git://git.whamcloud.com - fs/lustre-release.git/log

LU-15671 mds: do not send OST_CREATE transno interop

Send OST_CREATE RPCs from the MDS with no_resend and no_delay
when communicating with an old OST that does not support the
OBD_CONNECT2_REPLAY_RESEND. Likewise, the OST should not reply
to the MDS RPC with rq_transno set, or this will trigger:

osp_precreate_send() ASSERTION(req->rq_transno == 0) failed

This can be avoided if the MDS is upgraded before the OSS, but
will always be hit if OSS is upgraded first.

After 2.20.53 the MDS/OSS assume that this is always true, since
rolling upgrades are unsupported for larger version differences.

Test-Parameters: testgroup=rolling-upgrade-oss
Fixes: 63e17799a3 ("LU-8367 osp: enable replay for precreation request")
Signed-off-by: Andreas Dilger <adilger@whamcloud.com>
Signed-off-by: Sergey Cheremencev <scherementsev@ddn.com>
Change-Id: I1ab601a2f55540dd75cf24838f7cdb7f823ed42c
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/51056
Tested-by: Maloo <maloo@whamcloud.com>
Tested-by: jenkins <devops@whamcloud.com>
Reviewed-by: Alex Zhuravlev <bzzz@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>

LU-9798 tests: split server recovery tests

The Lustre test suite recovery-mds-scale has two tests; one
that fails over MDSs and a second test that fails over OSSs.
The default time for each of these tests to run is 24 hours.

Break up the two server recovery tests into separate test
suites so that they can be run independent of each other and
in parallel.

Collect commmon functions and global variables into a libary
used by all the recovery-*scale tests.

Test-Parameters: trivial env=SLOW=no,FAILURE_MODE=HARD \
iscsi=1 clientcount=4 osscount=2 ostcount=7 mdscount=2 \
mdtcount=2 austeroptions=-R failover=true \
testlist=recovery-mds-scale,recovery-oss-scale,recovery-random-scale

Signed-off-by: Alex Deiter <adeiter@tintri.com>
Change-Id: I4eda5265742270821fa10ab24f0e0637c00059d7
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/43461
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>

LU-16420 tests: move overstriping test to sanity-pfl

The overstriping test does not really need to be run for
every test run, so sanity-pfl is good enough. Also try
to see if moving this subtest out of sanity will stop
the test_51d failures seen since the test was landed.

Fixes: 97d29eb800 ("LU-13748 mdt: remove LASSERT in mdt_dump_lmm")
Test-Parameters: trivial testlist=sanity-pfl
Test-Parameters: testlist=sanity env=ONLY=0-51,HONOR_EXCEPT=true mdscount=2 mdtcount=4 ostcount=8
Test-Parameters: testlist=sanity env=ONLY=0-51,HONOR_EXCEPT=true mdscount=2 mdtcount=4 ostcount=8
Test-Parameters: testlist=sanity env=ONLY=0-51,HONOR_EXCEPT=true mdscount=2 mdtcount=4 ostcount=8
Test-Parameters: testlist=sanity env=ONLY=0-51,HONOR_EXCEPT=true mdscount=2 mdtcount=4 ostcount=8
Test-Parameters: testlist=sanity env=ONLY=0-51,HONOR_EXCEPT=true mdscount=2 mdtcount=4 ostcount=8
Test-Parameters: testlist=sanity env=ONLY=0-51,HONOR_EXCEPT=true mdscount=2 mdtcount=4 ostcount=8
Test-Parameters: testlist=sanity env=ONLY=0-51,HONOR_EXCEPT=true mdscount=2 mdtcount=4 ostcount=8
Test-Parameters: testlist=sanity env=ONLY=0-51,HONOR_EXCEPT=true mdscount=2 mdtcount=4 ostcount=8
Test-Parameters: testlist=sanity env=ONLY=0-51,HONOR_EXCEPT=true mdscount=2 mdtcount=4 ostcount=8
Test-Parameters: testlist=sanity env=ONLY=0-51,HONOR_EXCEPT=true mdscount=2 mdtcount=4 ostcount=8
Signed-off-by: Andreas Dilger <adilger@whamcloud.com>
Change-Id: I71ebc76fdcceb2b22994c09c3574cda4602540e5
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/51268
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Colin Faber <cfaber@ddn.com>
Reviewed-by: Alex Deiter <alex.deiter@gmail.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>

LU-16882 test: change ALWAYS_EXCEPT with function always_except

Change ALWAYS_EXCEPT with function always_except in replay-single.sh

Test-Parameters: trivial testlist=replay-single

Change-Id: I6fb4c350b1aa6f8c38920bf52e8ae42b64bdcd64
Signed-off-by: Wei Liu <sarah@whamcloud.com>
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/51242
Tested-by: Oleg Drokin <green@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: Alex Deiter <alex.deiter@gmail.com>

LU-16873 osd: update OI_Scrub file with new magic

The fix for LUS-11542 detects the format change correctly
but does not write new oi scrub file magic, so new mount
triggers the "oi files counter reset" again and again.

Fixes: 126275ba83 ("LU-16655 scrub: upgrade scrub_file from 2.12 format")
HPE-bug-id: LUS-11646
Signed-off-by: Alexander Zarochentsev <alexander.zarochentsev@hpe.com>
Change-Id: Ia13fcfaf0d8f2c4ee9331dd9fec0ff159d195186
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/51226
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: Andrew Perepechko <andrew.perepechko@hpe.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>

LU-16848 gss: drop legacy 'lgssd' and pipefs code

Get rid of the old 'lgssd' code which is unused
for a while, and not even buildable.

Test-Parameters: trivial
Test-Parameters: testlist=sanity-sec
Test-Parameters: kerberos=true testlist=sanity-krb5

Change-Id: I821cc5b625e97c7f5fdc758b166bd54e5e4b379e
Signed-off-by: Aurelien Degremont <adegremont@nvidia.com>
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/51032
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Sebastien Buisson <sbuisson@ddn.com>
Reviewed-by: Timothy Day <timday@amazon.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>

LU-16828 sanity: wrong argument for find in test_133g

> proc_regexp="/{proc,sys}/{fs,sys,kernel/debug}/{lustre,lnet}/"
> proc_dirs=$(eval ls -d $proc_regexp)

After running the upper commands, 'echo "$proc_dirs"' woud print
multiple lines like what sanity test 401a would print:

proc_dirs='/proc/fs/lustre/
/sys/fs/lustre/
/sys/kernel/debug/lnet/
/sys/kernel/debug/lustre/'

This will cause wrong command exectuted remotely in test_133g without
"-name" argument and badarea_io pipe after the command:

... |
xargs -n 1 find /proc/fs/lustre/
/sys/fs/lustre/
/sys/kernel/debug/lnet/
/sys/kernel/debug/lustre/ -name |
xargs -n 1 badarea_io

This patch fixes the problem.

Test-Parameters: trivial testlist=sanity env=ONLY=133g
Signed-off-by: Li Xi <lixi@ddn.com>
Change-Id: I93d611c34eecfa6d14d3f83a60faa3e637514128
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/50994
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: Alex Deiter <alex.deiter@gmail.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>

LU-16772 quota: protect lqe_glbl_data in qmt_site_recalc_cb

lqe_glbl_data should be protected with lqe_glbl_data_lock in
qmt_site_recalc_sb like it did in other places. Otherwise it
may cause following panic:

  BUG: unable to handle kernel NULL pointer at 00000000000000f8
  qmt_site_recalc_cb+0x2f8/0x790 [lquota]
  cfs_hash_for_each_tight+0x121/0x310 [libcfs]
  qmt_pool_recalc+0x372/0x9f0 [lquota]

Also protect lqe_glbl_data access with lqe_glbl_data_lock in
qmt_lvbo_free().

Fixes: 1dbcbd70f8 ("LU-15021 quota: protect lqe_glbl_data in lqe")
Signed-off-by: Sergey Cheremencev <scherementsev@ddn.com>
Change-Id: I030f14b02062151f1708a03ac7414a9991f798f6
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/50748
Tested-by: jenkins <devops@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>

LU-16850 socklnd: remove ksnr_myiface from ksock_conn_cb

Drop ksnr_myiface: it is no longer needed since socklnd
TCP bonding got removed. There's one interface per
connection cb per peer_ni, and it can be accessed as
net->ksnn_interface.ksni_index.

Fix setting of ksni_nroutes accordingly. Duplication of
interface index in conn_cb and ksnn_interface was causing
the assertion
ASSERTION( net->ksnn_interface.ksni_nroutes == 0 )
in ksocknal_shutdown() to fail if the corresponding
device is deregistered before lnd shutdown.

Modify test_214 of sanity-lnet to create connections so that
the scenario of socklnd shutdown with NI on a deregistered
interface is covered.

Fixes: 3c9282a6 ("LU-16378 lnet: handles unregister/register events)
Test-Parameters: trivial
Signed-off-by: Serguei Smirnov <ssmirnov@whamcloud.com>
Change-Id: I4de164c9e64aa770164a8320b9460fadce49aa06
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/51148
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Frank Sehr <fsehr@whamcloud.com>
Reviewed-by: Cyril Bordage <cbordage@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>

LU-16623 lod: handle object allocation consistently

Consistently handle the various OS_STATFS_* flags that indicate
an OST or MDT is full or otherwise marked ineligible for use.

Fix lod_statfs_check() so it skips MDTs with OS_STATFS_ENOINO
for allocating dir stripes instead of only checking OST targets.

In the LOD code, ltd_active=0 indicates that the device is not
usable for new object allocations for a variety of reasons. That
includes out of space or inodes, read-only, max_create_count=0,
or disconnected export, not *only* that the OSP is disconnected
from the OST as with imp_deactive.  Targets marked ltd_active=0
will not be counted in ld_active_tgt_count, so these OSTs will
not count toward stripe_count for stripe_count=-1 files.

Set flags = LOD_USES_DEFAULT_STRIPE in lod_qos_prep_create() for
stripe_count = -1 layouts and pass it to lod_stripe_count_min()
to avoid use of *all* OSTs when free space is imbalanced or OSTs
are not available, and be happy with allocations on 3/4 of OSTs.
It looks like this functionality was missed when object allocations
transitioned from the LOV to LOD module.  Put the LOV_USES_* into
an enum and rename to LOD_USES_* for consistency with current code.

Apply the lod.*.max_stripe_count limits to PFL components as well
as plain file layouts in lod_comp_entry_stripe_count().

Rename ltd_connecting to ltd_discon, since there is no guarantee
that this target is actually *connecting*, only that it is currently
disconnected.  Use ltd_discon in places that checked ltd_active to
decide if the OSP was disconnected from the OST, which shouldn't be
skipped just because the OST is full or has creates disabled.

Fixes: 7b124fef76 ("LU-4277 lod: handle os_state as a flag, check READONLY")
Fixes: 5b147e47de ("LU-11115 lod: skip max_create_count=0 OST in QoS and RR algorithms")
Fixes: c7f2e70a27 ("LU-1303 lod: QoS allocation policy")
Fixes: c1d0a355a6 ("LU-12624 lod: alloc dir stripes by QoS")
Fixes: 3c9580931d ("LU-9162 lod: option to set max stripe count per filesystem")
Signed-off-by: Andreas Dilger <adilger@whamcloud.com>
Signed-off-by: Sergey Cheremencev <scherementsev@ddn.com>
Change-Id: Ifb9443fe6c80b4d7f82b442060db7ac8423ebbe5
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/50250
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Reviewed-by: Patrick Farrell <pfarrell@whamcloud.com>

New tag 2.15.56

Change-Id: Ic2aee8234566ccc5b16e8c4fb1039365940965b8
Signed-off-by: Oleg Drokin <green@whamcloud.com>

LU-12610 osc: remove OBD_ -> CFS_ macros

Remove OBD macros that are simply redefinitions
of CFS macros.

Also, convert some spaces to tabs.

Test-Parameters: trivial
Signed-off-by: Timothy Day <timday@amazon.com>
Signed-off-by: Ben Evans <beevans@whamcloud.com>
Change-Id: Icb4f25f51515d833fed2c05581288cde719c1d08
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/51124
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: James Simmons <jsimmons@infradead.org>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>

LU-12610 target: remove OBD_ -> CFS_ macros

Remove OBD macros that are simply redefinitions
of CFS macros.

Test-Parameters: trivial
Signed-off-by: Timothy Day <timday@amazon.com>
Signed-off-by: Ben Evans <beevans@whamcloud.com>
Change-Id: I97e3f74d72d41558f293567b4609fa37aaa3b13d
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/51123
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: James Simmons <jsimmons@infradead.org>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>

LU-16830 lod: improve rr allocation

Roundrobin allocation uses atomic_inc() % ost_count for
generation OST index. When some OSTs are unavailable and
many threads make object creation, it could happen that
OST idx is the same for all attempts. For example with
4 OSTs configuration when 2 OSTs do faiover, estimation
of probability is 0.5^12=0.024%. The result is ENOSPC for
user application.

Let's try one by one OSTs for a last speed loop.

HPE-bug-id: LUS-11265
Signed-off-by: Alexander Boyko <alexander.boyko@hpe.com>
Change-Id: I325cf4ad706c9b0df64cf53792e77c1fad6f7739
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/50996
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Sergey Cheremencev <scherementsev@ddn.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>

LU-16799 tests: fix sanity-krb5

sanity-krb5.sh needs to be fixed in several ways.
It cannot assume that the Kerberos credentials cache is FILE, and
expect ccache files to be under /tmp/krb5cc_*.
The lsvcgssd daemon must be launched with -vvv flags for easier
debugging.
Keyring needs to be cleared appropriately after using 'lfs flushctx'.

Test-Parameters: trivial testlist=sanity-krb5 kerberos=true
Signed-off-by: Sebastien Buisson <sbuisson@ddn.com>
Change-Id: I31ca8d2d97e137c7ba9fa478d5432aeedb5135a8
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/50864
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Aurelien Degremont <adegremont@nvidia.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>

LU-9859 libcfs: move percpt_lock into lnet

lnet is the only users of percpt_lock - and there are only two such
locks!
So move the code into lnet, as part of deprecating libcfs.

Test-Parameters: trivial
Signed-off-by: Mr NeilBrown <neilb@suse.de>
Change-Id: Id7091e88cf61228aa031921747fb9c7b08214931
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/50832
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: James Simmons <jsimmons@infradead.org>
Reviewed-by: Arshad Hussain <arshad.hussain@aeoncomputing.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>

LU-16788 tests: sanity should remove temp files

during the test to fit OSTSIZE

Test-Parameters: trivial
Signed-off-by: Alex Zhuravlev <bzzz@whamcloud.com>
Change-Id: I2f1cfe0511061794d81d0349cf36a50f40470553
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/50819
Reviewed-by: Arshad Hussain <arshad.hussain@aeoncomputing.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>

LU-12610 misc: remove OBD_ -> CFS_ macros

Remove OBD macros that are simply redefinitions
of CFS macros.

Test-Parameters: trivial
Signed-off-by: Timothy Day <timday@amazon.com>
Signed-off-by: Ben Evans <beevans@whamcloud.com>
Change-Id: I15fe8aa22cb0203bed102a35361f4854ddaabecb
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/50809
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: James Simmons <jsimmons@infradead.org>
Reviewed-by: Neil Brown <neilb@suse.de>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>

LU-12610 osd: remove OBD_ -> CFS_ macros

Remove OBD macros that are simply redefinitions
of CFS macros.

The CFS macros are provided by libcfs.h, so add this
header where needed.

Test-Parameters: trivial
Signed-off-by: Timothy Day <timday@amazon.com>
Signed-off-by: Ben Evans <beevans@whamcloud.com>
Change-Id: Ia7f7ba611b98500ecf06137159649949a621476f
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/50805
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Arshad Hussain <arshad.hussain@aeoncomputing.com>
Reviewed-by: Neil Brown <neilb@suse.de>
Reviewed-by: James Simmons <jsimmons@infradead.org>
Reviewed-by: Oleg Drokin <green@whamcloud.com>

LU-16694 misc: remove, update old scripts

There are two old checkstack.pl in-tree. Remove both and pull down
a new one from upstream.

There's only one script in lustre/contrib (lustre_server.sh). It
is meant for ClusterLabs resource-agents. But the script hasn't
been maintained. Hence, remove it.

Test-Parameters: trivial
Signed-off-by: Timothy Day <timday@amazon.com>
Change-Id: Id52ff11a7fa525b7ef20656df77c66a728e2b77a
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/50720
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: Olaf Faaland <faaland1@llnl.gov>
Reviewed-by: Nathaniel Clark <nclark@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>

LU-16763 tests: add more logging to run-llog.sh

Add more logging to run-llog.sh. At the same time, add SPDX text
and fix some minor shellcheck warnings.

Test-Parameters: trivial testlist=sanity env=ONLY=60a,ONLY_REPEAT=10
Signed-off-by: Timothy Day <timday@amazon.com>
Change-Id: I317fac7b872be53a1094022cfcd7d130b4c79c0a
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/50719
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: Arshad Hussain <arshad.hussain@aeoncomputing.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>

LU-12610 ptlrpc: replace OBD_ -> CFS_ macros

Replace OBD macros that are simply redefinitions
of CFS macros.

Signed-off-by: Timothy Day <timday@amazon.com>
Signed-off-by: Ben Evans <beevans@whamcloud.com>
Change-Id: I634f364d33ac56de678d273d87c9ac54d1f8c1ef
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/50684
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: Arshad Hussain <arshad.hussain@aeoncomputing.com>
Reviewed-by: Neil Brown <neilb@suse.de>
Reviewed-by: James Simmons <jsimmons@infradead.org>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>

LU-11388 tests: replay-single/131b to refresh grants

so that the write (to be replayed after replay-barrier)
doesn't turn sync due to insufficient grant.

Test-Parameters: trivial testlist=replay-single env=ONLY=131b,ONLY_REPEAT=30
Fixes: cb3b2bb683 ("LU-11388 test: enable replay-single test_131b")
Signed-off-by: Alex Zhuravlev <bzzz@whamcloud.com>
Change-Id: If4656c1028b49c58eedd905abd0c329f3706f491
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/50661
Reviewed-by: Mikhail Pershin <mpershin@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Reviewed-by: Jian Yu <yujian@whamcloud.com>
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>

LU-16548 lnet: report actual timeout used by lnd

lnd_timeout value reported by lnetctl may be different
from what is actually used.
There's an lnd_timeout calculated as a function of transaction
timeout and retry_count. This is the value displayed by
"lnetctl global show". However, each LND may define its own
timeout by setting timeout module parameter to a positive value,
which overrides the higher-level lnd_timeout defined by LNet.
lnetctl net show -v will show the timeout value in the
lnd_tunables section.
The timeout for socklnd, o2iblnd and gnilnd is implemented.
A test for sock, ib and gni is included.

Test-Parameters: trivial
Signed-off-by: Frank Sehr <fsehr@whamcloud.com>
Change-Id: I85a107ba6f1259c577f74945b89fd695f191d514
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/50620
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Neil Brown <neilb@suse.de>
Reviewed-by: Serguei Smirnov <ssmirnov@whamcloud.com>
Reviewed-by: Cyril Bordage <cbordage@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>

LU-16517 build: pass extra configure options to "make debs"

While running "make debs", the configure command in debian/rules
ignores some user defined configure options. This patch fixes
the issue by adding the detection of the extra options into
debian/rules.

Test-Parameters: trivial clientdistro=ubuntu2204

Change-Id: Ia9db4e05abf33834cb3c853f4f0829dadc8d7400
Signed-off-by: Jian Yu <yujian@whamcloud.com>
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/50464
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Tested-by: Shuichi Ihara <sihara@ddn.com>
Reviewed-by: Shaun Tancheff <shaun.tancheff@hpe.com>
Reviewed-by: Shuichi Ihara <sihara@ddn.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>

LU-16505 tests: run_info() minor fix

Address the run_info() parameters correctly.

Fixes: a98728e4fd ("LU-15626 tests: Fix "error" reported by shellcheck for recovery-mds-scale")
Test-Parameters: trivial clientcount=5 mdtcount=2 mdscount=2 osscount=2 austeroptions=-R failover=true iscsi=1 env=SERVER_FAILOVER_PERIOD=1600,REQFAIL_PERCENT=100,SLOW=yes testlist=recovery-mds-scale
Signed-off-by: Elena Gryaznova <elena.gryaznova@hpe.com>
HPE-bug-id: LUS-11449
Change-Id: Ice8f5b0a85d66708942c7665f028fd6de66165a9
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/49771
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Reviewed-by: Arshad Hussain <arshad.hussain@aeoncomputing.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: Alex Deiter <alex.deiter@gmail.com>

LU-15971 llite: implicit default LMV inherit

With implicit default LMV inherit, the inherited default LMV is
not stored on disk, but maintained on client side.

Benefits:
* change of directory default LMV will be populated to all sub levels
  at runtime.
* default LMV will be packed into mkdir request, therefore MDT doesn't
  need to read it from disk, as will improve mkdir performance.

Caveats:
* to disable inherited default LMV on subdir, a default LMV need to be
  set on this subdir explicitly like this:
        "lfs setdirstripe -D -i <subdir_mdt_index> --max-inherit 0"

Changes on client side:
* update inherited default LMV after lookup/open/revalidate.
* pack default LMV in mkdir request.
* add "--raw" option for "lfs getdirstripe -D" to print default LMV
  stored in inode, if directory doesn't have default LMV, or its
  default LMV is implicitly inherited, nothing will be printed.

Changes on MDT side:
* use the default LMV from client in lod_ah_init() to mkdir.
* don't save inherited default LMV in mkdir.

Add sanityn 114.

Test-Parameters: clientversion=2.14 testlist=sanity mdtcount=4 mdscount=2 env=SANITY_EXCEPT="39l 39r 134b 150b 160a 205a 208 220 230e 230p 300g 807"
Test-Parameters: serverversion=2.14 testlist=sanity mdtcount=4 mdscount=2 env=SANITY_EXCEPT="27Cg 39r 65n 413a 413b 905"
Signed-off-by: Lai Siyao <lai.siyao@whamcloud.com>
Change-Id: Iae109a0ef35a273175c70dd0b394e721a5ce0c45
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/47789
Tested-by: Maloo <maloo@whamcloud.com>
Tested-by: jenkins <devops@whamcloud.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: Hongchao Zhang <hongchao@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>

LU-6142 lnet: use list_first_entry() where appropriate.

This patch changes
list_entry(foo.next, ...)
to
list_first_entry(&foo, ...)

in cases where 'foo' is a list head, rather than a list member.

Test-Parameters: trivial
Signed-off-by: Mr NeilBrown <neilb@suse.de>
Change-Id: Ied04412bf976d8fb219bb3c14c56879d2cf83ae7
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/50827
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Arshad Hussain <arshad.hussain@aeoncomputing.com>
Reviewed-by: James Simmons <jsimmons@infradead.org>
Reviewed-by: Oleg Drokin <green@whamcloud.com>

LU-10391 mdt: change md_perm.mp_nid to large nid

mp_nid in struct md_perm is now a large nid.

Signed-off-by: Mr NeilBrown <neilb@suse.de>
Change-Id: Id7502a5399191a36550162837cce37d3bfc9797e
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/50103
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: James Simmons <jsimmons@infradead.org>
Reviewed-by: Sebastien Buisson <sbuisson@ddn.com>
Reviewed-by: Nathaniel Clark <nclark@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>

LU-16785 build: Cleanup test for IS_ENCRYPTED macro

Suppress warning when -Wunused-value is enabled.

Test-Parameters: trivial
Signed-off-by: Shaun Tancheff <shaun.tancheff@hpe.com>
Change-Id: Iec387286fd8499bb13f43c89e671d7d7aa0de9e1
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/50818
Tested-by: Maloo <maloo@whamcloud.com>
Tested-by: jenkins <devops@whamcloud.com>
Reviewed-by: Sebastien Buisson <sbuisson@ddn.com>
Reviewed-by: Petros Koutoupis <petros.koutoupis@hpe.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>

LU-16807 ksocklnd: ksocklnd_ni_get_eth_intf_speed() must use only rtnl lock

A kernel with debug enable is reporting the following:

include/linux/inetdevice.h:221 suspicious rcu_dereference_check() usage!
other info that might help us debug this:
rcu_scheduler_active = 2, debug_locks = 1
5 locks held by lctl/42289:
#0: ffffffffa04263f8 ((libcfs_ioctl_list).rwsem){++++}-{3:3}, at: __blocking_notifier_call_chain+0x44/0xa0
#1: ffffffffa04fa7b0 (lnet_config_mutex){+.+.}-{3:3}, at: lnet_configure+0x1d/0xc0 [lnet]
#2: ffffffffa04f06e8 (the_lnet.ln_api_mutex){+.+.}-{3:3}, at: LNetNIInit+0x4c/0x960 [lnet]
#3: ffffffffa04f0788 (&the_lnet.ln_lnd_mutex){+.+.}-{3:3}, at: lnet_startup_lndnet+0x11e/0xa90 [lnet]
#4: ffffffff834a0cd0 (rtnl_mutex){+.+.}-{3:3}, at: rtnl_lock+0x1b/0x30

stack backtrace:
CPU: 2 PID: 42289 Comm: lctl Kdump: loaded Tainted: G        W  O     --------- -  - 4.18.0rh8.5-debug #2
Hardware name: Bochs Bochs, BIOS Bochs 01/01/2011
  Call Trace:
  ? dump_stack+0x119/0x18e
  ? lockdep_rcu_suspicious+0x141/0x14f
  ? ksocklnd_ni_get_eth_intf_speed.isra.1+0x308/0x360 [ksocklnd]
  ? mark_held_locks+0x6a/0xc0
  ? ktime_get_with_offset+0x233/0x2b0
  ? trace_hardirqs_on+0x4e/0x220
  ? ksocknal_tunables_setup+0xed/0x200 [ksocklnd]
  ? ksocknal_startup+0x4ff/0x1180 [ksocklnd]

The function __ethtool_get_link_ksettings() has a hard requirement
to have the in_dev device protected by the rtnl mutex. At the
same time we are aquiring in_dev using the rcu lock. The in_dev
needs to be stabilized by the same lock. So use rtnl functions
instead.

Test-Parameters: trivial testlist=sanity-lnet
Change-Id: Iec795d62eb33002950cc962f29f9b93905b3bb3f
Signed-off-by: James Simmons <jsimmons@infradead.org>
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/51028
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Reviewed-by: Chris Horn <chris.horn@hpe.com>
Reviewed-by: Neil Brown <neilb@suse.de>

LU-16751 gss: remove old patches for nfs-utils 1.0.*

Remove patches for an old version of nfs-utils. Remove the README
that suggests using them. lustre/utils/gss has an entire fork of
nfs-utils utils/gssd directory already.

Also, sk_utils.c shouldn't be executable.

Test-Parameters: trivial
Signed-off-by: Timothy Day <timday@amazon.com>
Change-Id: Ia47193073d2403125043d51db889d0ded41ea9b7
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/51003
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: Sebastien Buisson <sbuisson@ddn.com>
Reviewed-by: Aurelien Degremont <adegremont@nvidia.com>

LU-9329 utils: add large xattr support for lustre_c

lustre_rsync.c had the problem of not able to get the desired
xattr buffer size, thus is not able to support large xattr
(> PATH_MAX).

lustre-rsync-test:1A would fail if /tmp file system on the
test client supports large xattr.

The following lustre_rsync log shows that lgetxattr keeps
on failing on large xattr (user.foo):

(trusted.lma,14307984) rc=0x18
lsetxattr(), rc=0, errno=0
(trusted.lov,14307984) rc=0x38
lsetxattr(), rc=0, errno=0
(trusted.link,14307984) rc=0x2f
lsetxattr(), rc=0, errno=0
(trusted.som,14307984) rc=0x18
lsetxattr(), rc=0, errno=0
(user.foo,14307984) rc=0xffffffff
(lustre.lov,14307984) rc=0x38
lsetxattr(), rc=-1, errno=95

Test-Parameters: trivial testlist=lustre-rsync-test
Signed-off-by: Li Xi <lixi@ddn.com>
Change-Id: I3ff49721b88dd31aa8af76da8932d5004c82ea09
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/50997
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: Emoly Liu <emoly@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>

LU-16807 libcfs: give the tcd_lock types different classes.

There are three different trace contexts:
process, softirq, irq.
Each has its own lock (tcd_lock) which is locked as appropriate for
that context.

lockdep currently doesn't see that they are different and so deduces
that the different uses might lead to deadlocks.

So use separate calls to spin_lock_init() so that they each get a
separate lock class, and lockdep sees no problem.

Test-Parameters: trivial testlist=sanity-lnet
Signed-off-by: Mr NeilBrown <neilb@suse.de>
Change-Id: Icca7706d8e0d8ae8add4c540d2da090b53d7e65c
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/50992
Tested-by: Maloo <maloo@whamcloud.com>
Tested-by: jenkins <devops@whamcloud.com>
Reviewed-by: James Simmons <jsimmons@infradead.org>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: Chris Horn <chris.horn@hpe.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>

LU-16813 utils: move mirror_end initialization

Move initialization for mirror_end variable in
llapi_mirror_resync_many(), otherwise lfs mirror resync
may fail since mirror_end gets reset on each pass of
the loop.

Fixes: b0297a1056 ("LU-16518 lnet: fix clang build errors")
Test-Parameters: testlist=sanity-flr env=ONLY=42,ONLY_REPEAT=100
Signed-off-by: Timothy Day <timday@amazon.com>
Change-Id: I4edc078cd6e30d7a0ad84383b55b63b885a34d4b
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/50919
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Reviewed-by: Alex Zhuravlev <bzzz@whamcloud.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>

LU-16796 libcfs: Remove reference to LASSERT_ATOMIC_POS

This patch removes all reference to LASSERT_ATOMIC_POS macro.
Once all the access is removed it would be easier to just
toggle atomic_* API calls with recount_* counts.

Signed-off-by: Arshad Hussain <arshad.hussain@aeoncomputing.com>
Change-Id: I2051de3707106532259e51ec3e4c890c65836b1a
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/50881
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: James Simmons <jsimmons@infradead.org>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: Timothy Day <timday@amazon.com>
Reviewed-by: Neil Brown <neilb@suse.de>

LU-9859 libcfs: move cfs_expr_list_print to nidstrings.c

cfs_expr_list_print() is only called from nidstrings.c
so move it there and make it static.

Test-Parameters: trivial
Signed-off-by: Mr NeilBrown <neilb@suse.de>
Change-Id: Ie30aba19fd7935ba424c9212a81e7d0d91ad6f57
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/50834
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: James Simmons <jsimmons@infradead.org>
Reviewed-by: Oleg Drokin <green@whamcloud.com>

LU-12610 mdt: remove OBD_ -> CFS_ macros

Remove OBD macros that are simply redefinitions
of CFS macros.

Test-Parameters: trivial
Signed-off-by: Timothy Day <timday@amazon.com>
Signed-off-by: Ben Evans <beevans@whamcloud.com>
Change-Id: I89edc38316bb121849b24f881a8bafaf78038aa1
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/50806
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: Neil Brown <neilb@suse.de>
Reviewed-by: Arshad Hussain <arshad.hussain@aeoncomputing.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>

LU-16814 utils: Change llapi_root_path_open() signature

This patch changes llapi_root_path_open() first argument
form char * to const char * as the argument passed is
not modified.

This patch also fixes llapi_rmfid and llapi_root_path_open
man page.

Fixes: 5d93025240 ("LU-16427 lfs: rmfid does not print anything on error")
Reported-by: Thomas Bertschinger <bertschinger@lanl.gov>
Test-Parameters: trivial testlist=sanity
Signed-off-by: Arshad Hussain <arshad.hussain@aeoncomputing.com>
Change-Id: I81de8cb9280af91af8a2de8dbb51f8e82807220d
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/50909
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: Thomas Bertschinger <bertschinger@lanl.gov>
Reviewed-by: Oleg Drokin <green@whamcloud.com>

LU-12610 ldlm: replace OBD_ -> CFS_ macros

Replace OBD macros that are simply redefinitions
of CFS macros.

Test-Parameters: trivial
Signed-off-by: Timothy Day <timday@amazon.com>
Signed-off-by: Ben Evans <beevans@whamcloud.com>
Change-Id: I4d903f286f138152cff22df5cba411d2c9fcb4a8
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/50685
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: Patrick Farrell <pfarrell@whamcloud.com>
Reviewed-by: James Simmons <jsimmons@infradead.org>
Reviewed-by: Arshad Hussain <arshad.hussain@aeoncomputing.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>

LU-16748 llite: update comment of ll_swap_layouts_close

mdc_close_layout_swap_pack() does not exist, and the lease lock
handle is release in mdc_close_intent_pack(). This patch updates
the comment.

Also, mdt_close_swap_layouts() does not exist. This patch removes
its declaration.

Test-Parameters: trivial
Signed-off-by: Li Xi <lixi@ddn.com>
Change-Id: Iecd1754f627803e85f54a91d648e87e235106bd7
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/50660
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: Timothy Day <timday@amazon.com>

LU-16745 mdc: md_open_data should keep ref on close_req

md_open_data should keep a ref on mod_close_req,
otherwise the mod_close_req could be freed before
we try to access mod_close_req via md_open_data.

Change-Id: I621f7db389854326db298d99957a0bce43024b6e
Signed-off-by: Li Dongyang <dongyangli@ddn.com>
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/50656
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: Lai Siyao <lai.siyao@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>

LU-16829 gss: quiet noisy warnings

GSS code has plenty of debugging messages as warnings.
Update the code to make them just debugging messages and
nothing more.

Test-Parameters: trivial

Change-Id: I53c471d758b0309ae10bab000211fa0381cabf87
Signed-off-by: Aurelien Degremont <adegremont@nvidia.com>
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/50613
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Arshad Hussain <arshad.hussain@aeoncomputing.com>
Reviewed-by: Sebastien Buisson <sbuisson@ddn.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>

LU-16723 libcfs: refactor parser to be simpler

The parser code has a lot of unused complexity that can be
streamlined. Refactor the parser makes the interface simpler
from a development perspective and can provide a consistent
user experience for all of the lustre utilities.

All functions which are not used outside of the parser have
been made static. Functions which don't appear to be used at
all have been removed.

The file headers have been standardized and the SDX text
added. The function names have been changed to be more standard.

Test-Parameters: testlist=sanity env=ONLY=60a,ONLY_REPEAT=10
Signed-off-by: Timothy Day <timday@amazon.com>
Change-Id: I3354f213de424f51aef94c840083a4cb781d7aae
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/50583
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: Arshad Hussain <arshad.hussain@aeoncomputing.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>

LU-9859 libcfs: use round_up directly

The macro cfs_size_round() is just round_up(val, 8). Replace
cfs_size_round() with the Linux standard round_up().

Change-Id: I5a5ba4e663672c0b0bba5c99be9e0bece2dc3c87
Signed-off-by: James Simmons <jsimmons@infradead.org>
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/50545
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Arshad Hussain <arshad.hussain@aeoncomputing.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: Neil Brown <neilb@suse.de>
Reviewed-by: Oleg Drokin <green@whamcloud.com>

LU-16619 build: Ubuntu jammy 5.19 client support

Ubuntu 5.19 kernel removed lsmcontext_init() and changed
security_dentry_init_security to require struct context *

Linux kernel linux-hwe-5.19
LSM: Removed scaffolding function lsmcontext_init

Linux kernel linux-hwe-5.19
LSM: security_dentry_init_security with struct lsmcontext

Signed-off-by: Shaun Tancheff <shaun.tancheff@hpe.com>
Change-Id: Ib6479a2cd20df5e565ae6203e05df2afa3f3de31
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/50210
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Sebastien Buisson <sbuisson@ddn.com>
Reviewed-by: James Simmons <jsimmons@infradead.org>
Reviewed-by: Oleg Drokin <green@whamcloud.com>

LU-15668 osd-ldiskfs: fix osd_bio_private double free

In osd_do_bio(), if the IO is fragmented and bio_alloc()
fails to allocate new bio, the bio_private still holds
osd_bio_private for last bio and will be double freed
in osd_do_bio() and dio_integrity_complete_routine().

Test-Parameters: trivial
Change-Id: I42eaf95a85ec99a60359122054efb5beb0fb6104
Signed-off-by: Li Dongyang <dongyangli@ddn.com>
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/46879
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: Alex Zhuravlev <bzzz@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>

LU-16784 tests: fix path to lgss_sk

Find correct path to lgss_sk utility, by looking inside Lustre build
tree if command is not installed on the local node.

Test-Parameters: trivial
Test-Parameters: mdscount=2 mdtcount=4 osscount=1 ostcount=8 clientcount=2 testlist=sanity-sec env=SHARED_KEY=true
Signed-off-by: Sebastien Buisson <sbuisson@ddn.com>
Change-Id: I23920bb2a44d2ec7e9662e75c23bd5302d8dfee2
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/50825
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Sarah Liu <sarah@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>

LU-16804 tests: load CONFIG at beginning of init_test_env

In order to have all environment variables properly loaded, make
CONFIG loaded at the beginning of init_test_env().

Signed-off-by: Sebastien Buisson <sbuisson@ddn.com>
Change-Id: I1c3caa3d582c4b317ff3d0d10fc0103e046ddf17
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/50914
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Sarah Liu <sarah@whamcloud.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>

LU-15562 statahead: using try lock for batched RPCs

To avoid the possible deadlock between the batched statahead RPC
and rename()/migrate() operation, we use trylock to obtain the DLM
PR ibits lock for file attributes in a batched statahead RPC.
A failed trylock means that other users maybe modify the directory
simultaneously as the server only grants read lock to a client in
current Lustre design which is compatible with the PR lock for
attributes in stat()-ahead.
When a trylock failed, the MDT reports the conflict with the error
code -EBUSY, and the client stops the statahead immediately.

In this patch, we set "statahead_batch_max" with 64 to enable
batched statahead by default.

Signed-off-by: Qian Yingjin <qian@ddn.com>
Change-Id: I38394b1547e18ad156f94e49cd81aaef2f6fafb5
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/46549
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: Mikhail Pershin <mpershin@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>

LU-16826 lfsck: init rec_fid before declare_insert

lfsck_namespace_repair_dangling() doesn't init the record
buffer properly before calling dt_declare_insert() for the
case of local agent creation.

Test-parameters: trivial testlist=sanity-lfsck
HPE-bug-id: LUS-11609
Signed-off-by: Alexander Zarochentsev <alexander.zarochentsev@hpe.com>
Change-Id: Ibd0a44217e9ebcf469f7a817651e63214c218974
Reviewed-by: Andrew Perepechko <andrew.perepechko@hpe.com>
Reviewed-by: Shaun Tancheff <stancheff@cray.com>
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/50980
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Shaun Tancheff <shaun.tancheff@hpe.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>

LU-6142 libcfs: use list_first_entry() where appropriate.

This patch changes
list_entry(foo.next, ...)
to
list_first_entry(&foo, ...)

in cases where 'foo' is a list head, rather than a list member.
It also re-indents following lines where that is needed.

Test-Parameters: trivial
Signed-off-by: Mr NeilBrown <neilb@suse.de>
Change-Id: Ib7ec3f3ab7f34afb3b39fda0ff9d916514b062ff
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/50829
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Arshad Hussain <arshad.hussain@aeoncomputing.com>
Reviewed-by: James Simmons <jsimmons@infradead.org>
Reviewed-by: Oleg Drokin <green@whamcloud.com>

LU-6142 all: use list_first_entry() where appropriate.

Lustre already uses list_first_entry() in many places, but
it is not consistent. Let's make it consistent.
The patch was generated with

sed -i 's/list_entry(([^,]*)->next,/list_first_entry(1,/'
`git grep -l 'list_entry(.*->next' lustre/ lnet/ libcfs/ `

followed by some manual cleanup of indents, and adding
list_first_entry() to libcfs/include/libcfs/util/list.h

Test-Parameters: trivial
Signed-off-by: Mr NeilBrown <neilb@suse.de>
Change-Id: Id646fba1faf40282e66ede07c88c8db5ffadc211
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/50826
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Arshad Hussain <arshad.hussain@aeoncomputing.com>
Reviewed-by: James Simmons <jsimmons@infradead.org>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>

LU-10391 ptlrpc: switch sptlrpc_rule_set_choose to large nid

sptlrpc_rule_set_choose() and sptlrpc_target_choose_flavor() now take
a large nid. Only the net number is needed, so this is quite straight
forward.

Signed-off-by: Mr NeilBrown <neilb@suse.de>
Change-Id: Id4e8083d31c0393e2ef748babb6b851501b8d46f
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/50102
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Sebastien Buisson <sbuisson@ddn.com>
Reviewed-by: James Simmons <jsimmons@infradead.org>
Reviewed-by: Oleg Drokin <green@whamcloud.com>

LU-9680 lnet: collect data about routes by using Netlink

Migrate the LNet route API to use the Netlink API for
the case of collecting data about routes. This change also
allows large NID support for IPv6. Support for adding and
deleting routes with Netlink will be done in an follow on
patch.

Test-Parameters: trivial testlist=sanity-lnet
Change-Id: Ic8a9eb1f2571b14049a385698594849a585878ca
Signed-off-by: James Simmons <jsimmons@infradead.org>
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/49839
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Serguei Smirnov <ssmirnov@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Reviewed-by: Chris Horn <chris.horn@hpe.com>

LU-13805 llite: Fix return for non-queued aio

If an AIO fails or is completed synchronously (even
partially), the VFS will handle calling the completion
callback to finish the AIO, and so Lustre needs to return
the number of bytes successfully completed to the VFS.

This fixes a bug where if an AIO was racing with buffered
I/O, the AIO would fall back to buffered I/O, causing it to
complete before returning to the VFS rather than being
queued. In this case, Lustre would return 0 the VFS, and
the VFS would complete the AIO and report 0 bytes moved.

This fixes the logic for this.

Signed-off-by: Patrick Farrell <pfarrell@whamcloud.com>
Change-Id: I9306402201e2962bbff04a4264c37bd0f1eca7b7
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/49915
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Zhenyu Xu <bobijam@hotmail.com>
Reviewed-by: Qian Yingjin <qian@ddn.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>

LU-16717 mdt: resume dir migration with bad_type

LFSCK may set hash type to "none,bad_type" upon migration failure,
set it back to "fnv_1a_64,migrating,bad_type,fixed" to allow
migration resumption. fnv_1a_64 is set because it's the default hash
type, and now that we don't know the hash type in the original
migration command, just try with it.

LFSCK just add "bad_type" flag on such directory, so that such
migration can always be resumed in the future.

Add sanity 230z.

Signed-off-by: Lai Siyao <lai.siyao@whamcloud.com>
Change-Id: I19606aefcb9115e6724843785aea89a1c380e23f
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/50797
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: Hongchao Zhang <hongchao@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>

LU-16717 mdt: treat unknown hash type as sane type

Directory migration failure may leave directory hash type as
LMV_HASH_TYPE_UNKNOWN|LMV_HASH_FLAG_BAD_TYPE, which should be treated
as sane hash type on existing directories, otherwise such directories
can't be unlinked.

Add sanity 230y.

Signed-off-by: Lai Siyao <lai.siyao@whamcloud.com>
Change-Id: Ieffc0808d1db989d0bf9723f05cddb06f349e208
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/50796
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: Hongchao Zhang <hongchao@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>

LU-16634 misc: replace obsolete ioctl numbers

Renumber the OBD_IOC_BARRIER and IOC_OSC_SET_ACTIVE ioctl numbers
to use IOC type 'f' instead of 'g' and 'h', respectively. Add
compatibility code to handle their deprecation in the future.

Move already-deprecated ioctls OBD_IOC_GETNAME_OLD, OBD_GET_VERSION,
IOC_LIBCFS_GET_NI, and IOC_LIBCFS_PING to lustre_ioctl_old.h also.
Remove other already obsolete ioctl number definitions.

Add new llapi_ioctl() and llapi_ioctl_dev() helper functions to
handle fallback of ioctl commands to older versions when needed.

Signed-off-by: Andreas Dilger <adilger@whamcloud.com>
Change-Id: I519f23d34ce6197f3430a3825c5190cedb00cf14
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/50390
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Arshad Hussain <arshad.hussain@aeoncomputing.com>
Reviewed-by: Vitaliy Kuznetsov <vkuznetsov@ddn.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>

LU-16756 kernel: RHEL 9.2 server support

This patch makes changes to support RHEL 9.2 release
with kernel 5.14.0-284.11.1.el9_2 for Lustre server.

Test-Parameters: trivial env=SANITY_EXCEPT=27J fstype=ldiskfs \
clientdistro=el9.2 serverdistro=el9.2 testlist=sanity

Test-Parameters: trivial env=SANITY_EXCEPT=27J fstype=zfs \
clientdistro=el9.2 serverdistro=el9.2 testlist=sanity

Change-Id: I8fbffa2d9e48dabe62b6affadd11e7b03ae1aee7
Signed-off-by: Jian Yu <yujian@whamcloud.com>
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/50852
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Yang Sheng <ys@whamcloud.com>
Reviewed-by: Li Dongyang <dongyangli@ddn.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>

LU-16755 kernel: update RHEL 8.8 [4.18.0-477.10.1.el8_8]

Update RHEL 8.8 kernel to 4.18.0-477.10.1.el8_8.

Test-Parameters: trivial fstype=ldiskfs \
clientdistro=el8.8 serverdistro=el8.8 testlist=sanity

Test-Parameters: trivial fstype=zfs \
clientdistro=el8.8 serverdistro=el8.8 testlist=sanity

Change-Id: I6d7703512f9c5a8b686f06e94f32f0e51c9b2001
Signed-off-by: Jian Yu <yujian@whamcloud.com>
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/51051
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: Peter Jones <pjones@whamcloud.com>
Reviewed-by: Yang Sheng <ys@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>

LU-12353 ldiskfs: add ext4-dquot-commit-speedup patch to more series

Add ext4-dquot-commit-speedup.patch to RHEL 8.x and RHEL 9.x
ldiskfs patch series.

Test-Parameters: trivial clientdistro=el8.7 serverdistro=el8.7 testlist=sanity
Test-Parameters: trivial clientdistro=el8.8 serverdistro=el8.8 testlist=sanity
Test-Parameters: trivial clientdistro=el9.1 serverdistro=el9.1 testlist=sanity

Change-Id: Ib0ac325bde442b4eafedde9ba44984b02d5ea061
Fixes: dad25f258e50 ("LU-12353 ldiskfs: speedup quota journalling")
Signed-off-by: Jian Yu <yujian@whamcloud.com>
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/50853
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: Li Dongyang <dongyangli@ddn.com>

Revert "LU-15969 llite: add support for ->fileattr_get/set"

This causes Lu-15969

This reverts commit 1addb410788f0415a85bef72afff8533ac9da012.

Reason for revert: <INSERT REASONING HERE>

Change-Id: I074452fc6267a9f06931232f6b24372dd4211fa4
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/51093
Tested-by: Oleg Drokin <green@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>

LU-16835 target: server_ioctl() should return ENOTTY

server_ioctl() returns EOPNOTSUPP when ioctl cmd
is not supported. This will confuse e2fsprogs >= 1.47
as it's trying to get/set fs label using
FS_IOC_GETFSLABEL and FS_IOC_SETFSLABEL ioctl.

If the cmd is not supported we really should return
ENOTTY which means "Inappropriate ioctl for device".

Change-Id: Ibca0e6ad9a5c19cc7bd55ebd96cf32af05cf0fb6
Test-Parameters: trivial
Fixes: d5be104cc9 ("LU-11355 lustre: enable fstrim on lustre device")
Signed-off-by: Li Dongyang <dongyangli@ddn.com>
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/51072
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>

LU-16790 man: fix lctl-nodemap-modify man page

The section about the rbac property is malformed, regarding line feeds
in particular.
The forbid_encryption property is not mentioned.
Default values for admin, trusted, deny_unknown, audit_mode and
readonly_mount properties are not expressed correctly. These
properties grok 0 or 1.

Fixes: 22bef9b6c6 ("LU-16524 sec: add fscrypt_admin rbac role")
Test-Parameters: trivial
Signed-off-by: Sebastien Buisson <sbuisson@ddn.com>
Change-Id: I712e4d21b3038d87c50bb12206cbcc92ef4b615f
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/50918
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: Patrick Farrell <pfarrell@whamcloud.com>

LU-7495 utils: add --links option for lfs find

This adds a "--links" option for lfs find to filter files and
directories by the number of hard links. It also adds a printf format
'%n' to print the number of links for a file.

This commit also fixes '-l' as a short option for '--lazy' which was
added in 11aa7f8704c490b011f60f234c3ac9929ce76948 but the short option
did not work.

Signed-off-by: Thomas Bertschinger <bertschinger@lanl.gov>
Change-Id: I5d15bc290df8e8a08402f8d5cfa0a7139791b0a4
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/50886
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: Anjus George <georgea@ornl.gov>
Reviewed-by: Oleg Drokin <green@whamcloud.com>

LU-16803 build: Map uuid_le to guid_t for in-kernel APIs

Linux commit v6.2-rc5-72-g5e6a51787fef
uuid: Decouple guid_t and uuid_le types and respective macros

Removed uapi guid_t however sf_uuid is used in kernel which
must be a guid_t while uapi only has uuid_le.

Test-Parameters: trivial
Fixes: 20621c761b ("LU-16655 scrub: change sf_uuid to guid_t")
Signed-off-by: Shaun Tancheff <shaun.tancheff@hpe.com>
Change-Id: I38e062b0e5ce9b8c32f60e41b7c95781717e1f97
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/50885
Tested-by: Maloo <maloo@whamcloud.com>
Tested-by: jenkins <devops@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Reviewed-by: James Simmons <jsimmons@infradead.org>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>

LU-16796 libcfs: Remove unsed LASSERT_ATOMIC_* macros

This is the first up series of patchs which
replaces atomic_t with refcount_t

This patch removes unused LASSERT_ATOMIC_* macros.
There are still:
LASSERT_ATOMIC_GT
LASSERT_ATOMIC_GT_LT
LASSERT_ATOMIC_ZERO
LASSERT_ATOMIC_POS
macros being used which will be subsequently removed
in following patches. This helps in breaking up patch
logically as changes are large.

Test-Parameters: trivial
Signed-off-by: Arshad Hussain <arshad.hussain@aeoncomputing.com>
Change-Id: I5c41f978d10b193365ffba6ebeca9aab2cead99c
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/50870
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Reviewed-by: Neil Brown <neilb@suse.de>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: James Simmons <jsimmons@infradead.org>

LU-16797 lnet: improve numeric NID to CPT hashing

Testing shows that the sum-by-multiplication of nid bytes method for
hashing a NID to a CPT does not have good distribution for gnilnd
and kfilnd NIDs. For these LNDs, use the old hash_long() method.

Also modify lustre_lnet_calc_cpt_of_nid() to ensure the specified
number of CPTs is greater than 0. This avoids a divide by zero.

Test-Parameters: trivial
HPE-bug-id: LUS-11576
Signed-off-by: Chris Horn <chris.horn@hpe.com>
Change-Id: I77c6a5c3bc6b773e97f0a3c87182c31674d284e0
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/50862
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: Serguei Smirnov <ssmirnov@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>

LU-16795 build: Update ZFS version to 2.1.11

Update ZFS version to 2.1.11. The changes are listed in:
https://github.com/openzfs/zfs/releases/tag/zfs-2.1.11

Change-Id: I51de752fd82174586bcda7a9a42152b9fb2111bd
Signed-off-by: Jian Yu <yujian@whamcloud.com>
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/50856
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Minh Diep <mdiep@whamcloud.com>
Reviewed-by: Colin Faber <cfaber@ddn.com>
Reviewed-by: Nathaniel Clark <nclark@whamcloud.com>
Reviewed-by: xinliang <xinliang.liu@linaro.org>
Reviewed-by: Oleg Drokin <green@whamcloud.com>

LU-12610 obd: remove OBD_ -> CFS_ macros

Remove OBD macros that are simply redefinitions
of CFS macros.

Signed-off-by: Timothy Day <timday@amazon.com>
Signed-off-by: Ben Evans <beevans@whamcloud.com>
Change-Id: Id9e312a6074c5e11370f018afd3201d73b53e7e0
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/50808
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: James Simmons <jsimmons@infradead.org>
Reviewed-by: Arshad Hussain <arshad.hussain@aeoncomputing.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>

LU-12610 tests: remove OBD_ -> CFS_ macros

Remove OBD macros that are simply redefinitions
of CFS macros.

Test-Parameters: trivial
Signed-off-by: Timothy Day <timday@amazon.com>
Signed-off-by: Ben Evans <beevans@whamcloud.com>
Change-Id: Ieab1ffb7d83446f24fbf5921459ca7b6aaa7b0e0
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/50807
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: James Simmons <jsimmons@infradead.org>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: Arshad Hussain <arshad.hussain@aeoncomputing.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>

LU-12610 llite: remove OBD_ -> CFS_ macros

Remove OBD macros that are simply redefinitions
of CFS macros.

Signed-off-by: Timothy Day <timday@amazon.com>
Signed-off-by: Ben Evans <beevans@whamcloud.com>
Change-Id: I7bbcc3e1fda6418c258eb4d1c52b929a7cf72ed1
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/50804
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: James Simmons <jsimmons@infradead.org>
Reviewed-by: Arshad Hussain <arshad.hussain@aeoncomputing.com>
Reviewed-by: Neil Brown <neilb@suse.de>
Reviewed-by: Oleg Drokin <green@whamcloud.com>

LU-16529 test: verify the grant > 0 before test

In the test 84 in sanity-quota, the grant should be larger than 0
after the dd complete, this patch adds check for it and forces the
quota to be re-integrated during the check.

Signed-off-by: Hongchao Zhang <hongchao@whamcloud.com>
Change-Id: I9cddcf0c4afc12a11f3535792262ebb35a1e480e
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/50799
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: Sergey Cheremencev <scherementsev@ddn.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>

LU-16782 kernel: update RHEL 9.1 [5.14.0-162.23.1.el9_1]

Update RHEL 9.1 kernel to 5.14.0-162.23.1.el9_1.

Test-Parameters: trivial fstype=ldiskfs \
clientdistro=el9.1 serverdistro=el9.1 testlist=sanity

Change-Id: I961bac2129b98da2950694fa03e0bf47b780d85c
Signed-off-by: Jian Yu <yujian@whamcloud.com>
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/50785
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Colin Faber <cfaber@ddn.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>

LU-16775 tests: cleanup target after sanity-sec test_31

sanity-sec test_31 adds an LNet network tcp999, and associated
servicenode param on MDS target. This param must be cleared when
exiting the test, otherwise it can lead to incorrect client HA
behavior, trying to reach out to the fake service nodes.

Test-Parameters: trivial
Test-Parameters: mdscount=2 mdtcount=4 osscount=1 ostcount=8 clientcount=2 \
clientselinux testlist=sanity-sec env=SHARED_KEY=true,ONLY="31 61"
Fixes: c508c94268 ("LU-16557 client: -o network needs add_conn processing")
Signed-off-by: Sebastien Buisson <sbuisson@ddn.com>
Change-Id: If3a1926855bd23e9154c9a32b7a555e934e94565
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/50766
Reviewed-by: Sarah Liu <sarah@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>

LU-16232 scripts: clean specific MDTs update llogs

Modify remove_updatelog script as following:
- avoid bash-specific process substitution for compatibility
- add '-m <index,...> option to select only specific MDTs to
process through their update llogs only
- add test to check script actions

Signed-off-by: Mikhail Pershin <mpershin@whamcloud.com>
Change-Id: Icd935d5f4964d364f38b4962355ef8a5e679dbb3
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/50558
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: Colin Faber <cfaber@ddn.com>

LU-16712 cksum: fix generating T10PI guard tags for partial brw page

To get better performance, we allocate a page as the buffer for
T10PI guard tags, we fill the buffer going over every page for brw,
when the buffer is considered full, we use
cfs_crypto_hash_update_page() to update the hash and reuse the buffer.

It doesn't work when there's a page in the brw gets clipped, and the
checksum sector size is 512. For a page with PAGE_SIZE of 4096,
and offset at 1024, we will end up with 6 guard tags, and won't have
enough space in the very end of the buffer for a full brw page, which
needs 8.

Work out the number of guard tags for each page, update the
checksum hash and reuse the buffer when needed.

Change-Id: Ic591e63b24534f2a42b670669520895cb35a9546
Fixes: b1e7be00cb ("LU-10472 osc: add T10PI support for RPC checksum")
Signed-off-by: Li Dongyang <dongyangli@ddn.com>
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/50540
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: Li Xi <lixi@ddn.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>

LU-13748 mdt: remove LASSERT in mdt_dump_lmm()

Change LASSERT() in mdt_dump_lmm() into a CERROR(), since this
function is normally used for debugging and it doesn't help to
crash the MDS just when it complaining about some file layout.

Add sanity test_27Cg that triggered this initially.

Test-Parameters: trivial
Signed-off-by: Andreas Dilger <adilger@whamcloud.com>
Change-Id: I27f0b4489de48d6f197e0143589d76428d85e549
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/50532
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Patrick Farrell <pfarrell@whamcloud.com>
Reviewed-by: Vitaliy Kuznetsov <vkuznetsov@ddn.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>

LU-16707 scripts: update license-status

Improve license-status script to parse SPDX text and
to only look at files tracked by git. This will make
it easier to track which files still need SPDX.

Test-Parameters: trivial
Signed-off-by: Timothy Day <timday@amazon.com>
Change-Id: I4ece6c7af6133d3e6fa0771dbf49bc511c5f853c
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/50515
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: Arshad Hussain <arshad.hussain@aeoncomputing.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>

LU-16466 kfilnd: Allow custom traffic class per-NI

Allow a traffic class to be specified per-NI. Per-NI values can be
specified via yaml configuration or the lnetctl CLI. A new kfilnd
module parameter, traffic_class, defines the default traffic class
when a per-NI value is not otherwise specified.

Here are the valid values and their associated mapping to kfabric:
- best_effort -> KFI_TC_BEST_EFFORT
- low_latency -> KFI_TC_LOW_LATENCY
- dedicated_access -> KFI_TC_DEDICATED_ACCESS
- bulk_data -> KFI_TC_BULK_DATA
- scavenger -> KFI_TC_SCAVENGER
- network_ctrl -> KFI_TC_NETWORK_CTRL

The default value of the traffic_class kfilnd parameter is
"best_effort".

Here's an example yaml configuration:
net:
    - net type: kfi1
      local NI(s):
        - interfaces:
              0: cxi0
          tunables:
          lnd tunables:
              traffic_class: bulk_data
    - net type: kfi2
      local NI(s):
        - interfaces:
              0: cxi1
          tunables:
          lnd tunables:
              traffic_class: low_latency

Here's an example of setting per-NI values using the lnetctl CLI:

$ lnetctl net add --net kfi1 --if cxi0 --traffic-class bulk_data
$ lnetctl net add --net kfi2 --if cxi0 --traffic-class low_latency

Both the string and numeric representation of the traffic class is
shown in the output of lnetctl net show. e.g.

$ lnetctl net show -v | egrep -e kfi -e traffic
    - net type: kfi1
        - nid: 0@kfi1
              traffic_class: bulk_data
              traffic_class_num: 515
    - net type: kfi2
        - nid: 0@kfi2
              traffic_class: low_latency
              traffic_class_num: 513
$

This is simply a debug tool to ensure the feature is working
correctly. "traffic_class_num" is ignored if it is specified as
part of a yaml configuration, and it is ommitted from lnetctl export
output when the --backup option is specified.

HPE-bug-id: LUS-10197
Test-Parameters: trivial
Signed-off-by: Chris Horn <chris.horn@hpe.com>
Change-Id: I014a323e675d608ee5b506d8676fcc2cc78e8c4a
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/50490
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Ian Ziemba <ian.ziemba@hpe.com>
Reviewed-by: Ron Gredvig <ron.gredvig@hpe.com>
Reviewed-by: James Simmons <jsimmons@infradead.org>
Reviewed-by: Oleg Drokin <green@whamcloud.com>

LU-16801 lnet: Use dynamic allocation for LND tunables

Increasing size of lnet_ioctl_config_lnd_tunables can cause us to
trip Werror=frame-larger-than warning in lnet_net_cmd() and its
call to static function lnet_genl_parse_local_ni() (seen on
ubuntu2004 and ubuntu2204.) Dynamically allocate this memory instead.

Also modify how LND tunables are initialized in
handle_yaml_config_ni() to match what is done in
lnet_genl_parse_local_ni().

Test-Parameters: trivial
Signed-off-by: Chris Horn <chris.horn@hpe.com>
Change-Id: I6d169ac44521e98958feee70fc5ed73ffb8a7fcd
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/50872
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: James Simmons <jsimmons@infradead.org>
Reviewed-by: Serguei Smirnov <ssmirnov@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>

LU-9859 libcfs: remove workitem.

There are no more users of the "workitem" code so it can be removed.
Lustre uses Linux workqueues instead.

Test-Parameters: trivial
Signed-off-by: Mr NeilBrown <neilb@suse.de>
Change-Id: I392ab73cdd38bc2927e0626d83a6f838dfd8113c
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/50462
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: Aurelien Degremont <adegremont@nvidia.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>

LU-930 doc: move lnet pages, remove old docs

Purge ancient lnet documentation. It doesn't seem to
have been updated in at least 20 years, based on the
git history.

In it's place, put the lnetctl and lst man pages. This
is a more sensible place for the lnet man pages to
live. This patch is preliminary to them being split
up and updated.

Test-Parameters: trivial
Signed-off-by: Timothy Day <timday@amazon.com>
Change-Id: I33a3f38bd0f6803406a235f2177bab21dbf7546b
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/50387
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Reviewed-by: Chris Horn <chris.horn@hpe.com>
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>

LU-15969 llite: add support for ->fileattr_get/set

From Linux 5.13, FS_IOC_SETFLAGS and GETFLAGS aren't passed down to
the filesystem, we need ->fileattr_get/set inode_operations instead.

Signed-off-by: Mr NeilBrown <neilb@suse.de>
Change-Id: I7c13640012f3e3b6176bcdcc78a07f30e0667cbb
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/50649
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Shaun Tancheff <shaun.tancheff@hpe.com>
Reviewed-by: James Simmons <jsimmons@infradead.org>
Reviewed-by: Oleg Drokin <green@whamcloud.com>

LU-9325 obd: replace simple_strtoul()

Replace the use of simple_strtoul() in filename_is_volatile() with
sscanf(). This change also strengthens the checking of the format
of the volatile file's name.

Remove the use of simple_strtoul() from target_name2index() as well.

Change-Id: I4bf15af00d6cecdd46eba3cca5df0fbeb473d9ce
Signed-off-by: James Simmons <jsimmons@infradead.org>
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/49910
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: Arshad Hussain <arshad.hussain@aeoncomputing.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>

LU-16340 quota: notify only global lqe

Don't notify slaves with new limits when set new
limits to the pools. Do this only for lqes that
belong to the global pool.

The fix helps to avoid a case when slaves do not
apply new limit because slaves` data version is
greater than the one comes from the MDT. It was
possible, if set a lot of times different limits to PQ.
After that new limits from the global pool could not
be be applied:

qsd_upd_schedule()) lustre-OST0000: discarding glb
update from glimpse ver:7 local ver:203

For details about the problem see "check indexes versions"
test in sanity-quota.sh.

Add test 25 "check indexes versions" into sanity-quota.
Without the fix it reproduces above problem.

Fix checkpatch to don't print "Invalid vsprintf pointer
extension" for %px.

HPE-bug-id: LUS-10705
Change-Id: Idb091a10894e9db9f67d215baef2926723d6c65d
Signed-off-by: Sergey Cheremencev <sergey.cheremencev@hpe.com>
Reviewed-on: https://es-gerrit.dev.cray.com/160536
Reviewed-by: Andrew Perepechko <andrew.perepechko@hpe.com>
Reviewed-by: Alexander Boyko <c17825@cray.com>
Tested-by: Elena Gryaznova <elena.gryaznova@hpe.com>
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/49239
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: Alexander Boyko <alexander.boyko@hpe.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>

LU-16331 utils: fix 'lfs find -O <uuid>' with gaps

Fix the UUID parsing in llapi_get_target_uuids() so that the OST
UUIDs are put into the right slots when there is a gap in numbering.
Otherwise, "lfs find -O <uuid>" will not be able to find the given
UUID if it is the first OST after the gap.

Add test case for 'lfs find' and 'lfs getstripe' with large/sparse
OST indices in conf-sanity test_81.

Test-Parameters: trivial testlist=conf-sanity env=ONLY="81-82"
Signed-off-by: Andreas Dilger <adilger@whamcloud.com>
Change-Id: Ia0581f85f016c202514148114924509118a0f792
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/49207
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Jian Yu <yujian@whamcloud.com>
Reviewed-by: Feng Lei <flei@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>

LU-10026 csdc: reserve layout bits for compress component

Add layout bits for compress component layout.

* lcme_compr_type: compression type (gzip, lz4, lz4hc, lzo, etc.)
* lcme_compr_lvl: compression level (0=default, 1-15)
* lcme_compr_chunk_log_bits: chunk size = 2^(16+chunk_log_bits)

Component pattern:
* LOV_PATTERN_COMPRESS - file contains compressed data chunks and
       cannot be read by a client without decompression support.

Compress component flags:
* LCME_FL_COMPRESS - the component should be compressed with the
       compression algorithm stored in lcme_comp_type, at level
       lcme_comp_level, with chunk size 2^(16+lcme_chunk_log_bits)
* LCME_FL_PARTIAL - the component holds some uncompressed chunks due
       to poor write size/alignment, and may benefit from being
       recompressed as the full file data is available
* LCME_FL_NOCOMPR - the component should not be compressed because
       the data was found to be incompressible, or by user request

Signed-off-by: Bobi Jam <bobijam@whamcloud.com>
Change-Id: Idca22cca87b01bba8a5b3c85ca62044abe1d30eb
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/49170
Reviewed-by: Artem Blagodarenko <ablagodarenko@ddn.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>

LU-14294 tests: fixed NFS configuration issue

* Used the systemctl command to manage system services
* Used the same order of parameters to setup and cleanup NFS
* Used tab for indentation

Test-Parameters: trivial clientdistro=el7.9 \
testlist=parallel-scale-nfsv3,parallel-scale-nfsv4
Test-Parameters: clientdistro=el8.7 \
testlist=parallel-scale-nfsv3,parallel-scale-nfsv4
Test-Parameters: clientdistro=el9.0 \
testlist=parallel-scale-nfsv3,parallel-scale-nfsv4
Test-Parameters: clientdistro=sles12sp5 \
testlist=parallel-scale-nfsv3,parallel-scale-nfsv4
Test-Parameters: clientdistro=sles15sp4 \
testlist=parallel-scale-nfsv3,parallel-scale-nfsv4
Test-Parameters: clientdistro=ubuntu2004 \
testlist=parallel-scale-nfsv3,parallel-scale-nfsv4
Test-Parameters: clientdistro=ubuntu2204 \
testlist=parallel-scale-nfsv3,parallel-scale-nfsv4

Change-Id: I6b087035ac7524aa99c0facad48f8c3fb7444cbc
Signed-off-by: Alex Deiter <adeiter@tintri.com>
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/49062
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: Jian Yu <yujian@whamcloud.com>
Reviewed-by: James Simmons <jsimmons@infradead.org>
Reviewed-by: Oleg Drokin <green@whamcloud.com>

LU-16212 kfilnd: net and lnd tunables wrong

The kfilnd net and lnd tunables were not getting set properly in
many cases. Tunables not explcitly set were not set to the correct
default values.

Fixed logic in kfilnd_tunables_setup().

Improved error messages in kfilnd_startup() and
kfilnd_tunables_setup().

HPE-bug-id: LUS-11164
Test-Parameters: trivial
Signed-off-by: Ron Gredvig <ron.gredvig@hpe.com>
Change-Id: I49807b80466f5cbee65a9b3a7ee5ec2fe0c9e141
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/48779
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: James Simmons <jsimmons@infradead.org>
Reviewed-by: Ian Ziemba <ian.ziemba@hpe.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>

LU-13199 lustre: remove cl_{offset,index,page_size} helpers

These helpers could be replaced with PAGE_SIZE and PAGE_SHIFT calculation
directly which avoid CPU overhead.

Change-Id: I624136d4399a03e599f09f00a77b86de045f19e9
Signed-off-by: Wang Shilong <wshilong@ddn.com>
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/37426
Reviewed-by: Neil Brown <neilb@suse.de>
Reviewed-by: James Simmons <jsimmons@infradead.org>
Reviewed-by: Patrick Farrell <pfarrell@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>

LU-9859 lnet: convert selftest to use workqueues

Instead of the cfs workitem library, use workqueues.

As lnet wants to provide a cpu mask of allowed cpus, it
needs to be a WQ_UNBOUND work queue so that tasks can
run on cpus other than where they were submitted.
We use alloc_ordered_workqueue for lst_sched_serial (now called
lst_serial_wq) - "ordered" means the same as "serial" did.
We use cfs_cpt_bind_queue() for the other workqueues which sets up the
CPU mask as required.

An important difference with workqueues is that there is no equivalent
to cfs_wi_exit() which can be called in the action function and which
will ensure the function is not called again - and that the item is no
longer queued.

To provide similar semantics we treat swi_state == SWI_STATE_DONE as
meaning that the wi is complete and any further calls must be no-op.
We also call cancel_work_sync() (via swi_cancel_workitem()) before
freeing or reusing memory that held a work-item.

To ensure the same exclusion that cfs_wi_exit() provided the state is
set and tested under a lock - either crpc_lock, scd_lock, or tsi_lock
depending on which structure the wi is embedded in.

Another minor difference is that with workqueues the action function
returns void, not an int.

Also change SWI_STATE_* from #define to an enum. The only place these
values are ever stored is in one field in a struct.

Linux-commit: 6106c0f82481e686b337ee0c403821fb5c3c17ef
Linux-commit: 3fc0b7d3e0a4d37e4c60c2232df4500187a07232
Linux-commit: 7d70718de014ada7280bb011db8655e18ed935b1

Test-Parameters: trivial testlist=lnet-selftest
Change-Id: I5ccf1399ebbfdd4cab3696749bd1ec666147b757
Signed-off-by: Mr NeilBrown <neilb@suse.de>
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/36991
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: James Simmons <jsimmons@infradead.org>
Reviewed-by: Serguei Smirnov <ssmirnov@whamcloud.com>
Reviewed-by: Chris Horn <chris.horn@hpe.com>
Reviewed-by: Frank Sehr <fsehr@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>

LU-12273 obd: Reserve metadata overstriping flags

Reserve flag bits for metadata overstriping.

Test-Parameters: trivial
Signed-off-by: Patrick Farrell <farr0186@gmail.com>
Change-Id: I894b9420a4b08cceaccca6b3184ecb3bd22a680c
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/49707
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Qian Yingjin <qian@ddn.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>

LU-137 osd-ldiskfs: pass through resize ioctl

Pass through the EXT4_IOC_RESIZE_FS ioctl to the underlying ldiskfs
code so that it is possible to online resize MDT and OST filesystems.

When running resize2fs against a filesystem, it compares st_rdev of
the block device against st_dev of the mounted filesystem, so the
mounted Lustre stub filesystem needs to return proper stat information
from the ldiskfs root directory. Add in a server_getattr() method
to the server inode_operations. Using generic_fillattr() is enough,
we don't need the added complexity of calling ext4_getattr() (which
does not exist on directories for all kernel versions).

Change the OSD API from returning the superblock with dt_mnt_sb_get()
to returning the vfsmount with dt_mnt_get(), since it also contains
the superblock, but is more useful for calling some inode methods.

Signed-off-by: Andreas Dilger <adilger@whamcloud.com>
Change-Id: I934ae1f495bd15c6435be81b51ed04f0986c0322
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/20161
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Alex Deiter <alex.deiter@gmail.com>
Reviewed-by: Li Dongyang <dongyangli@ddn.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>

LU-6142 obd: change lmd flags to bitmap

Change lmd flags to an enum that is accessible with the Linux
bitmap API. This lays the foundation for creating a match table
for the server options for mounting.

Change-Id: If7906a9a3ba177b67d0cfbaa276a00a6ba9b7b6d
Signed-off-by: James Simmons <jsimmons@infradead.org>
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/49912
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Neil Brown <neilb@suse.de>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>

LU-15934 lod: renew the update llog

Skip and renew the update llog file while it was
corrupted.

Signed-off-by: Yang Sheng <ys@whamcloud.com>
Change-Id: I3491858dce42b4a8ed11db55ebbf8a12ef5f521d
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/49569
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: Lai Siyao <lai.siyao@whamcloud.com>
Reviewed-by: Mikhail Pershin <mpershin@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>

LU-11623 mdt: return UPDATE|PERM on open

This patch includes the following changes:
* try lock UPDATE|PERM ibits on open to speed up subsequent stat.
* open returns PR lock to client by default, because CR mode
  UPDATE|PERM ibits don't make sense since all modifications take
  PW lock.
* don't lock UPDATE|PERM ibits for PCC attach, otherwise these ibits
  revoke will cause file detach.
* update sanity-pcc 13a to make it fail on single client test if
  anything went wrong.
* update sanity-lfsck 31d because previously CR UPDATE lock is
  fetched, thus the test pass by mistake.

This should help common workloads with open followed by a stat
or other such operation.

Benchmark results:

This patch can significantly improve open-create + stat on the same
client.

This patch in combination with two others:

https://review.whamcloud.com/32157
https://review.whamcloud.com/33584

Improves the 'stat' side of open-create + stat by >10x.

Without patches (master branch commit 26a7abe):

mpirun -np 24 --allow-run-as-root /work/tools/bin/mdtest -n 50000 -d
/cache1/out/ -F -C -T -v -w 32k

   Operation                      Max            Min           Mean
   ---------                      ---            ---           ----
   File creation     :       3838.205       3838.204       3838.204
   File stat         :      33459.289      33459.249      33459.271
   File read         :          0.000          0.000          0.000
   File removal      :          0.000          0.000          0.000
   Tree creation     :       3146.841       3146.841       3146.841
   Tree removal      :          0.000          0.000          0.000

With the three patches:

mpirun -np 24 --allow-run-as-root /work/tools/bin/mdtest -n 50000 -d
/cache1/out/ -F -C -T -v -w 32k
SUMMARY rate: (of 1 iterations)
   Operation                      Max            Min           Mean
   ---------                      ---            ---           ----
   File creation     :       3822.440       3822.439       3822.440
   File stat         :     350620.140     350615.980     350617.193
   File read         :          0.000          0.000          0.000
   File removal      :          0.000          0.000          0.000
   Tree creation     :       2076.727       2076.727       2076.727
   Tree removal      :          0.000          0.000          0.000

Note 33K stats/second vs 350K stats/second.

ls -l time of the mdtest directory is also reduced from 23.5 seconds
to 5.8 seconds.

Change-Id: Ib3410629c190de6f74246a4a92f8216537fa2b95
Signed-off-by: Oleg Drokin <green@whamcloud.com>
Signed-off-by: Lai Siyao <lai.siyao@whamcloud.com>
Signed-off-by: Qian Yingjiin <qian@ddn.com>
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/33585
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Tested-by: Shuichi Ihara <sihara@ddn.com>
Reviewed-by: Mikhail Pershin <mpershin@whamcloud.com>
Reviewed-by: James Simmons <jsimmons@infradead.org>

LU-15529 mdt: optimize dir migration locking

Optimize dir migration locking and fix some deadlocks:
* don't lock all stripes of parent, but source and target parent
  stripes only.
* use mdt_rename_source_lock() to lock source, because directory
  stripes are not changed in migration.
* refactor migrate links locking code.
* pass spobj and tpobj to mdo_migrate() interface to avoid parsing
  parent directory layout in MDD layer again.
* never lock the same FID twice, which may lead to deadlock:
  . if link parent is local, don't hold local LOOKUP lock, but revoke
    only, because later we need to lock other ibits of sobj.
  . if sobj is plain directory, unlock sobj before locking tobj,
    because sobj will become a stripe of tobj during migration.
* enable striped directory migration in racer test.

Also update sanityn 80b to migrate directory and access it for one
minute and verify filesystem is not broken, though both migration and
directory access may fail in this period.

Test-Parameters: env=SLOW=y mdscount=2 mdtcount=4 testlist=racer,racer,racer
Signed-off-by: Lai Siyao <lai.siyao@whamcloud.com>
Change-Id: Ie9859df244529c986c2f3f032a49e3f9c89a2747
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/40891
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: Qian Yingjin <qian@ddn.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>