netdev - Re: [PATCH] usbnet: dereference after null check in usbnet_start_xmit() and __usbnet_read

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

Message-ID: <87vbcacta6.fsf@nemi.mork.no>
Date:	Thu, 20 Aug 2015 09:29:05 +0200
From:	Bjørn Mork <bjorn@...k.no>
To:	Vivek Kumar Bhagat <vivek.bhagat@...sung.com>
Cc:	"netdev\@vger.kernel.org" <netdev@...r.kernel.org>,
	Nitin Jhanwar <nitin.j@...sung.com>,
	HEMANSHU SRIVASTAVA <hemanshu.s@...sung.com>
Subject: Re: [PATCH] usbnet: dereference after null check in usbnet_start_xmit() and __usbnet_read_cmd()

Vivek Kumar Bhagat <vivek.bhagat@...sung.com> writes:

> Dear Bjorn,
>
>>>This is wrong.  There are usbnet minidrivers depending on info->tx_fixup
>>> being called with a NULL skb.
> Also, if dev_hard_start_xmit() ensures that skb can not be NULL in usbnet_start_xmit()
> then we should remove below check.
>     if (skb)      <--- This check is confusing which says skb can be NULL.
>                 skb_tx_timestamp(skb); 

No, that test is there because of the ugly hack in cdc_ncm.  It doesn't
go through dev_hard_start_xmit(), but calls usbnet_start_xmit() directly
with a NULL skb as a signal to itself.  Yes, I told you it was ugly ;)

I do agree that it would be nice to make this go away.  But until that
happens usbnet_start_xmit() has to deal with NULL skbs, forwarding them
to the tx_fixup hook.

Bjørn
--
To unsubscribe from this list: send the line "unsubscribe netdev" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html