netdev - Re: [PATCH] bnx2x: fix occasional statistics off-by-4GB error

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: <1363591634.4752.10.camel@lb-tlvb-eilong.il.broadcom.com>
Date:	Mon, 18 Mar 2013 09:27:14 +0200
From:	"Eilon Greenstein" <eilong@...adcom.com>
To:	"Eric Dumazet" <eric.dumazet@...il.com>
cc:	"David Miller" <davem@...emloft.net>, dmitry@...adcom.com,
	zenczykowski@...il.com, maze@...gle.com, netdev@...r.kernel.org,
	yuvalmin@...adcom.com
Subject: Re: [PATCH] bnx2x: fix occasional statistics off-by-4GB error

On Sun, 2013-03-17 at 16:17 -0700, Eric Dumazet wrote:
> On Sun, 2013-03-17 at 22:53 +0200, Eilon Greenstein wrote:
> 
> > This is not such a trivial issue - the HW/FW is guaranteeing the atomic
> > read and this is why we can always use 32b variables.
> > 
> 
> I am glad you qualify the following code as 'trivial'
> 

Eric - I did not mean that the solution is trivial, I meant that the
problem is trivial. Like the trivial statement that the code should not
contain any bugs :)


> All you really needed is to have normal host structures containing u64
> fields, and you only needed one single helper to convert from hw u32:u32
> to host u64.
> 
> Because gcc handles u64 operations very well, even on 32bit hosts, and
> it knows about the carry flag.

I was not clear when I said size-extension: most of the HW counters are
46bits and we are trying to keep 64bits counters in the SW, so simply
copying the data from the HW is not enough.



--
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