From: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
To: Mark Hurenkamp <mark.hurenkamp@xs4all.nl>
Cc: xen-devel@lists.xensource.com
Subject: Re: radeon in dom0/ivtv in domU: irq 16 nobody cared
Date: Mon, 19 Apr 2010 10:08:49 -0400 [thread overview]
Message-ID: <20100419140849.GA20776@phenom.dumpdata.com> (raw)
In-Reply-To: <201004170238.31591.mark.hurenkamp@xs4all.nl>
On Sat, Apr 17, 2010 at 02:38:31AM +0200, Mark Hurenkamp wrote:
> > > Anyhow, what I would like you test is to provide 'noirqdebug' flag on
> > > your Linux kernel command line. That should take care of the 'nobody
> > > cared IRQ'. Please test and see if you your machine is more or less
> > > sluggish.
> I have had my system up and running for several days now, and indeed
> the IRQ nobody cared has not occured. I do not see my system behave
> any more sluggish than before. All the scheduled recordings seem to have
> been recorded correctly (sometimes up to 3 in parallel).
Excellent. Then the patch I am working on will make your machine work
fine.
>
> > > Also, please do run this:
> > > while (true)
> > > do
> > > sleep 10
> > > cat /proc/interrupts | grep 16
> > > done
> > >
> > > during the lifecycle of the whole machine. I am curious to see whether
> > > you also get an IRQ storm (or if it is just a trickle - which is OK)
> > > after launching the DomU guest.
> Having done so, i notice that the count in dom0 and domU is identical,
> and seems to increase with about ~100 or so per second while domU is
> recording from /dev/video0. To me that does not seem unusual.
>
>
> Probably unrelated, but although i do have compiz running, and quite some eye-
> candy enabled in my dom0 Kde environement, the radeon driver still doesn't
> seem to generate any interupts.
>
>
> And probably also unrelated, dmesg in the domU shows a lot of these messages:
Ugh. No idea.
> xennet: skb rides the rocket: 20 frags
> Pid: 0, comm: swapper Not tainted 2.6.32mh25 #6
> Call Trace:
> <IRQ> [<ffffffffa00b941a>] xennet_start_xmit+0x7d/0x680 [xen_netfront]
> [<ffffffff8100eb49>] ? xen_force_evtchn_callback+0xd/0xf
> [<ffffffff8100f442>] ? check_events+0x12/0x20
> [<ffffffff8108fd61>] ? lock_acquired+0x255/0x264
> [<ffffffff813dcc11>] dev_hard_start_xmit+0x23b/0x2e6
> [<ffffffff813efcd3>] sch_direct_xmit+0x6a/0x178
> [<ffffffff813dd04d>] dev_queue_xmit+0x279/0x3e2
> [<ffffffff813dcf3d>] ? dev_queue_xmit+0x169/0x3e2
> [<ffffffff813faaf2>] ? rcu_read_unlock+0x2b/0x46
> [<ffffffff813fad61>] ? nf_hook_slow+0xcf/0xe0
> [<ffffffff8140f222>] ? ip_finish_output+0x0/0x71
> [<ffffffff8140e1b0>] ip_finish_output2+0x238/0x281
> [<ffffffff8140f291>] ip_finish_output+0x6f/0x71
> [<ffffffff8140f35f>] ip_output+0xcc/0xd5
> [<ffffffff8140d894>] dst_output+0x10/0x12
> [<ffffffff8140de17>] ip_local_out+0x23/0x27
> [<ffffffff8140e507>] ip_queue_xmit+0x30e/0x3a4
> [<ffffffff8100eb49>] ? xen_force_evtchn_callback+0xd/0xf
> [<ffffffff8100f442>] ? check_events+0x12/0x20
> [<ffffffff81421e1a>] tcp_transmit_skb+0x672/0x6b0
> [<ffffffff8108d003>] ? mark_lock+0x27/0x21e
> [<ffffffff81423a01>] tcp_write_xmit+0x7cd/0x8bd
> [<ffffffff814209ef>] ? tcp_established_options+0x33/0xae
> [<ffffffff81423b56>] __tcp_push_pending_frames+0x2f/0x5c
> [<ffffffff8141d352>] tcp_data_snd_check+0x29/0xef
> [<ffffffff8141f6b7>] tcp_rcv_established+0xd2/0x6a9
> [<ffffffff8100f442>] ? check_events+0x12/0x20
> [<ffffffff81426dec>] tcp_v4_do_rcv+0x1bf/0x367
> [<ffffffff814282df>] ? tcp_v4_rcv+0x2b8/0x66d
> [<ffffffff814282df>] ? tcp_v4_rcv+0x2b8/0x66d
> [<ffffffff81428482>] tcp_v4_rcv+0x45b/0x66d
> [<ffffffff8140b6d6>] ip_local_deliver_finish+0x154/0x200
> [<ffffffff8140b7f9>] ip_local_deliver+0x77/0x7e
> [<ffffffff8140b120>] ip_rcv_finish+0x3a9/0x3d5
> [<ffffffff8140b3d5>] ip_rcv+0x289/0x2c9
> [<ffffffff813dbd72>] netif_receive_skb+0x415/0x43f
> [<ffffffffa00ba56d>] xennet_poll+0xaf9/0xca4 [xen_netfront]
> [<ffffffff813dc546>] net_rx_action+0xc8/0x21a
> [<ffffffff81062f24>] __do_softirq+0x11c/0x1fd
> [<ffffffff8124dd9b>] ? _raw_spin_unlock+0x8f/0x98
> [<ffffffff81013f9c>] call_softirq+0x1c/0x30
> [<ffffffff81015523>] do_softirq+0x4b/0xa6
> [<ffffffff8106299b>] irq_exit+0x4a/0x99
> [<ffffffff812c08dd>] xen_evtchn_do_upcall+0x1af/0x1ec
> [<ffffffff81013fee>] xen_do_hypervisor_callback+0x1e/0x30
> <EOI> [<ffffffff810093aa>] ? hypercall_page+0x3aa/0x100b
> [<ffffffff810093aa>] ? hypercall_page+0x3aa/0x100b
> [<ffffffff8100eb7c>] ? xen_safe_halt+0x10/0x1a
> [<ffffffff8100c141>] ? xen_idle+0x51/0x5c
> [<ffffffff81011d14>] ? cpu_idle+0xb0/0x107
> [<ffffffff814678eb>] ? rest_init+0x7f/0x81
> [<ffffffff81986d76>] ? start_kernel+0x417/0x422
> [<ffffffff819862c1>] ? x86_64_start_reservations+0xac/0xb0
> [<ffffffff81989f01>] ? xen_start_kernel+0x648/0x64f
>
> And dmesg in dom0 (seems related to a chrome crash):
This is not good. Can you describe how to reproduce this.
> FIX bio-0: Marking all objects used
> chrome[12644]: segfault at 7f6cc3b5ad0c ip 0000000000c10c47 sp 00007fff88da2a80
> error 4 in chrome[400000+262f000]
> =============================================================================
> BUG bio-0: Poison overwritten
> -----------------------------------------------------------------------------
>
> INFO: 0xffff88002fcfe58b-0xffff88002fcfe58b. First byte 0xeb instead of 0x6b
> INFO: Allocated in mempool_alloc_slab+0x15/0x17 age=611 cpu=0 pid=10074
> INFO: Freed in mempool_free_slab+0x17/0x19 age=609 cpu=0 pid=0
> INFO: Slab 0xffffea000136c730 objects=32 used=31 fp=0xffff88002fcfe500
> flags=0x200000000040c3
> INFO: Object 0xffff88002fcfe500 @offset=1280 fp=0x(null)
>
> Bytes b4 0xffff88002fcfe4f0: 5a 5a 5a 5a 5a 5a 5a 5a 5a 5a 5a 5a 5a 5a 5a 5a
> ZZZZZZZZZZZZZZZZ
> Object 0xffff88002fcfe500: 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b
> kkkkkkkkkkkkkkkk
> Object 0xffff88002fcfe510: 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b
> kkkkkkkkkkkkkkkk
> Object 0xffff88002fcfe520: 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b
> kkkkkkkkkkkkkkkk
> Object 0xffff88002fcfe530: 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b
> kkkkkkkkkkkkkkkk
> Object 0xffff88002fcfe540: 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b
> kkkkkkkkkkkkkkkk
> Object 0xffff88002fcfe550: 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b > kkkkkkkkkkkkkkkk
> Object 0xffff88002fcfe560: 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b
> kkkkkkkkkkkkkkkk
> Object 0xffff88002fcfe570: 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b
> kkkkkkkkkkkkkkkk
> Object 0xffff88002fcfe580: 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b eb 6b 6b 6b 6b
> kkkkkkkkkkkëkkkk
> Object 0xffff88002fcfe590: 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b
> kkkkkkkkkkkkkkkk
> Object 0xffff88002fcfe5a0: 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b a5
> kkkkkkkkkkkkkkk¥
> Redzone 0xffff88002fcfe5b0: bb bb bb bb bb bb bb bb
> »»»»»»»»
> Padding 0xffff88002fcfe5f0: 5a 5a 5a 5a 5a 5a 5a 5a 5a 5a 5a 5a 5a 5a 5a 5a
> ZZZZZZZZZZZZZZZZ
> Pid: 914, comm: kjournald Tainted: G B W 2.6.32mh25 #6
> Call Trace:
> [<ffffffff8111aaf8>] print_trailer+0x140/0x149
> [<ffffffff8111b024>] check_bytes_and_report+0xba/0xeb
> [<ffffffff8111b11f>] check_object+0xca/0x1ae
> [<ffffffff8111d239>] __slab_alloc+0x3a6/0x454
> [<ffffffff810e80f6>] ? mempool_alloc_slab+0x15/0x17
> [<ffffffff810e80f6>] ? mempool_alloc_slab+0x15/0x17
> [<ffffffff8111d897>] kmem_cache_alloc+0xb4/0x173
> [<ffffffff810e80f6>] mempool_alloc_slab+0x15/0x17
> [<ffffffff810e81df>] mempool_alloc+0x6c/0x11e
> [<ffffffff8100eb49>] ? xen_force_evtchn_callback+0xd/0xf
> [<ffffffff8100eb49>] ? xen_force_evtchn_callback+0xd/0xf
> [<ffffffff81155924>] bio_alloc_bioset+0x35/0xc2
> [<ffffffff811559db>] bio_clone+0x2a/0x7c
> [<ffffffffa001b9c0>] make_request+0x51b/0x697 [raid1]
> [<ffffffff813952ec>] md_make_request+0xcf/0x10d
> [<ffffffff8107ff97>] ? up_read+0x2b/0x2f
> [<ffffffff81230263>] generic_make_request+0x2ca/0x323
> [<ffffffff81480738>] ? sub_preempt_count+0x9/0x83
> [<ffffffff81230389>] submit_bio+0xcd/0xea
> [<ffffffff81150e27>] submit_bh+0xf4/0x117
> [<ffffffff811d6f6d>] journal_commit_transaction+0x95b/0xeea
> [<ffffffff8147dcf0>] ? _spin_unlock_irqrestore+0x70/0x7e
> [<ffffffff8106be6d>] ? try_to_del_timer_sync+0x73/0x81
> [<ffffffff8106be7b>] ? del_timer_sync+0x0/0xa6
> [<ffffffff811da626>] kjournald+0xf0/0x23c
> [<ffffffff8107c1fa>] ? autoremove_wake_function+0x0/0x39
> [<ffffffff811da536>] ? kjournald+0x0/0x23c
> [<ffffffff8107bedc>] kthread+0x7f/0x87
> [<ffffffff81013e9a>] child_rip+0xa/0x20
> [<ffffffff810137d4>] ? restore_args+0x0/0x30
> [<ffffffff8100f442>] ? check_events+0x12/0x20
> [<ffffffff8100f442>] ? check_events+0x12/0x20
> [<ffffffff81013e90>] ? child_rip+0x0/0x20
> FIX bio-0: Restoring 0xffff88002fcfe58b-0xffff88002fcfe58b=0x6b
>
>
> Regards,
> Mark.
>
> _______________________________________________
> Xen-devel mailing list
> Xen-devel@lists.xensource.com
> http://lists.xensource.com/xen-devel
prev parent reply other threads:[~2010-04-19 14:08 UTC|newest]
Thread overview: 15+ messages / expand[flat|nested] mbox.gz Atom feed top
2010-04-07 22:45 radeon in dom0/ivtv in domU: irq 16 nobody cared Mark Hurenkamp
2010-04-08 0:19 ` Konrad Rzeszutek Wilk
2010-04-08 17:37 ` Konrad Rzeszutek Wilk
2010-04-08 18:45 ` Jeremy Fitzhardinge
2010-04-13 13:18 ` Konrad Rzeszutek Wilk
2010-04-13 18:30 ` Jeremy Fitzhardinge
2010-04-13 19:10 ` Konrad Rzeszutek Wilk
2010-04-13 19:21 ` Jeremy Fitzhardinge
2010-04-08 23:47 ` Mark Hurenkamp
2010-04-09 0:10 ` Mark Hurenkamp
2010-04-13 12:57 ` Konrad Rzeszutek Wilk
2010-04-13 20:05 ` Mark Hurenkamp
2010-04-13 20:43 ` Jeremy Fitzhardinge
2010-04-17 0:38 ` Mark Hurenkamp
2010-04-19 14:08 ` Konrad Rzeszutek Wilk [this message]
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=20100419140849.GA20776@phenom.dumpdata.com \
--to=konrad.wilk@oracle.com \
--cc=mark.hurenkamp@xs4all.nl \
--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 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.