From: Milan Broz <mbroz-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>
To: Ritesh Raj Sarraf <rrs-7WuBAv+fczCJ8c2fQYRYNw@public.gmane.org>
Cc: device-mapper development
<dm-devel-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>,
dm-crypt-4q3lyFh4P1g@public.gmane.org
Subject: Re: [dm-devel] dm-crypt low performance
Date: Tue, 13 Jan 2009 16:50:48 +0100 [thread overview]
Message-ID: <496CB858.70803@redhat.com> (raw)
In-Reply-To: <0ths36-dce.ln1-b1aPebNLFEbkkoF7R9+BZ0EOCMrvLtNR@public.gmane.org>
Ritesh Raj Sarraf wrote:
> I'm using dm-crypt on an external USB enclosure ide hdd. I copied 20Gb of
> data on to the encrypted device and am suffering severe performance penalty.
> After a couple of seconds, all processes involved in the I/O keep stalling.
> They periodically resume, just for a couple of seconds and then again stall.
>
> Running linux 2.6.26.
Please can you try it without dm-crypt involved?
Just copy 20Gb to your unencrypted disk attached through USB.
I saw this problem with plain USB attached storage some time ago too.
There is only one thread in dm-crypt and it is waiting for io_schedule,
so I think it is not dm-crypt who is blocking it, but the layer
below the dm-crypt.
Milan
> [15869.848387] kcryptd D f7044a0c 0 13656 2
> [15869.848387] f29d79c0 00000046 01c58000 f7044a0c c03c1430 c03c4740
> c03c4740 f29d79c0
> [15869.848387] f29d7b4c c201c740 00000001 00010001 00000001 f79b48d0
> f29d7b4c 00eb4b70
> [15869.848387] c01df9a6 c0135b72 00000000 00000000 00000000 000000ff
> c201c740 c201cb44
> [15869.848387] Call Trace:
> [15869.848387] [<c01df9a6>] cfq_may_queue+0x48/0xab
> [15869.848387] [<c0135b72>] getnstimeofday+0x32/0xaf
> [15869.848387] [<c02c1091>] io_schedule+0x4f/0x86
> [15869.848387] [<c01d734f>] get_request_wait+0xf2/0x16d
> [15869.848387] [<c013140a>] autoremove_wake_function+0x0/0x2d
> [15869.848387] [<c01d7789>] __make_request+0x2b7/0x33a
> [15869.848387] [<c01d6465>] generic_make_request+0x419/0x44a
> [15869.848387] [<f8c30563>] async_encrypt+0x2f/0x35 [crypto_blkcipher]
> [15869.848387] [<f8c659ce>] crypt_convert+0x20e/0x240 [dm_crypt]
> [15869.848387] [<f8c65c5d>] kcryptd_crypt+0x25d/0x339 [dm_crypt]
> [15869.848387] [<f8c65a00>] kcryptd_crypt+0x0/0x339 [dm_crypt]
> [15869.848387] [<f8c65a00>] kcryptd_crypt+0x0/0x339 [dm_crypt]
> [15869.848387] [<c012e818>] run_workqueue+0x73/0xed
> [15869.848387] [<c012e949>] worker_thread+0xb7/0xc3
> [15869.848387] [<c013140a>] autoremove_wake_function+0x0/0x2d
> [15869.848387] [<c012e892>] worker_thread+0x0/0xc3
> [15869.848387] [<c01311a6>] kthread+0x38/0x5d
> [15869.848387] [<c013116e>] kthread+0x0/0x5d
> [15869.848387] [<c01045af>] kernel_thread_helper+0x7/0x10
Milan
--
mbroz-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org
next prev parent reply other threads:[~2009-01-13 15:50 UTC|newest]
Thread overview: 10+ messages / expand[flat|nested] mbox.gz Atom feed top
2009-01-12 19:22 dm-crypt low performance Ritesh Raj Sarraf
[not found] ` <0ths36-dce.ln1-b1aPebNLFEbkkoF7R9+BZ0EOCMrvLtNR@public.gmane.org>
2009-01-13 15:50 ` Milan Broz [this message]
2009-01-13 17:37 ` Ritesh Raj Sarraf
2009-01-13 19:52 ` [dm-devel] " Ritesh Raj Sarraf
2009-01-13 19:52 ` Ritesh Raj Sarraf
2009-01-13 20:00 ` [linux-pm] " Alan Stern
2009-01-19 10:34 ` Ritesh Raj Sarraf
2009-01-19 16:23 ` [dm-devel] " Alan Stern
2009-01-19 10:34 ` Ritesh Raj Sarraf
2009-01-13 20:00 ` Alan Stern
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=496CB858.70803@redhat.com \
--to=mbroz-h+wxahxf7alqt0dzr+alfa@public.gmane.org \
--cc=dm-crypt-4q3lyFh4P1g@public.gmane.org \
--cc=dm-devel-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org \
--cc=rrs-7WuBAv+fczCJ8c2fQYRYNw@public.gmane.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.