diff --git a/fs/ceph/addr.c b/fs/ceph/addr.c index 29be367905a1..5a5751156f8b 100644 --- a/fs/ceph/addr.c +++ b/fs/ceph/addr.c @@ -407,6 +407,15 @@ static void ceph_netfs_issue_read(struct netfs_io_subrequest *subreq) struct page **pages; size_t page_off; + /* + * FIXME: io_iter.count needs to be corrected to aligned + * length. Otherwise, iov_iter_get_pages_alloc2() operates + * with the initial unaligned length value. As a result, + * ceph_msg_data_cursor_init() triggers BUG_ON() in the case + * if msg->sparse_read_total > msg->data_length. + */ + subreq->io_iter.count = len; + err = iov_iter_get_pages_alloc2(&subreq->io_iter, &pages, len, &page_off); if (err < 0) { doutc(cl, "%llx.%llx failed to allocate pages, %d\n", -- knotmirror 0.0.1 From 0abd87942e0c93964e93224836944712feba1d91 Mon Sep 17 00:00:00 2001 From: Dmitry Kandybka Date: Tue, 22 Apr 2025 12:32:04 +0300 Subject: [PATCH 2/5] ceph: fix possible integer overflow in ceph_zero_objects() In 'ceph_zero_objects', promote 'object_size' to 'u64' to avoid possible integer overflow. Compile tested only. Found by Linux Verification Center (linuxtesting.org) with SVACE. Signed-off-by: Dmitry Kandybka Reviewed-by: Viacheslav Dubeyko Signed-off-by: Ilya Dryomov --- fs/ceph/file.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/fs/ceph/file.c b/fs/ceph/file.c index 851d70200c6b..a7254cab44cc 100644 --- a/fs/ceph/file.c +++ b/fs/ceph/file.c @@ -2616,7 +2616,7 @@ static int ceph_zero_objects(struct inode *inode, loff_t offset, loff_t length) s32 stripe_unit = ci->i_layout.stripe_unit; s32 stripe_count = ci->i_layout.stripe_count; s32 object_size = ci->i_layout.object_size; - u64 object_set_size = object_size * stripe_count; + u64 object_set_size = (u64) object_size * stripe_count; u64 nearly, t; /* round offset up to next period boundary */ -- knotmirror 0.0.1 From d50eb28f2de5a35670ce10c0d34b9c9088f43f54 Mon Sep 17 00:00:00 2001 From: Viacheslav Dubeyko Date: Mon, 10 Feb 2025 15:01:58 -0800 Subject: [PATCH 3/5] ceph: cleanup hardcoded constants of file handle size The ceph/export.c contains very confusing logic of file handle size calculation based on hardcoded values. This patch makes the cleanup of this logic by means of introduction the named constants. Signed-off-by: Viacheslav Dubeyko Reviewed-by: Alex Markuze Signed-off-by: Ilya Dryomov --- fs/ceph/export.c | 21 +++++++++++++-------- 1 file changed, 13 insertions(+), 8 deletions(-) diff --git a/fs/ceph/export.c b/fs/ceph/export.c index 150076ced937..b2f2af104679 100644 --- a/fs/ceph/export.c +++ b/fs/ceph/export.c @@ -33,12 +33,19 @@ struct ceph_nfs_snapfh { u32 hash; } __attribute__ ((packed)); +#define BYTES_PER_U32 (sizeof(u32)) +#define CEPH_FH_BASIC_SIZE \ + (sizeof(struct ceph_nfs_fh) / BYTES_PER_U32) +#define CEPH_FH_WITH_PARENT_SIZE \ + (sizeof(struct ceph_nfs_confh) / BYTES_PER_U32) +#define CEPH_FH_SNAPPED_INODE_SIZE \ + (sizeof(struct ceph_nfs_snapfh) / BYTES_PER_U32) + static int ceph_encode_snapfh(struct inode *inode, u32 *rawfh, int *max_len, struct inode *parent_inode) { struct ceph_client *cl = ceph_inode_to_client(inode); - static const int snap_handle_length = - sizeof(struct ceph_nfs_snapfh) >> 2; + static const int snap_handle_length = CEPH_FH_SNAPPED_INODE_SIZE; struct ceph_nfs_snapfh *sfh = (void *)rawfh; u64 snapid = ceph_snap(inode); int ret; @@ -88,10 +95,8 @@ static int ceph_encode_fh(struct inode *inode, u32 *rawfh, int *max_len, struct inode *parent_inode) { struct ceph_client *cl = ceph_inode_to_client(inode); - static const int handle_length = - sizeof(struct ceph_nfs_fh) >> 2; - static const int connected_handle_length = - sizeof(struct ceph_nfs_confh) >> 2; + static const int handle_length = CEPH_FH_BASIC_SIZE; + static const int connected_handle_length = CEPH_FH_WITH_PARENT_SIZE; int type; if (ceph_snap(inode) != CEPH_NOSNAP) @@ -308,7 +313,7 @@ static struct dentry *ceph_fh_to_dentry(struct super_block *sb, if (fh_type != FILEID_INO32_GEN && fh_type != FILEID_INO32_GEN_PARENT) return NULL; - if (fh_len < sizeof(*fh) / 4) + if (fh_len < sizeof(*fh) / BYTES_PER_U32) return NULL; doutc(fsc->client, "%llx\n", fh->ino); @@ -427,7 +432,7 @@ static struct dentry *ceph_fh_to_parent(struct super_block *sb, if (fh_type != FILEID_INO32_GEN_PARENT) return NULL; - if (fh_len < sizeof(*cfh) / 4) + if (fh_len < sizeof(*cfh) / BYTES_PER_U32) return NULL; doutc(fsc->client, "%llx\n", cfh->parent_ino); -- knotmirror 0.0.1 From 72386d5245b249f5a0a8fabb881df7ad947b8ea4 Mon Sep 17 00:00:00 2001 From: Dennis Marttinen Date: Thu, 29 May 2025 17:45:12 +0000 Subject: [PATCH 4/5] ceph: set superblock s_magic for IMA fsmagic matching The CephFS kernel driver forgets to set the filesystem magic signature in its superblock. As a result, IMA policy rules based on fsmagic matching do not apply as intended. This causes a major performance regression in Talos Linux [1] when mounting CephFS volumes, such as when deploying Rook Ceph [2]. Talos Linux ships a hardened kernel with the following IMA policy (irrelevant lines omitted): [...] dont_measure fsmagic=0xc36400 # CEPH_SUPER_MAGIC [...] measure func=FILE_CHECK mask=^MAY_READ euid=0 measure func=FILE_CHECK mask=^MAY_READ uid=0 [...] Currently, IMA compares 0xc36400 == 0x0 for CephFS files, resulting in all files opened with O_RDONLY or O_RDWR getting measured with SHA512 on every open(2): 10 69990c87e8af323d47e2d6ae4... ima-ng sha512: /data/cephfs/test-file Since O_WRONLY is rare, this results in an order of magnitude lower performance than expected for practically all file operations. Properly setting CEPH_SUPER_MAGIC in the CephFS superblock resolves the regression. Tests performed on a 3x replicated Ceph v19.3.0 cluster across three i5-7200U nodes each equipped with one Micron 7400 MAX M.2 disk (BlueStore) and Gigabit ethernet, on Talos Linux v1.10.2: FS-Mark 3.3 Test: 500 Files, Empty Files/s > Higher Is Better 6.12.27-talos . 16.6 |==== +twelho patch . 208.4 |==================================================== FS-Mark 3.3 Test: 500 Files, 1KB Size Files/s > Higher Is Better 6.12.27-talos . 15.6 |======= +twelho patch . 118.6 |==================================================== FS-Mark 3.3 Test: 500 Files, 32 Sub Dirs, 1MB Size Files/s > Higher Is Better 6.12.27-talos . 12.7 |=============== +twelho patch . 44.7 |===================================================== IO500 [3] 2fcd6d6 results (benchmarks within variance omitted): | IO500 benchmark | 6.12.27-talos | +twelho patch | Speedup | |-------------------|----------------|----------------|-----------| | mdtest-easy-write | 0.018524 kIOPS | 1.135027 kIOPS | 6027.33 % | | mdtest-hard-write | 0.018498 kIOPS | 0.973312 kIOPS | 5161.71 % | | ior-easy-read | 0.064727 GiB/s | 0.155324 GiB/s | 139.97 % | | mdtest-hard-read | 0.018246 kIOPS | 0.780800 kIOPS | 4179.29 % | This applies outside of synthetic benchmarks as well, for example, the time to rsync a 55 MiB directory with ~12k of mostly small files drops from an unusable 10m5s to a reasonable 26s (23x the throughput). [1]: https://www.talos.dev/ [2]: https://www.talos.dev/v1.10/kubernetes-guides/configuration/ceph-with-rook/ [3]: https://github.com/IO500/io500 Cc: stable@vger.kernel.org Signed-off-by: Dennis Marttinen Reviewed-by: Viacheslav Dubeyko Signed-off-by: Ilya Dryomov --- fs/ceph/super.c | 1 + 1 file changed, 1 insertion(+) diff --git a/fs/ceph/super.c b/fs/ceph/super.c index f3951253e393..fc4cab8b7b77 100644 --- a/fs/ceph/super.c +++ b/fs/ceph/super.c @@ -1227,6 +1227,7 @@ static int ceph_set_super(struct super_block *s, struct fs_context *fc) s->s_time_min = 0; s->s_time_max = U32_MAX; s->s_flags |= SB_NODIRATIME | SB_NOATIME; + s->s_magic = CEPH_SUPER_MAGIC; ceph_fscrypt_set_ops(s); -- knotmirror 0.0.1 From b828b4bf29d10a3e505a76a39c4daea969e19dc9 Mon Sep 17 00:00:00 2001 From: Viacheslav Dubeyko Date: Mon, 2 Jun 2025 11:49:56 -0700 Subject: [PATCH 5/5] ceph: fix variable dereferenced before check in ceph_umount_begin() smatch warnings: fs/ceph/super.c:1042 ceph_umount_begin() warn: variable dereferenced before check 'fsc' (see line 1041) vim +/fsc +1042 fs/ceph/super.c void ceph_umount_begin(struct super_block *sb) { struct ceph_fs_client *fsc = ceph_sb_to_fs_client(sb); doutc(fsc->client, "starting forced umount\n"); ^^^^^^^^^^^ Dereferenced if (!fsc) ^^^^ Checked too late. return; fsc->mount_state = CEPH_MOUNT_SHUTDOWN; __ceph_umount_begin(fsc); } The VFS guarantees that the superblock is still alive when it calls into ceph via ->umount_begin(). Finally, we don't need to check the fsc and it should be valid. This patch simply removes the fsc check. Reported-by: kernel test robot Reported-by: Dan Carpenter Closes: https://lore.kernel.org/r/202503280852.YDB3pxUY-lkp@intel.com/ Signed-off-by: Viacheslav Dubeyko Reviewed by: Alex Markuze Signed-off-by: Ilya Dryomov --- fs/ceph/super.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) diff --git a/fs/ceph/super.c b/fs/ceph/super.c index fc4cab8b7b77..2b8438d8a324 100644 --- a/fs/ceph/super.c +++ b/fs/ceph/super.c @@ -1033,8 +1033,7 @@ void ceph_umount_begin(struct super_block *sb) struct ceph_fs_client *fsc = ceph_sb_to_fs_client(sb); doutc(fsc->client, "starting forced umount\n"); - if (!fsc) - return; + fsc->mount_state = CEPH_MOUNT_SHUTDOWN; __ceph_umount_begin(fsc); }