From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from mailman by lists.gnu.org with archive (Exim 4.43) id 1Llp5J-0000Yo-B6 for mharc-grub-devel@gnu.org; Mon, 23 Mar 2009 14:41:17 -0400 Received: from mailman by lists.gnu.org with tmda-scanned (Exim 4.43) id 1Llp5G-0000Y3-Su for grub-devel@gnu.org; Mon, 23 Mar 2009 14:41:14 -0400 Received: from exim by lists.gnu.org with spam-scanned (Exim 4.43) id 1Llp5C-0000Vq-3I for grub-devel@gnu.org; Mon, 23 Mar 2009 14:41:14 -0400 Received: from [199.232.76.173] (port=38260 helo=monty-python.gnu.org) by lists.gnu.org with esmtp (Exim 4.43) id 1Llp5B-0000Vl-Oa for grub-devel@gnu.org; Mon, 23 Mar 2009 14:41:09 -0400 Received: from mta-out.inet.fi ([195.156.147.13]:44037 helo=kirsi1.inet.fi) by monty-python.gnu.org with esmtp (Exim 4.60) (envelope-from ) id 1Llp5B-0001gq-4j for grub-devel@gnu.org; Mon, 23 Mar 2009 14:41:09 -0400 Received: from [192.168.1.102] (84.248.105.254) by kirsi1.inet.fi (8.5.014) id 49B910AD006D5E0E for grub-devel@gnu.org; Mon, 23 Mar 2009 20:41:08 +0200 Message-ID: <49C7D7B9.2020408@nic.fi> Date: Mon, 23 Mar 2009 20:40:57 +0200 From: =?ISO-8859-1?Q?Vesa_J=E4=E4skel=E4inen?= User-Agent: Thunderbird 2.0.0.21 (Windows/20090302) MIME-Version: 1.0 To: The development of GRUB 2 References: <49C7A64F.30004@gmail.com> <49C7B592.9090307@nic.fi> <49C7BA85.30604@gmail.com> In-Reply-To: <49C7BA85.30604@gmail.com> X-Enigmail-Version: 0.95.7 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit X-detected-operating-system: by monty-python.gnu.org: GNU/Linux 2.6 (newer, 3) Subject: Re: [Patch] Move normal.mod to conf/common.rmk X-BeenThere: grub-devel@gnu.org X-Mailman-Version: 2.1.5 Precedence: list Reply-To: The development of GRUB 2 List-Id: The development of GRUB 2 List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 23 Mar 2009 18:41:15 -0000 phcoder wrote: > Actually if it was up to me I would have removed setjmp altogether. IMHO > it's a bad programming technique. It's used only to launch rescue mode. > I would just call the rescue mode interpreter by function. Perhaps Bean > who is working on normal.mod splitting has even better idea. > Also IMO normal.mod should have no arch-dependent parts and setjmp.S is > an unfortunate example. If some platforms need a special service then it > could be provided by drivers and/or commands. I see no reason why > additional cpu or platform-dependent would appear in normal.mod setjmp > could even be put in a separate module but it's quite a waste because > ELF overhead is much bigger then the module itself. Another possibility > is ot compile it like Or.. one could move it to kernel. But that is not as such a kernel component. More like a system library functionality. Actually setjmp might be used in bitmap loaders.