From d93231a6bc8a452323d5fef16cca7107ce483a27 Mon Sep 17 00:00:00 2001 From: Luís Henriques Date: Fri, 3 Jun 2022 14:29:09 +0100 Subject: ceph: prevent a client from exceeding the MDS maximum xattr size MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit The MDS tries to enforce a limit on the total key/values in extended attributes. However, this limit is enforced only if doing a synchronous operation (MDS_OP_SETXATTR) -- if we're buffering the xattrs, the MDS doesn't have a chance to enforce these limits. This patch adds support for decoding the xattrs maximum size setting that is distributed in the mdsmap. Then, when setting an xattr, the kernel client will revert to do a synchronous operation if that maximum size is exceeded. While there, fix a dout() that would trigger a printk warning: [ 98.718078] ------------[ cut here ]------------ [ 98.719012] precision 65536 too large [ 98.719039] WARNING: CPU: 1 PID: 3755 at lib/vsprintf.c:2703 vsnprintf+0x5e3/0x600 ... Link: https://tracker.ceph.com/issues/55725 Signed-off-by: Luís Henriques Reviewed-by: Xiubo Li Signed-off-by: Ilya Dryomov --- include/linux/ceph/mdsmap.h | 1 + 1 file changed, 1 insertion(+) (limited to 'include/linux/ceph') diff --git a/include/linux/ceph/mdsmap.h b/include/linux/ceph/mdsmap.h index 523fd0452856..4c3e0648dc27 100644 --- a/include/linux/ceph/mdsmap.h +++ b/include/linux/ceph/mdsmap.h @@ -25,6 +25,7 @@ struct ceph_mdsmap { u32 m_session_timeout; /* seconds */ u32 m_session_autoclose; /* seconds */ u64 m_max_file_size; + u64 m_max_xattr_size; /* maximum size for xattrs blob */ u32 m_max_mds; /* expected up:active mds number */ u32 m_num_active_mds; /* actual up:active mds number */ u32 possible_max_rank; /* possible max rank index */ -- cgit v1.2.3 From 1b7587d69ea7b8c350195392bfd30a0f31374ae4 Mon Sep 17 00:00:00 2001 From: Xiubo Li Date: Mon, 6 Jun 2022 20:15:33 +0800 Subject: ceph: fix the incorrect comment for the ceph_mds_caps struct The incorrect comment is misleading. Acutally the last members in ceph_mds_caps strcut is a union for none export and export bodies. Signed-off-by: Xiubo Li Reviewed-by: Jeff Layton Signed-off-by: Ilya Dryomov --- include/linux/ceph/ceph_fs.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include/linux/ceph') diff --git a/include/linux/ceph/ceph_fs.h b/include/linux/ceph/ceph_fs.h index 86bf82dbd8b8..40740e234ce1 100644 --- a/include/linux/ceph/ceph_fs.h +++ b/include/linux/ceph/ceph_fs.h @@ -768,7 +768,7 @@ struct ceph_mds_caps { __le32 xattr_len; __le64 xattr_version; - /* filelock */ + /* a union of non-export and export bodies. */ __le64 size, max_size, truncate_size; __le32 truncate_seq; struct ceph_timespec mtime, atime, ctime; -- cgit v1.2.3 From b53aca4b460ae2ece453963ef01667ee8ee78f52 Mon Sep 17 00:00:00 2001 From: Xiubo Li Date: Thu, 9 Jun 2022 15:21:37 +0800 Subject: ceph: fix incorrect old_size length in ceph_mds_request_args The 'old_size' is a __le64 type since birth, not sure why the kclient incorrectly switched it to __le32. This change is okay won't break anything because union will always allocate more memory than the 'open' member needed. Rename 'file_replication' to 'pool' as ceph did. Though this 'open' struct may never be used in kclient in future, it's confusing when going through the ceph code. Signed-off-by: Xiubo Li Reviewed-by: Ilya Dryomov Signed-off-by: Ilya Dryomov --- include/linux/ceph/ceph_fs.h | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) (limited to 'include/linux/ceph') diff --git a/include/linux/ceph/ceph_fs.h b/include/linux/ceph/ceph_fs.h index 40740e234ce1..49586ff26152 100644 --- a/include/linux/ceph/ceph_fs.h +++ b/include/linux/ceph/ceph_fs.h @@ -433,9 +433,9 @@ union ceph_mds_request_args { __le32 stripe_unit; /* layout for newly created file */ __le32 stripe_count; /* ... */ __le32 object_size; - __le32 file_replication; - __le32 mask; /* CEPH_CAP_* */ - __le32 old_size; + __le32 pool; + __le32 mask; /* CEPH_CAP_* */ + __le64 old_size; } __attribute__ ((packed)) open; struct { __le32 flags; -- cgit v1.2.3 From a8af0d682ae0c9cf62dd0ad6afdb1480951d6a10 Mon Sep 17 00:00:00 2001 From: Jeff Layton Date: Thu, 30 Jun 2022 16:21:50 -0400 Subject: libceph: clean up ceph_osdc_start_request prototype This function always returns 0, and ignores the nofail boolean. Drop the nofail argument, make the function void return and fix up the callers. Signed-off-by: Jeff Layton Reviewed-by: Ilya Dryomov Signed-off-by: Ilya Dryomov --- drivers/block/rbd.c | 6 +++--- fs/ceph/addr.c | 33 +++++++++++++-------------------- fs/ceph/file.c | 32 +++++++++++++------------------- include/linux/ceph/osd_client.h | 5 ++--- net/ceph/osd_client.c | 15 ++++++--------- 5 files changed, 37 insertions(+), 54 deletions(-) (limited to 'include/linux/ceph') diff --git a/drivers/block/rbd.c b/drivers/block/rbd.c index ef9bc62e9afd..b4580b73479f 100644 --- a/drivers/block/rbd.c +++ b/drivers/block/rbd.c @@ -1297,7 +1297,7 @@ static void rbd_osd_submit(struct ceph_osd_request *osd_req) dout("%s osd_req %p for obj_req %p objno %llu %llu~%llu\n", __func__, osd_req, obj_req, obj_req->ex.oe_objno, obj_req->ex.oe_off, obj_req->ex.oe_len); - ceph_osdc_start_request(osd_req->r_osdc, osd_req, false); + ceph_osdc_start_request(osd_req->r_osdc, osd_req); } /* @@ -2081,7 +2081,7 @@ static int rbd_object_map_update(struct rbd_obj_request *obj_req, u64 snap_id, if (ret) return ret; - ceph_osdc_start_request(osdc, req, false); + ceph_osdc_start_request(osdc, req); return 0; } @@ -4768,7 +4768,7 @@ static int rbd_obj_read_sync(struct rbd_device *rbd_dev, if (ret) goto out_req; - ceph_osdc_start_request(osdc, req, false); + ceph_osdc_start_request(osdc, req); ret = ceph_osdc_wait_request(osdc, req); if (ret >= 0) ceph_copy_from_page_vector(pages, buf, 0, ret); diff --git a/fs/ceph/addr.c b/fs/ceph/addr.c index de12715c237b..ec76e77f8d4b 100644 --- a/fs/ceph/addr.c +++ b/fs/ceph/addr.c @@ -337,6 +337,7 @@ static void ceph_netfs_issue_read(struct netfs_io_subrequest *subreq) /* should always give us a page-aligned read */ WARN_ON_ONCE(page_off); len = err; + err = 0; osd_req_op_extent_osd_data_pages(req, 0, pages, len, 0, false, false); req->r_callback = finish_netfs_read; @@ -344,9 +345,7 @@ static void ceph_netfs_issue_read(struct netfs_io_subrequest *subreq) req->r_inode = inode; ihold(inode); - err = ceph_osdc_start_request(req->r_osdc, req, false); - if (err) - iput(inode); + ceph_osdc_start_request(req->r_osdc, req); out: ceph_osdc_put_request(req); if (err) @@ -620,9 +619,8 @@ static int writepage_nounlock(struct page *page, struct writeback_control *wbc) dout("writepage %llu~%llu (%llu bytes)\n", page_off, len, len); req->r_mtime = inode->i_mtime; - err = ceph_osdc_start_request(osdc, req, true); - if (!err) - err = ceph_osdc_wait_request(osdc, req); + ceph_osdc_start_request(osdc, req); + err = ceph_osdc_wait_request(osdc, req); ceph_update_write_metrics(&fsc->mdsc->metric, req->r_start_latency, req->r_end_latency, len, err); @@ -1150,8 +1148,7 @@ new_request: } req->r_mtime = inode->i_mtime; - rc = ceph_osdc_start_request(&fsc->client->osdc, req, true); - BUG_ON(rc); + ceph_osdc_start_request(&fsc->client->osdc, req); req = NULL; wbc->nr_to_write -= i; @@ -1692,9 +1689,8 @@ int ceph_uninline_data(struct file *file) } req->r_mtime = inode->i_mtime; - err = ceph_osdc_start_request(&fsc->client->osdc, req, false); - if (!err) - err = ceph_osdc_wait_request(&fsc->client->osdc, req); + ceph_osdc_start_request(&fsc->client->osdc, req); + err = ceph_osdc_wait_request(&fsc->client->osdc, req); ceph_osdc_put_request(req); if (err < 0) goto out_unlock; @@ -1735,9 +1731,8 @@ int ceph_uninline_data(struct file *file) } req->r_mtime = inode->i_mtime; - err = ceph_osdc_start_request(&fsc->client->osdc, req, false); - if (!err) - err = ceph_osdc_wait_request(&fsc->client->osdc, req); + ceph_osdc_start_request(&fsc->client->osdc, req); + err = ceph_osdc_wait_request(&fsc->client->osdc, req); ceph_update_write_metrics(&fsc->mdsc->metric, req->r_start_latency, req->r_end_latency, len, err); @@ -1908,15 +1903,13 @@ static int __ceph_pool_perm_get(struct ceph_inode_info *ci, osd_req_op_raw_data_in_pages(rd_req, 0, pages, PAGE_SIZE, 0, false, true); - err = ceph_osdc_start_request(&fsc->client->osdc, rd_req, false); + ceph_osdc_start_request(&fsc->client->osdc, rd_req); wr_req->r_mtime = ci->netfs.inode.i_mtime; - err2 = ceph_osdc_start_request(&fsc->client->osdc, wr_req, false); + ceph_osdc_start_request(&fsc->client->osdc, wr_req); - if (!err) - err = ceph_osdc_wait_request(&fsc->client->osdc, rd_req); - if (!err2) - err2 = ceph_osdc_wait_request(&fsc->client->osdc, wr_req); + err = ceph_osdc_wait_request(&fsc->client->osdc, rd_req); + err2 = ceph_osdc_wait_request(&fsc->client->osdc, wr_req); if (err >= 0 || err == -ENOENT) have |= POOL_READ; diff --git a/fs/ceph/file.c b/fs/ceph/file.c index b4e978420802..c3caa9bf9755 100644 --- a/fs/ceph/file.c +++ b/fs/ceph/file.c @@ -985,9 +985,8 @@ static ssize_t ceph_sync_read(struct kiocb *iocb, struct iov_iter *to, osd_req_op_extent_osd_data_pages(req, 0, pages, len, page_off, false, false); - ret = ceph_osdc_start_request(osdc, req, false); - if (!ret) - ret = ceph_osdc_wait_request(osdc, req); + ceph_osdc_start_request(osdc, req); + ret = ceph_osdc_wait_request(osdc, req); ceph_update_read_metrics(&fsc->mdsc->metric, req->r_start_latency, @@ -1250,7 +1249,7 @@ static void ceph_aio_retry_work(struct work_struct *work) req->r_inode = inode; req->r_priv = aio_req; - ret = ceph_osdc_start_request(req->r_osdc, req, false); + ceph_osdc_start_request(req->r_osdc, req); out: if (ret < 0) { req->r_result = ret; @@ -1387,9 +1386,8 @@ ceph_direct_read_write(struct kiocb *iocb, struct iov_iter *iter, continue; } - ret = ceph_osdc_start_request(req->r_osdc, req, false); - if (!ret) - ret = ceph_osdc_wait_request(&fsc->client->osdc, req); + ceph_osdc_start_request(req->r_osdc, req); + ret = ceph_osdc_wait_request(&fsc->client->osdc, req); if (write) ceph_update_write_metrics(metric, req->r_start_latency, @@ -1452,8 +1450,7 @@ ceph_direct_read_write(struct kiocb *iocb, struct iov_iter *iter, r_private_item); list_del_init(&req->r_private_item); if (ret >= 0) - ret = ceph_osdc_start_request(req->r_osdc, - req, false); + ceph_osdc_start_request(req->r_osdc, req); if (ret < 0) { req->r_result = ret; ceph_aio_complete_req(req); @@ -1566,9 +1563,8 @@ ceph_sync_write(struct kiocb *iocb, struct iov_iter *from, loff_t pos, false, true); req->r_mtime = mtime; - ret = ceph_osdc_start_request(&fsc->client->osdc, req, false); - if (!ret) - ret = ceph_osdc_wait_request(&fsc->client->osdc, req); + ceph_osdc_start_request(&fsc->client->osdc, req); + ret = ceph_osdc_wait_request(&fsc->client->osdc, req); ceph_update_write_metrics(&fsc->mdsc->metric, req->r_start_latency, req->r_end_latency, len, ret); @@ -2032,12 +2028,10 @@ static int ceph_zero_partial_object(struct inode *inode, } req->r_mtime = inode->i_mtime; - ret = ceph_osdc_start_request(&fsc->client->osdc, req, false); - if (!ret) { - ret = ceph_osdc_wait_request(&fsc->client->osdc, req); - if (ret == -ENOENT) - ret = 0; - } + ceph_osdc_start_request(&fsc->client->osdc, req); + ret = ceph_osdc_wait_request(&fsc->client->osdc, req); + if (ret == -ENOENT) + ret = 0; ceph_osdc_put_request(req); out: @@ -2339,7 +2333,7 @@ static ssize_t ceph_do_objects_copy(struct ceph_inode_info *src_ci, u64 *src_off if (IS_ERR(req)) ret = PTR_ERR(req); else { - ceph_osdc_start_request(osdc, req, false); + ceph_osdc_start_request(osdc, req); ret = ceph_osdc_wait_request(osdc, req); ceph_update_copyfrom_metrics(&fsc->mdsc->metric, req->r_start_latency, diff --git a/include/linux/ceph/osd_client.h b/include/linux/ceph/osd_client.h index cba8a6ffc329..fb6be72104df 100644 --- a/include/linux/ceph/osd_client.h +++ b/include/linux/ceph/osd_client.h @@ -507,9 +507,8 @@ extern struct ceph_osd_request *ceph_osdc_new_request(struct ceph_osd_client *, extern void ceph_osdc_get_request(struct ceph_osd_request *req); extern void ceph_osdc_put_request(struct ceph_osd_request *req); -extern int ceph_osdc_start_request(struct ceph_osd_client *osdc, - struct ceph_osd_request *req, - bool nofail); +void ceph_osdc_start_request(struct ceph_osd_client *osdc, + struct ceph_osd_request *req); extern void ceph_osdc_cancel_request(struct ceph_osd_request *req); extern int ceph_osdc_wait_request(struct ceph_osd_client *osdc, struct ceph_osd_request *req); diff --git a/net/ceph/osd_client.c b/net/ceph/osd_client.c index 9d82bb42e958..87b883c7bfd6 100644 --- a/net/ceph/osd_client.c +++ b/net/ceph/osd_client.c @@ -4578,15 +4578,12 @@ bad: /* * Register request, send initial attempt. */ -int ceph_osdc_start_request(struct ceph_osd_client *osdc, - struct ceph_osd_request *req, - bool nofail) +void ceph_osdc_start_request(struct ceph_osd_client *osdc, + struct ceph_osd_request *req) { down_read(&osdc->lock); submit_request(req, false); up_read(&osdc->lock); - - return 0; } EXPORT_SYMBOL(ceph_osdc_start_request); @@ -4756,7 +4753,7 @@ int ceph_osdc_unwatch(struct ceph_osd_client *osdc, if (ret) goto out_put_req; - ceph_osdc_start_request(osdc, req, false); + ceph_osdc_start_request(osdc, req); linger_cancel(lreq); linger_put(lreq); ret = wait_request_timeout(req, opts->mount_timeout); @@ -4827,7 +4824,7 @@ int ceph_osdc_notify_ack(struct ceph_osd_client *osdc, if (ret) goto out_put_req; - ceph_osdc_start_request(osdc, req, false); + ceph_osdc_start_request(osdc, req); ret = ceph_osdc_wait_request(osdc, req); out_put_req: @@ -5043,7 +5040,7 @@ int ceph_osdc_list_watchers(struct ceph_osd_client *osdc, if (ret) goto out_put_req; - ceph_osdc_start_request(osdc, req, false); + ceph_osdc_start_request(osdc, req); ret = ceph_osdc_wait_request(osdc, req); if (ret >= 0) { void *p = page_address(pages[0]); @@ -5120,7 +5117,7 @@ int ceph_osdc_call(struct ceph_osd_client *osdc, if (ret) goto out_put_req; - ceph_osdc_start_request(osdc, req, false); + ceph_osdc_start_request(osdc, req); ret = ceph_osdc_wait_request(osdc, req); if (ret >= 0) { ret = req->r_ops[0].rval; -- cgit v1.2.3