netdev.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Shmulik Ladkani <shmulik.ladkani@gmail.com>
To: Eyal Birger <eyal.birger@gmail.com>
Cc: davem@davemloft.net, willemb@google.com, edumazet@google.com,
	marcel@holtmann.org, netdev@vger.kernel.org
Subject: Re: [PATCH net-next 7/7] net: move skb->dropcount to skb->cb[]
Date: Thu, 26 Feb 2015 10:49:09 +0200	[thread overview]
Message-ID: <20150226104909.1f2c882d@halley> (raw)
In-Reply-To: <1424916612-744-8-git-send-email-eyal.birger@gmail.com>

On Thu, 26 Feb 2015 04:10:12 +0200 Eyal Birger <eyal.birger@gmail.com> wrote:
> diff --git a/include/net/sock.h b/include/net/sock.h
> index d462f5e..8807586 100644
> --- a/include/net/sock.h
> +++ b/include/net/sock.h
> @@ -2078,12 +2078,26 @@ static inline int sock_intr_errno(long timeo)
>  	return timeo == MAX_SCHEDULE_TIMEOUT ? -ERESTARTSYS : -EINTR;
>  }
>  
> +struct sock_skb_cb {
> +	u32 dropcount;
> +};
> +
> +/* Store sock_skb_cb at the end of skb->cb[] so protocol families
> + * using skb->cb[] would keep using it directly and utilize its
> + * alignement guarantee.
> + */
> +#define SOCK_SKB_CB_OFFSET ((FIELD_SIZEOF(struct sk_buff, cb) - \
> +			    sizeof(struct sock_skb_cb)))
> +
> +#define SOCK_SKB_CB(__skb) ((struct sock_skb_cb *)((__skb)->cb + \
> +			    SOCK_SKB_CB_OFFSET))
> +
>  #define sock_skb_cb_check_size(size) \
> -	BUILD_BUG_ON((size) > FIELD_SIZEOF(struct sk_buff, cb))
> +	BUILD_BUG_ON((size) > SOCK_SKB_CB_OFFSET)
>  

You didn't take care of aligning the 'sock_skb_cb'.
(Althogh in practive, dropcount is 4 and cb[] is 48 so you're golden).

How about:

struct sock_skb_cb {
	/* protocol families specifc CBs */
	u8 reserved[__SOCK_SKB_CB_OFFSET];
	struct __sock_skb_cb x;			// name me better!
};

Where '__SOCK_SKB_CB_OFFSET' and '__sock_skb_cb' are as follows:

struct __sock_skb_cb {	// name me better!
	u32 dropcount;
};

#define __SOCK_SKB_CB_OFFSET_UNALIGNED \
	(FIELD_SIZEOF(struct sk_buff, cb) - sizeof(struct __sock_skb_cb))

#define __SOCK_SKB_CB_OFFSET \
	(__SOCK_SKB_CB_OFFSET_UNALIGNED - \
	 (__SOCK_SKB_CB_OFFSET_UNALIGNED % __alignof__(struct __sock_skb_cb)))

This also takes care for alignement of '__sock_skb_cb x' within the CB.

Then,

#define sock_skb_cb_check_size(size) \
	BUILD_BUG_ON((size) > FIELD_SIZEOF(struct sock_skb_cb, reserved))

Regards,
Shmulik

  reply	other threads:[~2015-02-26  8:49 UTC|newest]

Thread overview: 30+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2015-02-26  2:10 [PATCH net-next 0/7] net: move skb->dropcount to skb->cb[] Eyal Birger
2015-02-26  2:10 ` [PATCH net-next 1/7] net: bluetooth: compact struct bt_skb_cb by inlining struct hci_req_ctrl Eyal Birger
2015-02-26  7:02   ` Shmulik Ladkani
2015-02-26 16:15   ` Marcel Holtmann
2015-02-26 21:22     ` Eyal Birger
2015-02-26  2:10 ` [PATCH net-next 2/7] net: bluetooth: compact struct bt_skb_cb by converting boolean fields to bit fields Eyal Birger
2015-02-26  7:30   ` Shmulik Ladkani
2015-02-26 13:22     ` Eyal Birger
2015-02-26 11:31   ` David Laight
2015-02-26 13:21     ` Eyal Birger
2015-02-26  2:10 ` [PATCH net-next 3/7] net: rxrpc: change call to sock_recv_ts_and_drops() on rxrpc recvmsg to sock_recv_timestamp() Eyal Birger
2015-02-26  4:40   ` Eric Dumazet
2015-02-26  7:30     ` Eyal Birger
2015-02-26 14:31       ` Eric Dumazet
2015-02-26 14:55         ` Eyal Birger
2015-02-26  2:10 ` [PATCH net-next 4/7] net: packet: use skb->dev as storage for skb orig len instead of skb->cb[] Eyal Birger
2015-02-26  4:03   ` Eyal Birger
2015-02-26 17:26   ` Willem de Bruijn
2015-02-26 18:38     ` Eyal Birger
2015-02-26 18:51       ` Willem de Bruijn
2015-02-26 18:55         ` Eyal Birger
2015-02-26  2:10 ` [PATCH net-next 5/7] net: use common macro for assering skb->cb[] available size in protocol families Eyal Birger
2015-02-26  2:10 ` [PATCH net-next 6/7] net: add common accessor for setting dropcount on packets Eyal Birger
2015-02-26  4:43   ` Eric Dumazet
2015-02-26  2:10 ` [PATCH net-next 7/7] net: move skb->dropcount to skb->cb[] Eyal Birger
2015-02-26  8:49   ` Shmulik Ladkani [this message]
2015-02-26 10:17     ` Eyal Birger
2015-02-26 10:48       ` Eyal Birger
2015-02-26 15:32         ` Shmulik Ladkani
2015-02-26 18:52           ` Eyal Birger

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=20150226104909.1f2c882d@halley \
    --to=shmulik.ladkani@gmail.com \
    --cc=davem@davemloft.net \
    --cc=edumazet@google.com \
    --cc=eyal.birger@gmail.com \
    --cc=marcel@holtmann.org \
    --cc=netdev@vger.kernel.org \
    --cc=willemb@google.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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).