hanchenye-llvm-project/llvm/lib/Target/Sparc
Venkatraman Govindaraju f80d72f149 Sparc: Add support for indirect branch and blockaddress in Sparc backend.
llvm-svn: 183094
2013-06-03 05:58:33 +00:00
..
MCTargetDesc Remove unused argument. 2013-05-10 18:16:59 +00:00
TargetInfo Move all of the header files which are involved in modelling the LLVM IR 2013-01-02 11:36:10 +00:00
CMakeLists.txt llvm/lib: [CMake] Add explicit dependency to intrinsics_gen. 2012-06-24 13:32:01 +00:00
DelaySlotFiller.cpp Sparc: Combine add/or/sethi instruction with restore if possible. 2013-06-02 21:48:17 +00:00
FPMover.cpp Use the new script to sort the includes of every file under lib. 2012-12-03 16:50:05 +00:00
LLVMBuild.txt
Makefile
README.txt Don't use %g0 to materialize 0 directly. 2013-05-19 21:47:13 +00:00
Sparc.h
Sparc.td
SparcAsmPrinter.cpp Sparc: Add support for indirect branch and blockaddress in Sparc backend. 2013-06-03 05:58:33 +00:00
SparcCallingConv.td Complete formal arguments for the SPARC v9 64-bit ABI. 2013-04-06 18:32:12 +00:00
SparcFrameLowering.cpp Sparc: Perform leaf procedure optimization by default 2013-06-02 02:24:27 +00:00
SparcFrameLowering.h [Sparc] Add support for leaf functions in sparc backend. 2013-05-29 04:46:31 +00:00
SparcISelDAGToDAG.cpp Track IR ordering of SelectionDAG nodes 2/4. 2013-05-25 02:42:55 +00:00
SparcISelLowering.cpp Sparc: Add support for indirect branch and blockaddress in Sparc backend. 2013-06-03 05:58:33 +00:00
SparcISelLowering.h Sparc: Add support for indirect branch and blockaddress in Sparc backend. 2013-06-03 05:58:33 +00:00
SparcInstr64Bit.td Sparc: When storing 0, use %g0 directly in the store instruction instead of 2013-06-03 00:21:54 +00:00
SparcInstrFormats.td Use i32 for all SPARC shift amounts, even in 64-bit mode. 2013-04-14 05:48:50 +00:00
SparcInstrInfo.cpp Implement spill and fill of I64Regs. 2013-05-20 00:53:25 +00:00
SparcInstrInfo.h Reorder includes in Target backends to following coding standards. Remove some superfluous forward declarations. 2012-03-17 18:46:09 +00:00
SparcInstrInfo.td Sparc: Add support for indirect branch and blockaddress in Sparc backend. 2013-06-03 05:58:33 +00:00
SparcMachineFunctionInfo.cpp
SparcMachineFunctionInfo.h [Sparc] Add support for leaf functions in sparc backend. 2013-05-29 04:46:31 +00:00
SparcRegisterInfo.cpp [Sparc] Generate correct code for leaf functions with stack objects 2013-06-01 04:51:18 +00:00
SparcRegisterInfo.h Add an I64Regs register class for 64-bit registers. 2013-04-02 04:08:54 +00:00
SparcRegisterInfo.td Make SubRegIndex size mandatory, following r183020. 2013-05-31 23:45:26 +00:00
SparcSelectionDAGInfo.cpp
SparcSelectionDAGInfo.h
SparcSubtarget.cpp [Sparc] Generate correct code for leaf functions with stack objects 2013-06-01 04:51:18 +00:00
SparcSubtarget.h [Sparc] Generate correct code for leaf functions with stack objects 2013-06-01 04:51:18 +00:00
SparcTargetMachine.cpp Remove the MachineMove class. 2013-05-13 01:16:13 +00:00
SparcTargetMachine.h Switch TargetTransformInfo from an immutable analysis pass that requires 2013-01-07 01:37:14 +00:00

README.txt

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

* Implement JIT support

* Use %g0 directly to materialize 0. No instruction is required.