From: Amir Vadai <amirv@mellanox.com>
To: Eric Dumazet <eric.dumazet@gmail.com>
Cc: David Laight <David.Laight@ACULAB.COM>,
"David S. Miller" <davem@davemloft.net>,
"netdev@vger.kernel.org" <netdev@vger.kernel.org>,
Yevgeny Petrilin <yevgenyp@mellanox.com>,
Or Gerlitz <ogerlitz@mellanox.com>,
Eugenia Emantayev <eugenia@mellanox.com>
Subject: Re: net/mlx4_en: Pad ethernet packets smaller than 17 bytes
Date: Thu, 27 Feb 2014 17:46:37 +0200 [thread overview]
Message-ID: <20140227154635.GA24226@mtl-eit-vdi-22.mtl.labs.mlnx> (raw)
In-Reply-To: <1393512255.26794.20.camel@edumazet-glaptop2.roam.corp.google.com>
On 27/02/14 06:44 -0800, Eric Dumazet wrote:
> On Thu, 2014-02-27 at 16:02 +0200, Amir Vadai wrote:
> >
> > Yes, skb will not be classified as inline-able if nr_frags != 1.
> > See is_inline()
Maybe I wasn't accurate enough. inline-able skb must have:
1. skb->len < inline_thold
2. nr_frags <= 1
Default value for inline_thold is 104 and it is a module parameter
that can be changed.
>
> How often this inline stuff ever happens ?
>
> It looks like the length of the frame needs to be shorter than 44 bytes,
> right ?
It should be shorter than 104 - inline_thold default (and max) value is 104.
I don't understand why you say 44.
>
> Minimal length for TCP is 54 bytes.
>
> I wonder why you tried to inline stuff that is probably never
> seen in real workload.
>
If you mean why did I try to send packets shorter than 17 using inline
- the answer is that it is easy to do it, so why not to?
Amir
next prev parent reply other threads:[~2014-02-27 15:46 UTC|newest]
Thread overview: 17+ messages / expand[flat|nested] mbox.gz Atom feed top
2014-02-27 12:26 [PATCH net-next 0/9] net/mlx4: Mellanox driver update 27-02-2014 Amir Vadai
2014-02-27 12:26 ` [PATCH net-next 1/9] net/mlx4_en: Fix UP limit in ieee_ets->prio_tc Amir Vadai
2014-02-27 12:26 ` [PATCH net-next 2/9] net/mlx4_en: Verify mlx4_en module parameters Amir Vadai
2014-02-27 12:27 ` [PATCH net-next 3/9] net/mlx4_en: Pad ethernet packets smaller than 17 bytes Amir Vadai
2014-02-27 13:08 ` David Laight
2014-02-27 14:02 ` Amir Vadai
2014-02-27 14:44 ` Eric Dumazet
2014-02-27 15:46 ` Amir Vadai [this message]
2014-02-27 20:36 ` David Miller
2014-02-27 21:47 ` Amir Vadai
2014-02-27 21:58 ` David Miller
2014-02-27 12:27 ` [PATCH net-next 4/9] net/mlx4_en: Move queue stopped/waked counters to be per ring Amir Vadai
2014-02-27 12:27 ` [PATCH net-next 5/9] net/mlx4: Replace mlx4_en_mac_to_u64() with mlx4_mac_to_u64() Amir Vadai
2014-02-27 12:27 ` [PATCH net-next 6/9] net/mlx4_en: Fix selftest failing on non 10G link speed Amir Vadai
2014-02-27 12:27 ` [PATCH net-next 7/9] net/mlx4_core: Fix sparse warning Amir Vadai
2014-02-27 12:27 ` [PATCH net-next 8/9] net/mlx4_en: Use union for BlueFlame WQE Amir Vadai
2014-02-27 12:27 ` [PATCH net-next 9/9] net/mlx4_en: Change Connect-X description in kconfig Amir Vadai
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20140227154635.GA24226@mtl-eit-vdi-22.mtl.labs.mlnx \
--to=amirv@mellanox.com \
--cc=David.Laight@ACULAB.COM \
--cc=davem@davemloft.net \
--cc=eric.dumazet@gmail.com \
--cc=eugenia@mellanox.com \
--cc=netdev@vger.kernel.org \
--cc=ogerlitz@mellanox.com \
--cc=yevgenyp@mellanox.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.