linux-kernel - Re: [REPORT] Request for reviewing crypto code wrt wait_for

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: <20210806114058.GA13896@gondor.apana.org.au>
Date:   Fri, 6 Aug 2021 19:40:58 +0800
From:   Herbert Xu <herbert@...dor.apana.org.au>
To:     Byungchul Park <byungchul.park@....com>
Cc:     torvalds@...ux-foundation.org, peterz@...radead.org,
        mingo@...hat.com, will@...nel.org, davem@...emloft.net,
        linux-crypto@...r.kernel.org, linux-kernel@...r.kernel.org,
        tglx@...utronix.de, rostedt@...dmis.org, joel@...lfernandes.org,
        alexander.levin@...rosoft.com, daniel.vetter@...ll.ch,
        chris@...is-wilson.co.uk, duyuyang@...il.com,
        johannes.berg@...el.com, tj@...nel.org, tytso@....edu,
        willy@...radead.org, david@...morbit.com, amir73il@...il.com,
        bfields@...ldses.org, gregkh@...uxfoundation.org,
        kernel-team@....com
Subject: Re: [REPORT] Request for reviewing crypto code wrt
 wait_for_completion()

On Fri, Aug 06, 2021 at 05:03:44PM +0900, Byungchul Park wrote:
> Hello crypto folks,
> 
> I developed a tool for tracking waiters and reporting if any of the
> events that the waiters are waiting for would never happen, say, a
> deadlock. Yes, it would look like Lockdep but more inclusive.
> 
> While I ran the tool(Dept: Dependency Tracker) on v5.4.96, I got some
> reports from the tool. One of them is related to crypto subsystem.
> Because I'm not that familiar with the code, I'd like to ask you guys to
> review the related code.
> 
> If I understand correctly, it doesn't actually cause deadlock but looks
> like a problematic code. I know you are not used to the format of the
> report from Dept so.. let me summerize the result.
> 
> The simplified call trace looks like when the problem araised :
> 
> THREAD A
> --------
> A1 crypto_alg_mod_lookup()
> A2    crypto_probing_notify(CRYPTO_MSG_ALG_REQUEST)
> A3       cryptomgr_schedule_probe()
> A4          kthread_run(cyptomgr_probe) ---> Start THREAD B
> 
> A5    crypto_larval_wait()
> A6       wait_for_completion_killable_timeout(c) /* waiting for B10 */

This larval would be an instantiation larval, and it can only be
woken up by thread B, not C.

> THREAD B
> --------
> B1 cryptomgr_probe()
> B2    pkcslpad_create()
> B3       crypto_wait_for_test()
> B4          crypto_probing_notify(CRYPTO_MSG_ALG_REGISTER)
> B5             cryptomgr_schedule_test()
> B6                kthread_run(cyptomgr_test) ---> Start THREAD C
> 
> B7    tmpl->alloc()
> B8    crupto_register_instance()
> B9          wait_for_completion_killable(c) /* waiting for C3 */
> B10   complete_all(c)

I presume you're talking about about the wait_for_completion from
crypto_wait_for_test, in which case it can only be woken by thread
C.  After which thread B will return to cryptomgr_probe and wake up
thread A.

> THREAD C
> --------
> C1 cryptomgr_test()
> C2    crypto_alg_tested()
> C3       complete_all(c)
> 
> ---
> 
> For example, in this situation, I think C3 could wake up both A6 and B9
> before THREAD B reaches B10 which is not desired by A6. Say, is it okay
> to wake up A6 with B7 ~ B9 having yet to complete?

AFAICS thread C only wakes up test larvals, not instantiation larvals.
Please let me know if you have any further issues.

Thanks,
-- 
Email: Herbert Xu <herbert@...dor.apana.org.au>
Home Page: http://gondor.apana.org.au/~herbert/
PGP Key: http://gondor.apana.org.au/~herbert/pubkey.txt