linux-kernel - Re: INFO: task hung in request_key

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

Message-ID: <e34e7b99-221f-661b-a299-6cdc35760545@i-love.sakura.ne.jp>
Date:   Wed, 1 Jul 2020 21:49:29 +0900
From:   Tetsuo Handa <penguin-kernel@...ove.sakura.ne.jp>
To:     Luis Chamberlain <mcgrof@...nel.org>,
        Andrew Morton <akpm@...ux-foundation.org>,
        "Eric W. Biederman" <ebiederm@...ssion.com>,
        Naresh Kamboju <naresh.kamboju@...aro.org>,
        Herbert Xu <herbert@...dor.apana.org.au>,
        Christoph Hellwig <hch@...radead.org>
Cc:     syzbot <syzbot+46c77dc7e98c732de754@...kaller.appspotmail.com>,
        dhowells@...hat.com, jarkko.sakkinen@...ux.intel.com,
        jmorris@...ei.org, keyrings@...r.kernel.org,
        linux-kernel@...r.kernel.org,
        linux-security-module@...r.kernel.org, serge@...lyn.com,
        syzkaller-bugs@...glegroups.com
Subject: Re: INFO: task hung in request_key_tag

On 2020/07/01 21:20, Luis Chamberlain wrote:
> On Wed, Jul 01, 2020 at 07:04:15PM +0900, Tetsuo Handa wrote:
>> I suspect commit 9e9b47d6bbe9df65 ("umh: fix processed error when UMH_WAIT_PROC is used").
>> Maybe the change in kernel/umh.c and/or security/keys/request_key.c made by that commit is
>> affecting call_usermodehelper_keys() == 0 case when complete_request_key() is called.
> 
> That patch has been dropped for now due to another reported issue
> bisected to it and even though we have not root caused that issue [0].
> 
> It would be good to have a reproducer for this reported issue as well,

Reproducer is not available yet.
But at least this patch is changing the behavior of

  call_usermodehelper_keys() == 0 && (test_bit(KEY_FLAG_USER_CONSTRUCT, &key->flags) || key_validate(key) < 0)

case from key_negate_and_link() to key_revoke() in complete_request_key().
Since test_and_clear_bit(KEY_FLAG_USER_CONSTRUCT, &key->flags) might be called from
key_reject_and_link() from key_negate_and_link(), this change might be the cause of
this hung task report.