Skip to content

Conversation

@simonjwinwood
Copy link

   "typ_name (typ_uinfo_t TYPE(8 word))  = ''word00010''"

and

   "typ_name (typ_uinfo_t TYPE(8 sword))  = ''sword00010''"

Signed-off-by: Simon Winwood <simonjwinwood@gmail.com>
@lsf37
Copy link
Member

lsf37 commented Jun 26, 2024

Can you please provide motivation for the change in the description so the we can figure out in the future why we did this?

The C proofs all seem to be fine, which is a good sign, the errors are all in AsmRefine. @mbrcknl might know more about these.

@simonjwinwood
Copy link
Author

So this is no longer true:

lemma typ_uinfo_t_signed_word:

    192 lemma typ_uinfo_t_signed_word:
    193   "typ_uinfo_t TYPE (('a :: len8) signed word) = typ_uinfo_t TYPE ('a word)"

@lsf37
Copy link
Member

lsf37 commented Jun 27, 2024

This might actually not be needed. It's used to prove the align_td lemma underneath, which should still be true, and then the ptr_inverse_safe lemma which I have no idea what it is. The definition talks about footprints, though, which should also still be the same. So both should hold.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants