linux-kernel - workqueue: Is flush_work() without INIT

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-next>] [day] [month] [year] [list]

Date:   Fri, 18 Jan 2019 20:58:49 +0900
From:   Tetsuo Handa <penguin-kernel@...ove.sakura.ne.jp>
To:     Tejun Heo <tj@...nel.org>
Cc:     LKML <linux-kernel@...r.kernel.org>
Subject: workqueue: Is flush_work() without INIT_WORK() OK?

Hello, Tejun.

syzbot is reporting "INFO: trying to register non-static key in __flush_work"
at https://syzkaller.appspot.com/bug?id=a5954455fcfa51c29ca2ab55b203076337e1c770 ,
and I think that the caller is passing a work_struct allocated by kzalloc() to
flush_work().

But I can't judge which problem (the caller side or the WQ side) without knowing
the behavior of

	struct work_struct work;
	memset(&work, 0, sizeof(work));
	flush_work(&work);

. Is this behavior defined as "safe and no-op"? If this should be "safe and no-op",
we need to check whether INIT_WORK() was called before starting lockdep checks
at __flush_work(). If this is not "safe and no-op", I want to emit a warning if
INIT_WORK() was not called.