From: Sasha Levin <sashal@kernel.org>
To: linux-kernel@vger.kernel.org, stable@vger.kernel.org
Cc: Trond Myklebust <trond.myklebust@hammerspace.com>,
Sasha Levin <sashal@kernel.org>,
linux-nfs@vger.kernel.org
Subject: [PATCH AUTOSEL 5.10 09/25] pNFS/NFSv4: Improve rejection of out-of-order layouts
Date: Tue, 2 Feb 2021 10:05:59 -0500 [thread overview]
Message-ID: <20210202150615.1864175-9-sashal@kernel.org> (raw)
In-Reply-To: <20210202150615.1864175-1-sashal@kernel.org>
From: Trond Myklebust <trond.myklebust@hammerspace.com>
[ Upstream commit d29b468da4f940bd2bff2628ba8d2d652671d244 ]
If a layoutget ends up being reordered w.r.t. a layoutreturn, e.g. due
to a layoutget-on-open not knowing a priori which file to lock, then we
must assume the layout is no longer being considered valid state by the
server.
Incrementally improve our ability to reject such states by using the
cached old stateid in conjunction with the plh_barrier to try to
identify them.
Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
fs/nfs/pnfs.c | 22 ++++++++++++++++------
1 file changed, 16 insertions(+), 6 deletions(-)
diff --git a/fs/nfs/pnfs.c b/fs/nfs/pnfs.c
index cbe7a82d55824..bbf1d34d79a44 100644
--- a/fs/nfs/pnfs.c
+++ b/fs/nfs/pnfs.c
@@ -994,7 +994,7 @@ pnfs_layout_stateid_blocked(const struct pnfs_layout_hdr *lo,
{
u32 seqid = be32_to_cpu(stateid->seqid);
- return !pnfs_seqid_is_newer(seqid, lo->plh_barrier);
+ return !pnfs_seqid_is_newer(seqid, lo->plh_barrier) && lo->plh_barrier;
}
/* lget is set to 1 if called from inside send_layoutget call chain */
@@ -1910,6 +1910,11 @@ static void nfs_layoutget_end(struct pnfs_layout_hdr *lo)
wake_up_var(&lo->plh_outstanding);
}
+static bool pnfs_is_first_layoutget(struct pnfs_layout_hdr *lo)
+{
+ return test_bit(NFS_LAYOUT_FIRST_LAYOUTGET, &lo->plh_flags);
+}
+
static void pnfs_clear_first_layoutget(struct pnfs_layout_hdr *lo)
{
unsigned long *bitlock = &lo->plh_flags;
@@ -2384,17 +2389,17 @@ pnfs_layout_process(struct nfs4_layoutget *lgp)
goto out_forget;
}
- if (!pnfs_layout_is_valid(lo)) {
- /* We have a completely new layout */
- pnfs_set_layout_stateid(lo, &res->stateid, lgp->cred, true);
- } else if (nfs4_stateid_match_other(&lo->plh_stateid, &res->stateid)) {
+ if (nfs4_stateid_match_other(&lo->plh_stateid, &res->stateid)) {
/* existing state ID, make sure the sequence number matches. */
if (pnfs_layout_stateid_blocked(lo, &res->stateid)) {
+ if (!pnfs_layout_is_valid(lo) &&
+ pnfs_is_first_layoutget(lo))
+ lo->plh_barrier = 0;
dprintk("%s forget reply due to sequence\n", __func__);
goto out_forget;
}
pnfs_set_layout_stateid(lo, &res->stateid, lgp->cred, false);
- } else {
+ } else if (pnfs_layout_is_valid(lo)) {
/*
* We got an entirely new state ID. Mark all segments for the
* inode invalid, and retry the layoutget
@@ -2407,6 +2412,11 @@ pnfs_layout_process(struct nfs4_layoutget *lgp)
pnfs_mark_matching_lsegs_return(lo, &lo->plh_return_segs,
&range, 0);
goto out_forget;
+ } else {
+ /* We have a completely new layout */
+ if (!pnfs_is_first_layoutget(lo))
+ goto out_forget;
+ pnfs_set_layout_stateid(lo, &res->stateid, lgp->cred, true);
}
pnfs_get_lseg(lseg);
--
2.27.0
next prev parent reply other threads:[~2021-02-02 16:10 UTC|newest]
Thread overview: 28+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-02-02 15:05 [PATCH AUTOSEL 5.10 01/25] af_key: relax availability checks for skb size calculation Sasha Levin
2021-02-02 15:05 ` [PATCH AUTOSEL 5.10 02/25] regulator: core: avoid regulator_resolve_supply() race condition Sasha Levin
2021-02-02 16:12 ` Mark Brown
2021-02-03 1:04 ` Sasha Levin
2021-02-03 12:12 ` Mark Brown
2021-02-02 15:05 ` [PATCH AUTOSEL 5.10 03/25] ASoC: wm_adsp: Fix control name parsing for multi-fw Sasha Levin
2021-02-02 15:05 ` [PATCH AUTOSEL 5.10 04/25] drm/nouveau/nvif: fix method count when pushing an array Sasha Levin
2021-02-02 15:05 ` [PATCH AUTOSEL 5.10 05/25] mac80211: 160MHz with extended NSS BW in CSA Sasha Levin
2021-02-02 15:05 ` [PATCH AUTOSEL 5.10 06/25] ASoC: Intel: Skylake: Zero snd_ctl_elem_value Sasha Levin
2021-02-02 15:05 ` [PATCH AUTOSEL 5.10 07/25] chtls: Fix potential resource leak Sasha Levin
2021-02-02 15:05 ` [PATCH AUTOSEL 5.10 08/25] pNFS/NFSv4: Try to return invalid layout in pnfs_layout_process() Sasha Levin
2021-02-02 15:05 ` Sasha Levin [this message]
2021-02-02 15:06 ` [PATCH AUTOSEL 5.10 10/25] ALSA: hda: intel-dsp-config: add PCI id for TGL-H Sasha Levin
2021-02-02 15:06 ` [PATCH AUTOSEL 5.10 11/25] ASoC: ak4458: correct reset polarity Sasha Levin
2021-02-02 15:06 ` [PATCH AUTOSEL 5.10 12/25] ASoC: Intel: sof_sdw: set proper flags for Dell TGL-H SKU 0A5E Sasha Levin
2021-02-02 15:06 ` [PATCH AUTOSEL 5.10 13/25] iwlwifi: mvm: skip power command when unbinding vif during CSA Sasha Levin
2021-02-02 15:06 ` [PATCH AUTOSEL 5.10 14/25] iwlwifi: mvm: take mutex for calling iwl_mvm_get_sync_time() Sasha Levin
2021-02-02 15:06 ` [PATCH AUTOSEL 5.10 15/25] iwlwifi: pcie: add a NULL check in iwl_pcie_txq_unmap Sasha Levin
2021-02-02 15:06 ` [PATCH AUTOSEL 5.10 16/25] iwlwifi: pcie: fix context info memory leak Sasha Levin
2021-02-02 15:06 ` [PATCH AUTOSEL 5.10 17/25] iwlwifi: mvm: invalidate IDs of internal stations at mvm start Sasha Levin
2021-02-02 15:06 ` [PATCH AUTOSEL 5.10 18/25] iwlwifi: pcie: add rules to match Qu with Hr2 Sasha Levin
2021-02-02 15:06 ` [PATCH AUTOSEL 5.10 19/25] iwlwifi: mvm: guard against device removal in reprobe Sasha Levin
2021-02-02 15:06 ` [PATCH AUTOSEL 5.10 20/25] iwlwifi: queue: bail out on invalid freeing Sasha Levin
2021-02-02 15:06 ` [PATCH AUTOSEL 5.10 21/25] SUNRPC: Move simple_get_bytes and simple_get_netobj into private header Sasha Levin
2021-02-02 15:06 ` [PATCH AUTOSEL 5.10 22/25] SUNRPC: Handle 0 length opaque XDR object data properly Sasha Levin
2021-02-02 15:06 ` [PATCH AUTOSEL 5.10 23/25] i2c: mediatek: Move suspend and resume handling to NOIRQ phase Sasha Levin
2021-02-02 15:06 ` [PATCH AUTOSEL 5.10 24/25] blk-cgroup: Use cond_resched() when destroy blkgs Sasha Levin
2021-02-02 15:06 ` [PATCH AUTOSEL 5.10 25/25] nvme-pci: add the DISABLE_WRITE_ZEROES quirk for a SPCC device Sasha Levin
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=20210202150615.1864175-9-sashal@kernel.org \
--to=sashal@kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-nfs@vger.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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox