Stream: Beginner Questions

Topic: SMT timeout issues


view this post on Zulip Ant S. (May 06 2026 at 12:39):

Sorry, meant to make a new thread

view this post on Zulip Ant S. (May 06 2026 at 12:40):

Ant S. said in #Beginner Questions > Getting value of a configuration parameter:

A follow-up from this thread|

I've had to recently downgrade my machine, and so one SMT query will complete nondeterministically -- sometimes it times out and sometimes it doesn't. This problem is a whole lot more persistent on my weaker machine. Setting declare [[ smt_timeout = 0 ]]does nothing, nor do big numbers like 10000 or -1. What do I do?

view this post on Zulip Mathias Fleury (May 06 2026 at 18:08):

There is an reconstruction_step_timeout which might help

view this post on Zulip Mathias Fleury (May 06 2026 at 18:09):

but you can also call sledgehammer again and try to find another faster proof

view this post on Zulip Ant S. (May 06 2026 at 18:10):

sledgehammer can't :(
But I'll try reconstruction ty

view this post on Zulip Mathias Fleury (May 06 2026 at 18:11):

try replacing the by (smt ....) by sledgehammer(add: ....) or even sledgehammer (...)

view this post on Zulip Ant S. (May 06 2026 at 18:11):

what does that do

view this post on Zulip Ant S. (May 06 2026 at 18:12):

Is that not what putting facts in a using clause is for

view this post on Zulip Mathias Fleury (May 06 2026 at 18:12):

that will force sledgehammer to use that facts (with add:) or only those facts

view this post on Zulip Ant S. (May 06 2026 at 18:12):

oh

view this post on Zulip Ant S. (May 06 2026 at 18:12):

whoa

view this post on Zulip Mathias Fleury (May 06 2026 at 18:12):

otherwise sledgehammer will use its own heuristics to select facts


Last updated: May 17 2026 at 06:46 UTC