Messages in this thread | | | Date | Fri, 13 Nov 2009 11:35:02 +0900 | From | KAMEZAWA Hiroyuki <> | Subject | Re: [PATCH] show per-process swap usage via procfs v3 |
| |
On Fri, 13 Nov 2009 10:51:12 +0900 KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com> wrote: > > > @@ -597,7 +600,9 @@ copy_one_pte(struct mm_struct *dst_mm, s > > > &src_mm->mmlist); > > > spin_unlock(&mmlist_lock); > > > } > > > - if (is_write_migration_entry(entry) && > > > + if (!non_swap_entry(entry)) > > > + rss[2]++; > > > + else if (is_write_migration_entry(entry) && > > > is_cow_mapping(vm_flags)) { > > > /* > > > > What are the implications for fork performance? > > This path is executed when page table entry contains a entry of > !pte_none() && !pte_present(). > > There are not very big chance to reach here.(this path is under unlikely()). >
[before] text data bss dec hex filename 6649003 3221828 10232816 20103647 132c1df vmlinux [after] text data bss dec hex filename 6649243 3221828 10232816 20103887 132c2cf vmlinux
Now, 240 bytes of text size..Hmm.
Thanks, -Kame
| |