[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-Id: <20081119.154153.233063184.davem@davemloft.net>
Date: Wed, 19 Nov 2008 15:41:53 -0800 (PST)
From: David Miller <davem@...emloft.net>
To: dada1@...mosbay.com
Cc: netdev@...r.kernel.org
Subject: Re: [RFC] question about inet_listen_lock(), and [PATCH] net:
listening_hash get a spinlock per bucket
From: Eric Dumazet <dada1@...mosbay.com>
Date: Mon, 17 Nov 2008 17:21:19 +0100
> I was looking at converting tcp ehash/listening_hash rwlocks to spinlocks, and found
> inet_listen_lock()/unlock() definition :
I've known about this weird "bump a counter and sleep" locking thing we do
for listening sockets, but I haven't been able yet to track down the history
and understand why we do this.
There really was a reason, and your suspicion about /proc/net/tcp might be
correct, and of course doesn't really apply these days.
But I don't know for sure yet, and I'd rather not apply thing until I do.
Give me some more time to do the history search, thanks.
--
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
Powered by blists - more mailing lists