Date: Fri, 29 Feb 2008 20:44:57 +0100
From: Ingo Molnar <mingo@elte.hu>
To: David Miller <davem@davemloft.net>
Cc: paulus@samba.org, alexey.zaytsev@gmail.com, tglx@linutronix.de,
       linux-kernel@vger.kernel.org
Subject: Re: printk timestamps jumping back and forth in 2.6.25-rc.
Message-ID: <20080229194457.GB21249@elte.hu>
References: <20080228201315.GA30736@elte.hu> <20080228201613.GA30999@elte.hu> <18375.59486.392919.984544@cargo.ozlabs.ibm.com> <20080229.104235.192308716.davem@davemloft.net>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <20080229.104235.192308716.davem@davemloft.net>
User-Agent: Mutt/1.5.17 (2007-11-01)
Sender: linux-kernel-owner@vger.kernel.org
Content-Length: 874
Lines: 23


* David Miller <davem@davemloft.net> wrote:

> From: Paul Mackerras <paulus@samba.org>
> Date: Fri, 29 Feb 2008 22:11:26 +1100
> 
> > On powerpc we use the timebase register, which is synchronized across
> > CPUs, so if this will add unnecessary overhead, we will want to way to
> > turn it off on powerpc (assuming this ends up going upstream).
> 
> Similarly on sparc64, all of our TSCs are fully synchronized so this 
> complexity is completely unnecessary.

i suspect for them the easiest way is to expose time_sync_thresh and let 
them set it to ULONG_MAX - which skips all synchronization overhead and 
complexity.

	Ingo
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/