llvm/lib/Target/Alpha
2006-07-11 17:58:07 +00:00
..
.cvsignore
Alpha.h
Alpha.td getCalleeSaveRegs and getCalleeSaveRegClasses are no long TableGen'd. 2006-05-18 00:12:58 +00:00
AlphaAsmPrinter.cpp 0 offsets for memory operands 2006-07-03 17:57:34 +00:00
AlphaCodeEmitter.cpp Change from using MachineRelocation ctors to using static methods 2006-05-03 20:30:20 +00:00
AlphaInstrFormats.td Let the alpha breakage begin. First Formals and RET. next Calls 2006-06-12 18:09:24 +00:00
AlphaInstrInfo.cpp these are copies too 2006-03-09 18:18:51 +00:00
AlphaInstrInfo.h
AlphaInstrInfo.td Let the alpha breakage begin. First Formals and RET. next Calls 2006-06-12 18:09:24 +00:00
AlphaISelDAGToDAG.cpp Ensure that dump calls that are associated with asserts are removed from 2006-07-11 17:58:07 +00:00
AlphaISelLowering.cpp These are already implemented 2006-07-03 18:00:29 +00:00
AlphaISelLowering.h inline asm, at least for floats 2006-06-21 13:37:27 +00:00
AlphaJITInfo.cpp Fix a purely hypothetical problem (for now): emitWord emits in the host 2006-05-02 19:14:47 +00:00
AlphaJITInfo.h
AlphaRegisterInfo.cpp Let the alpha breakage begin. First Formals and RET. next Calls 2006-06-12 18:09:24 +00:00
AlphaRegisterInfo.h getCalleeSaveRegs and getCalleeSaveRegClasses are no long TableGen'd. 2006-05-18 00:12:58 +00:00
AlphaRegisterInfo.td Add dwarf register numbering to register data. 2006-03-24 21:15:58 +00:00
AlphaRelocations.h Patches to make the LLVM sources more -pedantic clean. Patch provided 2006-05-24 17:04:05 +00:00
AlphaSchedule.td Alpha Scheduling classes 2006-03-09 17:16:45 +00:00
AlphaSubtarget.cpp
AlphaSubtarget.h Alpha Scheduling classes 2006-03-09 17:16:45 +00:00
AlphaTargetMachine.cpp Don't pass target name into TargetData anymore, it is never used or needed. 2006-06-16 18:22:52 +00:00
AlphaTargetMachine.h Refactor a bunch of includes so that TargetMachine.h doesn't have to include 2006-05-12 06:33:49 +00:00
Makefile
README.txt ignore ordered/unordered for now 2006-06-04 00:25:51 +00:00

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
Fix Ordered/Unordered FP stuff


%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
improve bytezap opertunities
ulong %foo(ulong %y) {
entry:
        %tmp = and ulong %y,  65535
        %tmp2 = shr ulong %tmp,  ubyte 3
        ret ulong %tmp2
}


compiles to a 3 instruction sequence without instcombine
        zapnot $16,3,$0
        srl $0,3,$0
        ret $31,($26),1
 
After instcombine you get
ulong %foo(ulong %y) {
entry:
        %tmp = shr ulong %y, ubyte 3            ; <ulong> [#uses=1]
        %tmp2 = and ulong %tmp, 8191            ; <ulong> [#uses=1]
        ret ulong %tmp2
}

which compiles to
        lda $0,8191($31)
        srl $16,3,$1
        and $1,$0,$0
        ret $31,($26),1