From mboxrd@z Thu Jan 1 00:00:00 1970 Date: Thu, 8 May 2008 16:34:53 +0200 From: Hans Rosenfeld Subject: Re: [PATCH] x86: fix PAE pmd_bad bootup warning Message-ID: <20080508143453.GE12654@escobedo.amd.com> References: <20080506124946.GA2146@elte.hu> <20080506202201.GB12654@escobedo.amd.com> <1210106579.4747.51.camel@nimitz.home.sr71.net> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <1210106579.4747.51.camel@nimitz.home.sr71.net> Sender: owner-linux-mm@kvack.org Return-Path: To: Dave Hansen Cc: Hugh Dickins , Ingo Molnar , Jeff Chua , Thomas Gleixner , "H. Peter Anvin" , Gabriel C , Arjan van de Ven , Nishanth Aravamudan , linux-kernel@vger.kernel.org, linux-mm@kvack.org List-ID: On Tue, May 06, 2008 at 01:42:59PM -0700, Dave Hansen wrote: > Could you post the code you're using to do this? I have to wonder if > you're leaving a fd open somewhere. Even if you rm the hugepage file, > it'll stay allocated if you have a fd open, or if *someone* is still > mapping it. A stripped-down program exposing the leak is attached. It doesn't fork or do any other fancy stuff, so I don't see a way it could leave any fd open. While trying to reproduce this, I noticed that the huge page wouldn't leak when I just mmapped it and exited without explicitly unmapping, as I described before. The huge page is leaked only when the /proc/self/pagemap entry for the huge page is read. The kernel I tested with was 2.6.26-rc1-00179-g3de2403. > Can you umount your hugetlbfs? Yes, but the pages remain lost. Hans -- %SYSTEM-F-ANARCHISM, The operating system has been overthrown -- 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-mm.org/ . Don't email: email@kvack.org