Correct some load / store instruction itinerary mistakes:
1. Cortex-A8 load / store multiplies can only issue on ALU0. 2. Eliminate A8_Issue, A8_LSPipe will correctly limit the load / store issues. 3. Correctly model all vld1 and vld2 variants. llvm-svn: 116134
Showing
- llvm/lib/Target/ARM/ARMInstrNEON.td 42 additions, 42 deletionsllvm/lib/Target/ARM/ARMInstrNEON.td
- llvm/lib/Target/ARM/ARMSchedule.td 12 additions, 0 deletionsllvm/lib/Target/ARM/ARMSchedule.td
- llvm/lib/Target/ARM/ARMScheduleA8.td 174 additions, 219 deletionsllvm/lib/Target/ARM/ARMScheduleA8.td
- llvm/lib/Target/ARM/ARMScheduleA9.td 175 additions, 45 deletionsllvm/lib/Target/ARM/ARMScheduleA9.td
- llvm/test/CodeGen/ARM/reg_sequence.ll 1 addition, 1 deletionllvm/test/CodeGen/ARM/reg_sequence.ll
Loading
Please register or sign in to comment