Date: Sat, 17 Apr 2010 18:35:07 +0000 (UTC) From: Alan Cox <alc@FreeBSD.org> To: src-committers@freebsd.org, svn-src-all@freebsd.org, svn-src-head@freebsd.org Subject: svn commit: r206768 - head/sys/vm Message-ID: <201004171835.o3HIZ7i3037483@svn.freebsd.org>
next in thread | raw e-mail | index | archive | help
Author: alc Date: Sat Apr 17 18:35:07 2010 New Revision: 206768 URL: http://svn.freebsd.org/changeset/base/206768 Log: In vm_object_backing_scan(), setting PG_REFERENCED on a page before sleeping on that page is nonsensical. Doing so reduces the likelihood that the page daemon will reclaim the page before the thread waiting in vm_object_backing_scan() is reawakened. However, it does not guarantee that the page is not reclaimed, so vm_object_backing_scan() restarts after reawakening. More importantly, this muddles the meaning of PG_REFERENCED. There is no reason to believe that the caller of vm_object_backing_scan() is going to use (i.e., access) the contents of the page. There is especially no reason to believe that an access is more likely because vm_object_backing_scan() had to sleep on the page. Discussed with: kib MFC after: 3 weeks Modified: head/sys/vm/vm_object.c Modified: head/sys/vm/vm_object.c ============================================================================== --- head/sys/vm/vm_object.c Sat Apr 17 18:18:46 2010 (r206767) +++ head/sys/vm/vm_object.c Sat Apr 17 18:35:07 2010 (r206768) @@ -1553,9 +1553,6 @@ vm_object_backing_scan(vm_object_t objec } } else if (op & OBSC_COLLAPSE_WAIT) { if ((p->oflags & VPO_BUSY) || p->busy) { - vm_page_lock_queues(); - vm_page_flag_set(p, PG_REFERENCED); - vm_page_unlock_queues(); VM_OBJECT_UNLOCK(object); p->oflags |= VPO_WANTED; msleep(p, VM_OBJECT_MTX(backing_object),
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?201004171835.o3HIZ7i3037483>