From: rick.macklem@gmail.com
To: linux-nfs@vger.kernel.org
Cc: Rick Macklem <rmacklem@uoguelph.ca>
Subject: [PATCH v1 13/17] Fix handling of POSIX ACLs with zero ACEs
Date: Tue, 30 Dec 2025 18:21:15 -0800 [thread overview]
Message-ID: <20251231022119.1714-14-rick.macklem@gmail.com> (raw)
In-Reply-To: <20251231022119.1714-1-rick.macklem@gmail.com>
From: Rick Macklem <rmacklem@uoguelph.ca>
The NFSv4.2 code does a SETATTR with a POSIX draft default ACL
of zero ACEs to delete the default ACL. This patch fixes handling
of this case.
Note that the server code does not handle the case where
a file system is of ACL_SCOPE_FILE OBJECT (which means that
NFS4 and POSIX draft ACLs are mixed as the "true form" ACL
for different file objects in the file system).
Signed-off-by: Rick Macklem <rmacklem@uoguelph.ca>
---
fs/nfsd/nfs4proc.c | 30 ++++++++++++++++++++++++------
1 file changed, 24 insertions(+), 6 deletions(-)
diff --git a/fs/nfsd/nfs4proc.c b/fs/nfsd/nfs4proc.c
index a172bd253086..1dab7d3aa11e 100644
--- a/fs/nfsd/nfs4proc.c
+++ b/fs/nfsd/nfs4proc.c
@@ -1215,12 +1215,30 @@ nfsd4_proc_setacl(struct svc_rqst *rqstp, svc_fh *fh, struct inode *inode,
inode_lock(inode);
- if (dpacl != NULL)
- error = set_posix_acl(&nop_mnt_idmap, fh->fh_dentry,
- ACL_TYPE_DEFAULT, dpacl);
- if (!error && pacl != NULL)
- error = set_posix_acl(&nop_mnt_idmap, fh->fh_dentry,
- ACL_TYPE_ACCESS, pacl);
+ if (dpacl != NULL) {
+ /* a_count == 0 means delete the ACL. */
+ if (dpacl->a_count > 0)
+ error = set_posix_acl(&nop_mnt_idmap, fh->fh_dentry,
+ ACL_TYPE_DEFAULT, dpacl);
+ else
+ error = set_posix_acl(&nop_mnt_idmap, fh->fh_dentry,
+ ACL_TYPE_DEFAULT, NULL);
+ }
+ if (!error && pacl != NULL) {
+ /*
+ * For any file system that is not ACL_SCOPE_FILE_OBJECT,
+ * a_count == 0 MUST reply nfserr_inval.
+ * For a file system that is ACL_SCOPE_FILE_OBJECT,
+ * a_count == 0 means delete the ACL.
+ * XXX File systems that are ACL_SCOPE_FILE_OBJECT
+ * are not yet supported.
+ */
+ if (pacl->a_count > 0)
+ error = set_posix_acl(&nop_mnt_idmap, fh->fh_dentry,
+ ACL_TYPE_ACCESS, pacl);
+ else
+ error = -EINVAL;
+ }
inode_unlock(inode);
return nfserrno(error);
--
2.49.0
next prev parent reply other threads:[~2025-12-31 2:22 UTC|newest]
Thread overview: 21+ messages / expand[flat|nested] mbox.gz Atom feed top
2025-12-31 2:21 [PATCH v1 00/17] Add NFSv4.2 POSIX ACL support rick.macklem
2025-12-31 2:21 ` [PATCH v1 01/17] Add definitions for the POSIX draft ACL attributes rick.macklem
2025-12-31 2:21 ` [PATCH v1 02/17] Add a new function to acquire the POSIX draft ACLs rick.macklem
2025-12-31 2:21 ` [PATCH v1 03/17] Add a function to set POSIX ACLs rick.macklem
2025-12-31 2:21 ` [PATCH v1 04/17] Add support for encoding/decoding POSIX draft ACLs rick.macklem
2025-12-31 2:21 ` [PATCH v1 05/17] Add a check for both POSIX and NFSv4 ACLs being set rick.macklem
2025-12-31 2:21 ` [PATCH v1 06/17] Add na_dpaclerr and na_paclerr for file creation rick.macklem
2025-12-31 2:21 ` [PATCH v1 07/17] Add support for POSIX draft ACLs " rick.macklem
2025-12-31 2:21 ` [PATCH v1 08/17] Add the arguments for decoding of POSIX ACLs rick.macklem
2025-12-31 2:21 ` [PATCH v1 09/17] Fix a couple of bugs in POSIX ACL decoding rick.macklem
2025-12-31 2:21 ` [PATCH v1 10/17] Improve correctness for the ACL_TRUEFORM attribute rick.macklem
2025-12-31 2:21 ` [PATCH v1 11/17] Make sort_pacl_range() global rick.macklem
2025-12-31 2:21 ` [PATCH v1 12/17] Call sort_pacl_range() for decoded POSIX draft ACLs rick.macklem
2025-12-31 2:21 ` rick.macklem [this message]
2025-12-31 2:21 ` [PATCH v1 14/17] Fix handling of zero length ACLs for file creation rick.macklem
2025-12-31 2:21 ` [PATCH v1 15/17] Do not allow (N)VERIFY to check POSIX ACL attributes rick.macklem
2025-12-31 2:21 ` [PATCH v1 16/17] Set the POSIX ACL attributes supported rick.macklem
2025-12-31 2:21 ` [PATCH v1 17/17] Change a bunch of function prefixes to nfsd42_ rick.macklem
2025-12-31 2:47 ` [PATCH v1 00/17] Add NFSv4.2 POSIX ACL support Chuck Lever
2025-12-31 2:49 ` Rick Macklem
2025-12-31 2:57 ` Rick Macklem
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20251231022119.1714-14-rick.macklem@gmail.com \
--to=rick.macklem@gmail.com \
--cc=linux-nfs@vger.kernel.org \
--cc=rmacklem@uoguelph.ca \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox