cfi: fix ordering of textual SSPUSH behavior, clarify ordering.

[dtzSiFive]

No description provided.

[github-actions]

Normative Rule Changes Detected

This PR modifies normatively tagged text. Please review the changes below to ensure they are intentional.

View Detected Changes

Normative Tag Change Report

riscv-spec Specification

================================================================================
Tag Changes Report
================================================================================

Reference file: ref/riscv-spec-norm-tags.json
Current file: build/riscv-spec-norm-tags.json
Modified 2 tags:
  * "norm:ss_push":
      Reference: "A shadow stack push operation is defined as decrement of the ssp by XLEN/8 followed by a store of th..."
      Current:   "A shadow stack push operation is defined as a store of the value in the link register to memory at o..."
  * "norm:sspush_c-sspush_decrement":
      Reference: "The ssp is decremented by SSPUSH and C.SSPUSH only if the store to the shadow stack completes succes..."
      Current:   "The ssp is decremented by SSPUSH and C.SSPUSH only if it is determined that the store to the shadow ..."

================================================================================
Summary: 2 total changes
  Added:    0
  Deleted:  0
  Modified: 2
================================================================================

---

What happens next:

• This comment is informational only and does not block merging
• When this PR is merged, a GitHub issue will be automatically created with the NormRules label for CSC tracking
• If these changes are unintentional, please update the PR before merging

How to update reference files (if needed):

make update-ref
git add ref/*.json
git commit -m "Update normative tag reference files"

Note: New tags (additions) are automatically added to the reference files when PRs are merged to main. Only modifications and deletions require manual review.

This comment was automatically generated by the normative tag check workflow.

[aswaterman]

I don't think this actually changes the memory ordering implications. Either way, the store depends on ssp. It doesn't really matter whether it depends on the incremented value or the original value.

[dtzSiFive]

The main change intent is to not indicate ssp is decremented first "followed by" the store, when the code listing has them in the other order.

But okay I see what you mean, if the update to ssp can be said to precede the store but doesn't commit until after regardless this change is unnecessary.

[aswaterman]

I think this is true, but if this follows from RVWMO's handling of implicit CSR accesses (which I think it does, but I haven't confirmed it), then I'd rather not add any new text. I'll leave this open for now.

[dtzSiFive]

Aye, I'll check and agreed. Thanks!