From: Eric Dumazet <eric.dumazet@gmail.com>
To: Ben Greear <greearb@candelatech.com>
Cc: Stephen Hemminger <shemminger@vyatta.com>,
Rick Jones <rick.jones2@hp.com>, Tom Parkin <tparkin@katalix.com>,
netdev@vger.kernel.org, David.Laight@ACULAB.COM,
James Chapman <jchapman@katalix.com>
Subject: Re: [PATCH v2] l2tp: use per-cpu variables for u64_stats updates
Date: Wed, 27 Jun 2012 23:50:49 +0200 [thread overview]
Message-ID: <1340833849.26242.186.camel@edumazet-glaptop> (raw)
In-Reply-To: <4FEB7DC0.9090404@candelatech.com>
On Wed, 2012-06-27 at 14:40 -0700, Ben Greear wrote:
> Notice that the netlink stats are claiming 64-bit and they are not
> (always) 64-bit.
There is no such claim.
netlink provides a 64bit generic binary holder, even for legacy drivers
still using 'unsigned long' stats, so obviously 32bit values on 32bit
kernels.
> That is a nice binary API that is still wrapping before its time
> in many cases. There may be good performance reasons for keeping
> some counters at 32-bit, but it plays merry hell with anyone wanting
> to optimize an application to poll less often for stats that are
> supposedly 64-bit.
We dont want hundred of patches to bring 64bit stats on legacy drivers.
Nobody cares, because its way too late to try to 'fix' this.
If you want your application running on linux-2.6.x, or linux-3.0, you
are forced to correctly detect each counter being 32 or 64, not because
netlink API is 64bit, but because a driver provides a 32 or 64 bit
value.
next prev parent reply other threads:[~2012-06-27 21:50 UTC|newest]
Thread overview: 21+ messages / expand[flat|nested] mbox.gz Atom feed top
2012-06-27 12:00 [PATCH v2] l2tp: use per-cpu variables for u64_stats updates Tom Parkin
2012-06-27 19:03 ` Eric Dumazet
2012-06-27 20:21 ` Rick Jones
2012-06-27 20:39 ` Eric Dumazet
2012-06-27 20:50 ` Stephen Hemminger
2012-06-27 20:58 ` Ben Greear
2012-06-27 21:20 ` Eric Dumazet
2012-06-27 21:31 ` Ben Greear
2012-06-27 21:35 ` Eric Dumazet
2012-06-27 23:01 ` Rick Jones
2012-06-27 23:09 ` David Miller
2012-06-27 23:39 ` Rick Jones
2012-06-28 5:00 ` Eric Dumazet
2012-06-28 8:24 ` Tom Parkin
2012-06-28 8:46 ` David Laight
2012-06-28 18:17 ` Ben Hutchings
2012-06-27 21:32 ` Eric Dumazet
2012-06-27 21:40 ` Ben Greear
2012-06-27 21:50 ` Eric Dumazet [this message]
2012-06-27 21:00 ` Rick Jones
2012-06-27 22:21 ` David Miller
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=1340833849.26242.186.camel@edumazet-glaptop \
--to=eric.dumazet@gmail.com \
--cc=David.Laight@ACULAB.COM \
--cc=greearb@candelatech.com \
--cc=jchapman@katalix.com \
--cc=netdev@vger.kernel.org \
--cc=rick.jones2@hp.com \
--cc=shemminger@vyatta.com \
--cc=tparkin@katalix.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