[all-commits] [llvm/llvm-project] 5a2539: [ARM] Add and update FMA tests. NFC

[David Green via All-commits]

  Branch: refs/heads/master
  Home:   https://github.com/llvm/llvm-project
  Commit: 5a253992212451be919978610332cc42d4b48d47
      https://github.com/llvm/llvm-project/commit/5a253992212451be919978610332cc42d4b48d47
  Author: David Green <david.green at arm.com>
  Date:   2020-01-05 (Sun, 05 Jan 2020)

  Changed paths:
    M llvm/test/CodeGen/ARM/cortex-a57-misched-vfma.ll
    A llvm/test/CodeGen/ARM/fp16-fusedMAC.ll
    M llvm/test/CodeGen/ARM/fusedMAC.ll

  Log Message:
  -----------
  [ARM] Add and update FMA tests. NFC


  Commit: c15a56f61a56e862c9613a334d1427638899942b
      https://github.com/llvm/llvm-project/commit/c15a56f61a56e862c9613a334d1427638899942b
  Author: David Green <david.green at arm.com>
  Date:   2020-01-05 (Sun, 05 Jan 2020)

  Changed paths:
    M llvm/lib/Target/ARM/ARMInstrVFP.td
    M llvm/test/CodeGen/ARM/fp16-fusedMAC.ll

  Log Message:
  -----------
  [ARM] Fill in FP16 FMA patterns

This adds fp16 variants of all the fma patterns in the ARM backend.

Differential Revision: https://reviews.llvm.org/D72138


  Commit: fb8c9a339a9d0b78370fbd814d62dd5779f1e196
      https://github.com/llvm/llvm-project/commit/fb8c9a339a9d0b78370fbd814d62dd5779f1e196
  Author: David Green <david.green at arm.com>
  Date:   2020-01-05 (Sun, 05 Jan 2020)

  Changed paths:
    M llvm/lib/Target/ARM/ARM.td
    M llvm/lib/Target/ARM/ARMISelLowering.cpp
    M llvm/lib/Target/ARM/ARMPredicates.td
    M llvm/lib/Target/ARM/ARMSubtarget.h
    M llvm/lib/Target/ARM/ARMTargetTransformInfo.h
    M llvm/test/CodeGen/ARM/cortex-a57-misched-vfma.ll
    M llvm/test/CodeGen/ARM/fp16-fullfp16.ll
    M llvm/test/CodeGen/ARM/fp16-fusedMAC.ll
    M llvm/test/CodeGen/Thumb2/float-intrinsics-double.ll
    M llvm/test/CodeGen/Thumb2/float-intrinsics-float.ll

  Log Message:
  -----------
  [ARM] Use isFMAFasterThanFMulAndFAdd for scalars as well as MVE vectors

This adds extra scalar handling to isFMAFasterThanFMulAndFAdd, allowing
the target independent code to handle more folds in more situations (for
example if the fast math flags are present, but the global
AllowFPOpFusion option isnt). It also splits apart the HasSlowFPVMLx
into HasSlowFPVFMx, to allow VFMA and VMLA to be controlled separately
if needed.

Differential Revision: https://reviews.llvm.org/D72139


Compare: https://github.com/llvm/llvm-project/compare/170de3de2eea...fb8c9a339a9d