Skip site navigation (1)Skip section navigation (2)
Date:      Fri, 26 Oct 2012 10:28:10 -0600
From:      Sebastian Kuzminsky <seb@lineratesystems.com>
To:        Gleb Smirnoff <glebius@freebsd.org>
Cc:        freebsd-net@freebsd.org
Subject:   Re: fragmentation problem in FreeBSD 7
Message-ID:  <CAN=597T93_=y2yYQ86Br876gOLh=S780Q6dhEP08AYv4-LSKHw@mail.gmail.com>
In-Reply-To: <20121026135354.GD70741@FreeBSD.org>
References:  <CAN=597Rb-ToBQuJ%2BYet9e25Hbt-QmLJPKUXGf1pFEbVsRvFONg@mail.gmail.com> <20121026135354.GD70741@FreeBSD.org>

next in thread | previous in thread | raw e-mail | index | archive | help
On Fri, Oct 26, 2012 at 7:53 AM, Gleb Smirnoff <glebius@freebsd.org> wrote:

> Thanks for submission!
>
> I'm about to commit the attached patch to head. Can you please review it?
> Haven't I missed anything important?
>

Looks good to me.  Thanks for also including the sw_csum cleanup, that code
had me scratching my head for a whole morning.  :-)


I have also moved from CSUM_DELAY_IP to CSUM_IP. AFAIU, the alias
> CSUM_DELAY_IP
> was made to match CSUM_DELAY_DATA. But to my point of view it makes it more
> difficult to understand code, because a person reading code sees different
> constants in the stack and in drivers. Since your change touches every line
> in the stack, that utilizes CSUM_DELAY_IP, I decided to consistently use
> CSUM_IP constant.
>

I agree with this.  I did not understand why the original code use
CSUM_DELAY_IP instead of the more obvious CSUM_IP, but i felt too timid to
change it ;-)



> Totus tuus, Glebius.
>

Et tuus, thanks for taking the time to review and clean up my patch!

-- 
Sebastian Kuzminsky
Linerate Systems



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?CAN=597T93_=y2yYQ86Br876gOLh=S780Q6dhEP08AYv4-LSKHw>