Redesign atomic store with hints intrinsics

[Lukacma]

This proposal is about redesigning atomic store with hint intrinsics, to make them more easily extensible.

Checklist: (mark with X those which apply)

• If an issue reporting the bug exists, I have mentioned it in the
  PR (do not bother creating the issue if all you want to do is
  fixing the bug yourself).
• [x ] I have added/updated the SPDX-FileCopyrightText lines on top
  of any file I have edited. Format is SPDX-FileCopyrightText: Copyright {year} {entity or name} <{contact informations}>
  (Please update existing copyright lines if applicable. You can
  specify year ranges with hyphen , as in 2017-2019, and use
  commas to separate gaps, as in 2018-2020, 2022).
• [ x] I have updated the Copyright section of the sources of the
  specification I have edited (this will show up in the text
  rendered in the PDF and other output format supported). The
  format is the same described in the previous item.
• I have run the CI scripts (if applicable, as they might be
  tricky to set up on non-*nix machines). The sequence can be
  found in the contribution
  guidelines. Don't
  worry if you cannot run these scripts on your machine, your
  patch will be automatically checked in the Actions of the pull
  request.
• [x ] I have added an item that describes the changes I have
  introduced in this PR in the section Changes for next
  release of the section Change Control/Document history
  of the document. Create Changes for next release if it does
  not exist. Notice that changes that are not modifying the
  content and rendering of the specifications (both HTML and PDF)
  do not need to be listed.
• [ x] When modifying content and/or its rendering, I have checked the
  correctness of the result in the PDF output (please refer to the
  instructions on how to build the PDFs
  locally).
• [ x] The variable draftversion is set to true in the YAML header
  of the sources of the specifications I have modified.
• Please DO NOT add my GitHub profile to the list of contributors
  in the README page of the project.

[CarolineConcatto]

What will be the behaviour if hint is not valid?

[Lukacma]

I guess that is undefined behaviour and will be implementation based. For LLVM we will throw error and reject.

[CarolineConcatto]

Should it throw an error or just ignore and still do the atomic store?

[Lukacma]

I would throw error. This seems like a user mistake and we shouldn't just let it silently pass imo.

[ricbal02]

I agree with throwing an error, the generic versions of these should be used when not using a hint

[paulwalker-arm]

Given the point of this change is to make the builtin more generic, it doesn't make sense to then add barriers to future uses. We've already said a toolchain is free to ignore the hint and that's effectively what an older toolchain should do if given a hint that only newer toolchains support. If we want to help users spot potential typos then we can add an "unknown hint will be ignored" diagnostic.

[ricbal02]

So for the atomic_store intrinsics here, it would be fine to have them work without the hint, however, for the atomic_fetch intrinsics added in #430, the assembly output is limited to the LSE variant as the hint does not work for other variants, so outline atomics will not be used. In that case we really don't want this to be seen as an alternative to using the non-hint atomics

[sallyarmneale]

Is it Hint or hint?

[CarolineConcatto]

What is the difference between Hint and hint besides the first letter being in capital or lower? Does it have different meaning for @sallyarmneale ?

[ricbal02]

LGTM

[rockdreamer]

LGTM otherwise

[rockdreamer]

LGTM