Skip to content
GitLab
Projects Groups Snippets
  • /
  • Help
    • Help
    • Support
    • Community forum
    • Submit feedback
    • Contribute to GitLab
  • Sign in / Register
  • Iris Iris
  • Project information
    • Project information
    • Activity
    • Labels
    • Members
  • Repository
    • Repository
    • Files
    • Commits
    • Branches
    • Tags
    • Contributors
    • Graph
    • Compare
  • Issues 171
    • Issues 171
    • List
    • Boards
    • Service Desk
    • Milestones
  • Merge requests 15
    • Merge requests 15
  • CI/CD
    • CI/CD
    • Pipelines
    • Jobs
    • Schedules
  • Deployments
    • Deployments
    • Releases
  • Analytics
    • Analytics
    • Value stream
    • CI/CD
    • Repository
  • Wiki
    • Wiki
  • Activity
  • Graph
  • Create a new issue
  • Jobs
  • Commits
  • Issue Boards
Collapse sidebar
  • Iris
  • IrisIris
  • Issues
  • #328
Closed
Open
Issue created Jun 23, 2020 by Ralf Jung@jungOwner

Add RA for auth of a heap

auth (gmap X Y) is a very frequently reoccurring RA, and finding all the right lemmas to compose for it can be challenging. After I saw people use gen_heap when really they want this RA, I am now convinced that we should have it in Iris. ;)

The only open question for me is, what is Y? (X is any countable type.)

  • We probably should have fractions, so that would be Y := frac * agree T. Even if you don't need fractions, just making it always "1" should not be hard to use (we should just make sure to have a lemma that from owning the "1" fraction twice, derives False).
  • @tchajed mentioned they also need something with agreement in a few places. So we could either also have a version with Y := agree T, or we could do the strictly more powerful thing (subsuming both of the above) and do Y := (frac * agree T) + agree T. I think with the right surface-level definitions, this is actually not harder to use than either of the two more specialized heaps.
Assignee
Assign to
Time tracking