From mboxrd@z Thu Jan 1 00:00:00 1970 Date: Tue, 30 Jan 2001 12:38:12 +0100 From: Rasmus Andersen Subject: Re: [PATCH] guard mm->rss with page_table_lock (241p11) Message-ID: <20010130123812.O3298@jaquet.dk> References: <20010131001737.C6620@metastasis.f00f.org> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: ; from riel@conectiva.com.br on Tue, Jan 30, 2001 at 09:23:27AM -0200 Sender: owner-linux-mm@kvack.org Return-Path: To: Rik van Riel Cc: Chris Wedgwood , "David S. Miller" , David Howells , linux-kernel@vger.kernel.org, linux-mm@kvack.org List-ID: On Tue, Jan 30, 2001 at 09:23:27AM -0200, Rik van Riel wrote: > Why bother ? > > In most places where we update mm->rss, we are *already* > holding the spinlock anyway, this correction is just for > a few places. > > The big patch Rasmus made seems to contain spin_lock(&foo) > in places where we already have the lock, leading to > instant SMP deadlock. I suspect Rasmus' patch should be > about half the size it is currently... After donning my brown paper bag yesterday I looked at the call-paths again and removed one more lock pair (the one in swapfile). The others seemed OK so I made a SMP-on-UP kernel and ran my usual stuff (X, mozilla, kernel compiles) alongside mmap001, mmap002 and misc001 with no ill effects. I will beat on it some more today and tomorrow, but if real SMP is needed for testing I need some help to do that. Regards, Rasmus -- To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to majordomo@kvack.org. For more info on Linux MM, see: http://www.linux.eu.org/Linux-MM/