From: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
To: stable@vger.kernel.org
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
patches@lists.linux.dev,
Zhang Tianci <zhangtianci.1997@bytedance.com>,
Jiachen Zhang <zhangjiachen.jaycee@bytedance.com>,
"Christian Brauner (Microsoft)" <brauner@kernel.org>,
Miklos Szeredi <mszeredi@redhat.com>
Subject: [PATCH 6.1 54/71] ovl: Use ovl mounters fsuid and fsgid in ovl_link()
Date: Mon, 2 Jan 2023 12:22:19 +0100 [thread overview]
Message-ID: <20230102110553.758530779@linuxfoundation.org> (raw)
In-Reply-To: <20230102110551.509937186@linuxfoundation.org>
From: Zhang Tianci <zhangtianci.1997@bytedance.com>
commit 5b0db51215e895a361bc63132caa7cca36a53d6a upstream.
There is a wrong case of link() on overlay:
$ mkdir /lower /fuse /merge
$ mount -t fuse /fuse
$ mkdir /fuse/upper /fuse/work
$ mount -t overlay /merge -o lowerdir=/lower,upperdir=/fuse/upper,\
workdir=work
$ touch /merge/file
$ chown bin.bin /merge/file // the file's caller becomes "bin"
$ ln /merge/file /merge/lnkfile
Then we will get an error(EACCES) because fuse daemon checks the link()'s
caller is "bin", it denied this request.
In the changing history of ovl_link(), there are two key commits:
The first is commit bb0d2b8ad296 ("ovl: fix sgid on directory") which
overrides the cred's fsuid/fsgid using the new inode. The new inode's
owner is initialized by inode_init_owner(), and inode->fsuid is
assigned to the current user. So the override fsuid becomes the
current user. We know link() is actually modifying the directory, so
the caller must have the MAY_WRITE permission on the directory. The
current caller may should have this permission. This is acceptable
to use the caller's fsuid.
The second is commit 51f7e52dc943 ("ovl: share inode for hard link")
which removed the inode creation in ovl_link(). This commit move
inode_init_owner() into ovl_create_object(), so the ovl_link() just
give the old inode to ovl_create_or_link(). Then the override fsuid
becomes the old inode's fsuid, neither the caller nor the overlay's
mounter! So this is incorrect.
Fix this bug by using ovl mounter's fsuid/fsgid to do underlying
fs's link().
Link: https://lore.kernel.org/all/20220817102952.xnvesg3a7rbv576x@wittgenstein/T
Link: https://lore.kernel.org/lkml/20220825130552.29587-1-zhangtianci.1997@bytedance.com/t
Signed-off-by: Zhang Tianci <zhangtianci.1997@bytedance.com>
Signed-off-by: Jiachen Zhang <zhangjiachen.jaycee@bytedance.com>
Reviewed-by: Christian Brauner (Microsoft) <brauner@kernel.org>
Fixes: 51f7e52dc943 ("ovl: share inode for hard link")
Cc: <stable@vger.kernel.org> # v4.8
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
---
fs/overlayfs/dir.c | 46 ++++++++++++++++++++++++++++++----------------
1 file changed, 30 insertions(+), 16 deletions(-)
--- a/fs/overlayfs/dir.c
+++ b/fs/overlayfs/dir.c
@@ -592,28 +592,42 @@ static int ovl_create_or_link(struct den
goto out_revert_creds;
}
- err = -ENOMEM;
- override_cred = prepare_creds();
- if (override_cred) {
+ if (!attr->hardlink) {
+ err = -ENOMEM;
+ override_cred = prepare_creds();
+ if (!override_cred)
+ goto out_revert_creds;
+ /*
+ * In the creation cases(create, mkdir, mknod, symlink),
+ * ovl should transfer current's fs{u,g}id to underlying
+ * fs. Because underlying fs want to initialize its new
+ * inode owner using current's fs{u,g}id. And in this
+ * case, the @inode is a new inode that is initialized
+ * in inode_init_owner() to current's fs{u,g}id. So use
+ * the inode's i_{u,g}id to override the cred's fs{u,g}id.
+ *
+ * But in the other hardlink case, ovl_link() does not
+ * create a new inode, so just use the ovl mounter's
+ * fs{u,g}id.
+ */
override_cred->fsuid = inode->i_uid;
override_cred->fsgid = inode->i_gid;
- if (!attr->hardlink) {
- err = security_dentry_create_files_as(dentry,
- attr->mode, &dentry->d_name, old_cred,
- override_cred);
- if (err) {
- put_cred(override_cred);
- goto out_revert_creds;
- }
+ err = security_dentry_create_files_as(dentry,
+ attr->mode, &dentry->d_name, old_cred,
+ override_cred);
+ if (err) {
+ put_cred(override_cred);
+ goto out_revert_creds;
}
put_cred(override_creds(override_cred));
put_cred(override_cred);
-
- if (!ovl_dentry_is_whiteout(dentry))
- err = ovl_create_upper(dentry, inode, attr);
- else
- err = ovl_create_over_whiteout(dentry, inode, attr);
}
+
+ if (!ovl_dentry_is_whiteout(dentry))
+ err = ovl_create_upper(dentry, inode, attr);
+ else
+ err = ovl_create_over_whiteout(dentry, inode, attr);
+
out_revert_creds:
revert_creds(old_cred);
return err;
next prev parent reply other threads:[~2023-01-02 11:25 UTC|newest]
Thread overview: 84+ messages / expand[flat|nested] mbox.gz Atom feed top
2023-01-02 11:21 [PATCH 6.1 00/71] 6.1.3-rc1 review Greg Kroah-Hartman
2023-01-02 11:21 ` [PATCH 6.1 01/71] eventpoll: add EPOLL_URING_WAKE poll wakeup flag Greg Kroah-Hartman
2023-01-02 11:21 ` [PATCH 6.1 02/71] eventfd: provide a eventfd_signal_mask() helper Greg Kroah-Hartman
2023-01-02 11:21 ` [PATCH 6.1 03/71] io_uring: pass in EPOLL_URING_WAKE for eventfd signaling and wakeups Greg Kroah-Hartman
2023-01-02 11:21 ` [PATCH 6.1 04/71] nvme-pci: fix doorbell buffer value endianness Greg Kroah-Hartman
2023-01-02 11:21 ` [PATCH 6.1 05/71] nvme-pci: fix mempool alloc size Greg Kroah-Hartman
2023-01-02 11:21 ` [PATCH 6.1 06/71] nvme-pci: fix page size checks Greg Kroah-Hartman
2023-01-02 11:21 ` [PATCH 6.1 07/71] ACPI: resource: do IRQ override on XMG Core 15 Greg Kroah-Hartman
2023-01-02 11:21 ` [PATCH 6.1 08/71] ACPI: resource: do IRQ override on Lenovo 14ALC7 Greg Kroah-Hartman
2023-01-02 11:21 ` [PATCH 6.1 09/71] ACPI: resource: Add Asus ExpertBook B2502 to Asus quirks Greg Kroah-Hartman
2023-01-02 11:21 ` [PATCH 6.1 10/71] ACPI: video: Fix Apple GMUX backlight detection Greg Kroah-Hartman
2023-01-02 11:21 ` [PATCH 6.1 11/71] block, bfq: fix uaf for bfqq in bfq_exit_icq_bfqq Greg Kroah-Hartman
2023-01-02 11:21 ` [PATCH 6.1 12/71] ata: ahci: Fix PCS quirk application for suspend Greg Kroah-Hartman
2023-01-02 11:21 ` [PATCH 6.1 13/71] nvme: fix the NVME_CMD_EFFECTS_CSE_MASK definition Greg Kroah-Hartman
2023-01-02 11:21 ` [PATCH 6.1 14/71] nvmet: dont defer passthrough commands with trivial effects to the workqueue Greg Kroah-Hartman
2023-01-02 11:21 ` [PATCH 6.1 15/71] fs/ntfs3: Validate BOOT record_size Greg Kroah-Hartman
2023-01-02 11:21 ` [PATCH 6.1 16/71] fs/ntfs3: Add overflow check for attribute size Greg Kroah-Hartman
2023-01-02 11:21 ` [PATCH 6.1 17/71] fs/ntfs3: Validate data run offset Greg Kroah-Hartman
2023-01-02 11:21 ` [PATCH 6.1 18/71] fs/ntfs3: Add null pointer check to attr_load_runs_vcn Greg Kroah-Hartman
2023-01-02 11:21 ` [PATCH 6.1 19/71] fs/ntfs3: Fix memory leak on ntfs_fill_super() error path Greg Kroah-Hartman
2023-01-02 11:21 ` [PATCH 6.1 20/71] fs/ntfs3: Add null pointer check for inode operations Greg Kroah-Hartman
2023-01-02 11:21 ` [PATCH 6.1 21/71] fs/ntfs3: Validate attribute name offset Greg Kroah-Hartman
2023-01-02 11:21 ` [PATCH 6.1 22/71] fs/ntfs3: Validate buffer length while parsing index Greg Kroah-Hartman
2023-01-02 11:21 ` [PATCH 6.1 23/71] fs/ntfs3: Validate resident attribute name Greg Kroah-Hartman
2023-01-02 11:21 ` [PATCH 6.1 24/71] fs/ntfs3: Fix slab-out-of-bounds read in run_unpack Greg Kroah-Hartman
2023-01-02 11:21 ` [PATCH 6.1 25/71] soundwire: dmi-quirks: add quirk variant for LAPBC710 NUC15 Greg Kroah-Hartman
2023-01-02 11:21 ` [PATCH 6.1 26/71] phy: sun4i-usb: Introduce port2 SIDDQ quirk Greg Kroah-Hartman
2023-01-02 11:21 ` [PATCH 6.1 27/71] phy: sun4i-usb: Add support for the H616 USB PHY Greg Kroah-Hartman
2023-01-02 11:21 ` [PATCH 6.1 28/71] fs/ntfs3: Validate index root when initialize NTFS security Greg Kroah-Hartman
2023-01-02 11:21 ` [PATCH 6.1 29/71] fs/ntfs3: Use __GFP_NOWARN allocation at wnd_init() Greg Kroah-Hartman
2023-01-02 11:21 ` [PATCH 6.1 30/71] fs/ntfs3: Use __GFP_NOWARN allocation at ntfs_fill_super() Greg Kroah-Hartman
2023-01-02 11:21 ` [PATCH 6.1 31/71] fs/ntfs3: Delete duplicate condition in ntfs_read_mft() Greg Kroah-Hartman
2023-01-02 11:21 ` [PATCH 6.1 32/71] fs/ntfs3: Fix slab-out-of-bounds in r_page Greg Kroah-Hartman
2023-01-02 11:21 ` [PATCH 6.1 33/71] objtool: Fix SEGFAULT Greg Kroah-Hartman
2023-01-02 11:21 ` [PATCH 6.1 34/71] iommu/mediatek: Fix crash on isr after kexec() Greg Kroah-Hartman
2023-01-02 11:22 ` [PATCH 6.1 35/71] powerpc/rtas: avoid device tree lookups in rtas_os_term() Greg Kroah-Hartman
2023-01-02 11:22 ` [PATCH 6.1 36/71] powerpc/rtas: avoid scheduling " Greg Kroah-Hartman
2023-01-02 11:22 ` [PATCH 6.1 37/71] rtc: msc313: Fix function prototype mismatch in msc313_rtc_probe() Greg Kroah-Hartman
2023-01-02 11:22 ` [PATCH 6.1 38/71] NFSD: fix use-after-free in __nfs42_ssc_open() Greg Kroah-Hartman
2023-01-02 11:22 ` [PATCH 6.1 39/71] kprobes: kretprobe events missing on 2-core KVM guest Greg Kroah-Hartman
2023-01-02 11:22 ` [PATCH 6.1 40/71] HID: multitouch: fix Asus ExpertBook P2 P2451FA trackpoint Greg Kroah-Hartman
2023-01-02 11:22 ` [PATCH 6.1 41/71] HID: plantronics: Additional PIDs for double volume key presses quirk Greg Kroah-Hartman
2023-01-02 11:22 ` [PATCH 6.1 42/71] futex: Fix futex_waitv() hrtimer debug object leak on kcalloc error Greg Kroah-Hartman
2023-01-02 11:22 ` [PATCH 6.1 43/71] rtmutex: Add acquire semantics for rtmutex lock acquisition slow path Greg Kroah-Hartman
2023-01-02 11:22 ` [PATCH 6.1 44/71] mm, mremap: fix mremap() expanding vma with addr inside vma Greg Kroah-Hartman
2023-01-02 11:22 ` [PATCH 6.1 45/71] mm/mempolicy: fix memory leak in set_mempolicy_home_node system call Greg Kroah-Hartman
2023-01-02 11:22 ` [PATCH 6.1 46/71] kmsan: export kmsan_handle_urb Greg Kroah-Hartman
2023-01-02 11:22 ` [PATCH 6.1 47/71] kmsan: include linux/vmalloc.h Greg Kroah-Hartman
2023-01-02 11:22 ` [PATCH 6.1 48/71] pstore: Properly assign mem_type property Greg Kroah-Hartman
2023-01-02 11:22 ` [PATCH 6.1 49/71] pstore/zone: Use GFP_ATOMIC to allocate zone buffer Greg Kroah-Hartman
2023-01-02 11:22 ` [PATCH 6.1 50/71] hfsplus: fix bug causing custom uid and gid being unable to be assigned with mount Greg Kroah-Hartman
2023-01-02 11:22 ` [PATCH 6.1 51/71] ACPI: x86: s2idle: Force AMD GUID/_REV 2 on HP Elitebook 865 Greg Kroah-Hartman
2023-01-02 11:22 ` [PATCH 6.1 52/71] ACPI: x86: s2idle: Stop using AMD specific codepath for Rembrandt+ Greg Kroah-Hartman
2023-01-02 11:22 ` [PATCH 6.1 53/71] binfmt: Fix error return code in load_elf_fdpic_binary() Greg Kroah-Hartman
2023-01-02 11:22 ` Greg Kroah-Hartman [this message]
2023-01-02 11:22 ` [PATCH 6.1 55/71] ovl: update ->f_iocb_flags when ovl_change_flags() modifies ->f_flags Greg Kroah-Hartman
2023-01-02 11:22 ` [PATCH 6.1 56/71] ALSA: line6: correct midi status byte when receiving data from podxt Greg Kroah-Hartman
2023-01-02 11:22 ` [PATCH 6.1 57/71] ALSA: line6: fix stack overflow in line6_midi_transmit Greg Kroah-Hartman
2023-01-02 11:22 ` [PATCH 6.1 58/71] ALSA: hda/hdmi: Static PCM mapping again with AMD HDMI codecs Greg Kroah-Hartman
2023-01-02 11:22 ` [PATCH 6.1 59/71] pnode: terminate at peers of source Greg Kroah-Hartman
2023-01-02 11:22 ` [PATCH 6.1 60/71] mfd: mt6360: Add bounds checking in Regmap read/write call-backs Greg Kroah-Hartman
2023-01-02 11:22 ` [PATCH 6.1 61/71] md: fix a crash in mempool_free Greg Kroah-Hartman
2023-01-02 11:22 ` [PATCH 6.1 62/71] mm, compaction: fix fast_isolate_around() to stay within boundaries Greg Kroah-Hartman
2023-01-02 11:22 ` [PATCH 6.1 63/71] f2fs: should put a page when checking the summary info Greg Kroah-Hartman
2023-01-02 11:22 ` [PATCH 6.1 64/71] f2fs: allow to read node block after shutdown Greg Kroah-Hartman
2023-01-02 11:22 ` [PATCH 6.1 65/71] block: Do not reread partition table on exclusively open device Greg Kroah-Hartman
2023-01-02 11:22 ` [PATCH 6.1 66/71] mmc: vub300: fix warning - do not call blocking ops when !TASK_RUNNING Greg Kroah-Hartman
2023-01-02 11:22 ` [PATCH 6.1 67/71] tpm: acpi: Call acpi_put_table() to fix memory leak Greg Kroah-Hartman
2023-01-02 11:22 ` [PATCH 6.1 68/71] tpm: tpm_crb: Add the missed " Greg Kroah-Hartman
2023-01-02 11:22 ` [PATCH 6.1 69/71] tpm: tpm_tis: " Greg Kroah-Hartman
2023-01-02 11:22 ` [PATCH 6.1 70/71] SUNRPC: Dont leak netobj memory when gss_read_proxy_verf() fails Greg Kroah-Hartman
2023-01-02 11:22 ` [PATCH 6.1 71/71] kcsan: Instrument memcpy/memset/memmove with newer Clang Greg Kroah-Hartman
2023-01-02 23:14 ` [PATCH 6.1 00/71] 6.1.3-rc1 review Rudi Heitbaum
2023-01-03 0:25 ` Shuah Khan
2023-01-03 1:13 ` Guenter Roeck
2023-01-03 8:40 ` Naresh Kamboju
2023-01-03 8:45 ` Naresh Kamboju
2023-01-03 8:59 ` Ron Economos
2023-01-03 10:34 ` Sudip Mukherjee (Codethink)
2023-01-03 12:08 ` Bagas Sanjaya
2023-01-03 13:22 ` Allen Pais
2023-01-03 15:33 ` Jon Hunter
2023-01-03 19:33 ` Florian Fainelli
2023-01-04 1:39 ` Justin Forbes
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=20230102110553.758530779@linuxfoundation.org \
--to=gregkh@linuxfoundation.org \
--cc=brauner@kernel.org \
--cc=mszeredi@redhat.com \
--cc=patches@lists.linux.dev \
--cc=stable@vger.kernel.org \
--cc=zhangjiachen.jaycee@bytedance.com \
--cc=zhangtianci.1997@bytedance.com \
/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