From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([2001:4830:134:3::10]:35121) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1dIwxe-0002MF-H2 for qemu-devel@nongnu.org; Thu, 08 Jun 2017 08:54:51 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1dIwxd-0001Y8-Lz for qemu-devel@nongnu.org; Thu, 08 Jun 2017 08:54:50 -0400 References: <20170531094330.1808-1-pbonzini@redhat.com> <20170608125231.GJ4145@stefanha-x1.localdomain> From: Paolo Bonzini Message-ID: Date: Thu, 8 Jun 2017 14:54:38 +0200 MIME-Version: 1.0 In-Reply-To: <20170608125231.GJ4145@stefanha-x1.localdomain> Content-Type: text/plain; charset=windows-1252 Content-Language: en-US Content-Transfer-Encoding: quoted-printable Subject: Re: [Qemu-devel] [RFC PATCH 00/11] Block layer thread-safety, part 2 List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: Stefan Hajnoczi Cc: qemu-devel@nongnu.org, kwolf@redhat.com, famz@redhat.com, qemu-block@nongnu.org, mreitz@redhat.com On 08/06/2017 14:52, Stefan Hajnoczi wrote: > On Wed, May 31, 2017 at 11:43:19AM +0200, Paolo Bonzini wrote: >> This part takes care of drivers and devices, making sure that they can >> accept concurrent I/O from multiple AioContext. Only RFC for now beca= use it >> depends on Kevin's QED conversion to coroutines, which is still in flu= x. >> >> The following drivers are thread-safe without using any QemuMutex/CoMu= tex: >> crypto, gluster, null, rbd, win32-aio. >> >> The following drivers already use mutexes for everything except possib= ly >> snapshots, which do not (yet?) need protection: bochs, cloop, dmg, qco= w, >> parallels, vhdx, vmdk, curl, iscsi, nfs. >> >> The following drivers already use mutexes for _almost_ everything: vpc >> (missing get_block_status), vdi (missing bitmap access), vvfat (missin= g >> commit), nbd (s->in_flight not protected), qcow2 (must call CoQueue AP= Is >> under CoMutex). They are fixed by patches 1-6. >> >> The following drivers must be changed to use CoMutex to protect intern= al >> data: qed (patches 7-9), sheepdog (patch 10). >> >> The following driver must be changed to support I/O from any AioContex= t: >> ssh. It is fixed by patch 11. >> >> Paolo >> >> Paolo Bonzini (11): >> qcow2: call CoQueue APIs under CoMutex >> coroutine-lock: add qemu_co_rwlock_downgrade and qemu_co_rwlock_upgr= ade >> vdi: make it thread-safe >> vpc: make it thread-safe >> vvfat: make it thread-safe >> nbd: make it thread-safe >> qed: move tail of qed_aio_write_main to qed_aio_write_{cow,alloc} >> block: invoke .bdrv_drain callback in coroutine context and from Aio= Context >> qed: protect table cache with CoMutex >> sheepdog: add queue_lock >> ssh: support I/O from any AioContext >> >> block/io.c | 42 +++++++--- >> block/nbd-client.c | 30 +++---- >> block/qcow2.c | 4 +- >> block/qed-cluster.c | 4 +- >> block/qed-l2-cache.c | 6 ++ >> block/qed-table.c | 24 ++++-- >> block/qed.c | 195 +++++++++++++++++++++++++++---------= --------- >> block/qed.h | 11 ++- >> block/sheepdog.c | 21 ++++- >> block/ssh.c | 24 ++++-- >> block/vdi.c | 48 +++++------ >> block/vpc.c | 20 ++--- >> block/vvfat.c | 8 +- >> include/block/block_int.h | 2 +- >> include/qemu/coroutine.h | 19 +++++ >> util/qemu-coroutine-lock.c | 35 ++++++++ >> 16 files changed, 326 insertions(+), 167 deletions(-) >> >> --=20 >> 2.13.0 >=20 > Reviewed-by: Stefan Hajnoczi Thanks, I'll repost once QED is converted to coroutines, and send part 3 when I have some time. Paolo