From: Daniel Stodden <daniel.stodden@citrix.com>
To: Jeremy Fitzhardinge <jeremy@goop.org>
Cc: Xen <xen-devel@lists.xensource.com>, Tom Kopec <tek@acm.org>
Subject: Re: blktap lockdep hiccup
Date: Mon, 6 Sep 2010 18:46:57 -0700 [thread overview]
Message-ID: <1283824017.3138.27.camel@agari.van.xensource.com> (raw)
In-Reply-To: <4C8597E5.4070802@goop.org>
On Mon, 2010-09-06 at 21:39 -0400, Jeremy Fitzhardinge wrote:
> On 09/03/2010 09:08 AM, Daniel Stodden wrote:
> > On Thu, 2010-09-02 at 18:46 -0400, Jeremy Fitzhardinge wrote:
> >> On 08/22/2010 11:54 PM, Daniel Stodden wrote:
> >>> Response processing doesn't really belong into hard irq context.
> >>>
> >>> Another potential problem this avoids is that switching interrupt cpu
> >>> affinity in Xen domains can presently lead to event loss, if
> >>> RING_FINAL_CHECK is run from hard irq context.
> >> I just got this warning from a 32-bit pv domain. I think it may relate
> >> to this change. The warning is
> > We clearly spin_lock_irqsave all through the blkif_do_interrupt frame.
> >
> > It follows that something underneath quite unconditionally chose to
> > reenable them again (?)
> >
> > Either: Can you add a bunch of similar WARN_ONs along that path?
> >
> > Or: This lock is quite coarse-grained. The lock only matters for queue
> > access, and we know irqs are reenabled, so no need for flags. In fact we
> > only need to spin_lock_irq around the __blk_end_ calls and
> > kick_pending_.
> >
> > But I don't immediately see what's to blame, so I'd be curious.
>
> I haven't got around to investigating this in more detail yet, but
> there's also this long-standing lockdep hiccup in blktap:
Ack. Let's fix that somewhere this week and see if we can clean up the
spin locking problem too then.
Daniel
> Starting auto Xen domains: lurch alloc irq_desc for 1235 on node 0
> alloc kstat_irqs on node 0
> block tda: sector-size: 512 capacity: 614400
> INFO: trying to register non-static key.
> the code is fine but needs lockdep annotation.
> turning off the locking correctness validator.
> Pid: 4266, comm: tapdisk2 Not tainted 2.6.32.21 #146
> Call Trace:
> [<ffffffff8107f0a4>] __lock_acquire+0x1df/0x16e5
> [<ffffffff8100f955>] ? xen_force_evtchn_callback+0xd/0xf
> [<ffffffff81010082>] ? check_events+0x12/0x20
> [<ffffffff810f0359>] ? apply_to_page_range+0x295/0x37d
> [<ffffffff81080677>] lock_acquire+0xcd/0xf1
> [<ffffffff810f0359>] ? apply_to_page_range+0x295/0x37d
> [<ffffffff810f0259>] ? apply_to_page_range+0x195/0x37d
> [<ffffffff81506f7d>] _spin_lock+0x31/0x40
> [<ffffffff810f0359>] ? apply_to_page_range+0x295/0x37d
> [<ffffffff810f0359>] apply_to_page_range+0x295/0x37d
> [<ffffffff812ab37c>] ? blktap_map_uaddr_fn+0x0/0x55
> [<ffffffff8100d0cf>] ? xen_make_pte+0x8a/0x8e
> [<ffffffff812ac34e>] blktap_device_process_request+0x43d/0x954
> [<ffffffff8100f955>] ? xen_force_evtchn_callback+0xd/0xf
> [<ffffffff81010082>] ? check_events+0x12/0x20
> [<ffffffff8100f955>] ? xen_force_evtchn_callback+0xd/0xf
> [<ffffffff81010082>] ? check_events+0x12/0x20
> [<ffffffff8107d687>] ? mark_held_locks+0x52/0x70
> [<ffffffff81506ddb>] ? _spin_unlock_irq+0x30/0x3c
> [<ffffffff8107d949>] ? trace_hardirqs_on_caller+0x125/0x150
> [<ffffffff812acba6>] blktap_device_run_queue+0x1c5/0x28f
> [<ffffffff812a0234>] ? unbind_from_irq+0x18/0x198
> [<ffffffff81010082>] ? check_events+0x12/0x20
> [<ffffffff812ab14d>] blktap_ring_poll+0x7c/0xc7
> [<ffffffff81124e9b>] do_select+0x387/0x584
> [<ffffffff81124b14>] ? do_select+0x0/0x584
> [<ffffffff811255de>] ? __pollwait+0x0/0xcc
> [<ffffffff811256aa>] ? pollwake+0x0/0x56
> [<ffffffff811256aa>] ? pollwake+0x0/0x56
> [<ffffffff811256aa>] ? pollwake+0x0/0x56
> [<ffffffff811256aa>] ? pollwake+0x0/0x56
> [<ffffffff8108059b>] ? __lock_acquire+0x16d6/0x16e5
> [<ffffffff8100f955>] ? xen_force_evtchn_callback+0xd/0xf
> [<ffffffff81010082>] ? check_events+0x12/0x20
> [<ffffffff8100f955>] ? xen_force_evtchn_callback+0xd/0xf
> [<ffffffff81010082>] ? check_events+0x12/0x20
> [<ffffffff8100f955>] ? xen_force_evtchn_callback+0xd/0xf
> [<ffffffff811252a4>] core_sys_select+0x20c/0x2da
> [<ffffffff811250d6>] ? core_sys_select+0x3e/0x2da
> [<ffffffff81010082>] ? check_events+0x12/0x20
> [<ffffffff8101006f>] ? xen_restore_fl_direct_end+0x0/0x1
> [<ffffffff81108661>] ? kmem_cache_free+0x18e/0x1c8
> [<ffffffff8141e912>] ? sock_destroy_inode+0x19/0x1b
> [<ffffffff811299bd>] ? destroy_inode+0x2f/0x44
> [<ffffffff8102ef22>] ? pvclock_clocksource_read+0x4b/0xa2
> [<ffffffff8100fe8b>] ? xen_clocksource_read+0x21/0x23
> [<ffffffff81010003>] ? xen_clocksource_get_cycles+0x9/0x16
> [<ffffffff81075700>] ? ktime_get_ts+0xb2/0xbf
> [<ffffffff811255b6>] sys_select+0x96/0xbe
> [<ffffffff81013d32>] system_call_fastpath+0x16/0x1b
> block tdb: sector-size: 512 capacity: 20971520
> block tdc: sector-size: 512 capacity: 146800640
> block tdd: sector-size: 512 capacity: 188743680
>
> J
>
next prev parent reply other threads:[~2010-09-07 1:46 UTC|newest]
Thread overview: 41+ messages / expand[flat|nested] mbox.gz Atom feed top
2010-08-23 6:54 Fix the occasional xen-blkfront deadlock, when irqbalancing Daniel Stodden
2010-08-23 6:54 ` [PATCH] blkfront: Move blkif_interrupt into a tasklet Daniel Stodden
2010-08-23 7:01 ` Daniel Stodden
2010-09-02 22:46 ` Jeremy Fitzhardinge
2010-09-02 23:08 ` Daniel Stodden
2010-09-07 1:39 ` blktap lockdep hiccup Jeremy Fitzhardinge
2010-09-07 1:46 ` Daniel Stodden [this message]
2010-09-08 2:03 ` [PATCH] blkfront: Move blkif_interrupt into a tasklet Jeremy Fitzhardinge
2010-09-08 2:21 ` Daniel Stodden
2010-09-08 6:37 ` Jeremy Fitzhardinge
2010-09-23 16:08 ` Andrew Jones
2010-09-23 16:23 ` Jeremy Fitzhardinge
2010-09-23 16:38 ` Paolo Bonzini
2010-09-23 18:36 ` Jeremy Fitzhardinge
2010-09-24 7:14 ` Andrew Jones
2010-09-24 18:50 ` Jeremy Fitzhardinge
2010-09-27 7:41 ` Andrew Jones
2010-09-27 9:46 ` Daniel Stodden
2010-09-27 10:21 ` Andrew Jones
2011-08-16 11:26 ` imammedo
2011-08-16 14:57 ` Konrad Rzeszutek Wilk
2011-08-17 2:38 ` Konrad Rzeszutek Wilk
2011-08-17 7:30 ` Paolo Bonzini
2011-08-17 9:07 ` Igor Mammedov
2011-08-24 15:36 ` Konrad Rzeszutek Wilk
2011-08-24 16:36 ` Igor Mammedov
2011-08-29 19:46 ` Konrad Rzeszutek Wilk
2011-08-31 23:47 ` [PATCH] xen: x86_32: do not enable iterrupts when returning from exception in interrupt context Igor Mammedov
2011-08-31 22:37 ` Jeremy Fitzhardinge
2011-09-01 8:19 ` Igor Mammedov
2011-09-01 11:46 ` [PATCH v2] " Igor Mammedov
2011-09-01 15:45 ` Konrad Rzeszutek Wilk
2011-09-01 16:46 ` Jeremy Fitzhardinge
2011-09-02 8:18 ` Igor Mammedov
2011-09-02 13:40 ` Konrad Rzeszutek Wilk
2011-09-02 14:01 ` [Xen-devel] " Igor Mammedov
2011-09-02 14:47 ` Konrad Rzeszutek Wilk
2011-09-06 9:16 ` Igor Mammedov
2011-09-02 9:19 ` Igor Mammedov
2011-09-02 10:00 ` Keir Fraser
2010-08-23 21:09 ` Fix the occasional xen-blkfront deadlock, when irqbalancing Jeremy Fitzhardinge
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=1283824017.3138.27.camel@agari.van.xensource.com \
--to=daniel.stodden@citrix.com \
--cc=jeremy@goop.org \
--cc=tek@acm.org \
--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).