From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([2001:4830:134:3::10]:45903) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1bxESw-0007Qk-Th for qemu-devel@nongnu.org; Thu, 20 Oct 2016 10:37:09 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1bxESt-0007gl-PY for qemu-devel@nongnu.org; Thu, 20 Oct 2016 10:37:06 -0400 Date: Thu, 20 Oct 2016 10:37:01 -0400 (EDT) From: Paolo Bonzini Message-ID: <1553719312.5624034.1476974221561.JavaMail.zimbra@redhat.com> In-Reply-To: <1476971169-31604-1-git-send-email-kwolf@redhat.com> References: <1476971169-31604-1-git-send-email-kwolf@redhat.com> MIME-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: 7bit Subject: Re: [Qemu-devel] [PATCH 0/9] block-backend: Use coroutine for flush/discard/ioctl List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: Kevin Wolf Cc: qemu-block@nongnu.org, mreitz@redhat.com, qemu-devel@nongnu.org ----- Original Message ----- > From: "Kevin Wolf" > To: qemu-block@nongnu.org > Cc: kwolf@redhat.com, mreitz@redhat.com, pbonzini@redhat.com, qemu-devel@nongnu.org > Sent: Thursday, October 20, 2016 3:46:00 PM > Subject: [PATCH 0/9] block-backend: Use coroutine for flush/discard/ioctl > > Paolo, this is my attempt at implementing what you were asking for last > Friday. > I converted blk_(co_)flush/pdiscard/ioctl so that all interfaces (coroutine, > AIO, sync) go through the same coroutine-based function already on the > BlockBackend level. Where it was reasonably easy, I also removed the > corresponding emulations from block/io.c IIUC, this should cover your > immediate > needs. > > > Function to remove this series leaves for another day: > > * bdrv_aio_flush (used by blkdebug, blkverify, qed) > * bdrv_flush (even more users) > * bdrv_pdiscard (used by qcow2) > > > BlockDriver callbacks to remove left for another day: > > * bdrv_aio_pdiscard (implemented by raw-posix and rbd) > * bdrv_aio_ioctl (implemented by raw-posix and iscsi) > > In both cases, raw-posix is trivial to covert, but iscsi and rbd feel rather > scary without a proper test setup. Thanks! Paolo