Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Inconsistent use of "most precise answer" on page 47 #47

Closed
mdempsky opened this issue May 14, 2020 · 0 comments
Closed

Inconsistent use of "most precise answer" on page 47 #47

mdempsky opened this issue May 14, 2020 · 0 comments

Comments

@mdempsky
Copy link
Contributor

At the end of section 8.2, it's claimed that "n -> Top" is the most precise answer for analyzing the doubling code. This is true if we're talking about abstract states valid at any point during execution, as n can be odd within the loop. But at termination, n will always be 0, so the most precise answer would be "n -> Even".

On the other hand, section 8.3 opens talking about how "x -> Odd" is the most precise answer for "x <- 0; x <- 1". But this only a sound answer if we're talking about analysis at termination, as opposed to any point during execution.

One way to resolve this inconsistency would be to qualify at the end of section 8.2 that flow-insensitive analysis "gives the most precise answer for x".

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

No branches or pull requests

1 participant