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:	Tue, 10 Aug 2010 10:55:02 -0700
From:	Philip Rakity <prakity@...vell.com>
To:	Lennert Buytenhek <buytenh@...tstofly.org>
CC:	Sachin Sanap <ssanap@...vell.com>,
	"netdev@...r.kernel.org" <netdev@...r.kernel.org>,
	Ashish Karkare <akarkare@...vell.com>,
	Prabhanjan Sarnaik <sarnaik@...vell.com>,
	"eric.y.miao@...il.com" <eric.y.miao@...il.com>,
	Mark Brown <markb@...vell.com>
Subject: Re: [PATCH v2]  net: add Fast Ethernet driver for PXA168.


We agree that extra header prepend space is needed.  The solution of globally defining SKB_NET_PAD to increase this area for all drivers is where I have some concern.  The solution certainly works but at the cost of extra space for drivers that do not need to do this.   The solution is fine by me and maybe the best answer is to increase the value in the standard linux implementation from 32 to 48 or 64 bytes. 


On Aug 10, 2010, at 10:30 AM, Lennert Buytenhek wrote:

> On Tue, Aug 10, 2010 at 09:24:23AM -0700, Philip Rakity wrote:
> 
>>>> * Headroom in SKB for 802.11 not included in the patch since that
>>>> varies based on 802.11 a/b/g/n.
>>> 
>>> I don't think this is true?
>>> 
>>> (The 11a/b/n on-the-air preambles are of different lengths (and are
>>> sent at different rates), but that isn't visible to software.)
>> 
>> 802.11b/g headers are larger and if not handled by firmware in an
>> offload engine are programmed by the driver.  There needs to be room
>> for 
>> 
>> a) Frame Control (2 bytes)
>> b) AID (2 bytes)
>> c) address-1 (6 bytes)
>> d) address-2 (6 bytes)
>> e) address-3 (6 bytes)
>> f) seqfrag  (2 bytes)
>> g) address-4 (6 bytes) 
>> 
>> or 30 bytes.
> 
> Depends on the driver and/or whether you are doing WDS.
> 
> mwl8k always needs 30 bytes due to hardware pecularities (while the
> 4-address-with-QoS header is 32 bytes, the QoS field is always passed
> in the TX descriptor), but other drivers might need 32 bytes (and not
> 30 bytes) for 4-address-with-QoS headers, or might not need 30 or 32
> bytes for transmitting 3-address or 2-address frames.
> 
> 
>> 11n needs more.
> 
> Really?  Pointer?
> 
> 
>> Some of these fields ARE visible to s/w.
> 
> Where did I say that they are not?
> 
> What I said was that the _preambles_ are not visible to software.

--
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

Powered by blists - more mailing lists