[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <1401464457.3645.109.camel@edumazet-glaptop2.roam.corp.google.com>
Date: Fri, 30 May 2014 08:40:57 -0700
From: Eric Dumazet <eric.dumazet@...il.com>
To: "fugang.duan@...escale.com" <fugang.duan@...escale.com>
Cc: "ezequiel.garcia@...e-electrons.com"
<ezequiel.garcia@...e-electrons.com>,
David Laight <David.Laight@...LAB.COM>,
"Frank.Li@...escale.com" <Frank.Li@...escale.com>,
"davem@...emloft.net" <davem@...emloft.net>,
"netdev@...r.kernel.org" <netdev@...r.kernel.org>,
"shawn.guo@...aro.org" <shawn.guo@...aro.org>,
"bhutchings@...arflare.com" <bhutchings@...arflare.com>,
"stephen@...workplumber.org" <stephen@...workplumber.org>
Subject: RE: [PATCH v1 4/6] net: fec: Increase buffer descriptor entry number
On Fri, 2014-05-30 at 15:08 +0000, fugang.duan@...escale.com wrote:
> If frag page data is not match the alignment for ethernet DMA controller, there need three descriptor for one MSS:
> One descriptor for headers, one for the first non-align bytes copied from frag page, one for the rest of frag page.
>
You could avoid the 2nd descriptor, by copying the unaligned part of the
payload into first descriptor (containing headers : about 66 bytes or so
you have room, if not, increase the 128 bytes room to 192 bytes)
> So one frame may cost descriptor number is: 3 x 45
May... but in general it would be closer of 2 * 45
>
> So the descriptors slots set to 512 is not big, just is reasonable. Do you think ?
I believe it is fine : that is about 5 64KB TSO packets.
--
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