1
0
mirror of https://github.com/RPCS3/llvm-mirror.git synced 2024-10-19 19:12:56 +02:00
llvm-mirror/lib/Target/Sparc
James Y Knight 2d0c86cc0b Support expanding partial-word cmpxchg to full-word cmpxchg in AtomicExpandPass.
Many CPUs only have the ability to do a 4-byte cmpxchg (or ll/sc), not 1
or 2-byte. For those, you need to mask and shift the 1 or 2 byte values
appropriately to use the 4-byte instruction.

This change adds support for cmpxchg-based instruction sets (only SPARC,
in LLVM). The support can be extended for LL/SC-based PPC and MIPS in
the future, supplanting the ISel expansions those architectures
currently use.

Tests added for the IR transform and SPARCv9.

Differential Revision: http://reviews.llvm.org/D21029

llvm-svn: 273025
2016-06-17 18:11:48 +00:00
..
AsmParser Don't pass a Reloc::Model to MC. 2016-05-18 11:58:50 +00:00
Disassembler This change adds co-processor condition branching and conditional traps to the Sparc back-end. 2016-03-09 18:20:21 +00:00
InstPrinter This change adds co-processor condition branching and conditional traps to the Sparc back-end. 2016-03-09 18:20:21 +00:00
MCTargetDesc Remove autoconf support 2016-01-26 21:29:08 +00:00
TargetInfo Remove autoconf support 2016-01-26 21:29:08 +00:00
CMakeLists.txt [Sparc][LEON] LEON Erratum fix. Insert NOP after LD or LDF instruction. 2016-05-23 10:56:36 +00:00
DelaySlotFiller.cpp [sparc] Remove some unused (and undefined) declarations. 2016-05-27 10:19:03 +00:00
LeonFeatures.td [Sparc][LEON] LEON Erratum fix. Insert NOP after LD or LDF instruction. 2016-05-23 10:56:36 +00:00
LeonPasses.cpp [sparc] Simplify a slow and verbose way of checking if a string starts with "ld". 2016-05-27 16:45:37 +00:00
LeonPasses.h [sparc] Move LEON passes into llvm namespace. 2016-05-27 10:06:27 +00:00
LLVMBuild.txt
README.txt Initial test commit only 2016-02-26 11:38:24 +00:00
Sparc.h This change adds co-processor condition branching and conditional traps to the Sparc back-end. 2016-03-09 18:20:21 +00:00
Sparc.td [Sparc][LEON] LEON Erratum fix. Insert NOP after LD or LDF instruction. 2016-05-23 10:56:36 +00:00
SparcAsmPrinter.cpp [sparc] Remove some unused (and undefined) declarations. 2016-05-27 10:19:03 +00:00
SparcCallingConv.td [Sparc] Implement i64 load/store support for 32-bit sparc. 2015-08-10 19:11:39 +00:00
SparcFrameLowering.cpp Unify XDEBUG and EXPENSIVE_CHECKS (into the latter), and add an option to the cmake build to enable them. 2016-04-29 15:22:48 +00:00
SparcFrameLowering.h Change eliminateCallFramePseudoInstr() to return an iterator 2016-03-31 18:33:38 +00:00
SparcInstr64Bit.td [SPARC] Fix 8 and 16-bit atomic load and store. 2016-05-23 20:33:00 +00:00
SparcInstrAliases.td This change adds co-processor condition branching and conditional traps to the Sparc back-end. 2016-03-09 18:20:21 +00:00
SparcInstrFormats.td [Sparc] This provides support for itineraries on Sparc. 2016-04-22 08:17:17 +00:00
SparcInstrInfo.cpp Pass DebugLoc and SDLoc by const ref. 2016-06-12 15:39:02 +00:00
SparcInstrInfo.h Pass DebugLoc and SDLoc by const ref. 2016-06-12 15:39:02 +00:00
SparcInstrInfo.td [SPARC] Fix 8 and 16-bit atomic load and store. 2016-05-23 20:33:00 +00:00
SparcInstrVIS.td
SparcISelDAGToDAG.cpp [Sparc] Enable more inline assembly constraints. 2016-05-20 09:03:01 +00:00
SparcISelLowering.cpp Support expanding partial-word cmpxchg to full-word cmpxchg in AtomicExpandPass. 2016-06-17 18:11:48 +00:00
SparcISelLowering.h Pass DebugLoc and SDLoc by const ref. 2016-06-12 15:39:02 +00:00
SparcMachineFunctionInfo.cpp
SparcMachineFunctionInfo.h
SparcMCInstLower.cpp [NFC] Header cleanup 2016-04-18 09:17:29 +00:00
SparcRegisterInfo.cpp Pass DebugLoc and SDLoc by const ref. 2016-06-12 15:39:02 +00:00
SparcRegisterInfo.h [sparc] Remove some unused (and undefined) declarations. 2016-05-27 10:19:03 +00:00
SparcRegisterInfo.td The patch adds missing registers and instructions to complete all the registers supported by the Sparc v8 manual. 2016-02-27 12:49:59 +00:00
SparcSchedule.td [Sparc][LEON] Add UMAC and SMAC instruction support for Sparc LEON subtargets 2016-05-09 11:55:15 +00:00
SparcSubtarget.cpp [Sparc][LEON] LEON Erratum fix. Insert NOP after LD or LDF instruction. 2016-05-23 10:56:36 +00:00
SparcSubtarget.h [Sparc][LEON] LEON Erratum fix. Insert NOP after LD or LDF instruction. 2016-05-23 10:56:36 +00:00
SparcTargetMachine.cpp [Sparc][LEON] LEON Erratum fix. Insert NOP after LD or LDF instruction. 2016-05-23 10:56:36 +00:00
SparcTargetMachine.h [Sparc][LEON] LEON Erratum fix. Insert NOP after LD or LDF instruction. 2016-05-23 10:56:36 +00:00
SparcTargetObjectFile.cpp
SparcTargetObjectFile.h
SparcTargetStreamer.h

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

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