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

LU-13992 llite: ASSERTION( last_oap_count > 0 ) failed

Punch uses o_blocks to send end of a region. So it
can be mixed with real blocks count on error.

Update blocks count only on success.

Change-Id: I86241c4e5723079b20401805b853d356130f58d9
HPE-bug-id: LUS-7407
Test-Parameters: fstype=zfs
Signed-off-by: Andriy Skulysh <c17819@cray.com>
Reviewed-by: Andrew Perepechko <c17827@cray.com>
Reviewed-by: Alexander Boyko <c17825@cray.com>
Tested-by: Elena Gryaznova <c17455@cray.com>
Reviewed-on: https://review.whamcloud.com/40050
Tested-by: jenkins <devops@whamcloud.com>
Reviewed-by: Alexander Boyko <alexander.boyko@hpe.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: Arshad Hussain <arshad.super@gmail.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>

LU-13986 target: fix possible liveloop in distribute_txn thd

A recent patch to update_trans.c changed how distribute_txn_thread()
waited for more work to do.

It previously had an explicit "wait_event()" which listed all the
conditions to wait for.  It would then recheck each condition and
possibly perform an appropriate action.

It was changed to check each condition only once (per loop).  If the
condition was true, the action would be performed and a flag set.  If
no conditions were true (indicated by flag), it would wait, otherwise
it would loop and recheck all condition.

One of the "if (condition) { do work }" stanzas in the loop tested a
condition that was *not* a condition that should wake up the loop.
"batchid" was not tested at all in the wait_event().  The flag
mentioned above was, however, set when that condition tested true.
This can cause the loop to spin indefinitely.

So remove the "__set_current_state(TASK_RUNNING);" so that the value
of batchid cannot stop the loop from sleeping (calling 'schedule()').

Signed-off-by: Mr NeilBrown <neilb@suse.de>
Change-Id: I124ee3e8250dc63fa927f72dc4d29ed3e7b53005
Reviewed-on: https://review.whamcloud.com/40043
Tested-by: jenkins <devops@whamcloud.com>
Reviewed-by: Andreas Dilger <adilger@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>

LU-13701 utils: fix usage of getopt_long in portals

Due to char being unsigned by default, using this type to store
the int return value of getopt_long causes parsing issues on arm
platform, seen with lctl net_drop_add options. Fix it.

Test-Parameters: trivial testgroup=review-ldiskfs-arm
Signed-off-by: Serguei Smirnov <ssmirnov@whamcloud.com>
Change-Id: Ibef6ba347a90d9ed3cd4bd9346ed9f79f0120b87
Reviewed-on: https://review.whamcloud.com/40036
Tested-by: jenkins <devops@whamcloud.com>
Reviewed-by: James Simmons <jsimmons@infradead.org>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Chris Horn <chris.horn@hpe.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>

LU-13985 lustre: seq_file .next functions must update *pos

A seq_file .next function must update *pos on EOF to a value which
will cause a subsequent ->start to also return EOF.
If it doesn't the last record of the file can be returned
twice to a 'read()'. Also the seq_file code will generate
a warning.

This patch fixes various ->next functions to always update
*pos.

Signed-off-by: Mr NeilBrown <neilb@suse.de>
Change-Id: Ia0c432cd50550ecde6b308cbc554b316fa03adae
Reviewed-on: https://review.whamcloud.com/40035
Tested-by: jenkins <devops@whamcloud.com>
Reviewed-by: Andreas Dilger <adilger@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>

LU-13982 tests: fix infinite loop in sanity test_184c

If the dd in test_184c fails to create the file, for example due to
ENOSPC, the subsequent "while" loops indefinitely.

So add a loop-count to ensure it stops eventually.

Also change the test from "-f" to "-s" so we wait for the file to be
non-empty.

Test-Parameters: trivial
Signed-off-by: Mr NeilBrown <neilb@suse.de>
Change-Id: I5051adf101f08856b97fa994f687b976fda84df4
Reviewed-on: https://review.whamcloud.com/40007
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: James Nunez <jnunez@whamcloud.com>

LU-13981 tests: use $TRUNCATE consistently and correctly.

Two changes:
1/ tests should should always use $TRUNCATE rather than just
    "truncate" to ensure the tests/truncate programs is used
    rather than anything in /bin /usr/bin.  This might not
    be needed in practice, but it is good to be consistent.

2/ The arguments provided should match what tests/truncate
    expects - "filename size".  Using "-s size filename"
    as expected by /usr/bin/truncate is confusing and
    error prone.

Signed-off-by: Mr NeilBrown <neilb@suse.de>
Change-Id: Ic0adc0148ea0ff625695dbfc4cf339d0eca70241
Reviewed-on: https://review.whamcloud.com/40005
Tested-by: jenkins <devops@whamcloud.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: James Simmons <jsimmons@infradead.org>
Reviewed-by: Oleg Drokin <green@whamcloud.com>

LU-13810 tests: increase limit for 1g

With wide striping file has objects at all
OSTs. As ZFS acquires several Kb for each
inode, each OST reserves minimum qunit 1M
even without write. At clean system with
8 OSTs, it acquires 8MB after file creation.
Increase pool limit from 10M to 20M to
make sanity-quota_1g pass at ZFS with 8 OSTs.

HPE-bug-id: LUS-9349
Test-Parameters: env=ONLY=1g testlist=sanity-quota
Change-Id: I1286e57aadebdd665c51d965220961b1f758c6f5
Signed-off-by: Sergey Cheremencev <sergey.cheremencev@hpe.com>
Reviewed-on: https://review.whamcloud.com/39917
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: Petros Koutoupis <petros.koutoupis@hpe.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>

LU-13948 tests: load modules after reboot_node

LOAD_MODULES_REMOTE should be taken into account
for FAILURE_MODE=HARD.

Test-parameters: envdefinitions="LOAD_MODULES_REMOTE=true"

Signed-off-by: Elena Gryaznova <elena.gryaznova@hpe.com>
HPE-bug-id: LUS-9283
Reviewed-by: Andriy Skulysh <c17819@cray.com>
Reviewed-by: Vladimir Saveliev <c17830@cray.com>
Change-Id: I85f0a2812ac3be4ac9645d3b165b7371504969f0
Reviewed-on: https://review.whamcloud.com/39829
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>

LU-13056 utils: Default MGS device llog_print/llog_catlist

Add the default device "MGS" for lctl llog_catlist, llog_print,
llog_check, llog_remove, llog_cancel and llog_info.

Example:
The two lines below are equivalent.

$ lctl --device MGS llog_catlist
$ lctl llog_catlist

Tests 123xx of conf_sanity have been modified to mix between the old
and new syntax.

lctl-llog_* man page have been modified to match the new syntax.

Signed-off-by: Etienne AUJAMES <eaujames@ddn.com>
Change-Id: If988dfc5d8171b2ca47bb68dd06f9d2953548cb2
Reviewed-on: https://review.whamcloud.com/38940
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: Emoly Liu <emoly@whamcloud.com>
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>

LU-12232 test: call dt_sync after dd

In test_6 of replay-ost-single, calling wait_mds_ost_sync
after "dd" to sync the write transactions.

Test-Parameters: trivial alwaysuploadlogs \
mdtfilesystemtype=zfs ostfilesystemtype=zfs \
testlist=replay-ost-single,replay-ost-single,replay-ost-single

Signed-off-by: Hongchao Zhang <hongchao@whamcloud.com>
Change-Id: I26f6b1fe903c8b7b3f4e5fb698baffc4c3b85130
Reviewed-on: https://review.whamcloud.com/36772
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: James Nunez <jnunez@whamcloud.com>
Reviewed-by: Wei Liu <sarah@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>

LU-8465 tests: enable pfsck if possible

To test pfsck widely, try enable pfsck by default
for Lustre tests.

Signed-off-by: Wang Shilong <wshilong@ddn.com>
Change-Id: I271422cf9d1f9cd0cc25c228c1f2df003e4f73f9
Reviewed-on: https://review.whamcloud.com/40070
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Tested-by: jenkins <devops@whamcloud.com>
Reviewed-by: Jian Yu <yujian@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>

LU-13017 tests: disable statahead_agl for sanity test_56ra

The sanity test_56ra can fail because statahead_agl can cause extra
glimpse request.

If a stat() systemcall is made after an AGL glimpse request is sent,
but before the reply has been received, the code handling the stat
cannot see that glimpse request and so will send another.  This
elevates the number of requests counted.

There is a parameter (statahead_agl) which make it easy to disable the
AGL, but it isn't implemented properly.  Specifically, inodes can
still be added to the sai_agls list when agl is disabled.  They will
never be removed, which causes an assertion to fail.

To clean this up, remove the sai_agl_valid flag, and use a test on
sai_task being non-NULL instead.  Also check agl_should_run() while
locked against ->sai_task changing, and before adding anything
to lli_agl_list.

We don't need the 'added' variable.  It is perfectly OK to wake_up the
sai_agl_task *before* adding to the list as long is that is all done
under the lock.  The task will wait for the lock before checking the
list, so it won't see it being empty.

Signed-off-by: Mr NeilBrown <neilb@suse.de>
Change-Id: I12c4e447104a86b3f48eaf57b6cf7ce4b41cc6de
Reviewed-on: https://review.whamcloud.com/39667
Reviewed-by: Lai Siyao <lai.siyao@whamcloud.com>
Reviewed-by: Yingjin Qian <qian@ddn.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>

LU-13920 hsm: process hsm_actions only after mdd setup

There is no guarantee that MDD setup is finished
at the moment when coordinator is started by
config params processing. If MDD setup is not finished,
hsm actions llog is not inited(mdd_hsm_actions_llog_init).
Hence hsm_pending_restore will not be called, i.e.
RESTORE requests will be sent to agents without taken
layout locks. I believe it may cause different problems.
I faced at least a kernel panic when llog includes
RESTORE request in ARS_WAITING that hasn't bee sent to
agent before failover. And the 2nd one RESTORE request
to the same fid was resent after recovery. Finally
agent handles to RESTORE to the same FID in parallel
that resulted in a panic with following bt:

BUG: unable to handle kernel NULL pointer dereference at 0000000000000040
IP: [<ffffffffc0b03bec>] thandle_get_sub_by_dt+0x14c/0x420 [ptlrpc]
...
[<ffffffffc1202732>] lod_sub_get_thandle+0x2f2/0x400 [lod]
[<ffffffffc1205021>] lod_sub_declare_xattr_set+0x61/0x300 [lod]
[<ffffffffc11db0d5>] lod_obj_stripe_replace_parent_fid_cb+0x245/0x450 [lod]
[<ffffffffc11eae0e>] lod_obj_for_each_stripe+0x11e/0x2d0 [lod]
[<ffffffffc11ebfe2>] lod_replace_parent_fid+0x2a2/0x390 [lod]
[<ffffffffc11dae90>] ? lod_attr_get+0x110/0x110 [lod]
[<ffffffffc11f8faf>] lod_declare_xattr_set+0x24f/0xf70 [lod]
[<ffffffffc077b251>] ? lprocfs_counter_sub+0xc1/0x130 [obdclass]
[<ffffffffc1091ff4>] mdo_declare_xattr_set+0x74/0x2b0 [mdd]
[<ffffffffc077b129>] ? lprocfs_counter_add+0xf9/0x160 [obdclass]
[<ffffffffc0fa2f7b>] ? osd_trans_create+0xbb/0x620 [osd_ldiskfs]
[<ffffffffc1094903>] mdd_declare_xattr_set+0x33/0x70 [mdd]
[<ffffffffc1094b4e>] mdd_object_pfid_replace+0x7e/0x1e0 [mdd]
[<ffffffffc109c2c6>] mdd_swap_layouts+0xa76/0x1dc0 [mdd]
[<ffffffffc10a5e1a>] ? mdd_trans_stop+0x3a/0x174 [mdd]
[<ffffffffc114b489>] hsm_cdt_request_completed.isra.14+0xc89/0xf50 [mdt]
[<ffffffffc077b129>] ? lprocfs_counter_add+0xf9/0x160 [obdclass]
[<ffffffffc114d844>] mdt_hsm_update_request_state+0x544/0x7b0 [mdt]
[<ffffffffc0a82277>] ? lustre_msg_buf+0x17/0x60 [ptlrpc]
[<ffffffffc10fec92>] ? ucred_set_audit_enabled.isra.15+0x22/0x60 [mdt]
[<ffffffffc112b98f>] mdt_hsm_progress+0x1ef/0x3f0 [mdt]
[<ffffffffc0aefcfa>] tgt_request_handle+0x96a/0x1640 [ptlrpc]
[<ffffffffc06cca9e>] ? libcfs_nid2str_r+0xfe/0x130 [lnet]
[<ffffffffc0a91466>] ptlrpc_server_handle_request+0x256/0xb10 [ptlrpc]
[<ffffffffc0a95fbc>] ptlrpc_main+0xb3c/0x14d0 [ptlrpc]
[<ffffffffc0a95480>] ? ptlrpc_register_service+0xf90/0xf90 [ptlrpc]
[<ffffffffa04c1c31>] kthread+0xd1/0xe0
[<ffffffffa04c1b60>] ? insert_kthread_work+0x40/0x40
[<ffffffffa0b76c37>] ret_from_fork_nospec_begin+0x21/0x21
[<ffffffffa04c1b60>] ? insert_kthread_work+0x40/0x40
Code: 74 29 4c 3b a0 50 ff ff ff 75 e4 4d 85 ed 74 1b bf 01 00 00 00 e8 c5 b8 ff ff 85 c0 0f 85 98 00 00 00 49 8b 45 00 e9 04 ff ff ff <49> 8b 44 24 40 48 8b 40 08 48 85 c0 0f 84 b3 02 00 00 4c 89 e6
RIP [<ffffffffc0b03bec>] thandle_get_sub_by_dt+0x14c/0x420 [ptlrpc]

Note, I faced this panic while testing https://review.whamcloud.com/#/c/38867/,
however I believe the same issue may exist even without 38867.

Patch makes mdt_hsm_cdt_start to wait until MDT initialization becomes
finished. Without this fix you should see below error in dmesg
each time after MDS restart if HSM is enbaled.
mdt_hsm_cdt_start()) lustre-MDT0000: cannot take the layout locks needed for registered restore: -2

Test-Parameters: testlist=sanity-hsm
Change-Id: I4c4edaa72a562576ea71d89a4b60618d099ec4f5
Signed-off-by: Sergey Cheremencev <sergey.cheremencev@hpe.com>
Reviewed-on: https://review.whamcloud.com/40028
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Ben Evans <beevans@whamcloud.com>
Reviewed-by: Nathan Rutman <nrutman@gmail.com>
Reviewed-by: Nikitas Angelinas <nikitas.angelinas@hpe.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>

LU-13006 jobid: enhance tests to check per-session jobids.

Lustre allows per-session jobids by writing to "jobid_this_session".

The upstream-linux client does *not* support jobids using the process
environment.

Update the sanity tests to recognize this.

1/ Allow setting jobid_var=USER to fail - if it fails, don't
   test use of envionment variables
2/ Check if "jobid_this_session tests - if it does, check
   that it is used for generating jobids.
3/ Some tests use jobid_var to test general assignment
   to config values.  Change those to use jobid_name if
   it is available.

Also if jobid_var is set to 'session' and jobid_name contains %j -
use jobid_name as is done when jobid_var is a variable name.

Signed-off-by: Mr NeilBrown <neilb@suse.de>
Change-Id: Ia7d01b8ffb9c6c910d2ce8f0615c802485604bf9
Reviewed-on: https://review.whamcloud.com/39656
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: Oleg Drokin <green@whamcloud.com>

LU-13977 tests: fix float comparison in sanity test_255a

  [ $a -gt $b ]
compares integers.

in ladvise_willread_performance(), average_ladvise and lowest_speedup
are calculated to 2 decimals, so they probably aren't integers.  So
this (nearly) always fails, but as the failure is not reported when a
VM is detected, the failure goes unnoticed.

The bash [[ ]] command can be used instead.  Its '>' operator compares
floats.

Test-Parameters: trivial
Signed-off-by: Mr NeilBrown <neilb@suse.de>
Change-Id: Ifc674d5804b1269cebce82893a08eace9ffd9be4
Reviewed-on: https://review.whamcloud.com/40001
Tested-by: jenkins <devops@whamcloud.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: James Simmons <jsimmons@infradead.org>

LU-13919 kernel: kernel update RHEL7.8 [3.10.0-1127.19.1.el7]

Update RHEL7.8 kernel to 3.10.0-1127.19.1.el7.

The extent status tree shrinker patches are removed from the
RHEL 7.8 kernel patch series because the following bug fix has
been included since version 3.10.0-1127.18.2.el7:

ext4: change LRU to round-robin in extent status tree shrinker (BZ#1847343)

Test-Parameters: clientdistro=el7.8 serverdistro=el7.8

Change-Id: I8172e79de239681c540ab0644b63f8172790a027
Signed-off-by: Jian Yu <yujian@whamcloud.com>
Reviewed-on: https://review.whamcloud.com/39964
Tested-by: jenkins <devops@whamcloud.com>
Reviewed-by: James Simmons <jsimmons@infradead.org>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Yang Sheng <ys@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>

LU-13511 obdclass: don't initialize obj for zero FID

Object with zero FID is used in stripe allocation, and it's
meaningless to initialize such object via lu_object_find_at(),
return error early to avoid assertion in lu_object_put().

Signed-off-by: Lai Siyao <lai.siyao@whamcloud.com>
Change-Id: Ia1bda3d01ff7552e94f31a9c928868652937d559
Reviewed-on: https://review.whamcloud.com/39792
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Stephane Thiell <sthiell@stanford.edu>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: Alex Zhuravlev <bzzz@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>

LU-11548 llite: increase readahead default values

It is commonly recommended to increase the readahead tunables
for clients to increase performance, since the current defaults
are too small, having been set several years ago for slower
networks and servers.

Increase the readahead defaults to better match values that are
recommended today:
- read_ahead_max_mb increased from 64MB to 1GB by default,
or 1/32 RAM, whichever is less
- read_ahead_per_file_max_mb is increased from 64MB to 256MB,
or 1/4 of read_ahead_max_mb, whichever is less

Modify the constant names to better match the variable and /proc
filenames.

Fix sanity test_101g to allow readahead to generate extra read
RPCs, as long as they are the expected size or larger.

Signed-off-by: Andreas Dilger <adilger@whamcloud.com>
Change-Id: Iec864788fa1979c27adad42e613d1bf03f3ebbe5
Reviewed-on: https://review.whamcloud.com/33400
Tested-by: jenkins <devops@whamcloud.com>
Reviewed-by: Wang Shilong <wshilong@whamcloud.com>
Reviewed-by: Yingjin Qian <qian@ddn.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>

LU-13745 llite: switch generic_file_splice_read() to use of ->read_iter()

... and kill the ->splice_read() instances that can be switched to it

* Special Note * Using generic_file_splice_read() breaks Lustre
but in my testing on RHEL7 not provding a ->splice_read, which
results in default_file_splice_read() to be used seems to work.

Linux-commit: 82c156f853840645604acd7c2cebcb75ed1b6652

Change-Id: I28bd2fe43fec25e881c78b41c58c59ac5e74eb49
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Reviewed-on: https://review.whamcloud.com/39272
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Wang Shilong <wshilong@whamcloud.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>

New tag 2.13.56

Change-Id: Iea1ae97f08b59719917e57bc1b2a554a010f76c4

LU-13969 - Updates to lustre-release yaml.sh

Updated output of release() function to standarize node.yml file
os_distribution parameter. Changes as follows:

RHEL   - use redhat-release first and os-release as backup as the latter may
         not include the full version (major/minor)
CENTOS - use centos-release first and os-release as backup, same as RHEL
SUSE   - use os-release instead of suse-release as the latter is deprecated
UBUNTU - use os-release

Removed parsing system-release and *-release as neither
option correctly outputs desired info

Removed "lustre_" references in node.yml file attributes, the default in Maloo
is to look for non-lustre prefixes first

Change-Id: Ia011f944aae53f31fcd3a539e846ea5aba7ec7c4
Signed-off-by: Lee Ochoa <lochoa@whamcloud.com>
Reviewed-on: https://review.whamcloud.com/39952
Tested-by: jenkins <devops@whamcloud.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: James Nunez <jnunez@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>

LU-13941 test: Restrict create_count to valid range

create_count range is restricted to:
[OST_MIN_PRECREATE=32, OST_MAX_PRECREATE=20000]

Enforce this restriction rather that halt with an error.

Test-Parameters: trivial testlist=parallel-scale
HPE-bug-id: LUS-5960
Fixes: 21501dedf64e ("LU-9780 tests: Testing Round-Robin allocation")
Signed-off-by: Shaun Tancheff <shaun.tancheff@hpe.com>
Change-Id: I125f2121a57d16a9031ed21938ce2f0b99b9dea8
Reviewed-on: https://review.whamcloud.com/39798
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: James Nunez <jnunez@whamcloud.com>

LU-13935 ofd: object removal is not handled properly

We should not check for object existence in
ofd_version_get_check() when we haven't got the lock yet.
ofd_attr_set() will check it later.

In ofd_destroy(), we should unlock if the object
does not exist, otherwise oti_w_locks assertion will fail.

Change-Id: I00f67d15c3268bcf55aafa88c088f2dbf55a470c
Signed-off-by: Andrew Perepechko <andrew.perepechko@hpe.com>
HPE-bug-id: LUS-9063,LUS-9282
Reviewed-on: https://review.whamcloud.com/39765
Reviewed-by: Alexander Zarochentsev <alexander.zarochentsev@hpe.com>
Reviewed-by: Alexander Boyko <alexander.boyko@hpe.com>
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>

LU-1538 tests: standardize test script init – full group

Standardize the initial Lustre test script initialization for
clarity and consistency for test suites in the full test group.

The LUSTRE path is already normalized in init_test_env(), so this
doesn't need to be done in the caller. Use $(...) subshells instead
of `...` in the affected lines. Remove NAME, SRCDIR, PATH, MULTIOP,
SETUP, CLEANUP, CHECKSTAT, TMP, SAVE_PWD, variable initialization,
since it is already done in init_test_env() or not needed in the test
scripts.

Move all definitions of ALWAYS_EXCEPT and SLOW to after
init_test_env() and init_logging() and call build_test_filter()
immediately after the ALWAYS_EXCEPT and SLOW definitions.

Test-Parameters: trivial
Test-Parameters: envdefinitions=SLOW=yes,ENABLE_QUOTA=yes clientcount=2 osscount=1 ostcount=7 mdscount=1 mdtcount=1 austeroptions=-R iscsi=1 testlist=metadata-updates,performance-sanity,racer
Test-Parameters: envdefinitions=SLOW=yes,ENABLE_QUOTA=yes clientcount=2 osscount=1 ostcount=7 mdscount=1 mdtcount=1 austeroptions=-R testlist=obdfilter-survey,sanity-benchmark,sanity-lsnapshot

Signed-off-by: James Nunez <jnunez@whamcloud.com>
Change-Id: I6e21a4a5d9e9215d5b452c4fd30467d9c007b5a5
Reviewed-on: https://review.whamcloud.com/36359
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Emoly Liu <emoly@whamcloud.com>
Reviewed-by: Jian Yu <yujian@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>

LU-12233 lnet: deadlock on LNet shutdown

Release ln_api_mutex during LNet shutdown while waiting
for zombie LNI to allow other threads to read the LNet
state updated by the shutdown and fall through, avoiding
the deadlock

Signed-off-by: Serguei Smirnov <ssmirnov@whamcloud.com>
Change-Id: Iaba11624d5b79bd0acb4add39f6153c55770440a
Reviewed-on: https://review.whamcloud.com/39933
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Chris Horn <chris.horn@hpe.com>
Reviewed-by: Amir Shehata <ashehata@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>

LU-13967 quota: change warning to cdebug

Changing CWARN to CDEBUG as it doesn't
point to any known problem.

Change-Id: Ia05a316a6ce1cc61899dd6662a9895946eee46aa
Signed-off-by: Sergey Cheremencev <sergey.cheremencev@hpe.com>
Reviewed-on: https://review.whamcloud.com/39921
Tested-by: jenkins <devops@whamcloud.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: James Nunez <jnunez@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>

LU-11631 obdclass: nlink is not set in struct obdo

Traditionally the nlink field is not set in obdo_from_la() and vice
versa, because it's used for communication between MDT and OST, and
this field is not used, but when DNE is enabled, this function is
used to getattr between MDTs, which will always see zero nlink.

Signed-off-by: Lai Siyao <lai.siyao@whamcloud.com>
Change-Id: I2a8e6a7b3f98864ec11bca7f7c2070af90b64ade
Reviewed-on: https://review.whamcloud.com/39896
Tested-by: jenkins <devops@whamcloud.com>
Reviewed-by: Mike Pershin <mpershin@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>

LU-13740 ldiskfs: add Ubuntu 20.04 LTS support

The latest Ubuntu LTS is based on a 5.4 Linux kernel. We can use
the previous work of generic 5.4 ldiskfs support with a small
change to the ext-pdirop.patch to make ldiskfs available to
this latest Ubuntu LTS.

Change-Id: I6d1106c3da984b508ccc77364e13adc692bddb60
Signed-off-by: James Simmons <jsimmons@infradead.org>
Reviewed-on: https://review.whamcloud.com/39779
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Li Dongyang <dongyangli@ddn.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>

LU-13840 utils: --pool instead of -o with lfs setquota

The use of "lfs setquota -o" as the short option for
specifying the quota pool is confusing, because "lfs quota -o"
is the short option for getting quota on an OST. Use only
long option --pool for lfs setquota.

Patch also changes -o with --pool for setquota in
sanity-quota.sh

Change-Id: I18a695dcbbcd668611b1301c48f4e10d354b2686
Signed-off-by: Sergey Cheremencev <sergey.cheremencev@hpe.com>
Reviewed-on: https://review.whamcloud.com/39715
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Tested-by: jenkins <devops@whamcloud.com>
Reviewed-by: Jian Yu <yujian@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>

LU-12397 osp: always set opd_new_connection

Flag opd_got_disconnected could be set back to 0
due to a race osp_precreate_thread vs osp_import_event.
Next ACTIVE event doesn't set opd_new_connection as
opd_got_disconnected also 0(i.e. import hasn't disconnected).
Such race is causing osp_precreate_thread to infinitly sleep
in wait despite сonnection state is FULL.

The patch always sets opd_new_connection flag on ACTIVE event
regardless value of opd_got_disconnected.

Patch is adding conf-sanity_101b test to race DISCON and ACTIVE
events. Without a fix the test causes to hung osp_precreate_thread
and as a result osp_precreate_reserve threads.

Change-Id: Iff41a2743f108679d5f70aca8e1c2108e979ac09
Cray-bug-id: LUS-7178
Signed-off-by: Sergey Cheremencev <c17829@cray.com>
Reviewed-on: https://es-gerrit.dev.cray.com/154883
Reviewed-by: Andriy Skulysh <c17819@cray.com>
Reviewed-by: Alexander Boyko <c17825@cray.com>
Tested-by: Elena Gryaznova <c17455@cray.com>
Reviewed-on: https://review.whamcloud.com/35078
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Tested-by: jenkins <devops@whamcloud.com>
Reviewed-by: Alexander Boyko <alexander.boyko@hpe.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>

LU-11518 osc: Do ELC on locks with no OSC object

Currently, osc_ldlm_weigh_ast weighs locks with no OSC
object in their ast data as "1", meaning the lock is not
considered for ELC.

This doesn't make much sense, since if there is no OSC
object, it's unlikely there's any data under the lock, so
it's actually a good candidate for ELC.

Signed-off-by: Patrick Farrell <pfarrell@whamcloud.com>
Change-Id: Ie832afbf2479f3a348e44d2c21992696830000ae
Reviewed-on: https://review.whamcloud.com/34584
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: Gu Zheng <gzheng@ddn.com>
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>

LU-13922 osd-ldiskfs: no need to add OI cache in readdir

It's a waste of time to call osd_add_oi_cache() in osd_it_ea_rec(),
because each dirent read will override it.

Signed-off-by: Lai Siyao <lai.siyao@whamcloud.com>
Change-Id: Iec701bf66153fdf2ba7a3f3b89565381215abf33
Reviewed-on: https://review.whamcloud.com/39782
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Tested-by: jenkins <devops@whamcloud.com>
Reviewed-by: Stephane Thiell <sthiell@stanford.edu>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>

LU-6142 lustre: don't take spinlock to read a 'long'.

Reading a 'long' (or unsigned long) is always an atomic operation.
There is never a need to take a spinlock to just read a single 'long'.

There are several procfs/debugfs/sysfs handlers which needlessly take
a spinlock for this purpose.

This patch:
- removes the taking of the spinlock
- changes the printf to scnprintf() as appropriate
- directly returns the value returned by scnprintf rather than
storing it in a variable
- accesses the 'long' as an arg to the scnprintf(), rather than
introducing a variabe to hold it.

Test-Parameters: trivial
Signed-off-by: Mr NeilBrown <neilb@suse.de>
Change-Id: If4a6454b46844864e1177536a9c7b91e4c97de86
Reviewed-on: https://review.whamcloud.com/39743
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: Arshad Hussain <arshad.super@gmail.com>
Reviewed-by: James Simmons <jsimmons@infradead.org>

LU-11518 ldlm: pool recalc forceful call

Let pool recalc to be able to be called forcefully independently of
the last recalc time;

Call the pool recalc forcefully on the lock decref instead of LRU
cancel to take into account the fresh SLV obtained from the server.

Call LRU recalc from after_reply if a significant SLV change occurs.
Add a sysfs attribute to control what 'a significant SLV change' is.

Signed-off-by: Vitaly Fertman <c17818@cray.com>
Change-Id: Iffeb8d73effdfc494f412422f285921aa4eb9811
HPE-bug-id: LUS-8678
Reviewed-on: https://es-gerrit.dev.cray.com/157134
Reviewed-by: Andriy Skulysh <c17819@cray.com>
Tested-by: Jenkins Build User <nssreleng@cray.com>
Reviewed-by: Alexey Lyashkov <c17817@cray.com>
Reviewed-on: https://review.whamcloud.com/39564
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Gu Zheng <gzheng@ddn.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>

LU-11518 ldlm: pool fixes

At the time the client side recalc period was increased up to 10secs
the grant & cancel rates started showing the speed not in seconds but
in tens of seconds.

At the pool initialization time, the server side recalc job should not
be delayed on client's recalc period.

It may happen an NS time is significant and comparable (or even more)
than the recalc period of the next NS (all the following NS's) in the
list. If the time has been already spent on the next NS, it does not
mean we want to double the delay for the original NS and recalc after
next N secs.

Make lock volume factor more fine grained (default is 100 now vs the
original 1): it is likely to cancel locks on clients twice faster than
server requested is too fast.

Protect missed pl_server_lock_volume update by the pool lock.

Replace ktime_get_real_seconds with ktime_get_seconds for the recal
interval.

Signed-off-by: Vitaly Fertman <c17818@cray.com>
Change-Id: Icba73209682a1b1d0d20c087581fad4f73ee3389
HPE-bug-id: LUS-8678
Reviewed-by: Andriy Skulysh <c17819@cray.com>
Reviewed-by: Alexey Lyashkov <c17817@cray.com>
Tested-by: Alexander Lezhoev <c17454@cray.com>
Reviewed-on: https://review.whamcloud.com/39563
Tested-by: jenkins <devops@whamcloud.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: Gu Zheng <gzheng@ddn.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>

LU-11518 ldlm: control lru_size for extent lock

We register ELC for extent locks to be canceled at enqueue time,
but it can't make positive effect to locks that have dirty pages
under it. To keep the semantics of lru_size, the client should
check how many unused locks are cached after adding a lock into
lru list. If it has already exceeded the hard limit
(ns_max_unused), the client will initiate async lock cancellation
process in batch mode (ns->ns_cancel_batch).

To do it, re-use the new batching LRU cancel functionality.

Wherever unlimited LRU cancel is called (not ELC), try to cancel in
batched mode.

And a new field named new sysfs attribute named *lru_cancel_batch*
is introduced into ldlm namespace to control the batch count.

Signed-off-by: Jinshan Xiong <jinshan.xiong@intel.com>
Signed-off-by: Shuichi Ihara <sihara@ddn.com>
Signed-off-by: Gu Zheng <gzheng@ddn.com>
Signed-off-by: Vitaly Fertman <c17818@cray.com>
Change-Id: Ib18b829372da8599ba872b5ac5ab7421661f942d
Reviewed-on: https://es-gerrit.dev.cray.com/157068
Reviewed-by: Andriy Skulysh <c17819@cray.com>
Reviewed-by: Alexey Lyashkov <c17817@cray.com>
Tested-by: Alexander Lezhoev <c17454@cray.com>
Reviewed-on: https://review.whamcloud.com/39562
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-13960 tests: correct usage of _var variable

In the setmodopts() function in functions.sh, the '_var'
variable is set and used. There is one use of the variable
'var' which should be '_var'. Change the use of 'var' to
'_var'.

Test-Parameters: trivial
Test-Parameters: testlist=conf-sanity env=ONLY=53 clientdistro=ubuntu2004 fstype=ldiskfs
Test-Parameters: testlist=conf-sanity env=ONLY=53 clientdistro=ubuntu1804 fstype=ldiskfs
Test-Parameters: testlist=conf-sanity env=ONLY=53 clientdistro=el7.8 fstype=ldiskfs
Signed-off-by: James Nunez <jnunez@whamcloud.com>
Change-Id: If524be1f0b4b2170a514a558256a5308c9a5e586
Reviewed-on: https://review.whamcloud.com/39891
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: John L. Hammond <jhammond@whamcloud.com>
Reviewed-by: Vikentsi Lapa <vlapa@whamcloud.com>

LU-13019 tests: replace usleep in test scripts

The "usleep" command is specific to RHEL, and is not available on
Ubuntu clients. Instead, "sleep" can be used with a fractional
second value to get the same effect more portably.

The "usleep" values are increased to be more reasonable lengths, as
a 100us sleep doesn't make sense after do_facet or other remote call
that may take seconds to complete. Also, in VM testing there may
be unpredictable delays, so it is better to be conservative.

Test-Parameters: trivial testlist=sanity-hsm,replay-single,replay-dual
Signed-off-by: Andreas Dilger <adilger@whamcloud.com>
Change-Id: Ic88c37d33ac17f46b9c1fb912112d318d9c3d07f
Reviewed-on: https://review.whamcloud.com/39889
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: James Simmons <jsimmons@infradead.org>
Reviewed-by: Emoly Liu <emoly@whamcloud.com>
Reviewed-by: Sebastien Buisson <sbuisson@ddn.com>
Reviewed-by: James Nunez <jnunez@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>

LU-13944 tests: add ability to set clients parameters

Patch adds the ability to set clients parameters via
CLIENT_LCTL_SETPARAM_PARAM.
Example:
CLIENT_LCTL_SETPARAM_PARAM="lbug_on_eviction=1 debug=-1 "
sets lbug_on_eviction=1 and debug=-1 on all clients.

Test-Parameters: trivial
HPE-bug-id: LUS-9052
Signed-off-by: Elena Gryaznova <elena.gryaznova@hpe.com>
Reviewed-by: Vladimir Saveliev <c17830@cray.com>
Reviewed-by: Andriy Skulysh <c17819@cray.com>
Change-Id: Ifb06e7992c2f1db94f0a1503cfc31e00d4674804
Reviewed-on: https://review.whamcloud.com/39816
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>

LU-13940 llite: it_lock_bits should be bit-wise tested

If lock_mode is not set then ensure that it_lock_bits has
the MDS_INODELOCK_OPEN bit set

Fixes: 5db0e2de00 ("LU-7403 llite: flatten struct lookup_intent")
HPE-bug-id: LUS-9198
Signed-off-by: Shaun Tancheff <shaun.tancheff@hpe.com>
Change-Id: Icccaa59cd5b47a7d3f97569a452f34b33e905409
Reviewed-on: https://review.whamcloud.com/39797
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Petros Koutoupis <petros.koutoupis@hpe.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: Yingjin Qian <qian@ddn.com>

LU-13013 osd: do not count credits for mapped blocks

this should help to save credits if an application
overwrites using many tiny fragments.

Signed-off-by: Alex Zhuravlev <bzzz@whamcloud.com>
Change-Id: Ice6541672339f2ba9c4e3d91403a756be323baab
Reviewed-on: https://review.whamcloud.com/36866
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>

LU-13900 clio: don't call aio_complete() in lustre upon errors

See following codes in aio_ret():
"
static inline ssize_t aio_ret(struct kiocb *req, ssize_t ret)
{
        switch (ret) {
        case -EIOCBQUEUED:
                return ret;
        case -ERESTARTSYS:
        case -ERESTARTNOINTR:
        case -ERESTARTNOHAND:
        case -ERESTART_RESTARTBLOCK:
                /*
                 * There's no easy way to restart the
                 * syscall since other AIO's may be already running.
                 * Just fail this IO with EINTR.
                 */
                ret = -EINTR;
                /*FALLTHRU*/
        default:
                aio_complete(req, ret, 0);
                return 0;
        }
}
"
VFS will call aio_complete() if ret is not -EIOCBQUEUED,
this could happen when we don't pass user buffer as page
alignment or some other errors happen in Lustre.

So in Lustre, we need be careful to handle this case to avoid double
aio_complete() called.

Fixes: d1dde ("LU-4198 clio: AIO support for direct IO")
Signed-off-by: Wang Shilong <wshilong@ddn.com>
Change-Id: I2137844437e91a4e269f7584c3a233e5fc81b0df
Reviewed-on: https://review.whamcloud.com/39636
Reviewed-by: Bobi Jam <bobijam@hotmail.com>
Tested-by: jenkins <devops@whamcloud.com>
Reviewed-by: Yingjin Qian <qian@ddn.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>

LU-6142 lov: guard against class_exp2obd() returning NULL.

class_exp2obd() can return NULL. Sometimes this is tested for,
sometimes not.
Add appropriate tested so that a NULL return is never dereferenced.

Test-Parameters: trivial
Signed-off-by: Mr NeilBrown <neilb@suse.de>
Change-Id: Ib78514e8e709971ae6d08a11186f77d6c950b1a2
Reviewed-on: https://review.whamcloud.com/39379
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>

LU-6142 lov: discard unused lov_dump_lmm* functions

lov_dump_lmm() and lov_dump_lmm_v3() are never used, so remove them.

So mark lov_lsm_pack_v1v3() and lov_lsm_pack_foreign() as static.

Test-Parameters: trivial
Signed-off-by: Mr NeilBrown <neilb@suse.de>
Change-Id: I013c265819e7e139a9878e9e0aa18022519bd2fe
Reviewed-on: https://review.whamcloud.com/39378
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: Lai Siyao <lai.siyao@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>

LU-12275 sec: verify dir is empty when setting enc policy

When setting an encryption policy on a directory that we want to
be encrypted, make sure it is empty.
Otherwise return -ENOTEMPTY from server to client doing the ioctl.

Also add test to exercise the ioctl interface to get/set encryption
policies on directories, thanks to the fscrypt userspace tool.
And test the ability to lock an encrypted directory.

Fixes: 40d91eafe257 ("LU-12275 sec: atomicity of encryption context getting/setting")
Test-Parameters: testlist=sanity-sec envdefinitions=ONLY="36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 52 53 54" 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 52 53 54" clientdistro=el8.1 fstype=zfs mdscount=2 mdtcount=4
Signed-off-by: Sebastien Buisson <sbuisson@ddn.com>
Change-Id: Iabffd24800314b12908f505d7e66492be7292a16
Reviewed-on: https://review.whamcloud.com/39617
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Tested-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: Lai Siyao <lai.siyao@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>

LU-13645 ldlm: don't use a locks without l_ast_data

Partially initialized lock (without l_ast_data to be init)
caused a fail with blocking ast, as discard from page cache
skipped, and stale data will read later with fast read.
Slow read have chance to attach this lock to right IO, but
it don’t true always, so that should disabled, until lock
will have l_ast_data set always for DoM and Lock Ahead locks.

HPE-bugid: LUS-8750
Change-Id: I2c5180c8044a12d7bd8f5f1c871447ca8b47a8ff
Signed-off-by: Alexey Lyashkov <c17817@cray.com>
Reviewed-on: https://review.whamcloud.com/39318
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: Vitaly Fertman <vitaly.fertman@hpe.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>

LU-10360 mgc: Use IR for client->MDS/OST connections

When a target registers with the MGS, the MGS sends an IR log
to the client to speed up recovery.

The IR log contains the updated NID information on the target
which just registered.

This patch allows clients to update their imports with the latest
NIDs for the targets reported in the IR log. It also allows clients
to create new connections for targets which were not added via
the config log.

For example if a target reboots and comes up with a new NID, then
the client can continue using it.

This functionality is disabled by default and can be enabled by
setting a new file system specific module parameter, dynamic_nids.

lctl set_param mgc.*.dynamic_nids=1

This parameters will need to be set on the clients and the MGS

Signed-off-by: Amir Shehata <ashehata@whamcloud.com>
Change-Id: I9e85438c39a46d4950ac50309c1bbae2a96d7d5d
Reviewed-on: https://review.whamcloud.com/39613
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>

LU-13921 utils: new output format for ofd_access_log_reader

prefix every field with a tag to make it extensible:
o=l-OST0000 f=[0x1:0x1:0x0] t=1598363594 b=0 e=4096 s=4096 s=1 n=1 r=w

Signed-off-by: Alex Zhuravlev <bzzz@whamcloud.com>
Change-Id: Iab7d02852a1eb43d2259a4e914bafe87332749a6
Reviewed-on: https://review.whamcloud.com/39728
Reviewed-by: John L. Hammond <jhammond@whamcloud.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: Gu Zheng <gzheng@ddn.com>
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>

LU-13763 osc: don't allow negative grants

Add check in the osc_init_grant() to prevent possible
underflow of cl_avail_grant and report error if it happens

Signed-off-by: Mikhail Pershin <mpershin@whamcloud.com>
Change-Id: Idcd25ed427c23735e1cdc70359bace43b5b9d886
Reviewed-on: https://review.whamcloud.com/39827
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: Vladimir Saveliev <c17830@cray.com>
Reviewed-by: Li Dongyang <dongyangli@ddn.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>

LU-13943 tests: load modules for list_nids getting

If modules not loaded it does not make sence to call
host_nids_address() -> $LCTL list_nids

Test-Parameters: trivial testlist=lnet-selftest
HPE-bug-id: LUS-9265
Signed-off-by: Elena Gryaznova <elena.gryaznova@hpe.com>
Reviewed-by: Vladimir Saveliev <c17830@cray.com>
Reviewed-by: Alexander Zarochentsev <c17826@cray.com>
Change-Id: I9eb431cd293e8b4831d433ceda752b03fd78c581
Reviewed-on: https://review.whamcloud.com/39815
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Alexander Zarochentsev <alexander.zarochentsev@hpe.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>

LU-13902 config: add test for /usr/include/libiberty/

The Makefile currently assumes that -I/usr/include/libiberty
is only needed on UBUNTU. But UBUNTU is only evaludate if
modules are being built (it is based on the base kernel found).
So with --disable-modules, -I/usr/include/liberty is never
selected.

There is no need for this indirection. It is easy to check if
/usr/inclue/libiberty exists, and to use if it if does.
So do that.

Signed-off-by: Mr NeilBrown <neilb@suse.de>
Change-Id: I6fb482256df3c7ce537dd4d451a9da27b8d8d3ea
Reviewed-on: https://review.whamcloud.com/39749
Tested-by: jenkins <devops@whamcloud.com>
Reviewed-by: James Simmons <jsimmons@infradead.org>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>

LU-13909 tests: get lustre_inode_cache count from sysfs

If slab cache merging is enabled (which is the default for newer
kernels), then there may not be a dedicated 'lustre_inode_cache'
entry in /proc/slabinfo (it may be aliased with 'signal_cache' and
'ptlrpc_cache' and/or other slabs of the same size).  This may result
in num_inodes() returning nothing to the caller, causing sanity
test_76() to not actually check anything useful.

Instead of looking for the number of cached inodes in /proc/slabinfo,
get this count from /sys/kernel/slab/lustre_inode_cache/objects if
available.  While it may not hold the actual number of cached inodes
due to slab sharing, sanity test_76 only cares about the relative
increase in cached inodes before and after running the test.

Due to slab sharing, the margin for error in counting cached objects
in the "lustre_inode_cache" slab can be fairly large (it may also be
shared with 'ptlrpc_cache' and increase significantly when sending
many RPCs).  The main goal of test_76 is to ensure there are not
persistently more objects cached in this slab compared to before
the create/unlink loop was run.

Test-Parameters: trivial testlist=sanity env=ONLY=76,ONLY_REPEAT=50
Signed-off-by: Andreas Dilger <adilger@whamcloud.com>
Change-Id: I11066680b7add1ec1e0b4d15ff0ef7a3a666b484
Reviewed-on: https://review.whamcloud.com/39735
Tested-by: jenkins <devops@whamcloud.com>
Reviewed-by: Yingjin Qian <qian@ddn.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Neil Brown <neilb@suse.de>
Reviewed-by: Oleg Drokin <green@whamcloud.com>

LU-10810 osd: implement lseek method in OSD

Introduce lseek support in dt_object methods and at OSD level

This is server support for SEEK_HOLE and SEEK_DATA distributed
request, it returns a resulting offset or error code

ZFS support is added as well but it is not yet available due to
missing export of dmu_offset_next() function.

Patch adds also osd_quasi_file() common method to create fake
file structure to use in inode i_fop calls

Signed-off-by: Mikhail Pershin <mpershin@whamcloud.com>
Change-Id: I3802743e31c94d2de219b72a540c5df9c72f1897
Reviewed-on: https://review.whamcloud.com/39706
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Bobi Jam <bobijam@hotmail.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>

LU-13918 lfs: let lfs mirror extend to use prefer flag

so that lfs mirror extend -N --flags=prefer create new
mirror with prefer flag set and new IO go to the just
created mirror.

Signed-off-by: Alex Zhuravlev <bzzz@whamcloud.com>
Change-Id: I2eb771fce3c14491eb193b053f3e97cc689b29c3
Reviewed-on: https://review.whamcloud.com/39698
Tested-by: jenkins <devops@whamcloud.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: Gu Zheng <gzheng@ddn.com>
Reviewed-by: Yingjin Qian <qian@ddn.com>
Tested-by: Maloo <maloo@whamcloud.com>

LU-13904 tests: don't assume echo_client is a module.

The upstream lustre client can be compiled with all code
in the kernel - no modules.
So testing if the module is loaded gets incorrect results.
So add a test to see if the functionalty is already present,
and don't load a module if none is needed.

Signed-off-by: Mr NeilBrown <neilb@suse.de>
Change-Id: Id630107d57f473dfdc21687a4964e1ca4c83efe4
Reviewed-on: https://review.whamcloud.com/39655
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: Oleg Drokin <green@whamcloud.com>

LU-13783 osc: handle removal of NR_UNSTABLE_NFS

In Linux 5.8 the NR_UNSTABLE_NFS page counters are go. All pages that
have been writen but are not yet safe are now counted in NR_WRITEBACK.

So change osc_page to count in NR_WRITEBACK, but if NR_UNSTABLE_NFS
still exists in the kernel, use a #define to direct the updates to
that counter.

Signed-off-by: Mr NeilBrown <neilb@suse.de>
Change-Id: I49cbc267fafaee949f45b2e559511aedcf4d8fed
Reviewed-on: https://review.whamcloud.com/39260
Tested-by: jenkins <devops@whamcloud.com>
Reviewed-by: Shaun Tancheff <shaun.tancheff@hpe.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: James Simmons <jsimmons@infradead.org>
Reviewed-by: Oleg Drokin <green@whamcloud.com>

LU-1538 tests: standardize test script init – misc tests

Standardize the initial Lustre test script initialization for
clarity and consistency test suites are that are not normally
run during patch and branch testing.

The LUSTRE path is already normalized in init_test_env(), so this
doesn't need to be done in the caller. Use $(...) subshells instead
of `...` in the affected lines. Remove NAME, SRCDIR, PATH, MULTIOP,
SETUP, CLEANUP, CHECKSTAT, TMP, SAVE_PWD, variable initialization,
since it is already done in init_test_env() or not needed in the test
scripts.

Remove MACHINEFILE from dom-performance because it is now
included init_test_env().

Move all definitions of ALWAYS_EXCEPT and SLOW to after
init_test_env() and init_logging() and call build_test_filter()
immediately after the ALWAYS_EXCEPT and SLOW definitions.

Test-Parameters: trivial envdefinitions=SLOW=yes,ENABLE_QUOTA=yes \
clientcount=2 osscount=1 ostcount=7 mdscount=1 mdtcount=1 \
austeroptions=-R testlist=lfsck-performance

Test-Parameters: envdefinitions=SLOW=yes,ENABLE_QUOTA=yes \
clientcount=2 osscount=1 ostcount=7 mdscount=1 mdtcount=1 \
austeroptions=-R testlist=sgpdd-survey

Test-Parameters: envdefinitions=SLOW=yes,ENABLE_QUOTA=yes \
clientcount=2 osscount=1 ostcount=7 mdscount=1 mdtcount=1 \
austeroptions=-R clientselinux testlist=sanity-selinux

Signed-off-by: James Nunez <jnunez@whamcloud.com>
Change-Id: I21280eb10de4d70ec3d7b4d6c8b8e976fb0821cc
Reviewed-on: https://review.whamcloud.com/36372
Tested-by: jenkins <devops@whamcloud.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Emoly Liu <emoly@whamcloud.com>
Reviewed-by: Wei Liu <sarah@whamcloud.com>

LU-1538 tests: standardize test script init – parallel-scale

Standardize the initial Lustre test script initialization for
clarity and consistency for all parallel-scale tests.

The LUSTRE path is already normalized in init_test_env(), so this
doesn't need to be done in the caller. Use $(...) subshells instead
of `...` in the affected lines. Remove NAME, SRCDIR, PATH, MULTIOP,
SETUP, CLEANUP, CHECKSTAT, TMP, SAVE_PWD, variable initialization,
since it is already done in init_test_env() or not needed in the test
scripts.

Move all definitions of ALWAYS_EXCEPT and SLOW to after
init_test_env() and init_logging() and call build_test_filter()
immediately after the ALWAYS_EXCEPT and SLOW definitions.

Test-Parameters: trivial
Test-Parameters: envdefinitions=SLOW=yes,ENABLE_QUOTA=yes clientdistro=el7.8 clientcount=2 osscount=1 ostcount=7 mdscount=1 mdtcount=1 austeroptions=-R testlist=parallel-scale-nfsv4
Test-Parameters: envdefinitions=SLOW=yes,ENABLE_QUOTA=yes clientdistro=el7.8 clientcount=2 osscount=1 ostcount=7 mdscount=1 mdtcount=1 austeroptions=-R testlist=parallel-scale,parallel-scale-nfsv3
Signed-off-by: James Nunez <jnunez@whamcloud.com>
Change-Id: I5690f17033ca92e5578c579e81cbcc024ca7f770
Reviewed-on: https://review.whamcloud.com/36355
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Emoly Liu <emoly@whamcloud.com>
Reviewed-by: Wei Liu <sarah@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>

LU-8066 osc: restore cur_dirty_grant_bytes

A previous patch which moved some attrs from /proc to /sys
inadvertently removed 'cur_dirty_grant_bytes'.
This patch restores it in /sys.

Test-Parameters: trivial
Fixes: 32fb31f3bf3d ("LU-8066 osc: move suitable values from procfs to sysfs")
Signed-off-by: Mr NeilBrown <neilb@suse.de>
Change-Id: Ic6cff067d3634d1b1406dad4eb156ad1967042c9
Reviewed-on: https://review.whamcloud.com/39740
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: Shaun Tancheff <shaun.tancheff@hpe.com>

LU-6142 lustre: all rhashtable_params should be static.

pools_hash_params and nid_hash_params are only used in the file where
they are declared - which is best for rhashtable params.
However they are not declared 'static', and one is even EXPORTed.

Change them both to 'static' and remove the EXPORT.

Test-Parameters: trivial
Signed-off-by: Mr NeilBrown <neilb@suse.de>
Change-Id: Ib71f8a7c3bfa38fa54b3653d51786a3d8da6117f
Reviewed-on: https://review.whamcloud.com/39730
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>

LU-6142 lov: don't use inline for operations functions.

These functions have their address taken and stored in an 'operations'
structure. So they cannot possibly be compiled "inline".
So remove the "inline" declaration.

Test-Parameters: trivial
Signed-off-by: Mr NeilBrown <neilb@suse.de>
Change-Id: Ie6712048bd40a4518adaa510dd7da0202b394c1f
Reviewed-on: https://review.whamcloud.com/39376
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Lai Siyao <lai.siyao@whamcloud.com>
Reviewed-by: James Simmons <jsimmons@infradead.org>
Reviewed-by: Shaun Tancheff <shaun.tancheff@hpe.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>

LU-6142 libcfs: remove last use of container_of0

Remove the last use of container_of0 in osd_ldiskfs. Looking at
the code using container_of is safe to use since the handle
shouldn't be NULL or PTR_ERR.

Change-Id: Id018543997884825d1c3cee76de3a520fc3b7c18
Signed-off-by: James Simmons <jsimmons@infradead.org>
Reviewed-on: https://review.whamcloud.com/39321
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Neil Brown <neilb@suse.de>
Reviewed-by: Alex Zhuravlev <bzzz@whamcloud.com>
Reviewed-by: Shaun Tancheff <shaun.tancheff@hpe.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>

LU-13127 ptlrpc: avoid ISO C90 mixed declaration

Avoid forbidden mixed declaration in lustre_msg_calc_cksum()

Test-Parameters: trivial
Fixes: 726897c87c42 ("LU-13127 ptlrpc: don't require CONFIG_CRYPTO_CRC32")
Signed-off-by: Mikhail Pershin <mpershin@whamcloud.com>
Change-Id: I180eefd979331f3fa206aa9b1d54c8e7b17bcbee
Reviewed-on: https://review.whamcloud.com/39677
Tested-by: jenkins <devops@whamcloud.com>
Reviewed-by: James Simmons <jsimmons@infradead.org>
Reviewed-by: Neil Brown <neilb@suse.de>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>

LU-13359 quota: call rhashtable_lookup near params decl

rhashtable_lookup() is an inline function which depends - for
performancs - on the 'rhashtable_params' being visible and
consnt. So it should only be called in the same file that
declared the params.

A recent patch make pools_hash_params an external variable and calls
rhashtable_lookup from a separate file, which will break the
optimisation.

So add lov_pool_find() and use it to maintainer optimization.

Fixes: 6b9f849fd5f4 ("LU-13359 quota: make used for pool correct")
Signed-off-by: Mr NeilBrown <neilb@suse.de>
Change-Id: Ieeb3b080491f5b2c9c825885fe7a42f4a8599a2a
Reviewed-on: https://review.whamcloud.com/39676
Reviewed-by: James Simmons <jsimmons@infradead.org>
Reviewed-by: Sergey Cheremencev <sergey.cheremencev@hpe.com>
Reviewed-by: Petros Koutoupis <petros.koutoupis@hpe.com>
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>

LU-12275 sec: encryption with different client PAGE_SIZE

In order to properly handle encryption/decryption on clients that have
a PAGE_SIZE != LUSTRE_ENCRYPTION_UNIT_SIZE (typically aarch64/ppc64),
a few adjustements are necessary:
- when encrypting, do not proceed with PAGE_SIZE as encryption length.
  Instead, round up to a multiple of LUSTRE_ENCRYPTION_UNIT_SIZE.
  On aarch64/ppc64, it avoids encrypting way beyond
  LUSTRE_ENCRYPTION_UNIT_SIZE when the page is not full.
- when decrypting, do not proceed with PAGE_SIZE as decryption length.
  Instead, do LUSTRE_ENCRYPTION_UNIT_SIZE length at a time. It enables
  proper detection of 'all 0s' sent by servers for content beyond file
  size.

Regarding tests, add sanity-sec test_53 to exercise encryption from
clients with different PAGE_SIZE.
The trick to achieve this with AT is to expect the client to have 64KB
PAGE_SIZE, and the servers to have 4KB PAGE_SIZE, and then mount a
client from the MDS node.
This also means code running on server side needs to have client
encryption support enabled, so CentOS/RHEL 8 at least.

Test-Parameters: trivial
Test-Parameters: clientarch=aarch64 clientcount=1 clientdistro=el8.1 serverdistro=el8.1 testlist=sanity-sec env=ONLY="36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 52 53" fstype=ldiskfs mdscount=2 mdtcount=4
Test-Parameters: clientarch=aarch64 clientcount=1 clientdistro=el8.1 serverdistro=el8.1 testlist=sanity-sec env=ONLY="36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 52 53" fstype=zfs mdscount=2 mdtcount=4
Test-Parameters: clientarch=x86_64 clientcount=1 clientdistro=el8.1 serverdistro=el8.1 testlist=sanity-sec env=ONLY="36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 52 53" fstype=ldiskfs mdscount=2 mdtcount=4
Test-Parameters: clientarch=x86_64 clientcount=1 clientdistro=el8.1 serverdistro=el8.1 testlist=sanity-sec env=ONLY="36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 52 53" fstype=zfs mdscount=2 mdtcount=4
Test-Parameters: clientdistro=el8.1 testlist=sanity-sec env=ONLY="36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52" mdscount=2 mdtcount=4
Signed-off-by: Sebastien Buisson <sbuisson@ddn.com>
Change-Id: Iee4b4d9e70c2e8c8e12061c39400cf6a8c03bac3
Reviewed-on: https://review.whamcloud.com/39315
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: Wang Shilong <wshilong@whamcloud.com>

LU-12295 mdd: don't LBUG() if dir nlink is wrong

Sometimes dir nlink may not be correctly decreased: subdir is remote,
when it's unlinked, its dirent is removed, but parent nlink decrease
failed.

Don't assert this in osd_destroy(), but print an error message and
continue since we've checked directory is empty.

Add OBD_FAIL_OSD_REF_DEL to simulate the error above.

Add sanity 48f.

Signed-off-by: Lai Siyao <lai.siyao@whamcloud.com>
Change-Id: I483aaf7a62b7761868b5e2af8dbfa92929fda78c
Reviewed-on: https://review.whamcloud.com/39734
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Alex Zhuravlev <bzzz@whamcloud.com>

LU-13605 lnet: Do not overwrite destination when routing

MR path selection in a routed environment is supposed to allow the
originator of a message to set the final destination NID. On a
multi-hop route, intermediate routers execute the same code path as
the message originator (i.e. the remote send cases). This causes
them to overwrite the destination NID when forwarding the message.

Check the msg_routing flag to determine whether we should set the
final destination NID (i.e. LNet peer NI).

A somewhat related issue is that because intermediate routers are not
selecting a destination lpni, they need to pick the next-hop lpni
based on the destination NID's remote net.

Test-Parameters: trivial
Fixes: 9dfdc2238be ("LU-13035 lnet: fix remote peer ni selection")
HPE-bug-id: LUS-8919
Signed-off-by: Chris Horn <chris.horn@hpe.com>
Change-Id: Id2fbbc5d8da347e971bbb8ad2779e80f75e29dd7
Reviewed-on: https://review.whamcloud.com/38731
Tested-by: jenkins <devops@whamcloud.com>
Reviewed-by: Serguei Smirnov <ssmirnov@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Shaun Tancheff <shaun.tancheff@hpe.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>

LU-13502 lnet: Add response tracking param to lnetctl

Add support for the lnet_response_tracking parameter to lnetctl.

Test-Parameters: trivial testlist=sanity-lnet
HPE-bug-id: LUS-8827
Signed-off-by: Chris Horn <hornc@cray.com>
Change-Id: I952a415e27582a7a6d920bfeb16618766c0235da
Reviewed-on: https://review.whamcloud.com/38514
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Neil Brown <neilb@suse.de>
Reviewed-by: Petros Koutoupis <petros.koutoupis@hpe.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>

LU-13187 osd-ldiskfs: don't enforce max dir size limit on IAM objects

Add ext4-no-max-dir-size-limit-for-iam-objects.patch to introduce new
inode state EXT4_STATE_IAM and use it to mark IAM objects.

Test-Parameter: testlist=sanity env=ONLY=129,ONLY_REPEAT=100
Change-Id: I3bcc5435ea07edb9fa265dcd8e3261d849495f00
Signed-off-by: Li Dongyang <dongyangli@ddn.com>
Reviewed-on: https://review.whamcloud.com/39823
Tested-by: jenkins <devops@whamcloud.com>
Reviewed-by: Neil Brown <neilb@suse.de>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: James Simmons <jsimmons@infradead.org>
Tested-by: Maloo <maloo@whamcloud.com>

LU-13827 utils: ofd_access_batch to print top hot files

with -F option one can specify fraction (in %%) of hot
files to be printed.

Signed-off-by: Alex Zhuravlev <bzzz@whamcloud.com>
Change-Id: I131c47bb2252da88c9afce757c5264b8d2011d1a
Reviewed-on: https://review.whamcloud.com/39529
Reviewed-by: John L. Hammond <jhammond@whamcloud.com>
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: Gu Zheng <gzheng@ddn.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>

LU-13376 utils: add batching to ofd_access_log_reader

Add interval based batching to ofd_access_log_reader. Add option to
control the batch interval, offset within the interval, and batch
output file.

Signed-off-by: John L. Hammond <jhammond@whamcloud.com>
Change-Id: I057e9616f4ec198dbf0c7c82a93a1e45907e7a42
Reviewed-on: https://review.whamcloud.com/38035
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Gu Zheng <gzheng@ddn.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>

LU-13945 utils: add includes for copy_file_range()

In lstddef.h include the needed headers to support the compat
definition of copy_file_range().

Test-Parameters: trivial
Signed-off-by: John L. Hammond <jhammond@whamcloud.com>
Change-Id: I83544cbbeb6407f4c3bb9fa3bd2a1297f2b2a2dc
Reviewed-on: https://review.whamcloud.com/39821
Reviewed-by: James Simmons <jsimmons@infradead.org>
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Alex Zhuravlev <bzzz@whamcloud.com>
Reviewed-by: Gu Zheng <gzheng@ddn.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>

LU-12661 tests: skip sanity 817 for kernel 4.12+

Skip the NFS exec mode bug for kernels 4.12 and later, since
this is also being hit on SLES kernel 4.12.14+ and not just 4.14.

Test-Parameters: trivial
Fixes: 4fed33473ca2 ("LU-12661 tests: skip sanity 817 if kernel >= 4.14")
Signed-off-by: Andreas Dilger <adilger@whamcloud.com>
Change-Id: Ibc4ffda72bd7827e250c4583c760505b8f3ebbe5
Reviewed-on: https://review.whamcloud.com/39838
Reviewed-by: Li Dongyang <dongyangli@ddn.com>
Reviewed-by: James Nunez <jnunez@whamcloud.com>
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>

LU-13599 mdt: add test for rs_lock limit exceeding

The check conditions in mdt_link_parents_lock() considers there
can be total 6 local locks but in fact when object has hard links
then it is regular file and it may have maximum 5 local locks.

Patch updates that check conditions for rs_locks as small
improvement and ports test for rs_locks limit from 2.12.

Signed-off-by: Mikhail Pershin <mpershin@whamcloud.com>
Change-Id: I98cca84825ce5789094fbceb5d1f7975410d134b
Reviewed-on: https://review.whamcloud.com/39194
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Lai Siyao <lai.siyao@whamcloud.com>
Reviewed-by: Stephane Thiell <sthiell@stanford.edu>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>

LU-13915 ldiskfs: Avoid atomic operation while bitmap prefetch

It is expensive since test_and_set_bit is a atomic operation. So
use test_bit while bitmap prefetch to avoid call it frequently.

Signed-off-by: Yang Sheng <ys@whamcloud.com>
Change-Id: I2ff2c39f1dd3b351462ed66cbd3ebb36e6af4bea
Reviewed-on: https://review.whamcloud.com/39697
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: Artem Blagodarenko <artem.blagodarenko@hpe.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>

LU-13909 llite: prune invalid dentries

When file LOOKUP lock is canceled on client, mark its dentries
invalid, and also prune them to avoid OOM, to achieve this,
ll_invalidate_aliases() is renamed to ll_prune_aliases(), the latter
calls d_prune_aliases() to prune unused invalid dentries.

The same for negative dentries when parent UPDATE lock is canceled,
rename ll_invalidate_negative_children() to
ll_prune_negative_children().

Since now unused invalid dentries will always be pruned, it's not
necessary to call __d_drop() in d_lustre_invalidate().

It's redundant to take i_lock before d_lustre_invalidate() in
ll_inode_revalidate() because d_lustre_invalidate() takes d_lock,
remove it.

Signed-off-by: Lai Siyao <lai.siyao@whamcloud.com>
Change-Id: Ib0ae57537e31ba9269e042b94bc5fbe7cb263a50
Reviewed-on: https://review.whamcloud.com/39685
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>
Reviewed-by: Yingjin Qian <qian@ddn.com>

LU-4671 tests: give multiop a chance to exit.

If 'multiop' is still running after a test complete, test-framework.sh
reports a failure.
test_43A signals multiop asking it to exit as the last thing it does.
If there is any delay in multip being schedule, test-framework will
see it and report an error - this is a false negative.

So use 'wait' to wait for multiop to respond to the signal.

Test-Parameters: trivial testlist=sanity envdefinitions=ONLY="43A"
Signed-off-by: Mr NeilBrown <neilb@suse.de>
Change-Id: Ic94d3c89bd98c9f6ef2d5bee22aac1e39116bf11
Reviewed-on: https://review.whamcloud.com/39665
Tested-by: jenkins <devops@whamcloud.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: Jian Yu <yujian@whamcloud.com>
Reviewed-by: James Nunez <jnunez@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>

LU-13896 lnet: Fix reference leak in lnet_select_pathway

We call lnet_nid2peerni_locked() to lookup the peer NI for the message
originator. lnet_nid2peerni_locked() takes a reference on the peer NI
that is never dropped.

Test-Parameters: trivial
Fixes: b0e8ab1a5f ("LU-13606 lnet: Allow router to forward to healthier NID")
HPE-bug-id: LUS-9185
Signed-off-by: Chris Horn <chris.horn@hpe.com>
Change-Id: Ie0e8f215d7becfbf33f905a1806da8513798ee8d
Reviewed-on: https://review.whamcloud.com/39603
Tested-by: jenkins <devops@whamcloud.com>
Reviewed-by: Neil Brown <neilb@suse.de>
Reviewed-by: Serguei Smirnov <ssmirnov@whamcloud.com>
Reviewed-by: Amir Shehata <ashehata@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>

LU-13810 tests: OST Pool Quotas with wide striping

All previous tests in sanity-quota except DOM,SEL
and PFL specific tests always use stripe count 1.
Check that block hard limit set to OST Pool Quotas
work properly with wide stripe files.

HPE-bug-id: LUS-8600
Test-Parameters: env=ONLY=1g testlist=sanity-quota
Change-Id: Ia6ebb21adb0fff18c6a9e0a36b1e50cb0f1d212a
Signed-off-by: Sergey Cheremencev <sergey.cheremencev@hpe.com>
Reviewed-on: https://review.whamcloud.com/39469
Tested-by: jenkins <devops@whamcloud.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Petros Koutoupis <petros.koutoupis@hpe.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>

LU-13735 lnet: Loosen restrictions on LNet Health params

The functions that set various LNet Health related parameters require
that the parameters be set in a specific order depending on whether
health is enabled or disabled. This is not user-friendly.
- Don't overwrite lnet_transaction_timeout when health is being
   enabled or disabled.
- Don't overwrite lnet_retry_count when health is being enabled
   (still set it to zero when health is disabled).
- Allow lnet_retry_count to be set to 0 when health is disabled
- Correct off-by-one error in transaction_to_set() to ensure
   lnet_transaction_timeout is greater than lnet_retry_count

HPE-bug-id: LUS-8995
Signed-off-by: Chris Horn <chris.horn@hpe.com>
Change-Id: Ic8ca7862543fc667fdf85844e05146c78bf48cd1
Reviewed-on: https://review.whamcloud.com/39228
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>

LU-13151 mdt: add parent FID to Changelog recordss

Use the link EA to add the parent FID to ChangeLog records, including
MTIME, TRUNC, and SATTR.

Some tools that maintain copies of filesystem metadata in an external
database monitor changelogs for changes to the filesystem, in order to
determine files that need to be rescanned. This can result in a large
number of small updates to the external database that can reduce the
tool's ingest performance. It might be beneficial to instead track and
scan complete directories that contain modified files and update the
external database using bulk operations. Adding the parent FID to
MTIME changelogs allows to more efficiently determine the parent
directories for some types of file data modifications, by issuing
OBD_IOC_FID2PATH once for each parent FID, instead of once for each
file FID.

Signed-off-by: Nikitas Angelinas <nikitas.angelinas@hpe.com>
Cray-bug-id: LUS-7986
Change-Id: I0c88271e706ef8202910e9461e5ae9f6dcbe0bdd
Reviewed-on: https://review.whamcloud.com/37264
Tested-by: jenkins <devops@whamcloud.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Nathan Rutman <nrutman@gmail.com>
Reviewed-by: Sebastien Buisson <sbuisson@ddn.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>

LU-11986 lnet: don't read debugfs lnet stats when shutting down

A race exist on shutdown with an external application reading
the debugfs file containing lnet stats which causes an kernel
crash.

[  257.192117] BUG: unable to handle kernel paging request at fffffffffffffff0
[  257.194859] IP: [<ffffffffc0bb95c6>] cfs_percpt_number+0x6/0x10 [libcfs]
[  257.196863] PGD 7c14067 PUD 7c16067 PMD 0
[  257.198665] Oops: 0000 [#1] SMP
[  257.200431] Modules linked in: ksocklnd(OE) lnet(OE) libcfs(OE) dm_service_time iscsi_tcp libiscsi_tcp libiscsi scsi_transport_iscsi sunrpc zfs(POE) zunicode(POE) zavl(POE) icp(POE) zcommon(POE) znvpair(POE) spl(OE) ppdev iosf_mbi crc32_pclmul crc32c_intel ghash_clmulni_intel aesni_intel lrw gf128mul glue_helper ablk_helper cryptd pcspkr sg e1000 video parport_pc parport i2c_piix4 dm_multipath dm_mod ip_tables xfs libcrc32c sd_mod crc_t10dif crct10dif_generic ata_generic pata_acpi crct10dif_pclmul crct10dif_common ata_piix serio_raw libata [last unloaded: obdclass]
[  257.222895] CPU: 0 PID: 7331 Comm: lctl Tainted: P           OE  ------------   3.10.0-957.el7_lustre.x86_64 #1
[  257.229312] Hardware name: innotek GmbH VirtualBox/VirtualBox, BIOS VirtualBox 12/01/2006
[  257.233659] task: ffff9c9fbaf15140 ti: ffff9c9fbabcc000 task.ti: ffff9c9fbabcc000
[  257.238388] RIP: 0010:[<ffffffffc0bb95c6>]  [<ffffffffc0bb95c6>] cfs_percpt_number+0x6/0x10 [libcfs]
[  257.243851] RSP: 0018:ffff9c9fbabcfdb0  EFLAGS: 00010296
[  257.246400] RAX: 0000000000000000 RBX: ffff9c9fba2a5200 RCX: 0000000000000000
[  257.250304] RDX: 0000000000000001 RSI: 00000000ffffffff RDI: 0000000000000000
[  257.253677] RBP: ffff9c9fbabcfdd0 R08: 000000000001f120 R09: ffff9c9fbe001700
[  257.257073] R10: ffffffffc0c376db R11: 0000000000000246 R12: 0000000000000000
[  257.260339] R13: 0000000000000000 R14: 0000000000001000 R15: ffff9c9fba2a5200
[  257.263204] FS:  00007fbdc89c6740(0000) GS:ffff9c9fbfc00000(0000) knlGS:0000000000000000
[  257.266409] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[  257.269105] CR2: fffffffffffffff0 CR3: 0000000022e36000 CR4: 00000000000606f0
[  257.272529] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
[  257.275209] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
[  257.277936] Call Trace:
[  257.279245]  [<ffffffffc0c0a88b>] ? lnet_counters_get_common+0xeb/0x150 [lnet]
[  257.283071]  [<ffffffffc0c0a95c>] lnet_counters_get+0x6c/0x150 [lnet]
[  257.286224]  [<ffffffffc0c3771b>] __proc_lnet_stats+0xfb/0x810 [lnet]
[  257.288975]  [<ffffffffc0ba6602>] lprocfs_call_handler+0x22/0x50 [libcfs]
[  257.292387]  [<ffffffffc0c36bf5>] proc_lnet_stats+0x25/0x30 [lnet]
[  257.295184]  [<ffffffffc0ba665d>] lnet_debugfs_read+0x2d/0x40 [libcfs]

The solution is to only allow reading of the lnet stats when the
lnet state is LNET_STATE_RUNNING.

Test-Parameters: trivial testlist=sanity-lnet
Change-Id: I8720a51ec358e4f6ae121acb34cc23020054ab84
Signed-off-by: James Simmons <jsimmons@infradead.org>
Reviewed-on: https://review.whamcloud.com/39404
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: Nathaniel Clark <nclark@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>

LU-12275 sec: ldiskfs not aware of client-side encryption

In osd-ldiskfs, always remove S_ENCRYPTED from inode flags,
because ldiskfs must not be aware of client-side encryption status.
This info is just stored into LMA so that it can be forwared to client
side.

Signed-off-by: Sebastien Buisson <sbuisson@ddn.com>
Change-Id: Ief08c059b04b8c7349d725b50b2094183eabc4d3
Reviewed-on: https://review.whamcloud.com/39558
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Wang Shilong <wshilong@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>

LU-12275 sec: restrict fallocate on encrypted files

For now, ll_fallocate only supports standard preallocation.
Anyway, encrypted inodes can't handle collapse range or zero range or
insert range since we would need to re-encrypt blocks with a different
IV or XTS tweak (which are based on the logical block number).
So make sure we return -EOPNOTSUPP in this case, like what ext4 does.

Test-Parameters: trivial
Signed-off-by: Sebastien Buisson <sbuisson@ddn.com>
Change-Id: Ia34cd04df9f297ac54109ed385b037fe282954d7
Reviewed-on: https://review.whamcloud.com/39220
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: Arshad Hussain <arshad.super@gmail.com>
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>

LU-12275 sec: O_DIRECT for encrypted file

Add O_DIRECT support for encrypted files.
By default, fscrypt does not support O_DIRECT because it needs
pagecache pages to proceed.
With Lustre, we can make use of pages being used for sending RPCs.
They can be twisted so that they have a proper mapping and index,
suitable for encryption/decryption.

One of the benefits of O_DIRECT support for encrypted files is that
we get support for mirroring at the same time.

Test-Parameters: testlist=sanity-sec envdefinitions=ONLY="36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 52" 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 52" clientdistro=el8.1 fstype=zfs mdscount=2 mdtcount=4
Signed-off-by: Sebastien Buisson <sbuisson@ddn.com>
Change-Id: I12f61c44b55f3a454f38016200f81eb735ab8f18
Reviewed-on: https://review.whamcloud.com/38967
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: Wang Shilong <wshilong@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>

LU-13910 mdt: 0 for success in mdt_path_current()

0 should be returned if no error is found in mdt_path_current(),
otherwise, the non-zero value will be treated as an error in
mdc_ioc_fid2path() and null will be returned by "lfs fid2path".

sanity.sh test_226c is added to verify this patch.

Signed-off-by: Emoly Liu <emoly@whamcloud.com>
Change-Id: I1c10b023da9bbbb908dfb691fcea6e84ced67a8d
Reviewed-on: https://review.whamcloud.com/39688
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>

LU-13903 tests: skip test_410 if modules weren't built

test_410 requires a special module, which is only built if
all modules are being built. So if only the user-space code
was built, this test will fail.

So make the test contitional on the module existing.

Test-Parameters: trivial
Signed-off-by: Mr NeilBrown <neilb@suse.de>
Change-Id: I235894c72b51f627c01ee08e850a59933b49e033
Reviewed-on: https://review.whamcloud.com/39652
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>

LU-13608 out: don't return einprogress error

When out_handle proccess an update request it could happened
that file doesn't exist, osd_fid_lookup triggers scrub and
returns EINPROGRESS. Remote MDT would process EINPROGRESS at
ptlrpc layer and resend a request in loop, and MDT recovery
would be blocked.

The fix adds fid to OI for ENOENT, like it was before the LU-7782.
So the second attempt with the same fid will return ENOENT.

Signed-off-by: Alexander Boyko <alexander.boyko@hpe.com>
HPE-bug-id: LUS-9062
Change-Id: Ib9a1753234ccc773e9b9529195ebfa6e5a8c101c
Reviewed-on: https://review.whamcloud.com/39538
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Andriy Skulysh <c17819@cray.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>

LU-13314 utils: fix lfs find time calculation margin

Allow a larger margin when checking files that are years old.

Re-enable sanity test_56ob.

Signed-off-by: Andreas Dilger <adilger@whamcloud.com>
Change-Id: Id74e4b737ebb3a2b721d3e3b6d79dffe703ebbe5
Reviewed-on: https://review.whamcloud.com/39433
Tested-by: jenkins <devops@whamcloud.com>
Reviewed-by: Jian Yu <yujian@whamcloud.com>
Reviewed-by: Yingjin Qian <qian@ddn.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>

LU-13773 tests: subscript failure propagation

When a test script calls another test script and there is a
failure in the called test script, the failure is not
propagated up to the main/calling test suite and Maloo is
not registering the failed test. An example of this is
sanity-dom calls sanity and sanityn, but, if a sanity test
fails, Maloo does not recognize the sanity failure.

Signed-off-by: James Nunez <jnunez@whamcloud.com>
Change-Id: I1914ecbece469cc1faffdffa4c980241ba3020b2
Reviewed-on: https://review.whamcloud.com/39409
Reviewed-by: Wei Liu <sarah@whamcloud.com>
Reviewed-by: Jian Yu <yujian@whamcloud.com>
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>

LU-13733 llite: report client stats sumsq

Commit cd8fb tries to account sumsq for every client operation, but
lprocfs_counter_init() did not init them properly, also add a test
case to verify new format of client stats.

Fixes: cd8fb1e8d300 ("LU-13597 ofd: add more information to job_stats")
Signed-off-by: Wang Shilong <wshilong@ddn.com>
Change-Id: I4c7fc4c3958fdab757a9755cc851836971a4b700
Reviewed-on: https://review.whamcloud.com/39223
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: Emoly Liu <emoly@whamcloud.com>
Reviewed-by: Patrick Farrell <farr0186@gmail.com>
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>

LU-13481 test: run sanity 33h with more files

Run sanity 33h with more files to avoid failure with small samples.

Test-parameters: trivial testlist=sanity env=ONLY=33h,ONLY_REPEAT=500

Signed-off-by: Lai Siyao <lai.siyao@whamcloud.com>
Change-Id: If14a852289ffc5907637a9dac6ed48890ced5586
Reviewed-on: https://review.whamcloud.com/39219
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: James Nunez <jnunez@whamcloud.com>

LU-12780 osd: use native kthreads for scrub.

The 'scrub' thread has nothing to do with ptlrpc, so
using ptlrpc_thread is strange - and not necessary.  This patch
switches to native kthread interfaces.

A particular advantage of the switch is seen in the convertion of
thread_is_running().
This is uses in two completely different ways.
1/ The thread itself needs to know if it has been asked to stop.
  It currently uss thread_is_running().  After the patch it
  uses kthread_should_stop().

2/ Other code needs to know if the thread is still active or not.  It
  previously used thread_is_running(), so it looked just like the
  first case.  Now it checks a new flag ->os_running, which is
  set precisely when SVC_RUNNING was set, and cleared when any other
  status was set.

As the thread can stop itself (e.g if osd_scrub_prep fails) or can be
asked to stop (scrub_stop()), we need to avoid confusion between the
two.  This is achieved by calling 'xchg(&scrub->os_task,NULL)'.
If the thread finds that to be non-NULL, it has stopped itself
and can just exit.  If scrub_stop() finds it to be non-NULL,
it calls kthread_should_stop() to stop the thread.

Instead of using the waitqueue in the 'struct ptlrpc_thread', we use
wake_up_var() and wait_var_event() on the 'scrub' pointer.  This is
used both to tell the thread there is work to do, and for callers to
wait for the task to make progress, or to exit.

Signed-off-by: Mr NeilBrown <neilb@suse.de>
Change-Id: Ib2f1845151734687e89a1b0d6a5135a5a4ba6e5c
Reviewed-on: https://review.whamcloud.com/38824
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: James Simmons <jsimmons@infradead.org>
Reviewed-by: Alex Zhuravlev <bzzz@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>

LU-13846 llite: move iov iter forward by ourself

Newer kernel will reward iov iter back to original
position for direct IO, see following codes:

        iov_iter_revert(from, write_len -
                        iov_iter_count(from));--------->here
out:
        return written;
}
EXPORT_SYMBOL(generic_file_direct_write);

This break assumptions from Lustre and caused problem
when Lustre need split one IO to several io loop, considering
4M block IO for 1 MiB stripe file, it will submit first 1MiB IO
4 times and caused data corruptions finally.

Since generic kernel varies from different kernel versions,
we'd better fix this problem by move iov iter forward by
Lustre itself.

Added a new test cases aiocp.c is copied from xfstests,
with codes style cleanups to make checkpatch.pl happy.

Change-Id: Iab5d8f1bb0e74ed49c821c81b734c68770edf4a8
Signed-off-by: Wang Shilong <wshilong@ddn.com>
Reviewed-on: https://review.whamcloud.com/39565
Tested-by: jenkins <devops@whamcloud.com>
Reviewed-by: Bobi Jam <bobijam@hotmail.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Yingjin Qian <qian@ddn.com>
Reviewed-by: Neil Brown <neilb@suse.de>
Reviewed-by: Oleg Drokin <green@whamcloud.com>

LU-13835 llite: reuse same cl_dio_aio for one IO

IO might be restarted if layout changed, this might
cause ki_complete() called several times for one IO.

Fixes: d1dded6 ("LU-4198 clio: AIO support for direct IO")
Fixes: 84c3e85 ("LU-13697 llite: fix short io for AIO")
Signed-off-by: Wang Shilong <wshilong@ddn.com>
Change-Id: I791b8f9d82fae6822d38293ba22adf74560c9dce
Reviewed-on: https://review.whamcloud.com/39542
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>

LU-13533 utils: ext4lazyinit should be disabled

lazyinit gets more than 24H on typical OST installation
and produces writes inside read. This influences to benchmark
tests that are usually executed just after cluster installation
is complete. Testing shows, disabling that feature adds ~30 sec
to formating OST drive so that is not a noticeable time during
install.

Explicitly send lazy_itable_init and lazy_journal_init in
conf_sanity 116 to avoid out of disk space.

HPE-bug-id: LUS-3358
Signed-off-by: Artem Blagodarenko <artem.blagodarenko@hpe.com>
Signed-off-by: Alexey Lyashkov <alexey.lyashkov@hpe.com>
Change-Id: I95bfa43d11cd67c890b036aa0b71fae4c1eea37d
Reviewed-on: https://review.whamcloud.com/38534
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: Alex Zhuravlev <bzzz@whamcloud.com>
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>

LU-13502 lnet: Conditionally attach rspt in LNetPut & LNetGet

Create a function to interpret the message type and md options to
determine whether response tracking should be enabled for a particular
PUT or GET.

Use that function in LNetPut and LNetGet to determine whether we
attach the response tracker.

HPE-bug-id: LUS-8827
Signed-off-by: Chris Horn <hornc@cray.com>
Change-Id: Ic5f2d3dedc3e773b0ec7866cccf6db9d15dc752a
Reviewed-on: https://review.whamcloud.com/38452
Tested-by: jenkins <devops@whamcloud.com>
Reviewed-by: Neil Brown <neilb@suse.de>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Petros Koutoupis <petros.koutoupis@hpe.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>

LU-10428 lnet: call event handlers without res_lock

Currently event handlers are called with the lnet_res_lock()
(a spinlock) held.  This is a problem if the handler wants
to take a mutex, allocate memory, or sleep for some other
reason.

The lock is needed because handlers for a given md need to
be serialized.  At the very least, the final event which
reports that the md is "unlinked" needs to be called last,
after any other events have been handled.

Instead of using a spinlock to ensure this ordering, we can
use a flag bit in the md.

- Before considering whether to send an event we wait for the flag bit
  to be cleared.  This ensures serialization.
- Also wait for the flag to be cleared before final freeing of the md.
- If this is not an unlink event and we need to call the handler, we
  set the flag bit before dropping lnet_res_lock().  This
  ensures the not further events will happen, and that the md
  won't be freed - so we can still clear the flag.
- use wait_var_event to wait for the flag it to be cleared,
  and wake_up_var() to signal a wakeup.  After wait_var_event()
  returns, we need to take the spinlock and check again.

Signed-off-by: Mr NeilBrown <neilb@suse.de>
Change-Id: I4dada92c4c06547bdc567838d129a8851d7de3bd
Reviewed-on: https://review.whamcloud.com/37068
Tested-by: jenkins <devops@whamcloud.com>
Reviewed-by: James Simmons <jsimmons@infradead.org>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Chris Horn <chris.horn@hpe.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>

LU-13127 ptlrpc: prefer crc32_le() over CryptoAPI

Prefer to call the crc32_le() library function directly if available,
instead of cfs_crypto_hash(CFS_HASH_ALG_CRC32). It is about 10x faster
for the 156-byte struct ptlrpc_body being checked in this function.
A test of small buffers in that compares the two implementations, run
on a 2.9GHz Core i7-7820 shows the difference is significant here:

  buffer size   156 bytes   1536 bytes   4096 bytes     1 MiB
  -----------+------------+------------+-----------+-----------
  cfs_crypto |  182 MiB/s | 1794 MiB/s | 4163 MB/s | 9631 MiB/s
  crc32_le   | 1947 MiB/s | 1871 MiB/s | 1867 MB/s | 1823 MiB/s

This corresponds to 10x faster or 1/10 as many cycles for ptlrpc_body.
The CryptoAPI speed crosses over around 1536 bytes, which is still 10x
larger than the ptlrpc_body size, so it is unlikely to be faster here.

Signed-off-by: Andreas Dilger <adilger@whamcloud.com>
Change-Id: I116fd6c148f15660dd7b7faefb86f9dd603ebbe5
Reviewed-on: https://review.whamcloud.com/39614
Reviewed-by: Sebastien Buisson <sbuisson@ddn.com>
Tested-by: jenkins <devops@whamcloud.com>
Reviewed-by: Alexey Lyashkov <alexey.lyashkov@hpe.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>