From: Roopa Prabhu <roopa@cumulusnetworks•com>
To: David Miller <davem@davemloft•net>
Cc: netdev@vger•kernel.org, jhs@mojatatu•com, tgraf@suug•ch,
nicolas.dichtel@6wind•com
Subject: Re: [PATCH net-next v5] rtnetlink: add new RTM_GETSTATS message to dump link stats
Date: Mon, 18 Apr 2016 21:43:32 -0700 [thread overview]
Message-ID: <5715B774.9070208@cumulusnetworks.com> (raw)
In-Reply-To: <20160418.234137.26692086759328024.davem@davemloft.net>
On 4/18/16, 8:41 PM, David Miller wrote:
> From: Roopa Prabhu <roopa@cumulusnetworks•com>
> Date: Mon, 18 Apr 2016 14:10:19 -0700
>
>> This patch adds a new RTM_GETSTATS message to query link stats via
>> netlink from the kernel. RTM_NEWLINK also dumps stats today, but
>> RTM_NEWLINK returns a lot more than just stats and is expensive in
>> some cases when frequent polling for stats from userspace is a
>> common operation.
> I'm holding off on this until we sort out the 64-bit netlink
> attribute alignment issue.
sure,
>
> Meanwhile, I'll some kind of a fix into the tree for the
> rtnl_fill_stats() change so that it doesn't cause unaligned
> accesses.
>
> I just tested out a clever idea, where for architectures where
> unaligned accesses is a problem, we insert a zero length NOP attribute
> before the 64-bit stats. This makes it properly aligned. A quick
> hack patch just passed testing on my sparc64 box, but I'll go over it
> some more.
>
> diff --git a/include/uapi/linux/if_link.h b/include/uapi/linux/if_link.h
> index bb3a90b..5ffdcb3 100644
> --- a/include/uapi/linux/if_link.h
> +++ b/include/uapi/linux/if_link.h
> @@ -155,6 +155,7 @@ enum {
> IFLA_PROTO_DOWN,
> IFLA_GSO_MAX_SEGS,
> IFLA_GSO_MAX_SIZE,
> + IFLA_PAD,
> __IFLA_MAX
> };
>
> diff --git a/net/core/rtnetlink.c b/net/core/rtnetlink.c
> index a7a3d34..b192576 100644
> --- a/net/core/rtnetlink.c
> +++ b/net/core/rtnetlink.c
> @@ -1052,6 +1052,15 @@ static noinline_for_stack int rtnl_fill_stats(struct sk_buff *skb,
> struct rtnl_link_stats64 *sp;
> struct nlattr *attr;
>
> + /* Add a zero length NOP attribute so that the nla_data()
> + * of the IFLA_STATS64 will be 64-bit aligned.
> + */
> +#ifndef HAVE_EFFICIENT_UNALIGNED_ACCESS
> + attr = nla_reserve(skb, IFLA_PAD, 0);
> + if (!attr)
> + return -EMSGSIZE;
> +#endif
> +
> attr = nla_reserve(skb, IFLA_STATS64,
> sizeof(struct rtnl_link_stats64));
> if (!attr)
>
that is cleaver :)
thanks!
next prev parent reply other threads:[~2016-04-19 4:43 UTC|newest]
Thread overview: 57+ messages / expand[flat|nested] mbox.gz Atom feed top
2016-04-18 21:10 [PATCH net-next v5] rtnetlink: add new RTM_GETSTATS message to dump link stats Roopa Prabhu
2016-04-18 21:35 ` Eric Dumazet
2016-04-19 0:57 ` David Miller
2016-04-19 1:48 ` David Miller
2016-04-19 2:22 ` Eric Dumazet
2016-04-19 2:40 ` Roopa Prabhu
2016-04-19 3:49 ` David Miller
2016-04-19 3:52 ` David Miller
2016-04-19 10:09 ` Johannes Berg
2016-04-19 10:48 ` Emmanuel Grumbach
2016-04-19 18:23 ` David Miller
2016-04-19 19:41 ` Johannes Berg
2016-04-20 1:53 ` David Ahern
2016-04-20 7:32 ` Johannes Berg
2016-04-20 12:48 ` Jiri Benc
2016-04-20 13:17 ` Johannes Berg
2016-04-20 13:34 ` Jiri Benc
2016-04-20 20:13 ` Johannes Berg
2016-04-19 2:30 ` roopa
2016-04-19 3:41 ` David Miller
2016-04-19 4:17 ` Eric Dumazet
2016-04-19 4:32 ` Eric Dumazet
2016-04-19 5:03 ` David Miller
2016-04-19 18:31 ` David Miller
2016-04-19 18:45 ` Eric Dumazet
2016-04-19 18:47 ` Eric Dumazet
2016-04-19 19:08 ` Nicolas Dichtel
2016-04-19 23:50 ` David Miller
2016-04-20 3:54 ` Roopa Prabhu
2016-04-20 8:57 ` [PATCH net-next 0/4] libnl: enhance API to ease 64bit alignment for attribute Nicolas Dichtel
2016-04-20 8:57 ` [PATCH net-next 1/4] netlink: fix test alignment in nla_align_64bit() Nicolas Dichtel
2016-04-20 9:33 ` Eric Dumazet
2016-04-20 9:44 ` Nicolas Dichtel
2016-04-20 9:57 ` Eric Dumazet
2016-04-20 10:14 ` Nicolas Dichtel
2016-04-20 14:31 ` [PATCH net-next] net: fix HAVE_EFFICIENT_UNALIGNED_ACCESS typos Eric Dumazet
2016-04-20 15:03 ` David Miller
2016-04-20 8:57 ` [PATCH net-next 2/4] libnl: add more helpers to align attribute on 64-bit Nicolas Dichtel
2016-04-20 8:57 ` [PATCH net-next 3/4] ipmr: align RTA_MFC_STATS " Nicolas Dichtel
2016-04-20 8:57 ` [PATCH net-next 4/4] ip6mr: " Nicolas Dichtel
2016-04-21 16:58 ` [PATCH net-next v2 0/4] libnl: enhance API to ease 64bit alignment for attribute Nicolas Dichtel
2016-04-21 16:58 ` [PATCH net-next v2 1/4] libnl: add more helpers to align attributes on 64-bit Nicolas Dichtel
2016-04-21 16:58 ` [PATCH net-next v2 2/4] rtnl: use the new API to align IFLA_STATS* Nicolas Dichtel
2016-04-21 16:58 ` [PATCH net-next v2 3/4] ipmr: align RTA_MFC_STATS on 64-bit Nicolas Dichtel
2016-04-21 16:58 ` [PATCH net-next v2 4/4] ip6mr: " Nicolas Dichtel
2016-04-21 18:28 ` [PATCH net-next v2 0/4] libnl: enhance API to ease 64bit alignment for attribute David Miller
2016-04-21 22:00 ` Nicolas Dichtel
2016-04-22 5:31 ` David Miller
2016-04-19 19:05 ` [PATCH net-next v5] rtnetlink: add new RTM_GETSTATS message to dump link stats Roopa Prabhu
2016-04-19 22:49 ` David Miller
2016-04-20 3:53 ` Roopa Prabhu
2016-04-19 4:43 ` Roopa Prabhu [this message]
2016-04-19 7:45 ` Nicolas Dichtel
2016-04-19 16:00 ` David Miller
2016-04-19 8:26 ` Nicolas Dichtel
2016-04-19 19:55 ` Paul Moore
2016-04-19 20:40 ` Roopa Prabhu
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=5715B774.9070208@cumulusnetworks.com \
--to=roopa@cumulusnetworks$(echo .)com \
--cc=davem@davemloft$(echo .)net \
--cc=jhs@mojatatu$(echo .)com \
--cc=netdev@vger$(echo .)kernel.org \
--cc=nicolas.dichtel@6wind$(echo .)com \
--cc=tgraf@suug$(echo .)ch \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
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