1
0
mirror of https://github.com/RPCS3/llvm-mirror.git synced 2024-11-24 19:52:54 +01:00
llvm-mirror/lib/Target/Sparc
Jim Grosbach 61c5ce1bde Add register-reuse to frame-index register scavenging. When a target uses
a virtual register to eliminate a frame index, it can return that register
and the constant stored there to PEI to track. When scavenging to allocate
for those registers, PEI then tracks the last-used register and value, and
if it is still available and matches the value for the next index, reuses
the existing value rather and removes the re-materialization instructions.
Fancier tracking and adjustment of scavenger allocations to keep more
values live for longer is possible, but not yet implemented and would likely
be better done via a different, less special-purpose, approach to the
problem.

eliminateFrameIndex() is modified so the target implementations can return
the registers they wish to be tracked for reuse.

ARM Thumb1 implements and utilizes the new mechanism. All other targets are
simply modified to adjust for the changed eliminateFrameIndex() prototype.

llvm-svn: 83467
2009-10-07 17:12:56 +00:00
..
AsmPrinter Instead of printing unnecessary basic block labels as labels in 2009-10-06 17:38:38 +00:00
TargetInfo Normalize target registration code. 2009-07-31 18:16:53 +00:00
CMakeLists.txt Normalize makefile comments and sort cmake file lists. 2009-08-31 13:05:24 +00:00
DelaySlotFiller.cpp
FPMover.cpp remove various std::ostream version of printing methods from 2009-08-23 03:41:05 +00:00
Makefile Add TargetInfo libraries for all targets. 2009-07-15 06:35:19 +00:00
README.txt
Sparc.h Add new helpers for registering targets. 2009-07-25 06:49:55 +00:00
Sparc.td
SparcCallingConv.td
SparcInstrFormats.td
SparcInstrInfo.cpp several major improvements to the sparc backend: support for weak linkage 2009-09-15 17:46:24 +00:00
SparcInstrInfo.h several major improvements to the sparc backend: support for weak linkage 2009-09-15 17:46:24 +00:00
SparcInstrInfo.td several major improvements to the sparc backend: support for weak linkage 2009-09-15 17:46:24 +00:00
SparcISelDAGToDAG.cpp Rename getTargetNode to getMachineNode, for consistency with the 2009-09-25 18:54:59 +00:00
SparcISelLowering.cpp Fix PR4926. When target hook EmitInstrWithCustomInserter() insert new basic blocks and update CFG, it should also inform sdisel of the changes so the phi source operands will come from the right basic blocks. 2009-09-19 09:51:03 +00:00
SparcISelLowering.h Enhance EmitInstrWithCustomInserter() so target can specify CFG changes that sdisel will use to properly complete phi nodes. 2009-09-18 21:02:19 +00:00
SparcMachineFunctionInfo.h add missing file 2009-09-15 18:03:13 +00:00
SparcMCAsmInfo.cpp Add source debug information to the Sparc code generator. 2009-09-08 12:47:30 +00:00
SparcMCAsmInfo.h Rename TargetAsmInfo (and its subclasses) to MCAsmInfo. 2009-08-22 20:48:53 +00:00
SparcRegisterInfo.cpp Add register-reuse to frame-index register scavenging. When a target uses 2009-10-07 17:12:56 +00:00
SparcRegisterInfo.h Add register-reuse to frame-index register scavenging. When a target uses 2009-10-07 17:12:56 +00:00
SparcRegisterInfo.td several major improvements to the sparc backend: support for weak linkage 2009-09-15 17:46:24 +00:00
SparcSubtarget.cpp Normalize Subtarget constructors to take a target triple string instead of 2009-08-02 22:11:08 +00:00
SparcSubtarget.h Normalize Subtarget constructors to take a target triple string instead of 2009-08-02 22:11:08 +00:00
SparcTargetMachine.cpp Rename TargetAsmInfo (and its subclasses) to MCAsmInfo. 2009-08-22 20:48:53 +00:00
SparcTargetMachine.h Change TargetAsmInfo to be constructed via TargetRegistry from a Target+Triple 2009-08-12 07:22:17 +00:00

To-do
-----

* Keep the address of the constant pool in a register instead of forming its
  address all of the time.
* We can fold small constant offsets into the %hi/%lo references to constant
  pool addresses as well.
* When in V9 mode, register allocate %icc[0-3].
* Add support for isel'ing UMUL_LOHI instead of marking it as Expand.
* Emit the 'Branch on Integer Register with Prediction' instructions.  It's
  not clear how to write a pattern for this though:

float %t1(int %a, int* %p) {
        %C = seteq int %a, 0
        br bool %C, label %T, label %F
T:
        store int 123, int* %p
        br label %F
F:
        ret float undef
}

codegens to this:

t1:
        save -96, %o6, %o6
1)      subcc %i0, 0, %l0
1)      bne .LBBt1_2    ! F
        nop
.LBBt1_1:       ! T
        or %g0, 123, %l0
        st %l0, [%i1]
.LBBt1_2:       ! F
        restore %g0, %g0, %g0
        retl
        nop

1) should be replaced with a brz in V9 mode.

* Same as above, but emit conditional move on register zero (p192) in V9 
  mode.  Testcase:

int %t1(int %a, int %b) {
        %C = seteq int %a, 0
        %D = select bool %C, int %a, int %b
        ret int %D
}

* Emit MULX/[SU]DIVX instructions in V9 mode instead of fiddling 
  with the Y register, if they are faster.

* Codegen bswap(load)/store(bswap) -> load/store ASI

* Implement frame pointer elimination, e.g. eliminate save/restore for 
  leaf fns.
* Fill delay slots