eliminate the last of the parallel's!
[oota-llvm.git] / lib / Target /
2010-03-27 Chris Lattnereliminate the last of the parallel's!
2010-03-27 Johnny ChenAdd NVMulSLFrm to represent "3-register multiply with...
2010-03-27 Chris Lattnereliminate almost all the rest of the x86-32 parallels.
2010-03-27 Jim GrosbachThumb2 storeFrom/LoadToStackSlot() need to handle tGPR...
2010-03-26 Johnny ChenRemove the duplicate multiclass N3VSh_QHSD and use...
2010-03-26 Johnny ChenAdd NVExtFrm to represent NEON Vector Extract Instructi...
2010-03-26 Johnny ChenAdd N3RegVShFrm to represent 3-Register Vector Shift...
2010-03-26 Jim Grosbachvldm/vstm can only do up to 16 double-word registers...
2010-03-26 Johnny ChenAdd N3RegFrm to represent "NEON 3 vector register forma...
2010-03-26 Evan ChengDo not sibcall if stack needs to be dynamically aligned.
2010-03-26 Evan ChengAllow trivial sibcall of vararg callee when no argument...
2010-03-26 Johnny ChenAdd N2RegVShLFrm and N2RegVShRFrm formats so that the...
2010-03-25 Jim Grosbachswitch the flag for using NEON for SP floating point...
2010-03-25 Jim Grosbachneed to fix 'make check' tests first. revert for a...
2010-03-25 Jim Grosbachswitch the flag for using NEON for SP floating point...
2010-03-25 Johnny ChenRemoved instruction class NI from ARMInstrFormats.td.
2010-03-25 Jim Grosbachswitch the use-vml[as] instructions flag to a subtarget...
2010-03-25 Johnny ChenAdd NVDupLnFrm and change NVDupLane class to use that...
2010-03-25 Jim GrosbachARM cortex-a8 doesn't do vmla/vmls well. disable them...
2010-03-25 Johnny ChenAdd NVCVTFrm (NEON Convert with fractional bits immedia...
2010-03-25 Daniel DunbarFix -Asserts warning, again.
2010-03-25 Jakob Stoklund OlesenTag SSE2 integer instructions as SSEPackedInt.
2010-03-25 Jakob Stoklund OlesenTeach TableGen to understand X.Y notation in the TSFlag...
2010-03-25 Jakob Stoklund OlesenAdd a late SSEDomainFix pass that twiddles SSE instruct...
2010-03-25 Johnny ChenAdded a new instruction class NVDupLane to be inherited...
2010-03-25 Bob WilsonReapply Kevin's change 94440, now that Chris has fixed...
2010-03-25 Chris Lattnereliminate a bunch more parallels now that scheduling
2010-03-25 Evan ChengDisable folding loads into tail call in 32-bit PIC...
2010-03-24 Bob WilsonSpeculatively revert this to see if it fixes buildbot...
2010-03-24 Kevin EnderbyAdded the Advanced Encryption Standard (AES) Instructions.
2010-03-24 Jim GrosbachMake the use of the vmla and vmls VFP instructions...
2010-03-24 Kevin EnderbyFixed the SS42AI template for the SSE 4.2 instructions...
2010-03-24 Nate BegemanPer chris's request, add some comments.
2010-03-24 Johnny ChenTrivial formating change.
2010-03-24 Nate BegemanBUILD_VECTOR was missing out on some prime opportunitie...
2010-03-24 Johnny ChenReverted r99326 which added NVdVmVCVTFrm, and later...
2010-03-24 Johnny ChenReverted r99376. The disassembler will deal with the...
2010-03-24 Jim Grosbachtweak the arm if conversion heuristic
2010-03-24 Johnny ChenMark VMOVDneon and VMOVQ as having the N2RegFrm form...
2010-03-24 Chris LattnerSwitch INC8r to defining its pattern in terms of X86inc...
2010-03-24 Johnny ChenRenamed NVdVmImmFrm and NVdVmVCVTFrm to the more proper...
2010-03-24 Chris Lattnerswitch SDTBinaryArithWithFlags to be a multiple-result...
2010-03-24 Chris LattnerSwitch SDTUnaryArithWithFlags to being modeled as a...
2010-03-24 Chris Lattnerremove 64-bit or_is_add parallels.
2010-03-24 Chris Lattnerremove useless or_is_add parallel's.
2010-03-24 Chris Lattnerreduce nesting.
2010-03-24 Jim Grosbachtry being more permissive for if-conversion on ARM...
2010-03-23 Jakob Stoklund OlesenRevert "Add a late SSEDomainFix pass that twiddles...
2010-03-23 Chris Lattner[llvm_void_ty] is no longer needed for result types,
2010-03-23 Jakob Stoklund OlesenAdd a late SSEDomainFix pass that twiddles SSE instruct...
2010-03-23 Johnny ChenRenamed NVdImmFrm to N1RegModImmFrm.
2010-03-23 Johnny ChenFix typo in the comment for N3VX class.
2010-03-23 Johnny ChenAdd comment.
2010-03-23 Johnny ChenAdd New NEON Format NVdVmVCVTFrm.
2010-03-23 Johnny ChenAdd New NEON Format NVdVmImmFrm.
2010-03-23 Evan ChengTeach isSafeToClobberEFLAGS to ignore dbg_value's....
2010-03-23 Bob WilsonFix VLDMQ and VSTMQ instructions to use the correct...
2010-03-23 Bob WilsonFix bad indentation, 80-column violations, and trailing...
2010-03-23 Johnny ChenAdd New NEON Format NVdImmFrm.
2010-03-23 Bob WilsonRename some instructions to match the corresponding...
2010-03-23 Bob WilsonChange VST1 instructions for loading Q register values...
2010-03-23 Bob WilsonChange VLD1 instructions for loading Q register values...
2010-03-23 Daniel DunbarMC: Add TargetAsmBackend::MayNeedRelaxation, for checki...
2010-03-23 Daniel DunbarMC: Add TargetAsmBackend::WriteNopData and use to elimi...
2010-03-23 Daniel DunbarMC: Add TargetAsmBackend::RelaxInstruction callback...
2010-03-22 Bob WilsonRename one more NEON instruction that I missed earlier.
2010-03-22 Bob WilsonRegroup some instructions. No functional change.
2010-03-22 Bob WilsonRename some VLD1/VST1 instructions to match the impleme...
2010-03-22 Bob WilsonRemove some redundant instruction classes.
2010-03-22 Bob WilsonRefactor instruction encoding arguments for VLDnLN...
2010-03-22 Jakob Stoklund OlesenCompletely remove Blackfin patterns that thought JustCC...
2010-03-22 Jeffrey YasskinDon't leak a MachineInstruction from Thumb1InstrInfo...
2010-03-20 Daniel DunbarMC/X86: Fix an MCOperand link, when we parsing shrld...
2010-03-20 Daniel DunbarTargetRegistry: Fix create{AsmInfo,MCDisassembler}...
2010-03-20 Bob Wilsonpr6652: Use LDM to restore PC to the return address...
2010-03-20 Bob WilsonRe-commit r98683 ("remove redundant writeback flag...
2010-03-20 Bob WilsonAdd instruction variants for VST2, VST3, and VST4 ...
2010-03-20 Bob WilsonAdd variants of VST2, VST3 and VST4 with address regist...
2010-03-20 Bob WilsonAdd instructions for double-spaced VST3 and VST4 withou...
2010-03-20 Bob WilsonAdd VST1 instructions with address register writeback.
2010-03-20 Bob WilsonAdd instruction variants for VLD2, VLD3, and VLD4 ...
2010-03-20 Bob WilsonTidy some more comments and whitespace.
2010-03-20 Bob WilsonAdd variants of VLD2, VLD3 and VLD4 with address regist...
2010-03-20 Bob WilsonTidy some comments and whitespace for consistency.
2010-03-20 Bob WilsonRename some instructions for consistency and sanity...
2010-03-20 Bob WilsonAdd instructions for double-spaced VLD3 and VLD4 withou...
2010-03-20 Bob WilsonAdd VLD1 instructions with address register writeback.
2010-03-20 Benjamin KramerPIC16: Simplify code by using a std::set<std::string...
2010-03-20 Bob WilsonRevert the rest of 98679.
2010-03-20 Bob WilsonFix a very bad typo. Since the register number was...
2010-03-20 Evan ChengIf call result is in ST0 and it is not being passed...
2010-03-20 Johnny ChenAdd NLdStFrm Format.
2010-03-19 Johnny ChenRevert r98679. The disassembler will be updated to...
2010-03-19 Bob WilsonRevert this change, since it was causing ARM performanc...
2010-03-19 Chris Lattnerremove the patterns that I commented out in r98930...
2010-03-19 Kevin EnderbyFixed the encoding problems of the crc32 instructions...
2010-03-19 Daniel DunbarMC/X86: Rename alternate spellings of {ADD64,CMP64...
2010-03-19 Johnny ChenRenumber LdStExFrm from 28 to 11 and shift the existing...
2010-03-19 Daniel DunbarMC: Add TargetAsmBackend::createObjectWriter.
2010-03-19 Daniel DunbarMCCodeEmitter: Add target independent fixup flag for...
next