Re: [PATCH v5 4/4] add profile information for invalidated page

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



Hi Mel,

On Sat, Feb 19, 2011 at 1:58 AM, Mel Gorman <mel@xxxxxxxxx> wrote:
> On Fri, Feb 18, 2011 at 12:08:22AM +0900, Minchan Kim wrote:
>> This patch adds profile information about invalidated page reclaim.
>> It's just for profiling for test so it is never for merging.
>>
>> Acked-by: Rik van Riel <riel@xxxxxxxxxx>
>> Cc: KOSAKI Motohiro <kosaki.motohiro@xxxxxxxxxxxxxx>
>> Cc: Wu Fengguang <fengguang.wu@xxxxxxxxx>
>> Cc: Johannes Weiner <hannes@xxxxxxxxxxx>
>> Cc: Nick Piggin <npiggin@xxxxxxxxx>
>> Cc: Mel Gorman <mel@xxxxxxxxx>
>> Signed-off-by: Minchan Kim <minchan.kim@xxxxxxxxx>
>> ---
>> Âinclude/linux/vmstat.h | Â Â4 ++--
>> Âmm/swap.c       Â|  Â3 +++
>> Âmm/vmstat.c      Â|  Â3 +++
>> Â3 files changed, 8 insertions(+), 2 deletions(-)
>>
>> diff --git a/include/linux/vmstat.h b/include/linux/vmstat.h
>> index 833e676..c38ad95 100644
>> --- a/include/linux/vmstat.h
>> +++ b/include/linux/vmstat.h
>> @@ -30,8 +30,8 @@
>>
>> Âenum vm_event_item { PGPGIN, PGPGOUT, PSWPIN, PSWPOUT,
>> Â Â Â Â Â Â Â FOR_ALL_ZONES(PGALLOC),
>> - Â Â Â Â Â Â PGFREE, PGACTIVATE, PGDEACTIVATE,
>> - Â Â Â Â Â Â PGFAULT, PGMAJFAULT,
>> + Â Â Â Â Â Â PGFREE, PGACTIVATE, PGDEACTIVATE, PGINVALIDATE,
>> + Â Â Â Â Â Â PGRECLAIM, PGFAULT, PGMAJFAULT,
>> Â Â Â Â Â Â Â FOR_ALL_ZONES(PGREFILL),
>> Â Â Â Â Â Â Â FOR_ALL_ZONES(PGSTEAL),
>> Â Â Â Â Â Â Â FOR_ALL_ZONES(PGSCAN_KSWAPD),
>> diff --git a/mm/swap.c b/mm/swap.c
>> index 0a33714..980c17b 100644
>> --- a/mm/swap.c
>> +++ b/mm/swap.c
>> @@ -397,6 +397,7 @@ static void lru_deactivate(struct page *page, struct zone *zone)
>> Â Â Â Â Â Â Â Â* is _really_ small and Âit's non-critical problem.
>> Â Â Â Â Â Â Â Â*/
>> Â Â Â Â Â Â Â SetPageReclaim(page);
>> + Â Â Â Â Â Â __count_vm_event(PGRECLAIM);
>> Â Â Â } else {
>> Â Â Â Â Â Â Â /*
>> Â Â Â Â Â Â Â Â* The page's writeback ends up during pagevec
>
> Is this name potentially misleading?
>
> Pages that are reclaimed are accounted for with _steal. It's not particularly
> obvious but that's the name it was given. I'd worry that an administrator that
> was not aware of *_steal would read pgreclaim as "pages that were reclaimed"
> when this is not necessarily the case.
>
> Is there a better name for this? pginvalidate_deferred
> or pginvalidate_delayed maybe?
>

Yep. Your suggestion is fair enough. But as I said in description,
It's just for testing for my profiling, not merging so I didn't care
about it.
I don't think we need new vmstat of pginvalidate.

-- 
Kind regards,
Minchan Kim

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@xxxxxxxxxx  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Fight unfair telecom internet charges in Canada: sign http://stopthemeter.ca/
Don't email: <a href


[Index of Archives]     [Linux ARM Kernel]     [Linux ARM]     [Linux Omap]     [Fedora ARM]     [IETF Annouce]     [Bugtraq]     [Linux]     [Linux OMAP]     [Linux MIPS]     [ECOS]     [Asterisk Internet PBX]     [Linux API]