lists.openwall.net   lists  /  announce  owl-users  owl-dev  john-users  john-dev  passwdqc-users  yescrypt  popa3d-users  /  oss-security  kernel-hardening  musl  sabotage  tlsify  passwords  /  crypt-dev  xvendor  /  Bugtraq  Full-Disclosure  linux-kernel  linux-netdev  linux-ext4  linux-hardening  linux-cve-announce  PHC 
Open Source and information security mailing list archives
 
Hash Suite: Windows password security audit tool. GUI, reports in PDF.
[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <s5hy3qhe1o2.wl-tiwai@suse.de>
Date:   Fri, 18 Aug 2017 10:10:53 +0200
From:   Takashi Iwai <tiwai@...e.de>
To:     "Stephen Barber" <smbarber@...omium.org>
Cc:     <alsa-devel@...a-project.org>, "Dylan Reid" <dgreid@...omium.org>,
        "Bhumika Goyal" <bhumirks@...il.com>,
        "Con Kolivas" <con@...ivas.org>,
        "Jaroslav Kysela" <perex@...ex.cz>, <linux-kernel@...r.kernel.org>
Subject: Re: [PATCH] ALSA: usb-audio: don't retry snd_usb_ctl_msg after timeout

On Fri, 18 Aug 2017 00:17:46 +0200,
Stephen Barber wrote:
> 
> A few calls to snd_usb_ctl_msg wrap the function in a retry loop. In
> the worst case, the timeout for snd_usb_ctl_msg is 5 seconds, which when
> retried 10 times (for example, if a device is removed) could cause a
> probe to hang for ~50 seconds.
> 
> Example stack trace from 3.14 which triggered a hung task timeout:
> Call Trace:
>  [<ffffffffa2c1f720>] ? inet6_set_link_af.part.35+0x12/0x12
>  [<ffffffffa2c20309>] schedule+0x6e/0x70
>  [<ffffffffa2c1f81c>] schedule_timeout+0xfc/0x13c
>  [<ffffffffa2667bbc>] ? rcu_read_unlock_sched_notrace+0x17/0x17
>  [<ffffffffa2c20d68>] __wait_for_common+0x153/0x190
>  [<ffffffffa2c20d68>] ? __wait_for_common+0x153/0x190
>  [<ffffffffa26890e5>] ? wake_up_state+0x12/0x12
>  [<ffffffffa2c20e0e>] wait_for_completion_timeout+0x1d/0x1f
>  [<ffffffffa2a07c70>] usb_start_wait_urb+0x93/0xf1
>  [<ffffffffa2a07daf>] usb_control_msg+0xe1/0x11d
>  [<ffffffffc02cd254>] snd_usb_ctl_msg+0x9c/0xf1 [snd_usb_audio]
>  [<ffffffffc02ce191>] snd_usb_mixer_set_ctl_value+0x124/0xab1 [snd_usb_audio]
>  [<ffffffffc02ce230>] snd_usb_mixer_set_ctl_value+0x1c3/0xab1 [snd_usb_audio]
>  [<ffffffffc02ce58e>] snd_usb_mixer_set_ctl_value+0x521/0xab1 [snd_usb_audio]
>  [<ffffffffc02cee88>] snd_usb_mixer_add_control+0x36a/0x1264 [snd_usb_audio]
>  [<ffffffffc02cf323>] snd_usb_mixer_add_control+0x805/0x1264 [snd_usb_audio]
>  [<ffffffffa2a06e11>] ? usb_free_urb+0x1a/0x1c
>  [<ffffffffc02cfcf7>] snd_usb_mixer_add_control+0x11d9/0x1264 [snd_usb_audio]
>  [<ffffffffc02d000f>] snd_usb_create_mixer+0xbc/0x286 [snd_usb_audio]
>  [<ffffffffc02cac18>] 0xffffffffc02cac17
>  [<ffffffffa2a0aaf1>] usb_probe_interface+0x17c/0x21c
>  [<ffffffffa29a65bc>] driver_probe_device+0xae/0x1fa
>  [<ffffffffa29a6767>] __device_attach_driver+0x5f/0x66
>  [<ffffffffa29a6708>] ? driver_probe_device+0x1fa/0x1fa
>  [<ffffffffa29a4a60>] bus_for_each_drv+0x87/0xaa
>  [<ffffffffa29a688a>] __device_attach+0x9d/0x101
>  [<ffffffffa29a6913>] device_initial_probe+0x13/0x15
>  [<ffffffffa29a5ae6>] bus_probe_device+0x33/0x96
>  [<ffffffffa29a3d19>] device_add+0x328/0x547
>  [<ffffffffa2a09355>] usb_set_configuration+0x624/0x674
>  [<ffffffffa2a11949>] generic_probe+0x45/0x77
>  [<ffffffffa2a0a962>] usb_probe_device+0x2d/0x40
>  [<ffffffffa29a65bc>] driver_probe_device+0xae/0x1fa
>  [<ffffffffa29a6767>] __device_attach_driver+0x5f/0x66
>  [<ffffffffa29a6708>] ? driver_probe_device+0x1fa/0x1fa
>  [<ffffffffa29a4a60>] bus_for_each_drv+0x87/0xaa
>  [<ffffffffa29a688a>] __device_attach+0x9d/0x101
>  [<ffffffffa29a6913>] device_initial_probe+0x13/0x15
>  [<ffffffffa29a5ae6>] bus_probe_device+0x33/0x96
>  [<ffffffffa29a3d19>] device_add+0x328/0x547
>  [<ffffffffa29030bc>] ? add_device_randomness+0x111/0x130
>  [<ffffffffa2a00967>] usb_new_device+0x2a2/0x3c0
>  [<ffffffffa2a02ddc>] hub_thread+0xa3d/0xeed
>  [<ffffffffa2c2010d>] ? __schedule+0x41e/0x5ac
>  [<ffffffffa26957ce>] ? finish_wait+0x62/0x62
>  [<ffffffffa2a0239f>] ? usb_reset_device+0x16a/0x16a
>  [<ffffffffa267b255>] kthread+0x108/0x110
>  [<ffffffffa267b14d>] ? __kthread_parkme+0x67/0x67
>  [<ffffffffa2c23b2c>] ret_from_fork+0x7c/0xb0
>  [<ffffffffa267b14d>] ? __kthread_parkme+0x67/0x67
> 
> Signed-off-by: Stephen Barber <smbarber@...omium.org>

Applied, thanks.


Takashi

Powered by blists - more mailing lists

Powered by Openwall GNU/*/Linux Powered by OpenVZ