llvm-6502/lib/Target/SparcV8
2004-10-29 21:57:16 +00:00
..
DelaySlotFiller.cpp
FPMover.cpp
Makefile Change name of target lib to conform to new naming scheme. 2004-10-29 21:57:16 +00:00
README.txt
SparcV8.h
SparcV8.td
SparcV8AsmPrinter.cpp
SparcV8CodeEmitter.cpp * Add baseline structural JIT code, but disable the JIT to allow llvm-gcc builds 2004-10-19 19:49:42 +00:00
SparcV8InstrFormats.td
SparcV8InstrInfo.cpp
SparcV8InstrInfo.h
SparcV8InstrInfo.td
SparcV8ISelSimple.cpp Add support for unreachable and undef 2004-10-17 02:42:42 +00:00
SparcV8JITInfo.h
SparcV8RegisterInfo.cpp Remove dependency on MRegisterInfo::getRegClass 2004-10-29 21:42:27 +00:00
SparcV8RegisterInfo.h
SparcV8RegisterInfo.td
SparcV8TargetMachine.cpp
SparcV8TargetMachine.h

SparcV8 backend skeleton
------------------------

This directory houses a 32-bit SPARC V8 backend employing a expander-based
instruction selector.  It is not yet functionally complete.  Watch
this space for more news coming soon!

Current expected test failures
------------------------------

SingleSource/Benchmarks: 
fldry heapsort objinst Oscar Queens chomp misr pi whetstone

SingleSource/UnitTests:
C++Catch SimpleC++Test 2002-05-02-ArgumentTest 2002-05-02-ManyArguments
2002-05-03-NotTest 2002-05-19-DivTest 2002-10-12-StructureArgs
2002-10-12-StructureArgsSimple 2003-05-07-VarArgs 2003-05-26-Shorts
2003-05-31-CastToBool 2003-05-31-LongShifts 2003-07-09-LoadShorts
2003-07-09-SignedArgs 2003-08-05-CastFPToUint 2003-08-11-VaListArg
StructModifyTest

To-do
-----

* support calling functions with more than 6 args
* support 64-bit integer (long, ulong) arguments to functions
  - use libc procedures instead of open-coding for:
    __div64 __mul64 __rem64 __udiv64 __umul64 __urem64
* support setcc on longs
* support basic binary operations on longs
* support casting <=32-bit integers, bools to long
* support casting 64-bit integers to FP types

$Date$