From nobody Wed Nov 5 10:26:00 2025 Delivered-To: importer@patchew.org Received-SPF: pass (zoho.com: domain of gnu.org designates 208.118.235.17 as permitted sender) client-ip=208.118.235.17; envelope-from=qemu-devel-bounces+importer=patchew.org@nongnu.org; helo=lists.gnu.org; Authentication-Results: mx.zohomail.com; spf=pass (zoho.com: domain of gnu.org designates 208.118.235.17 as permitted sender) smtp.mailfrom=qemu-devel-bounces+importer=patchew.org@nongnu.org; dmarc=fail(p=none dis=none) header.from=virtuozzo.com Return-Path: Received: from lists.gnu.org (lists.gnu.org [208.118.235.17]) by mx.zohomail.com with SMTPS id 1533664146420427.17646302380183; Tue, 7 Aug 2018 10:49:06 -0700 (PDT) Received: from localhost ([::1]:40244 helo=lists.gnu.org) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1fn66T-0003Wc-7G for importer@patchew.org; Tue, 07 Aug 2018 13:49:05 -0400 Received: from eggs.gnu.org ([2001:4830:134:3::10]:42939) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1fn60u-0007Tk-6Q for qemu-devel@nongnu.org; Tue, 07 Aug 2018 13:43:22 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1fn60r-00021P-50 for qemu-devel@nongnu.org; Tue, 07 Aug 2018 13:43:20 -0400 Received: from relay.sw.ru ([185.231.240.75]:45000) by eggs.gnu.org with esmtps (TLS1.0:DHE_RSA_AES_256_CBC_SHA1:32) (Exim 4.71) (envelope-from ) id 1fn60q-0001xV-Ml; Tue, 07 Aug 2018 13:43:17 -0400 Received: from vz-out.virtuozzo.com ([185.231.240.5] helo=kvm.sw.ru) by relay.sw.ru with esmtp (Exim 4.90_1) (envelope-from ) id 1fn60m-0003lt-PI; Tue, 07 Aug 2018 20:43:12 +0300 From: Vladimir Sementsov-Ogievskiy To: qemu-devel@nongnu.org, qemu-block@nongnu.org Date: Tue, 7 Aug 2018 20:43:08 +0300 Message-Id: <20180807174311.32454-5-vsementsov@virtuozzo.com> X-Mailer: git-send-email 2.11.1 In-Reply-To: <20180807174311.32454-1-vsementsov@virtuozzo.com> References: <20180807174311.32454-1-vsementsov@virtuozzo.com> X-detected-operating-system: by eggs.gnu.org: GNU/Linux 3.x [fuzzy] X-Received-From: 185.231.240.75 Subject: [Qemu-devel] [PATCH 4/7] qcow2: async scheme for qcow2_co_preadv X-BeenThere: qemu-devel@nongnu.org X-Mailman-Version: 2.1.21 Precedence: list List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Cc: kwolf@redhat.com, den@openvz.org, vsementsov@virtuozzo.com, mreitz@redhat.com Errors-To: qemu-devel-bounces+importer=patchew.org@nongnu.org Sender: "Qemu-devel" X-ZohoMail: RDMRC_1 RSF_0 Z_629925259 SPT_0 Content-Transfer-Encoding: quoted-printable MIME-Version: 1.0 Content-Type: text/plain; charset="utf-8" Start several async requests instead of read chunk by chunk. Signed-off-by: Vladimir Sementsov-Ogievskiy --- block/qcow2.c | 208 ++++++++++++++++++++++++++++++++++++++++++++++++++++++= ++-- 1 file changed, 204 insertions(+), 4 deletions(-) diff --git a/block/qcow2.c b/block/qcow2.c index 5e7f2ee318..a0df8d4e50 100644 --- a/block/qcow2.c +++ b/block/qcow2.c @@ -1869,6 +1869,197 @@ out: return ret; } =20 +typedef struct Qcow2WorkerTask { + uint64_t file_cluster_offset; + uint64_t offset; + uint64_t bytes; + uint64_t bytes_done; +} Qcow2WorkerTask; + +typedef int (*Qcow2DoWorkFunc)(BlockDriverState *bs, QEMUIOVector *qiov, + Qcow2WorkerTask *task); + +typedef struct Qcow2RWState { + BlockDriverState *bs; + QEMUIOVector *qiov; + uint64_t bytes; + int ret; + bool waiting_one; + bool waiting_all; + bool finalize; + Coroutine *co; + QSIMPLEQ_HEAD(, Qcow2Worker) free_workers; + QSIMPLEQ_HEAD(, Qcow2Worker) busy_workers; + int online_workers; + Qcow2DoWorkFunc do_work_func; +} Qcow2RWState; + +typedef struct Qcow2Worker { + Qcow2RWState *rws; + Coroutine *co; + Qcow2WorkerTask task; + bool busy; + QSIMPLEQ_ENTRY(Qcow2Worker) entry; +} Qcow2Worker; +#define QCOW2_MAX_WORKERS 64 + +static coroutine_fn void qcow2_rw_worker(void *opaque); +static Qcow2Worker *qcow2_new_worker(Qcow2RWState *rws) +{ + Qcow2Worker *w =3D g_new0(Qcow2Worker, 1); + w->rws =3D rws; + w->co =3D qemu_coroutine_create(qcow2_rw_worker, w); + + return w; +} + +static void qcow2_free_worker(Qcow2Worker *w) +{ + g_free(w); +} + +static coroutine_fn void qcow2_rw_worker(void *opaque) +{ + Qcow2Worker *w =3D opaque; + Qcow2RWState *rws =3D w->rws; + + rws->online_workers++; + + while (!rws->finalize) { + int ret =3D rws->do_work_func(rws->bs, rws->qiov, &w->task); + if (ret < 0 && rws->ret =3D=3D 0) { + rws->ret =3D ret; + } + + if (rws->waiting_all || rws->ret < 0) { + break; + } + + w->busy =3D false; + QSIMPLEQ_REMOVE(&rws->busy_workers, w, Qcow2Worker, entry); + QSIMPLEQ_INSERT_TAIL(&rws->free_workers, w, entry); + if (rws->waiting_one) { + rws->waiting_one =3D false; + /* we must unset it here, to prevent queuing rws->co in several + * workers (it may happen if other worker already waits us on = mutex, + * so it will be entered after our yield and before rws->co en= ter) + * + * TODO: rethink this comment, as here (and in other places in= the + * file) we moved from qemu_coroutine_add_next to aio_co_wake. + */ + aio_co_wake(rws->co); + } + + qemu_coroutine_yield(); + } + + if (w->busy) { + w->busy =3D false; + QSIMPLEQ_REMOVE(&rws->busy_workers, w, Qcow2Worker, entry); + } + qcow2_free_worker(w); + rws->online_workers--; + + if (rws->waiting_all && rws->online_workers =3D=3D 0) { + aio_co_wake(rws->co); + } +} + +static coroutine_fn void qcow2_rws_add_task(Qcow2RWState *rws, + uint64_t file_cluster_offset, + uint64_t offset, + uint64_t bytes, + uint64_t bytes_done) +{ + Qcow2Worker *w; + + assert(rws->co =3D=3D qemu_coroutine_self()); + + if (bytes_done =3D=3D 0 && bytes =3D=3D rws->bytes) { + Qcow2WorkerTask task =3D { + .file_cluster_offset =3D file_cluster_offset, + .offset =3D offset, + .bytes =3D bytes, + .bytes_done =3D bytes_done + }; + rws->ret =3D rws->do_work_func(rws->bs, rws->qiov, &task); + return; + } + + if (!QSIMPLEQ_EMPTY(&rws->free_workers)) { + w =3D QSIMPLEQ_FIRST(&rws->free_workers); + QSIMPLEQ_REMOVE_HEAD(&rws->free_workers, entry); + } else if (rws->online_workers < QCOW2_MAX_WORKERS) { + w =3D qcow2_new_worker(rws); + } else { + rws->waiting_one =3D true; + qemu_coroutine_yield(); + assert(!rws->waiting_one); /* already unset by worker */ + + w =3D QSIMPLEQ_FIRST(&rws->free_workers); + QSIMPLEQ_REMOVE_HEAD(&rws->free_workers, entry); + } + w->busy =3D true; + QSIMPLEQ_INSERT_TAIL(&rws->busy_workers, w, entry); + + w->task.file_cluster_offset =3D file_cluster_offset; + w->task.offset =3D offset; + w->task.bytes =3D bytes; + w->task.bytes_done =3D bytes_done; + + qemu_coroutine_enter(w->co); +} + +static void qcow2_init_rws(Qcow2RWState *rws, BlockDriverState *bs, + QEMUIOVector *qiov, uint64_t bytes, + Qcow2DoWorkFunc do_work_func) +{ + memset(rws, 0, sizeof(*rws)); + rws->bs =3D bs; + rws->qiov =3D qiov; + rws->bytes =3D bytes; + rws->co =3D qemu_coroutine_self(); + rws->do_work_func =3D do_work_func; + QSIMPLEQ_INIT(&rws->free_workers); + QSIMPLEQ_INIT(&rws->busy_workers); +} + +static void qcow2_finalize_rws(Qcow2RWState *rws) +{ + assert(rws->co =3D=3D qemu_coroutine_self()); + + /* kill waiting workers */ + rws->finalize =3D true; + while (!QSIMPLEQ_EMPTY(&rws->free_workers)) { + Qcow2Worker *w =3D QSIMPLEQ_FIRST(&rws->free_workers); + QSIMPLEQ_REMOVE_HEAD(&rws->free_workers, entry); + qemu_coroutine_enter(w->co); + } + + /* wait others */ + if (rws->online_workers > 0) { + rws->waiting_all =3D true; + qemu_coroutine_yield(); + rws->waiting_all =3D false; + } + + assert(rws->online_workers =3D=3D 0); + assert(QSIMPLEQ_EMPTY(&rws->free_workers)); + assert(QSIMPLEQ_EMPTY(&rws->busy_workers)); +} + +static coroutine_fn int qcow2_co_preadv_normal_task(BlockDriverState *bs, + QEMUIOVector *qiov, + Qcow2WorkerTask *task) +{ + return qcow2_co_preadv_normal(bs, + task->file_cluster_offset, + task->offset, + task->bytes, + qiov, + task->bytes_done); +} + static coroutine_fn int qcow2_co_preadv(BlockDriverState *bs, uint64_t off= set, uint64_t bytes, QEMUIOVector *qiov, int flags) @@ -1880,12 +2071,15 @@ static coroutine_fn int qcow2_co_preadv(BlockDriver= State *bs, uint64_t offset, uint64_t cluster_offset =3D 0; uint64_t bytes_done =3D 0; QEMUIOVector hd_qiov; + Qcow2RWState rws =3D {0}; + + qcow2_init_rws(&rws, bs, qiov, bytes, qcow2_co_preadv_normal_task); =20 qemu_iovec_init(&hd_qiov, qiov->niov); =20 qemu_co_mutex_lock(&s->lock); =20 - while (bytes !=3D 0) { + while (bytes !=3D 0 && rws.ret =3D=3D 0) { =20 /* prepare next request */ cur_bytes =3D MIN(bytes, INT_MAX); @@ -1942,9 +2136,10 @@ static coroutine_fn int qcow2_co_preadv(BlockDriverS= tate *bs, uint64_t offset, case QCOW2_CLUSTER_NORMAL: qemu_co_mutex_unlock(&s->lock); =20 - ret =3D qcow2_co_preadv_normal(bs, cluster_offset, - offset, cur_bytes, qiov, bytes_do= ne); - if (ret < 0) { + qcow2_rws_add_task(&rws, cluster_offset, offset, cur_bytes, + bytes_done); + if (rws.ret < 0) { + ret =3D rws.ret; goto fail_nolock; } =20 @@ -1967,6 +2162,11 @@ fail: qemu_co_mutex_unlock(&s->lock); =20 fail_nolock: + qcow2_finalize_rws(&rws); + if (ret =3D=3D 0) { + ret =3D rws.ret; + } + qemu_iovec_destroy(&hd_qiov); =20 return ret; --=20 2.11.1