https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=267621
Bug ID: 267621
Summary: panic: page ... already unswappable
Product: Base System
Version: CURRENT
Hardware: Any
OS: Any
Status: New
Severity: Affects Only Me
Priority: ---
Component: kern
Assignee: [email protected]
Reporter: [email protected]
We saw
panic: page ... already unswappable
on a downstream codebase. The page was wired. Another thread in physio() ->
vm_fault_quick_hold_pages() -> vm_page_wire_mapped() had just wired it.
Can we simply relax the assertion? Or should we test for a wired page (with
appropriate synchronization)?
This is trivially reproduced by adding vm_page_wire(mc[0]) at the top of
vm_pageout_flush(), which seems to be a fair repro due to the lack of
synchronization.
--
You are receiving this mail because:
You are the assignee for the bug.
