Messages in this thread | | | Date | Wed, 23 Nov 2011 15:41:11 +0900 | From | Minchan Kim <> | Subject | Re: [PATCH] vmscan: add task name to warn_scan_unevictable() messages |
| |
On Tue, Nov 22, 2011 at 10:32:45PM -0800, David Rientjes wrote: > On Wed, 23 Nov 2011, Minchan Kim wrote: > > > > diff --git a/mm/vmscan.c b/mm/vmscan.c > > > index a1893c0..29d163e 100644 > > > --- a/mm/vmscan.c > > > +++ b/mm/vmscan.c > > > @@ -3448,9 +3448,10 @@ void scan_mapping_unevictable_pages(struct address_space *mapping) > > > static void warn_scan_unevictable_pages(void) > > > { > > > printk_once(KERN_WARNING > > > - "The scan_unevictable_pages sysctl/node-interface has been " > > > + "%s: The scan_unevictable_pages sysctl/node-interface has been " > > > "disabled for lack of a legitimate use case. If you have " > > > - "one, please send an email to linux-mm@kvack.org.\n"); > > > + "one, please send an email to linux-mm@kvack.org.\n", > > > + current->comm); > > > } > > > > Just nitpick: > > How about using WARN_ONCE instead of custom warning? > > It can show more exact call path as well as comm. > > I guess it's more noticible to users. > > Anyway, either is okay to me. > > > > When I used WARN_ONCE() to notify users that /proc/pid/oom_adj was > deprecated, people complained that it triggered userspace log parsers > thinking that there's a serious problem and it adds a taint flag so it got > reverted. I'd recommend keeping it printk_once().
printk_once is better in case of not serious WARNING once I listen your opinion.
Reviewed-by: Minchan Kim <minchan.kim@gmail.com>
-- Kind regards, Minchan Kim
| |