Revert "kobject: don't block for each kobject_uevent".
authorHugh Dickins <hugh@veritas.com>
Thu, 16 Apr 2009 20:55:29 +0000 (21:55 +0100)
committerLinus Torvalds <torvalds@linux-foundation.org>
Thu, 16 Apr 2009 21:41:24 +0000 (14:41 -0700)
commit05f54c13cd0c33694eec39a265475c5d6cf223cf
treef3bdff25fd2e7f405ec159bdabd9ef1166d1f22d
parent35c80d5f400f68f2eccf3069d1c068e154bde9c9
Revert "kobject: don't block for each kobject_uevent".

This reverts commit f520360d93cdc37de5d972dac4bf3bdef6a7f6a7.

Tetsuo Handa, running a kernel with CONFIG_DEBUG_PAGEALLOC=y and
CONFIG_UEVENT_HELPER_PATH=/sbin/hotplug, has been hitting RCU detected
CPU stalls: it's been spinning in the loop where do_execve() counts up
the args (but why wasn't fixup_exception working? dunno).

The recent change, switching kobject_uevent_env() from UMH_WAIT_EXEC
to UMH_NO_WAIT, is broken: the exec uses args on the local stack here,
and an env which is kfreed as soon as call_usermodehelper() returns.
It very much needs to wait for the exec to be done.

An alternative would be to keep the UMH_NO_WAIT, and complicate the code
to allocate and free these resources correctly? but no, as GregKH
pointed out when making the commit, CONFIG_UEVENT_HELPER_PATH="" is a
much better optimization - though some distros are still saying
/sbin/hotplug in their .config, yet with no such binary in their initrd
or their root.

Reported-by: Tetsuo Handa <penguin-kernel@i-love.sakura.ne.jp>
Signed-off-by: Hugh Dickins <hugh@veritas.com>
Acked-by: Greg Kroah-Hartman <gregkh@suse.de>
Acked-by: Arjan van de Ven <arjan@linux.intel.com>
Acked-by: Will Newton <will.newton@gmail.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
lib/kobject_uevent.c