From patchwork Mon Feb 18 14:02:57 2019 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Stefano Garzarella X-Patchwork-Id: 10818061 Return-Path: Received: from mail.wl.linuxfoundation.org (pdx-wl-mail.web.codeaurora.org [172.30.200.125]) by pdx-korg-patchwork-2.web.codeaurora.org (Postfix) with ESMTP id AD6C31390 for ; Mon, 18 Feb 2019 14:08:48 +0000 (UTC) Received: from mail.wl.linuxfoundation.org (localhost [127.0.0.1]) by mail.wl.linuxfoundation.org (Postfix) with ESMTP id 98DC62ACDD for ; Mon, 18 Feb 2019 14:08:48 +0000 (UTC) Received: by mail.wl.linuxfoundation.org (Postfix, from userid 486) id 89D1F2ACF8; Mon, 18 Feb 2019 14:08:48 +0000 (UTC) X-Spam-Checker-Version: SpamAssassin 3.3.1 (2010-03-16) on pdx-wl-mail.web.codeaurora.org X-Spam-Level: X-Spam-Status: No, score=-2.9 required=2.0 tests=BAYES_00,MAILING_LIST_MULTI autolearn=ham version=3.3.1 Received: from lists.gnu.org (lists.gnu.org [209.51.188.17]) (using TLSv1 with cipher AES256-SHA (256/256 bits)) (No client certificate requested) by mail.wl.linuxfoundation.org (Postfix) with ESMTPS id D62812ACDD for ; Mon, 18 Feb 2019 14:08:47 +0000 (UTC) Received: from localhost ([127.0.0.1]:59121 helo=lists.gnu.org) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1gvjbD-0003qu-3Z for patchwork-qemu-devel@patchwork.kernel.org; Mon, 18 Feb 2019 09:08:47 -0500 Received: from eggs.gnu.org ([209.51.188.92]:59832) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1gvjYY-0001cB-A5 for qemu-devel@nongnu.org; Mon, 18 Feb 2019 09:06:08 -0500 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1gvjYV-0005KH-UI for qemu-devel@nongnu.org; Mon, 18 Feb 2019 09:06:02 -0500 Received: from mx1.redhat.com ([209.132.183.28]:60820) by eggs.gnu.org with esmtps (TLS1.0:DHE_RSA_AES_256_CBC_SHA1:32) (Exim 4.71) (envelope-from ) id 1gvjYL-0005Ay-Bu; Mon, 18 Feb 2019 09:05:49 -0500 Received: from smtp.corp.redhat.com (int-mx07.intmail.prod.int.phx2.redhat.com [10.5.11.22]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by mx1.redhat.com (Postfix) with ESMTPS id 476D87A481; Mon, 18 Feb 2019 14:05:40 +0000 (UTC) Received: from steredhat.redhat.com (ovpn-116-155.ams2.redhat.com [10.36.116.155]) by smtp.corp.redhat.com (Postfix) with ESMTP id 012431001F52; Mon, 18 Feb 2019 14:05:22 +0000 (UTC) From: Stefano Garzarella To: qemu-devel@nongnu.org Date: Mon, 18 Feb 2019 15:02:57 +0100 Message-Id: <20190218140301.197408-7-sgarzare@redhat.com> In-Reply-To: <20190218140301.197408-1-sgarzare@redhat.com> References: <20190218140301.197408-1-sgarzare@redhat.com> MIME-Version: 1.0 X-Scanned-By: MIMEDefang 2.84 on 10.5.11.22 X-Greylist: Sender IP whitelisted, not delayed by milter-greylist-4.5.16 (mx1.redhat.com [10.5.110.27]); Mon, 18 Feb 2019 14:05:40 +0000 (UTC) X-detected-operating-system: by eggs.gnu.org: GNU/Linux 2.2.x-3.x [generic] X-Received-From: 209.132.183.28 Subject: [Qemu-devel] [PATCH v5 06/10] virtio-blk: add DISCARD and WRITE_ZEROES features 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: Kevin Wolf , Laurent Vivier , Thomas Huth , Eduardo Habkost , qemu-block@nongnu.org, "Michael S. Tsirkin" , Jason Wang , Stefan Hajnoczi , "Dr . David Alan Gilbert" , Max Reitz , Paolo Bonzini Errors-To: qemu-devel-bounces+patchwork-qemu-devel=patchwork.kernel.org@nongnu.org Sender: "Qemu-devel" X-Virus-Scanned: ClamAV using ClamSMTP This patch adds the support of DISCARD and WRITE_ZEROES commands, that have been introduced in the virtio-blk protocol to have better performance when using SSD backend. We support only one segment per request since multiple segments are not widely used and there are no userspace APIs that allow applications to submit multiple segments in a single call. Reviewed-by: Michael S. Tsirkin Reviewed-by: Stefan Hajnoczi Signed-off-by: Stefano Garzarella --- hw/block/virtio-blk.c | 184 +++++++++++++++++++++++++++++++++ include/hw/virtio/virtio-blk.h | 2 + 2 files changed, 186 insertions(+) diff --git a/hw/block/virtio-blk.c b/hw/block/virtio-blk.c index 8798d13bc4..c159a3d5f7 100644 --- a/hw/block/virtio-blk.c +++ b/hw/block/virtio-blk.c @@ -169,6 +169,30 @@ out: aio_context_release(blk_get_aio_context(s->conf.conf.blk)); } +static void virtio_blk_discard_write_zeroes_complete(void *opaque, int ret) +{ + VirtIOBlockReq *req = opaque; + VirtIOBlock *s = req->dev; + bool is_write_zeroes = (virtio_ldl_p(VIRTIO_DEVICE(s), &req->out.type) & + ~VIRTIO_BLK_T_BARRIER) == VIRTIO_BLK_T_WRITE_ZEROES; + + aio_context_acquire(blk_get_aio_context(s->conf.conf.blk)); + if (ret) { + if (virtio_blk_handle_rw_error(req, -ret, false, is_write_zeroes)) { + goto out; + } + } + + virtio_blk_req_complete(req, VIRTIO_BLK_S_OK); + if (is_write_zeroes) { + block_acct_done(blk_get_stats(s->blk), &req->acct); + } + virtio_blk_free_request(req); + +out: + aio_context_release(blk_get_aio_context(s->conf.conf.blk)); +} + #ifdef __linux__ typedef struct { @@ -502,6 +526,84 @@ static bool virtio_blk_sect_range_ok(VirtIOBlock *dev, return true; } +static uint8_t virtio_blk_handle_discard_write_zeroes(VirtIOBlockReq *req, + struct virtio_blk_discard_write_zeroes *dwz_hdr, bool is_write_zeroes) +{ + VirtIOBlock *s = req->dev; + VirtIODevice *vdev = VIRTIO_DEVICE(s); + uint64_t sector; + uint32_t num_sectors, flags, max_sectors; + uint8_t err_status; + int bytes; + + sector = virtio_ldq_p(vdev, &dwz_hdr->sector); + num_sectors = virtio_ldl_p(vdev, &dwz_hdr->num_sectors); + flags = virtio_ldl_p(vdev, &dwz_hdr->flags); + max_sectors = is_write_zeroes ? s->conf.max_write_zeroes_sectors : + s->conf.max_discard_sectors; + + /* + * max_sectors is at most BDRV_REQUEST_MAX_SECTORS, this check + * make us sure that "num_sectors << BDRV_SECTOR_BITS" can fit in + * the integer variable. + */ + if (unlikely(num_sectors > max_sectors)) { + err_status = VIRTIO_BLK_S_IOERR; + goto err; + } + + bytes = num_sectors << BDRV_SECTOR_BITS; + + if (unlikely(!virtio_blk_sect_range_ok(s, sector, bytes))) { + err_status = VIRTIO_BLK_S_IOERR; + goto err; + } + + /* + * The device MUST set the status byte to VIRTIO_BLK_S_UNSUPP for discard + * and write zeroes commands if any unknown flag is set. + */ + if (unlikely(flags & ~VIRTIO_BLK_WRITE_ZEROES_FLAG_UNMAP)) { + err_status = VIRTIO_BLK_S_UNSUPP; + goto err; + } + + if (is_write_zeroes) { /* VIRTIO_BLK_T_WRITE_ZEROES */ + int blk_aio_flags = 0; + + if (flags & VIRTIO_BLK_WRITE_ZEROES_FLAG_UNMAP) { + blk_aio_flags |= BDRV_REQ_MAY_UNMAP; + } + + block_acct_start(blk_get_stats(s->blk), &req->acct, bytes, + BLOCK_ACCT_WRITE); + + blk_aio_pwrite_zeroes(s->blk, sector << BDRV_SECTOR_BITS, + bytes, blk_aio_flags, + virtio_blk_discard_write_zeroes_complete, req); + } else { /* VIRTIO_BLK_T_DISCARD */ + /* + * The device MUST set the status byte to VIRTIO_BLK_S_UNSUPP for + * discard commands if the unmap flag is set. + */ + if (unlikely(flags & VIRTIO_BLK_WRITE_ZEROES_FLAG_UNMAP)) { + err_status = VIRTIO_BLK_S_UNSUPP; + goto err; + } + + blk_aio_pdiscard(s->blk, sector << BDRV_SECTOR_BITS, bytes, + virtio_blk_discard_write_zeroes_complete, req); + } + + return VIRTIO_BLK_S_OK; + +err: + if (is_write_zeroes) { + block_acct_invalid(blk_get_stats(s->blk), BLOCK_ACCT_WRITE); + } + return err_status; +} + static int virtio_blk_handle_request(VirtIOBlockReq *req, MultiReqBuffer *mrb) { uint32_t type; @@ -603,6 +705,47 @@ static int virtio_blk_handle_request(VirtIOBlockReq *req, MultiReqBuffer *mrb) virtio_blk_free_request(req); break; } + /* + * VIRTIO_BLK_T_DISCARD and VIRTIO_BLK_T_WRITE_ZEROES are defined with + * VIRTIO_BLK_T_OUT flag set. We masked this flag in the switch statement, + * so we must mask it for these requests, then we will check if it is set. + */ + case VIRTIO_BLK_T_DISCARD & ~VIRTIO_BLK_T_OUT: + case VIRTIO_BLK_T_WRITE_ZEROES & ~VIRTIO_BLK_T_OUT: + { + struct virtio_blk_discard_write_zeroes dwz_hdr; + size_t out_len = iov_size(out_iov, out_num); + bool is_write_zeroes = (type & ~VIRTIO_BLK_T_BARRIER) == + VIRTIO_BLK_T_WRITE_ZEROES; + uint8_t err_status; + + /* + * Unsupported if VIRTIO_BLK_T_OUT is not set or the request contains + * more than one segment. + */ + if (unlikely(!(type & VIRTIO_BLK_T_OUT) || + out_len > sizeof(dwz_hdr))) { + virtio_blk_req_complete(req, VIRTIO_BLK_S_UNSUPP); + virtio_blk_free_request(req); + return 0; + } + + if (unlikely(iov_to_buf(out_iov, out_num, 0, &dwz_hdr, + sizeof(dwz_hdr)) != sizeof(dwz_hdr))) { + virtio_error(vdev, "virtio-blk discard/write_zeroes header" + " too short"); + return -1; + } + + err_status = virtio_blk_handle_discard_write_zeroes(req, &dwz_hdr, + is_write_zeroes); + if (err_status != VIRTIO_BLK_S_OK) { + virtio_blk_req_complete(req, err_status); + virtio_blk_free_request(req); + } + + break; + } default: virtio_blk_req_complete(req, VIRTIO_BLK_S_UNSUPP); virtio_blk_free_request(req); @@ -782,6 +925,24 @@ static void virtio_blk_update_config(VirtIODevice *vdev, uint8_t *config) blkcfg.alignment_offset = 0; blkcfg.wce = blk_enable_write_cache(s->blk); virtio_stw_p(vdev, &blkcfg.num_queues, s->conf.num_queues); + if (virtio_has_feature(s->host_features, VIRTIO_BLK_F_DISCARD)) { + virtio_stl_p(vdev, &blkcfg.max_discard_sectors, + s->conf.max_discard_sectors); + virtio_stl_p(vdev, &blkcfg.discard_sector_alignment, + blk_size >> BDRV_SECTOR_BITS); + /* + * We support only one segment per request since multiple segments + * are not widely used and there are no userspace APIs that allow + * applications to submit multiple segments in a single call. + */ + virtio_stl_p(vdev, &blkcfg.max_discard_seg, 1); + } + if (virtio_has_feature(s->host_features, VIRTIO_BLK_F_WRITE_ZEROES)) { + virtio_stl_p(vdev, &blkcfg.max_write_zeroes_sectors, + s->conf.max_write_zeroes_sectors); + blkcfg.write_zeroes_may_unmap = 1; + virtio_stl_p(vdev, &blkcfg.max_write_zeroes_seg, 1); + } memcpy(config, &blkcfg, s->config_size); } @@ -973,6 +1134,25 @@ static void virtio_blk_device_realize(DeviceState *dev, Error **errp) return; } + if (virtio_has_feature(s->host_features, VIRTIO_BLK_F_DISCARD) && + (!conf->max_discard_sectors || + conf->max_discard_sectors > BDRV_REQUEST_MAX_SECTORS)) { + error_setg(errp, "invalid max-discard-sectors property (%" PRIu32 ")" + ", must be between 1 and %d", + conf->max_discard_sectors, (int)BDRV_REQUEST_MAX_SECTORS); + return; + } + + if (virtio_has_feature(s->host_features, VIRTIO_BLK_F_WRITE_ZEROES) && + (!conf->max_write_zeroes_sectors || + conf->max_write_zeroes_sectors > BDRV_REQUEST_MAX_SECTORS)) { + error_setg(errp, "invalid max-write-zeroes-sectors property (%" PRIu32 + "), must be between 1 and %d", + conf->max_write_zeroes_sectors, + (int)BDRV_REQUEST_MAX_SECTORS); + return; + } + virtio_blk_set_config_size(s, s->host_features); virtio_init(vdev, "virtio-blk", VIRTIO_ID_BLOCK, s->config_size); @@ -1050,6 +1230,10 @@ static Property virtio_blk_properties[] = { VIRTIO_BLK_F_DISCARD, true), DEFINE_PROP_BIT64("write-zeroes", VirtIOBlock, host_features, VIRTIO_BLK_F_WRITE_ZEROES, true), + DEFINE_PROP_UINT32("max-discard-sectors", VirtIOBlock, + conf.max_discard_sectors, BDRV_REQUEST_MAX_SECTORS), + DEFINE_PROP_UINT32("max-write-zeroes-sectors", VirtIOBlock, + conf.max_write_zeroes_sectors, BDRV_REQUEST_MAX_SECTORS), DEFINE_PROP_END_OF_LIST(), }; diff --git a/include/hw/virtio/virtio-blk.h b/include/hw/virtio/virtio-blk.h index 7877ae67ae..cddcfbebe9 100644 --- a/include/hw/virtio/virtio-blk.h +++ b/include/hw/virtio/virtio-blk.h @@ -38,6 +38,8 @@ struct VirtIOBlkConf uint32_t request_merging; uint16_t num_queues; uint16_t queue_size; + uint32_t max_discard_sectors; + uint32_t max_write_zeroes_sectors; }; struct VirtIOBlockDataPlane;