Date: Wed, 10 Nov 2004 12:03:13 +0100
From: Ingo Molnar <mingo@elte.hu>
To: Andrew Morton <akpm@osdl.org>
Cc: Thomas Hood <jdthood@yahoo.co.uk>, linux-kernel@vger.kernel.org
Subject: Re: [PATCH] Documentation/preempt-locking.txt clarification
Message-ID: <20041110110313.GA5493@elte.hu>
References: <1073302283.1903.85.camel@thanatos.hubertnet> <1074561880.26456.26.camel@localhost> <1100074907.3654.780.camel@thanatos> <20041110005742.35828d2b.akpm@osdl.org>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <20041110005742.35828d2b.akpm@osdl.org>
User-Agent: Mutt/1.4.1i
Sender: linux-kernel-owner@vger.kernel.org
Content-Length: 866
Lines: 20


* Andrew Morton <akpm@osdl.org> wrote:

> But I don't see why that's needed: if the preempt command came from
> another CPU then this CPU will take the cross-CPU interrupt as soon as
> interrupts are enabled.  And the preempt command couldn't have come
> from _this_ CPU, because it had interrupts disabled.

this CPU can easily trigger a reschedule by _waking_ another task
synchronously (from within the irqs-off section) and hence causing
itself to preempt. I'd not say it's common, but it can happen in quite
unexpected places, so i'd encourage us to keep the more cautious advice
that is in the current text.

	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/