From mboxrd@z Thu Jan 1 00:00:00 1970 From: Keith Owens Date: Fri, 26 Mar 2004 03:39:09 +0000 Subject: Re: [Lse-tech] Re: [PATCH] [0/6] HUGETLB memory commitment Message-Id: <5310.1080272349@kao2.melbourne.sgi.com> List-Id: In-Reply-To: Your message of "Fri, 26 Mar 2004 14:28:26 +0530." <20040326085826.GA3332@in.ibm.com> MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit To: suparna@in.ibm.com Cc: Andrew Morton , apw@shadowen.org, anton@samba.org, sds@epoch.ncsc.mil, ak@suse.de, raybry@sgi.com, lse-tech@lists.sourceforge.net, linux-ia64@vger.kernel.org, linux-kernel@vger.kernel.org, mbligh@aracnet.com On Fri, 26 Mar 2004 14:28:26 +0530, Suparna Bhattacharya wrote: >On Thu, Mar 25, 2004 at 04:22:32PM -0800, Andrew Morton wrote: >> Keith Owens wrote: >> > >> > FWIW, lkcd (crash dump) treats hugetlb pages as normal kernel pages and >> > dumps them, which is pointless and wastes a lot of time. To avoid >> > dumping these pages in lkcd, I had to add a PG_hugetlb flag. lkcd runs > >This should already be fixed in recent versions of lkcd. It uses a >little bit of trickery to avoid an extra page flag -- hugetlb pages are >detected as "in use" as well as reserved, unlike other reserved pages >which helps identify them. Are you sure that this works for hugetlb pages that have been preallocated but not yet mapped? AFAICT the hugetlb pages start off as reserved with a zero usecount.