[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <CAOQ4uxhjWwTdENS2GqmOxtx4hdbv=N4f90iLVuxHNgH=NLem9w@mail.gmail.com>
Date: Wed, 7 Jan 2026 21:43:25 +0100
From: Amir Goldstein <amir73il@...il.com>
To: Chunsheng Luo <luochunsheng@...c.edu>
Cc: miklos@...redi.hu, bschubert@....com, linux-unionfs@...r.kernel.org,
linux-kernel@...r.kernel.org, linux-fsdevel <linux-fsdevel@...r.kernel.org>
Subject: Re: [PATCH] overlayfs: mask d_type high bits before whiteout check
On Wed, Jan 7, 2026 at 4:46 AM Chunsheng Luo <luochunsheng@...c.edu> wrote:
>
> Commit c31f91c6af96 ("fuse: don't allow signals to interrupt getdents
> copying") introduced the use of high bits in d_type as flags. However,
> overlayfs was not adapted to handle this change.
>
> In ovl_cache_entry_new(), the code checks if d_type == DT_CHR to
> determine if an entry might be a whiteout. When fuse is used as the
> lower layer and sets high bits in d_type, this comparison fails,
> causing whiteout files to not be recognized properly and resulting in
> incorrect overlayfs behavior.
>
> Fix this by masking out the high bits with S_DT_MASK before checking.
>
> Fixes: c31f91c6af96 ("fuse: don't allow signals to interrupt getdents copying")
> Link: https://github.com/containerd/stargz-snapshotter/issues/2214
> Signed-off-by: Chunsheng Luo <luochunsheng@...c.edu>
Hi Chunsheng,
Thanks for the report and the suggested fix.
This time overlayfs was surprised by unexpected d_type flags and next
time it could be another user.
I prefer to fix this in a more profound way -
Instead of making overlafys aware of d_type flags, require the users that
use the d_type flags to opt-in for them.
Please test/review the attached patch.
Thanks,
Amir.
> ---
> fs/overlayfs/readdir.c | 15 +++++++++++++++
> 1 file changed, 15 insertions(+)
>
> diff --git a/fs/overlayfs/readdir.c b/fs/overlayfs/readdir.c
> index 160960bb0ad0..a2ac47458bf9 100644
> --- a/fs/overlayfs/readdir.c
> +++ b/fs/overlayfs/readdir.c
> @@ -246,6 +246,9 @@ static int ovl_fill_lowest(struct ovl_readdir_data *rdd,
> {
> struct ovl_cache_entry *p;
>
> + /* Mask out high bits that may be used (e.g., fuse) */
> + d_type &= S_DT_MASK;
> +
> p = ovl_cache_entry_find(rdd->root, c_name, c_len);
> if (p) {
> list_move_tail(&p->l_node, &rdd->middle);
> @@ -316,6 +319,9 @@ static bool ovl_fill_merge(struct dir_context *ctx, const char *name,
> char *cf_name = NULL;
> int c_len = 0, ret;
>
> + /* Mask out high bits that may be used (e.g., fuse) */
> + d_type &= S_DT_MASK;
> +
> if (ofs->casefold)
> c_len = ovl_casefold(rdd, name, namelen, &cf_name);
>
> @@ -632,6 +638,9 @@ static bool ovl_fill_plain(struct dir_context *ctx, const char *name,
> struct ovl_readdir_data *rdd =
> container_of(ctx, struct ovl_readdir_data, ctx);
>
> + /* Mask out high bits that may be used (e.g., fuse) */
> + d_type &= S_DT_MASK;
> +
> rdd->count++;
> p = ovl_cache_entry_new(rdd, name, namelen, NULL, 0, ino, d_type);
> if (p == NULL) {
> @@ -755,6 +764,9 @@ static bool ovl_fill_real(struct dir_context *ctx, const char *name,
> struct dir_context *orig_ctx = rdt->orig_ctx;
> bool res;
>
> + /* Mask out high bits that may be used (e.g., fuse) */
> + d_type &= S_DT_MASK;
> +
> if (rdt->parent_ino && strcmp(name, "..") == 0) {
> ino = rdt->parent_ino;
> } else if (rdt->cache) {
> @@ -1144,6 +1156,9 @@ static bool ovl_check_d_type(struct dir_context *ctx, const char *name,
> struct ovl_readdir_data *rdd =
> container_of(ctx, struct ovl_readdir_data, ctx);
>
> + /* Mask out high bits that may be used (e.g., fuse) */
> + d_type &= S_DT_MASK;
> +
> /* Even if d_type is not supported, DT_DIR is returned for . and .. */
> if (!strncmp(name, ".", namelen) || !strncmp(name, "..", namelen))
> return true;
> --
> 2.43.0
>
View attachment "0001-readdir-require-opt-in-for-d_type-flags.patch" of type "text/x-patch" (2921 bytes)
Powered by blists - more mailing lists