From: "Marcin Mirosław" <marcin@mejor.pl>
To: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
Cc: xen-devel@lists.xensource.com
Subject: Re: Re: [b.g.o.358549] blkfront: Move blkif_interrupt into a tasklet.
Date: Tue, 31 May 2011 13:28:44 +0200 [thread overview]
Message-ID: <4DE4D0EC.9060508@mejor.pl> (raw)
In-Reply-To: <20110524131353.GA10513@dumpdata.com>
W dniu 24.05.2011 15:13, Konrad Rzeszutek Wilk pisze:
> On Sat, May 21, 2011 at 03:57:13PM +0200, Marcin Mirosław wrote:
>> Hi!
>> I've noticed kernel threads "flush-254:9" does 100%CPU utilization.
>> Device 254-9 is logical volume created via LVM.
>> I don't know if this problem is related to earlier problem, but it
>> happens on the same kernel as mentioned issue.
>
> What kernel? What is your dom0 kernel and your domU kernel? What
> is the LVM volume? Is it snapshot? Is it on top of a SATA device?
> SSD?
Unfortunately i can't say much about dom0. Admin is very busy and didn't
answer me. I only know domU is placed at one logical volume (lvm).
About domU:
- i didn't noticed any problem with kernels <2.6.37
- with kernels >=2.6.37 i noticed two problems, one of them is mentioned
earlier (WARNING: at block/blk-core.c:239 blk_start_queue+0x20/0x32() ),
second one is "100% of CPU utilization"
- i'm not using lvm snapshot (or other features)
- domU works as PV
- zgrep -i xen /proc/config.gz
CONFIG_XEN=y
CONFIG_XEN_PVHVM=y
CONFIG_XEN_MAX_DOMAIN_MEMORY=8
CONFIG_XEN_SAVE_RESTORE=y
CONFIG_XEN_BLKDEV_FRONTEND=y
CONFIG_XEN_NETDEV_FRONTEND=y
CONFIG_HVC_XEN=y
CONFIG_XEN_BALLOON=y
CONFIG_XEN_DEV_EVTCHN=y
CONFIG_XENFS=y
CONFIG_XEN_COMPAT_XENFS=y
CONFIG_XEN_SYS_HYPERVISOR=y
CONFIG_XEN_PLATFORM_PCI=m
( kernel: 2.6.36-hardened-r9 )
- grep -i xen /usr/src/linux-2.6.38-hardened-r5/.config | grep -v "^#"
CONFIG_XEN=y
CONFIG_XEN_DOM0=y
CONFIG_XEN_PRIVILEGED_GUEST=y
CONFIG_XEN_PVHVM=y
CONFIG_XEN_MAX_DOMAIN_MEMORY=128
CONFIG_XEN_SAVE_RESTORE=y
CONFIG_PCI_XEN=y
CONFIG_XEN_PCIDEV_FRONTEND=y
CONFIG_XEN_BLKDEV_FRONTEND=y
CONFIG_XEN_NETDEV_FRONTEND=y
CONFIG_HVC_XEN=y
CONFIG_XEN_BALLOON=y
CONFIG_XEN_DEV_EVTCHN=y
CONFIG_XEN_BACKEND=y
CONFIG_XENFS=y
CONFIG_XEN_COMPAT_XENFS=y
CONFIG_XEN_SYS_HYPERVISOR=y
CONFIG_XEN_XENBUS_FRONTEND=y
CONFIG_XEN_PLATFORM_PCI=m
CONFIG_SWIOTLB_XEN=y
(problematic kernel)
> Do give more details please.
I'm sorry for inconvenience.
> And what is 'bgo'? I presume a bugzilla, but you hadn't provided
> a link.
This is gentoo bugzilla: http://bugs.gentoo.org/358549
Regards,
Marcin
next prev parent reply other threads:[~2011-05-31 11:28 UTC|newest]
Thread overview: 12+ messages / expand[flat|nested] mbox.gz Atom feed top
2011-05-04 14:08 [b.g.o.358549] blkfront: Move blkif_interrupt into a tasklet Marcin Mirosław
2011-05-05 21:22 ` Konrad Rzeszutek Wilk
2011-05-05 22:57 ` Marcin Mirosław
2011-05-05 23:03 ` Daniel Stodden
2011-05-05 23:07 ` Marcin Mirosław
2011-05-21 13:57 ` Marcin Mirosław
2011-05-24 13:13 ` Konrad Rzeszutek Wilk
2011-05-31 11:28 ` Marcin Mirosław [this message]
2011-06-27 14:13 ` Konrad Rzeszutek Wilk
2011-06-27 15:04 ` Marcin Mirosław
2011-06-27 17:30 ` Konrad Rzeszutek Wilk
2011-06-27 22:01 ` Marcin Mirosław
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=4DE4D0EC.9060508@mejor.pl \
--to=marcin@mejor.pl \
--cc=konrad.wilk@oracle.com \
--cc=xen-devel@lists.xensource.com \
/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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).