From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from mailman by lists.gnu.org with tmda-scanned (Exim 4.43) id 1L9QB0-00038W-Oc for qemu-devel@nongnu.org; Sun, 07 Dec 2008 15:24:26 -0500 Received: from exim by lists.gnu.org with spam-scanned (Exim 4.43) id 1L9QAx-00035Y-9c for qemu-devel@nongnu.org; Sun, 07 Dec 2008 15:24:26 -0500 Received: from [199.232.76.173] (port=52055 helo=monty-python.gnu.org) by lists.gnu.org with esmtp (Exim 4.43) id 1L9QAw-00035T-V8 for qemu-devel@nongnu.org; Sun, 07 Dec 2008 15:24:23 -0500 Received: from bart.se.axis.com ([195.60.68.10]:38144) by monty-python.gnu.org with esmtp (Exim 4.60) (envelope-from ) id 1L9QAw-0001Iv-LY for qemu-devel@nongnu.org; Sun, 07 Dec 2008 15:24:22 -0500 Received: from bart.se.axis.com (bart.se.axis.com [127.0.0.1]) by bart.se.axis.com (Postfix) with ESMTP id DDCC36425C for ; Sun, 7 Dec 2008 21:24:21 +0100 (CET) Received: from axis.com (edgar.se.axis.com [10.93.151.1]) by bart.se.axis.com (Postfix) with ESMTP id C5D6B64221 for ; Sun, 7 Dec 2008 21:24:21 +0100 (CET) Date: Sun, 7 Dec 2008 21:24:21 +0100 From: "Edgar E. Iglesias" Subject: Re: [Qemu-devel] MMIO address changes Message-ID: <20081207202421.GF1167@edgar.se.axis.com> References: <200812011859.35859.paul@codesourcery.com> <20081203121745.GI31803@edgar.se.axis.com> <200812031403.54060.paul@codesourcery.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <200812031403.54060.paul@codesourcery.com> Reply-To: qemu-devel@nongnu.org List-Id: qemu-devel.nongnu.org List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: qemu-devel@nongnu.org Cc: "Edgar E. Iglesias" , pbrook@codesourcery.com On Wed, Dec 03, 2008 at 02:03:53PM +0000, Paul Brook wrote: > > > I've tried to be fairly thorough with the changes, and tested what I can. > > > However it's possible I missed or broke something, so please test your > > > favourite targets. > > > > FWIW the etrax-fs machine works OK although I am seeing a noticeable > > slow-down after the patch. > > Strange. I wouldn't expect significant slowdown. > AFAIK the hot-path code should be pretty much unchanged (if anything faster > because the IO handlers don't need to adjust the address). Even the TLB fill > handler overhead (tlb_set_page_exec) should be negligible. Hello Paul, I went through your patch and AFAICT it looks fine. The only thing I saw that may cause slowdowns was the size increase of PageDesc and subpage. In case anybody want's to reproduce this, just start the etrax image I've posted before, login trough telnet and do a ls -alR /, you'll see a very noticeable difference. Best regards Edgar