Chris Lattner 5d056952ba optimize single MBB loops better. In particular, produce: ...

LBB1_57:        #bb207.i
        movl 72(%esp), %ecx
        movb (%ecx,%eax), %cl
        movl 80(%esp), %edx
        movb %cl, 1(%edx,%eax)
        incl %eax
        cmpl $143, %eax
        jne LBB1_57     #bb207.i
        jmp LBB1_64     #cond_next255.i

intead of:

LBB1_57:        #bb207.i
        movl 72(%esp), %ecx
        movb (%ecx,%eax), %cl
        movl 80(%esp), %edx
        movb %cl, 1(%edx,%eax)
        incl %eax
        cmpl $143, %eax
        je LBB1_64      #cond_next255.i
        jmp LBB1_57     #bb207.i

This eliminates a branch per iteration of the loop.  This hurted PPC
particularly, because the extra branch meant another dispatch group for each
iteration of the loop.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@31530 91177308-0d34-0410-b5e6-96231b3b80d8

2006-11-08 01:03:21 +00:00

..

SelectionDAG

Add post-indexed load / store transformations.

2006-11-07 09:03:05 +00:00

AsmPrinter.cpp

Add a printSetLabel that takes two id's.

2006-11-01 09:23:08 +00:00

BranchFolding.cpp

optimize single MBB loops better. In particular, produce:

2006-11-08 01:03:21 +00:00

DwarfWriter.cpp

1. Add a pass to fold debug label instructions so a debug info client can detect

2006-11-07 19:33:46 +00:00

ELFWriter.cpp

For PR786:

2006-11-02 20:25:50 +00:00

IntrinsicLowering.cpp

For PR950:

2006-10-20 07:07:24 +00:00

LiveInterval.cpp

For PR786:

2006-11-02 20:25:50 +00:00

LiveIntervalAnalysis.cpp

Enable improved spilling costs by default. This speeds up viterbi on x86

2006-11-07 18:04:58 +00:00

LiveVariables.cpp

For PR786:

2006-11-02 20:25:50 +00:00

LLVMTargetMachine.cpp

1. Add a pass to fold debug label instructions so a debug info client can detect

2006-11-07 19:33:46 +00:00

MachineBasicBlock.cpp

add moveBefore/moveAfter helper methods

2006-10-24 00:02:26 +00:00

MachineDebugInfo.cpp

Accidently reran commit.

2006-11-07 21:58:55 +00:00

MachineFunction.cpp

const'ify jump table stuff

2006-10-28 18:17:09 +00:00

MachineInstr.cpp

be more aggressive about matching identical instructions.

2006-10-25 18:08:14 +00:00

MachinePassRegistry.cpp

Final polish on machine pass registries.

2006-08-02 12:30:23 +00:00

MachOWriter.cpp

Behold, more work on relocations. Things are looking pretty good now.

2006-09-10 23:03:44 +00:00

Makefile

this will work better

2006-11-03 19:15:55 +00:00

Passes.cpp

Work around a bug in gcc 3.3.5, reported by a user

2006-08-03 00:16:56 +00:00

PHIElimination.cpp

"Once more into the breach, dear friends, once more, or fill the wall up

2006-09-28 07:10:24 +00:00

PhysRegTracker.h

…

PrologEpilogInserter.cpp

TargetRegisterClass specifies the desired spill alignment. However, it cannot be honored if stack alignment is smaller.

2006-09-28 18:52:32 +00:00

RegAllocLinearScan.cpp

Use correct value for float HUGH_VAL.

2006-11-07 12:25:45 +00:00

RegAllocLocal.cpp

Fix UnitTests/2005-05-12-Int64ToFP.c with llc-beta. In particular, do not

2006-09-19 18:02:01 +00:00

RegAllocSimple.cpp

Rename

2006-11-01 23:18:32 +00:00

TwoAddressInstructionPass.cpp

Two-address instructions no longer have to be A := A op C. Now any pair of dest / src operands can be tied together.

2006-11-01 23:06:55 +00:00

UnreachableBlockElim.cpp

eliminate RegisterOpt. It does the same thing as RegisterPass.

2006-08-27 22:42:52 +00:00

VirtRegMap.cpp

Fixed some spiller bugs exposed by the recent two-address code changes. Now

2006-11-04 00:21:55 +00:00

VirtRegMap.h

Fix a long-standing wart in the code generator: two-address instruction lowering

2006-09-05 02:12:02 +00:00