From mboxrd@z Thu Jan 1 00:00:00 1970 From: William Lee Irwin III Date: Mon, 09 Aug 2004 19:12:49 +0000 Subject: Re: Hugetlb demanding paging for -mm tree Message-Id: <20040809191249.GT11200@holomorphy.com> List-Id: References: <20040806210750.GT17188@holomorphy.com> <200408091819.i79IJ3Y12216@unix-os.sc.intel.com> In-Reply-To: <200408091819.i79IJ3Y12216@unix-os.sc.intel.com> MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit To: "Chen, Kenneth W" Cc: 'Hirokazu Takahashi' , linux-kernel@vger.kernel.org, linux-ia64@vger.kernel.org, "Seth, Rohit" William Lee Irwin III wrote on Friday, August 06, 2004 2:08 PM >> update_mmu_cache() does not appear to check the size of the translation >> to be established in many architectures. e.g. on arch/ia64/ it does >> flush_icache_range(addr, addr + PAGE_SIZE) unconditionally, and only >> sets PG_arch_1 on a single struct page. Similar comments apply to >> sparc64 and ppc64; I didn't check any others. On Mon, Aug 09, 2004 at 11:19:04AM -0700, Chen, Kenneth W wrote: > I suppose this is fixable in update_mmu_cache() where it can check the > type of pte and do appropriate sizing and other things. ia64 would have > to check the address instead of looking at the pte. Yes, it's just a fair amount of document-hunting since there isn't always easily cut-and-pasteable stuff (e.g. ITAG_MASK for larger page sizes was omitted from the #ifdefs on sparc64). As for ia64 checking addresses... ew, can't we just use long format VHPT? The virtual placement constraints are nasty. -- wli