netdev - Re: [PATCH net-next-2.6] net: sock_def_readable() and friends RCU conversion

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Date:	Sat, 01 May 2010 06:56:04 +0200
From:	Eric Dumazet <eric.dumazet@...il.com>
To:	David Miller <davem@...emloft.net>
Cc:	hadi@...erus.ca, xiaosuo@...il.com, therbert@...gle.com,
	shemminger@...tta.com, netdev@...r.kernel.org, eilong@...adcom.com,
	bmb@...enacr.com
Subject: Re: [PATCH net-next-2.6] net: sock_def_readable() and friends RCU
 conversion

Le vendredi 30 avril 2010 à 16:35 -0700, David Miller a écrit :
> From: Eric Dumazet <eric.dumazet@...il.com>
> Date: Thu, 29 Apr 2010 23:01:49 +0200
> 
> > [PATCH net-next-2.6] net: sock_def_readable() and friends RCU conversion
> 
> So what's the difference between call_rcu() freeing this little waitqueue
> struct and doing it for the entire socket?
> 
> We'll still be doing an RCU call every socket destroy, and now we also have
> a new memory allocation/free per connection.
> 
> This has to show up in things like 'lat_connect' and friends, does it not?

Difference is this structure is small, one cache line at most.

So the cost of call_rcu() on this structure, with the well known cache
miss is very much reduced.

The thing that might cost is the smp_mb(), because it translate to a
"mfence" instruction, and it appears to cost more than a a regular
"lock ..."

Unfortunatly, oprofile doesnt work anymore on my bl460c machine after
last BIOS upgrade... Oh well...



--
To unsubscribe from this list: send the line "unsubscribe netdev" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html