Pack parent FID in getattr request if OBD_CONNECT2_GETATTR_PFID is
enabled, otherwise fill it with target FID for backward compatibility.

Fixes: f9a2da63

 ("LU-13437 mdt: don't fetch LOOKUP lock for remot...")
Test-Parameters: clientversion=2.12 testlist=sanity env=SANITY_EXCEPT="27M 151 156"
Test-Parameters: serverversion=2.12 testlist=sanity env=SANITY_EXCEPT="56 165 205b"
Signed-off-by: Lai Siyao <lai.siyao@whamcloud.com>
Change-Id: Idcf8388b65dee1f0a09a53b240ce8303f3c6ff75
Reviewed-on: https://review.whamcloud.com/39290


Tested-by: jenkins <devops@whamcloud.com>
Reviewed-by: Neil Brown <neilb@suse.de>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>

Before the MR feature was implemented, it was not possible to
configure multiple interfaces on the same LNet, so the ip2nets
syntax did not allow for this. Now that we have MR feature, we should
allow it to be configured via ip2nets syntax. e.g.

o2ib(ib0) 10.10.10.1
o2ib(ib1) 10.10.10.2

A test is added for configuring LNet with kernel ip2nets parameter.

setup_netns() refactored to facilitate the new test.

cleanup_lnet() is modified to check whether lnet module is loaded
before attempting lnetctl lnet unconfigured otherwise sanity-lnet.sh
could exit with rc 234 on cleanup.

Test-Parameters: trivial testlist=sanity-lnet
HPE-bug-id: LUS-9046
Signed-off-by: Chris Horn <chris.horn@hpe.com>
Change-Id: Iafc3882035269073fd7e4abb53d138d9267f6e21
Reviewed-on: https://review.whamcloud.com/39227


Reviewed-by: James Simmons <jsimmons@infradead.org>
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Neil Brown <neilb@suse.de>
Reviewed-by: Oleg Drokin <green@whamcloud.com>

This refcount is not explicitly initialized, so is implicitly
initialized to zero.  This prohibits the use
lnet_peer_ni_addref_locked() for taking the first reference,
so a couple of places open-code the atomic_inc just in case.

There is code in lnet_peer_add_nid() which drops a reference before
accessing the structure.  This isn't actually wrong, but it looks
weird.

lnet_destroy_peer_ni_locked() makes assumptions about the content of
the structure, so it cannot be used on a partially initialized
structure.

All these special cases make the code harder to understand.  This
patch cleans this up:

- lpni_refcount is now initialized to one, so the called for
  lnet_peer_ni_alloc() now owns a reference and must be sure
  to release it.
- lnet_peer_attach_peer_ni() now consumes a reference to
  the lpni.  A pointer returned by lnet_peer_ni_alloc()
  is most often passed to lnet_peer_attach_peer_ni() so
  these to changes largely cancel each other out - not completely
- The two 'atomic_inc' calls are changed to
  'lnet_peer_ni_addref_locked().
- A LIBCFS_FREE() is replaced by lnet_peer_ni_decref_locked(),
  and that function is improved to cope with lpni_hashlist
  being empty, or ->lpni_net being NULL.
- lnet_peer_add_nid() now holds a reference on the lpni until
  it don't need it any more, then explicity drops it.

This should make no functional change, but should make the code a
little less confusing.

Signed-off-by: Mr NeilBrown <neilb@suse.de>
Change-Id: Iec312e637d1e7b6eb14f2c363843403dd5cf8e8f
Reviewed-on: https://review.whamcloud.com/39120


Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Chris Horn <chris.horn@hpe.com>
Reviewed-by: James Simmons <jsimmons@infradead.org>
Reviewed-by: Oleg Drokin <green@whamcloud.com>

lnet_print_hdr() is unused, and has not been used in git history since
at least 2004.
So remove it.

Test-Parameters: trivial
Signed-off-by: Mr NeilBrown <neilb@suse.de>
Change-Id: I0d72726b69f205a8c62ae4dbf1423f6e745db5fe
Reviewed-on: https://review.whamcloud.com/39358


Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: James Simmons <jsimmons@infradead.org>
Reviewed-by: Chris Horn <chris.horn@hpe.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>

Noe of ksocknal_query(), ksocknal_notify(), and
ksocknal_lib_bind_thread_to_cpu() exist, so don't declare them.

Test-Parameters: trivial
Signed-off-by: Mr NeilBrown <neilb@suse.de>
Change-Id: Ice87c317f116bda9c04dcaa285bc7ba47be219ca
Reviewed-on: https://review.whamcloud.com/39326


Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: James Simmons <jsimmons@infradead.org>
Reviewed-by: Alexey Lyashkov <alexey.lyashkov@hpe.com>
Reviewed-by: Chris Horn <chris.horn@hpe.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>

Edge (final-hop) routers typically use the non-multi-rail destination
(NMR_DST) send case. i.e. they treat the destination as
non-multi-rail. The reason for this is that we do not want routers to
modify the destination peer interface selected by the message
originator. As a result of using the NMR_DST send case, edge routers
set a preferred NI, and then continue to use that NI, because it's
preferred, even if the NI goes down and the router has other healthy
interfaces available to it. Routers do not need to use the preferred
NI selection logic when they are forwarding a message, so modify the
NMR_DST algorithm to allow routers to select any suitable local NI.

Test-Parameters: trivial
HPE-bug-id: LUS-9045
Signed-off-by: Chris Horn <chris.horn@hpe.com>
Change-Id: Iae0fb47d58a70f640d316a8c85cf3058ca2f82eb
Reviewed-on: https://review.whamcloud.com/39168


Reviewed-by: Serguei Smirnov <ssmirnov@whamcloud.com>
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Neil Brown <neilb@suse.de>
Reviewed-by: Oleg Drokin <green@whamcloud.com>

Add the appropriate option to the MD used for LNet pings and pushes
to ensure that these are always tracked via LNet's response tracking
mechanism, regardless of the value of lnet_response_tracking
variable.

Test-Parameters: trivial
HPE-bug-id: LUS-8827
Signed-off-by: Chris Horn <hornc@cray.com>
Change-Id: I13d8ee42ccbb00c85843f64314b1f953d679a0dc
Reviewed-on: https://review.whamcloud.com/38451


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: James Simmons <jsimmons@infradead.org>
Reviewed-by: Oleg Drokin <green@whamcloud.com>

Add lnet_response_tracking parameter which will be used to control
the behavior of LNet response tracking.

Test-Parameters: trivial
HPE-bug-id: LUS-8827
Signed-off-by: Chris Horn <hornc@cray.com>
Change-Id: I9c5be488673bbaa3c3cb983fe099d2203c1d9fa7
Reviewed-on: https://review.whamcloud.com/38449


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: Oleg Drokin <green@whamcloud.com>

Change-Id: Iefb108c0dc97b0c69407e07ca08bcfc14f7dbfe2
Signed-off-by: Oleg Drokin <green@whamcloud.com>

Add OBD_CONNECT2_GETATTR_PFID connect flag to pack parent FID in
getattr request, which will be used to check whether target is
remote object, if so, don't take LOOKUP lock, otherwise client
may see stale directory entries.

Test-parameters: trivial

Signed-off-by: Lai Siyao <lai.siyao@whamcloud.com>
Change-Id: Ibdf880934456f255f83cd4bac9d61ab5e1ed7330
Reviewed-on: https://review.whamcloud.com/39289


Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: Neil Brown <neilb@suse.de>
Reviewed-by: Oleg Drokin <green@whamcloud.com>

In RHEL 8.2 kernel 4.18.0-193.el8, VM_FAULT_RETRY is
defined as an enumeration constant in linux/mm_types.h
instead of a macro in linux/mm.h. This patch adds
autoconf macros to check if VM_FAULT_RETRY is defined
at configure time.

Test-Parameters: clientdistro=el8.2 serverdistro=el8.2 \
testlist=sanity

Test-Parameters: clientdistro=el8.1 serverdistro=el8.1 \
testlist=sanity

Fixes: 2e813f3e

 ("LU-13731 llite: include linux/mm_types.h for VM_FAULT_RETRY")
Change-Id: I2fdae7b62a53e447a7eb979787bdbd79423b787d
Signed-off-by: Jian Yu <yujian@whamcloud.com>
Reviewed-on: https://review.whamcloud.com/39281


Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: Ben Menadue <ben.menadue@anu.edu.au>
Reviewed-by: Oleg Drokin <green@whamcloud.com>

This patch is to match the root path in function get_root_path()
correctly. For example, if the mount point is /mnt/lustre, the
following root path formats are acceptable:
- /mnt/lustre
- /mnt/lustre/*

sanity.sh test_154A/247d are modified to verify this patch.

Signed-off-by: Emoly Liu <emoly@whamcloud.com>
Change-Id: If705dd341b273d462aeba280fa27d5608b5f3b7c
Reviewed-on: https://review.whamcloud.com/39225


Tested-by: jenkins <devops@whamcloud.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Jian Yu <yujian@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>

In osc_brw_fini_request(), page needs decryption only if it
is not empty. To check this, use ZERO_PAGE macro available
for all architectures, and compare with memcmp.
It will likely be faster/more efficient than comparing the
words by hand as may use optimized CPU instructions or ASM code.

Signed-off-by: Sebastien Buisson <sbuisson@ddn.com>
Change-Id: I5e04b72790e8acbceb1989ba3659e170c0b11192
Reviewed-on: https://review.whamcloud.com/38918


Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: James Simmons <jsimmons@infradead.org>
Tested-by: jenkins <devops@whamcloud.com>
Reviewed-by: Mike Pershin <mpershin@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>

On client side, data read from DoM files do not go through the OSC
layer. So implement file decryption in ll_dom_finish_open() right
after file data has been put in cache pages.
On server side, DoM file size needs to be properly set on MDT when
content is encrypted. Pages are full of encrypted data, but inode size
must be apparent, clear text object size.
For reads of DoM encrypted files to work proprely, we also need to
make sure we send whole encryption units to client side.
Also add sanity-sec test_50 to exercise encryption of DoM files.

Test-Parameters: testlist=sanity-sec envdefinitions=ONLY="36 37 38 39 40 41 42 43 44 45 46 47 48 49 50" clientdistro=el8.1 fstype=ldiskfs mdscount=2 mdtcount=4
Test-Parameters: testlist=sanity-sec envdefinitions=ONLY="36 37 38 39 40 41 42 43 44 45 46 47 48 49 50" clientdistro=el8.1 fstype=zfs mdscount=2 mdtcount=4
Signed-off-by: Sebastien Buisson <sbuisson@ddn.com>
Change-Id: I7721ca4085373a7a01b2062c37458a7136e646e0
Reviewed-on: https://review.whamcloud.com/38702


Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: Mike Pershin <mpershin@whamcloud.com>
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>

In case some buffers need to be moved because of segment growth
from req_capsule_server_grow(), just set buflen to old length
before actually calling lustre_grow_msg().

Signed-off-by: Sebastien Buisson <sbuisson@ddn.com>
Change-Id: I6707927a0f24c0637dbc79aa91788122a84ab8c4
Reviewed-on: https://review.whamcloud.com/38701


Reviewed-by: Mike Pershin <mpershin@whamcloud.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>

Add sanity-quota_71a that does write to a file
consisted of 2 components on different OSTs(each OST
relates to unique pool). Check that limits in quota
pools work properly. sanity-quota_71b does the same
but for a file with SEL.

Test-Parameters: envdefinitions=ONLY=71 testlist=sanity-quota
Change-Id: I835bec4c9b21c287142e1df9b4cfe797ec68fbef
Signed-off-by: Sergey Cheremencev <sergey.cheremencev@hpe.com>
Reviewed-on: https://review.whamcloud.com/38661


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>

Encryption layer needs to set an encryption context on files and dirs
that are encrypted. This context is stored as an extended attribute,
that then needs to be fetched upon metadata ops like lookup, getattr,
open, truncate, and layout.

With this patch we send encryption context to the MDT along with
create RPCs. This closes the insecure window between creation and
setting of the encryption context, and saves a setxattr request.

This patch also introduces a way to have the MDT return encryption
context upon granted lock reply, making the encryption context
retrieval atomic, and sparing the client an additional getxattr
request.

Test-Parameters: testlist=sanity-sec envdefinitions=ONLY="36 37 38 39 40 41 42 43 44 45 46 47 48 49" clientdistro=el8.1 fstype=ldiskfs mdscount=2 mdtcount=4
Test-Parameters: testlist=sanity-sec envdefinitions=ONLY="36 37 38 39 40 41 42 43 44 45 46 47 48 49" clientdistro=el8.1 fstype=zfs mdscount=2 mdtcount=4
Test-Parameters: clientversion=2.12 env=SANITY_EXCEPT="27M 56ra 151 156 802"
Test-Parameters: serverversion=2.12 env=SANITY_EXCEPT="56oc 56od 165a 165b 165d 205b"
Test-Parameters: serverversion=2.12 clientdistro=el8.1 env=SANITYN_EXCEPT=106,SANITY_EXCEPT="56oc 56od 165a 165b 165d 205b"
Signed-off-by: Sebastien Buisson <sbuisson@ddn.com>
Change-Id: I45599cdff13d5587103aff6edd699abcda6cb8f4
Reviewed-on: https://review.whamcloud.com/38430


Tested-by: jenkins <devops@whamcloud.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Mike Pershin <mpershin@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>

Force file/directory name encryption policy to null on newly created
inodes. This is required because first implementation step of client
side encryption only supports content encryption, and not names.
This imposes to force usage of embedded llcrypt lib to the detriment
of in-kernel fscrypt lib, even if the kernel provides it.

This patch will have to be reverted when name encryption is
implemented.

Test-Parameters: testlist=sanity-sec envdefinitions=ONLY="36 37 38 39 40 41 42 43 44 45 46 47 48" clientdistro=el8.1 fstype=ldiskfs mdscount=2 mdtcount=4
Test-Parameters: testlist=sanity-sec envdefinitions=ONLY="36 37 38 39 40 41 42 43 44 45 46 47 48" clientdistro=el8.1 fstype=zfs mdscount=2 mdtcount=4
Signed-off-by: Sebastien Buisson <sbuisson@ddn.com>
Change-Id: Ia697a29006507278c218088d7c3a5e5ade620a15
Reviewed-on: https://review.whamcloud.com/38882


Reviewed-by: James Simmons <jsimmons@infradead.org>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>

no writes to random Lustre source locations as they could be readonly

Test-Parameters: trivial
Test-Parameters: fstype=ldiskfs testlist=pjdfstest
Test-Parameters: fstype=zfs testlist=pjdfstest
Change-Id: Icd262a698390eadf4b53cd5d311bc6c2a561a79e
Signed-off-by: Oleg Drokin <green@whamcloud.com>
Reviewed-on: https://review.whamcloud.com/39338


Reviewed-by: James Nunez <jnunez@whamcloud.com>
Reviewed-by: Wei Liu <sarah@whamcloud.com>
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>

Introduce a "null" algorithm for client side filename encryption,
which is basically a no-op.
This is needed because first implementation step only supports
content encryption, and not names. So give the ability to support
encryption policies that have a 'filenames_encryption_mode' property
internally set to LLCRYPT_MODE_NULL.

Signed-off-by: Sebastien Buisson <sbuisson@ddn.com>
Change-Id: I3470f89f227b3a03e56766fe1ba5e36ae92ec27b
Reviewed-on: https://review.whamcloud.com/38881


Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Yingjin Qian <qian@ddn.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>

This iteration re-used the serial tests internal name for the
parallel test variant. This name clashing causes builds to fail.

Remedy this by adding a postfix tag the internal unique identifier
of the converted parallel tests.

Test-Parameters: trivial
Signed-off-by: Shaun Tancheff <shaun.tancheff@hpe.com>
Change-Id: Ia39bf2e0004abd10a0c7b146eb2ef1cf62e6d891
Reviewed-on: https://review.whamcloud.com/39161


Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: James Simmons <jsimmons@infradead.org>
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Patrick Farrell <farr0186@gmail.com>

some files are created with a sel layout if RACER_ENABLE_SEL is set

Test-Parameters: testlist=racer envdefinitions="RACER_ENABLE_SEL=true"
Signed-off-by: Vitaly Fertman <c17818@cray.com>
Change-Id: I1f699988fdd0a0dbb19b71d6ea353d68326d0d5b
Reviewed-by: Alexander Zarochentsev <c17826@cray.com>
Reviewed-by: Elena Gryaznova <c17455@cray.com>
Tested-by: Elena Gryaznova <c17455@cray.com>
Reviewed-on: https://es-gerrit.dev.cray.com/156106
HPE-bug-id: LUS-7591
Reviewed-on: https://review.whamcloud.com/37959


Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>

lod_statfs_and_check() may consider an OST as OK, but SEL needs its
extension size to fit the free space and the OST may turn out to be
low-on-space, thus inappropriate OST may be chosen, SEL cannot use
it and ENOSPC occurs.

Take the extension size into account at the object allocation time.
If none of OSTs is able to allocate an object, try to allocate in old
manner instead of returning ENOSPC from lod_qos_prep_create().

Adjust tests to work on smaller extension sizes to fit into OSTs
used for sanity tests to enable the new logic above.

Signed-off-by: Vitaly Fertman <c17818@cray.com>
Change-Id: If5159200a72a68a7261aa7031e58a1ac6a8e3f24
Reviewed-by: Alexey Lyashkov <c17817@cray.com>
Reviewed-by: Andriy Skulysh <c17819@cray.com>
Tested-by: Elena Gryaznova <c17455@cray.com>
Reviewed-on: https://es-gerrit.dev.cray.com/156372
HPE-bug-id: LUS-8157
Reviewed-on: https://review.whamcloud.com/37961


Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Alexey Lyashkov <alexey.lyashkov@hpe.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>

Linux 5.6 introduces proc_ops with v5.5-8862-gd56c0d45f0e2
proc: decouple proc from VFS with "struct proc_ops"

Separate debugfs usage and procfs usage to prepare for the divergence
of debugfs using file_operations and procfs using proc_ops

HPE-bug-id: LUS-8589
Signed-off-by: Shaun Tancheff <shaun.tancheff@hpe.com>
Change-Id: I1746e563b55a9e89f90ac01843c304fe6b690d8b
Reviewed-on: https://review.whamcloud.com/37834


Reviewed-by: Petros Koutoupis <petros.koutoupis@hpe.com>
Reviewed-by: Neil Brown <neilb@suse.de>
Reviewed-by: James Simmons <jsimmons@infradead.org>
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>

DIO reads in Lustre are protected by Lustre range locking
and do not need the inode mutex.  This code was removed
in LU-1669, the range lock was added for DIO reads in
LU-6227, and then the mutex was accidentally re-introduced
in LU-6260.

Remove it again.

Test-Parameters: envdefinitions=ONLY=16 testlist=sanityn,sanityn,sanityn,sanityn,sanityn,sanityn,sanityn,sanityn
Fixes: 98883bd3

 ("LU-6260 llite: add support for direct IO api changes")
Signed-off-by: Patrick Farrell <farr0186@gmail.com>
Change-Id: I5b3570e83a4b4ff36d9a22bc6bd3be5d5f991924
Reviewed-on: https://review.whamcloud.com/37419


Reviewed-by: Wang Shilong <wshilong@whamcloud.com>
Tested-by: jenkins <devops@whamcloud.com>
Reviewed-by: Neil Brown <neilb@suse.de>
Tested-by: Shuichi Ihara <sihara@ddn.com>
Reviewed-by: James Simmons <jsimmons@infradead.org>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>

This new C file contains acl related code so it can be
conditionally compiled.

Also remove conditional code around the call to mdc_unpack_acl(), as
those tests are already performed inside mdc_unpack_acl().

In the Makefile, use CONFIG_FS_POSIX_ACL to control conditional
compilation as CONFIG_LUSTRE_FS_POSIX_ACL is not available, but the
two are semantically equivalent in this context.

This removes all #ifdef of CONFIG_LUSTRE_FS_POSIX_ACL from C files in
'mdc'.

Signed-off-by: Mr NeilBrown <neilb@suse.de>
Change-Id: Ic9060bee4a2ad55580d8879fe32fee01b1cb8884
Reviewed-on: https://review.whamcloud.com/39292


Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: James Simmons <jsimmons@infradead.org>
Reviewed-by: Lai Siyao <lai.siyao@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>

This function needs to be called from the obdclass modules,
but is currently defined in a module that depends on that module.
The get around this interdependence, a global variable
  ptlrpc_put_connection_superhack
is used to make a pointer to the function available.

Rather than this hack, we can make ptlrpc_connection_put()
static-inline.  This does expose some details of ptlrpc to obdclass,
but there is already a fairly tight connection.

Also change the return value to 'void' as it is never used,
and don't bother checking for NULL before calling, as the
function has its own test for NULL.

Signed-off-by: Mr NeilBrown <neilb@suse.de>
Change-Id: I37333b78e410e8c46bddc468c31bed61dd9e7b33
Reviewed-on: https://review.whamcloud.com/39291


Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Sebastien Buisson <sbuisson@ddn.com>
Reviewed-by: Alex Zhuravlev <bzzz@whamcloud.com>
Reviewed-by: James Simmons <jsimmons@infradead.org>
Reviewed-by: Oleg Drokin <green@whamcloud.com>

obd_statfs() can call lmv_statfs() with ->obd_dev_mutex helds.
lmv_statfs will then call obd_statfs() on a different device
and ->obd_dev_mutex will be taken again.  This is a *different*
mutex, but lockdep cannot see the difference, so it complains.

We can tell lockdep not to worry in this case using
mutex_lock_interruptible_nested().

Test-Parameters: trivial
Signed-off-by: Mr NeilBrown <neilb@suse.de>
Change-Id: I0776407b722dd29ab1321289953b63f76fce7ceb
Reviewed-on: https://review.whamcloud.com/39248


Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: James Simmons <jsimmons@infradead.org>
Reviewed-by: Shaun Tancheff <shaun.tancheff@hpe.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>

Linux 5.3 changed struct cache_detail readers to writers.
As this mechanism is used by GSS authentication in Lustre via SunRPC,
we need to make sure lsvcgssd daemon does open
/proc/net/rpc/auth.sptlrpc.init/channel in R+W mode.

It also affects CentOS/RHEL 7.8, as the kernel commit was ported to
these distros.

Test-Parameters: trivial
Test-Parameters: env=SHARED_KEY=true,SANITY_EXCEPT="56w 405" mdscount=2 mdtcount=4 osscount=2 ostcount=8 testlist=sanity,recovery-small,sanity-sec
Signed-off-by: Sebastien Buisson <sbuisson@ddn.com>
Change-Id: If88802d4f2bc3168dda4f79fe57f2f44ac7ef39e
Reviewed-on: https://review.whamcloud.com/39297


Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Petros Koutoupis <petros.koutoupis@hpe.com>
Reviewed-by: John L. Hammond <jhammond@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>

This patch fixes another case where the truncate write intent
extent is set incorrectly.  This may cause errors when truncating
PFL files to exactly the boundary between two extents.

Fixes: c32c7401

 ("LU-12586 lov: Correct write_intent end for trunc")
Signed-off-by: Bobi Jam <bobijam@whamcloud.com>
Change-Id: I40f14a48c53bfd1e6442e4414ee30f9eb159fc02
Reviewed-on: https://review.whamcloud.com/38412


Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Wang Shilong <wshilong@whamcloud.com>
Reviewed-by: Li Dongyang <dongyangli@ddn.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>

It might be possible during OST mounting, two targets reach
iam_format_guess() at the same time, if @initialized is 0,
they both access iam_lxx_format_init(), however list operation
inside is not protected by any locking which cause list corruptions
finally.

We could fix this by doing formats registration in module init,
since there are only two formats, just remove pointless list.

Signed-off-by: Wang Shilong <wshilong@ddn.com>
Change-Id: I6dd5a4d1297792b47fb4b94052465a7e0f9123aa
Reviewed-on: https://review.whamcloud.com/39213


Tested-by: jenkins <devops@whamcloud.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: John L. Hammond <jhammond@whamcloud.com>
Reviewed-by: Yingjin Qian <qian@ddn.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>

The rtr_nid is specified for all REPLY/ACK. However it is possible
for the route through the gateway specified by rtr_nid to be removed.
In this case we don't want to use it. We should lookup alternative
paths.

This patch checks if the peer looked up is indeed a gateway. If it's
not a gateway then we attempt to find another path. There is no need
to fail right away. It's not a hard requirement to fail if the default
rtr_nid is not valid.

Test-Parameters: trivial
Signed-off-by: Amir Shehata <ashehata@whamcloud.com>
Change-Id: Ic1c93b7c6c2c8060e2cfeb8fb1cb875dbc3010f7
Reviewed-on: https://review.whamcloud.com/39175


Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Chris Horn <chris.horn@hpe.com>
Reviewed-by: Serguei Smirnov <ssmirnov@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>

This variable is treated like an atomic_t, but a global spinlock is
used to protect updates - and also unnecessarily to protect reads.

Change to atomic_t and avoid using the spinlock.

Signed-off-by: Mr NeilBrown <neilb@suse.de>
Change-Id: Id94d280875a9e115dc077253c49e97a725dc91e1
Reviewed-on: https://review.whamcloud.com/39121


Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: James Simmons <jsimmons@infradead.org>
Reviewed-by: Chris Horn <chris.horn@hpe.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>

The problem is currently AIO could not handle i/o size > stripe size:

We need cl io loop to handle io across stripes, since -EIOCBQUEUED is
returned for AIO, io loop will be stopped thus short io happen.

The patch try to fix the problem by making IO engine aware of
special error, and it will be proceed to finish all IO requests.

Fixes: d1dde ("LU-4198 clio: AIO support for direct IO")
Signed-off-by: Wang Shilong <wshilong@ddn.com>
Change-Id: I1885e0fc510571417d888249b381f9c2f130ca5a
Reviewed-on: https://review.whamcloud.com/39104


Tested-by: jenkins <devops@whamcloud.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Bobi Jam <bobijam@hotmail.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>

param->fp_xxx_index is the number of the specified targets, not
the specific index, so it can't be passed into llapi_obd_statfs()
directly to get the statfs information. Instead, all the targets
listed in param->fp_xxx_indexes should be passed one by one.

Also, $mdt_idx in sanity.sh test_56rb should be initialized with
the correct mdt index.

Signed-off-by: Emoly Liu <emoly@whamcloud.com>
Change-Id: Idffdddac1b5b249aa903b97912c767826f3b146c
Reviewed-on: https://review.whamcloud.com/38959


Reviewed-by: Mike Pershin <mpershin@whamcloud.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>

strtoul function return a 64bits value on a 64bits system, so an
overflow occurs when we store user value into a quota/project
structure.

This commit apply the same 32 bits verification for "lfs" project,
quota,setquota and find commands on uid, gid and project id arguments.

Signed-off-by: Etienne AUJAMES <eaujames@ddn.com>
Change-Id: I809e9ac55d4bc676c20b18c6c198a69eaba9cff6
Reviewed-on: https://review.whamcloud.com/38938


Reviewed-by: Wang Shilong <wshilong@whamcloud.com>
Tested-by: jenkins <devops@whamcloud.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>

In ptlrpc_pinger_main, if the process to ping the recoverable
clients or obd_update_maxusage takes too long time, it could
be stuck in endless loop because of the negative value returned
by pinger_check_timeout.

Change-Id: Ib7fc22b3cc31255223bc2be60224ced1a3585f87
Signed-off-by: Hongchao Zhang <hongchao@whamcloud.com>
Reviewed-on: https://review.whamcloud.com/38915


Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Olaf Faaland-LLNL <faaland1@llnl.gov>
Reviewed-by: Oleg Drokin <green@whamcloud.com>

The gnilnd receives node health information asynchronous from any tx
failure, so aliveness of lpni as reported by lnet_is_peer_ni_alive()
may not match what LND is telling us. Use existing reset flag to
set cached NI status down so we can be sure that remote NIs are
correctly set down.

Test-Parameters: trivial
HPE-bug-id: LUS-8897
Signed-off-by: Chris Horn <chris.horn@hpe.com>
Change-Id: I1ab36b63d83fb35803eb13a330d698cfa49f17e9
Reviewed-on: https://review.whamcloud.com/38862


Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Amir Shehata <ashehata@whamcloud.com>
Reviewed-by: Serguei Smirnov <ssmirnov@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>

LNetMEUnlink is not particularly useful, and exposing it as an LNet
interface only provides the opportunity for it to be misused.

Every successful call to LNetMEAttach() is followed by a call to
LNetMDAttach().  If that call succeeds, the ME is owned by
the MD and the caller mustn't touch it again.
If the call fails, the caller is currently required to call
LNetMEUnlink(), which all callers do, and these are the only places
that LNetMEUnlink() are called.

As LNetMDAttach() knows when it will fail, it can unlink the ME itself
and save the caller the effort.
This allows LNetMEUnlink() to be removed which simplifies
the LNet interface.

LNetMEUnlink() is also used in in ptl_send_rpc() in a situation where
ptl_send_buf() fails.  In this case both the ME and the MD need to be
unlinked, as as they are interconnected, LNetMEUnlink() or
LNetMDUnlink() can equally do the job.  So change it to use
LNetMDUnlink().

LNetMEUnlink() is primarily a call the lnet_me_unlink(). It also
 - has some handling if ->me_md is not NULL, but that is never the
   case
 - takes the lnet_res_lock().  However LNetMDAttach() already
   takes that lock.
So none of this functionality is useful to LNetMDAttach().
On failure, it can call lnet_me_unlink() directly while ensuring
it still has the lock.

This patch:
 - moves the calls to lnet_md_validate() into lnet_md_build()
 - changes LNetMDAttach() to always take the lnet_res_lock(),
   and to call lnet_me_unlink() on failure.
 - removes all calls to LNetMEUnlink() and sometimes simplifies
   surrounding code.
 - changes lnet_md_link() to 'void' as it only ever returns
   '0', and thus simplify error handling in LNetMDAttach()
   and LNetMDBind()

Signed-off-by: Mr NeilBrown <neilb@suse.de>
Change-Id: Ied4e8fb544dbe1b32df7dc70439161dc74366a1d
Reviewed-on: https://review.whamcloud.com/38646


Reviewed-by: Yang Sheng <ys@whamcloud.com>
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>

Now that Lustre supports a separate file system for the server
targets we can update mount.lustre to try "lustre_tgt" for the
server backend. For backwards compatibility if using "lustre_tgt"
fails try the original "lustre" file system type.

Change-Id: Ica08ff2abc3ad06c78d6d435db7e2fa3897e037e
Signed-off-by: James Simmons <jsimmons@infradead.org>
Reviewed-on: https://review.whamcloud.com/38799


Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Yang Sheng <ys@whamcloud.com>