From: Nishanth Aravamudan <nacc@linux.vnet.ibm.com>
To: netdev@vger.kernel.org
Cc: Matt Carlson <mcarlson@broadcom.com>,
Michael Chan <mchan@broadcom.com>,
Grant Likely <grant.likely@secretlab.ca>,
Rob Herring <rob.herring@calxeda.com>,
linux-kernel@vger.kernel.org
Subject: tg3: possible irq lock inversion dependency detected
Date: Fri, 19 Apr 2013 12:01:51 -0700 [thread overview]
Message-ID: <20130419190150.GA15355@linux.vnet.ibm.com> (raw)
Running 3.9-rc7-ish, tripped the following (also being seen in FC19
alpha) on ppc64:
[ 117.026196] =========================================================
[ 117.026216] [ INFO: possible irq lock inversion dependency detected ]
[ 117.026238] 3.9.0-rc7+ #8 Not tainted
[ 117.026251] ---------------------------------------------------------
[ 117.026271] swapper/7/0 just changed the state of lock:
[ 117.026286] (&(&tp->lock)->rlock){+.-...}, at: [<c00000000064effc>] .tg3_timer+0x9c/0x10f0
[ 117.026334] but this lock took another, SOFTIRQ-unsafe lock in the past:
[ 117.026353] (devtree_lock){+.+...}
and interrupts could create inverse lock ordering between them.
[ 117.026389]
other info that might help us debug this:
[ 117.026409] Chain exists of:
&(&tp->lock)->rlock --> pci_lock --> devtree_lock
[ 117.026452] Possible interrupt unsafe locking scenario:
[ 117.026472] CPU0 CPU1
[ 117.026488] ---- ----
[ 117.026503] lock(devtree_lock);
[ 117.026521] local_irq_disable();
[ 117.026540] lock(&(&tp->lock)->rlock);
[ 117.026567] lock(pci_lock);
[ 117.026591] <Interrupt>
[ 117.026600] lock(&(&tp->lock)->rlock);
[ 117.026621]
*** DEADLOCK ***
[ 117.026644] 1 lock held by swapper/7/0:
[ 117.026657] #0: ((&tp->timer)){+.-...}, at: [<c0000000000bbc50>] .call_timer_fn+0x0/0x3e0
[ 117.026698]
the shortest dependencies between 2nd lock and 1st lock:
[ 117.026722] -> (devtree_lock){+.+...} ops: 13864154431488 {
[ 117.026758] HARDIRQ-ON-W at:
[ 117.026772] SOFTIRQ-ON-W at:
[ 117.026786] INITIAL USE at:
[ 117.026799] }
[ 117.026809] ... key at: [<c00000000159b208>] devtree_lock+0x18/0x48
[ 117.026835] ... acquired at:
[ 117.026857] -> (pci_lock){......} ops: 3955664879616 {
[ 117.026893] INITIAL USE at:
[ 117.026906] }
[ 117.026915] ... key at: [<c00000000157abe8>] pci_lock+0x18/0x48
[ 117.026942] ... acquired at:
[ 117.026963] -> (&(&tp->lock)->rlock){+.-...} ops: 154618822656 {
[ 117.027001] HARDIRQ-ON-W at:
[ 117.027015] IN-SOFTIRQ-W at:
[ 117.027029] INITIAL USE at:
[ 117.027042] }
[ 117.027051] ... key at: [<c0000000023a0311>] __key.48770+0x0/0x8
[ 117.027078] ... acquired at:
[ 117.027100]
stack backtrace:
[ 117.027116] Call Trace:
[ 117.027129] [c0000003a7c7ad10] [c000000000016ee0] .show_stack+0xd0/0x1f0 (unreliable)
[ 117.027161] [c0000003a7c7ade0] [c000000000137d78] .print_irq_inversion_bug+0x298/0x2f0
[ 117.027189] [c0000003a7c7ae80] [c000000000137e70] .check_usage_forwards+0xa0/0x150
[ 117.027217] [c0000003a7c7af70] [c000000000138b68] .mark_lock+0x258/0x7b0
[ 117.027242] [c0000003a7c7b030] [c0000000001396f8] .__lock_acquire+0x638/0x1c20
[ 117.027270] [c0000003a7c7b1b0] [c00000000013b480] .lock_acquire+0xb0/0x250
[ 117.027296] [c0000003a7c7b290] [c00000000093a1cc] ._raw_spin_lock+0x5c/0xc0
[ 117.027321] [c0000003a7c7b320] [c00000000064effc] .tg3_timer+0x9c/0x10f0
[ 117.027345] [c0000003a7c7b3d0] [c0000000000bbd0c] .call_timer_fn+0xbc/0x3e0
[ 117.027369] [c0000003a7c7b4b0] [c0000000000bc328] .run_timer_softirq+0x2f8/0x440
[ 117.027398] [c0000003a7c7b5c0] [c0000000000afef8] .__do_softirq+0x1b8/0x540
[ 117.027423] [c0000003a7c7b6f0] [c0000000000b04e8] .irq_exit+0xe8/0x100
[ 117.027448] [c0000003a7c7b770] [c000000000020464] .timer_interrupt+0x174/0x510
[ 117.027477] [c0000003a7c7b830] [c0000000000024f4] decrementer_common+0x174/0x180
[ 117.027509] --- Exception: 901 at .plpar_hcall_norets+0x84/0xd4
LR = .check_and_cede_processor+0x48/0x80
[ 117.027544] [c0000003a7c7bb20] [c000000000085168] .check_and_cede_processor+0x18/0x80 (unreliable)
[ 117.027577] [c0000003a7c7bb90] [c000000000085258] .dedicated_cede_loop+0x88/0x150
[ 117.027606] [c0000003a7c7bc50] [c00000000075808c] .cpuidle_enter+0x2c/0x40
[ 117.027632] [c0000003a7c7bcc0] [c0000000007589bc] .cpuidle_idle_call+0xfc/0x4e0
[ 117.027660] [c0000003a7c7bd70] [c000000000074be8] .pSeries_idle+0x18/0x40
[ 117.027685] [c0000003a7c7bde0] [c000000000018ec0] .cpu_idle+0x180/0x360
[ 117.027711] [c0000003a7c7bea0] [c00000000095391c] .start_secondary+0x534/0x53c
[ 117.027739] [c0000003a7c7bf90] [c0000000000097fc] .start_secondary_prolog+0x10/0x14
next reply other threads:[~2013-04-19 19:02 UTC|newest]
Thread overview: 4+ messages / expand[flat|nested] mbox.gz Atom feed top
2013-04-19 19:01 Nishanth Aravamudan [this message]
2013-04-19 19:23 ` tg3: possible irq lock inversion dependency detected Rob Herring
2013-04-19 22:21 ` Nishanth Aravamudan
2013-04-22 9:30 ` Thomas Gleixner
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=20130419190150.GA15355@linux.vnet.ibm.com \
--to=nacc@linux.vnet.ibm.com \
--cc=grant.likely@secretlab.ca \
--cc=linux-kernel@vger.kernel.org \
--cc=mcarlson@broadcom.com \
--cc=mchan@broadcom.com \
--cc=netdev@vger.kernel.org \
--cc=rob.herring@calxeda.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.