[PATCH] D15015: [AArch64] Add ARMv8.2-A FP16 vector instructions

Oliver Stannard via llvm-commits llvm-commits at lists.llvm.org
Thu Nov 26 05:10:34 PST 2015


olista01 created this revision.
olista01 added a reviewer: t.p.northover.
olista01 added a subscriber: llvm-commits.
olista01 set the repository for this revision to rL LLVM.
Herald added subscribers: rengolin, aemerson.

ARMv8.2-A adds 16-bit floating point versions of all existing SIMD
floating-point instructions. This is an optional extension, so all of
these instructions require the FeatureFullFP16 subtarget feature.

Note that VFP without SIMD is not a valid combination for any version of
ARMv8-A, but I have ensured that these instructions all depend on both
FeatureNEON and FeatureFullFP16 for consistency.

The ".2h" vector type specifier is now legal (for the scalar pairwise
reduction instructions), so some unrelated tests have been modified as
different error messages are emitted. This is not a problem as the
invalid operands are still caught.

Repository:
  rL LLVM

http://reviews.llvm.org/D15015

Files:
  lib/Target/AArch64/AArch64InstrFormats.td
  lib/Target/AArch64/AArch64InstrInfo.td
  lib/Target/AArch64/AArch64RegisterInfo.td
  lib/Target/AArch64/AsmParser/AArch64AsmParser.cpp
  test/MC/AArch64/arm64-advsimd.s
  test/MC/AArch64/armv8.1a-rdma.s
  test/MC/AArch64/fullfp16-diagnostics.s
  test/MC/AArch64/fullfp16-neon-neg.s
  test/MC/AArch64/neon-2velem.s
  test/MC/AArch64/neon-aba-abd.s
  test/MC/AArch64/neon-across.s
  test/MC/AArch64/neon-add-pairwise.s
  test/MC/AArch64/neon-add-sub-instructions.s
  test/MC/AArch64/neon-compare-instructions.s
  test/MC/AArch64/neon-diagnostics.s
  test/MC/AArch64/neon-facge-facgt.s
  test/MC/AArch64/neon-frsqrt-frecp.s
  test/MC/AArch64/neon-max-min-pairwise.s
  test/MC/AArch64/neon-max-min.s
  test/MC/AArch64/neon-mla-mls-instructions.s
  test/MC/AArch64/neon-scalar-abs.s
  test/MC/AArch64/neon-scalar-by-elem-mla.s
  test/MC/AArch64/neon-scalar-by-elem-mul.s
  test/MC/AArch64/neon-scalar-cvt.s
  test/MC/AArch64/neon-scalar-fp-compare.s
  test/MC/AArch64/neon-scalar-mul.s
  test/MC/AArch64/neon-scalar-recip.s
  test/MC/AArch64/neon-scalar-reduce-pairwise.s
  test/MC/AArch64/neon-simd-misc.s
  test/MC/AArch64/neon-simd-shift.s
  test/MC/Disassembler/AArch64/fullfp16-neon-neg.txt

-------------- next part --------------
A non-text attachment was scrubbed...
Name: D15015.41236.patch
Type: text/x-patch
Size: 219561 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20151126/bd662bb7/attachment-0001.bin>


More information about the llvm-commits mailing list