linux-kernel - stop_con_cpu_nowait() usage in set

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>] [day] [month] [year] [list]

Message-ID: <20110329130855.GA16492@htj.dyndns.org>
Date:	Tue, 29 Mar 2011 15:08:55 +0200
From:	Tejun Heo <tj@...nel.org>
To:	trenn@...ell.com, suresh.b.siddha@...el.com
Cc:	linux-kernel@...r.kernel.org
Subject: stop_con_cpu_nowait() usage in set_mtrr()

Hello, Suresh, Thomas.

This is about the openSUSE complete system freeze at start bug Thomas
just pinged about - https://bugzilla.novell.com/show_bug.cgi?id=672008

I looked through the set_mtrr() code and it's implementing its own
stop_machine() using stop_one_cpu_nowait().  This isn't safe.  There
is only one stop cpu execution context per cpu and the custom
implementation can race against the standard stop_machine() - it can
deadlock with each holding some of the execution contexts and neither
stepping away.

Suresh, is there any reason why set_mtrr() is using its own
synchronization instead of stop_machine()?

Thanks.

-- 
tejun
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/