2008-02-12 |
Nate Begeman | SSE4.1 64b integer insert/extract pattern support |
tree | commitdiff |
2008-02-12 |
Evan Cheng | Only using x86-64 rip relative addressing in non-staic... |
tree | commitdiff |
2008-02-12 |
Evan Cheng | Update comment. |
tree | commitdiff |
2008-02-12 |
Evan Cheng | Unbreak various insert_vector_elt and extract_vector_el... |
tree | commitdiff |
2008-02-11 |
Nate Begeman | Enable SSE4 codegen and pattern matching. |
tree | commitdiff |
2008-02-10 |
Nate Begeman | xmm0 variable blends |
tree | commitdiff |
2008-02-10 |
Dan Gohman | Rename MRegisterInfo to TargetRegisterInfo. |
tree | commitdiff |
2008-02-09 |
Nate Begeman | memopv16i8 had wrong alignment requirement, would have... |
tree | commitdiff |
2008-02-09 |
Nate Begeman | Skeleton of insert and extract matching, more to come |
tree | commitdiff |
2008-02-08 |
Evan Cheng | It's not always safe to fold movsd into xorpd, etc... |
tree | commitdiff |
2008-02-08 |
Dale Johannesen | 64-bit (MMX) vectors do not need restrictive alignment. |
tree | commitdiff |
2008-02-08 |
Dan Gohman | Avoid needlessly casting away const qualifiers. |
tree | commitdiff |
2008-02-08 |
Evan Cheng | Added missing entries in X86 load / store folding tables. |
tree | commitdiff |
2008-02-07 |
Dan Gohman | Follow Chris' suggestion; change the PseudoSourceValue... |
tree | commitdiff |
2008-02-07 |
Dan Gohman | Add SourceValue information for outgoing argument store... |
tree | commitdiff |
2008-02-07 |
Evan Cheng | Fix a x86-64 codegen deficiency. Allow gv + offset... |
tree | commitdiff |
2008-02-07 |
Evan Cheng | In some cases, e.g. ADD32ri, no transformation is made... |
tree | commitdiff |
2008-02-06 |
Dan Gohman | Re-apply the memory operand changes, with a fix for... |
tree | commitdiff |
2008-02-05 |
Dale Johannesen | Implement sseregparm. |
tree | commitdiff |
2008-02-04 |
Evan Cheng | Dwarf requires variable entries to be in the source... |
tree | commitdiff |
2008-02-04 |
Nate Begeman | The rest of the SSE4.1 intrinsic patterns that are... |
tree | commitdiff |
2008-02-04 |
Nate Begeman | Some more SSE 4.1 intrinsic patterns. |
tree | commitdiff |
2008-02-03 |
Nate Begeman | SSE 4.1 Intrinsics and detection |
tree | commitdiff |
2008-02-02 |
Evan Cheng | Get rid of the annoying blank lines before labels. |
tree | commitdiff |
2008-02-02 |
Nick Lewycky | Don't use uninitialized values. Fixes vec_align.ll... |
tree | commitdiff |
2008-02-02 |
Evan Cheng | SDIsel processes llvm.dbg.declare by recording the... |
tree | commitdiff |
2008-02-02 |
Evan Cheng | Frame index can be negative. |
tree | commitdiff |
2008-02-01 |
Evan Cheng | Remove the nasty LABEL hack with a much less evil one... |
tree | commitdiff |
2008-01-31 |
Evan Cheng | Revert 46556 and 46585. Dan please fix the PseudoSource... |
tree | commitdiff |
2008-01-31 |
Evan Cheng | Add an extra operand to LABEL nodes which distinguishes... |
tree | commitdiff |
2008-01-31 |
Evan Cheng | Add x86 specific getFrameIndexOffset(). This fixes... |
tree | commitdiff |
2008-01-31 |
Dan Gohman | Avoid unnecessarily casting away const. |
tree | commitdiff |
2008-01-31 |
Dan Gohman | Rename ISD::FLT_ROUNDS to ISD::FLT_ROUNDS_ to avoid... |
tree | commitdiff |
2008-01-31 |
Dan Gohman | Create a new class, MemOperand, for describing memory... |
tree | commitdiff |
2008-01-30 |
Evan Cheng | Treat the label for the first @llvm.dbg.stoppoint the... |
tree | commitdiff |
2008-01-30 |
Evan Cheng | Even though InsertAtEndOfBasicBlock is an ugly hack... |
tree | commitdiff |
2008-01-30 |
Evan Cheng | Skip over the label which marks the beginning of the... |
tree | commitdiff |
2008-01-29 |
Evan Cheng | Work in progress. This patch *fixes* x86-64 calls which... |
tree | commitdiff |
2008-01-29 |
Dale Johannesen | Handle 'X' constraint in asm's better. |
tree | commitdiff |
2008-01-27 |
Chris Lattner | add a note |
tree | commitdiff |
2008-01-27 |
Chris Lattner | Use fldz and fld1 for long double constants instead... |
tree | commitdiff |
2008-01-26 |
Chris Lattner | Add some notes. |
tree | commitdiff |
2008-01-26 |
Chris Lattner | Remove some code for inferring alignment info from... |
tree | commitdiff |
2008-01-26 |
Bill Wendling | If there's no instructions being emitted on X86 for... |
tree | commitdiff |
2008-01-25 |
Chris Lattner | optimize fxor like for |
tree | commitdiff |
2008-01-25 |
Chris Lattner | Add target-specific dag combines for FAND(x,0) and... |
tree | commitdiff |
2008-01-25 |
Anton Korobeynikov | Provide correct DWARF register numbering for debug... |
tree | commitdiff |
2008-01-24 |
Chris Lattner | Significantly simplify and improve handling of FP funct... |
tree | commitdiff |
2008-01-23 |
Evan Cheng | Let each target decide byval alignment. For X86, it... |
tree | commitdiff |
2008-01-23 |
Duncan Sands | The last pieces needed for loading arbitrary |
tree | commitdiff |
2008-01-23 |
Dale Johannesen | Honor explicit section information on Darwin. |
tree | commitdiff |
2008-01-22 |
Evan Cheng | SSE varargs arguments are passed in memory. |
tree | commitdiff |
2008-01-20 |
Anton Korobeynikov | Honour ByVal parameter attribute for name decoration |
tree | commitdiff |
2008-01-20 |
Anton Korobeynikov | Remove Darwin'ism |
tree | commitdiff |
2008-01-20 |
Anton Korobeynikov | Enable PIC codegen on x86-64/linux |
tree | commitdiff |
2008-01-19 |
Duncan Sands | Need to handle any 'nest' parameter before integer |
tree | commitdiff |
2008-01-18 |
Chris Lattner | make a method public |
tree | commitdiff |
2008-01-17 |
Dale Johannesen | Revert the part of 45848 that treated weak globals |
tree | commitdiff |
2008-01-17 |
Chris Lattner | This commit changes: |
tree | commitdiff |
2008-01-17 |
Chris Lattner | * Introduce a new SelectionDAG::getIntPtrConstant method |
tree | commitdiff |
2008-01-16 |
Duncan Sands | Trampoline support for x86-64. This looks like |
tree | commitdiff |
2008-01-16 |
Chris Lattner | make it more clear that this predicate only applies... |
tree | commitdiff |
2008-01-16 |
Chris Lattner | introduce a isTypeInSSEReg predicate, which allows... |
tree | commitdiff |
2008-01-16 |
Chris Lattner | My previous commit had an incomplete message, it should... |
tree | commitdiff |
2008-01-16 |
Chris Lattner | make the 'fp return in ST(0)' optimization smart enough to |
tree | commitdiff |
2008-01-16 |
Chris Lattner | various whitespace cleanups, no functionality change. |
tree | commitdiff |
2008-01-15 |
Dale Johannesen | Fix and enable EH for x86-64 Darwin. Adds |
tree | commitdiff |
2008-01-15 |
Chris Lattner | rename SDTRet -> SDTNone. |
tree | commitdiff |
2008-01-15 |
Chris Lattner | no need to expand ISD::TRAP to X86ISD::TRAP, just match... |
tree | commitdiff |
2008-01-15 |
Anton Korobeynikov | Fix JIT encoding of trap/ud2 instruction |
tree | commitdiff |
2008-01-15 |
Anton Korobeynikov | For PR1839: add initial support for __builtin_trap... |
tree | commitdiff |
2008-01-15 |
Evan Cheng | Rename CCIfStruct to CCIfByVal and CCStructAssign to... |
tree | commitdiff |
2008-01-15 |
Evan Cheng | Both x86-32 and x86-64 handle byval parameter attributes. |
tree | commitdiff |
2008-01-14 |
Chris Lattner | Improve the FP stackifier to decide all on its own... |
tree | commitdiff |
2008-01-13 |
Duncan Sands | Whitespace tweak. |
tree | commitdiff |
2008-01-12 |
Evan Cheng | Code clean up. |
tree | commitdiff |
2008-01-12 |
Chris Lattner | fix a wordo that gordon noticed :) |
tree | commitdiff |
2008-01-12 |
Chris Lattner | Any x86 instruction that reads from an invariant locati... |
tree | commitdiff |
2008-01-11 |
Chris Lattner | rename MachineInstr::setInstrDescriptor -> setDesc |
tree | commitdiff |
2008-01-11 |
Chris Lattner | remove xchg and shift-reg-by-1 instructions, which... |
tree | commitdiff |
2008-01-11 |
Chris Lattner | add a note, remove a done deed. |
tree | commitdiff |
2008-01-11 |
Arnold Schwaighofer | hrm - correct spelling. |
tree | commitdiff |
2008-01-11 |
Arnold Schwaighofer | Improve tail call optimized call's argument lowering... |
tree | commitdiff |
2008-01-11 |
Arnold Schwaighofer | Correct a copy and paste error. |
tree | commitdiff |
2008-01-11 |
Evan Cheng | Rename Int_CVTSI642SSr* to Int_CVTSI2SS64r* for naming... |
tree | commitdiff |
2008-01-11 |
Chris Lattner | more flags set right |
tree | commitdiff |
2008-01-11 |
Chris Lattner | add some missing flags. |
tree | commitdiff |
2008-01-11 |
Dale Johannesen | Weak things initialized to 0 don't go in bss on Darwin. |
tree | commitdiff |
2008-01-10 |
Chris Lattner | Simplify the side effect stuff a bit more and make... |
tree | commitdiff |
2008-01-10 |
Chris Lattner | IMPLICIT_USE and IMPLICIT_DEF are dead, remove them. |
tree | commitdiff |
2008-01-10 |
Chris Lattner | Start inferring side effect information more aggressive... |
tree | commitdiff |
2008-01-10 |
Chris Lattner | rename X86InstrX86-64.td -> X86Instr64bit.td |
tree | commitdiff |
2008-01-10 |
Chris Lattner | remove explicit sets of 'neverHasSideEffects' that... |
tree | commitdiff |
2008-01-10 |
Chris Lattner | get def use info more correct. |
tree | commitdiff |
2008-01-10 |
Chris Lattner | verify that the frame index is immutable before remat... |
tree | commitdiff |
2008-01-10 |
Evan Cheng | Mark byval parameter stack objects mutable for now. |
tree | commitdiff |
2008-01-10 |
Dale Johannesen | Emit unused EH frames for weak definitions on Darwin, |
tree | commitdiff |
2008-01-10 |
Evan Cheng | Do not use the stack pointer directly, issue a copyfrom... |
tree | commitdiff |
2008-01-10 |
Evan Cheng | Remove comments that do not correspond to anything... |
tree | commitdiff |
2008-01-09 |
Chris Lattner | add a testcase |
tree | commitdiff |
next |