From patchwork Wed Sep 5 14:53:22 2018 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Ilya Dryomov X-Patchwork-Id: 10589009 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 246ED920 for ; Wed, 5 Sep 2018 14:53:54 +0000 (UTC) Received: from mail.wl.linuxfoundation.org (localhost [127.0.0.1]) by mail.wl.linuxfoundation.org (Postfix) with ESMTP id 0BE7429703 for ; Wed, 5 Sep 2018 14:53:54 +0000 (UTC) Received: by mail.wl.linuxfoundation.org (Postfix, from userid 486) id 001D62A368; Wed, 5 Sep 2018 14:53:53 +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=-8.0 required=2.0 tests=BAYES_00,DKIM_SIGNED, DKIM_VALID,DKIM_VALID_AU,FREEMAIL_FROM,MAILING_LIST_MULTI,RCVD_IN_DNSWL_HI autolearn=ham version=3.3.1 Received: from vger.kernel.org (vger.kernel.org [209.132.180.67]) by mail.wl.linuxfoundation.org (Postfix) with ESMTP id 68CE629703 for ; Wed, 5 Sep 2018 14:53:53 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1727604AbeIETYY (ORCPT ); Wed, 5 Sep 2018 15:24:24 -0400 Received: from mail-wm0-f65.google.com ([74.125.82.65]:52824 "EHLO mail-wm0-f65.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1726335AbeIETYY (ORCPT ); Wed, 5 Sep 2018 15:24:24 -0400 Received: by mail-wm0-f65.google.com with SMTP id y139-v6so8250313wmc.2 for ; Wed, 05 Sep 2018 07:53:50 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20161025; h=from:to:subject:date:message-id:in-reply-to:references; bh=/nL3pexej0/CXrqldX5GQUXR43Dtwuh5Dv94KEYUCMQ=; b=EknexVHZUzB6HJ7aVuFAHZC80nUmT+kz52vVZEeiS1EBB1LS9g/JVYKQkT2b25EuFE HfY7BF56G2ZPjxdeVp5eCkVPqlj5Z8EHfZWeE9lBMQluvtbnz9vpoC22J7dBG+XmgRKe reQjlJU8vRg8OhmUWJHStDZg05WiJZMlgEPAsESGkry3oV3WYMqbm1ot8ECZ/rA7eViR MjkNAJwg4R7q45tu6ey1YX7bqYdVIcijOArMUFA2fX+Xpo1+OTBATvIQgxV4haWbsHzO aFW7nkhfSgG+xFWc72566UFGHv7HS/2URg3rPs4ceSHTC1vlEOh5DNjm7hdNJPKGgvaL NZxQ== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20161025; h=x-gm-message-state:from:to:subject:date:message-id:in-reply-to :references; bh=/nL3pexej0/CXrqldX5GQUXR43Dtwuh5Dv94KEYUCMQ=; b=f03h7D/sPJ1lKySV+l0vxxi+V83jORG9SsAgb36+MjNZ3hfm7XLOw7bqqigKJvMrEB XOQQ0XJ3Y55UhV3o+Kx50QbQB6iXPUABOiklIeXtgdmblLSvkLhCyGl/+yiPCCMlsPNs 3zZcSECDGFotm6d28S+KhqWXvvxMAumtj8Vk2pkoiJxOgHTZ35amoLayyOjLC1zO3CQY axyGRKuN96A/Tmfolhu65rVOhxZpfOdGpn9Q6JeI1YrRfYwf/7u7l82OYIVzYoakeM5o ZUpBB/lTosOARW6xcTsKSmHgNKuLMGnuCJzRqX408nQqFk4UHbJO/z/FHzsuIxjmQ1sl 4/5A== X-Gm-Message-State: APzg51DCNHPVOqDKdlVSlwi0PsOf+cg0ruuZx34aUhNKoUfVfoRjOsTE C2Fa6ivSPSH0B8lHJXHA2E//jC8W X-Google-Smtp-Source: ANB0VdarxSxsBNP0ym2S8oNOOVuHS/AQRE2ff8PeAZc4LIewA3jqezjvpZud4n6thlejPtV+Ant6IA== X-Received: by 2002:a1c:8b13:: with SMTP id n19-v6mr488179wmd.118.1536159229237; Wed, 05 Sep 2018 07:53:49 -0700 (PDT) Received: from orange.redhat.com ([213.175.37.12]) by smtp.gmail.com with ESMTPSA id j44-v6sm2517315wre.40.2018.09.05.07.53.48 for (version=TLS1_2 cipher=ECDHE-RSA-CHACHA20-POLY1305 bits=256/256); Wed, 05 Sep 2018 07:53:48 -0700 (PDT) From: Ilya Dryomov To: ceph-devel@vger.kernel.org Subject: [PATCH v2 2/2] rbd: support cloning across namespaces Date: Wed, 5 Sep 2018 16:53:22 +0200 Message-Id: <20180905145322.7311-3-idryomov@gmail.com> X-Mailer: git-send-email 2.14.4 In-Reply-To: <20180905145322.7311-1-idryomov@gmail.com> References: <20180905145322.7311-1-idryomov@gmail.com> Sender: ceph-devel-owner@vger.kernel.org Precedence: bulk List-ID: X-Mailing-List: ceph-devel@vger.kernel.org X-Virus-Scanned: ClamAV using ClamSMTP If parent_get class method is not supported by the OSDs, fall back to the legacy class method and assume that the parent is in the default (i.e. "") namespace. The "use the child's image namespace" workaround is no longer needed because creating images within namespaces will require parent_get aware OSDs. Signed-off-by: Ilya Dryomov --- drivers/block/rbd.c | 111 +++++++++++++++++++++++++++++++++++++++++++++------- 1 file changed, 97 insertions(+), 14 deletions(-) diff --git a/drivers/block/rbd.c b/drivers/block/rbd.c index 6116a2eb7f51..a5addea0fade 100644 --- a/drivers/block/rbd.c +++ b/drivers/block/rbd.c @@ -4207,11 +4207,13 @@ static ssize_t rbd_parent_show(struct device *dev, count += sprintf(&buf[count], "%s" "pool_id %llu\npool_name %s\n" + "pool_ns %s\n" "image_id %s\nimage_name %s\n" "snap_id %llu\nsnap_name %s\n" "overlap %llu\n", !count ? "" : "\n", /* first? */ spec->pool_id, spec->pool_name, + spec->pool_ns ?: "", spec->image_id, spec->image_name ?: "(unknown)", spec->snap_id, spec->snap_name, rbd_dev->parent_overlap); @@ -4586,12 +4588,89 @@ static int rbd_dev_v2_features(struct rbd_device *rbd_dev) struct parent_image_info { u64 pool_id; + const char *pool_ns; const char *image_id; u64 snap_id; + bool has_overlap; u64 overlap; }; +/* + * The caller is responsible for @pii. + */ +static int decode_parent_image_spec(void **p, void *end, + struct parent_image_info *pii) +{ + u8 struct_v; + u32 struct_len; + int ret; + + ret = ceph_start_decoding(p, end, 1, "ParentImageSpec", + &struct_v, &struct_len); + if (ret) + return ret; + + ceph_decode_64_safe(p, end, pii->pool_id, e_inval); + pii->pool_ns = ceph_extract_encoded_string(p, end, NULL, GFP_KERNEL); + if (IS_ERR(pii->pool_ns)) { + ret = PTR_ERR(pii->pool_ns); + pii->pool_ns = NULL; + return ret; + } + pii->image_id = ceph_extract_encoded_string(p, end, NULL, GFP_KERNEL); + if (IS_ERR(pii->image_id)) { + ret = PTR_ERR(pii->image_id); + pii->image_id = NULL; + return ret; + } + ceph_decode_64_safe(p, end, pii->snap_id, e_inval); + return 0; + +e_inval: + return -EINVAL; +} + +static int __get_parent_info(struct rbd_device *rbd_dev, + struct page *req_page, + struct page *reply_page, + struct parent_image_info *pii) +{ + struct ceph_osd_client *osdc = &rbd_dev->rbd_client->client->osdc; + size_t reply_len = PAGE_SIZE; + void *p, *end; + int ret; + + ret = ceph_osdc_call(osdc, &rbd_dev->header_oid, &rbd_dev->header_oloc, + "rbd", "parent_get", CEPH_OSD_FLAG_READ, + req_page, sizeof(u64), reply_page, &reply_len); + if (ret) + return ret == -EOPNOTSUPP ? 1 : ret; + + p = page_address(reply_page); + end = p + reply_len; + ret = decode_parent_image_spec(&p, end, pii); + if (ret) + return ret; + + ret = ceph_osdc_call(osdc, &rbd_dev->header_oid, &rbd_dev->header_oloc, + "rbd", "parent_overlap_get", CEPH_OSD_FLAG_READ, + req_page, sizeof(u64), reply_page, &reply_len); + if (ret) + return ret; + + p = page_address(reply_page); + end = p + reply_len; + ceph_decode_8_safe(&p, end, pii->has_overlap, e_inval); + if (pii->has_overlap) + ceph_decode_64_safe(&p, end, pii->overlap, e_inval); + + return 0; + +e_inval: + return -EINVAL; +} + /* * The caller is responsible for @pii. */ @@ -4621,6 +4700,7 @@ static int __get_parent_info_legacy(struct rbd_device *rbd_dev, return ret; } ceph_decode_64_safe(&p, end, pii->snap_id, e_inval); + pii->has_overlap = true; ceph_decode_64_safe(&p, end, pii->overlap, e_inval); return 0; @@ -4648,7 +4728,10 @@ static int get_parent_info(struct rbd_device *rbd_dev, p = page_address(req_page); ceph_encode_64(&p, rbd_dev->spec->snap_id); - ret = __get_parent_info_legacy(rbd_dev, req_page, reply_page, pii); + ret = __get_parent_info(rbd_dev, req_page, reply_page, pii); + if (ret > 0) + ret = __get_parent_info_legacy(rbd_dev, req_page, reply_page, + pii); __free_page(req_page); __free_page(reply_page); @@ -4669,10 +4752,11 @@ static int rbd_dev_v2_parent_info(struct rbd_device *rbd_dev) if (ret) goto out_err; - dout("%s pool_id %llu image_id %s snap_id %llu overlap %llu\n", - __func__, pii.pool_id, pii.image_id, pii.snap_id, pii.overlap); + dout("%s pool_id %llu pool_ns %s image_id %s snap_id %llu has_overlap %d overlap %llu\n", + __func__, pii.pool_id, pii.pool_ns, pii.image_id, pii.snap_id, + pii.has_overlap, pii.overlap); - if (pii.pool_id == CEPH_NOPOOL) { + if (pii.pool_id == CEPH_NOPOOL || !pii.has_overlap) { /* * Either the parent never existed, or we have * record of it but the image got flattened so it no @@ -4681,6 +4765,10 @@ static int rbd_dev_v2_parent_info(struct rbd_device *rbd_dev) * overlap to 0. The effect of this is that all new * requests will be treated as if the image had no * parent. + * + * If !pii.has_overlap, the parent image spec is not + * applicable. It's there to avoid duplication in each + * snapshot record. */ if (rbd_dev->parent_overlap) { rbd_dev->parent_overlap = 0; @@ -4708,20 +4796,14 @@ static int rbd_dev_v2_parent_info(struct rbd_device *rbd_dev) */ if (!rbd_dev->parent_spec) { parent_spec->pool_id = pii.pool_id; + if (pii.pool_ns && *pii.pool_ns) { + parent_spec->pool_ns = pii.pool_ns; + pii.pool_ns = NULL; + } parent_spec->image_id = pii.image_id; pii.image_id = NULL; parent_spec->snap_id = pii.snap_id; - /* TODO: support cloning across namespaces */ - if (rbd_dev->spec->pool_ns) { - parent_spec->pool_ns = kstrdup(rbd_dev->spec->pool_ns, - GFP_KERNEL); - if (!parent_spec->pool_ns) { - ret = -ENOMEM; - goto out_err; - } - } - rbd_dev->parent_spec = parent_spec; parent_spec = NULL; /* rbd_dev now owns this */ } @@ -4746,6 +4828,7 @@ static int rbd_dev_v2_parent_info(struct rbd_device *rbd_dev) out: ret = 0; out_err: + kfree(pii.pool_ns); kfree(pii.image_id); rbd_spec_put(parent_spec); return ret;