All of lore.kernel.org
 help / color / mirror / Atom feed
* [-] drop_caches-add-some-documentation-and-info-messsge.patch removed from -mm tree
@ 2013-07-10 20:25 akpm
  2013-07-11  7:36 ` Michal Hocko
  0 siblings, 1 reply; 10+ messages in thread
From: akpm @ 2013-07-10 20:25 UTC (permalink / raw)
  To: mm-commits, kosaki.motohiro, kamezawa.hiroyu, dave, mhocko

Subject: [-] drop_caches-add-some-documentation-and-info-messsge.patch removed from -mm tree
To: mhocko@suse.cz,dave@linux.vnet.ibm.com,kamezawa.hiroyu@jp.fujitsu.com,kosaki.motohiro@jp.fujitsu.com,mm-commits@vger.kernel.org
From: akpm@linux-foundation.org
Date: Wed, 10 Jul 2013 13:25:03 -0700


The patch titled
     Subject: drop_caches: add some documentation and info message
has been removed from the -mm tree.  Its filename was
     drop_caches-add-some-documentation-and-info-messsge.patch

This patch was dropped because it has gone stale

------------------------------------------------------
From: Michal Hocko <mhocko@suse.cz>
Subject: drop_caches: add some documentation and info message

I would like to resurrect Dave's patch.  The last time it was posted was
here https://lkml.org/lkml/2010/9/16/250 and there didn't seem to be any
strong opposition.

Kosaki was worried about possible excessive logging when somebody drops
caches too often (but then he claimed he didn't have a strong opinion on
that) but I would say opposite.  If somebody does that then I would really
like to know that from the log when supporting a system because it almost
for sure means that there is something fishy going on.  It is also worth
mentioning that only root can write drop caches so this is not an flooding
attack vector.

I am bringing that up again because this can be really helpful when
chasing strange performance issues which (surprise surprise) turn out to
be related to artificially dropped caches done because the admin thinks
this would help...

I have just refreshed the original patch on top of the current mm tree
but I could live with KERN_INFO as well if people think that KERN_NOTICE
is too hysterical.

: From: Dave Hansen <dave@linux.vnet.ibm.com>
: Date: Fri, 12 Oct 2012 14:30:54 +0200
: 
: There is plenty of anecdotal evidence and a load of blog posts
: suggesting that using "drop_caches" periodically keeps your system
: running in "tip top shape".  Perhaps adding some kernel
: documentation will increase the amount of accurate data on its use.
: 
: If we are not shrinking caches effectively, then we have real bugs.
: Using drop_caches will simply mask the bugs and make them harder
: to find, but certainly does not fix them, nor is it an appropriate
: "workaround" to limit the size of the caches.
: 
: It's a great debugging tool, and is really handy for doing things
: like repeatable benchmark runs.  So, add a bit more documentation
: about it, and add a little KERN_NOTICE.  It should help developers
: who are chasing down reclaim-related bugs.

[mhocko@suse.cz: refreshed to current -mm tree]
[akpm@linux-foundation.org: checkpatch fixes]
Signed-off-by: Dave Hansen <dave@linux.vnet.ibm.com>
Reviewed-by: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Signed-off-by: Michal Hocko <mhocko@suse.cz>
Acked-by: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Acked-by: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
---

 Documentation/sysctl/vm.txt |   33 +++++++++++++++++++++++++++------
 fs/drop_caches.c            |    2 ++
 2 files changed, 29 insertions(+), 6 deletions(-)

diff -puN Documentation/sysctl/vm.txt~drop_caches-add-some-documentation-and-info-messsge Documentation/sysctl/vm.txt
--- a/Documentation/sysctl/vm.txt~drop_caches-add-some-documentation-and-info-messsge
+++ a/Documentation/sysctl/vm.txt
@@ -169,18 +169,39 @@ Setting this to zero disables periodic w
 
 drop_caches
 
-Writing to this will cause the kernel to drop clean caches, dentries and
-inodes from memory, causing that memory to become free.
+Writing to this will cause the kernel to drop clean caches, as well as
+reclaimable slab objects like dentries and inodes.  Once dropped, their
+memory becomes free.
 
 To free pagecache:
 	echo 1 > /proc/sys/vm/drop_caches
-To free dentries and inodes:
+To free reclaimable slab objects (includes dentries and inodes):
 	echo 2 > /proc/sys/vm/drop_caches
-To free pagecache, dentries and inodes:
+To free slab objects and pagecache:
 	echo 3 > /proc/sys/vm/drop_caches
 
-As this is a non-destructive operation and dirty objects are not freeable, the
-user should run `sync' first.
+This is a non-destructive operation and will not free any dirty objects.
+To increase the number of objects freed by this operation, the user may run
+`sync' prior to writing to /proc/sys/vm/drop_caches.  This will minimize the
+number of dirty objects on the system and create more candidates to be
+dropped.
+
+This file is not a means to control the growth of the various kernel caches
+(inodes, dentries, pagecache, etc...)  These objects are automatically
+reclaimed by the kernel when memory is needed elsewhere on the system.
+
+Use of this file can cause performance problems.  Since it discards cached
+objects, it may cost a significant amount of I/O and CPU to recreate the
+dropped objects, especially if they were under heavy use.  Because of this,
+use outside of a testing or debugging environment is not recommended.
+
+You may see informational messages in your kernel log when this file is
+used:
+
+	cat (1234): dropped kernel caches: 3
+
+These are informational only.  They do not mean that anything is wrong
+with your system.
 
 ==============================================================
 
diff -puN fs/drop_caches.c~drop_caches-add-some-documentation-and-info-messsge fs/drop_caches.c
--- a/fs/drop_caches.c~drop_caches-add-some-documentation-and-info-messsge
+++ a/fs/drop_caches.c
@@ -59,6 +59,8 @@ int drop_caches_sysctl_handler(ctl_table
 	if (ret)
 		return ret;
 	if (write) {
+		printk(KERN_NOTICE "%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)
_

Patches currently in -mm which might be from mhocko@suse.cz are

origin.patch
linux-next.patch
include-linux-schedh-dont-use-task-pid-tgid-in-same_thread_group-has_group_leader_pid.patch
inode-convert-inode-lru-list-to-generic-lru-list-code-inode-move-inode-to-a-different-list-inside-lock.patch
list_lru-per-node-list-infrastructure-fix-broken-lru_retry-behaviour.patch
list_lru-remove-special-case-function-list_lru_dispose_all.patch
xfs-convert-dquot-cache-lru-to-list_lru-fix-dquot-isolation-hang.patch
list_lru-dynamically-adjust-node-arrays-super-fix-for-destroy-lrus.patch
staging-lustre-ldlm-convert-to-shrinkers-to-count-scan-api.patch
staging-lustre-obdclass-convert-lu_object-shrinker-to-count-scan-api.patch
staging-lustre-ptlrpc-convert-to-new-shrinker-api.patch
staging-lustre-libcfs-cleanup-linux-memh.patch
staging-lustre-replace-num_physpages-with-totalram_pages.patch


^ permalink raw reply	[flat|nested] 10+ messages in thread

* Re: [-] drop_caches-add-some-documentation-and-info-messsge.patch removed from -mm tree
  2013-07-10 20:25 [-] drop_caches-add-some-documentation-and-info-messsge.patch removed from -mm tree akpm
@ 2013-07-11  7:36 ` Michal Hocko
  2013-07-11 12:39   ` Michal Hocko
  2013-07-11 21:50   ` Andrew Morton
  0 siblings, 2 replies; 10+ messages in thread
From: Michal Hocko @ 2013-07-11  7:36 UTC (permalink / raw)
  To: akpm; +Cc: mm-commits, kosaki.motohiro, kamezawa.hiroyu, dave, linux-mm

Hi Andrew,

On Wed 10-07-13 13:25:03, Andrew Morton wrote:
[...]
> This patch was dropped because it has gone stale

Is there really a strong reason to not take this patch? It turned out
being useful for us. "drop_caches will solve your performance problems"
cargo cult is still alive.

I would turn this into a trace point but that would be much weaker
because the one who is debugging an issue would have to think about
enabling it before the affected workload starts. Which is not possible
quite often. Having logs and looking at them afterwards is so
_convinient_.

The code impact is really small as well and systems which should receive
as few message as possible shouldn't run at loglevel as high as
KERN_NOTICE anyway.

So what makes you hate this so much?

FWIW we are having this patch in our enterprise servers for the above
mentioned reasons but I think it might be useful for others as well. If
you think that this debugging aid is not worth it I can live with that
and keep it out of tree in our kernels.

> From: Michal Hocko <mhocko@suse.cz>
> Subject: drop_caches: add some documentation and info message
> 
> I would like to resurrect Dave's patch.  The last time it was posted was
> here https://lkml.org/lkml/2010/9/16/250 and there didn't seem to be any
> strong opposition.
> 
> Kosaki was worried about possible excessive logging when somebody drops
> caches too often (but then he claimed he didn't have a strong opinion on
> that) but I would say opposite.  If somebody does that then I would really
> like to know that from the log when supporting a system because it almost
> for sure means that there is something fishy going on.  It is also worth
> mentioning that only root can write drop caches so this is not an flooding
> attack vector.
> 
> I am bringing that up again because this can be really helpful when
> chasing strange performance issues which (surprise surprise) turn out to
> be related to artificially dropped caches done because the admin thinks
> this would help...
> 
> I have just refreshed the original patch on top of the current mm tree
> but I could live with KERN_INFO as well if people think that KERN_NOTICE
> is too hysterical.
> 
> : From: Dave Hansen <dave@linux.vnet.ibm.com>
> : Date: Fri, 12 Oct 2012 14:30:54 +0200
> : 
> : There is plenty of anecdotal evidence and a load of blog posts
> : suggesting that using "drop_caches" periodically keeps your system
> : running in "tip top shape".  Perhaps adding some kernel
> : documentation will increase the amount of accurate data on its use.
> : 
> : If we are not shrinking caches effectively, then we have real bugs.
> : Using drop_caches will simply mask the bugs and make them harder
> : to find, but certainly does not fix them, nor is it an appropriate
> : "workaround" to limit the size of the caches.
> : 
> : It's a great debugging tool, and is really handy for doing things
> : like repeatable benchmark runs.  So, add a bit more documentation
> : about it, and add a little KERN_NOTICE.  It should help developers
> : who are chasing down reclaim-related bugs.
> 
> [mhocko@suse.cz: refreshed to current -mm tree]
> [akpm@linux-foundation.org: checkpatch fixes]
> Signed-off-by: Dave Hansen <dave@linux.vnet.ibm.com>
> Reviewed-by: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
> Signed-off-by: Michal Hocko <mhocko@suse.cz>
> Acked-by: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
> Acked-by: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
> ---
> 
>  Documentation/sysctl/vm.txt |   33 +++++++++++++++++++++++++++------
>  fs/drop_caches.c            |    2 ++
>  2 files changed, 29 insertions(+), 6 deletions(-)
> 
> diff -puN Documentation/sysctl/vm.txt~drop_caches-add-some-documentation-and-info-messsge Documentation/sysctl/vm.txt
> --- a/Documentation/sysctl/vm.txt~drop_caches-add-some-documentation-and-info-messsge
> +++ a/Documentation/sysctl/vm.txt
> @@ -169,18 +169,39 @@ Setting this to zero disables periodic w
>  
>  drop_caches
>  
> -Writing to this will cause the kernel to drop clean caches, dentries and
> -inodes from memory, causing that memory to become free.
> +Writing to this will cause the kernel to drop clean caches, as well as
> +reclaimable slab objects like dentries and inodes.  Once dropped, their
> +memory becomes free.
>  
>  To free pagecache:
>  	echo 1 > /proc/sys/vm/drop_caches
> -To free dentries and inodes:
> +To free reclaimable slab objects (includes dentries and inodes):
>  	echo 2 > /proc/sys/vm/drop_caches
> -To free pagecache, dentries and inodes:
> +To free slab objects and pagecache:
>  	echo 3 > /proc/sys/vm/drop_caches
>  
> -As this is a non-destructive operation and dirty objects are not freeable, the
> -user should run `sync' first.
> +This is a non-destructive operation and will not free any dirty objects.
> +To increase the number of objects freed by this operation, the user may run
> +`sync' prior to writing to /proc/sys/vm/drop_caches.  This will minimize the
> +number of dirty objects on the system and create more candidates to be
> +dropped.
> +
> +This file is not a means to control the growth of the various kernel caches
> +(inodes, dentries, pagecache, etc...)  These objects are automatically
> +reclaimed by the kernel when memory is needed elsewhere on the system.
> +
> +Use of this file can cause performance problems.  Since it discards cached
> +objects, it may cost a significant amount of I/O and CPU to recreate the
> +dropped objects, especially if they were under heavy use.  Because of this,
> +use outside of a testing or debugging environment is not recommended.
> +
> +You may see informational messages in your kernel log when this file is
> +used:
> +
> +	cat (1234): dropped kernel caches: 3
> +
> +These are informational only.  They do not mean that anything is wrong
> +with your system.
>  
>  ==============================================================
>  
> diff -puN fs/drop_caches.c~drop_caches-add-some-documentation-and-info-messsge fs/drop_caches.c
> --- a/fs/drop_caches.c~drop_caches-add-some-documentation-and-info-messsge
> +++ a/fs/drop_caches.c
> @@ -59,6 +59,8 @@ int drop_caches_sysctl_handler(ctl_table
>  	if (ret)
>  		return ret;
>  	if (write) {
> +		printk(KERN_NOTICE "%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)
> _
> 
> Patches currently in -mm which might be from mhocko@suse.cz are
> 
> origin.patch
> linux-next.patch
> include-linux-schedh-dont-use-task-pid-tgid-in-same_thread_group-has_group_leader_pid.patch
> inode-convert-inode-lru-list-to-generic-lru-list-code-inode-move-inode-to-a-different-list-inside-lock.patch
> list_lru-per-node-list-infrastructure-fix-broken-lru_retry-behaviour.patch
> list_lru-remove-special-case-function-list_lru_dispose_all.patch
> xfs-convert-dquot-cache-lru-to-list_lru-fix-dquot-isolation-hang.patch
> list_lru-dynamically-adjust-node-arrays-super-fix-for-destroy-lrus.patch
> staging-lustre-ldlm-convert-to-shrinkers-to-count-scan-api.patch
> staging-lustre-obdclass-convert-lu_object-shrinker-to-count-scan-api.patch
> staging-lustre-ptlrpc-convert-to-new-shrinker-api.patch
> staging-lustre-libcfs-cleanup-linux-memh.patch
> staging-lustre-replace-num_physpages-with-totalram_pages.patch
> 

-- 
Michal Hocko
SUSE Labs

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org.  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>

^ permalink raw reply	[flat|nested] 10+ messages in thread

* Re: [-] drop_caches-add-some-documentation-and-info-messsge.patch removed from -mm tree
  2013-07-11  7:36 ` Michal Hocko
@ 2013-07-11 12:39   ` Michal Hocko
  2013-07-11 15:34     ` Dave Hansen
  2013-07-11 21:50   ` Andrew Morton
  1 sibling, 1 reply; 10+ messages in thread
From: Michal Hocko @ 2013-07-11 12:39 UTC (permalink / raw)
  To: akpm; +Cc: mm-commits, kosaki.motohiro, kamezawa.hiroyu, dave.hansen, linux-mm

[fix up Dave's email]

On Thu 11-07-13 09:36:44, Michal Hocko wrote:
> Hi Andrew,
> 
> On Wed 10-07-13 13:25:03, Andrew Morton wrote:
> [...]
> > This patch was dropped because it has gone stale
> 
> Is there really a strong reason to not take this patch? It turned out
> being useful for us. "drop_caches will solve your performance problems"
> cargo cult is still alive.
> 
> I would turn this into a trace point but that would be much weaker
> because the one who is debugging an issue would have to think about
> enabling it before the affected workload starts. Which is not possible
> quite often. Having logs and looking at them afterwards is so
> _convinient_.
> 
> The code impact is really small as well and systems which should receive
> as few message as possible shouldn't run at loglevel as high as
> KERN_NOTICE anyway.
> 
> So what makes you hate this so much?
> 
> FWIW we are having this patch in our enterprise servers for the above
> mentioned reasons but I think it might be useful for others as well. If
> you think that this debugging aid is not worth it I can live with that
> and keep it out of tree in our kernels.
> 
> > From: Michal Hocko <mhocko@suse.cz>
> > Subject: drop_caches: add some documentation and info message
> > 
> > I would like to resurrect Dave's patch.  The last time it was posted was
> > here https://lkml.org/lkml/2010/9/16/250 and there didn't seem to be any
> > strong opposition.
> > 
> > Kosaki was worried about possible excessive logging when somebody drops
> > caches too often (but then he claimed he didn't have a strong opinion on
> > that) but I would say opposite.  If somebody does that then I would really
> > like to know that from the log when supporting a system because it almost
> > for sure means that there is something fishy going on.  It is also worth
> > mentioning that only root can write drop caches so this is not an flooding
> > attack vector.
> > 
> > I am bringing that up again because this can be really helpful when
> > chasing strange performance issues which (surprise surprise) turn out to
> > be related to artificially dropped caches done because the admin thinks
> > this would help...
> > 
> > I have just refreshed the original patch on top of the current mm tree
> > but I could live with KERN_INFO as well if people think that KERN_NOTICE
> > is too hysterical.
> > 
> > : From: Dave Hansen <dave@linux.vnet.ibm.com>
> > : Date: Fri, 12 Oct 2012 14:30:54 +0200
> > : 
> > : There is plenty of anecdotal evidence and a load of blog posts
> > : suggesting that using "drop_caches" periodically keeps your system
> > : running in "tip top shape".  Perhaps adding some kernel
> > : documentation will increase the amount of accurate data on its use.
> > : 
> > : If we are not shrinking caches effectively, then we have real bugs.
> > : Using drop_caches will simply mask the bugs and make them harder
> > : to find, but certainly does not fix them, nor is it an appropriate
> > : "workaround" to limit the size of the caches.
> > : 
> > : It's a great debugging tool, and is really handy for doing things
> > : like repeatable benchmark runs.  So, add a bit more documentation
> > : about it, and add a little KERN_NOTICE.  It should help developers
> > : who are chasing down reclaim-related bugs.
> > 
> > [mhocko@suse.cz: refreshed to current -mm tree]
> > [akpm@linux-foundation.org: checkpatch fixes]
> > Signed-off-by: Dave Hansen <dave@linux.vnet.ibm.com>
> > Reviewed-by: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
> > Signed-off-by: Michal Hocko <mhocko@suse.cz>
> > Acked-by: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
> > Acked-by: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
> > Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
> > ---
> > 
> >  Documentation/sysctl/vm.txt |   33 +++++++++++++++++++++++++++------
> >  fs/drop_caches.c            |    2 ++
> >  2 files changed, 29 insertions(+), 6 deletions(-)
> > 
> > diff -puN Documentation/sysctl/vm.txt~drop_caches-add-some-documentation-and-info-messsge Documentation/sysctl/vm.txt
> > --- a/Documentation/sysctl/vm.txt~drop_caches-add-some-documentation-and-info-messsge
> > +++ a/Documentation/sysctl/vm.txt
> > @@ -169,18 +169,39 @@ Setting this to zero disables periodic w
> >  
> >  drop_caches
> >  
> > -Writing to this will cause the kernel to drop clean caches, dentries and
> > -inodes from memory, causing that memory to become free.
> > +Writing to this will cause the kernel to drop clean caches, as well as
> > +reclaimable slab objects like dentries and inodes.  Once dropped, their
> > +memory becomes free.
> >  
> >  To free pagecache:
> >  	echo 1 > /proc/sys/vm/drop_caches
> > -To free dentries and inodes:
> > +To free reclaimable slab objects (includes dentries and inodes):
> >  	echo 2 > /proc/sys/vm/drop_caches
> > -To free pagecache, dentries and inodes:
> > +To free slab objects and pagecache:
> >  	echo 3 > /proc/sys/vm/drop_caches
> >  
> > -As this is a non-destructive operation and dirty objects are not freeable, the
> > -user should run `sync' first.
> > +This is a non-destructive operation and will not free any dirty objects.
> > +To increase the number of objects freed by this operation, the user may run
> > +`sync' prior to writing to /proc/sys/vm/drop_caches.  This will minimize the
> > +number of dirty objects on the system and create more candidates to be
> > +dropped.
> > +
> > +This file is not a means to control the growth of the various kernel caches
> > +(inodes, dentries, pagecache, etc...)  These objects are automatically
> > +reclaimed by the kernel when memory is needed elsewhere on the system.
> > +
> > +Use of this file can cause performance problems.  Since it discards cached
> > +objects, it may cost a significant amount of I/O and CPU to recreate the
> > +dropped objects, especially if they were under heavy use.  Because of this,
> > +use outside of a testing or debugging environment is not recommended.
> > +
> > +You may see informational messages in your kernel log when this file is
> > +used:
> > +
> > +	cat (1234): dropped kernel caches: 3
> > +
> > +These are informational only.  They do not mean that anything is wrong
> > +with your system.
> >  
> >  ==============================================================
> >  
> > diff -puN fs/drop_caches.c~drop_caches-add-some-documentation-and-info-messsge fs/drop_caches.c
> > --- a/fs/drop_caches.c~drop_caches-add-some-documentation-and-info-messsge
> > +++ a/fs/drop_caches.c
> > @@ -59,6 +59,8 @@ int drop_caches_sysctl_handler(ctl_table
> >  	if (ret)
> >  		return ret;
> >  	if (write) {
> > +		printk(KERN_NOTICE "%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)
> > _
> > 
> > Patches currently in -mm which might be from mhocko@suse.cz are
> > 
> > origin.patch
> > linux-next.patch
> > include-linux-schedh-dont-use-task-pid-tgid-in-same_thread_group-has_group_leader_pid.patch
> > inode-convert-inode-lru-list-to-generic-lru-list-code-inode-move-inode-to-a-different-list-inside-lock.patch
> > list_lru-per-node-list-infrastructure-fix-broken-lru_retry-behaviour.patch
> > list_lru-remove-special-case-function-list_lru_dispose_all.patch
> > xfs-convert-dquot-cache-lru-to-list_lru-fix-dquot-isolation-hang.patch
> > list_lru-dynamically-adjust-node-arrays-super-fix-for-destroy-lrus.patch
> > staging-lustre-ldlm-convert-to-shrinkers-to-count-scan-api.patch
> > staging-lustre-obdclass-convert-lu_object-shrinker-to-count-scan-api.patch
> > staging-lustre-ptlrpc-convert-to-new-shrinker-api.patch
> > staging-lustre-libcfs-cleanup-linux-memh.patch
> > staging-lustre-replace-num_physpages-with-totalram_pages.patch
> > 
> 
> -- 
> Michal Hocko
> SUSE Labs
> 
> --
> To unsubscribe, send a message with 'unsubscribe linux-mm' in
> the body to majordomo@kvack.org.  For more info on Linux MM,
> see: http://www.linux-mm.org/ .
> Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>

-- 
Michal Hocko
SUSE Labs

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org.  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>

^ permalink raw reply	[flat|nested] 10+ messages in thread

* Re: [-] drop_caches-add-some-documentation-and-info-messsge.patch removed from -mm tree
  2013-07-11 12:39   ` Michal Hocko
@ 2013-07-11 15:34     ` Dave Hansen
  2013-07-11 15:42       ` Michal Hocko
  0 siblings, 1 reply; 10+ messages in thread
From: Dave Hansen @ 2013-07-11 15:34 UTC (permalink / raw)
  To: Michal Hocko; +Cc: akpm, kosaki.motohiro, kamezawa.hiroyu, linux-mm

On 07/11/2013 05:39 AM, Michal Hocko wrote:
>> I would turn this into a trace point but that would be much weaker
>> because the one who is debugging an issue would have to think about
>> enabling it before the affected workload starts. Which is not possible
>> quite often. Having logs and looking at them afterwards is so
>> _convinient_.

It would also be a lot weaker than the printk, but we could always add a
counter for this stuff and at least dump it out in /proc/vmstat.  We
wouldn't know who was doing it, but we'd at least know someone _was_
doing it.  It would also have a decent chance of getting picked up by
existing log collection systems.

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org.  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>

^ permalink raw reply	[flat|nested] 10+ messages in thread

* Re: [-] drop_caches-add-some-documentation-and-info-messsge.patch removed from -mm tree
  2013-07-11 15:34     ` Dave Hansen
@ 2013-07-11 15:42       ` Michal Hocko
  2013-07-11 15:45         ` Dave Hansen
  0 siblings, 1 reply; 10+ messages in thread
From: Michal Hocko @ 2013-07-11 15:42 UTC (permalink / raw)
  To: Dave Hansen; +Cc: akpm, kosaki.motohiro, kamezawa.hiroyu, linux-mm

On Thu 11-07-13 08:34:45, Dave Hansen wrote:
> On 07/11/2013 05:39 AM, Michal Hocko wrote:
> >> I would turn this into a trace point but that would be much weaker
> >> because the one who is debugging an issue would have to think about
> >> enabling it before the affected workload starts. Which is not possible
> >> quite often. Having logs and looking at them afterwards is so
> >> _convinient_.
> 
> It would also be a lot weaker than the printk, but we could always add a
> counter for this stuff and at least dump it out in /proc/vmstat.  We
> wouldn't know who was doing it, but we'd at least know someone _was_
> doing it.  It would also have a decent chance of getting picked up by
> existing log collection systems.

But wouldn't be a counter more intrusive code wise? Dunno, but printk
serves it purpose and it doesn't add much to the code.

-- 
Michal Hocko
SUSE Labs

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org.  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>

^ permalink raw reply	[flat|nested] 10+ messages in thread

* Re: [-] drop_caches-add-some-documentation-and-info-messsge.patch removed from -mm tree
  2013-07-11 15:42       ` Michal Hocko
@ 2013-07-11 15:45         ` Dave Hansen
  0 siblings, 0 replies; 10+ messages in thread
From: Dave Hansen @ 2013-07-11 15:45 UTC (permalink / raw)
  To: Michal Hocko; +Cc: akpm, kosaki.motohiro, kamezawa.hiroyu, linux-mm

On 07/11/2013 08:42 AM, Michal Hocko wrote:
> On Thu 11-07-13 08:34:45, Dave Hansen wrote:
>> On 07/11/2013 05:39 AM, Michal Hocko wrote:
>>>> I would turn this into a trace point but that would be much weaker
>>>> because the one who is debugging an issue would have to think about
>>>> enabling it before the affected workload starts. Which is not possible
>>>> quite often. Having logs and looking at them afterwards is so
>>>> _convinient_.
>>
>> It would also be a lot weaker than the printk, but we could always add a
>> counter for this stuff and at least dump it out in /proc/vmstat.  We
>> wouldn't know who was doing it, but we'd at least know someone _was_
>> doing it.  It would also have a decent chance of getting picked up by
>> existing log collection systems.
> 
> But wouldn't be a counter more intrusive code wise? Dunno, but printk
> serves it purpose and it doesn't add much to the code.

Yeah, I prefer the printk too.  I'd rather see a vmstat entry than a
tracepoint, though.

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org.  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>

^ permalink raw reply	[flat|nested] 10+ messages in thread

* Re: [-] drop_caches-add-some-documentation-and-info-messsge.patch removed from -mm tree
  2013-07-11  7:36 ` Michal Hocko
  2013-07-11 12:39   ` Michal Hocko
@ 2013-07-11 21:50   ` Andrew Morton
  2013-07-12 11:50     ` Michal Hocko
  1 sibling, 1 reply; 10+ messages in thread
From: Andrew Morton @ 2013-07-11 21:50 UTC (permalink / raw)
  To: Michal Hocko; +Cc: kosaki.motohiro, kamezawa.hiroyu, dave, linux-mm

On Thu, 11 Jul 2013 09:36:44 +0200 Michal Hocko <mhocko@suse.cz> wrote:

> On Wed 10-07-13 13:25:03, Andrew Morton wrote:
> [...]
> > This patch was dropped because it has gone stale
> 
> Is there really a strong reason to not take this patch? 

I flushed out a whole bunch of MM patches which had been floating
around in indecisive limbo.

I don't recall all the review issues surrounding this one.  If you
think the patch is still good, please resend and ensure that the
changelog adequately addresses all the issues which were raised, so we
don't just take another trip around the loop.

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org.  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>

^ permalink raw reply	[flat|nested] 10+ messages in thread

* Re: [-] drop_caches-add-some-documentation-and-info-messsge.patch removed from -mm tree
  2013-07-11 21:50   ` Andrew Morton
@ 2013-07-12 11:50     ` Michal Hocko
  2013-07-12 12:25       ` Borislav Petkov
  2013-07-12 19:08       ` KOSAKI Motohiro
  0 siblings, 2 replies; 10+ messages in thread
From: Michal Hocko @ 2013-07-12 11:50 UTC (permalink / raw)
  To: Andrew Morton
  Cc: kosaki.motohiro, kamezawa.hiroyu, linux-mm, Borislav Petkov, Dave Hansen

On Thu 11-07-13 14:50:34, Andrew Morton wrote:
> On Thu, 11 Jul 2013 09:36:44 +0200 Michal Hocko <mhocko@suse.cz> wrote:
> 
> > On Wed 10-07-13 13:25:03, Andrew Morton wrote:
> > [...]
> > > This patch was dropped because it has gone stale
> > 
> > Is there really a strong reason to not take this patch? 
> 
> I flushed out a whole bunch of MM patches which had been floating
> around in indecisive limbo.
> 
> I don't recall all the review issues surrounding this one.

Kosaki was concerned about annoying number of messages if somebody drops
caches too often (https://lkml.org/lkml/2010/9/20/450). As I noted in
the changelog
"
    Kosaki was worried about possible excessive logging when somebody drops
    caches too often (but then he claimed he didn't have a strong opinion on
    that) but I would say opposite.  If somebody does that then I would really
    like to know that from the log when supporting a system because it almost
    for sure means that there is something fishy going on.  It is also worth
    mentioning that only root can write drop caches so this is not an flooding
    attack vector.
"

Kosaki then Acked the patch.

You were worried (http://lkml.indiana.edu/hypermail/linux/kernel/1210.3/00605.html)
about people hating us because they are using this as a solution to
their issues. I concur that most of those are just hacks that found
their way into scripts looong time agon and stayed there.

Boris then noted (http://lkml.indiana.edu/hypermail/linux/kernel/1210.3/00659.html)
that he is using drop_caches to make s2ram faster but as others noted
this just adds the overhead to the resume path so it might work only for
certain use cases so a user space solution is more appropriate and
Boris' use case really sounds valid.

As a compromise I can lower the log level. Would KERN_INFO work for
you? Or even KERN_DEBUG?

I still find printk less intrusive than fiddling with vmstat counters.
-- 
Michal Hocko
SUSE Labs

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org.  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>

^ permalink raw reply	[flat|nested] 10+ messages in thread

* Re: [-] drop_caches-add-some-documentation-and-info-messsge.patch removed from -mm tree
  2013-07-12 11:50     ` Michal Hocko
@ 2013-07-12 12:25       ` Borislav Petkov
  2013-07-12 19:08       ` KOSAKI Motohiro
  1 sibling, 0 replies; 10+ messages in thread
From: Borislav Petkov @ 2013-07-12 12:25 UTC (permalink / raw)
  To: Michal Hocko
  Cc: Andrew Morton, kosaki.motohiro, kamezawa.hiroyu, linux-mm,
	Borislav Petkov, Dave Hansen

On Fri, Jul 12, 2013 at 01:50:28PM +0200, Michal Hocko wrote:
> Boris then noted
> (http://lkml.indiana.edu/hypermail/linux/kernel/1210.3/00659.html)
> that he is using drop_caches to make s2ram faster but as others noted
> this just adds the overhead to the resume path so it might work only
> for certain use cases so a user space solution is more appropriate and
> Boris' use case really sounds valid.

FWIW, I still use it. :-)

And we recently validated anew, a good use case for drop_caches which
was actually already mentioned - repeatable benchmark runs. In this
case, we show how *not* to use it in those benchmark runs. :)

http://marc.info/?l=linux-kernel&m=137276096923390

Thanks.

-- 
Regards/Gruss,
    Boris.

Sent from a fat crate under my desk. Formatting is fine.
--

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org.  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>

^ permalink raw reply	[flat|nested] 10+ messages in thread

* Re: [-] drop_caches-add-some-documentation-and-info-messsge.patch removed from -mm tree
  2013-07-12 11:50     ` Michal Hocko
  2013-07-12 12:25       ` Borislav Petkov
@ 2013-07-12 19:08       ` KOSAKI Motohiro
  1 sibling, 0 replies; 10+ messages in thread
From: KOSAKI Motohiro @ 2013-07-12 19:08 UTC (permalink / raw)
  To: Michal Hocko
  Cc: Andrew Morton, kosaki.motohiro, kamezawa.hiroyu, linux-mm,
	Borislav Petkov, Dave Hansen, kosaki.motohiro

(7/12/13 7:50 AM), Michal Hocko wrote:
> On Thu 11-07-13 14:50:34, Andrew Morton wrote:
>> On Thu, 11 Jul 2013 09:36:44 +0200 Michal Hocko <mhocko@suse.cz> wrote:
>>
>>> On Wed 10-07-13 13:25:03, Andrew Morton wrote:
>>> [...]
>>>> This patch was dropped because it has gone stale
>>>
>>> Is there really a strong reason to not take this patch?
>>
>> I flushed out a whole bunch of MM patches which had been floating
>> around in indecisive limbo.
>>
>> I don't recall all the review issues surrounding this one.
>
> Kosaki was concerned about annoying number of messages if somebody drops
> caches too often (https://lkml.org/lkml/2010/9/20/450). As I noted in
> the changelog
> "
>      Kosaki was worried about possible excessive logging when somebody drops
>      caches too often (but then he claimed he didn't have a strong opinion on
>      that) but I would say opposite.  If somebody does that then I would really
>      like to know that from the log when supporting a system because it almost
>      for sure means that there is something fishy going on.  It is also worth
>      mentioning that only root can write drop caches so this is not an flooding
>      attack vector.
> "
>
> Kosaki then Acked the patch.
>
> You were worried (http://lkml.indiana.edu/hypermail/linux/kernel/1210.3/00605.html)
> about people hating us because they are using this as a solution to
> their issues. I concur that most of those are just hacks that found
> their way into scripts looong time agon and stayed there.
>
> Boris then noted (http://lkml.indiana.edu/hypermail/linux/kernel/1210.3/00659.html)
> that he is using drop_caches to make s2ram faster but as others noted
> this just adds the overhead to the resume path so it might work only for
> certain use cases so a user space solution is more appropriate and
> Boris' use case really sounds valid.
>
> As a compromise I can lower the log level. Would KERN_INFO work for
> you? Or even KERN_DEBUG?
>
> I still find printk less intrusive than fiddling with vmstat counters.

Michal,

It's ok to go IMHO. However, please open new thread w/ a rebased patch. Many developers
don't pay attention a lot to -mm automatic notifications.

Thanks.


--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org.  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>

^ permalink raw reply	[flat|nested] 10+ messages in thread

end of thread, other threads:[~2013-07-12 19:08 UTC | newest]

Thread overview: 10+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2013-07-10 20:25 [-] drop_caches-add-some-documentation-and-info-messsge.patch removed from -mm tree akpm
2013-07-11  7:36 ` Michal Hocko
2013-07-11 12:39   ` Michal Hocko
2013-07-11 15:34     ` Dave Hansen
2013-07-11 15:42       ` Michal Hocko
2013-07-11 15:45         ` Dave Hansen
2013-07-11 21:50   ` Andrew Morton
2013-07-12 11:50     ` Michal Hocko
2013-07-12 12:25       ` Borislav Petkov
2013-07-12 19:08       ` KOSAKI Motohiro

This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.