From: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
To: linux-kernel@vger.kernel.org
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
stable@vger.kernel.org, Takashi Iwai <tiwai@suse.de>,
Sasha Levin <sashal@kernel.org>,
"Kirill A . Shutemov" <kirill.shutemov@linux.intel.com>
Subject: [PATCH 4.9 62/62] ALSA: timer: Fix mutex deadlock at releasing card
Date: Mon, 4 Nov 2019 22:45:24 +0100 [thread overview]
Message-ID: <20191104212001.419204692@linuxfoundation.org> (raw)
In-Reply-To: <20191104211901.387893698@linuxfoundation.org>
From: Takashi Iwai <tiwai@suse.de>
[ Upstream commit a39331867335d4a94b6165e306265c9e24aca073 ]
When a card is disconnected while in use, the system waits until all
opened files are closed then releases the card. This is done via
put_device() of the card device in each device release code.
The recently reported mutex deadlock bug happens in this code path;
snd_timer_close() for the timer device deals with the global
register_mutex and it calls put_device() there. When this timer
device is the last one, the card gets freed and it eventually calls
snd_timer_free(), which has again the protection with the global
register_mutex -- boom.
Basically put_device() call itself is race-free, so a relative simple
workaround is to move this put_device() call out of the mutex. For
achieving that, in this patch, snd_timer_close_locked() got a new
argument to store the card device pointer in return, and each caller
invokes put_device() with the returned object after the mutex unlock.
Reported-and-tested-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>
Cc: <stable@vger.kernel.org>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
sound/core/timer.c | 24 +++++++++++++++++-------
1 file changed, 17 insertions(+), 7 deletions(-)
diff --git a/sound/core/timer.c b/sound/core/timer.c
index 6eb4e97662d9c..19d90aa082184 100644
--- a/sound/core/timer.c
+++ b/sound/core/timer.c
@@ -239,7 +239,8 @@ static int snd_timer_check_master(struct snd_timer_instance *master)
return 0;
}
-static int snd_timer_close_locked(struct snd_timer_instance *timeri);
+static int snd_timer_close_locked(struct snd_timer_instance *timeri,
+ struct device **card_devp_to_put);
/*
* open a timer instance
@@ -251,6 +252,7 @@ int snd_timer_open(struct snd_timer_instance **ti,
{
struct snd_timer *timer;
struct snd_timer_instance *timeri = NULL;
+ struct device *card_dev_to_put = NULL;
int err;
mutex_lock(®ister_mutex);
@@ -274,7 +276,7 @@ int snd_timer_open(struct snd_timer_instance **ti,
list_add_tail(&timeri->open_list, &snd_timer_slave_list);
err = snd_timer_check_slave(timeri);
if (err < 0) {
- snd_timer_close_locked(timeri);
+ snd_timer_close_locked(timeri, &card_dev_to_put);
timeri = NULL;
}
goto unlock;
@@ -326,7 +328,7 @@ int snd_timer_open(struct snd_timer_instance **ti,
timeri = NULL;
if (timer->card)
- put_device(&timer->card->card_dev);
+ card_dev_to_put = &timer->card->card_dev;
module_put(timer->module);
goto unlock;
}
@@ -336,12 +338,15 @@ int snd_timer_open(struct snd_timer_instance **ti,
timer->num_instances++;
err = snd_timer_check_master(timeri);
if (err < 0) {
- snd_timer_close_locked(timeri);
+ snd_timer_close_locked(timeri, &card_dev_to_put);
timeri = NULL;
}
unlock:
mutex_unlock(®ister_mutex);
+ /* put_device() is called after unlock for avoiding deadlock */
+ if (card_dev_to_put)
+ put_device(card_dev_to_put);
*ti = timeri;
return err;
}
@@ -351,7 +356,8 @@ EXPORT_SYMBOL(snd_timer_open);
* close a timer instance
* call this with register_mutex down.
*/
-static int snd_timer_close_locked(struct snd_timer_instance *timeri)
+static int snd_timer_close_locked(struct snd_timer_instance *timeri,
+ struct device **card_devp_to_put)
{
struct snd_timer *timer = NULL;
struct snd_timer_instance *slave, *tmp;
@@ -403,7 +409,7 @@ static int snd_timer_close_locked(struct snd_timer_instance *timeri)
timer->hw.close(timer);
/* release a card refcount for safe disconnection */
if (timer->card)
- put_device(&timer->card->card_dev);
+ *card_devp_to_put = &timer->card->card_dev;
module_put(timer->module);
}
@@ -415,14 +421,18 @@ static int snd_timer_close_locked(struct snd_timer_instance *timeri)
*/
int snd_timer_close(struct snd_timer_instance *timeri)
{
+ struct device *card_dev_to_put = NULL;
int err;
if (snd_BUG_ON(!timeri))
return -ENXIO;
mutex_lock(®ister_mutex);
- err = snd_timer_close_locked(timeri);
+ err = snd_timer_close_locked(timeri, &card_dev_to_put);
mutex_unlock(®ister_mutex);
+ /* put_device() is called after unlock for avoiding deadlock */
+ if (card_dev_to_put)
+ put_device(card_dev_to_put);
return err;
}
EXPORT_SYMBOL(snd_timer_close);
--
2.20.1
next prev parent reply other threads:[~2019-11-04 22:25 UTC|newest]
Thread overview: 72+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-11-04 21:44 [PATCH 4.9 00/62] 4.9.199-stable review Greg Kroah-Hartman
2019-11-04 21:44 ` [PATCH 4.9 01/62] dm snapshot: use mutex instead of rw_semaphore Greg Kroah-Hartman
2019-11-04 21:44 ` [PATCH 4.9 02/62] dm snapshot: introduce account_start_copy() and account_end_copy() Greg Kroah-Hartman
2019-11-04 21:44 ` [PATCH 4.9 03/62] dm snapshot: rework COW throttling to fix deadlock Greg Kroah-Hartman
2019-11-04 21:44 ` [PATCH 4.9 04/62] dm: Use kzalloc for all structs with embedded biosets/mempools Greg Kroah-Hartman
2019-11-04 21:44 ` [PATCH 4.9 05/62] sc16is7xx: Fix for "Unexpected interrupt: 8" Greg Kroah-Hartman
2019-11-04 21:44 ` [PATCH 4.9 06/62] HID: i2c-hid: add Direkt-Tek DTLAPY133-1 to descriptor override Greg Kroah-Hartman
2019-11-04 21:44 ` [PATCH 4.9 07/62] x86/cpu: Add Atom Tremont (Jacobsville) Greg Kroah-Hartman
2019-11-04 21:44 ` [PATCH 4.9 08/62] HID: i2c-hid: Add Odys Winbook 13 to descriptor override Greg Kroah-Hartman
2019-11-04 21:44 ` [PATCH 4.9 09/62] scripts/setlocalversion: Improve -dirty check with git-status --no-optional-locks Greg Kroah-Hartman
2019-11-04 21:44 ` [PATCH 4.9 10/62] usb: handle warm-reset port requests on hub resume Greg Kroah-Hartman
2019-11-04 21:44 ` [PATCH 4.9 11/62] rtc: pcf8523: set xtal load capacitance from DT Greg Kroah-Hartman
2019-11-04 21:44 ` [PATCH 4.9 12/62] exec: load_script: Do not exec truncated interpreter path Greg Kroah-Hartman
2019-11-04 21:44 ` [PATCH 4.9 13/62] iio: fix center temperature of bmc150-accel-core Greg Kroah-Hartman
2019-11-04 21:44 ` [PATCH 4.9 14/62] perf map: Fix overlapped map handling Greg Kroah-Hartman
2019-11-04 21:44 ` [PATCH 4.9 15/62] perf jevents: Fix period for Intel fixed counters Greg Kroah-Hartman
2019-11-04 21:44 ` [PATCH 4.9 16/62] staging: rtl8188eu: fix null dereference when kzalloc fails Greg Kroah-Hartman
2019-11-04 21:44 ` [PATCH 4.9 17/62] RDMA/iwcm: Fix a lock inversion issue Greg Kroah-Hartman
2019-11-04 21:44 ` [PATCH 4.9 18/62] gpio: max77620: Use correct unit for debounce times Greg Kroah-Hartman
2019-11-04 21:44 ` [PATCH 4.9 19/62] fs: cifs: mute -Wunused-const-variable message Greg Kroah-Hartman
2019-11-04 21:44 ` [PATCH 4.9 20/62] serial: mctrl_gpio: Check for NULL pointer Greg Kroah-Hartman
2019-11-04 21:44 ` [PATCH 4.9 21/62] efi/cper: Fix endianness of PCIe class code Greg Kroah-Hartman
2019-11-04 21:44 ` [PATCH 4.9 22/62] efi/x86: Do not clean dummy variable in kexec path Greg Kroah-Hartman
2019-11-04 21:44 ` [PATCH 4.9 23/62] ocfs2: clear zero in unaligned direct IO Greg Kroah-Hartman
2019-11-04 21:44 ` [PATCH 4.9 24/62] fs: ocfs2: fix possible null-pointer dereferences in ocfs2_xa_prepare_entry() Greg Kroah-Hartman
2019-11-04 21:44 ` [PATCH 4.9 25/62] fs: ocfs2: fix a possible null-pointer dereference in ocfs2_write_end_nolock() Greg Kroah-Hartman
2019-11-04 21:44 ` [PATCH 4.9 26/62] fs: ocfs2: fix a possible null-pointer dereference in ocfs2_info_scan_inode_alloc() Greg Kroah-Hartman
2019-11-04 21:44 ` [PATCH 4.9 27/62] MIPS: fw: sni: Fix out of bounds init of o32 stack Greg Kroah-Hartman
2019-11-04 21:44 ` [PATCH 4.9 28/62] NFSv4: Fix leak of clp->cl_acceptor string Greg Kroah-Hartman
2019-11-04 21:44 ` [PATCH 4.9 29/62] s390/uaccess: avoid (false positive) compiler warnings Greg Kroah-Hartman
2019-11-04 21:44 ` [PATCH 4.9 30/62] tracing: Initialize iter->seq after zeroing in tracing_read_pipe() Greg Kroah-Hartman
2019-11-04 21:44 ` [PATCH 4.9 31/62] USB: legousbtower: fix a signedness bug in tower_probe() Greg Kroah-Hartman
2019-11-04 21:44 ` [PATCH 4.9 32/62] thunderbolt: Use 32-bit writes when writing ring producer/consumer Greg Kroah-Hartman
2019-11-04 21:44 ` [PATCH 4.9 33/62] ath6kl: fix a NULL-ptr-deref bug in ath6kl_usb_alloc_urb_from_pipe() Greg Kroah-Hartman
2019-11-04 21:44 ` [PATCH 4.9 34/62] fuse: flush dirty data/metadata before non-truncate setattr Greg Kroah-Hartman
2019-11-04 21:44 ` [PATCH 4.9 35/62] fuse: truncate pending writes on O_TRUNC Greg Kroah-Hartman
2019-11-04 21:44 ` [PATCH 4.9 36/62] ALSA: bebob: Fix prototype of helper function to return negative value Greg Kroah-Hartman
2019-11-04 21:44 ` [PATCH 4.9 37/62] UAS: Revert commit 3ae62a42090f ("UAS: fix alignment of scatter/gather segments") Greg Kroah-Hartman
2019-11-05 14:31 ` Oliver Neukum
2019-11-06 0:11 ` Sasha Levin
2019-11-06 7:45 ` Oliver Neukum
2019-11-06 11:17 ` Greg Kroah-Hartman
2019-11-04 21:45 ` [PATCH 4.9 38/62] USB: gadget: Reject endpoints with 0 maxpacket value Greg Kroah-Hartman
2019-11-04 21:45 ` [PATCH 4.9 39/62] usb-storage: Revert commit 747668dbc061 ("usb-storage: Set virt_boundary_mask to avoid SG overflows") Greg Kroah-Hartman
2019-11-04 21:45 ` [PATCH 4.9 40/62] USB: ldusb: fix ring-buffer locking Greg Kroah-Hartman
2019-11-04 21:45 ` [PATCH 4.9 41/62] USB: ldusb: fix control-message timeout Greg Kroah-Hartman
2019-11-04 21:45 ` [PATCH 4.9 42/62] USB: serial: whiteheat: fix potential slab corruption Greg Kroah-Hartman
2019-11-04 21:45 ` [PATCH 4.9 43/62] USB: serial: whiteheat: fix line-speed endianness Greg Kroah-Hartman
2019-11-04 21:45 ` [PATCH 4.9 44/62] HID: i2c-hid: add Trekstor Primebook C11B to descriptor override Greg Kroah-Hartman
2019-11-04 21:45 ` [PATCH 4.9 45/62] HID: Fix assumption that devices have inputs Greg Kroah-Hartman
2019-11-04 21:45 ` [PATCH 4.9 46/62] HID: fix error message in hid_open_report() Greg Kroah-Hartman
2019-11-04 21:45 ` [PATCH 4.9 47/62] nl80211: fix validation of mesh path nexthop Greg Kroah-Hartman
2019-11-04 21:45 ` [PATCH 4.9 48/62] s390/cmm: fix information leak in cmm_timeout_handler() Greg Kroah-Hartman
2019-11-04 21:45 ` [PATCH 4.9 49/62] rtlwifi: Fix potential overflow on P2P code Greg Kroah-Hartman
2019-11-04 21:45 ` [PATCH 4.9 50/62] dmaengine: cppi41: Fix cppi41_dma_prep_slave_sg() when idle Greg Kroah-Hartman
2019-11-04 21:45 ` [PATCH 4.9 51/62] llc: fix sk_buff leak in llc_sap_state_process() Greg Kroah-Hartman
2019-11-04 21:45 ` [PATCH 4.9 52/62] llc: fix sk_buff leak in llc_conn_service() Greg Kroah-Hartman
2019-11-04 21:45 ` [PATCH 4.9 53/62] bonding: fix potential NULL deref in bond_update_slave_arr Greg Kroah-Hartman
2019-11-04 21:45 ` [PATCH 4.9 54/62] net: usb: sr9800: fix uninitialized local variable Greg Kroah-Hartman
2019-11-04 21:45 ` [PATCH 4.9 55/62] sch_netem: fix rcu splat in netem_enqueue() Greg Kroah-Hartman
2019-11-04 21:45 ` [PATCH 4.9 56/62] sctp: fix the issue that flags are ignored when using kernel_connect Greg Kroah-Hartman
2019-11-04 21:45 ` [PATCH 4.9 57/62] sctp: not bind the socket in sctp_connect Greg Kroah-Hartman
2019-11-04 21:45 ` [PATCH 4.9 58/62] xfs: Correctly invert xfs_buftarg LRU isolation logic Greg Kroah-Hartman
2019-11-04 21:45 ` [PATCH 4.9 59/62] ALSA: timer: Follow standard EXPORT_SYMBOL() declarations Greg Kroah-Hartman
2019-11-04 21:45 ` [PATCH 4.9 60/62] ALSA: timer: Limit max instances per timer Greg Kroah-Hartman
2019-11-04 21:45 ` [PATCH 4.9 61/62] ALSA: timer: Simplify error path in snd_timer_open() Greg Kroah-Hartman
2019-11-04 21:45 ` Greg Kroah-Hartman [this message]
2019-11-05 5:56 ` [PATCH 4.9 00/62] 4.9.199-stable review kernelci.org bot
2019-11-05 7:05 ` Naresh Kamboju
2019-11-05 14:24 ` Guenter Roeck
2019-11-05 17:01 ` shuah
2019-11-05 23:37 ` Jon Hunter
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=20191104212001.419204692@linuxfoundation.org \
--to=gregkh@linuxfoundation.org \
--cc=kirill.shutemov@linux.intel.com \
--cc=linux-kernel@vger.kernel.org \
--cc=sashal@kernel.org \
--cc=stable@vger.kernel.org \
--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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox