| lists.openwall.net | lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC | |
|
Open Source and information security mailing list archives
| ||
|
Message-ID: <02e4f1d6-f16e-4c0f-89d3-c75eea93b96f@oracle.com>
Date: Mon, 15 Dec 2025 11:03:58 -0500
From: Chuck Lever <chuck.lever@...cle.com>
To: Al Viro <viro@...iv.linux.org.uk>, Hugh Dickins <hughd@...gle.com>
Cc: Miklos Szeredi <miklos@...redi.hu>,
Christian Brauner
<brauner@...nel.org>,
Andrew Morton <akpm@...ux-foundation.org>,
Baolin Wang <baolin.wang@...ux.alibaba.com>,
linux-fsdevel@...r.kernel.org, linux-kernel@...r.kernel.org,
linux-mm@...ck.org, Linus Torvalds <torvalds@...ux-foundation.org>
Subject: Re: [RFC][PATCH 2/2] shmem: fix recovery on rename failures
On 12/13/25 10:30 PM, Al Viro wrote:
> maple_tree insertions can fail if we are seriously short on memory;
> simple_offset_rename() does not recover well if it runs into that.
> The same goes for simple_offset_rename_exchange().
>
> Moreover, shmem_whiteout() expects that if it succeeds, the caller will
> progress to d_move(), i.e. that shmem_rename2() won't fail past the
> successful call of shmem_whiteout().
>
> Not hard to fix, fortunately - mtree_store() can't fail if the index we
> are trying to store into is already present in the tree as a singleton.
>
> For simple_offset_rename_exchange() that's enough - we just need to be
> careful about the order of operations.
>
> For simple_offset_rename() solution is to preinsert the target into the
> tree for new_dir; the rest can be done without any potentially failing
> operations.
>
> That preinsertion has to be done in shmem_rename2() rather than in
> simple_offset_rename() itself - otherwise we'd need to deal with the
> possibility of failure after successful shmem_whiteout().
>
> Fixes: a2e459555c5f ("shmem: stable directory offsets")
> Signed-off-by: Al Viro <viro@...iv.linux.org.uk>
> ---
> fs/libfs.c | 50 +++++++++++++++++++---------------------------
> include/linux/fs.h | 2 +-
> mm/shmem.c | 18 ++++++++++++-----
> 3 files changed, 35 insertions(+), 35 deletions(-)
>
> diff --git a/fs/libfs.c b/fs/libfs.c
> index 9264523be85c..591eb649ebba 100644
> --- a/fs/libfs.c
> +++ b/fs/libfs.c
> @@ -346,22 +346,22 @@ void simple_offset_remove(struct offset_ctx *octx, struct dentry *dentry)
> * User space expects the directory offset value of the replaced
> * (new) directory entry to be unchanged after a rename.
> *
> - * Returns zero on success, a negative errno value on failure.
> + * Caller must have grabbed a slot for new_dentry in the maple_tree
> + * associated with new_dir, even if dentry is negative.
> */
> -int simple_offset_rename(struct inode *old_dir, struct dentry *old_dentry,
> - struct inode *new_dir, struct dentry *new_dentry)
> +void simple_offset_rename(struct inode *old_dir, struct dentry *old_dentry,
> + struct inode *new_dir, struct dentry *new_dentry)
> {
> struct offset_ctx *old_ctx = old_dir->i_op->get_offset_ctx(old_dir);
> struct offset_ctx *new_ctx = new_dir->i_op->get_offset_ctx(new_dir);
> long new_offset = dentry2offset(new_dentry);
>
> - simple_offset_remove(old_ctx, old_dentry);
> + if (WARN_ON(!new_offset))
> + return;
>
> - if (new_offset) {
> - offset_set(new_dentry, 0);
> - return simple_offset_replace(new_ctx, old_dentry, new_offset);
> - }
> - return simple_offset_add(new_ctx, old_dentry);
> + simple_offset_remove(old_ctx, old_dentry);
> + offset_set(new_dentry, 0);
> + WARN_ON(simple_offset_replace(new_ctx, old_dentry, new_offset));
> }
>
> /**
> @@ -388,31 +388,23 @@ int simple_offset_rename_exchange(struct inode *old_dir,
> long new_index = dentry2offset(new_dentry);
> int ret;
>
> - simple_offset_remove(old_ctx, old_dentry);
> - simple_offset_remove(new_ctx, new_dentry);
> + if (WARN_ON(!old_index || !new_index))
> + return -EINVAL;
>
> - ret = simple_offset_replace(new_ctx, old_dentry, new_index);
> - if (ret)
> - goto out_restore;
> + ret = mtree_store(&new_ctx->mt, new_index, old_dentry, GFP_KERNEL);
> + if (WARN_ON(ret))
> + return ret;
>
> - ret = simple_offset_replace(old_ctx, new_dentry, old_index);
> - if (ret) {
> - simple_offset_remove(new_ctx, old_dentry);
> - goto out_restore;
> + ret = mtree_store(&old_ctx->mt, old_index, new_dentry, GFP_KERNEL);
> + if (WARN_ON(ret)) {
> + mtree_store(&new_ctx->mt, new_index, new_dentry, GFP_KERNEL);
Under extreme memory pressure, this mtree_store() might also fail?
> + return ret;
> }
>
> - ret = simple_rename_exchange(old_dir, old_dentry, new_dir, new_dentry);
> - if (ret) {
> - simple_offset_remove(new_ctx, old_dentry);
> - simple_offset_remove(old_ctx, new_dentry);
> - goto out_restore;
> - }
> + offset_set(old_dentry, new_index);
> + offset_set(new_dentry, old_index);
> + simple_rename_exchange(old_dir, old_dentry, new_dir, new_dentry);
> return 0;
> -
> -out_restore:
> - (void)simple_offset_replace(old_ctx, old_dentry, old_index);
> - (void)simple_offset_replace(new_ctx, new_dentry, new_index);
> - return ret;
> }
>
> /**
> diff --git a/include/linux/fs.h b/include/linux/fs.h
> index 04ceeca12a0d..f5c9cf28c4dc 100644
> --- a/include/linux/fs.h
> +++ b/include/linux/fs.h
> @@ -3247,7 +3247,7 @@ struct offset_ctx {
> void simple_offset_init(struct offset_ctx *octx);
> int simple_offset_add(struct offset_ctx *octx, struct dentry *dentry);
> void simple_offset_remove(struct offset_ctx *octx, struct dentry *dentry);
> -int simple_offset_rename(struct inode *old_dir, struct dentry *old_dentry,
> +void simple_offset_rename(struct inode *old_dir, struct dentry *old_dentry,
> struct inode *new_dir, struct dentry *new_dentry);
> int simple_offset_rename_exchange(struct inode *old_dir,
> struct dentry *old_dentry,
> diff --git a/mm/shmem.c b/mm/shmem.c
> index d3edc809e2e7..4232f8a39a43 100644
> --- a/mm/shmem.c
> +++ b/mm/shmem.c
> @@ -4039,6 +4039,7 @@ static int shmem_rename2(struct mnt_idmap *idmap,
> struct inode *inode = d_inode(old_dentry);
> int they_are_dirs = S_ISDIR(inode->i_mode);
> int error;
> + int had_offset = false;
>
> if (flags & ~(RENAME_NOREPLACE | RENAME_EXCHANGE | RENAME_WHITEOUT))
> return -EINVAL;
> @@ -4050,16 +4051,23 @@ static int shmem_rename2(struct mnt_idmap *idmap,
> if (!simple_empty(new_dentry))
> return -ENOTEMPTY;
>
> + error = simple_offset_add(shmem_get_offset_ctx(new_dir), new_dentry);
> + if (error == -EBUSY)
> + had_offset = true;
> + else if (unlikely(error))
> + return error;
> +
> if (flags & RENAME_WHITEOUT) {
> error = shmem_whiteout(idmap, old_dir, old_dentry);
> - if (error)
> + if (error) {
> + if (!had_offset)
> + simple_offset_remove(shmem_get_offset_ctx(new_dir),
> + new_dentry);
> return error;
> + }
> }
>
> - error = simple_offset_rename(old_dir, old_dentry, new_dir, new_dentry);
> - if (error)
> - return error;
> -
> + simple_offset_rename(old_dir, old_dentry, new_dir, new_dentry);
> if (d_really_is_positive(new_dentry)) {
> (void) shmem_unlink(new_dir, new_dentry);
> if (they_are_dirs) {
Reviewed-by: Chuck Lever <chuck.lever@...cle.com>
--
Chuck Lever
Powered by blists - more mailing lists