All of lore.kernel.org
 help / color / mirror / Atom feed
From: Ben Greear <greearb@candelatech.com>
To: Neil Horman <nhorman@tuxdriver.com>
Cc: Eric Dumazet <eric.dumazet@gmail.com>,
	Jiri Pirko <jpirko@redhat.com>,
	netdev@vger.kernel.org, Alexey Dobriyan <adobriyan@gmail.com>,
	"David S. Miller" <davem@davemloft.net>,
	robert.olsson@its.uu.se
Subject: Re: [PATCH] pktgen: Clone skb to avoid corruption of skbs in ndo_start_xmit methods
Date: Wed, 20 Jul 2011 09:33:20 -0700	[thread overview]
Message-ID: <4E270350.7050904@candelatech.com> (raw)
In-Reply-To: <20110720151827.GD12349@hmsreliant.think-freely.org>

On 07/20/2011 08:18 AM, Neil Horman wrote:
> On Wed, Jul 20, 2011 at 06:24:15AM +0200, Eric Dumazet wrote:
>> Le mardi 19 juillet 2011 à 22:07 -0400, Neil Horman a écrit :
>>>>
>>> I think this is a good idea.  It lets pktgen dynamically make the clone/share
>>> decision dynamically and only impacts performance for those systems.
>>>
>>
>> Just let pktgen refuse to use clone_skb command for these devices.
>>
> copy that, This is by no means final, but what do you think of this?  If its
> agreeable to you, Ben, et al. I can add this to my local tree and start auditing
> all the drivers that may need to have the flag set.
>
> Regards
> Neil
>
>
> diff --git a/include/linux/if.h b/include/linux/if.h
> index 3bc63e6..ae904fe 100644
> --- a/include/linux/if.h
> +++ b/include/linux/if.h
> @@ -76,6 +76,7 @@
>   #define IFF_BRIDGE_PORT	0x4000		/* device used as bridge port */
>   #define IFF_OVS_DATAPATH	0x8000	/* device used as Open vSwitch
>   					 * datapath port */
> +#define IFF_CANT_SHARE_SKB	0x10000	/* Device can't share skbs in tx path */
>
>   #define IF_GET_IFACE	0x0001		/* for querying only */
>   #define IF_GET_PROTO	0x0002
> diff --git a/net/core/pktgen.c b/net/core/pktgen.c
> index f76079c..bf6d88d 100644
> --- a/net/core/pktgen.c
> +++ b/net/core/pktgen.c
> @@ -1071,6 +1071,9 @@ static ssize_t pktgen_if_write(struct file *file,
>   		if (len<  0)
>   			return len;
>
> +		if (pkt_dev->priv_flags&  IFF_CANT_SHARE_SKB)
> +			return -EOPNOTSUPP;
> +
>   		i += len;
>   		pkt_dev->clone_skb = value;
>

Please only return an error if value > 1.

Also, if there is any place where user can configure pkt_dev,
you would want code there to check for the CANT_SHARE_SKB flag
and force clone_skb to zero if user changes to a different
device that cannot share skbs.

Thanks,
Ben

-- 
Ben Greear <greearb@candelatech.com>
Candela Technologies Inc  http://www.candelatech.com


  parent reply	other threads:[~2011-07-20 16:33 UTC|newest]

Thread overview: 59+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2011-07-19 19:52 [PATCH] pktgen: Clone skb to avoid corruption of skbs in ndo_start_xmit methods Neil Horman
2011-07-19 20:02 ` Eric Dumazet
2011-07-19 20:17   ` Joe Perches
2011-07-19 20:29   ` Jiri Pirko
2011-07-19 20:41     ` Eric Dumazet
2011-07-19 21:01       ` Ben Greear
2011-07-20  0:19       ` Neil Horman
2011-07-20  0:25         ` Rick Jones
2011-07-20 15:23           ` Loke, Chetan
2011-07-20  0:43         ` Eric Dumazet
2011-07-20  0:52           ` Ben Greear
2011-07-20  2:07             ` Neil Horman
2011-07-20  4:19               ` Ben Greear
2011-07-20  4:24               ` Eric Dumazet
2011-07-20 15:17                 ` Loke, Chetan
2011-07-20 15:18                 ` Neil Horman
2011-07-20 15:30                   ` Eric Dumazet
2011-07-20 15:39                     ` Neil Horman
2011-07-20 15:44                       ` Eric Dumazet
2011-07-20 16:19                         ` Neil Horman
2011-07-20 15:35                   ` Michał Mirosław
2011-07-20 15:40                     ` Neil Horman
2011-07-20 16:08                       ` Michał Mirosław
2011-07-20 16:18                         ` Neil Horman
2011-07-20 16:37                     ` Ben Greear
2011-07-20 16:33                   ` Ben Greear [this message]
2011-07-20 16:36                     ` Neil Horman
2011-07-21 22:01                   ` David Miller
2011-07-21 22:14                     ` Ben Greear
2011-07-21 22:19                       ` David Miller
2011-07-21 22:26                         ` Ben Greear
2011-07-21 23:50                         ` Neil Horman
2011-07-22  0:08                           ` David Miller
2011-07-22  1:37                             ` Neil Horman
2011-07-20  1:59           ` Neil Horman
2011-07-25 19:45 ` [PATCH 0/2] pktgen: Clone skb to avoid corruption of skbs in ndo_start_xmit methods (v2) Neil Horman
2011-07-25 19:45   ` [PATCH 1/2] net: add IFF_SKB_TX_SHARED flag to priv_flags Neil Horman
2011-07-25 20:11     ` Eric Dumazet
2011-07-26 14:54       ` Neil Horman
2011-07-25 19:45   ` [PATCH 2/2] net: Audit drivers to identify those needing IFF_TX_SKB_SHARING cleared Neil Horman
2011-07-26 18:34     ` Krzysztof Halasa
2011-07-26 16:05 ` [PATCH 0/2] pktgen: Clone skb to avoid corruption of skbs in ndo_start_xmit methods (v3) Neil Horman
2011-07-26 16:05   ` [PATCH 1/2] net: add IFF_SKB_TX_SHARED flag to priv_flags Neil Horman
2011-07-28  5:42     ` David Miller
2011-07-28  8:15     ` Robert Olsson
2011-07-28 10:50       ` Neil Horman
2011-07-26 16:05   ` [PATCH 2/2] net: Audit drivers to identify those needing IFF_TX_SKB_SHARING cleared Neil Horman
2011-07-28  5:42     ` David Miller
2011-07-29  6:16   ` [PATCH 0/2] pktgen: Clone skb to avoid corruption of skbs in ndo_start_xmit methods (v3) Michael S. Tsirkin
2011-07-29 11:19     ` Neil Horman
2011-08-17 15:07   ` Ben Hutchings
2011-08-22  0:27     ` Neil Horman
2011-08-22 16:17       ` Ben Hutchings
2011-08-22 17:33         ` Ben Hutchings
2011-08-22 18:14         ` Neil Horman
2011-08-23 14:01           ` Ben Hutchings
2011-08-23 15:13             ` Neil Horman
2011-08-23 15:53               ` Ben Hutchings
2011-08-23 16:21                 ` Neil Horman

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=4E270350.7050904@candelatech.com \
    --to=greearb@candelatech.com \
    --cc=adobriyan@gmail.com \
    --cc=davem@davemloft.net \
    --cc=eric.dumazet@gmail.com \
    --cc=jpirko@redhat.com \
    --cc=netdev@vger.kernel.org \
    --cc=nhorman@tuxdriver.com \
    --cc=robert.olsson@its.uu.se \
    /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.