From: Kent Overstreet <kmo@daterainc.com>
To: Josep Lladonosa <jlladono@gmail.com>
Cc: Gabriel de Perthuis <g2p.code@gmail.com>,
"linux-bcache@vger.kernel.org" <linux-bcache@vger.kernel.org>
Subject: Re: bcache and memory at boot. Bug?
Date: Mon, 10 Feb 2014 14:46:48 -0800 [thread overview]
Message-ID: <20140210224648.GC2362@kmo> (raw)
In-Reply-To: <CAPBO7TZBgfa2CftYZYv2+dTteyBYJYyXD8vuw7KQNN5xexFjpQ@mail.gmail.com>
On Mon, Feb 10, 2014 at 02:15:14PM +0100, Josep Lladonosa wrote:
> On 10 February 2014 12:40, Gabriel de Perthuis <g2p.code@gmail.com> wrote:
> >> Hello,
> >>
> >> I am using bcache in my laptop (RAM 32 GiB). I have been using it
> >> perfectly for a month. ext4 + bcache.
> >>
> >> Yesterday, system did not want do to a normal startup. Messages where
> >> from udev out of memory (and udevadm killed?) and system was not
> >> mounting / (/dev/bcache0).
> >>
> >> [ 6.823222] bcache-register invoked oom-killer: gfp_mask=0x82d2,
> >> order=0, oom_score_adj=0
> >
> > Build your kernel with CONFIG_COMPACTION. I've added it to the FAQ:
> > http://bcache.evilpiepirate.org/~kent/bcache/FAQ/
>
>
> Hello Gabriel,
>
> Thanks for your reply.
>
> My .config file has already:
> ...
> CONFIG_COMPACTION=y
I looked at the allocations done in bcache_device_init(), those aren't high
order page allocations (they're vmalloc allocations) so compaction won't help
here.
> ...
>
> I have been digging around to find information and sometimes bugs come
> related to unsigned variables and their operations when storing large
> numbers (reserved memory?). Kernel is 64 bits. Memory is 32 GiB.
> Perhaps it could happen that the allocated memory is not totally
> freed.
>
> When kernel has been able to boot with 32 GiB (having bcache0 messed
> up and bcache1 mountable), it seems that bcache has more than 16 GiB
> allocated and kept.
Woah, really? If so that's bad - I don't see how that follows (or am I
misreading) from your numbers, can you show a before and after when you bring
bcache up?
>
> -Free memory with kernel 3.11.0-rc7 with 32GiB and bcache0 (halved?)
> and bcache1 (mounted)
>
> $ head -2 /proc/meminfo
> MemTotal: 32913872 kB
> MemFree: 12376640 kB
>
>
> - Free memory with kernel 3.12.9 with mem=16G and bcache0 (mounted)
>
> $ head -2 /proc/meminfo
> MemTotal: 15602144 kB
> MemFree: 14005372 kB
>
>
>
> >
> >
> > --
> > To unsubscribe from this list: send the line "unsubscribe linux-bcache" in
> > the body of a message to majordomo@vger.kernel.org
> > More majordomo info at http://vger.kernel.org/majordomo-info.html
>
>
>
> --
> --
> Salutacions...Josep
> --
> --
> To unsubscribe from this list: send the line "unsubscribe linux-bcache" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at http://vger.kernel.org/majordomo-info.html
next prev parent reply other threads:[~2014-02-10 22:46 UTC|newest]
Thread overview: 7+ messages / expand[flat|nested] mbox.gz Atom feed top
[not found] <CAPBO7TYBMn=oU8K-6EGzw2LofwHPHm742GbzgGYmdbHt5g7sNQ@mail.gmail.com>
2014-02-08 8:53 ` Fwd: bcache and memory at boot. Bug? Josep Lladonosa
2014-02-10 11:40 ` Gabriel de Perthuis
2014-02-10 13:15 ` Josep Lladonosa
2014-02-10 22:46 ` Kent Overstreet [this message]
2014-02-11 10:56 ` Josep Lladonosa
[not found] ` <1392117811.5355.YahooMailNeo@web181506.mail.ne1.yahoo.com>
[not found] ` <CAPBO7TbeuHqur3oi9q82d6_MBG9dw=XcZ7yQyV6iSEx2uDuiMQ@mail.gmail.com>
2014-02-26 16:03 ` Fwd: " Josep Lladonosa
2014-02-26 16:38 ` Josep Lladonosa
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=20140210224648.GC2362@kmo \
--to=kmo@daterainc.com \
--cc=g2p.code@gmail.com \
--cc=jlladono@gmail.com \
--cc=linux-bcache@vger.kernel.org \
/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
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
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.