After sysctl vm.idlezero_enable=1, many processes wait for [*vm page queue free mutex] and don't respond. It seems that the vm_pagezero() thread loops in vm_phys_zero_pages_idle() forever with mtx_lock(&vm_page_queue_free_mtx). How-To-Repeat: sysctl vm.idlezero_enable=1 find / -type f -print | xargs grep -l hogehogehoge
Also happens on 10.1. Just reproduced on a FreeBSD/amd64 VM with 256MB RAM. All commands hang as soon as you do a 'sysctl vm.idlezero_enable=1'
The same on 10.1-STABLE r284558M.
For bugs matching the following conditions: - Status == In Progress - Assignee == "bugs@FreeBSD.org" - Last Modified Year <= 2017 Do - Set Status to "Open"
idlezero is no more.