

llvm-svn: 26077

llvm-svn: 26076

constant pool entries.

llvm-svn: 26075

llvm-svn: 26074

llvm-svn: 26073

llvm-svn: 26072

llvm-svn: 26071

llvm-svn: 26070

llvm-svn: 26069

llvm-svn: 26068

SDOperand Select(SDOperand N);
to
void Select(SDOperand &Result, SDOperand N);

llvm-svn: 26067

llvm-svn: 26066

llvm-svn: 26065

Move the methods out of line.
Remove unused Debug.h stuff.
Teach getNumBytesForInstruction to know the size of an inline asm.

llvm-svn: 26064

llvm-svn: 26063

llvm-svn: 26062

Patch by Marco Matthies, thanks!

llvm-svn: 26061

llvm-svn: 26060

compilation, C++ compilation, and linker options.
e.g. This is the options I use for testing on my x86 iMac:
nice ./NightlyTest.pl -release -cflags "-Os -DNDEBUG -fomit-frame-pointer" -cxxflags "-Os -DNDEBUG -finline-functions -felide-constructors -fomit-frame-pointer"

llvm-svn: 26057

optimization where we reduce the number of bits in AND masks when possible.

llvm-svn: 26056

llvm-svn: 26055

llvm-svn: 26054

llvm-svn: 26053