On Wed, 2013-08-28 at 10:12 +0800, Gu Zheng wrote:

On 08/28/2013 05:38 AM, Toshi Kani wrote:

 :

quoted

quoted

What about changing device_hotplug_lock and acpi_scan_lock to rwsem? like the
attached one(With a preliminary test, it also can make the splat go away).:)

I am curious how msleep(10) & restart_syscall() work in the change
below.  Doesn't the msleep() make s_active held longer time, which can
lead the thread holding device_hotplug_lock to wait it for deletion?

Yes, but it can avoid busy waiting. 

I know, but it's kinda unfortunate to sleep with s_active held in this
situation.  But I am fine with the 5ms Rafael used in this latest
patchset since this is a rare case anyway.

quoted

Also, does restart_syscall() release s_active and reopen this file
again?

Sure, it just set a TIF_SIGPENDING flag and return an -ERESTARTNOINTR error, s_active/file
will be released/closed in the failed path. And when do_signal() catches the -ERESTARTNOINTR,
it will change the regs to restart the syscall.

I see.  This is a clever functionality. 

Thanks,
-Toshi