From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from [140.186.70.92] (port=47164 helo=eggs.gnu.org) by lists.gnu.org with esmtp (Exim 4.43) id 1P6rR1-0004XH-El for qemu-devel@nongnu.org; Fri, 15 Oct 2010 17:03:29 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1P6rQx-0001TN-3a for qemu-devel@nongnu.org; Fri, 15 Oct 2010 17:03:24 -0400 Received: from moutng.kundenserver.de ([212.227.17.9]:62992) by eggs.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1P6rQw-0001TB-OW for qemu-devel@nongnu.org; Fri, 15 Oct 2010 17:03:23 -0400 Message-ID: <4CB8C196.8030403@mail.berlios.de> Date: Fri, 15 Oct 2010 23:03:18 +0200 From: Stefan Weil MIME-Version: 1.0 References: <4CB48A51.8040508@mail.berlios.de> In-Reply-To: Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Subject: [Qemu-devel] Re: Where's gpxe-eepro100-80862449.rom ? List-Id: qemu-devel.nongnu.org List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: Markus Armbruster Cc: qemu-devel@nongnu.org Am 13.10.2010 09:13, schrieb Markus Armbruster: > Stefan Weil writes: > > >> Am 12.10.2010 14:41, schrieb Markus Armbruster: >> >>> Commit db667a12 added a reference to ROM file >>> gpxe-eepro100-80862449.rom, but no such file. Intentional? >>> >> Yes. See >> http://lists.nongnu.org/archive/html/qemu-devel/2010-05/msg00418.html >> >> Citation from my former mail: >> >> Yes, it is missing. I did not add it for several reasons: >> >> * The new variant i82801 (which needs the missing ROM) is currently >> not available >> via model=i82801 (it can be selected by experts using other methods). >> > It's available via -device i82801. -net nic,model= is obsolescent. > > >> * Experts can get the missing ROM from ROM-o-matic. >> >> * All three ROMs differ only in one or two bytes (device id). That's >> a waste >> of binary space (56832 bytes!). Therefore I plan to remove all but >> the first >> ROM file. The missing ones can be created by make or patched on the >> fly by >> QEMU during load. Maybe it is also possible to create ROM files >> which support >> more than one vendor/device entry - that would be the best solution. >> > No objections to creating by make or on load, but I'd be just fine with > the duplicated ROMs. Disk space waste in the order of 100KiB is lost in > the noise nowadays. My qemu/.git is>100MiB, and my build tree for all > targets ~1.5GiB. > > >> There is still nothing new regarding these last points >> because I did not spend any time on them. >> >> Do you think there is urgent need for a gpxe-eepro100-80862449.rom binary? >> > Well, "-device i82801" complains because it misses this binary. Do we > want to ship that way? > I just sent two patches which create the rom data on load. So -device i82801 no longer complains but gets the boot information from dhcp (tested with i386-softmmu). Your build tree will be made smaller by at least 50 KB :-) Do you think these patches should be added to stable-0.13, too? Regards, Stefan