[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20110506171701.GJ11636@one.firstfloor.org>
Date: Fri, 6 May 2011 19:17:01 +0200
From: Andi Kleen <andi@...stfloor.org>
To: Eric Dumazet <eric.dumazet@...il.com>
Cc: Andi Kleen <andi@...stfloor.org>,
Thomas Gleixner <tglx@...utronix.de>,
john stultz <johnstul@...ibm.com>,
lkml <linux-kernel@...r.kernel.org>,
Paul Mackerras <paulus@...ba.org>,
"Paul E. McKenney" <paulmck@...ux.vnet.ibm.com>,
Anton Blanchard <anton@...ba.org>, Ingo Molnar <mingo@...e.hu>
Subject: Re: [RFC] time: xtime_lock is held too long
> Unless I am very mistaken, it cannot explain my original report of up to
> 24.000 instructions waiting in ktime_get()
Yes it's a different problem. Sorry for hijacking your thread.
>
> BTW, how many consecutive prefetch() a cpu can 'store/execute' before
> throwing away hints ?
It depends on the CPU. Older ones had a small number only. On the
newer ones it tends to be limited by the various queues to memory,
which are a shared resource (so it also depends on what the other
threads/cores are doing)
I normally assume as a rule of thumb < 50
-Andi
--
ak@...ux.intel.com -- Speaking for myself only.
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
Powered by blists - more mailing lists