linux-kernel - A potential data race in drivers/scsi/mvumi.ko

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: <cb9e90a8-2f68-dba2-9126-a0c90973a65d@ispras.ru>
Date:	Fri, 12 Aug 2016 15:39:32 +0300
From:	Pavel Andrianov <andrianov@...ras.ru>
To:	"James E.J. Bottomley" <jejb@...ux.vnet.ibm.com>
Cc:	"Martin K. Petersen" <martin.petersen@...cle.com>,
	linux-scsi@...r.kernel.org, linux-kernel@...r.kernel.org,
	Vaishali Thakkar <vaishali.thakkar@...cle.com>,
	ldv-project@...uxtesting.org
Subject: A potential data race in drivers/scsi/mvumi.ko

Hi!

There is a potential data race in drivers/scsi/mvumi.ko.
Regard such case:

Thread 1					Thread 2
...
-> mvumi_reset_host_9500 -
    is called without any locking
   -> mvumi_wait_for_outstanding
     ->mvumi_start
       ->mvumi_check_handshake
         -> mvumi_handshake_event
           -> mvumi_handshake
             ->mvumi_hs_build_page

			   interrupt comes
					      -> mvumi_isr_handler
						-> mvumi_handshake
						  -> mvumi_hs_build_page

In this case the same data mhba->handshake_page is modified from two 
threads. Likely the first thread should acquire the same spinlock 
mhba->shost->host_lock as the second thread.

-- 
Pavel Andrianov
Linux Verification Center, ISPRAS
web: http://linuxtesting.org
e-mail: andrianov@...ras.ru