Date: Sun, 24 Aug 2008 23:52:52 -0700 From: "Rao, Nikhil" <nikhil.rao@intel.com> To: <freebsd-hackers@freebsd.org> Subject: pmap->pm_stats.resident_count Message-ID: <12A5C15467D5B94F8E0FF265D9498ADD038AA0E9@orsmsx419.amr.corp.intel.com> In-Reply-To: <12A5C15467D5B94F8E0FF265D9498ADD0383F1CA@orsmsx419.amr.corp.intel.com> References: <20080818120021.751FB106568E@hub.freebsd.org> <12A5C15467D5B94F8E0FF265D9498ADD0383F1CA@orsmsx419.amr.corp.intel.com>
next in thread | previous in thread | raw e-mail | index | archive | help
vm_daemon() looks at pmap->pm_stats.resident_count in the pmap structure using vmspace_resident_count() which calls pmap_resident_count(), pm_stats.resident_count is also incremented for device memory either if the device memory has been mapped using pmap_object_init_pt or using pmap_enter. Since these pages do not belong to memory that is managed by the kernel, they should not really be accounted in the if check below in vm_pageout.c ? if (limit >=3D 0 && size >=3D limit) { 1593 vm_pageout_map_deactivate_pages( 1594 &p->p_vmspace->vm_map, limit); 1595 } Is my understanding correct ? What do the members of this list think ? Thanks, Nikhil
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?12A5C15467D5B94F8E0FF265D9498ADD038AA0E9>