netdev - Re: [RFC PATCH] bnx2x: fix tx queue locking and memory barriers

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: <20100225164015.3991ab5e@dhcp-lab-109.englab.brq.redhat.com>
Date:	Thu, 25 Feb 2010 16:40:15 +0100
From:	Stanislaw Gruszka <sgruszka@...hat.com>
To:	David Miller <davem@...emloft.net>
Cc:	netdev@...r.kernel.org, vladz@...adcom.com, eilong@...adcom.com,
	dhowells@...hat.com
Subject: Re: [RFC PATCH] bnx2x: fix tx queue locking and memory barriers

On Thu, 25 Feb 2010 02:18:22 -0800 (PST)
David Miller <davem@...emloft.net> wrote:

> > Memory barriers here IMHO, prevent to make queue permanently stopped
> > when on one cpu bnx2x_tx_int() make queue empty, whereas on other
> > cpu bnx2x_start_xmit() see it full and make stop it, such cause
> > queue will be stopped forever.
> 
> Instead of having an opinion, please show the exact sequence
> of events that can lead to this situation.  With such facts
> inhand, you will have no need for an opinion :-)

Ok, here is the story:

Queue (4000 elements) is almost full: fp->tx_bd_prod = 3980, fp->tx_bd_cons = 0,
Packets ware already transmitted by device, but that was not reported to
the driver because interrupts where disabled. We are transmitting new skb
with 20 fragments.

cpu0: (in bnx2x_poll)                   cpu1: (transferring data)

bnx2x_tx_int():                         bnx2x_start_xmit():
local fp->tx_bd_cons = 3980;            send more data to device
return

bnx2x_tx_int():                         local fp->tx_bd_prod = 4000
local fp->tx_bd_cons = 4000;            local fp->tx_bd_cons still 0
queue not stopped                       no avail space in queue
return;                                 stop queue
                                        smp_mb() - not paired, cpu1 does not "see" cpu0 caches changes
                                        local fp->tx_bd_cons still 0
                                        no wake
					
Finally queue is sopped and device will not generate interrupt nor
call bnx2x_tx_int() from bnx2x_poll() since bnx2x_has_tx_work() will
return false.
--
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