netdev
[Top] [All Lists]

Re: [PATCH 2.6.11-rc4-mm1] connector: Add a fork connector

To: Guillaume Thouvenin <guillaume.thouvenin@xxxxxxxx>
Subject: Re: [PATCH 2.6.11-rc4-mm1] connector: Add a fork connector
From: Paul Jackson <pj@xxxxxxx>
Date: Wed, 2 Mar 2005 06:51:52 -0800
Cc: akpm@xxxxxxxx, linux-kernel@xxxxxxxxxxxxxxx, johnpol@xxxxxxxxxxx, elsa-devel@xxxxxxxxxxxxxxxxxxxxx, jlan@xxxxxxxxxxxx, gh@xxxxxxxxxx, efocht@xxxxxxxxxxxx, netdev@xxxxxxxxxxx, kaigai@xxxxxxxxxxxxx
In-reply-to: <1109753292.8422.117.camel@xxxxxxxxxxxxxxxxxxxxxxxx>
Organization: SGI
References: <1109240677.1738.196.camel@xxxxxxxxxxxxxxxxxxxxxxxx> <1109753292.8422.117.camel@xxxxxxxxxxxxxxxxxxxxxxxx>
Sender: netdev-bounce@xxxxxxxxxxx
Guillaume wrote:
> 
>   I also run the lmbench and results are send in response to another
> thread "A common layer for Accounting packages". When fork connector is
> turned off the overhead is negligible. 

Good.

If I read this code right:
>
> +static inline void fork_connector(pid_t parent, pid_t child)
> +{
> +     static DEFINE_SPINLOCK(cn_fork_lock);
> +     static __u32 seq;   /* used to test if message is lost */
> +
> +     if (cn_fork_enable) {

then the code executed if the fork connector is off is a call to an
inline function that tests an integer, finds it zero, and returns.

This is sufficiently little code that I for one would hardly
even need lmbench to be comfortable that fork() wasn't impacted
seriously, in the case that the fork connector is disabled.

-- 
                  I won't rest till it's the best ...
                  Programmer, Linux Scalability
                  Paul Jackson <pj@xxxxxxx> 1.650.933.1373, 1.925.600.0401

<Prev in Thread] Current Thread [Next in Thread>