All of lore.kernel.org
 help / color / mirror / Atom feed
From: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
To: linux-kernel@vger.kernel.org
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
	stable@vger.kernel.org, Yue Cui <cuiyue-fnst@fujitsu.com>,
	Anna Schumaker <Anna.Schumaker@Netapp.com>,
	Trond Myklebust <trond.myklebust@hammerspace.com>,
	Sasha Levin <sashal@kernel.org>
Subject: [PATCH 5.19 09/38] NFSv4.2: Update mode bits after ALLOCATE and DEALLOCATE
Date: Wed, 21 Sep 2022 17:45:53 +0200	[thread overview]
Message-ID: <20220921153646.597985540@linuxfoundation.org> (raw)
In-Reply-To: <20220921153646.298361220@linuxfoundation.org>

From: Anna Schumaker <Anna.Schumaker@Netapp.com>

[ Upstream commit d7a5118635e725d195843bda80cc5c964d93ef31 ]

The fallocate call invalidates suid and sgid bits as part of normal
operation. We need to mark the mode bits as invalid when using fallocate
with an suid so these will be updated the next time the user looks at them.

This fixes xfstests generic/683 and generic/684.

Reported-by: Yue Cui <cuiyue-fnst@fujitsu.com>
Fixes: 913eca1aea87 ("NFS: Fallocate should use the nfs4_fattr_bitmap")
Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
 fs/nfs/internal.h  | 25 +++++++++++++++++++++++++
 fs/nfs/nfs42proc.c |  9 +++++++--
 fs/nfs/write.c     | 25 -------------------------
 3 files changed, 32 insertions(+), 27 deletions(-)

diff --git a/fs/nfs/internal.h b/fs/nfs/internal.h
index 8f8cd6e2d4db..597e3ce3f148 100644
--- a/fs/nfs/internal.h
+++ b/fs/nfs/internal.h
@@ -604,6 +604,31 @@ static inline gfp_t nfs_io_gfp_mask(void)
 	return GFP_KERNEL;
 }
 
+/*
+ * Special version of should_remove_suid() that ignores capabilities.
+ */
+static inline int nfs_should_remove_suid(const struct inode *inode)
+{
+	umode_t mode = inode->i_mode;
+	int kill = 0;
+
+	/* suid always must be killed */
+	if (unlikely(mode & S_ISUID))
+		kill = ATTR_KILL_SUID;
+
+	/*
+	 * sgid without any exec bits is just a mandatory locking mark; leave
+	 * it alone.  If some exec bits are set, it's a real sgid; kill it.
+	 */
+	if (unlikely((mode & S_ISGID) && (mode & S_IXGRP)))
+		kill |= ATTR_KILL_SGID;
+
+	if (unlikely(kill && S_ISREG(mode)))
+		return kill;
+
+	return 0;
+}
+
 /* unlink.c */
 extern struct rpc_task *
 nfs_async_rename(struct inode *old_dir, struct inode *new_dir,
diff --git a/fs/nfs/nfs42proc.c b/fs/nfs/nfs42proc.c
index 068c45b3bc1a..6dab9e408372 100644
--- a/fs/nfs/nfs42proc.c
+++ b/fs/nfs/nfs42proc.c
@@ -78,10 +78,15 @@ static int _nfs42_proc_fallocate(struct rpc_message *msg, struct file *filep,
 
 	status = nfs4_call_sync(server->client, server, msg,
 				&args.seq_args, &res.seq_res, 0);
-	if (status == 0)
+	if (status == 0) {
+		if (nfs_should_remove_suid(inode)) {
+			spin_lock(&inode->i_lock);
+			nfs_set_cache_invalid(inode, NFS_INO_INVALID_MODE);
+			spin_unlock(&inode->i_lock);
+		}
 		status = nfs_post_op_update_inode_force_wcc(inode,
 							    res.falloc_fattr);
-
+	}
 	if (msg->rpc_proc == &nfs4_procedures[NFSPROC4_CLNT_ALLOCATE])
 		trace_nfs4_fallocate(inode, &args, status);
 	else
diff --git a/fs/nfs/write.c b/fs/nfs/write.c
index 5d7e1c206184..4212473c69ee 100644
--- a/fs/nfs/write.c
+++ b/fs/nfs/write.c
@@ -1497,31 +1497,6 @@ void nfs_commit_prepare(struct rpc_task *task, void *calldata)
 	NFS_PROTO(data->inode)->commit_rpc_prepare(task, data);
 }
 
-/*
- * Special version of should_remove_suid() that ignores capabilities.
- */
-static int nfs_should_remove_suid(const struct inode *inode)
-{
-	umode_t mode = inode->i_mode;
-	int kill = 0;
-
-	/* suid always must be killed */
-	if (unlikely(mode & S_ISUID))
-		kill = ATTR_KILL_SUID;
-
-	/*
-	 * sgid without any exec bits is just a mandatory locking mark; leave
-	 * it alone.  If some exec bits are set, it's a real sgid; kill it.
-	 */
-	if (unlikely((mode & S_ISGID) && (mode & S_IXGRP)))
-		kill |= ATTR_KILL_SGID;
-
-	if (unlikely(kill && S_ISREG(mode)))
-		return kill;
-
-	return 0;
-}
-
 static void nfs_writeback_check_extend(struct nfs_pgio_header *hdr,
 		struct nfs_fattr *fattr)
 {
-- 
2.35.1




  parent reply	other threads:[~2022-09-21 15:48 UTC|newest]

Thread overview: 42+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2022-09-21 15:45 [PATCH 5.19 00/38] 5.19.11-rc1 review Greg Kroah-Hartman
2022-09-21 15:45 ` [PATCH 5.19 01/38] of: fdt: fix off-by-one error in unflatten_dt_nodes() Greg Kroah-Hartman
2022-09-21 15:45 ` [PATCH 5.19 02/38] pinctrl: qcom: sc8180x: Fix gpio_wakeirq_map Greg Kroah-Hartman
2022-09-21 15:45 ` [PATCH 5.19 03/38] pinctrl: qcom: sc8180x: Fix wrong pin numbers Greg Kroah-Hartman
2022-09-21 15:45 ` [PATCH 5.19 04/38] pinctrl: rockchip: Enhance support for IRQ_TYPE_EDGE_BOTH Greg Kroah-Hartman
2022-09-21 15:45 ` [PATCH 5.19 05/38] pinctrl: sunxi: Fix name for A100 R_PIO Greg Kroah-Hartman
2022-09-21 15:45 ` [PATCH 5.19 06/38] SUNRPC: Fix call completion races with call_decode() Greg Kroah-Hartman
2022-09-21 15:45 ` [PATCH 5.19 07/38] NFSv4: Turn off open-by-filehandle and NFS re-export for NFSv4.0 Greg Kroah-Hartman
2022-09-21 15:45 ` [PATCH 5.19 08/38] gpio: mpc8xxx: Fix support for IRQ_TYPE_LEVEL_LOW flow_type in mpc85xx Greg Kroah-Hartman
2022-09-21 15:45 ` Greg Kroah-Hartman [this message]
2022-09-21 15:45 ` [PATCH 5.19 10/38] Revert "SUNRPC: Remove unreachable error condition" Greg Kroah-Hartman
2022-09-21 15:45 ` [PATCH 5.19 11/38] drm/panel-edp: Fix delays for Innolux N116BCA-EA1 Greg Kroah-Hartman
2022-09-21 15:45 ` [PATCH 5.19 12/38] drm/meson: Correct OSD1 global alpha value Greg Kroah-Hartman
2022-09-21 15:45 ` [PATCH 5.19 13/38] drm/meson: Fix OSD1 RGB to YCbCr coefficient Greg Kroah-Hartman
2022-09-21 15:45 ` [PATCH 5.19 14/38] drm/rockchip: vop2: Fix eDP/HDMI sync polarities Greg Kroah-Hartman
2022-09-21 15:45 ` [PATCH 5.19 15/38] drm/i915/vdsc: Set VDSC PIC_HEIGHT before using for DP DSC Greg Kroah-Hartman
2022-09-21 15:46 ` [PATCH 5.19 16/38] drm/i915/guc: Dont update engine busyness stats too frequently Greg Kroah-Hartman
2022-09-21 15:46 ` [PATCH 5.19 17/38] drm/i915/guc: Cancel GuC engine busyness worker synchronously Greg Kroah-Hartman
2022-09-21 15:46 ` [PATCH 5.19 18/38] block: blk_queue_enter() / __bio_queue_enter() must return -EAGAIN for nowait Greg Kroah-Hartman
2022-09-21 15:46 ` [PATCH 5.19 19/38] parisc: ccio-dma: Add missing iounmap in error path in ccio_probe() Greg Kroah-Hartman
2022-09-21 15:46 ` [PATCH 5.19 20/38] of/device: Fix up of_dma_configure_id() stub Greg Kroah-Hartman
2022-09-21 15:46 ` [PATCH 5.19 21/38] io_uring/msg_ring: check file type before putting Greg Kroah-Hartman
2022-09-21 15:46 ` [PATCH 5.19 22/38] cifs: revalidate mapping when doing direct writes Greg Kroah-Hartman
2022-09-21 15:46 ` [PATCH 5.19 23/38] cifs: dont send down the destination address to sendmsg for a SOCK_STREAM Greg Kroah-Hartman
2022-09-21 15:46 ` [PATCH 5.19 24/38] cifs: always initialize struct msghdr smb_msg completely Greg Kroah-Hartman
2022-09-21 15:46 ` [PATCH 5.19 25/38] blk-lib: fix blkdev_issue_secure_erase Greg Kroah-Hartman
2022-09-21 15:46 ` [PATCH 5.19 26/38] parisc: Allow CONFIG_64BIT with ARCH=parisc Greg Kroah-Hartman
2022-09-21 15:46 ` [PATCH 5.19 27/38] tools/include/uapi: Fix <asm/errno.h> for parisc and xtensa Greg Kroah-Hartman
2022-09-21 15:46 ` [PATCH 5.19 28/38] drm/i915/gt: Fix perf limit reasons bit positions Greg Kroah-Hartman
2022-09-21 15:46 ` [PATCH 5.19 29/38] drm/i915: Set correct domains values at _i915_vma_move_to_active Greg Kroah-Hartman
2022-09-21 15:46 ` [PATCH 5.19 30/38] drm/amdgpu: make sure to init common IP before gmc Greg Kroah-Hartman
2022-09-21 15:46 ` [PATCH 5.19 31/38] drm/amdgpu: Dont enable LTR if not supported Greg Kroah-Hartman
2022-09-21 15:46 ` [PATCH 5.19 32/38] drm/amdgpu: move nbio ih_doorbell_range() into ih code for vega Greg Kroah-Hartman
2022-09-21 15:46 ` [PATCH 5.19 33/38] drm/amdgpu: move nbio sdma_doorbell_range() into sdma " Greg Kroah-Hartman
2022-09-21 15:46 ` [PATCH 5.19 34/38] net: Find dst with sks xfrm policy not ctl_sk Greg Kroah-Hartman
2022-09-21 15:46 ` [PATCH 5.19 35/38] dt-bindings: apple,aic: Fix required item "apple,fiq-index" in affinity description Greg Kroah-Hartman
2022-09-21 15:46 ` [PATCH 5.19 36/38] cgroup: Add missing cpus_read_lock() to cgroup_attach_task_all() Greg Kroah-Hartman
2022-09-21 15:46 ` [PATCH 5.19 37/38] ALSA: hda/sigmatel: Keep power up while beep is enabled Greg Kroah-Hartman
2022-09-21 15:46 ` [PATCH 5.19 38/38] ALSA: hda/sigmatel: Fix unused variable warning for beep power change Greg Kroah-Hartman
2022-09-22  7:11 ` [PATCH 5.19 00/38] 5.19.11-rc1 review Bagas Sanjaya
2022-09-22  7:14   ` Bagas Sanjaya
2022-09-22 16:42 ` Guenter Roeck

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=20220921153646.597985540@linuxfoundation.org \
    --to=gregkh@linuxfoundation.org \
    --cc=Anna.Schumaker@Netapp.com \
    --cc=cuiyue-fnst@fujitsu.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=sashal@kernel.org \
    --cc=stable@vger.kernel.org \
    --cc=trond.myklebust@hammerspace.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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.