[PATCH] D88215: Add llvm.loop.unroll.threshold metadata

[Michael Kruse via Phabricator via llvm-commits]

Meinersbur added a comment.

Unlike the reviewers in D84779 <https://reviews.llvm.org/D84779>, I would be OK with it being target specific. We already have precedence for this in form of the `amdgpu-unroll-threshold` function attribute and this just applied the same information more specifically to loops. I don't see the issue of non-generality being raised in D68873 <https://reviews.llvm.org/D68873>.

I'd object to an `llvm.loop.unroll.threshold` metadata without even the intend to define/implement it to other targets as well. I think the threshold number itself is inherently target-specific, so I think it is difficult to define "threshold of 10000" to be comparable between AMD, X86, PowerPC, etc.

Sorry you're caught between reviewer's opinions.


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D88215/new/

https://reviews.llvm.org/D88215