lists.openwall.net   lists  /  announce  owl-users  owl-dev  john-users  john-dev  passwdqc-users  yescrypt  popa3d-users  /  oss-security  kernel-hardening  musl  sabotage  tlsify  passwords  /  crypt-dev  xvendor  /  Bugtraq  Full-Disclosure  linux-kernel  linux-netdev  linux-ext4  linux-hardening  linux-cve-announce  PHC 
Open Source and information security mailing list archives
 
Hash Suite: Windows password security audit tool. GUI, reports in PDF.
[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Date:	Mon, 12 Dec 2011 09:26:41 +0800
From:	Tao Ma <tm@....ma>
To:	Minchan Kim <minchan.kim@...il.com>
CC:	linux-mm@...ck.org, linux-kernel@...r.kernel.org,
	Mel Gorman <mel@....ul.ie>, Rik van Riel <riel@...hat.com>,
	Johannes Weiner <hannes@...xchg.org>,
	Christoph Hellwig <hch@...radead.org>,
	KAMEZAWA Hiroyuki <kamezawa.hiroyu@...fujitsu.com>,
	KOSAKI Motohiro <kosaki.motohiro@...fujitsu.com>,
	Andrea Arcangeli <aarcange@...hat.com>,
	Andrew Morton <akpm@...ux-foundation.org>
Subject: Re: [PATCH V2] vmscan/trace: Add 'active' and 'file' info to trace_mm_vmscan_lru_isolate.

On 12/12/2011 09:23 AM, Minchan Kim wrote:
> On Mon, Dec 12, 2011 at 10:19 AM, Tao Ma <tm@....ma> wrote:
>> On 12/12/2011 08:59 AM, Minchan Kim wrote:
>>> Hi Tao,
>>>
>>> On Sun, Dec 11, 2011 at 11:46 PM, Tao Ma <tm@....ma> wrote:
>>>> From: Tao Ma <boyu.mt@...bao.com>
>>>>
>>>> In trace_mm_vmscan_lru_isolate, we don't output 'active' and 'file'
>>>> information to the trace event and it is a bit inconvenient for the
>>>> user to get the real information(like pasted below).
>>>> mm_vmscan_lru_isolate: isolate_mode=2 order=0 nr_requested=32 nr_scanned=32
>>>> nr_taken=32 contig_taken=0 contig_dirty=0 contig_failed=0
>>>>
>>>> So this patch adds these 2 info to the trace event and it now looks like:
>>>> mm_vmscan_lru_isolate: isolate_mode=2 order=0 nr_requested=32 nr_scanned=32
>>>> nr_taken=32 contig_taken=0 contig_dirty=0 contig_failed=0 active=1 file=0
>>>>
>>>> Cc: Mel Gorman <mel@....ul.ie>
>>>> Cc: Rik van Riel <riel@...hat.com>
>>>> Cc: Johannes Weiner <hannes@...xchg.org>
>>>> Cc: Christoph Hellwig <hch@...radead.org>
>>>> Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@...fujitsu.com>
>>>> Cc: KOSAKI Motohiro <kosaki.motohiro@...fujitsu.com>
>>>> Cc: Andrea Arcangeli <aarcange@...hat.com>
>>>> Cc: Andrew Morton <akpm@...ux-foundation.org>
>>>> Signed-off-by: Tao Ma <boyu.mt@...bao.com>
>>>> ---
>>>>  include/trace/events/vmscan.h |   25 +++++++++++++++++--------
>>>>  mm/memcontrol.c               |    2 +-
>>>>  mm/vmscan.c                   |    6 +++---
>>>>  3 files changed, 21 insertions(+), 12 deletions(-)
>>>>
>>>> diff --git a/include/trace/events/vmscan.h b/include/trace/events/vmscan.h
>>>> index edc4b3d..82bc49c 100644
>>>> --- a/include/trace/events/vmscan.h
>>>> +++ b/include/trace/events/vmscan.h
>>>> @@ -266,9 +266,10 @@ DECLARE_EVENT_CLASS(mm_vmscan_lru_isolate_template,
>>>>                unsigned long nr_lumpy_taken,
>>>>                unsigned long nr_lumpy_dirty,
>>>>                unsigned long nr_lumpy_failed,
>>>> -               isolate_mode_t isolate_mode),
>>>> +               isolate_mode_t isolate_mode,
>>>> +               int active, int file),
>>>>
>>>> -       TP_ARGS(order, nr_requested, nr_scanned, nr_taken, nr_lumpy_taken, nr_lumpy_dirty, nr_lumpy_failed, isolate_mode),
>>>> +       TP_ARGS(order, nr_requested, nr_scanned, nr_taken, nr_lumpy_taken, nr_lumpy_dirty, nr_lumpy_failed, isolate_mode, active, file),
>>>>
>>>>        TP_STRUCT__entry(
>>>>                __field(int, order)
>>>> @@ -279,6 +280,8 @@ DECLARE_EVENT_CLASS(mm_vmscan_lru_isolate_template,
>>>>                __field(unsigned long, nr_lumpy_dirty)
>>>>                __field(unsigned long, nr_lumpy_failed)
>>>>                __field(isolate_mode_t, isolate_mode)
>>>> +               __field(int, active)
>>>> +               __field(int, file)
>>>>        ),
>>>>
>>>>        TP_fast_assign(
>>>> @@ -290,9 +293,11 @@ DECLARE_EVENT_CLASS(mm_vmscan_lru_isolate_template,
>>>>                __entry->nr_lumpy_dirty = nr_lumpy_dirty;
>>>>                __entry->nr_lumpy_failed = nr_lumpy_failed;
>>>>                __entry->isolate_mode = isolate_mode;
>>>> +               __entry->active = active;
>>>> +               __entry->file = file;
>>>>        ),
>>>>
>>>> -       TP_printk("isolate_mode=%d order=%d nr_requested=%lu nr_scanned=%lu nr_taken=%lu contig_taken=%lu contig_dirty=%lu contig_failed=%lu",
>>>> +       TP_printk("isolate_mode=%d order=%d nr_requested=%lu nr_scanned=%lu nr_taken=%lu contig_taken=%lu contig_dirty=%lu contig_failed=%lu active=%d file=%d",
>>>>                __entry->isolate_mode,
>>>>                __entry->order,
>>>>                __entry->nr_requested,
>>>> @@ -300,7 +305,9 @@ DECLARE_EVENT_CLASS(mm_vmscan_lru_isolate_template,
>>>>                __entry->nr_taken,
>>>>                __entry->nr_lumpy_taken,
>>>>                __entry->nr_lumpy_dirty,
>>>> -               __entry->nr_lumpy_failed)
>>>> +               __entry->nr_lumpy_failed,
>>>> +               __entry->active,
>>>> +               __entry->file)
>>>>  );
>>>>
>>>>  DEFINE_EVENT(mm_vmscan_lru_isolate_template, mm_vmscan_lru_isolate,
>>>> @@ -312,9 +319,10 @@ DEFINE_EVENT(mm_vmscan_lru_isolate_template, mm_vmscan_lru_isolate,
>>>>                unsigned long nr_lumpy_taken,
>>>>                unsigned long nr_lumpy_dirty,
>>>>                unsigned long nr_lumpy_failed,
>>>> -               isolate_mode_t isolate_mode),
>>>> +               isolate_mode_t isolate_mode,
>>>> +               int active, int file),
>>>>
>>>> -       TP_ARGS(order, nr_requested, nr_scanned, nr_taken, nr_lumpy_taken, nr_lumpy_dirty, nr_lumpy_failed, isolate_mode)
>>>> +       TP_ARGS(order, nr_requested, nr_scanned, nr_taken, nr_lumpy_taken, nr_lumpy_dirty, nr_lumpy_failed, isolate_mode, active, file)
>>>>
>>>>  );
>>>>
>>>> @@ -327,9 +335,10 @@ DEFINE_EVENT(mm_vmscan_lru_isolate_template, mm_vmscan_memcg_isolate,
>>>>                unsigned long nr_lumpy_taken,
>>>>                unsigned long nr_lumpy_dirty,
>>>>                unsigned long nr_lumpy_failed,
>>>> -               isolate_mode_t isolate_mode),
>>>> +               isolate_mode_t isolate_mode,
>>>> +               int active, int file),
>>>>
>>>> -       TP_ARGS(order, nr_requested, nr_scanned, nr_taken, nr_lumpy_taken, nr_lumpy_dirty, nr_lumpy_failed, isolate_mode)
>>>> +       TP_ARGS(order, nr_requested, nr_scanned, nr_taken, nr_lumpy_taken, nr_lumpy_dirty, nr_lumpy_failed, isolate_mode, active, file)
>>>>
>>>>  );
>>>>
>>>> diff --git a/mm/memcontrol.c b/mm/memcontrol.c
>>>> index 6aff93c..246fbce 100644
>>>> --- a/mm/memcontrol.c
>>>> +++ b/mm/memcontrol.c
>>>> @@ -1249,7 +1249,7 @@ unsigned long mem_cgroup_isolate_pages(unsigned long nr_to_scan,
>>>>        *scanned = scan;
>>>>
>>>>        trace_mm_vmscan_memcg_isolate(0, nr_to_scan, scan, nr_taken,
>>>> -                                     0, 0, 0, mode);
>>>> +                                     0, 0, 0, mode, active, file);
>>>>
>>>>        return nr_taken;
>>>>  }
>>>> diff --git a/mm/vmscan.c b/mm/vmscan.c
>>>> index f54a05b..97955ca 100644
>>>> --- a/mm/vmscan.c
>>>> +++ b/mm/vmscan.c
>>>> @@ -1103,7 +1103,7 @@ int __isolate_lru_page(struct page *page, isolate_mode_t mode, int file)
>>>>  static unsigned long isolate_lru_pages(unsigned long nr_to_scan,
>>>>                struct list_head *src, struct list_head *dst,
>>>>                unsigned long *scanned, int order, isolate_mode_t mode,
>>>> -               int file)
>>>> +               int active, int file)
>>>>  {
>>>>        unsigned long nr_taken = 0;
>>>>        unsigned long nr_lumpy_taken = 0;
>>>> @@ -1221,7 +1221,7 @@ static unsigned long isolate_lru_pages(unsigned long nr_to_scan,
>>>>                        nr_to_scan, scan,
>>>>                        nr_taken,
>>>>                        nr_lumpy_taken, nr_lumpy_dirty, nr_lumpy_failed,
>>>> -                       mode);
>>>> +                       mode, active, file);
>>>>        return nr_taken;
>>>>  }
>>>>
>>>> @@ -1237,7 +1237,7 @@ static unsigned long isolate_pages_global(unsigned long nr,
>>>>        if (file)
>>>>                lru += LRU_FILE;
>>>>        return isolate_lru_pages(nr, &z->lru[lru].list, dst, scanned, order,
>>>> -                                                               mode, file);
>>>> +                                                       mode, active, file);
>>>
>>> I guess you want to count exact scanning number of which lru list.
>>> But It's impossible now since we do lumpy reclaim so that trace's
>>> result is mixed by active/inactive list scanning.
>>> And I don't like adding new argument for just trace although it's trivial.
>> yeah, I know we do lumpy reclaim, but it has no hint about whether it is
>> a file or anon lru. So I think we at least need a 'file=[0/1]' in this
>> trace event.
>>>
>>> I think 'mode' is more proper rather than  specific 'active'.
>>> The 'mode' can achieve your goal without passing new argument "active".
>> sorry, but how can we find the real relationship between 'mode' and
>> 'active'? I am not quite familiar with this field. So if you can
>> explicit describe it, I am fine to drop this field.
> 
> 
> mode is following as,
> 
> /* Isolate inactive pages */
> #define ISOLATE_INACTIVE        ((__force fmode_t)0x1)
> /* Isolate active pages */
> #define ISOLATE_ACTIVE          ((__force fmode_t)0x2)
> /* Isolate clean file */
> #define ISOLATE_CLEAN           ((__force fmode_t)0x4)
> /* Isolate unmapped file */
> #define ISOLATE_UNMAPPED        ((__force fmode_t)0x8)
> 
> For example,
> 
> If mode is 0x1, it means we are isolating inactive.
> If mode is 0x2, it means we are isolating active.
> If mode is 0x4|0x2, it mean we are isolating clear pages in active list.
> If mode is 0x8|0x2, it mean we are isolate unmapped page in active list.
cool, thanks for the detailed explanation.
So I will drop this field, and only add 'file' to the trace event.

Thanks
Tao
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

Powered by blists - more mailing lists