From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([208.118.235.92]:38103) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1UG7fO-00082Q-7m for qemu-devel@nongnu.org; Thu, 14 Mar 2013 08:53:55 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1UG7fM-0005IK-Ng for qemu-devel@nongnu.org; Thu, 14 Mar 2013 08:53:54 -0400 Received: from mx1.redhat.com ([209.132.183.28]:28757) by eggs.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1UG7fM-0005I2-Fq for qemu-devel@nongnu.org; Thu, 14 Mar 2013 08:53:52 -0400 Date: Thu, 14 Mar 2013 13:53:34 +0100 From: Kevin Wolf Message-ID: <20130314125334.GD2485@dhcp-200-207.str.redhat.com> References: <1363000996-13221-1-git-send-email-xiawenc@linux.vnet.ibm.com> <1363000996-13221-2-git-send-email-xiawenc@linux.vnet.ibm.com> <513E1934.4030908@redhat.com> <513EB6B3.9050609@linux.vnet.ibm.com> <20130313124128.GG2309@dhcp-200-207.str.redhat.com> <87li9r9n8f.fsf@blackfin.pond.sub.org> <20130314090141.GB2485@dhcp-200-207.str.redhat.com> <87mwu6w5a4.fsf@blackfin.pond.sub.org> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <87mwu6w5a4.fsf@blackfin.pond.sub.org> Subject: Re: [Qemu-devel] [PATCH V9 01/14] block: move bdrv_snapshot_find() to block/snapshot.c List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: Markus Armbruster Cc: qemu-devel@nongnu.org, stefanha@gmail.com, pbonzini@redhat.com, Wenchao Xia , lcapitulino@redhat.com Am 14.03.2013 um 13:10 hat Markus Armbruster geschrieben: > Kevin Wolf writes: > > But you have to do it right. This specific patch would introduce a > > copyright violation. It's really not that hard to conform to the terms > > of the MIT license, but that doesn't mean that you can ignore it. There > > is exactly one requirement and it reads like this: > > > > The above copyright notice and this permission notice shall be > > included in all copies or substantial portions of the Software. > > That's why I pointed to resources and examples on how to do it properly. > > > (I'm still waiting for a patch to blockdev.c, for which you did it > > wrong, by the way) > > Oops, that one fell through the cracks. Patch coming. Thanks. > >> Of course, the stronger license still has to be compatible with GPLv2, > >> so we can accept the result into QEMU. > >> > >> If a subsystem has additional requirements on licenses, its maintainers > >> will explain them to you. For what it's worth, substantial parts of the > >> block layer are already GPLv2+. > > > > What parts exactly? As long as there are plans for a libqblock and as > > long as it doesn't seem completely impossible to have it under LGPL, I > > will ask to use either MIT or LGPL for block layer code (this doesn't > > apply to qemu-only code that isn't used in the tools - in this sense, > > things like blockdev.c are not part of the block layer) > > $ git-grep -lw GPL block block* > block-migration.c > block/blkverify.c > block/gluster.c > block/linux-aio.c > block/raw-aio.h > block/rbd.c > block/sheepdog.c > blockdev-nbd.c > blockdev.c Luckily, none of these are really critical for a libqblock library, even though they would be nice to have. If we can't license such a library as LGPL, we would lose the most important potential user, which is libvirt. In which case I probably wouldn't want to bother with providing a library at all. Kevin