[all-commits] [llvm/llvm-project] 9679fb: [LLVM][CodeGen][SME] Don't use non-streaming zeroi...

[Paul Walker via All-commits]

  Branch: refs/heads/main
  Home:   https://github.com/llvm/llvm-project
  Commit: 9679fbfea993cb2e815fe0fa5eefde310f917349
      https://github.com/llvm/llvm-project/commit/9679fbfea993cb2e815fe0fa5eefde310f917349
  Author: Paul Walker <paul.walker at arm.com>
  Date:   2026-09-10 (Thu, 10 Sep 2026)

  Changed paths:
    M llvm/lib/Target/AArch64/AArch64SVEInstrInfo.td
    M llvm/test/CodeGen/AArch64/sve-bf16-arith.ll
    M llvm/test/CodeGen/AArch64/sve-bf16-compares.ll
    M llvm/test/CodeGen/AArch64/sve-bf16-converts.ll
    M llvm/test/CodeGen/AArch64/sve-bf16-int-converts.ll
    M llvm/test/CodeGen/AArch64/sve-bf16-reductions.ll
    M llvm/test/CodeGen/AArch64/sve-bf16-rounding.ll

  Log Message:
  -----------
  [LLVM][CodeGen][SME] Don't use non-streaming zeroing instruction when in streaming mode. (#222353)

When selecting scalable vector bf16-f32 converts we emit zip(0,vec).
Unfortunately, the typical zero-latency zeroing instruction is not
available when in streaming mode.



To unsubscribe from these emails, change your notification settings at https://github.com/llvm/llvm-project/settings/notifications