[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <54AE43E3.60209@suse.cz>
Date: Thu, 08 Jan 2015 09:46:27 +0100
From: Vlastimil Babka <vbabka@...e.cz>
To: Joonsoo Kim <iamjoonsoo.kim@....com>
CC: Andrew Morton <akpm@...ux-foundation.org>,
Mel Gorman <mgorman@...e.de>,
David Rientjes <rientjes@...gle.com>, linux-mm@...ck.org,
linux-kernel@...r.kernel.org
Subject: Re: [PATCH 1/3] mm/compaction: enhance trace output to know more
about compaction internals
On 01/08/2015 09:18 AM, Joonsoo Kim wrote:
> On Tue, Jan 06, 2015 at 10:05:39AM +0100, Vlastimil Babka wrote:
>> On 12/03/2014 08:52 AM, Joonsoo Kim wrote:
>> > It'd be useful to know where the both scanner is start. And, it also be
>> > useful to know current range where compaction work. It will help to find
>> > odd behaviour or problem on compaction.
>>
>> Overall it looks good, just two questions:
>> 1) Why change the pfn output to hexadecimal with different printf layout and
>> change the variable names and? Is it that better to warrant people having to
>> potentially modify their scripts parsing the old output?
>
> Deciaml output has really bad readability since we manage all pages by order
> of 2 which is well represented by hexadecimal. With hex output, we can
> easily notice whether we move out from one pageblock to another one.
OK. I don't have any strong objection, maybe Mel should comment on this as the
author of most of the tracepoints? But if it happens, I think converting the old
tracepoints to new hexadecimal format should be a separate patch from adding the
new ones.
>> 2) Would it be useful to also print in the mm_compaction_isolate_template based
>> tracepoints, pfn of where the particular scanner left off a block prematurely?
>> It doesn't always match start_pfn + nr_scanned.
>
> With start_pfn and end_pfn, detailed analysis is possible. We can know pageblock
> where we actually scan and isolate and how much pages we try in that
> pageblock and can guess why it doesn't become freepage with pageblock
> order roughly.
>
> nr_scanned is just different metric. end_pfn don't need to match with
> start_pfn + nr_scanned.
Well that's part of my point. end_pfn is the end of the pageblock. nr_scanned
might be lower than end_pfn - start_pfn, because we terminate in the middle of
the pageblock. But it might be also lower, because we e.g. skip higher-order
free pages. So we don't recognize where we terminated early.
> Thanks.
>
> --
> To unsubscribe, send a message with 'unsubscribe linux-mm' in
> the body to majordomo@...ck.org. For more info on Linux MM,
> see: http://www.linux-mm.org/ .
> Don't email: <a href=mailto:"dont@...ck.org"> email@...ck.org </a>
>
--
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