Match VTRN, VZIP, and VUZP shuffles. Restore the tests for these operations,
[oota-llvm.git] / lib / Target / ARM / ARMISelLowering.cpp
2009-08-21 Bob WilsonMatch VTRN, VZIP, and VUZP shuffles. Restore the tests...
2009-08-21 Anton KorobeynikovFix some typos and use type-based isel for VZIP/VUZP...
2009-08-21 Anton KorobeynikovAdd lowering of ARM 4-element shuffles to multiple...
2009-08-21 Anton KorobeynikovAdd nodes & dummy matchers for some v{zip,uzp,trn}...
2009-08-21 Anton KorobeynikovExpand EXTRACT_SUBVECTOR
2009-08-21 Anton KorobeynikovProvide vext.{16,32}
2009-08-21 Anton KorobeynikovUse masks not nodes for vector shuffle predicates....
2009-08-19 Bob WilsonAdd support for Neon VEXT (vector extract) shuffles.
2009-08-15 Bill WendlingReapply r79127. It was fixed by d0k.
2009-08-15 Bill WendlingRevert r79127. It was causing compilation errors.
2009-08-15 Evan ChengChange allowsUnalignedMemoryAccesses to take type argum...
2009-08-15 Evan ChengTurn on if-conversion for thumb2.
2009-08-14 Anton KorobeynikovAllow targets to specify their choice of calling conven...
2009-08-14 Evan ChengAdd Thumb2 lsr hooks.
2009-08-14 Evan Cheng80 col violation.
2009-08-14 Bob WilsonNow that all the legal Neon shuffles (or at least the...
2009-08-14 Bob WilsonCreate a new ARM-specific DAG node, VDUP, to represent...
2009-08-14 Bob WilsonDuring legalization, change Neon vdup_lane operations...
2009-08-13 Owen AndersonPush LLVMContexts through the IntegerType APIs.
2009-08-13 Bob WilsonAdd a fixme message about canonicalizing floating-point...
2009-08-13 Bob WilsonRevert r78852 for now. I want to do this differently...
2009-08-13 Bob WilsonAdd a comment to describe why vector shuffles are legal...
2009-08-13 Bob WilsonUse cast<> instead of dyn_cast<> in places where the...
2009-08-12 Bob WilsonRecognize Neon VDUP shuffles during legalization instea...
2009-08-12 Bob WilsonRecognize Neon VREV shuffles during legalization instea...
2009-08-12 Jim GrosbachAdd catch block handling to SjLj exception handling.
2009-08-12 Evan ChengShrink Thumb2 movcc instructions.
2009-08-11 Owen AndersonSplit EVT into MVT and EVT, the former representing...
2009-08-11 Jim GrosbachWhitespace cleanup. Remove trailing whitespace.
2009-08-11 Bob WilsonUse vAny type to get rid of Neon intrinsics that differ...
2009-08-11 Jim GrosbachSjLj based exception handling unwinding support. This...
2009-08-10 Owen AndersonRename MVT to EVT, in preparation for splitting SimpleV...
2009-08-10 Owen AndersonSimpleValueType-ify a few more methods on TargetLowering.
2009-08-10 Owen AndersonContinue the SimpleValueType-ification.
2009-08-07 Evan ChengIt turns out most of the thumb2 instructions are not...
2009-08-06 Bob WilsonImplement Neon VST[234] operations.
2009-08-05 Anton KorobeynikovRemove redundand checks: the only way to have, e.g...
2009-08-05 Anton KorobeynikovUnbreak the stuff, this is ugly, but we cannot do bette...
2009-08-05 Anton KorobeynikovMissed pieces for ARM HardFP ABI.
2009-08-05 Dan GohmanMajor calling convention code refactoring.
2009-08-05 Bob WilsonChange DAG nodes for Neon VLD2/3/4 operations to return...
2009-08-04 Bob WilsonLower Neon VLD* intrinsics to custom DAG nodes, and...
2009-08-04 Bob WilsonMinor cleanup. No functional changes intended.
2009-08-03 Bob WilsonLower CONCAT_VECTOR during legalization instead of...
2009-08-02 Chris Lattnerconvert ctors/dtors section to be in TLOF instead of
2009-08-01 Evan ChengFix Thumb2 function call isel. Thumb1 and Thumb2 should...
2009-07-31 Chris Lattnerrefactor section construction in TLOF to be through...
2009-07-30 Bob WilsonLower a 128-bit BUILD_VECTOR with 2 elements to a pair...
2009-07-29 Evan ChengOptimize Thumb2 jumptable to use tbb / tbh when all...
2009-07-28 Evan ChengIn thumb2 mode, add pc is unpredictable. Use add +...
2009-07-28 Chris Lattnerthe apple "ld_classic" linker doesn't support .literal1...
2009-07-28 Chris LattnerRip all of the global variable lowering logic out of...
2009-07-26 Bob WilsonAdd support for ARM Neon VREV instructions.
2009-07-25 Evan ChengChange Thumb2 jumptable codegen to one that uses two...
2009-07-22 Owen AndersonGet rid of the Pass+Context magic.
2009-07-15 Chris Lattnerfix an arm codegen bug (the same as PR4482 on ppc)...
2009-07-14 Bob WilsonRemove an extra space.
2009-07-14 Torok Edwinllvm_unreachable->llvm_unreachable(0), LLVM_UNREACHABLE...
2009-07-13 Bob WilsonFix comment typos.
2009-07-11 Torok Edwinassert(0) -> LLVM_UNREACHABLE.
2009-07-09 Owen AndersonThread LLVMContext through MVT and related parts of...
2009-07-08 David GoodwinUse common code for both ARM and Thumb-2 instruction...
2009-07-08 Torok EdwinStart converting to new error handling API.
2009-07-08 Nick LewyckyRemove the vicmp and vfcmp instructions. Because we...
2009-07-07 Evan ChengAdd some more Thumb2 multiplication instructions.
2009-07-03 Tilmann SchellerAdd NumFixedArgs attribute to CallSDNode which indicate...
2009-07-03 Evan ChengAdd thumb2 sign / zero extend with rotate instructions.
2009-07-02 Evan ChengThumb2 pre/post indexed loads.
2009-07-02 Evan Cheng80 col violation.
2009-07-01 Bill WendlingUpdate comments to make it clear that the function...
2009-06-30 Bill WendlingAdd an "alignment" field to the MachineFunction object...
2009-06-29 David GoodwinRename ARMcmpNZ to ARMcmpZ and use it to represent...
2009-06-26 David GoodwinThumb-2 has CLZ.
2009-06-22 Bob WilsonAdd support for ARM's Advanced SIMD (NEON) instruction...
2009-06-19 Evan ChengAdd comments.
2009-06-19 Evan ChengShould be using Bcc (average) latency to determine...
2009-06-19 Evan ChengLatency information for ARM v6. It's rough and not...
2009-06-18 Evan ChengRemove UseThumbBacktraces. Just check if subtarget...
2009-06-16 Anton KorobeynikovAddress review comments: add 3 ARM calling conventions.
2009-06-08 Anton KorobeynikovThe attached patches implement most of the ARM AAPCS...
2009-05-22 Bob WilsonOnly 64-bit targets support TImode libcalls. Disable...
2009-05-18 Bob WilsonFix pr4202: Disable CodePlacementOpt for ARM. The...
2009-05-14 Jim GrosbachUpdate the names of the exception handling sjlj instrin...
2009-05-13 Evan ChengRun code placement optimization for targets that want...
2009-05-12 Jim GrosbachAdd support for GCC compatible builtin setjmp and longj...
2009-04-25 Bob WilsonChange LowerCallResult method so that CCValAssign:...
2009-04-24 Bob WilsonAdjust a comment to reflect what the code does. Splitt...
2009-04-24 Bob WilsonFix up some problems with getCopyToReg and getCopyFromR...
2009-04-20 Bob WilsonMove duplicated AddLiveIn function from X86 and ARM...
2009-04-17 Bob WilsonMove the AddLiveIn function definition closer to its...
2009-04-17 Bob WilsonRearrange code to reduce indentation.
2009-04-17 Bob WilsonClean up formatting, remove trailing whitespace, fix...
2009-04-17 Bob WilsonUse CallConvLower.h and TableGen descriptions of the...
2009-04-08 Bob WilsonFix PR3795: Apply Dan's suggested fix for
2009-04-07 Jim GrosbachPR2985 / <rdar://problem/6584986>
2009-04-01 Bob WilsonFix PR3862: Recognize some ARM-specific constraints...
2009-03-20 Bob WilsonFix a few more indentation problems and an 80-column...
2009-03-20 Bob WilsonNo functional changes. Fix indentation and whitespace...
2009-03-13 Evan ChengFix some significant problems with constant pools that...
2009-03-12 Chris LattnerMove 3 "(add (select cc, 0, c), x) -> (select cc, x...
next