From: Peter Zijlstra <peterz@infradead.org>
To: Herbert Xu <herbert@gondor.apana.org.au>
Cc: David Miller <davem@davemloft.net>,
eric.dumazet@gmail.com, elena.reshetova@intel.com,
keescook@chromium.org, netdev@vger.kernel.org,
bridge@lists.linux-foundation.org, linux-kernel@vger.kernel.org,
kuznet@ms2.inr.ac.ru, jmorris@namei.org, kaber@trash.net,
stephen@networkplumber.org, ishkamiel@gmail.com,
dwindsor@gmail.com, akpm@linux-foundation.org
Subject: Re: [PATCH 07/17] net: convert sock.sk_refcnt from atomic_t to refcount_t
Date: Mon, 20 Mar 2017 14:40:17 +0100 [thread overview]
Message-ID: <20170320134017.h3c2jrsnd4guuyu7@hirez.programming.kicks-ass.net> (raw)
In-Reply-To: <20170320132713.GA26954@gondor.apana.org.au>
On Mon, Mar 20, 2017 at 09:27:13PM +0800, Herbert Xu wrote:
> On Mon, Mar 20, 2017 at 02:23:57PM +0100, Peter Zijlstra wrote:
> >
> > So what bench/setup do you want ran?
>
> You can start by counting how many cycles an atomic op takes
> vs. how many cycles this new code takes.
On what uarch?
I think I tested hand coded asm version and it ended up about double the
cycles for a cmpxchg loop vs the direct instruction on an IVB-EX (until
the memory bus saturated, at which point they took the same). Newer
parts will of course have different numbers,
Can't we run some iperf on a 40gbe fiber loop or something? It would be
very useful to have an actual workload we can run.
next prev parent reply other threads:[~2017-03-20 13:40 UTC|newest]
Thread overview: 54+ messages / expand[flat|nested] mbox.gz Atom feed top
2017-03-16 15:28 [PATCH 00/17] net subsystem refcount conversions Elena Reshetova
2017-03-16 15:28 ` [PATCH 01/17] net: convert neighbour.refcnt from atomic_t to refcount_t Elena Reshetova
2017-03-16 15:28 ` [PATCH 02/17] net: convert neigh_params.refcnt " Elena Reshetova
2017-03-16 15:28 ` [PATCH 03/17] net: convert nf_bridge_info.use " Elena Reshetova
2017-03-16 15:28 ` [PATCH 04/17] net: convert sk_buff.users " Elena Reshetova
2017-03-16 15:28 ` [PATCH 05/17] net: convert sk_buff_fclones.fclone_ref " Elena Reshetova
2017-03-16 15:28 ` [PATCH 06/17] net: convert sock.sk_wmem_alloc " Elena Reshetova
2017-03-16 15:28 ` [PATCH 07/17] net: convert sock.sk_refcnt " Elena Reshetova
2017-03-16 16:58 ` Eric Dumazet
2017-03-16 17:38 ` Kees Cook
2017-03-16 19:10 ` David Miller
2017-03-17 7:42 ` Reshetova, Elena
2017-03-17 16:13 ` Eric Dumazet
2017-03-18 16:47 ` Herbert Xu
2017-03-19 1:21 ` David Miller
2017-03-20 10:39 ` Peter Zijlstra
2017-03-20 13:16 ` Herbert Xu
2017-03-20 13:23 ` Peter Zijlstra
2017-03-20 13:27 ` Herbert Xu
2017-03-20 13:40 ` Peter Zijlstra [this message]
2017-03-20 14:51 ` Eric Dumazet
2017-03-20 14:59 ` Eric Dumazet
2017-03-20 16:18 ` Eric Dumazet
2017-03-20 16:34 ` Eric Dumazet
2017-03-20 14:59 ` Peter Zijlstra
2017-03-21 20:49 ` Kees Cook
2017-03-21 21:23 ` Eric Dumazet
2017-03-21 22:36 ` David Miller
2017-03-21 23:51 ` Kees Cook
2017-03-22 2:03 ` Eric Dumazet
2017-03-22 12:25 ` Peter Zijlstra
2017-03-22 13:22 ` Eric Dumazet
2017-03-22 14:33 ` Peter Zijlstra
2017-03-22 14:54 ` Eric Dumazet
2017-03-22 15:08 ` Peter Zijlstra
2017-03-22 15:22 ` Eric Dumazet
2017-03-22 16:51 ` Peter Zijlstra
2017-03-22 19:08 ` Kees Cook
2017-03-22 12:11 ` Peter Zijlstra
2017-03-20 14:10 ` David Laight
2017-03-20 14:28 ` Peter Zijlstra
2017-03-20 15:00 ` David Laight
2017-03-16 15:28 ` [PATCH 08/17] net: convert sk_filter.refcnt " Elena Reshetova
2017-03-16 16:04 ` Daniel Borkmann
2017-03-17 8:02 ` Reshetova, Elena
2017-03-16 15:28 ` [PATCH 09/17] net: convert ip_mc_list.refcnt " Elena Reshetova
2017-03-16 15:29 ` [PATCH 10/17] net: convert in_device.refcnt " Elena Reshetova
2017-03-16 15:29 ` [PATCH 11/17] net: convert netpoll_info.refcnt " Elena Reshetova
2017-03-16 15:29 ` [PATCH 12/17] net: convert unix_address.refcnt " Elena Reshetova
2017-03-16 15:29 ` [PATCH 13/17] net: convert fib_rule.refcnt " Elena Reshetova
2017-03-16 15:29 ` [PATCH 14/17] net: convert inet_frag_queue.refcnt " Elena Reshetova
2017-03-16 15:29 ` [PATCH 15/17] net: convert net.passive " Elena Reshetova
2017-03-16 15:29 ` [PATCH 16/17] net: convert netlbl_lsm_cache.refcount " Elena Reshetova
2017-03-16 15:29 ` [PATCH 17/17] net: convert packet_fanout.sk_ref " Elena Reshetova
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=20170320134017.h3c2jrsnd4guuyu7@hirez.programming.kicks-ass.net \
--to=peterz@infradead.org \
--cc=akpm@linux-foundation.org \
--cc=bridge@lists.linux-foundation.org \
--cc=davem@davemloft.net \
--cc=dwindsor@gmail.com \
--cc=elena.reshetova@intel.com \
--cc=eric.dumazet@gmail.com \
--cc=herbert@gondor.apana.org.au \
--cc=ishkamiel@gmail.com \
--cc=jmorris@namei.org \
--cc=kaber@trash.net \
--cc=keescook@chromium.org \
--cc=kuznet@ms2.inr.ac.ru \
--cc=linux-kernel@vger.kernel.org \
--cc=netdev@vger.kernel.org \
--cc=stephen@networkplumber.org \
/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
all inboxes | Powered by JetHome®