netdev - Re: [PATCH net RFC 2/2] ixgbe: fix sleep bug caused by napi_disable inside local_bh

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: <1380657915.22910.7.camel@jekeller-desk1.amr.corp.intel.com>
Date:	Tue, 1 Oct 2013 20:05:15 +0000
From:	"Keller, Jacob E" <jacob.e.keller@...el.com>
To:	Yuval Mintz <yuvalmin@...adcom.com>
CC:	Francois Romieu <romieu@...zoreil.com>,
	"netdev@...r.kernel.org" <netdev@...r.kernel.org>,
	"Duyck, Alexander H" <alexander.h.duyck@...el.com>,
	Hyong-Youb Kim <hykim@...i.com>,
	Dmitry Kravkov <dmitry@...adcom.com>,
	"Amir Vadai" <amirv@...lanox.com>,
	Eliezer Tamir <eliezer.tamir@...ux.intel.com>
Subject: Re: [PATCH net RFC 2/2] ixgbe: fix sleep bug caused by napi_disable
 inside local_bh_disable()d context

On Tue, 2013-10-01 at 12:11 +0000, Yuval Mintz wrote:
> > > > I have to move the local_bh_disable in order to put napi_disable
> > > outside
> > > > of the call since napi_disable could sleep, causing a scheduling while
> > > > atomic BUG.
> > >
> > > I am in violent agreement with this part.
> > > --
> > > Ueimor
> > 
> > Regards,
> > Jake
> > --
> 
> It seem like we've hit the same issue with the bnx2x driver.
> Is there anything new about the RFC?
> 
> Thanks,
> Yuval
> 
> 
> 
> 
> 

The napi_disable call for might_sleep() is the same. The solution in the
ixgbe driver is different. I completely re-wrote the segment about how
to disable the q_vector by adding a new state, rather than abusing the
QV_LOCKED_NAPI state. In addition I refactored it so that the
qv_lock_napi used spin_lock_bh() instead of plain spin_lock(), and
changed it so that we didn't need the local_bh_disable() call in
ixgbe_napi_disable_all.

This is a much cleaner solution than what I originally proposed.

Regards,
Jake