All of lore.kernel.org
 help / color / mirror / Atom feed
From: Nicolas Boichat <drinkcat@chromium.org>
To: Mark Brown <broonie@kernel.org>
Cc: Mauro Carvalho Chehab <mchehab@osg.samsung.com>,
	Antti Palosaari <crope@iki.fi>, Ingo Molnar <mingo@redhat.com>,
	Arjan van de Ven <arjan@linux.intel.com>,
	Lars-Peter Clausen <lars@metafoo.de>,
	Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
	linux-kernel@vger.kernel.org, Bard Liao <bardliao@realtek.com>,
	Oder Chiou <oder_chiou@realtek.com>,
	Liam Girdwood <lgirdwood@gmail.com>,
	Jaroslav Kysela <perex@perex.cz>, Takashi Iwai <tiwai@suse.de>,
	alsa-devel@alsa-project.org,
	Anatol Pomozov <anatol.pomozov@gmail.com>
Subject: [RFC PATCH 2/2] ASoC: rt5677: Add lockdep class to silence lockdep warnings
Date: Thu, 25 Jun 2015 17:35:04 +0800	[thread overview]
Message-ID: <1435224904-35517-2-git-send-email-drinkcat@chromium.org> (raw)
In-Reply-To: <1435224904-35517-1-git-send-email-drinkcat@chromium.org>

Accessing rt5677->regmap requires read/write operations on
another regmap (rt5677->regmap_physical), which causes the
lockdep detector to throw a false warning, as both regmaps
are using the same lockdep class by default. Introduce a new
class for rt5677->regmap to silence this warning.

Sample warning:
[    2.569449] =============================================
[    2.569451] [ INFO: possible recursive locking detected ]
[    2.569454] 3.18.0 #311 Tainted: G S
[    2.569456] ---------------------------------------------
[    2.569458] swapper/0/1 is trying to acquire lock:
[    2.569469]  (&map->mutex){+.+...}, at: [<ffffffc00037dba0>]
regmap_lock_mutex+0x10/0x18
[    2.569470]
[    2.569470] but task is already holding lock:
[    2.569476]  (&map->mutex){+.+...}, at: [<ffffffc00037dba0>]
regmap_lock_mutex+0x10/0x18
[    2.569478]
[    2.569478] other info that might help us debug this:
[    2.569479]  Possible unsafe locking scenario:
[    2.569479]
[    2.569480]        CPU0
[    2.569481]        ----
[    2.569484]   lock(&map->mutex);
[    2.569486]   lock(&map->mutex);
[    2.569487]
[    2.569487]  *** DEADLOCK ***
[    2.569487]
[    2.569489]  May be due to missing lock nesting notation
[    2.569489]
[    2.569491] 3 locks held by swapper/0/1:
[    2.569499]  #0:  (&dev->mutex){......}, at: [<ffffffc000369e80>]
__driver_attach+0x38/0x98
[    2.569505]  #1:  (&dev->mutex){......}, at: [<ffffffc000369ea0>]
__driver_attach+0x58/0x98
[    2.569512]  #2:  (&map->mutex){+.+...}, at: [<ffffffc00037dba0>]
regmap_lock_mutex+0x10/0x18
[    2.569513]
[    2.569513] stack backtrace:
[    2.569517] CPU: 0 PID: 1 Comm: swapper/0 Tainted: G S
3.18.0 #311
[    2.569521] Call trace:
[    2.569527] [<ffffffc000089198>] dump_backtrace+0x0/0x108
[    2.569530] [<ffffffc0000892b0>] show_stack+0x10/0x1c
[    2.569535] [<ffffffc0005e09b4>] dump_stack+0x74/0x94
[    2.569540] [<ffffffc0000d5d1c>] __lock_acquire+0x73c/0x1910
[    2.569544] [<ffffffc0000d7674>] lock_acquire+0xec/0x128
[    2.569548] [<ffffffc0005e3870>] mutex_lock_nested+0x58/0x354
[    2.569551] [<ffffffc00037db9c>] regmap_lock_mutex+0xc/0x18
[    2.569554] [<ffffffc00037ef50>] regmap_read+0x34/0x68
[    2.569559] [<ffffffc0004ddc84>] rt5677_read+0x9c/0xb4
[    2.569562] [<ffffffc00037ee8c>] _regmap_read+0xa8/0x138
[    2.569565] [<ffffffc00037ef60>] regmap_read+0x44/0x68
[    2.569569] [<ffffffc0004dd6a4>] rt5677_i2c_probe+0x25c/0x4a4
[    2.569574] [<ffffffc00042984c>] i2c_device_probe+0xfc/0x130
[    2.569577] [<ffffffc000369c58>] driver_probe_device+0xd4/0x23c
[    2.569580] [<ffffffc000369eb0>] __driver_attach+0x68/0x98
[    2.569584] [<ffffffc000368dc8>] bus_for_each_dev+0x70/0x90
[    2.569588] [<ffffffc0003697e4>] driver_attach+0x1c/0x28
[    2.569591] [<ffffffc000369400>] bus_add_driver+0xd8/0x1e0
[    2.569594] [<ffffffc00036a7f0>] driver_register+0xbc/0x10c
[    2.569598] [<ffffffc00042a4bc>] i2c_register_driver+0x48/0xac
[    2.569601] [<ffffffc0008d523c>] rt5677_i2c_driver_init+0x14/0x20
[    2.569605] [<ffffffc0000828dc>] do_one_initcall+0xf4/0x18c
[    2.569609] [<ffffffc0008a4ae8>] kernel_init_freeable+0x144/0x1e4
[    2.569613] [<ffffffc0005de3a4>] kernel_init+0x10/0xd4

Signed-off-by: Nicolas Boichat <drinkcat@chromium.org>
---
 sound/soc/codecs/rt5677.c | 5 +++++
 1 file changed, 5 insertions(+)

diff --git a/sound/soc/codecs/rt5677.c b/sound/soc/codecs/rt5677.c
index 31d969a..3007d13 100644
--- a/sound/soc/codecs/rt5677.c
+++ b/sound/soc/codecs/rt5677.c
@@ -4998,6 +4998,7 @@ static const struct regmap_config rt5677_regmap_physical = {
 	.num_ranges = ARRAY_SIZE(rt5677_ranges),
 };
 
+static struct lock_class_key rt5677_regmap_key;
 static const struct regmap_config rt5677_regmap = {
 	.reg_bits = 8,
 	.val_bits = 16,
@@ -5015,6 +5016,10 @@ static const struct regmap_config rt5677_regmap = {
 	.num_reg_defaults = ARRAY_SIZE(rt5677_reg),
 	.ranges = rt5677_ranges,
 	.num_ranges = ARRAY_SIZE(rt5677_ranges),
+
+	/* Silence incorrect lockdep warnings, as rt5677_regmap_physical is
+	 * accessed from within rt5677_regmap. */
+	.lockdep_lock_class_key = &rt5677_regmap_key,
 };
 
 static const struct i2c_device_id rt5677_i2c_id[] = {
-- 
2.4.3.573.g4eafbef


  reply	other threads:[~2015-06-25  9:36 UTC|newest]

Thread overview: 32+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2015-06-25  9:35 [RFC PATCH 1/2] regmap: add configurable lock class key for lockdep Nicolas Boichat
2015-06-25  9:35 ` Nicolas Boichat [this message]
2015-06-25 13:21 ` Arjan van de Ven
2015-06-25 14:29   ` Mark Brown
2015-06-25 15:03   ` Lars-Peter Clausen
2015-06-25 15:03     ` Lars-Peter Clausen
2015-06-25 15:33     ` Mark Brown
2015-06-25 15:47       ` Lars-Peter Clausen
2015-06-25 15:47         ` Lars-Peter Clausen
2015-06-25 16:08         ` Mark Brown
2015-06-26  3:16           ` Nicolas Boichat
2015-06-26  3:16             ` Nicolas Boichat
2015-06-29 12:51             ` Nicolas Boichat
2015-06-29 12:59               ` Lars-Peter Clausen
2015-06-29 12:59                 ` Lars-Peter Clausen
2015-06-29 14:03                 ` Nicolas Boichat
2015-06-29 14:18                   ` Lars-Peter Clausen
2015-06-29 15:34                     ` Mark Brown
2015-06-30  4:56                       ` Nicolas Boichat
2015-06-30  4:56                         ` Nicolas Boichat
2015-06-30 11:02                         ` [alsa-devel] " Lars-Peter Clausen
2015-06-29 14:22                   ` Mark Brown
2015-06-29 14:35                     ` Arjan van de Ven
2015-06-29 15:32                       ` Mark Brown
2015-06-29 15:36                         ` Arjan van de Ven
2015-06-29 16:02                           ` Mark Brown
2015-06-29 14:19                 ` Mark Brown
2015-06-25 14:49 ` Mark Brown
2015-06-25 15:59 ` Lars-Peter Clausen
2015-06-26  2:34   ` Nicolas Boichat
2015-06-26  2:34     ` Nicolas Boichat
2015-06-26  8:31     ` Lars-Peter Clausen

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=1435224904-35517-2-git-send-email-drinkcat@chromium.org \
    --to=drinkcat@chromium.org \
    --cc=alsa-devel@alsa-project.org \
    --cc=anatol.pomozov@gmail.com \
    --cc=arjan@linux.intel.com \
    --cc=bardliao@realtek.com \
    --cc=broonie@kernel.org \
    --cc=crope@iki.fi \
    --cc=gregkh@linuxfoundation.org \
    --cc=lars@metafoo.de \
    --cc=lgirdwood@gmail.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=mchehab@osg.samsung.com \
    --cc=mingo@redhat.com \
    --cc=oder_chiou@realtek.com \
    --cc=perex@perex.cz \
    --cc=tiwai@suse.de \
    /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.