netdev - RE: Intel 82574L hang when sending short ethernet packets at 100BaseT

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]

Date:	Wed, 29 Aug 2012 20:29:26 +0000
From:	"Dave, Tushar N" <tushar.n.dave@...el.com>
To:	Kelvie Wong <kelvie@...e.org>
CC:	"netdev@...r.kernel.org" <netdev@...r.kernel.org>,
	"e1000-devel@...ts.sourceforge.net" 
	<e1000-devel@...ts.sourceforge.net>,
	Kelvie Wong <kwong@...ldtech.com>
Subject: RE: Intel 82574L hang when sending short ethernet packets at
 100BaseT

>-----Original Message-----
>From: kelvie@...il.com [mailto:kelvie@...il.com] On Behalf Of Kelvie Wong
>Sent: Wednesday, August 29, 2012 1:17 PM
>To: Dave, Tushar N
>Cc: netdev@...r.kernel.org; e1000-devel@...ts.sourceforge.net; Kelvie Wong
>Subject: Re: Intel 82574L hang when sending short ethernet packets at
>100BaseT
>
>On Wed, Aug 29, 2012 at 1:13 PM, Dave, Tushar N <tushar.n.dave@...el.com>
>wrote:
>>>
>>>Anyway, here is the reproduction tool:
>>
>> I see, "Adding enough zero pad bytes to bring frame size up to 17 bytes
>causes the reset to no longer occur" - Is this true?
>>
>> -Tushar
>
>Yes, that is correct. We have been doing that ourselves as a workaround.

I'm already aware of this issue on PCI/PCI-X parts and e1000 SF driver does have this WA implemented. The patch for upstream e1000 driver is submitted too which will be accepted soon.
I will check with team and see if we need similar WA for PCIe part(s)!

-Tushar