stable.vger.kernel.org archive mirror
 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, Scott Mayhew <smayhew@redhat.com>,
	Trond Myklebust <trond.myklebust@primarydata.com>
Subject: [PATCH 4.9 68/86] NFS: Fix unstable write completion
Date: Fri, 16 Mar 2018 16:23:31 +0100	[thread overview]
Message-ID: <20180316152321.959682435@linuxfoundation.org> (raw)
In-Reply-To: <20180316152317.167709497@linuxfoundation.org>

4.9-stable review patch.  If anyone has any objections, please let me know.

------------------

From: Trond Myklebust <trond.myklebust@primarydata.com>

commit c4f24df942a181699c5bab01b8e5e82b925f77f3 upstream.

We do want to respect the FLUSH_SYNC argument to nfs_commit_inode() to
ensure that all outstanding COMMIT requests to the inode in question are
complete. Currently we may exit early from both nfs_commit_inode() and
nfs_write_inode() even if there are COMMIT requests in flight, or unstable
writes on the commit list.

In order to get the right semantics w.r.t. sync_inode(), we don't need
to have nfs_commit_inode() reset the inode dirty flags when called from
nfs_wb_page() and/or nfs_wb_all(). We just need to ensure that
nfs_write_inode() leaves them in the right state if there are outstanding
commits, or stable pages.

Reported-by: Scott Mayhew <smayhew@redhat.com>
Fixes: dc4fd9ab01ab ("nfs: don't wait on commit in nfs_commit_inode()...")
Cc: stable@vger.kernel.org # v4.14+
Signed-off-by: Trond Myklebust <trond.myklebust@primarydata.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

---
 fs/nfs/write.c |   83 +++++++++++++++++++++++++++++----------------------------
 1 file changed, 43 insertions(+), 40 deletions(-)

--- a/fs/nfs/write.c
+++ b/fs/nfs/write.c
@@ -1847,40 +1847,43 @@ int nfs_generic_commit_list(struct inode
 	return status;
 }
 
-int nfs_commit_inode(struct inode *inode, int how)
+static int __nfs_commit_inode(struct inode *inode, int how,
+		struct writeback_control *wbc)
 {
 	LIST_HEAD(head);
 	struct nfs_commit_info cinfo;
 	int may_wait = how & FLUSH_SYNC;
-	int error = 0;
-	int res;
+	int ret, nscan;
 
 	nfs_init_cinfo_from_inode(&cinfo, inode);
 	nfs_commit_begin(cinfo.mds);
-	res = nfs_scan_commit(inode, &head, &cinfo);
-	if (res)
-		error = nfs_generic_commit_list(inode, &head, how, &cinfo);
+	for (;;) {
+		ret = nscan = nfs_scan_commit(inode, &head, &cinfo);
+		if (ret <= 0)
+			break;
+		ret = nfs_generic_commit_list(inode, &head, how, &cinfo);
+		if (ret < 0)
+			break;
+		ret = 0;
+		if (wbc && wbc->sync_mode == WB_SYNC_NONE) {
+			if (nscan < wbc->nr_to_write)
+				wbc->nr_to_write -= nscan;
+			else
+				wbc->nr_to_write = 0;
+		}
+		if (nscan < INT_MAX)
+			break;
+		cond_resched();
+	}
 	nfs_commit_end(cinfo.mds);
-	if (res == 0)
-		return res;
-	if (error < 0)
-		goto out_error;
-	if (!may_wait)
-		goto out_mark_dirty;
-	error = wait_on_commit(cinfo.mds);
-	if (error < 0)
-		return error;
-	return res;
-out_error:
-	res = error;
-	/* Note: If we exit without ensuring that the commit is complete,
-	 * we must mark the inode as dirty. Otherwise, future calls to
-	 * sync_inode() with the WB_SYNC_ALL flag set will fail to ensure
-	 * that the data is on the disk.
-	 */
-out_mark_dirty:
-	__mark_inode_dirty(inode, I_DIRTY_DATASYNC);
-	return res;
+	if (ret || !may_wait)
+		return ret;
+	return wait_on_commit(cinfo.mds);
+}
+
+int nfs_commit_inode(struct inode *inode, int how)
+{
+	return __nfs_commit_inode(inode, how, NULL);
 }
 EXPORT_SYMBOL_GPL(nfs_commit_inode);
 
@@ -1890,11 +1893,11 @@ int nfs_write_inode(struct inode *inode,
 	int flags = FLUSH_SYNC;
 	int ret = 0;
 
-	/* no commits means nothing needs to be done */
-	if (!nfsi->commit_info.ncommit)
-		return ret;
-
 	if (wbc->sync_mode == WB_SYNC_NONE) {
+		/* no commits means nothing needs to be done */
+		if (!nfsi->commit_info.ncommit)
+			goto check_requests_outstanding;
+
 		/* Don't commit yet if this is a non-blocking flush and there
 		 * are a lot of outstanding writes for this mapping.
 		 */
@@ -1905,16 +1908,16 @@ int nfs_write_inode(struct inode *inode,
 		flags = 0;
 	}
 
-	ret = nfs_commit_inode(inode, flags);
-	if (ret >= 0) {
-		if (wbc->sync_mode == WB_SYNC_NONE) {
-			if (ret < wbc->nr_to_write)
-				wbc->nr_to_write -= ret;
-			else
-				wbc->nr_to_write = 0;
-		}
-		return 0;
-	}
+	ret = __nfs_commit_inode(inode, flags, wbc);
+	if (!ret) {
+		if (flags & FLUSH_SYNC)
+			return 0;
+	} else if (nfsi->commit_info.ncommit)
+		goto out_mark_dirty;
+
+check_requests_outstanding:
+	if (!atomic_read(&nfsi->commit_info.rpcs_out))
+		return ret;
 out_mark_dirty:
 	__mark_inode_dirty(inode, I_DIRTY_DATASYNC);
 	return ret;

  parent reply	other threads:[~2018-03-16 15:23 UTC|newest]

Thread overview: 88+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2018-03-16 15:22 [PATCH 4.9 00/86] 4.9.88-stable review Greg Kroah-Hartman
2018-03-16 15:22 ` [PATCH 4.9 01/86] RDMA/ucma: Limit possible option size Greg Kroah-Hartman
2018-03-16 15:22 ` [PATCH 4.9 02/86] RDMA/ucma: Check that user doesnt overflow QP state Greg Kroah-Hartman
2018-03-16 15:22 ` [PATCH 4.9 03/86] RDMA/mlx5: Fix integer overflow while resizing CQ Greg Kroah-Hartman
2018-03-16 15:22 ` [PATCH 4.9 05/86] scsi: qla2xxx: Fix NULL pointer crash due to active timer for ABTS Greg Kroah-Hartman
2018-03-16 15:22 ` [PATCH 4.9 06/86] drm/i915: Always call to intel_display_set_init_power() in resume_early Greg Kroah-Hartman
2018-03-16 15:22 ` [PATCH 4.9 07/86] workqueue: Allow retrieval of current tasks work struct Greg Kroah-Hartman
2018-03-16 15:22 ` [PATCH 4.9 08/86] drm: Allow determining if current task is output poll worker Greg Kroah-Hartman
2018-03-16 15:22 ` [PATCH 4.9 09/86] drm/nouveau: Fix deadlock on runtime suspend Greg Kroah-Hartman
2018-03-16 15:22 ` [PATCH 4.9 10/86] drm/radeon: " Greg Kroah-Hartman
2018-03-16 15:22 ` [PATCH 4.9 11/86] drm/amdgpu: " Greg Kroah-Hartman
2018-03-16 15:22 ` [PATCH 4.9 12/86] drm/amdgpu: Notify sbios device ready before send request Greg Kroah-Hartman
2018-03-16 15:22 ` [PATCH 4.9 17/86] MIPS: BMIPS: Do not mask IPIs during suspend Greg Kroah-Hartman
2018-03-16 15:22 ` [PATCH 4.9 18/86] MIPS: ath25: Check for kzalloc allocation failure Greg Kroah-Hartman
2018-03-16 15:22 ` [PATCH 4.9 19/86] MIPS: OCTEON: irq: Check for null return on kzalloc allocation Greg Kroah-Hartman
2018-03-16 15:22 ` [PATCH 4.9 20/86] Input: matrix_keypad - fix race when disabling interrupts Greg Kroah-Hartman
2018-03-16 15:22 ` [PATCH 4.9 21/86] loop: Fix lost writes caused by missing flag Greg Kroah-Hartman
2018-03-16 15:22 ` [PATCH 4.9 22/86] virtio_ring: fix num_free handling in error case Greg Kroah-Hartman
2018-03-16 15:22 ` [PATCH 4.9 23/86] KVM: s390: fix memory overwrites when not using SCA entries Greg Kroah-Hartman
2018-03-16 15:22 ` [PATCH 4.9 24/86] kbuild: Handle builtin dtb file names containing hyphens Greg Kroah-Hartman
2018-03-16 15:22 ` [PATCH 4.9 25/86] IB/mlx5: Fix incorrect size of klms in the memory region Greg Kroah-Hartman
2018-03-16 15:22 ` [PATCH 4.9 26/86] bcache: fix crashes in duplicate cache device register Greg Kroah-Hartman
2018-03-16 15:22 ` [PATCH 4.9 27/86] bcache: dont attach backing with duplicate UUID Greg Kroah-Hartman
2018-03-16 15:22 ` [PATCH 4.9 28/86] x86/MCE: Serialize sysfs changes Greg Kroah-Hartman
2018-03-16 15:22 ` [PATCH 4.9 29/86] perf tools: Fix trigger class trigger_on() Greg Kroah-Hartman
2018-03-16 15:22 ` [PATCH 4.9 31/86] ALSA: hda/realtek: Limit mic boost on T480 Greg Kroah-Hartman
2018-03-16 15:22 ` [PATCH 4.9 32/86] ALSA: hda/realtek - Fix dock line-out volume on Dell Precision 7520 Greg Kroah-Hartman
2018-03-16 15:22 ` [PATCH 4.9 33/86] ALSA: hda/realtek - Make dock sound work on ThinkPad L570 Greg Kroah-Hartman
2018-03-16 15:22 ` [PATCH 4.9 36/86] ALSA: hda: add dock and led support for HP EliteBook 820 G3 Greg Kroah-Hartman
2018-03-16 15:23 ` [PATCH 4.9 37/86] ALSA: hda: add dock and led support for HP ProBook 640 G2 Greg Kroah-Hartman
2018-03-16 15:23 ` [PATCH 4.9 38/86] nospec: Kill array_index_nospec_mask_check() Greg Kroah-Hartman
2018-03-16 15:23 ` [PATCH 4.9 39/86] nospec: Include <asm/barrier.h> dependency Greg Kroah-Hartman
2018-03-16 15:23 ` [PATCH 4.9 40/86] Revert "x86/retpoline: Simplify vmexit_fill_RSB()" Greg Kroah-Hartman
2018-03-16 15:23 ` [PATCH 4.9 41/86] x86/speculation: Use IBRS if available before calling into firmware Greg Kroah-Hartman
2018-03-16 15:23 ` [PATCH 4.9 42/86] x86/retpoline: Support retpoline builds with Clang Greg Kroah-Hartman
2018-03-16 15:23 ` [PATCH 4.9 43/86] x86/speculation, objtool: Annotate indirect calls/jumps for objtool Greg Kroah-Hartman
2018-03-16 15:23 ` [PATCH 4.9 44/86] x86/boot, objtool: Annotate indirect jump in secondary_startup_64() Greg Kroah-Hartman
2018-03-16 15:23 ` [PATCH 4.9 45/86] x86/speculation: Move firmware_restrict_branch_speculation_*() from C to CPP Greg Kroah-Hartman
2018-03-16 15:23 ` [PATCH 4.9 46/86] x86/paravirt, objtool: Annotate indirect calls Greg Kroah-Hartman
2018-03-16 15:23 ` [PATCH 4.9 47/86] watchdog: hpwdt: SMBIOS check Greg Kroah-Hartman
2018-03-16 15:23 ` [PATCH 4.9 48/86] watchdog: hpwdt: Check source of NMI Greg Kroah-Hartman
2018-03-16 15:23 ` [PATCH 4.9 49/86] watchdog: hpwdt: fix unused variable warning Greg Kroah-Hartman
2018-03-16 15:23 ` [PATCH 4.9 50/86] watchdog: hpwdt: Remove legacy NMI sourcing Greg Kroah-Hartman
2018-03-16 15:23 ` [PATCH 4.9 51/86] ARM: omap2: hide omap3_save_secure_ram on non-OMAP3 builds Greg Kroah-Hartman
2018-03-16 15:23 ` [PATCH 4.9 52/86] Input: tca8418_keypad - remove double read of key event register Greg Kroah-Hartman
2018-03-16 15:23 ` [PATCH 4.9 53/86] [media] tc358743: fix register i2c_rd/wr function fix Greg Kroah-Hartman
2018-03-16 15:23 ` [PATCH 4.9 54/86] netfilter: add back stackpointer size checks Greg Kroah-Hartman
2018-03-16 15:23 ` [PATCH 4.9 55/86] netfilter: x_tables: fix missing timer initialization in xt_LED Greg Kroah-Hartman
2018-03-16 15:23 ` [PATCH 4.9 56/86] netfilter: nat: cope with negative port range Greg Kroah-Hartman
2018-03-16 15:23 ` [PATCH 4.9 57/86] netfilter: IDLETIMER: be syzkaller friendly Greg Kroah-Hartman
2018-03-16 15:23 ` [PATCH 4.9 58/86] netfilter: ebtables: CONFIG_COMPAT: dont trust userland offsets Greg Kroah-Hartman
2018-03-16 15:23 ` [PATCH 4.9 59/86] netfilter: bridge: ebt_among: add missing match size checks Greg Kroah-Hartman
2018-03-16 15:23 ` [PATCH 4.9 60/86] netfilter: ipv6: fix use-after-free Write in nf_nat_ipv6_manip_pkt Greg Kroah-Hartman
2018-03-16 15:23 ` [PATCH 4.9 61/86] netfilter: x_tables: pass xt_counters struct instead of packet counter Greg Kroah-Hartman
2018-03-16 15:23 ` [PATCH 4.9 62/86] netfilter: x_tables: pass xt_counters struct to counter allocator Greg Kroah-Hartman
2018-03-16 15:23 ` [PATCH 4.9 63/86] netfilter: x_tables: pack percpu counter allocations Greg Kroah-Hartman
2018-03-16 15:23 ` [PATCH 4.9 64/86] ext4: inplace xattr block update fails to deduplicate blocks Greg Kroah-Hartman
2018-03-16 15:23 ` [PATCH 4.9 65/86] ubi: Fix race condition between ubi volume creation and udev Greg Kroah-Hartman
2018-03-16 15:23 ` [PATCH 4.9 66/86] scsi: qla2xxx: Replace fcport alloc with qla2x00_alloc_fcport Greg Kroah-Hartman
2018-03-16 15:23 ` [PATCH 4.9 67/86] NFS: Fix an incorrect type in struct nfs_direct_req Greg Kroah-Hartman
2018-03-16 15:23 ` Greg Kroah-Hartman [this message]
2018-03-16 15:23 ` [PATCH 4.9 69/86] x86/module: Detect and skip invalid relocations Greg Kroah-Hartman
2018-03-16 15:23 ` [PATCH 4.9 70/86] x86: Treat R_X86_64_PLT32 as R_X86_64_PC32 Greg Kroah-Hartman
2018-03-16 15:23 ` [PATCH 4.9 71/86] ASoC: sgtl5000: Fix suspend/resume Greg Kroah-Hartman
2018-03-16 15:23 ` [PATCH 4.9 72/86] ASoC: rt5651: Fix regcache sync errors on resume Greg Kroah-Hartman
2018-03-16 15:23 ` [PATCH 4.9 73/86] serial: sh-sci: prevent lockup on full TTY buffers Greg Kroah-Hartman
2018-03-16 15:23 ` [PATCH 4.9 74/86] tty/serial: atmel: add new version check for usart Greg Kroah-Hartman
2018-03-16 15:23 ` [PATCH 4.9 75/86] uas: fix comparison for error code Greg Kroah-Hartman
2018-03-16 15:23 ` [PATCH 4.9 76/86] staging: comedi: fix comedi_nsamples_left Greg Kroah-Hartman
2018-03-16 15:23 ` [PATCH 4.9 77/86] staging: android: ashmem: Fix lockdep issue during llseek Greg Kroah-Hartman
2018-03-16 15:23 ` [PATCH 4.9 78/86] USB: storage: Add JMicron bridge 152d:2567 to unusual_devs.h Greg Kroah-Hartman
2018-03-16 15:23 ` [PATCH 4.9 79/86] usbip: vudc: fix null pointer dereference on udc->lock Greg Kroah-Hartman
2018-03-16 15:23 ` [PATCH 4.9 80/86] usb: quirks: add control message delay for 1b1c:1b20 Greg Kroah-Hartman
2018-03-16 15:23 ` [PATCH 4.9 81/86] usb: usbmon: Read text within supplied buffer size Greg Kroah-Hartman
2018-03-16 15:23 ` [PATCH 4.9 82/86] usb: gadget: f_fs: Fix use-after-free in ffs_fs_kill_sb() Greg Kroah-Hartman
2018-03-16 15:23 ` [PATCH 4.9 83/86] serial: 8250_pci: Add Brainboxes UC-260 4 port serial device Greg Kroah-Hartman
2018-03-16 15:23 ` [PATCH 4.9 84/86] serial: core: mark port as initialized in autoconfig Greg Kroah-Hartman
2018-03-16 15:23 ` [PATCH 4.9 85/86] earlycon: add reg-offset to physical address before mapping Greg Kroah-Hartman
2018-03-16 15:23 ` [PATCH 4.9 86/86] PCI: dwc: Fix enumeration end when reaching root subordinate Greg Kroah-Hartman
2018-03-17 10:18 ` [PATCH 4.9 00/86] 4.9.88-stable review Naresh Kamboju
2018-03-18 10:27   ` Greg Kroah-Hartman
2018-03-20 23:49     ` Ben Hutchings
2018-03-21 13:32       ` Greg Kroah-Hartman
2018-03-21 17:50         ` Naresh Kamboju
2018-03-22  8:19           ` Greg Kroah-Hartman
2018-03-22 17:47             ` Naresh Kamboju
2018-03-17 14:41 ` Guenter Roeck
2018-03-18 10:27   ` Greg Kroah-Hartman

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=20180316152321.959682435@linuxfoundation.org \
    --to=gregkh@linuxfoundation.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=smayhew@redhat.com \
    --cc=stable@vger.kernel.org \
    --cc=trond.myklebust@primarydata.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;
as well as URLs for NNTP newsgroup(s).