Re: [PATCH] make lock/release

netdev

Re: [PATCH] make lock/release_sock fast calls

To:	Stephen Hemminger <shemminger@xxxxxxxx>
Subject:	Re: [PATCH] make lock/release_sock fast calls
From:	Christoph Hellwig <hch@xxxxxxxxxxxxx>
Date:	Thu, 1 Apr 2004 05:45:21 +0100
Cc:	"David S. Miller" <davem@xxxxxxxxxx>, netdev@xxxxxxxxxxx
In-reply-to:	<20040331112407.66309027@xxxxxxxxxxxxxxxxxxxxx>; from shemminger@xxxxxxxx on Wed, Mar 31, 2004 at 11:24:07AM -0800
References:	<20040331112407.66309027@xxxxxxxxxxxxxxxxxxxxx>
Sender:	netdev-bounce@xxxxxxxxxxx
User-agent:	Mutt/1.2.5.1i

On Wed, Mar 31, 2004 at 11:24:07AM -0800, Stephen Hemminger wrote:
> lock_sock and release_sock are no longer inline, but they are called in
> the critical path of many operations, so it makes sense to use register based
> fastcall.

Not sure I agree.  If you care for performance you compile with
CONFIG_REGPARAM set, no?

[More with this subject...]

<Prev in Thread]	Current Thread	[Next in Thread>
[PATCH] make lock/release_sock fast calls, Stephen Hemminger Re: [PATCH] make lock/release_sock fast calls, Christoph Hellwig <= Re: [PATCH] make lock/release_sock fast calls, David S. Miller

Previous by Date:	Re: [RFC] Use RCU for fib_rules, Paul McKenney
Next by Date:	Re: [PATCH]dump interface IPv6 multicast/anycast addresses through netlink, YOSHIFUJI Hideaki / 吉藤英明
Previous by Thread:	[PATCH] make lock/release_sock fast calls, Stephen Hemminger
Next by Thread:	Re: [PATCH] make lock/release_sock fast calls, David S. Miller
Indexes:	[Date] [Thread] [Top] [All Lists]