From 620239d9a32e9fe27c9204ec11e40058671aeeb6 Mon Sep 17 00:00:00 2001 From: Jeff Layton Date: Mon, 25 Apr 2022 15:54:27 -0400 Subject: ceph: fix setting of xattrs on async created inodes Currently when we create a file, we spin up an xattr buffer to send along with the create request. If we end up doing an async create however, then we currently pass down a zero-length xattr buffer. Fix the code to send down the xattr buffer in req->r_pagelist. If the xattrs span more than a page, however give up and don't try to do an async create. Cc: stable@vger.kernel.org URL: https://bugzilla.redhat.com/show_bug.cgi?id=2063929 Fixes: 9a8d03ca2e2c ("ceph: attempt to do async create when possible") Reported-by: John Fortin Reported-by: Sri Ramanujam Signed-off-by: Jeff Layton Reviewed-by: Xiubo Li Signed-off-by: Ilya Dryomov --- fs/ceph/file.c | 16 +++++++++++++--- 1 file changed, 13 insertions(+), 3 deletions(-) diff --git a/fs/ceph/file.c b/fs/ceph/file.c index 6c9e837aa1d3..8c8226c0feac 100644 --- a/fs/ceph/file.c +++ b/fs/ceph/file.c @@ -629,9 +629,15 @@ static int ceph_finish_async_create(struct inode *dir, struct dentry *dentry, iinfo.change_attr = 1; ceph_encode_timespec64(&iinfo.btime, &now); - iinfo.xattr_len = ARRAY_SIZE(xattr_buf); - iinfo.xattr_data = xattr_buf; - memset(iinfo.xattr_data, 0, iinfo.xattr_len); + if (req->r_pagelist) { + iinfo.xattr_len = req->r_pagelist->length; + iinfo.xattr_data = req->r_pagelist->mapped_tail; + } else { + /* fake it */ + iinfo.xattr_len = ARRAY_SIZE(xattr_buf); + iinfo.xattr_data = xattr_buf; + memset(iinfo.xattr_data, 0, iinfo.xattr_len); + } in.ino = cpu_to_le64(vino.ino); in.snapid = cpu_to_le64(CEPH_NOSNAP); @@ -743,6 +749,10 @@ int ceph_atomic_open(struct inode *dir, struct dentry *dentry, err = ceph_security_init_secctx(dentry, mode, &as_ctx); if (err < 0) goto out_ctx; + /* Async create can't handle more than a page of xattrs */ + if (as_ctx.pagelist && + !list_is_singular(&as_ctx.pagelist->head)) + try_async = false; } else if (!d_in_lookup(dentry)) { /* If it's not being looked up, it's negative */ return -ENOENT; -- cgit v1.2.3 From 642d51fb0775a41dd6bb3d99b5a40c24df131c20 Mon Sep 17 00:00:00 2001 From: Xiubo Li Date: Thu, 5 May 2022 18:53:09 +0800 Subject: ceph: check folio PG_private bit instead of folio->private The pages in the file mapping maybe reclaimed and reused by other subsystems and the page->private maybe used as flags field or something else, if later that pages are used by page caches again the page->private maybe not cleared as expected. Here will check the PG_private bit instead of the folio->private. Cc: stable@vger.kernel.org URL: https://tracker.ceph.com/issues/55421 Signed-off-by: Xiubo Li Reviewed-by: Luis Henriques Reviewed-by: Jeff Layton Signed-off-by: Ilya Dryomov --- fs/ceph/addr.c | 11 +++++++---- 1 file changed, 7 insertions(+), 4 deletions(-) diff --git a/fs/ceph/addr.c b/fs/ceph/addr.c index aa25bffd4823..b6edcf89a429 100644 --- a/fs/ceph/addr.c +++ b/fs/ceph/addr.c @@ -85,7 +85,7 @@ static bool ceph_dirty_folio(struct address_space *mapping, struct folio *folio) if (folio_test_dirty(folio)) { dout("%p dirty_folio %p idx %lu -- already dirty\n", mapping->host, folio, folio->index); - BUG_ON(!folio_get_private(folio)); + VM_BUG_ON_FOLIO(!folio_test_private(folio), folio); return false; } @@ -122,7 +122,7 @@ static bool ceph_dirty_folio(struct address_space *mapping, struct folio *folio) * Reference snap context in folio->private. Also set * PagePrivate so that we get invalidate_folio callback. */ - BUG_ON(folio_get_private(folio)); + VM_BUG_ON_FOLIO(folio_test_private(folio), folio); folio_attach_private(folio, snapc); return ceph_fscache_dirty_folio(mapping, folio); @@ -150,7 +150,7 @@ static void ceph_invalidate_folio(struct folio *folio, size_t offset, } WARN_ON(!folio_test_locked(folio)); - if (folio_get_private(folio)) { + if (folio_test_private(folio)) { dout("%p invalidate_folio idx %lu full dirty page\n", inode, folio->index); @@ -729,8 +729,11 @@ static void writepages_finish(struct ceph_osd_request *req) /* clean all pages */ for (i = 0; i < req->r_num_ops; i++) { - if (req->r_ops[i].op != CEPH_OSD_OP_WRITE) + if (req->r_ops[i].op != CEPH_OSD_OP_WRITE) { + pr_warn("%s incorrect op %d req %p index %d tid %llu\n", + __func__, req->r_ops[i].op, req, i, req->r_tid); break; + } osd_data = osd_req_op_extent_osd_data(req, i); BUG_ON(osd_data->type != CEPH_OSD_DATA_TYPE_PAGES); -- cgit v1.2.3