projects / oota-llvm.git / search
? search:
summary | shortlog | log | commit | commitdiff | tree
first ⋅ prev ⋅ next
CodeGen: assert an instruction is being inserted with the correct iterator.
2014-11-14 Tim NorthoverCodeGen: assert an instruction is being inserted with...
commit | commitdiff | tree
2014-11-13 Tim NorthoverARM: allow constpool entry to be moved to the user...
commit | commitdiff | tree
2014-11-13 Tim NorthoverARM: avoid duplicating branches during constant islands.
 commit | commitdiff | tree
2014-11-13 Tim NorthoverARM: add @llvm.arm.space intrinsic for testing ConstantIslands.
 commit | commitdiff | tree
2014-11-05 Tim NorthoverARM: try to add extra CS-register whenever stack alignment...
commit | commitdiff | tree
2014-11-05 Tim NorthoverARM/Dwarf: correctly align stack before callee-saved...
commit | commitdiff | tree
2014-11-04 Tim NorthoverDocs: give binutils/gold instructions for CMake too.
 commit | commitdiff | tree
2014-11-02 Tim NorthoverDocs: update va_arg example with valid x86_64 va_list...
commit | commitdiff | tree
2014-10-30 Tim NorthoverARM: test default values for TAG_CPU_unaligned_access...
commit | commitdiff | tree
2014-10-28 Tim NorthoverAArch64: enable Cortex-A57 FP balancing on Cortex-A53.
 commit | commitdiff | tree
2014-10-23 Tim NorthoverScheduleDAG: record PhysReg dependencies represented...
commit | commitdiff | tree
2014-10-20 Tim NorthoverARM: rework Thumb1 frame index rewriting
 commit | commitdiff | tree
2014-10-15 Tim NorthoverARM: drop check for triple that's no longer used.
 commit | commitdiff | tree
2014-10-14 Tim NorthoverARM: remove ARM/Thumb distinction for preferred alignment.
 commit | commitdiff | tree
2014-10-14 Tim NorthoverARM: allow misaligned local variables in Thumb1 mode.
 commit | commitdiff | tree
2014-10-14 Tim NorthoverARM: set preferred aggregate alignment to 32 universally.
 commit | commitdiff | tree
2014-10-06 Tim NorthoverARM: silence unused variable warning
 commit | commitdiff | tree
2014-10-06 Tim NorthoverARM: remove dead InstPrinting code
 commit | commitdiff | tree
2014-10-01 Tim NorthoverARM: yes it can (as of r218789)
 commit | commitdiff | tree
2014-10-01 Tim NorthoverARM: allow copying of CPSR when all else fails.
 commit | commitdiff | tree
2014-09-10 Tim NorthoverARM: don't size-reduce STMs using the LR register.
 commit | commitdiff | tree
2014-09-09 Tim Northoverllvm-objdump: don't crash when __compact_unwind has...
commit | commitdiff | tree
2014-09-05 Tim NorthoverARM: cover all sub-architecture enumerators to keep...
commit | commitdiff | tree
2014-09-04 Tim NorthoverAArch64: fix vector-immediate BIC/ORR on big-endian...
commit | commitdiff | tree
2014-09-04 Tim NorthoverAArch64: fix big-endian immediate materialisation
 commit | commitdiff | tree
2014-08-29 Tim NorthoverAArch64: only try to get operand of a known node.
 commit | commitdiff | tree
2014-08-29 Tim NorthoverAArch64: skip select/setcc combine in complex case.
 commit | commitdiff | tree
2014-08-18 Tim NorthoverTableGen: allow use of uint64_t for available features...
commit | commitdiff | tree
2014-08-15 Tim NorthoverARM: implement MRS/MSR (banked reg) system instructions.
 commit | commitdiff | tree
2014-08-12 Tim Northoverllvm-objdump: print contents of MachO __unwind_info...
commit | commitdiff | tree
2014-08-11 Tim NorthoverAArch64: add support for dynamic-loader relocations
 commit | commitdiff | tree
2014-08-11 Tim Northoverllvm-readobj: zero out timestamp in COFF auto-generated...
commit | commitdiff | tree
2014-08-11 Tim Northoverllvm-objdump: fix remaining use of %x format specifier...
commit | commitdiff | tree
2014-08-08 Tim NorthoverAArch64: avoid deleting the current iterator in a loop.
 commit | commitdiff | tree
2014-08-08 Tim Northoverllvm-objdump: add missing % in format specifier.
 commit | commitdiff | tree
2014-08-08 Tim Northoverllvm-objdump: use portable format specifiers for info.
 commit | commitdiff | tree
2014-08-08 Tim NorthoverAArch64: stop trying to take control of all UnknownArch...
commit | commitdiff | tree
2014-08-06 Tim NorthoverARM: do not generate BLX instructions on Cortex-M CPUs.
 commit | commitdiff | tree
2014-08-06 Tim NorthoverARM-MachO: materialize callee address correctly on...
commit | commitdiff | tree
2014-08-01 Tim Northoverllvm-objdump: implement printing for MachO __compact_unwind...
commit | commitdiff | tree
2014-07-29 Tim NorthoverCodeGenPrep: fall back to MVT::Other if instruction...
commit | commitdiff | tree
2014-07-29 Tim NorthoverARM: add __aeabi_d2h for truncation on AEABI systems
 commit | commitdiff | tree
2014-07-29 Tim NorthoverARM: fix @llvm.convert.from.fp16 on softfloat targets.
 commit | commitdiff | tree
2014-07-27 Tim NorthoverAArch64: fix conversion of 'J' inline asm constraints.
 commit | commitdiff | tree
2014-07-24 Tim NorthoverAArch64: refactor ReconstructShuffle function
 commit | commitdiff | tree
2014-07-23 Tim NorthoverARM: spot SBFX-compatbile code expressed with sign_extend_inreg
 commit | commitdiff | tree
2014-07-23 Tim NorthoverARM: add patterns for [su]xta[bh] from just a shift.
 commit | commitdiff | tree
2014-07-23 Tim NorthoverAArch64: remove "arm64_be" support in favour of "aarch64_be".
 commit | commitdiff | tree
2014-07-23 Tim NorthoverAArch64: remove arm64 triple enumerator.
 commit | commitdiff | tree
2014-07-22 Tim NorthoverX86: drop relocations on __eh_frame sections globally.
 commit | commitdiff | tree
2014-07-21 Tim NorthoverCodeGen: emit IR-level f16 conversion intrinsics as...
commit | commitdiff | tree
2014-07-18 Tim NorthoverAArch64: implement efficient f16 bitcasts
 commit | commitdiff | tree
2014-07-18 Tim NorthoverNVPTX: support fpext/fptrunc to and from f16.
 commit | commitdiff | tree
2014-07-18 Tim NorthoverR600: support fpext/fptrunc operations to and from...
commit | commitdiff | tree
2014-07-18 Tim NorthoverAArch64: support f16 extend/trunc operations.
 commit | commitdiff | tree
2014-07-18 Tim NorthoverX86: support fpext/fptrunc operations to and from 16...
commit | commitdiff | tree
2014-07-18 Tim NorthoverARM: support legalisation of "fptrunc ... to half"...
commit | commitdiff | tree
2014-07-18 Tim NorthoverCodeGen: soften f16 type by default instead of marking...
commit | commitdiff | tree
2014-07-18 Tim NorthoverR600: rename misleading fp16 test.
 commit | commitdiff | tree
2014-07-18 Tim NorthoverR600: support f16 -> f64 conversion intrinsic.
 commit | commitdiff | tree
2014-07-18 Tim NorthoverNVPTX: support direct f16 <-> f64 conversions via intrinsics.
 commit | commitdiff | tree
2014-07-17 Tim NorthoverARM: support direct f16 <-> f64 conversions
 commit | commitdiff | tree
2014-07-17 Tim NorthoverCodeGen: generate single libcall for fptrunc -> f16...
commit | commitdiff | tree
2014-07-17 Tim NorthoverX86: support double extension of f16 type.
 commit | commitdiff | tree
2014-07-17 Tim NorthoverCodeGen: extend f16 conversions to permit types > float.
 commit | commitdiff | tree
2014-07-16 Tim NorthoverCodeGen: don't form illegail EXTLOAD operations.
 commit | commitdiff | tree
2014-07-16 Tim NorthoverConvert test to CHECK-LABEL
 commit | commitdiff | tree
2014-07-15 Tim NorthoverAArch64: fall back to generic code for out of range...
commit | commitdiff | tree
2014-07-14 Tim NorthoverX86: remove temporary atomicrmw used during lowering.
 commit | commitdiff | tree
2014-07-14 Tim NorthoverAArch64: remove unnecessary pseudo-instruction.
 commit | commitdiff | tree
2014-07-10 Tim NorthoverAArch64: correctly fast-isel i8 & i16 multiplies
 commit | commitdiff | tree
2014-07-09 Tim NorthoverUse simpler constructor for range adapter.
 commit | commitdiff | tree
2014-07-09 Tim NorthoverGeneric: add range-adapter for option parsing.
 commit | commitdiff | tree
2014-07-07 Tim Northover[testing]: lld generally lives in tools/, so fix llvm...
commit | commitdiff | tree
2014-07-07 Tim NorthoverX86: revert unintentional change to X86FastISel.
 commit | commitdiff | tree
2014-07-07 Tim NorthoverCodeGen: it turns out that NAND is not the same thing...
commit | commitdiff | tree
2014-07-04 Tim NorthoverARM: when falling back to scattered relocs, keep the...
commit | commitdiff | tree
2014-07-04 Tim Northoverllvm-readobj: fix MachO relocatoin printing a bit.
 commit | commitdiff | tree
2014-07-01 Tim NorthoverX86: remove atomic instructions *after* we've iterated...
commit | commitdiff | tree
2014-07-01 Tim NorthoverX86: delegate expanding atomic libcalls to generic...
commit | commitdiff | tree
2014-07-01 Tim NorthoverAArch64: fix comment typo
 commit | commitdiff | tree
2014-07-01 Tim NorthoverX86: expand atomics in IR instead of as MachineInstrs.
 commit | commitdiff | tree
2014-06-30 Tim Northovermacho-dump: add code to print LC_ID_DYLIB load commands.
 commit | commitdiff | tree
2014-06-23 Tim NorthoverARM: mark UBFX as not allowing PC.
 commit | commitdiff | tree
2014-06-18 Tim NorthoverDAG: move sret demotion into most basic LowerCallTo...
commit | commitdiff | tree
2014-06-17 Tim NorthoverAArch64: estimate inline asm length during branch relaxation
 commit | commitdiff | tree
2014-06-16 Tim NorthoverARM: implement correct atomic operations on v7M
 commit | commitdiff | tree
2014-06-15 Tim NorthoverLegalizeDAG: make sure cast is unsigned before using...
commit | commitdiff | tree
2014-06-15 Tim NorthoverAArch64: improve handling & modelling of FP_TO_XINT...
commit | commitdiff | tree
2014-06-15 Tim NorthoverAArch64: improve vector [su]itofp handling.
 commit | commitdiff | tree
2014-06-13 Tim NorthoverX86: lower ATOMIC_CMP_SWAP_WITH_SUCCESS directly
 commit | commitdiff | tree
2014-06-13 Tim NorthoverAtomics: make use of the "cmpxchg weak" instruction.
 commit | commitdiff | tree
2014-06-13 Tim NorthoverAtomics: switch direction of cmpxchg comparison
 commit | commitdiff | tree
2014-06-13 Tim NorthoverSCCP: update for cmpxchg returning { iN, i1 } now.
 commit | commitdiff | tree
2014-06-13 Tim NorthoverDocs: remove extra {} around result types.
 commit | commitdiff | tree
2014-06-13 Tim NorthoverDocs: fix grammar error in description
 commit | commitdiff | tree
2014-06-13 Tim NorthoverIR: add "cmpxchg weak" variant to support permitted...
commit | commitdiff | tree
2014-06-13 Tim NorthoverCPP backend: set volatile property on atomic instructions.
 commit | commitdiff | tree
2014-06-11 Tim NorthoverX86: add stringy name for X86ISD::LCMPXCHG16_DAG
 commit | commitdiff | tree
2014-06-10 Tim NorthoverAArch64: disallow x30 & x29 as the destination for...
commit | commitdiff | tree
next
C/C++ LLVM-based compilers that forbids OOTA behaviors