On 07/31/2013 10:17:08 PM, Andrew Morton wrote:
On Wed, 31 Jul 2013 23:11:50 -0400 KOSAKI Motohiro
<kosaki.motohiro@xxxxxxxxxxxxxx> wrote:
> >> --- a/fs/drop_caches.c
> >> +++ b/fs/drop_caches.c
> >> @@ -59,6 +59,8 @@ int drop_caches_sysctl_handler(ctl_table
*table, int write,
> >> if (ret)
> >> return ret;
> >> if (write) {
> >> + printk(KERN_INFO "%s (%d): dropped kernel
caches: %d\n",
> >> + current->comm, task_pid_nr(current),
sysctl_drop_caches);
> >> if (sysctl_drop_caches & 1)
> >> iterate_supers(drop_pagecache_sb, NULL);
> >> if (sysctl_drop_caches & 2)
> >
> > How about we do
> >
> > if (!(sysctl_drop_caches & 4))
> > printk(....)
> >
> > so people can turn it off if it's causing problems?
>
> The best interface depends on the purpose. If you want to detect
crazy application,
> we can't assume an application co-operate us. So, I doubt this
works.
You missed the "!". I'm proposing that setting the new bit 2 will
permit people to prevent the new printk if it is causing them
problems.
Or an alternative for those planning to patch it down to a KERN_DEBUG
locally.
I'd be surprised if anybody who does this sees the printk and thinks
"hey, I'll dig into the VM's balancing logic and come up to speed on
the tradeoffs sufficient to contribute to kernel development" because
of something in dmesg. Anybody actually annoyed by it will chop out the
printk (you barely need to know C to do that), the rest won't notice.
Rob
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@xxxxxxxxx. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href