From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([2001:4830:134:3::10]:32949) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1VbR9F-0003zV-7w for qemu-devel@nongnu.org; Wed, 30 Oct 2013 04:29:11 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1VbR99-0002Kv-7Z for qemu-devel@nongnu.org; Wed, 30 Oct 2013 04:29:05 -0400 Received: from mx1.redhat.com ([209.132.183.28]:29948) by eggs.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1VbR98-0002Ko-V4 for qemu-devel@nongnu.org; Wed, 30 Oct 2013 04:28:59 -0400 Date: Wed, 30 Oct 2013 09:28:46 +0100 From: Stefan Hajnoczi Message-ID: <20131030082846.GC10847@stefanha-thinkpad.redhat.com> References: <1381233491-17019-1-git-send-email-pl@kamp.de> <1381233491-17019-15-git-send-email-pl@kamp.de> <20131018123812.GB19041@stefanha-thinkpad.redhat.com> <52612E47.9050004@redhat.com> <52613709.6040604@kamp.de> <52613C8D.5060708@redhat.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: Subject: Re: [Qemu-devel] [PATCHv4 14/17] block/get_block_status: fix BDRV_BLOCK_ZERO for unallocated blocks List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: Peter Lieven Cc: "kwolf@redhat.com" , "anthony@codemonkey.ws" , Stefan Hajnoczi , "qemu-devel@nongnu.org" , "ronniesahlberg@gmail.com" , Paolo Bonzini On Fri, Oct 18, 2013 at 09:10:43PM +0200, Peter Lieven wrote: > > > > Am 18.10.2013 um 15:50 schrieb Paolo Bonzini : > > > > Il 18/10/2013 15:26, Peter Lieven ha scritto: > >>> > >>> > >>> - bdrv_discard_zeroes for bdrv_has_discard_write_zeroes > >> This would conform to the linux ioctl BLKDISCARDZEROES. > >> However, we need the write_zeroes operation for a guarantee > >> that zeroes are return. > > > > Yes. I'm fine with the current names actually, just thinking loudly. > > > >>> - bdrv_unallocated_blocks_are_zero for bdrv_has_discard_zeroes > >>> > >>> But I'm not sure why we have different BlockDriver APIs. I'd rather put > >>> the new flags in BlockDriverInfo, and make the new functions simple > >>> wrappers around bdrv_get_info. I think I proposed that before, maybe I > >>> wasn't clear or I was misunderstood. > >> I think Kevin wanted to have special functions for this. > > > > Yes, but I think he referred to block.c functions not BlockDriver functions. > > Ok, if Stefan and Kevin agree i will change it once more. I Would also like some Feedback on the new names for the functions and changed description. I can send a respin next week then. (Catching up with old mails) Fine here. Stefan