From mboxrd@z Thu Jan 1 00:00:00 1970 Date: Fri, 6 Jul 2007 09:20:22 +0900 From: KAMEZAWA Hiroyuki Subject: Re: [BUGFIX][PATCH] DO flush icache before set_pte() on ia64. Message-Id: <20070706092022.b9b5fbcc.kamezawa.hiroyu@jp.fujitsu.com> In-Reply-To: <20070706071853.9434deae.kamezawa.hiroyu@jp.fujitsu.com> References: <20070704150504.423f6c54.kamezawa.hiroyu@jp.fujitsu.com> <20070705181308.GB8320@stroyan.net> <20070706071853.9434deae.kamezawa.hiroyu@jp.fujitsu.com> Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-linux-mm@kvack.org Return-Path: To: KAMEZAWA Hiroyuki Cc: Mike Stroyan , linux-ia64@vger.kernel.org, linux-kernel@vger.kernel.org, tony.luck@intel.com, linux-mm@kvack.org, clameter@sgi.com, y-goto@jp.fujitsu.com, dmosberger@gmail.com, hugh@veritas.com, nickpiggin@yahoo.com.au List-ID: On Fri, 6 Jul 2007 07:18:53 +0900 KAMEZAWA Hiroyuki wrote: > On Thu, 5 Jul 2007 12:13:09 -0600 > Mike Stroyan wrote: > > The L3 cache is involved in the HP-UX defect description because the > > earlier HP-UX patch PHKL_33781 added flushing of the instruction cache > > when an executable mapping was removed. Linux never added that > > unsuccessfull attempt at montecito cache coherency. In the current > > linux situation it can execute old cache lines straight from L2 icache. > > > Hmm... I couldn't understand "why icache includes old lines in a new page." > This happens at > - a file is newly loaded into page-cache. > - only on NFS. > - happens very *often* if the program is unlucky. > > So I wrote my understainding as I think. > I'll remove reference to HP-UX in the next post. And rewrite all description. > > > > The only defect that I see in the current implementation of > > lazy_mmu_prot_update() is that it is called too late in some > > functions that are already calling it. Are your large changes > > attempting to correct other defects? Or are you simplifying > > away potentially valuable code because you don't understand it? > > > I know your *simple* patch in April wasn't included. So I wrote this. > In April thread, commenter's advices was "implement flush_icache_page()" I think. > If you have a better patch, please post. > I'll check callers of lazy_mmu_prot_update() again and remove uncecessary calls. But, basically, i-cache flush will be necessary when VM_EXEC is on. PG_arch_1 will help us for optimization. -Kame -- 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