All of lore.kernel.org
 help / color / mirror / Atom feed
From: Michal Hocko <mhocko@kernel.org>
To: Vlastimil Babka <vbabka@suse.cz>
Cc: Andrew Morton <akpm@linux-foundation.org>,
	David Rientjes <rientjes@google.com>,
	Johannes Weiner <hannes@cmpxchg.org>,
	linux-mm@kvack.org, LKML <linux-kernel@vger.kernel.org>
Subject: Re: [PATCH 3/3] oom, trace: add compaction retry tracepoint
Date: Wed, 14 Dec 2016 19:11:34 +0100	[thread overview]
Message-ID: <20161214181133.GB16763@dhcp22.suse.cz> (raw)
In-Reply-To: <60cfb7ca-fb95-7a34-bae2-9b7c49119573@suse.cz>

On Wed 14-12-16 18:28:38, Vlastimil Babka wrote:
> On 12/14/2016 03:53 PM, Michal Hocko wrote:
> > From: Michal Hocko <mhocko@suse.com>
> > 
> > Higher order requests oom debugging is currently quite hard. We do have
> > some compaction points which can tell us how the compaction is operating
> > but there is no trace point to tell us about compaction retry logic.
> > This patch adds a one which will have the following format
> > 
> >             bash-3126  [001] ....  1498.220001: compact_retry: order=9 priority=COMPACT_PRIO_SYNC_LIGHT compaction_result=withdrawn retries=0 max_retries=16 should_retry=0
> > 
> > we can see that the order 9 request is not retried even though we are in
> > the highest compaction priority mode becase the last compaction attempt
> > was withdrawn. This means that compaction_zonelist_suitable must have
> > returned false and there is no suitable zone to compact for this request
> > and so no need to retry further.
> > 
> > another example would be
> >            <...>-3137  [001] ....    81.501689: compact_retry: order=9 priority=COMPACT_PRIO_SYNC_LIGHT compaction_result=failed retries=0 max_retries=16 should_retry=0
> > 
> > in this case the order-9 compaction failed to find any suitable
> > block. We do not retry anymore because this is a costly request
> > and those do not go below COMPACT_PRIO_SYNC_LIGHT priority.
> > 
> > Signed-off-by: Michal Hocko <mhocko@suse.com>
> > ---
> >  include/trace/events/mmflags.h | 26 ++++++++++++++++++++++++++
> >  include/trace/events/oom.h     | 39 +++++++++++++++++++++++++++++++++++++++
> >  mm/page_alloc.c                | 22 ++++++++++++++++------
> >  3 files changed, 81 insertions(+), 6 deletions(-)
> > 
> > diff --git a/include/trace/events/mmflags.h b/include/trace/events/mmflags.h
> > index 7e4cfede873c..aa4caa6914a9 100644
> > --- a/include/trace/events/mmflags.h
> > +++ b/include/trace/events/mmflags.h
> > @@ -187,8 +187,32 @@ IF_HAVE_VM_SOFTDIRTY(VM_SOFTDIRTY,	"softdirty"	)		\
> >  	EM( COMPACT_NO_SUITABLE_PAGE,	"no_suitable_page")	\
> >  	EM( COMPACT_NOT_SUITABLE_ZONE,	"not_suitable_zone")	\
> >  	EMe(COMPACT_CONTENDED,		"contended")
> > +
> > +/* High-level compaction status feedback */
> > +#define COMPACTION_FAILED	1
> > +#define COMPACTION_WITHDRAWN	2
> > +#define COMPACTION_PROGRESS	3
> > +
> > +#define compact_result_to_feedback(result)	\
> > +({						\
> > + 	enum compact_result __result = result;	\
> > +	(compaction_failed(__result)) ? COMPACTION_FAILED : \
> > +		(compaction_withdrawn(__result)) ? COMPACTION_WITHDRAWN : COMPACTION_PROGRESS; \
> > +})
> 
> It seems you forgot to actually use this "function" (sorry, didn't notice
> earlier) so currently it's translating enum compact_result directly into the
> failed/withdrawn/progress strings, which is wrong.

You are right. I've screwed while integrating the enum translation part.

> The correct place for the result->feedback conversion should be
> TP_fast_assign, so __entry->result should become __entry->feedback. It's too
> late in TP_printk, as userspace tools (e.g. trace-cmd) won't know the
> functions that compact_result_to_feedback() uses.

Thanks. The follow up fix should be
---
>From 82c9064698594d694b052d3847c905e90becc7c5 Mon Sep 17 00:00:00 2001
From: Michal Hocko <mhocko@suse.com>
Date: Wed, 14 Dec 2016 19:09:10 +0100
Subject: [PATCH] fold me "oom, trace: add compaction retry tracepoint"

Do not forget to translate compact_result into a highlevel constants
as per Vlastimil

Signed-off-by: Michal Hocko <mhocko@suse.com>
---
 include/trace/events/oom.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/include/trace/events/oom.h b/include/trace/events/oom.h
index e9d690665b7a..38baeb27221a 100644
--- a/include/trace/events/oom.h
+++ b/include/trace/events/oom.h
@@ -94,7 +94,7 @@ TRACE_EVENT(compact_retry,
 	TP_fast_assign(
 		__entry->order = order;
 		__entry->priority = priority;
-		__entry->result = result;
+		__entry->result = compact_result_to_feedback(result);
 		__entry->retries = retries;
 		__entry->max_retries = max_retries;
 		__entry->ret = ret;
-- 
2.10.2

-- 
Michal Hocko
SUSE Labs

WARNING: multiple messages have this Message-ID
From: Michal Hocko <mhocko@kernel.org>
To: Vlastimil Babka <vbabka@suse.cz>
Cc: Andrew Morton <akpm@linux-foundation.org>,
	David Rientjes <rientjes@google.com>,
	Johannes Weiner <hannes@cmpxchg.org>,
	linux-mm@kvack.org, LKML <linux-kernel@vger.kernel.org>
Subject: Re: [PATCH 3/3] oom, trace: add compaction retry tracepoint
Date: Wed, 14 Dec 2016 19:11:34 +0100	[thread overview]
Message-ID: <20161214181133.GB16763@dhcp22.suse.cz> (raw)
In-Reply-To: <60cfb7ca-fb95-7a34-bae2-9b7c49119573@suse.cz>

On Wed 14-12-16 18:28:38, Vlastimil Babka wrote:
> On 12/14/2016 03:53 PM, Michal Hocko wrote:
> > From: Michal Hocko <mhocko@suse.com>
> > 
> > Higher order requests oom debugging is currently quite hard. We do have
> > some compaction points which can tell us how the compaction is operating
> > but there is no trace point to tell us about compaction retry logic.
> > This patch adds a one which will have the following format
> > 
> >             bash-3126  [001] ....  1498.220001: compact_retry: order=9 priority=COMPACT_PRIO_SYNC_LIGHT compaction_result=withdrawn retries=0 max_retries=16 should_retry=0
> > 
> > we can see that the order 9 request is not retried even though we are in
> > the highest compaction priority mode becase the last compaction attempt
> > was withdrawn. This means that compaction_zonelist_suitable must have
> > returned false and there is no suitable zone to compact for this request
> > and so no need to retry further.
> > 
> > another example would be
> >            <...>-3137  [001] ....    81.501689: compact_retry: order=9 priority=COMPACT_PRIO_SYNC_LIGHT compaction_result=failed retries=0 max_retries=16 should_retry=0
> > 
> > in this case the order-9 compaction failed to find any suitable
> > block. We do not retry anymore because this is a costly request
> > and those do not go below COMPACT_PRIO_SYNC_LIGHT priority.
> > 
> > Signed-off-by: Michal Hocko <mhocko@suse.com>
> > ---
> >  include/trace/events/mmflags.h | 26 ++++++++++++++++++++++++++
> >  include/trace/events/oom.h     | 39 +++++++++++++++++++++++++++++++++++++++
> >  mm/page_alloc.c                | 22 ++++++++++++++++------
> >  3 files changed, 81 insertions(+), 6 deletions(-)
> > 
> > diff --git a/include/trace/events/mmflags.h b/include/trace/events/mmflags.h
> > index 7e4cfede873c..aa4caa6914a9 100644
> > --- a/include/trace/events/mmflags.h
> > +++ b/include/trace/events/mmflags.h
> > @@ -187,8 +187,32 @@ IF_HAVE_VM_SOFTDIRTY(VM_SOFTDIRTY,	"softdirty"	)		\
> >  	EM( COMPACT_NO_SUITABLE_PAGE,	"no_suitable_page")	\
> >  	EM( COMPACT_NOT_SUITABLE_ZONE,	"not_suitable_zone")	\
> >  	EMe(COMPACT_CONTENDED,		"contended")
> > +
> > +/* High-level compaction status feedback */
> > +#define COMPACTION_FAILED	1
> > +#define COMPACTION_WITHDRAWN	2
> > +#define COMPACTION_PROGRESS	3
> > +
> > +#define compact_result_to_feedback(result)	\
> > +({						\
> > + 	enum compact_result __result = result;	\
> > +	(compaction_failed(__result)) ? COMPACTION_FAILED : \
> > +		(compaction_withdrawn(__result)) ? COMPACTION_WITHDRAWN : COMPACTION_PROGRESS; \
> > +})
> 
> It seems you forgot to actually use this "function" (sorry, didn't notice
> earlier) so currently it's translating enum compact_result directly into the
> failed/withdrawn/progress strings, which is wrong.

You are right. I've screwed while integrating the enum translation part.

> The correct place for the result->feedback conversion should be
> TP_fast_assign, so __entry->result should become __entry->feedback. It's too
> late in TP_printk, as userspace tools (e.g. trace-cmd) won't know the
> functions that compact_result_to_feedback() uses.

Thanks. The follow up fix should be
---

  reply	other threads:[~2016-12-14 18:11 UTC|newest]

Thread overview: 29+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2016-12-14 14:53 [PATCH 0/3] mm, oom: add oom detection tracepoints Michal Hocko
2016-12-14 14:53 ` Michal Hocko
2016-12-14 14:53 ` [PATCH 1/3] mm, trace: extract COMPACTION_STATUS and ZONE_TYPE to a common header Michal Hocko
2016-12-14 14:53   ` Michal Hocko
2016-12-14 17:14   ` Vlastimil Babka
2016-12-14 17:14     ` Vlastimil Babka
2016-12-14 17:32   ` kbuild test robot
2016-12-14 19:02     ` Michal Hocko
2016-12-14 19:02       ` Michal Hocko
2016-12-14 14:53 ` [PATCH 2/3] oom, trace: Add oom detection tracepoints Michal Hocko
2016-12-14 14:53   ` Michal Hocko
2016-12-14 17:17   ` Vlastimil Babka
2016-12-14 17:17     ` Vlastimil Babka
2016-12-14 14:53 ` [PATCH 3/3] oom, trace: add compaction retry tracepoint Michal Hocko
2016-12-14 14:53   ` Michal Hocko
2016-12-14 17:28   ` Vlastimil Babka
2016-12-14 17:28     ` Vlastimil Babka
2016-12-14 18:11     ` Michal Hocko [this message]
2016-12-14 18:11       ` Michal Hocko
2016-12-15  8:18       ` Vlastimil Babka
2016-12-15  8:18         ` Vlastimil Babka
2016-12-20 13:01 [PATCH 0/3 v2] mm, oom: add oom detection tracepoints Michal Hocko
2016-12-20 13:01 ` [PATCH 3/3] oom, trace: add compaction retry tracepoint Michal Hocko
2016-12-20 13:01   ` Michal Hocko
2017-01-04 10:47   ` Vlastimil Babka
2017-01-04 10:47     ` Vlastimil Babka
2017-01-04 10:56     ` Michal Hocko
2017-01-04 10:56       ` Michal Hocko
2017-01-04 14:49       ` Vlastimil Babka
2017-01-04 14:49         ` Vlastimil Babka

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20161214181133.GB16763@dhcp22.suse.cz \
    --to=mhocko@kernel.org \
    --cc=akpm@linux-foundation.org \
    --cc=hannes@cmpxchg.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-mm@kvack.org \
    --cc=rientjes@google.com \
    --cc=vbabka@suse.cz \
    --subject='Re: [PATCH 3/3] oom, trace: add compaction retry tracepoint' \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link

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.