From mboxrd@z Thu Jan 1 00:00:00 1970 From: Chris Wedgwood Date: Sun, 22 Feb 2004 05:27:59 +0000 Subject: Re: HUGEPAGE SIZE a boottime option Message-Id: <20040222052759.GA14666@dingdong.cryptoapps.com> List-Id: References: <20040220010731.GA28820@sgi.com> In-Reply-To: <20040220010731.GA28820@sgi.com> MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit To: linux-ia64@vger.kernel.org On Fri, Feb 20, 2004 at 11:36:38AM -0800, Seth, Rohit wrote: > Ken is currently in the process of providing the support for boot > time configuration for huge page size. What's wrong with Jack's patch? > As long as there are only two supported page sizes (PAGE_SIZE and > HPAGE_SIZE), it is possible to achieve the boot time configuration > of HUGE_PAGE_SIZE without needing to patch the kernel (and possibly > better packing the vhpt fault handler code). We will ship the patch > out to the list. Jack's code doesn't have this limitation. IMO it's a more sane way to go. Why should people be restricted to a limited set (two) of hugepage sizes? This seems especially important for vendor kernels where people want to use the same kernel across a variety of different machines.