Re: Warnings from PM QoS plist usage

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



On Wed, Jul 14, 2010 at 10:35:10AM +0100, Mark Brown wrote:

CCing in the mailing list like I meant to on the original post, sorry:

> When running with 'pm_qos: Reimplement using plists' and
> CONFIG_DEBUG_PI_LIST enabled ALSA (and presumably other PM QoS users)
> generate tracebacks like this whenever they use PM QoS:
> 
> [  118.250000] ------------[ cut here ]------------
> [  118.250000] WARNING: at lib/plist.c:57 plist_check_head+0x2c/0x44()
> [  118.250000] Modules linked in: snd_soc_fnord snd_soc_s3c64xx_i2s snd_soc_s3c_6
> [  118.250000] [<c0034aac>] (unwind_backtrace+0x0/0xec) from [<c004a8f0>] (warn_)
> [  118.250000] [<c004a8f0>] (warn_slowpath_common+0x4c/0x7c) from [<c004a93c>] ()
> [  118.250000] [<c004a93c>] (warn_slowpath_null+0x1c/0x24) from [<c0187c2c>] (pl)
> [  118.250000] [<c0187c2c>] (plist_check_head+0x2c/0x44) from [<c0187cf0>] (plis)
> [  118.250000] [<c0187cf0>] (plist_add+0x18/0xbc) from [<c006755c>] (update_targ)
> [  118.250000] [<c006755c>] (update_target+0xb8/0x124) from [<c025336c>] (snd_pc)
> [  118.250000] [<c025336c>] (snd_pcm_hw_params+0x2cc/0x310) from [<c0253780>] (s)
> [  118.250000] [<c0253780>] (snd_pcm_common_ioctl1+0x1f8/0x10ac) from [<c0254e20)
> [  118.250000] [<c0254e20>] (snd_pcm_playback_ioctl1+0x3d8/0x3fc) from [<c00c7e5)
> [  118.250000] [<c00c7e50>] (vfs_ioctl+0x2c/0x70) from [<c00c8534>] (do_vfs_ioct)
> [  118.250000] [<c00c8534>] (do_vfs_ioctl+0x4d0/0x524) from [<c00c85c0>] (sys_io)
> [  118.250000] [<c00c85c0>] (sys_ioctl+0x38/0x5c) from [<c002eec0>] (ret_fast_sy)
> [  118.250000] ---[ end trace 810ec758dcd5bc5a ]---                              
> This is because the plists introduced by the above commit are being
> initialised with the _PLIST_HEAD_INIT() rather than PLIST_HEAD_INIT()
> and don't have the associated locks which the debug code relies on.
> 
> I've not looked at the PM QoS code at all so I don't know what the
> appropriate place to fix this is - do we perhaps want to fix the debug
> code to cope with no locks?  Looking at the plist.h code it seems fairly
> clear that there's currently no expectation that users should use
> _PLIST_INIT_HEAD() directly.

-- 
"You grabbed my hand and we fell into it, like a daydream - or a fever."
_______________________________________________
linux-pm mailing list
linux-pm@xxxxxxxxxxxxxxxxxxxxxxxxxx
https://lists.linux-foundation.org/mailman/listinfo/linux-pm


[Index of Archives]     [Linux ACPI]     [Netdev]     [Ethernet Bridging]     [Linux Wireless]     [CPU Freq]     [Kernel Newbies]     [Fedora Kernel]     [Security]     [Linux for Hams]     [Netfilter]     [Bugtraq]     [Yosemite News]     [MIPS Linux]     [ARM Linux]     [Linux RAID]     [Linux Admin]     [Samba]

  Powered by Linux