From patchwork Thu Jun 20 17:46:03 2013 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 8bit X-Patchwork-Submitter: =?utf-8?q?Marc-Andr=C3=A9_Lureau?= X-Patchwork-Id: 253048 Return-Path: X-Original-To: incoming@patchwork.ozlabs.org Delivered-To: patchwork-incoming@bilbo.ozlabs.org Received: from lists.gnu.org (lists.gnu.org [IPv6:2001:4830:134:3::11]) (using TLSv1 with cipher AES256-SHA (256/256 bits)) (Client did not present a certificate) by ozlabs.org (Postfix) with ESMTPS id 2E72C2C00A1 for ; Fri, 21 Jun 2013 03:58:00 +1000 (EST) Received: from localhost ([::1]:46502 helo=lists.gnu.org) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1Upj4K-0002hy-DL for incoming@patchwork.ozlabs.org; Thu, 20 Jun 2013 13:54:48 -0400 Received: from eggs.gnu.org ([2001:4830:134:3::10]:60551) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1Upj39-0000gf-1l for qemu-devel@nongnu.org; Thu, 20 Jun 2013 13:53:38 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1Upj36-0004Ml-4p for qemu-devel@nongnu.org; Thu, 20 Jun 2013 13:53:34 -0400 Received: from mail-wg0-x229.google.com ([2a00:1450:400c:c00::229]:58821) by eggs.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1UpiwR-0002CZ-6t for qemu-devel@nongnu.org; Thu, 20 Jun 2013 13:46:39 -0400 Received: by mail-wg0-f41.google.com with SMTP id y10so1105726wgg.0 for ; Thu, 20 Jun 2013 10:46:38 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113; h=from:to:cc:subject:date:message-id:x-mailer:in-reply-to:references :mime-version:content-type:content-transfer-encoding; bh=2zUnwEIw0O7EdGiGWblfv3LLTuy6jNHn3qVE8WyPlow=; b=ZdFWX28U/xpGBm5gP6R9VGkGL40sbQPUV89F5s7dNA/6bYNXF/GoFqmuKPnnDWEDVc i2p0XVRWui/0dg9+ZD/NqNxlVlbLOvzhyFFaagkhWmTVtNOm+/KkzsI7otdTjxQQKt5I by2cAODPrb6OwMNeqAsvPV8b7x7MaxjjNOxGJYWhFmThQslaIhCw32dRKNZqqs3Gn4Hv s4yGBq3zeoeAks9bz2ALX/mb2diO/88WA34IZmwSw/mgtnDIDtqemqkv33qEa6U2eLrG F1ZGXpySXV569vpybckK0MdaCPrXRV8aRbguz52vrTWDuQGZT/icDjXR9nUj5xrWAbRl yLcA== X-Received: by 10.194.243.226 with SMTP id xb2mr6702951wjc.67.1371750398436; Thu, 20 Jun 2013 10:46:38 -0700 (PDT) Received: from localhost (158.red-80-26-177.adsl.dynamic.ccgg.telefonica.net. [80.26.177.158]) by mx.google.com with ESMTPSA id x13sm2062181wib.3.2013.06.20.10.46.36 for (version=TLSv1.2 cipher=RC4-SHA bits=128/128); Thu, 20 Jun 2013 10:46:37 -0700 (PDT) From: "=?UTF-8?q?Marc-Andr=C3=A9=20Lureau?=" To: qemu-devel@nongnu.org Date: Thu, 20 Jun 2013 19:46:03 +0200 Message-Id: <1371750371-18491-5-git-send-email-marcandre.lureau@redhat.com> X-Mailer: git-send-email 1.8.3.rc1.49.g8d97506 In-Reply-To: <1371750371-18491-1-git-send-email-marcandre.lureau@redhat.com> References: <1371750371-18491-1-git-send-email-marcandre.lureau@redhat.com> MIME-Version: 1.0 X-detected-operating-system: by eggs.gnu.org: Error: Malformed IPv6 address (bad octet value). X-Received-From: 2a00:1450:400c:c00::229 Cc: spice-devel@lists.freedesktop.org, =?UTF-8?q?Marc-Andr=C3=A9=20Lureau?= Subject: [Qemu-devel] [PATCH 04/12] Split nbd block client code X-BeenThere: qemu-devel@nongnu.org X-Mailman-Version: 2.1.14 Precedence: list List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: qemu-devel-bounces+incoming=patchwork.ozlabs.org@nongnu.org Sender: qemu-devel-bounces+incoming=patchwork.ozlabs.org@nongnu.org Signed-off-by: Marc-André Lureau --- block/Makefile.objs | 2 +- block/nbd-client.c | 386 +++++++++++++++++++++++++++++++++++++++++++++++++++ block/nbd-client.h | 52 +++++++ block/nbd.c | 387 ++++------------------------------------------------ 4 files changed, 469 insertions(+), 358 deletions(-) create mode 100644 block/nbd-client.c create mode 100644 block/nbd-client.h diff --git a/block/Makefile.objs b/block/Makefile.objs index 2981654..5890b5c 100644 --- a/block/Makefile.objs +++ b/block/Makefile.objs @@ -10,7 +10,7 @@ block-obj-$(CONFIG_POSIX) += raw-posix.o block-obj-$(CONFIG_LINUX_AIO) += linux-aio.o ifeq ($(CONFIG_POSIX),y) -block-obj-y += nbd.o sheepdog.o +block-obj-y += nbd.o nbd-client.o sheepdog.o block-obj-$(CONFIG_LIBISCSI) += iscsi.o block-obj-$(CONFIG_CURL) += curl.o block-obj-$(CONFIG_RBD) += rbd.o diff --git a/block/nbd-client.c b/block/nbd-client.c new file mode 100644 index 0000000..6d5f39c --- /dev/null +++ b/block/nbd-client.c @@ -0,0 +1,386 @@ +/* + * QEMU Block driver for NBD + * + * Copyright (C) 2008 Bull S.A.S. + * Author: Laurent Vivier + * + * Some parts: + * Copyright (C) 2007 Anthony Liguori + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN + * THE SOFTWARE. + */ +#include "nbd-client.h" +#include "qemu/sockets.h" + +#define HANDLE_TO_INDEX(bs, handle) ((handle) ^ ((uint64_t)(intptr_t)bs)) +#define INDEX_TO_HANDLE(bs, index) ((index) ^ ((uint64_t)(intptr_t)bs)) + +static void nbd_reply_ready(void *opaque) +{ + NbdClientSession *s = opaque; + uint64_t i; + int ret; + + if (s->reply.handle == 0) { + /* No reply already in flight. Fetch a header. It is possible + * that another thread has done the same thing in parallel, so + * the socket is not readable anymore. + */ + ret = nbd_receive_reply(s->sock, &s->reply); + if (ret == -EAGAIN) { + return; + } + if (ret < 0) { + s->reply.handle = 0; + goto fail; + } + } + + /* There's no need for a mutex on the receive side, because the + * handler acts as a synchronization point and ensures that only + * one coroutine is called until the reply finishes. */ + i = HANDLE_TO_INDEX(s, s->reply.handle); + if (i >= MAX_NBD_REQUESTS) { + goto fail; + } + + if (s->recv_coroutine[i]) { + qemu_coroutine_enter(s->recv_coroutine[i], NULL); + return; + } + +fail: + for (i = 0; i < MAX_NBD_REQUESTS; i++) { + if (s->recv_coroutine[i]) { + qemu_coroutine_enter(s->recv_coroutine[i], NULL); + } + } +} + +static void nbd_restart_write(void *opaque) +{ + NbdClientSession *s = opaque; + + qemu_coroutine_enter(s->send_coroutine, NULL); +} + +static int nbd_have_request(void *opaque) +{ + NbdClientSession *s = opaque; + + return s->in_flight > 0; +} + +static int nbd_co_send_request(NbdClientSession *s, + struct nbd_request *request, + QEMUIOVector *qiov, int offset) +{ + int rc, ret; + + qemu_co_mutex_lock(&s->send_mutex); + s->send_coroutine = qemu_coroutine_self(); + qemu_aio_set_fd_handler(s->sock, nbd_reply_ready, nbd_restart_write, + nbd_have_request, s); + if (qiov) { + if (!s->is_unix) { + socket_set_cork(s->sock, 1); + } + rc = nbd_send_request(s->sock, request); + if (rc >= 0) { + ret = qemu_co_sendv(s->sock, qiov->iov, qiov->niov, + offset, request->len); + if (ret != request->len) { + rc = -EIO; + } + } + if (!s->is_unix) { + socket_set_cork(s->sock, 0); + } + } else { + rc = nbd_send_request(s->sock, request); + } + qemu_aio_set_fd_handler(s->sock, nbd_reply_ready, NULL, + nbd_have_request, s); + s->send_coroutine = NULL; + qemu_co_mutex_unlock(&s->send_mutex); + return rc; +} + +static void nbd_co_receive_reply(NbdClientSession *s, + struct nbd_request *request, struct nbd_reply *reply, + QEMUIOVector *qiov, int offset) +{ + int ret; + + /* Wait until we're woken up by the read handler. TODO: perhaps + * peek at the next reply and avoid yielding if it's ours? */ + qemu_coroutine_yield(); + *reply = s->reply; + if (reply->handle != request->handle) { + reply->error = EIO; + } else { + if (qiov && reply->error == 0) { + ret = qemu_co_recvv(s->sock, qiov->iov, qiov->niov, + offset, request->len); + if (ret != request->len) { + reply->error = EIO; + } + } + + /* Tell the read handler to read another header. */ + s->reply.handle = 0; + } +} + +static void nbd_coroutine_end(NbdClientSession *s, + struct nbd_request *request) +{ + int i = HANDLE_TO_INDEX(s, request->handle); + s->recv_coroutine[i] = NULL; + if (s->in_flight-- == MAX_NBD_REQUESTS) { + qemu_co_mutex_unlock(&s->free_sema); + } +} + +static void nbd_coroutine_start(NbdClientSession *s, + struct nbd_request *request) +{ + int i; + + /* Poor man semaphore. The free_sema is locked when no other request + * can be accepted, and unlocked after receiving one reply. */ + if (s->in_flight >= MAX_NBD_REQUESTS - 1) { + qemu_co_mutex_lock(&s->free_sema); + assert(s->in_flight < MAX_NBD_REQUESTS); + } + s->in_flight++; + + for (i = 0; i < MAX_NBD_REQUESTS; i++) { + if (s->recv_coroutine[i] == NULL) { + s->recv_coroutine[i] = qemu_coroutine_self(); + break; + } + } + + assert(i < MAX_NBD_REQUESTS); + request->handle = INDEX_TO_HANDLE(s, i); +} + +static int nbd_co_readv_1(NbdClientSession *client, int64_t sector_num, + int nb_sectors, QEMUIOVector *qiov, + int offset) +{ + struct nbd_request request; + struct nbd_reply reply; + ssize_t ret; + + request.type = NBD_CMD_READ; + request.from = sector_num * 512; + request.len = nb_sectors * 512; + + nbd_coroutine_start(client, &request); + ret = nbd_co_send_request(client, &request, NULL, 0); + if (ret < 0) { + reply.error = -ret; + } else { + nbd_co_receive_reply(client, &request, &reply, qiov, offset); + } + nbd_coroutine_end(client, &request); + + return -reply.error; + +} + +static int nbd_co_writev_1(NbdClientSession *client, int64_t sector_num, + int nb_sectors, QEMUIOVector *qiov, + int offset) +{ + struct nbd_request request; + struct nbd_reply reply; + ssize_t ret; + + request.type = NBD_CMD_WRITE; + if (!bdrv_enable_write_cache(client->bs) && + (client->nbdflags & NBD_FLAG_SEND_FUA)) { + request.type |= NBD_CMD_FLAG_FUA; + } + + request.from = sector_num * 512; + request.len = nb_sectors * 512; + + nbd_coroutine_start(client, &request); + ret = nbd_co_send_request(client, &request, qiov, offset); + if (ret < 0) { + reply.error = -ret; + } else { + nbd_co_receive_reply(client, &request, &reply, NULL, 0); + } + nbd_coroutine_end(client, &request); + + return -reply.error; +} + +/* qemu-nbd has a limit of slightly less than 1M per request. Try to + * remain aligned to 4K. */ +#define NBD_MAX_SECTORS 2040 + +int nbd_client_session_co_readv(NbdClientSession *client, int64_t sector_num, + int nb_sectors, QEMUIOVector *qiov) +{ + int offset = 0; + int ret; + + while (nb_sectors > NBD_MAX_SECTORS) { + ret = nbd_co_readv_1(client, sector_num, NBD_MAX_SECTORS, qiov, offset); + if (ret < 0) { + return ret; + } + offset += NBD_MAX_SECTORS * 512; + sector_num += NBD_MAX_SECTORS; + nb_sectors -= NBD_MAX_SECTORS; + } + + return nbd_co_readv_1(client, sector_num, nb_sectors, qiov, offset); +} + +int nbd_client_session_co_writev(NbdClientSession *client, int64_t sector_num, + int nb_sectors, QEMUIOVector *qiov) +{ + int offset = 0; + int ret; + + while (nb_sectors > NBD_MAX_SECTORS) { + ret = nbd_co_writev_1(client, sector_num, + NBD_MAX_SECTORS, qiov, offset); + if (ret < 0) { + return ret; + } + offset += NBD_MAX_SECTORS * 512; + sector_num += NBD_MAX_SECTORS; + nb_sectors -= NBD_MAX_SECTORS; + } + + return nbd_co_writev_1(client, sector_num, nb_sectors, qiov, offset); +} + +int nbd_client_session_co_flush(NbdClientSession *client) +{ + struct nbd_request request; + struct nbd_reply reply; + ssize_t ret; + + if (!(client->nbdflags & NBD_FLAG_SEND_FLUSH)) { + return 0; + } + + request.type = NBD_CMD_FLUSH; + if (client->nbdflags & NBD_FLAG_SEND_FUA) { + request.type |= NBD_CMD_FLAG_FUA; + } + + request.from = 0; + request.len = 0; + + nbd_coroutine_start(client, &request); + ret = nbd_co_send_request(client, &request, NULL, 0); + if (ret < 0) { + reply.error = -ret; + } else { + nbd_co_receive_reply(client, &request, &reply, NULL, 0); + } + nbd_coroutine_end(client, &request); + + return -reply.error; +} + +int nbd_client_session_co_discard(NbdClientSession *client, int64_t sector_num, + int nb_sectors) +{ + struct nbd_request request; + struct nbd_reply reply; + ssize_t ret; + + if (!(client->nbdflags & NBD_FLAG_SEND_TRIM)) { + return 0; + } + + request.type = NBD_CMD_TRIM; + request.from = sector_num * 512; + request.len = nb_sectors * 512; + + nbd_coroutine_start(client, &request); + ret = nbd_co_send_request(client, &request, NULL, 0); + if (ret < 0) { + reply.error = -ret; + } else { + nbd_co_receive_reply(client, &request, &reply, NULL, 0); + } + nbd_coroutine_end(client, &request); + + return -reply.error; +} + +static void nbd_teardown_connection(NbdClientSession *client) +{ + struct nbd_request request; + + request.type = NBD_CMD_DISC; + request.from = 0; + request.len = 0; + nbd_send_request(client->sock, &request); + + qemu_aio_set_fd_handler(client->sock, NULL, NULL, NULL, NULL); + closesocket(client->sock); + client->sock = -1; +} + +void nbd_client_session_close(NbdClientSession *client) +{ + nbd_teardown_connection(client); + g_free(client->export_name); +} + +int nbd_client_session_init(NbdClientSession *client, + BlockDriverState *bs, int sock) +{ + int ret; + + ret = nbd_receive_negotiate(sock, client->export_name, + &client->nbdflags, &client->size, + &client->blocksize); + if (ret < 0) { + logout("Failed to negotiate with the NBD server\n"); + closesocket(sock); + return ret; + } + + qemu_co_mutex_init(&client->send_mutex); + qemu_co_mutex_init(&client->free_sema); + client->bs = bs; + client->sock = sock; + + /* Now that we're connected, set the socket to be non-blocking and + * kick the reply mechanism. */ + qemu_set_nonblock(client->sock); + qemu_aio_set_fd_handler(client->sock, nbd_reply_ready, NULL, + nbd_have_request, client); + + logout("Established connection with NBD server\n"); + return 0; +} diff --git a/block/nbd-client.h b/block/nbd-client.h new file mode 100644 index 0000000..c1a7871 --- /dev/null +++ b/block/nbd-client.h @@ -0,0 +1,52 @@ +#ifndef NBD_CLIENT_H +#define NBD_CLIENT_H + +#include "qemu-common.h" +#include "block/nbd.h" +#include "block/block_int.h" + +/* #define DEBUG_NBD */ + +#if defined(DEBUG_NBD) +#define logout(fmt, ...) \ + fprintf(stderr, "nbd\t%-24s" fmt, __func__, ##__VA_ARGS__) +#else +#define logout(fmt, ...) ((void)0) +#endif + +#define MAX_NBD_REQUESTS 16 + +typedef struct NbdClientSession { + CoMutex send_mutex; + CoMutex free_sema; + + int sock; + uint32_t nbdflags; + off_t size; + size_t blocksize; + + Coroutine *send_coroutine; + int in_flight; + + Coroutine *recv_coroutine[MAX_NBD_REQUESTS]; + struct nbd_reply reply; + + char *export_name; /* An NBD server may export several devices */ + bool is_unix; + + BlockDriverState *bs; +} NbdClientSession; + +int nbd_client_session_init(NbdClientSession *client, + BlockDriverState *bs, int sock); +void nbd_client_session_close(NbdClientSession *client); + +int nbd_client_session_co_discard(NbdClientSession *client, int64_t sector_num, + int nb_sectors); +int nbd_client_session_co_flush(NbdClientSession *client); +int nbd_client_session_co_writev(NbdClientSession *client, int64_t sector_num, + int nb_sectors, QEMUIOVector *qiov); +int nbd_client_session_co_readv(NbdClientSession *client, int64_t sector_num, + int nb_sectors, QEMUIOVector *qiov); + +#endif /* NBD_CLIENT_H */ diff --git a/block/nbd.c b/block/nbd.c index 30e3b78..79ba0a6 100644 --- a/block/nbd.c +++ b/block/nbd.c @@ -26,10 +26,8 @@ * THE SOFTWARE. */ -#include "qemu-common.h" -#include "block/nbd.h" +#include "nbd-client.h" #include "qemu/uri.h" -#include "block/block_int.h" #include "qemu/module.h" #include "qemu/sockets.h" #include "qapi/qmp/qjson.h" @@ -38,41 +36,14 @@ #include #include -#define EN_OPTSTR ":exportname=" - -/* #define DEBUG_NBD */ - -#if defined(DEBUG_NBD) -#define logout(fmt, ...) \ - fprintf(stderr, "nbd\t%-24s" fmt, __func__, ##__VA_ARGS__) -#else -#define logout(fmt, ...) ((void)0) -#endif - -#define MAX_NBD_REQUESTS 16 -#define HANDLE_TO_INDEX(bs, handle) ((handle) ^ ((uint64_t)(intptr_t)bs)) -#define INDEX_TO_HANDLE(bs, index) ((index) ^ ((uint64_t)(intptr_t)bs)) - typedef struct BDRVNBDState { - int sock; - uint32_t nbdflags; - off_t size; - size_t blocksize; - - CoMutex send_mutex; - CoMutex free_sema; - Coroutine *send_coroutine; - int in_flight; - - Coroutine *recv_coroutine[MAX_NBD_REQUESTS]; - struct nbd_reply reply; + NbdClientSession client; - bool is_unix; QemuOpts *socket_opts; - - char *export_name; /* An NBD server may export several devices */ } BDRVNBDState; +#define EN_OPTSTR ":exportname=" + static int nbd_parse_uri(const char *filename, QDict *options) { URI *uri; @@ -218,9 +189,9 @@ static int nbd_config(BDRVNBDState *s, QDict *options) "be used at the same time."); return -EINVAL; } - s->is_unix = true; + s->client.is_unix = true; } else if (qdict_haskey(options, "host")) { - s->is_unix = false; + s->client.is_unix = false; } else { return -EINVAL; } @@ -238,171 +209,20 @@ static int nbd_config(BDRVNBDState *s, QDict *options) qemu_opt_set_number(s->socket_opts, "port", NBD_DEFAULT_PORT); } - s->export_name = g_strdup(qdict_get_try_str(options, "export")); - if (s->export_name) { + s->client.export_name = g_strdup(qdict_get_try_str(options, "export")); + if (s->client.export_name) { qdict_del(options, "export"); } return 0; } - -static void nbd_coroutine_start(BDRVNBDState *s, struct nbd_request *request) -{ - int i; - - /* Poor man semaphore. The free_sema is locked when no other request - * can be accepted, and unlocked after receiving one reply. */ - if (s->in_flight >= MAX_NBD_REQUESTS - 1) { - qemu_co_mutex_lock(&s->free_sema); - assert(s->in_flight < MAX_NBD_REQUESTS); - } - s->in_flight++; - - for (i = 0; i < MAX_NBD_REQUESTS; i++) { - if (s->recv_coroutine[i] == NULL) { - s->recv_coroutine[i] = qemu_coroutine_self(); - break; - } - } - - assert(i < MAX_NBD_REQUESTS); - request->handle = INDEX_TO_HANDLE(s, i); -} - -static int nbd_have_request(void *opaque) -{ - BDRVNBDState *s = opaque; - - return s->in_flight > 0; -} - -static void nbd_reply_ready(void *opaque) -{ - BDRVNBDState *s = opaque; - uint64_t i; - int ret; - - if (s->reply.handle == 0) { - /* No reply already in flight. Fetch a header. It is possible - * that another thread has done the same thing in parallel, so - * the socket is not readable anymore. - */ - ret = nbd_receive_reply(s->sock, &s->reply); - if (ret == -EAGAIN) { - return; - } - if (ret < 0) { - s->reply.handle = 0; - goto fail; - } - } - - /* There's no need for a mutex on the receive side, because the - * handler acts as a synchronization point and ensures that only - * one coroutine is called until the reply finishes. */ - i = HANDLE_TO_INDEX(s, s->reply.handle); - if (i >= MAX_NBD_REQUESTS) { - goto fail; - } - - if (s->recv_coroutine[i]) { - qemu_coroutine_enter(s->recv_coroutine[i], NULL); - return; - } - -fail: - for (i = 0; i < MAX_NBD_REQUESTS; i++) { - if (s->recv_coroutine[i]) { - qemu_coroutine_enter(s->recv_coroutine[i], NULL); - } - } -} - -static void nbd_restart_write(void *opaque) -{ - BDRVNBDState *s = opaque; - qemu_coroutine_enter(s->send_coroutine, NULL); -} - -static int nbd_co_send_request(BDRVNBDState *s, struct nbd_request *request, - QEMUIOVector *qiov, int offset) -{ - int rc, ret; - - qemu_co_mutex_lock(&s->send_mutex); - s->send_coroutine = qemu_coroutine_self(); - qemu_aio_set_fd_handler(s->sock, nbd_reply_ready, nbd_restart_write, - nbd_have_request, s); - if (qiov) { - if (!s->is_unix) { - socket_set_cork(s->sock, 1); - } - rc = nbd_send_request(s->sock, request); - if (rc >= 0) { - ret = qemu_co_sendv(s->sock, qiov->iov, qiov->niov, - offset, request->len); - if (ret != request->len) { - rc = -EIO; - } - } - if (!s->is_unix) { - socket_set_cork(s->sock, 0); - } - } else { - rc = nbd_send_request(s->sock, request); - } - qemu_aio_set_fd_handler(s->sock, nbd_reply_ready, NULL, - nbd_have_request, s); - s->send_coroutine = NULL; - qemu_co_mutex_unlock(&s->send_mutex); - return rc; -} - -static void nbd_co_receive_reply(BDRVNBDState *s, struct nbd_request *request, - struct nbd_reply *reply, - QEMUIOVector *qiov, int offset) -{ - int ret; - - /* Wait until we're woken up by the read handler. TODO: perhaps - * peek at the next reply and avoid yielding if it's ours? */ - qemu_coroutine_yield(); - *reply = s->reply; - if (reply->handle != request->handle) { - reply->error = EIO; - } else { - if (qiov && reply->error == 0) { - ret = qemu_co_recvv(s->sock, qiov->iov, qiov->niov, - offset, request->len); - if (ret != request->len) { - reply->error = EIO; - } - } - - /* Tell the read handler to read another header. */ - s->reply.handle = 0; - } -} - -static void nbd_coroutine_end(BDRVNBDState *s, struct nbd_request *request) -{ - int i = HANDLE_TO_INDEX(s, request->handle); - s->recv_coroutine[i] = NULL; - if (s->in_flight-- == MAX_NBD_REQUESTS) { - qemu_co_mutex_unlock(&s->free_sema); - } -} - static int nbd_establish_connection(BlockDriverState *bs) { BDRVNBDState *s = bs->opaque; int sock; - int ret; - off_t size; - size_t blocksize; - if (s->is_unix) { + if (s->client.is_unix) { sock = unix_socket_outgoing(qemu_opt_get(s->socket_opts, "path")); } else { sock = tcp_socket_outgoing_opts(s->socket_opts); @@ -417,50 +237,13 @@ static int nbd_establish_connection(BlockDriverState *bs) return -errno; } - /* NBD handshake */ - ret = nbd_receive_negotiate(sock, s->export_name, &s->nbdflags, &size, - &blocksize); - if (ret < 0) { - logout("Failed to negotiate with the NBD server\n"); - closesocket(sock); - return ret; - } - - /* Now that we're connected, set the socket to be non-blocking and - * kick the reply mechanism. */ - qemu_set_nonblock(sock); - qemu_aio_set_fd_handler(sock, nbd_reply_ready, NULL, - nbd_have_request, s); - - s->sock = sock; - s->size = size; - s->blocksize = blocksize; - - logout("Established connection with NBD server\n"); - return 0; -} - -static void nbd_teardown_connection(BlockDriverState *bs) -{ - BDRVNBDState *s = bs->opaque; - struct nbd_request request; - - request.type = NBD_CMD_DISC; - request.from = 0; - request.len = 0; - nbd_send_request(s->sock, &request); - - qemu_aio_set_fd_handler(s->sock, NULL, NULL, NULL, NULL); - closesocket(s->sock); + return sock; } static int nbd_open(BlockDriverState *bs, QDict *options, int flags) { BDRVNBDState *s = bs->opaque; - int result; - - qemu_co_mutex_init(&s->send_mutex); - qemu_co_mutex_init(&s->free_sema); + int result, sock; /* Pop the config into our state object. Exit if invalid. */ result = nbd_config(s, options); @@ -471,172 +254,62 @@ static int nbd_open(BlockDriverState *bs, QDict *options, int flags) /* establish TCP connection, return error if it fails * TODO: Configurable retry-until-timeout behaviour. */ - result = nbd_establish_connection(bs); - - return result; -} - -static int nbd_co_readv_1(BlockDriverState *bs, int64_t sector_num, - int nb_sectors, QEMUIOVector *qiov, - int offset) -{ - BDRVNBDState *s = bs->opaque; - struct nbd_request request; - struct nbd_reply reply; - ssize_t ret; - - request.type = NBD_CMD_READ; - request.from = sector_num * 512; - request.len = nb_sectors * 512; - - nbd_coroutine_start(s, &request); - ret = nbd_co_send_request(s, &request, NULL, 0); - if (ret < 0) { - reply.error = -ret; - } else { - nbd_co_receive_reply(s, &request, &reply, qiov, offset); - } - nbd_coroutine_end(s, &request); - return -reply.error; - -} - -static int nbd_co_writev_1(BlockDriverState *bs, int64_t sector_num, - int nb_sectors, QEMUIOVector *qiov, - int offset) -{ - BDRVNBDState *s = bs->opaque; - struct nbd_request request; - struct nbd_reply reply; - ssize_t ret; - - request.type = NBD_CMD_WRITE; - if (!bdrv_enable_write_cache(bs) && (s->nbdflags & NBD_FLAG_SEND_FUA)) { - request.type |= NBD_CMD_FLAG_FUA; + sock = nbd_establish_connection(bs); + if (sock < 0) { + return sock; } - request.from = sector_num * 512; - request.len = nb_sectors * 512; - - nbd_coroutine_start(s, &request); - ret = nbd_co_send_request(s, &request, qiov, offset); - if (ret < 0) { - reply.error = -ret; - } else { - nbd_co_receive_reply(s, &request, &reply, NULL, 0); - } - nbd_coroutine_end(s, &request); - return -reply.error; + /* NBD handshake */ + return nbd_client_session_init(&s->client, bs, sock); } -/* qemu-nbd has a limit of slightly less than 1M per request. Try to - * remain aligned to 4K. */ -#define NBD_MAX_SECTORS 2040 - static int nbd_co_readv(BlockDriverState *bs, int64_t sector_num, int nb_sectors, QEMUIOVector *qiov) { - int offset = 0; - int ret; - while (nb_sectors > NBD_MAX_SECTORS) { - ret = nbd_co_readv_1(bs, sector_num, NBD_MAX_SECTORS, qiov, offset); - if (ret < 0) { - return ret; - } - offset += NBD_MAX_SECTORS * 512; - sector_num += NBD_MAX_SECTORS; - nb_sectors -= NBD_MAX_SECTORS; - } - return nbd_co_readv_1(bs, sector_num, nb_sectors, qiov, offset); + BDRVNBDState *s = bs->opaque; + + return nbd_client_session_co_readv(&s->client, sector_num, + nb_sectors, qiov); } static int nbd_co_writev(BlockDriverState *bs, int64_t sector_num, int nb_sectors, QEMUIOVector *qiov) { - int offset = 0; - int ret; - while (nb_sectors > NBD_MAX_SECTORS) { - ret = nbd_co_writev_1(bs, sector_num, NBD_MAX_SECTORS, qiov, offset); - if (ret < 0) { - return ret; - } - offset += NBD_MAX_SECTORS * 512; - sector_num += NBD_MAX_SECTORS; - nb_sectors -= NBD_MAX_SECTORS; - } - return nbd_co_writev_1(bs, sector_num, nb_sectors, qiov, offset); + BDRVNBDState *s = bs->opaque; + + return nbd_client_session_co_writev(&s->client, sector_num, + nb_sectors, qiov); } static int nbd_co_flush(BlockDriverState *bs) { BDRVNBDState *s = bs->opaque; - struct nbd_request request; - struct nbd_reply reply; - ssize_t ret; - - if (!(s->nbdflags & NBD_FLAG_SEND_FLUSH)) { - return 0; - } - - request.type = NBD_CMD_FLUSH; - if (s->nbdflags & NBD_FLAG_SEND_FUA) { - request.type |= NBD_CMD_FLAG_FUA; - } - - request.from = 0; - request.len = 0; - nbd_coroutine_start(s, &request); - ret = nbd_co_send_request(s, &request, NULL, 0); - if (ret < 0) { - reply.error = -ret; - } else { - nbd_co_receive_reply(s, &request, &reply, NULL, 0); - } - nbd_coroutine_end(s, &request); - return -reply.error; + return nbd_client_session_co_flush(&s->client); } static int nbd_co_discard(BlockDriverState *bs, int64_t sector_num, int nb_sectors) { BDRVNBDState *s = bs->opaque; - struct nbd_request request; - struct nbd_reply reply; - ssize_t ret; - if (!(s->nbdflags & NBD_FLAG_SEND_TRIM)) { - return 0; - } - request.type = NBD_CMD_TRIM; - request.from = sector_num * 512; - request.len = nb_sectors * 512; - - nbd_coroutine_start(s, &request); - ret = nbd_co_send_request(s, &request, NULL, 0); - if (ret < 0) { - reply.error = -ret; - } else { - nbd_co_receive_reply(s, &request, &reply, NULL, 0); - } - nbd_coroutine_end(s, &request); - return -reply.error; + return nbd_client_session_co_discard(&s->client, sector_num, + nb_sectors); } static void nbd_close(BlockDriverState *bs) { BDRVNBDState *s = bs->opaque; - g_free(s->export_name); - qemu_opts_del(s->socket_opts); - nbd_teardown_connection(bs); + qemu_opts_del(s->socket_opts); + nbd_client_session_close(&s->client); } static int64_t nbd_getlength(BlockDriverState *bs) { BDRVNBDState *s = bs->opaque; - return s->size; + return s->client.size; } static BlockDriver bdrv_nbd = {