Whamcloud - gitweb
fs/lustre-release.git
3 years agoLU-7734 lnet: Multi-Rail local_ni/peer_ni selection
Amir Shehata [Tue, 5 Jan 2016 00:02:25 +0000 (16:02 -0800)]
LU-7734 lnet: Multi-Rail local_ni/peer_ni selection

This patch implements the local_ni/peer_ni selection algorithm.
It adds APIs to the peer module to encapsulate
iterating through the peer_nis in a peer and creating a peer.

Signed-off-by: Amir Shehata <amir.shehata@intel.com>
Change-Id: Ifc0e5ebf84ab25753adfcfcb433b024100f35ace
Reviewed-on: http://review.whamcloud.com/18383
Reviewed-by: Doug Oucharek <doug.s.oucharek@intel.com>
Reviewed-by: Olaf Weber <olaf@sgi.com>
Tested-by: Jenkins
Tested-by: Doug Oucharek <doug.s.oucharek@intel.com>
3 years agoLU-7734 lnet: Multi-Rail peer split
Amir Shehata [Sat, 12 Dec 2015 04:02:54 +0000 (20:02 -0800)]
LU-7734 lnet: Multi-Rail peer split

Split the peer structure into peer/peer_net/peer_ni, as
described in the Multi-Rail HLD.

Removed deathrow list in peers, instead peers are immediately
deleted. deathrow complicates memory management for peers to
little gain.

Moved to LNET_LOCK_EX for any operations which will modify the
peer tables. And CPT locks for any operatios which read the peer
tables. Therefore there is no need to use lnet_cpt_of_nid() to
calculate the CPT of the peer NID, instead we use lnet_nid_cpt_hash()
to distribute peers across multiple CPTs.

It is no longe true that peers and NIs would exist on
the same CPT. In the new design peers and NIs don't have a 1-1
relationship. You can send to the same peer from several NIs, which
can exist on separate CPTs

Signed-off-by: Amir Shehata <amir.shehata@intel.com>
Change-Id: Ida41d830d38d0ab2bb551476e4a8866d52a25fe2
Reviewed-on: http://review.whamcloud.com/18293
Tested-by: Jenkins
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: Olaf Weber <olaf@sgi.com>
Reviewed-by: Doug Oucharek <doug.s.oucharek@intel.com>
3 years agoLU-7734 lnet: Multi-Rail local NI split
Amir Shehata [Sat, 12 Dec 2015 04:02:54 +0000 (20:02 -0800)]
LU-7734 lnet: Multi-Rail local NI split

This patch allows the configuration of multiple NIs under one Net.
It is now possible to have multiple NIDs on the same network:
   Ex: <ip1>@tcp, <ip2>@tcp.
This can be configured using the following syntax:
   Ex: tcp(eth0, eth1)

The data structures for the example above can be visualized
as follows

               NET(tcp)
                |
        -----------------
        |               |
      NI(eth0)        NI(eth1)

For more details refer to the Mult-Rail Requirements and HLD
documents

Signed-off-by: Amir Shehata <amir.shehata@intel.com>
Change-Id: Id7c73b9b811a3082b61e53b9e9f95743188cbd51
Reviewed-on: http://review.whamcloud.com/18274
Tested-by: Jenkins
Reviewed-by: Doug Oucharek <doug.s.oucharek@intel.com>
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: Olaf Weber <olaf@sgi.com>
3 years agoNew tag 2.9.52 2.9.52 v2_9_52 v2_9_52_0
Oleg Drokin [Tue, 24 Jan 2017 05:26:21 +0000 (00:26 -0500)]
New tag 2.9.52

Change-Id: I7fd2714e14825f5966751a6d0a66313c1e3088b5
Signed-off-by: Oleg Drokin <oleg.drokin@intel.com>
3 years agoLU-8587 utils: Fix incorrect indenting in llapi_hsm_log_ct_progress 70/24570/3
Oleg Drokin [Sun, 1 Jan 2017 18:45:06 +0000 (13:45 -0500)]
LU-8587 utils: Fix incorrect indenting in llapi_hsm_log_ct_progress

gcc6 highlights this case of incorrect indenting:
        if (progress_type == CT_RUNNING)
                rc = llapi_json_add_item(&json_items, "current_bytes",
                                         LLAPI_JSON_BIGNUM, &current);
                if (rc < 0)
                        goto err;

Just add the braces around, though logic-wise it's all fine.

Change-Id: I770857fe2f9ce29817558247ce0987842b8d06b4
Signed-off-by: Alex Zhuravlev <bzzz@whamcloud.com>
Signed-off-by: Oleg Drokin <oleg.drokin@intel.com>
Reviewed-on: https://review.whamcloud.com/24570
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: John L. Hammond <john.hammond@intel.com>
Reviewed-by: Dmitry Eremin <dmitry.eremin@intel.com>
3 years agoLU-8996 kernel: kernel update RHEL6.8 [2.6.32-642.13.1.el6] 78/24878/2
Bob Glossman [Tue, 10 Jan 2017 17:06:03 +0000 (09:06 -0800)]
LU-8996 kernel: kernel update RHEL6.8 [2.6.32-642.13.1.el6]

Update RHEL6.8 kernel to 2.6.32-642.13.1.el6

Test-Parameters: trivial clientdistro=el6.8 mdsdistro=el6.8 ossdistro=el6.8 \
  mdsfilesystemtype=ldiskfs mdtfilesystemtype=ldiskfs \
  ostfilesystemtype=ldiskfs testgroup=review-ldiskfs

Signed-off-by: Bob Glossman <bob.glossman@intel.com>
Change-Id: I3d8ffa55ff050503a69c4db260ac6b915564349a
Reviewed-on: https://review.whamcloud.com/24878
Tested-by: Jenkins
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: Yang Sheng <yang.sheng@intel.com>
Reviewed-by: Minh Diep <minh.diep@intel.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>
3 years agoLU-8929 lfsck: dumper gets current position properly 69/24869/2
Fan Yong [Tue, 27 Sep 2016 15:12:12 +0000 (23:12 +0800)]
LU-8929 lfsck: dumper gets current position properly

It is normal that the LFSCK iteration has been done when the
dump hanlder found the status as LS_SCANNING_PHASE1, it may
because of race, or the LFSCK failed to update the status.
Under such cases, the dump handler will use the position in
the last checkpoint as the current position. It may be not
100% accurate, but not serious issue.

Signed-off-by: Fan Yong <fan.yong@intel.com>
Change-Id: I672258baa9d0b0aa8ec12249c13b2b147a274ab4
Reviewed-on: https://review.whamcloud.com/24869
Tested-by: Jenkins
Reviewed-by: Andreas Dilger <andreas.dilger@intel.com>
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: Lai Siyao <lai.siyao@intel.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>
3 years agoLU-7481 utils: label lustre device correctly 45/24845/3
Hongchao Zhang [Sat, 8 Oct 2016 19:49:33 +0000 (03:49 +0800)]
LU-7481 utils: label lustre device correctly

Currently, the device label will be read before mounting Lustre,
the flags LDD_F_VIRGIN and LDD_F_WRITECONF will be set according
to the label (the corresponding original flags containing in the
lustre_disk_data will be ignored). But the device label could be
changed during mount by the journal recovery, and the device should
be also labeled to indicate the target device is supposed to start.

Test-Parameters: testlist=conf-sanity,conf-sanity,conf-sanity

Change-Id: I2df1d81f764a7d1ffa26afb0197d137c057a25e9
Signed-off-by: Hongchao Zhang <hongchao.zhang@intel.com>
Reviewed-on: https://review.whamcloud.com/24845
Reviewed-by: Andreas Dilger <andreas.dilger@intel.com>
Tested-by: Jenkins
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: Jian Yu <jian.yu@intel.com>
Reviewed-by: Fan Yong <fan.yong@intel.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>
3 years agoLU-8932 lnet: define new network driver ptl4lnd 68/24768/3
Gregoire Pichon [Tue, 13 Dec 2016 13:41:42 +0000 (14:41 +0100)]
LU-8932 lnet: define new network driver ptl4lnd

Assign an ID to the new network driver ptl4lnd developped by Bull
that implements a LND based on Portals 4 API. It is intended to be
used with BXI, the Bull interconnect hardware.

Signed-off-by: Gregoire Pichon <gregoire.pichon@bull.net>
Change-Id: I38e505916899dc7f01b3ad0372c9f068fa06f308
Reviewed-on: https://review.whamcloud.com/24768
Tested-by: Jenkins
Reviewed-by: Doug Oucharek <doug.s.oucharek@intel.com>
Reviewed-by: Andreas Dilger <andreas.dilger@intel.com>
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: James Simmons <uja.ornl@yahoo.com>
3 years agoLU-7429 tests: generate dangling name entry properly 63/24763/2
Fan Yong [Sun, 25 Sep 2016 00:24:10 +0000 (08:24 +0800)]
LU-7429 tests: generate dangling name entry properly

There may be some creation after the dangling injection, such as
create ".lustre/lost+found/MDTxxxx" for LFSCK, create some quota
related local files. These creations may reuse the just released
local object/inode that is referenced by the dangling name entry.
That will fail the dangling injection as to the subsequent LFSCK
will not find dangling name entry. So before deleting the target
object for the dangling name entry, remove some other objects to
avoid the target object being reused by some potential creations.

Signed-off-by: Fan Yong <fan.yong@intel.com>
Change-Id: I0e1bb60c1095119e10009fe2a8ce38687e3e7692
Reviewed-on: https://review.whamcloud.com/24763
Tested-by: Jenkins
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: Jian Yu <jian.yu@intel.com>
Reviewed-by: Lai Siyao <lai.siyao@intel.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>
3 years agoLU-4423 ptlrpc: use 64-bit times for ptlrpc_sec 10/24710/3
Arnd Bergmann [Wed, 11 Jan 2017 01:50:19 +0000 (20:50 -0500)]
LU-4423 ptlrpc: use 64-bit times for ptlrpc_sec

Here we use an unsigned long to store the timeout for gc,
which is probably safe until 2106, but this patch converts it
to use ktime_get_real_seconds() and time64_t for consistency.

Linux-commit: 8cc980713ec9e6847896891c54562ad815c33424

Change-Id: I9c66ac818239debe676b78fbee5764cd5b69028c
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Oleg Drokin <green@linuxhacker.ru>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Signed-off-by: James Simmons <uja.ornl@yahoo.com>
Reviewed-on: https://review.whamcloud.com/24710
Tested-by: Jenkins
Reviewed-by: Andreas Dilger <andreas.dilger@intel.com>
Reviewed-by: Sebastien Buisson <sbuisson@ddn.com>
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: Dmitry Eremin <dmitry.eremin@intel.com>
3 years agoLU-4423 ptlrpc: use 64-bit times in ptlrpc_enc_page_pool 08/24708/2
Arnd Bergmann [Wed, 4 Jan 2017 21:40:12 +0000 (16:40 -0500)]
LU-4423 ptlrpc: use 64-bit times in ptlrpc_enc_page_pool

ptlrpc_enc_page_pool computes time deltas using 'long' values from
get_seconds(). This is probably safe beyond y2038, but it's better
to go use monotonic times and 64-bit here for consistency.

Linux-commit: 80018a9edbc3180ae31a7197f9dacab975a7f5e2

Change-Id: I15511b04d8b4f7353ce109d8d9bb7887f551d880
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Oleg Drokin <green@linuxhacker.ru>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Signed-off-by: James Simmons <uja.ornl@yahoo.com>
Reviewed-on: https://review.whamcloud.com/24708
Tested-by: Jenkins
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: Andreas Dilger <andreas.dilger@intel.com>
Reviewed-by: Sebastien Buisson <sbuisson@ddn.com>
Reviewed-by: Dmitry Eremin <dmitry.eremin@intel.com>
3 years agoLU-4423 lnet: Better cookie gen 82/24682/6
Tina Ruchandani [Thu, 5 Jan 2017 14:53:30 +0000 (09:53 -0500)]
LU-4423 lnet: Better cookie gen

api-ni.c uses do_gettimeofday to get a 'cookie' or timestamp.
This patch replaces it with ktime_get_ns for the following reasons:

1. ktime_get_ns returns a __u64 which is safer than 'struct timeval'
   which will overflow on 32-bit systems in year 2038 and beyond.
2. Improved resolution: nsecs instead of usecs.
3. Reduced compute: ktime_get_ns is faster than the multiply/add
   combination used in this function

Linux-commit: 9056be30542bfff51190bdda67088f319cf4c9f5

Drop unneeded wrapper function Remove the function
lnet_create_interface_cookie() and replace its call
with the function ktime_get_ns().

Linux-commit: 7bcd831b8579212303ec7c30e975432b914493dc

The ln_interface_cookie is used to ensure that a node can tell whether
the following sequence of events has happened:

node sends GET or PUT to peer
node is rebooted
peer sends REPLY or ACK to node

The ln_interface_cookie is set once, when LNet starts, and remains
unchanged afterwards. To avoid accidentally obtaining the same cookie
after a reboot, the code generated ths cookie using ktime_get_ns().
Once generated, the value of the cookie is not interpreted, only
compared for equality. Olaf Weber reported that due to the use of
ktime_get_ns() a small chance exist of generating a cookie of identical
value across reboots. Using ktime_get_real_ns() removes any chance of
this from happening.

Change-Id: I159a0ff2573afb87f279a8e8f282b0ac076d9bf3
Signed-off-by: Tina Ruchandani <ruchandani.tina@gmail.com>
Signed-off-by: Shivani Bhardwaj <shivanib134@gmail.com>
Suggested-by: Arnd Bergmann <arnd@arndb.de>
Reviewed-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Signed-off-by: James Simmons <uja.ornl@yahoo.com>
Reviewed-on: https://review.whamcloud.com/24682
Reviewed-by: Doug Oucharek <doug.s.oucharek@intel.com>
Reviewed-by: Olaf Weber <olaf@sgi.com>
Tested-by: Jenkins
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: Dmitry Eremin <dmitry.eremin@intel.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>
3 years agoLU-4423 libcfs: Use swap() in cfs_hash_bd_order() 76/24576/2
Amitoj Kaur Chawla [Mon, 2 Jan 2017 00:19:16 +0000 (19:19 -0500)]
LU-4423 libcfs: Use swap() in cfs_hash_bd_order()

Use swap() function instead of using a temporary variable for swapping
two variables.

The Coccinelle semantic patch used to make this change is as follows:
//<smpl>
@@
type T;
T a,b,c;
@@
- a = b;
- b = c;
- c = a;
+ swap(b, c);
//<smpl>

Signed-off-by: Amitoj Kaur Chawla <amitoj1606@gmail.com>
Change-Id: I19aa52fe4e05fed2e03c1d8515731a5ce01b3d09
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Signed-off-by: James Simmons <uja.ornl@yahoo.com>
Reviewed-on: https://review.whamcloud.com/24576
Tested-by: Jenkins
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: Andreas Dilger <andreas.dilger@intel.com>
Reviewed-by: Dmitry Eremin <dmitry.eremin@intel.com>
3 years agoLU-4315 docs: Fix Makefile.am to have one man page per line 71/24371/4
Steve Guminski [Thu, 15 Dec 2016 15:20:05 +0000 (10:20 -0500)]
LU-4315 docs: Fix Makefile.am to have one man page per line

The man pages for lfs(1) and lctl(8) are quite large. Splitting
them into one page per subcommand will allow for more detailed
information for each subcommand.

This patch modifies the Makefile.am so that the source man pages
are listed one per line.  This will make it easier for subsequent
patches to add new pages.  Existing pages that were missing from
the Makefile.am have been added to it, and an obsolete page has
been removed from the Makefile.am and been deleted.

Test-Parameters: trivial
Signed-off-by: Steve Guminski <stephenx.guminski@intel.com>
Change-Id: If312fff9bc5e68176caf0a70a51876e69b1614d8
Reviewed-on: https://review.whamcloud.com/24371
Tested-by: Jenkins
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: Andreas Dilger <andreas.dilger@intel.com>
Reviewed-by: Bob Glossman <bob.glossman@intel.com>
Reviewed-by: James Simmons <uja.ornl@yahoo.com>
3 years agoLU-8926 llite: reduce jobstats race window 53/24253/5
Patrick Farrell [Tue, 13 Dec 2016 15:43:34 +0000 (09:43 -0600)]
LU-8926 llite: reduce jobstats race window

In the current code, lli_jobid is set to zero on every call
to lustre_get_jobid.  This causes problems, because it's
used asynchronously to set the job id in RPCs, and some
RPCs will falsely get no jobid set.  (For small IO sizes,
this can be up to 60% of RPCs.)

It would be very expensive to put hard synchronization
between this and every outbound RPC, and it's OK to very
rarely get an RPC without correct job stats info.

This patch only updates the lli_jobid when the job id has
changed, which leaves only a very small window for reading
an inconsistent job id.

Signed-off-by: Patrick Farrell <paf@cray.com>
Change-Id: I6c3a7f8683dc5f5d467940920938db18b0c20462
Reviewed-on: https://review.whamcloud.com/24253
Tested-by: Jenkins
Reviewed-by: Andreas Dilger <andreas.dilger@intel.com>
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: Chris Horn <hornc@cray.com>
3 years agoLU-8855 llite: return small device numbers for compat stat() 77/23877/3
John L. Hammond [Mon, 21 Nov 2016 15:22:52 +0000 (09:22 -0600)]
LU-8855 llite: return small device numbers for compat stat()

The compat_sys_*stat*() syscalls will fail unless the devices majors
and minors are both less than 256. So in ll_getattr_it(), if we are in
32 bit compat mode then coerce the device numbers in to the expected
format.

Signed-off-by: John L. Hammond <john.hammond@intel.com>
Change-Id: I1bf13258902e13c76b9ebf3476fd1767712de0b3
Reviewed-on: https://review.whamcloud.com/23877
Tested-by: Jenkins
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>
Reviewed-by: Andreas Dilger <andreas.dilger@intel.com>
3 years agoLU-8821 mdt: avoid double find in mdt_path_current() 01/23701/2
John L. Hammond [Thu, 12 Nov 2015 15:49:09 +0000 (09:49 -0600)]
LU-8821 mdt: avoid double find in mdt_path_current()

In mdt_path_current() avoid finding the object we are already holding
a reference to.

Signed-off-by: John L. Hammond <john.hammond@intel.com>
Change-Id: Iae4796047d2c5b02989d29baf2e7620545f7e45c
Reviewed-on: https://review.whamcloud.com/23701
Tested-by: Jenkins
Reviewed-by: Jinshan Xiong <jinshan.xiong@intel.com>
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: Andreas Dilger <andreas.dilger@intel.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>
3 years agoLU-8066 obdclass: move lustre sysctl to sysfs 28/23428/6
Oleg Drokin [Sun, 1 Jan 2017 19:11:04 +0000 (14:11 -0500)]
LU-8066 obdclass: move lustre sysctl to sysfs

Backport from upstream the changes to port lustre
systctl to sysfs. Needed to re-export the function
lprocfs_read_frac_helper for later work. The
following patches were backported:

Fix class_procfs_init error return value. Dan Carpenter noticed
that procfs conversion patches introduced a bug where should
kobject_create_and_add, an error is not returned from
class_procfs_init.

Linux-commit: 3c4872f94359ed38a1392c0a9238c48a9aee6f8f

Move sysctl timeout to sysfs. This is the first step of
moving lustre sysctls from /proc/sys/lustre to /sys/fs/lustre

Linux-commit: e2424a1265f2772b66f068c205256e2aef5f74a0

Move max_dirty_mb from sysctl to sysfs. max_dirty_mb is
now a parameter in /sys/fs/lustre.

Linux-commit: df476a4d5de09d9324b108fc9c5ff2c00a0850d0

Move debug controls to sysfs. debug_peer_on_timeout,
dump_on_timeout and dump_on_eviction controls from
/proc/sys/lustre to /sys/fs/lustre

Linux-commit: 9e7fa14935901bcd09576b2866d5dd15f69caf83

Move AT controls from sysctl to sysfs. Adaptive Timeouts
controls are being moved from /proc/sys/lustre to
/sys/fs/lustre

Linux-commit: bcef118e7ed67e28edcaab9be9ca11412176c540

Signed-off-by: Oleg Drokin <green@linuxhacker.ru>
Change-Id: Id1b00bebf9ecca5284e9c71f4c0f91e56cbf391b
Reported-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Signed-off-by: James Simmons <uja.ornl@yahoo.com>
Reviewed-on: https://review.whamcloud.com/23428
Tested-by: Jenkins
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: Andreas Dilger <andreas.dilger@intel.com>
Reviewed-by: Dmitry Eremin <dmitry.eremin@intel.com>
3 years agoLU-4423 ldlm: use 64-bit time for pl_recalc 50/23350/4
Arnd Bergmann [Sun, 1 Jan 2017 16:41:33 +0000 (11:41 -0500)]
LU-4423 ldlm: use 64-bit time for pl_recalc

The ldlm pool calculates elapsed time by comparing the previous and
current get_seconds() values, which is unsafe on 32-bit machines
after 2038.

This changes the code to use time64_t and ktime_get_real_seconds(),
keeping the 'real' instead of 'monotonic' time because of the
debug prints.

Linux-commit: 8f83409cf2382c968f96877368cd5b542b92af1d
Linux-commit: b8cb86fd95bb461c3496e1f4b4083b198c963a9c

Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Change-Id: I81cca5b529dbf5615cf46461ad1c9179fdee7835
Signed-off-by: Oleg Drokin <green@linuxhacker.ru>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Signed-off-by: James Simmons <uja.ornl@yahoo.com>
Reviewed-on: https://review.whamcloud.com/23350
Tested-by: Jenkins
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: Andreas Dilger <andreas.dilger@intel.com>
Reviewed-by: Dmitry Eremin <dmitry.eremin@intel.com>
3 years agoLU-8710 ptlrpc: use current CPU instead of harcoded 0 05/23305/3
Dmitry Eremin [Fri, 21 Oct 2016 12:38:46 +0000 (15:38 +0300)]
LU-8710 ptlrpc: use current CPU instead of harcoded 0

fix crash if CPU 0 disabled.

Change-Id: I8ac5a10f544a1c8fc454bc64a6bb1d3607240be9
Signed-off-by: Dmitry Eremin <dmitry.eremin@intel.com>
Reviewed-on: https://review.whamcloud.com/23305
Tested-by: Jenkins
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: Doug Oucharek <doug.s.oucharek@intel.com>
Reviewed-by: Andreas Dilger <andreas.dilger@intel.com>
3 years agoLU-8703 libcfs: remove usless CPU partition code 03/23303/3
Dmitry Eremin [Fri, 21 Oct 2016 11:56:31 +0000 (14:56 +0300)]
LU-8703 libcfs: remove usless CPU partition code

 * remove scratch buffer and mutex which guard it.
 * remove global cpumask and spinlock which guard it.
 * remove cpt_version for checking CPUs state change during setup
   because of just disable CPUs state change during setup.
 * remove whole global struct cfs_cpt_data cpt_data.
 * remove few unused APIs.

Change-Id: I0cc853d57952e76cf32801838a19e6872905aaa0
Signed-off-by: Dmitry Eremin <dmitry.eremin@intel.com>
Reviewed-on: https://review.whamcloud.com/23303
Tested-by: Jenkins
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: James Simmons <uja.ornl@yahoo.com>
Reviewed-by: Doug Oucharek <doug.s.oucharek@intel.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>
3 years agoLU-4423 obd: use ktime_t for calculating elapsed time 46/23146/9
Arnd Bergmann [Tue, 10 Jan 2017 23:16:11 +0000 (18:16 -0500)]
LU-4423 obd: use ktime_t for calculating elapsed time

process_param2_config() tries to print how much time has passed
across a call_usermodehelper() function, and uses struct timeval
for that.

We want to remove this structure, so this is better expressed
in terms of ktime_t and ktime_us_delta().

Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Change-Id: I90c61d9d49ee0d500772f1b370790e37859f18b2
Signed-off-by: Oleg Drokin <green@linuxhacker.ru>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Signed-off-by: James Simmons <uja.ornl@yahoo.com>
Reviewed-on: https://review.whamcloud.com/23146
Tested-by: Jenkins
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: Andreas Dilger <andreas.dilger@intel.com>
Reviewed-by: Dmitry Eremin <dmitry.eremin@intel.com>
3 years agoLU-7471 tests: Modified make_custom_file_for_progress fn 46/17346/7
Aditya Pandit [Tue, 24 Nov 2015 09:35:40 +0000 (15:05 +0530)]
LU-7471 tests: Modified make_custom_file_for_progress fn

After executing tests like test_200 and if there is not
enough space the function make_custom_file_for_progress was returning
1 on error but this error was not getting caught properly.
Modified the code to catch the error properly.

Test-Parameters: trivial testlist=sanity-hsm

Seagate-bug-id: MRP-3026
Signed-off-by: Mikhail V. Pridushchenko <mikhail.v.pridushchenko@seagate.com>
Signed-off-by: Aditya Pandit <aditya.pandit@seagate.com>
Change-Id: I0db45e327d6f49d066b5c631096a459b8acd2758
Signed-off-by: Ashish Purkar <ashish.purkar@seagate.com>
Reviewed-on: https://review.whamcloud.com/17346
Reviewed-by: Emoly Liu <emoly.liu@intel.com>
Tested-by: Jenkins
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: Andreas Dilger <andreas.dilger@intel.com>
3 years agoLU-3764 tests: clean up sanity test_116a code style 82/7882/4
Andreas Dilger [Tue, 8 Oct 2013 18:56:06 +0000 (12:56 -0600)]
LU-3764 tests: clean up sanity test_116a code style

Clean up the code style for sanity.sh test_116a():
- proper indentation
- don't use $ for variables inside $((...))
- one statement per line

Test-Parameters: trivial
Signed-off-by: Andreas Dilger <andreas.dilger@intel.com>
Change-Id: I5fe502b85365e00ce34b90f8fabfb58320a07e51
Reviewed-on: https://review.whamcloud.com/7882
Tested-by: Jenkins
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: James Nunez <james.a.nunez@intel.com>
Reviewed-by: Steve Guminski <stephenx.guminski@intel.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>
3 years agoLU-9019 osd : remove struct timeval use in osd-ldiskfs 96/24896/2
James Simmons [Fri, 13 Jan 2017 21:39:23 +0000 (16:39 -0500)]
LU-9019 osd : remove struct timeval use in osd-ldiskfs

For brw_stats_show change the output to use a timespec64
type to avoid the overflow. Also change the format to
print the sub-second portion as 9 digits (nanoseconds)
for clarity, rather than printing six digits without
leading zeroes.

Both osd_write_prep() and osd_read_prep() want to report
the time it took to perform its operations in the
osd_get_page counter. This is currently done with
a call to do_gettimeofday with struct timeval which is
not 2038 safe. Move this operation to 64 bit time
handling.

Change-Id: I457ac799d855d2596220b6e0d0c5039e8b00021f
Signed-off-by: James Simmons <uja.ornl@yahoo.com>
Reviewed-on: https://review.whamcloud.com/24896
Tested-by: Jenkins
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: Fan Yong <fan.yong@intel.com>
Reviewed-by: Alex Zhuravlev <alexey.zhuravlev@intel.com>
Reviewed-by: Dmitry Eremin <dmitry.eremin@intel.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>
3 years agoLU-8562 osp: osp_precreate_thread gets stuck after disconnect 58/24758/3
Ned Bass [Sat, 7 Jan 2017 01:43:47 +0000 (17:43 -0800)]
LU-8562 osp: osp_precreate_thread gets stuck after disconnect

osp_precreate_thread() can get stuck because d->opd_got_disconnected
never gets reset. When opd_got_disconnected is set,
osp_precreate_cleanup_orphans() returns early with EAGAIN and can't
clear d->opd_pre_recovering. And because d->opd_pre_recovering can't
be cleared we always break out of the while loop where
d->opd_got_disconnected normally gets reset. So
osp_precreate_cleanup_orphans() is stuck always failing.

Signed-off-by: Ned Bass <bass6@llnl.gov>
Change-Id: I0b4f4e2e55e7a8d7ffae633a4d3c578b4a484ae2
Reviewed-on: https://review.whamcloud.com/24758
Tested-by: Jenkins
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: Sergey Cheremencev <sergey.cheremencev@seagate.com>
Reviewed-by: Alex Zhuravlev <alexey.zhuravlev@intel.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>
3 years agoLU-8840 osp: handle EA cache properly 82/23782/10
Fan Yong [Thu, 22 Sep 2016 08:54:55 +0000 (16:54 +0800)]
LU-8840 osp: handle EA cache properly

For success case, dt_xattr_get() should return the EA size
instead of zero. If such EA does not exist, return -ENODATA.

More code cleanup for OSP EA cache to avoid potential reference
leak, buffer overflow, and so on.

Signed-off-by: Fan Yong <fan.yong@intel.com>
Change-Id: I352b99b1ed08f1b15bdb8da2bf28689ae2d61c23
Reviewed-on: https://review.whamcloud.com/23782
Tested-by: Jenkins
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: Lai Siyao <lai.siyao@intel.com>
Reviewed-by: Alex Zhuravlev <alexey.zhuravlev@intel.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>
3 years agoLU-6245 libcfs: remove wrappers for timer functions 33/23733/4
James Simmons [Tue, 10 Jan 2017 23:21:33 +0000 (18:21 -0500)]
LU-6245 libcfs: remove wrappers for timer functions

Use the linux timer_list api directly instead of going
throught a abstraction layer in libcfs.

Signed-off-by: James Simmons <uja.ornl@yahoo.com>
Change-Id: Ida47b2746e68b7428cbc30b40fbdc8810b7d05f7
Reviewed-on: https://review.whamcloud.com/23733
Tested-by: Jenkins
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: Dmitry Eremin <dmitry.eremin@intel.com>
Reviewed-by: Andreas Dilger <andreas.dilger@intel.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>
3 years agoLU-8732 tests: lnet-selftest Error inserting modules 50/23250/2
Abrarahmed Momin [Tue, 22 Mar 2016 12:09:47 +0000 (17:39 +0530)]
LU-8732 tests: lnet-selftest Error inserting modules

Prevent locally derived module option MODOPTS_LIBCFS being
passed as part of environment variable to remote nodes.
This issue is more prominent in interop setups that have
different version LIBCFS library supporting different module
parameters.

Test-Parameters: trivial testlist=lnet-selftest
Signed-off-by: Abrarahmed Momin <abrar.habib@seagate.com>
Signed-off-by: Arshad Hussain <arshad.hussain@seagate.com>
Change-Id: Ie56103c55611bb62a7dfb314526c48781ab20bc4
Seagate-bug-id: MRP-2675
Reviewed-by: Ujjwal Lanjewar <ujjwal.lanjewar@seagate.com>
Reviewed-by: Ashish Purkar <ashish.purkar@seagate.com>
Reviewed-by: Elena V. Gryaznova <elena.gryaznova@seagate.com>
Tested-by: Elena V. Gryaznova <elena.gryaznova@seagate.com>
Reviewed-on: https://review.whamcloud.com/23250
Tested-by: Jenkins
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: Doug Oucharek <doug.s.oucharek@intel.com>
Reviewed-by: Andreas Dilger <andreas.dilger@intel.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>
3 years agoLU-9019 lnet: simplify lnet_eq_wait_locked 47/23147/9
James Simmons [Fri, 13 Jan 2017 22:54:34 +0000 (17:54 -0500)]
LU-9019 lnet: simplify lnet_eq_wait_locked

We can simplify the code by taking advantage of the behavior
of schedule_timeout_interruptible(). Instead of testing if
tms is less than zero we can pass in a signed long that
schedule_timeout_interruptible is expecting and for the case
of no timeout we can pass in MAX_SCHEDULE_TIMEOUT.

Change-Id: I15126aab500e975ec34ee270d3a1f9b83cf50902
Signed-off-by: James Simmons <uja.ornl@yahoo.com>
Reviewed-on: https://review.whamcloud.com/23147
Reviewed-by: Doug Oucharek <doug.s.oucharek@intel.com>
Tested-by: Jenkins
Reviewed-by: Dmitry Eremin <dmitry.eremin@intel.com>
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>
3 years agoLU-6142 lnet: code style cleanups, improve error handling 69/18469/5
James Simmons [Tue, 10 Jan 2017 23:11:34 +0000 (18:11 -0500)]
LU-6142 lnet: code style cleanups, improve error handling

During submission of the dynamic LNet work to the upstream
client Dan Carpenter suggested several improvements and a
few fixes. They are:

1) Rename the goto labels to be more clear.

2) Replace CLASSERT with BUILD_BUG_ON

3) Rework the loops to reduce the code indentation level
   and to make them more clear.

4) Add test to ensure the ioctl data is really smaller
   than what is reported in the ioctl header.

5) Fix one off error in lnet_get_peer_info

6) Handle errors instead of sucesses in lnet_dyn_[un]configure

Change-Id: If003d313ae587fd6311d9dfb419e8c6f59ef4705
Suggested-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: James Simmons <uja.ornl@yahoo.com>
Reviewed-on: https://review.whamcloud.com/18469
Tested-by: Jenkins
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: Doug Oucharek <doug.s.oucharek@intel.com>
Reviewed-by: Olaf Weber <olaf@sgi.com>
Reviewed-by: Dmitry Eremin <dmitry.eremin@intel.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>
3 years agoLU-7115 fld: don't retry for no LWP device 41/17041/13
Yang Sheng [Wed, 4 Nov 2015 08:36:37 +0000 (16:36 +0800)]
LU-7115 fld: don't retry for no LWP device

Just try again for LWP import since other client
import may run into deadloop.

Signed-off-by: Yang Sheng <yang.sheng@intel.com>
Change-Id: If1759b5ad7daa06abe6231669ad92e006ecb3059
Reviewed-on: https://review.whamcloud.com/17041
Tested-by: Jenkins
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: Andreas Dilger <andreas.dilger@intel.com>
Reviewed-by: wangdi <di.wang@intel.com>
Reviewed-by: Niu Yawei <yawei.niu@intel.com>
Reviewed-by: Fan Yong <fan.yong@intel.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>
3 years agoLU-8994 lmv: honour the specified stripe index 77/24777/3
Di Wang [Mon, 9 Jan 2017 15:17:09 +0000 (10:17 -0500)]
LU-8994 lmv: honour the specified stripe index

when creating the striped directory, specified
stripe index should always be used even the parent
has default stripe index.

Signed-off-by: Di Wang <di.wang@intel.com>
Change-Id: Ief00af3790282a0887b9d28c581ccf0070cc758f
Reviewed-on: https://review.whamcloud.com/24777
Reviewed-by: Andreas Dilger <andreas.dilger@intel.com>
Reviewed-by: Fan Yong <fan.yong@intel.com>
Tested-by: Jenkins
Reviewed-by: Lai Siyao <lai.siyao@intel.com>
Tested-by: Maloo <hpdd-maloo@intel.com>
3 years agoLU-8753 osp: add rpc generation 64/24364/6
Di Wang [Wed, 14 Dec 2016 22:13:30 +0000 (17:13 -0500)]
LU-8753 osp: add rpc generation

Add rpc generation to make sure current update
request will not be sent until the remote llog
object got refresh.

Signed-off-by: Di Wang <di.wang@intel.com>
Change-Id: Iae678686b522d545b69510444805a1e411acfcfe
Reviewed-on: https://review.whamcloud.com/24364
Reviewed-by: Fan Yong <fan.yong@intel.com>
Tested-by: Jenkins
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: Andreas Dilger <andreas.dilger@intel.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>
3 years agoLU-8871 kernel: kernel upgrade [SLES12 SP2 4.4.21-84] 57/24257/12
Bob Glossman [Thu, 8 Dec 2016 16:49:51 +0000 (08:49 -0800)]
LU-8871 kernel: kernel upgrade [SLES12 SP2 4.4.21-84]

Major linux version upgrade, from linux 3.12 to linux 4.4.
All new ldiskfs and kernel patches.
All new target and config files.

Test-Parameters: clientdistro=sles12sp2 testgroup=review-ldiskfs \
  mdsdistro=sles12sp2 ossdistro=sles12sp2 mdsfilesystemtype=ldiskfs \
  mdtfilesystemtype=ldiskfs ostfilesystemtype=ldiskfs

Signed-off-by: Bob Glossman <bob.glossman@intel.com>
Change-Id: I354197d5426ebde3da6c814bfb7f97f0f04113eb
Reviewed-on: https://review.whamcloud.com/24257
Tested-by: Jenkins
Reviewed-by: Yang Sheng <yang.sheng@intel.com>
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: Minh Diep <minh.diep@intel.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>
3 years agoLU-4423 ptlrpc: use 64-bit times for ptlrpc sec expiry 09/24709/2
Arnd Bergmann [Wed, 4 Jan 2017 21:41:46 +0000 (16:41 -0500)]
LU-4423 ptlrpc: use 64-bit times for ptlrpc sec expiry

The exp_flvr_expire and imp_sec_expire are defined as 'unsigned long',
which doesn't overflow until 2106, but to be on the safe side, this
changes the code to use time64_t like we do everywhere else.

Linux-commit: 986ef135c2afce959971b0086fc04e0f9f654120

Change-Id: Id1a560504e9a9c673e0880ac4e3e80938b3789bc
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Oleg Drokin <green@linuxhacker.ru>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Signed-off-by: James Simmons <uja.ornl@yahoo.com>
Reviewed-on: https://review.whamcloud.com/24709
Tested-by: Jenkins
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: Andreas Dilger <andreas.dilger@intel.com>
Reviewed-by: Sebastien Buisson <sbuisson@ddn.com>
Reviewed-by: John L. Hammond <john.hammond@intel.com>
Reviewed-by: Dmitry Eremin <dmitry.eremin@intel.com>
3 years agoLU-8945 ptlrpc : remove userland usage from ptlrpc 96/24396/4
James Simmons [Tue, 27 Dec 2016 15:08:15 +0000 (10:08 -0500)]
LU-8945 ptlrpc : remove userland usage from ptlrpc

The reason for __REQ_LAYOUT_USER__ was to expose a
section of code in layout.c to userland for a utility
similar to wireshark. This was done before wireshark
existed but now that it does we no longer need to do
this type of hack. This also reduces lustre_acl.h to
strictly a kernel header now.

Test-Parameters: trivial

Change-Id: I649e36947789b8ebcb1204b65ac68b55a1d6ea49
Signed-off-by: James Simmons <uja.ornl@yahoo.com>
Reviewed-on: https://review.whamcloud.com/24396
Tested-by: Jenkins
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: Dmitry Eremin <dmitry.eremin@intel.com>
Reviewed-by: Andreas Dilger <andreas.dilger@intel.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>
3 years agoLU-8835 osc: handle 64 bit time properly in osc_cache_too_much 14/23814/5
James Simmons [Tue, 10 Jan 2017 22:40:40 +0000 (17:40 -0500)]
LU-8835 osc: handle 64 bit time properly in osc_cache_too_much

Use 64 bit time for cl_lru_last_used and the function
osc_cache_too_much(). The use of 64 bit time will
introduce an expensive 64 bit division operation. Since
the time lapse being calculated in osc_cache_too_much
will never be more than seventy years we can cast the
time lapse to an long and we can perform a normal 32 bit
divison operation instead.

Signed-off-by: James Simmons <uja.ornl@yahoo.com>
Change-Id: Ic38db2d22436fc2a8aa431562e4f5d22ec18d842
Reviewed-on: https://review.whamcloud.com/23814
Reviewed-by: Andreas Dilger <andreas.dilger@intel.com>
Tested-by: Jenkins
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: Dmitry Eremin <dmitry.eremin@intel.com>
3 years agoLU-6210 lod: Change positional struct initializers to C99 54/23754/4
Steve Guminski [Mon, 14 Nov 2016 18:12:28 +0000 (13:12 -0500)]
LU-6210 lod: Change positional struct initializers to C99

This patch makes no functional changes.  Struct initializers in the
lod directory that use C89 or GCC-only syntax are updated to C99
syntax.

The C99 syntax prevents incorrect initialization if values are
accidently placed in the wrong position, allows changes in the struct
definition, and clears any members that are not given an explicit
value.

The following struct initializers have been updated:

lustre/lod/lod_dev.c:
struct lu_seq_range range
struct l_wait_info lwi
lustre/lod/lproc_lod.c:
static struct lprocfs_vars lprocfs_lod_osd_vars[]

Test-Parameters: trivial
Signed-off-by: Steve Guminski <stephenx.guminski@intel.com>
Change-Id: I6533b9cbfd9a98e903ba8e0f5c5d8e725226366d
Reviewed-on: https://review.whamcloud.com/23754
Tested-by: Jenkins
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: Nathaniel Clark <nathaniel.l.clark@intel.com>
Reviewed-by: James Simmons <uja.ornl@yahoo.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>
3 years agoLU-8066 obd: Add debugfs root 27/23427/4
Oleg Drokin [Sun, 1 Jan 2017 15:59:10 +0000 (10:59 -0500)]
LU-8066 obd: Add debugfs root

This is just plumbing for migrating remaining procfs to
debugfs support

Linux-commit: 5c8c82f63a11c07a0687d2c71411166017012689

ldebugfs_remove is usually called on directories with files passed in
as attributes, so simple debugfs_remove failes on them as not empty
Switch to debugfs_remove_recursive.

This fixes a number of problems where a new filesystem is mounted after
being unmounted first.

Linux-commit: 6a491f2b80f2806221ba3a5a3e26fbe945f82d83

Turns out we mistakenly export some pretty-wide-reaching debugfs
functions as EXPORT_SYMBOL instead of EXPORT_SYMBOL_GPL as we should,
so this patch rectifies the situation.

Linux-commit: da33f1dd9f69c5a5d23e1254906b027af3362e36

Move /proc/fs/lustre/devices to debugfs. The devices file prints out
status information about all obd devices in the system in human
readable form.

Linux-commit: 4361a048743f900bb0710bd7cb36a650d7bef93a

Signed-off-by: Dmitry Eremin <dmitry.eremin@intel.com>
Change-Id: Iefe6ee69b607f627c5682513128706bf0c1c3e09
Signed-off-by: Oleg Drokin <oleg.drokin@intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Signed-off-by: James Simmons <uja.ornl@yahoo.com>
Reviewed-on: https://review.whamcloud.com/23427
Tested-by: Jenkins
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: Jian Yu <jian.yu@intel.com>
Reviewed-by: Emoly Liu <emoly.liu@intel.com>
3 years agoLU-8288 lfsck: handle dangling LOV EA reference 62/21562/17
Fan Yong [Sat, 10 Sep 2016 11:30:42 +0000 (19:30 +0800)]
LU-8288 lfsck: handle dangling LOV EA reference

Originally, the layout LFSCK logic of handling dangling LOV EA
reference is as following:

During the first phase scanning, if the layout LFSCK find that
some LOV EA entry references an OST-object that does not exist,
then it will repair the inconsistency based on the LFSCK start
parameter "-c" option. If "-c" option is specified, the layout
LFSCK will think the OST-object lost, then it will create the
lost OST-object with the FID that is stored in the LOV EA slot.
But such repairing may be incorrect. Because the LOV EA may be
corrupted as to the LOV EA is invalid. Means the OST-object is
still on the OST. When moves to the second stage scaning, the
layout LFSCK will find the orphan OST-object that claims to be
as one of the MDT-object's stripe. And if someone has already
modified the new created OST-object before finding the orphan,
then the layout LFSCK cannot recover the original data back.

To avoid above trouble, the patch introduces new start option:
"--delay-create-ostobj" or short described as "-d". It allows
the layout LFSCK to postpone creating the "lost" OST-object
until all the orphan OST-objects handled. It will record the
dangling references in some new introduced layout LFSCK trace
files on disk during the layout LFSCK first stage scanning,
then travel those traces file after all orphan OST-objects
handled in the second-stage scanning. The side-effect of such
option is that as long as one OST does not join the layout
LFSCK or fail to complete the scanning, then reparing dangling
LOV EA will be skipped. For a large system with a lot of OSTs,
such condition may be a bit strict. The default value is 'off'.

Signed-off-by: Fan Yong <fan.yong@intel.com>
Change-Id: Ic222e1ad20c8011aa5f41cc43171d017ab5c464f
Reviewed-on: https://review.whamcloud.com/21562
Tested-by: Jenkins
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: Lai Siyao <lai.siyao@intel.com>
Reviewed-by: Andreas Dilger <andreas.dilger@intel.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>
3 years agoLU-6142 lnet: change wire protocol typedefs to proper structure 66/24566/3
James Simmons [Sun, 1 Jan 2017 16:12:59 +0000 (11:12 -0500)]
LU-6142 lnet: change wire protocol typedefs to proper structure

The upstream kernel requires proper structures so
convert all the LNet wire protocols typedefs in
the LNet core.

Test-Parameters: trivial

Change-Id: I0287e0d16a1faa7407dc63f9cd607b94ff49d980
Signed-off-by: James Simmons <uja.ornl@yahoo.com>
Reviewed-on: https://review.whamcloud.com/24566
Tested-by: Jenkins
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: Doug Oucharek <doug.s.oucharek@intel.com>
Reviewed-by: Olaf Weber <olaf@sgi.com>
Reviewed-by: Andreas Dilger <andreas.dilger@intel.com>
Reviewed-by: Dmitry Eremin <dmitry.eremin@intel.com>
3 years agoLU-8752 lnet: Stop MLX5 triggering a dump_cqe 06/24306/2
Doug Oucharek [Mon, 12 Dec 2016 17:31:37 +0000 (09:31 -0800)]
LU-8752 lnet: Stop MLX5 triggering a dump_cqe

We have found that MLX5 will trigger a dump_cqe if we don't
invalidate the rkey on a newly alloated MR for FastReg usage.

This fix just tags the MR as invalid on its creation if we are
using FastReg and that will force it to do an invalidate of the
rkey on first usage.

Test-Parameters: trivial
Signed-off-by: Doug Oucharek <doug.s.oucharek@intel.com>
Change-Id: Id0de4f799d70c58011746520b20c8cfca6a29e6a
Reviewed-on: https://review.whamcloud.com/24306
Tested-by: Jenkins
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: Dmitry Eremin <dmitry.eremin@intel.com>
Reviewed-by: Amir Shehata <amir.shehata@intel.com>
Reviewed-by: James Simmons <uja.ornl@yahoo.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>
3 years agoLU-8569 lfsck: handle linkEA overflow 41/23741/10
Fan Yong [Fri, 23 Sep 2016 05:00:47 +0000 (13:00 +0800)]
LU-8569 lfsck: handle linkEA overflow

If the linkEA is marked as overflow (32 bits timestamp in header),
and even if some hard links are removed, we still cannot clear the
overflow timestamp, because the missed hard links entries are still
not in the linkEA. It is the namespace LFSCK's duty to add the missed
entries into the linkEA (if possible) and clear the overflow timestamp
if all the hard links entries are in the linkEA. But it is possible
that there are some new hard links added during the namespace LFSCK
scanning. So the namespace LFSCK needs to compare the linkEA overflow
timestamp with the namespace LFSCK latest reset time (scan the system
from the beginning) in the 2nd-stage scanning. If the latter one is
newer, then means all the hard links entries have been in the linkEA,
so the linkEA overflow timestamp can be cleared.

To avoid the trouble caused by clock drift among MDTs, the linkEA
overflow timestamp will be set as the MDT local time on which the
object resides even if the set linkEA operation is sponsored by
some remote MDT. So we can directly compare the linkEA overflow
time with the namespace LFSCK latest reset time.

The old interfaces between MDD/OUT and LFSCK via lfsck_in_notify()
for the linkEA overflow event are obsoleted and removed.

Signed-off-by: Fan Yong <fan.yong@intel.com>
Change-Id: Ia2c3f9d0a0ecb0b3285041f1365cb4e075a07dda
Reviewed-on: https://review.whamcloud.com/23741
Tested-by: Jenkins
Reviewed-by: Andreas Dilger <andreas.dilger@intel.com>
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: Lai Siyao <lai.siyao@intel.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>
3 years agoLU-7812 tests: address race condition for sanity:120{e,f} 11/24811/6
Jinshan Xiong [Wed, 11 Jan 2017 02:07:19 +0000 (18:07 -0800)]
LU-7812 tests: address race condition for sanity:120{e,f}

By adding a delay between lock cancel and stat, it can avoid the
problem that the locks are still held by ldlm lock thread while
'cancel_lru_locks osc' is called.

Test-Parameters: trivial envdefinitions="ONLY=120e" testlist=sanity,sanity,sanity,sanity,sanity,sanity,sanity,sanity,sanity,sanity
Signed-off-by: Jinshan Xiong <jinshan.xiong@intel.com>
Change-Id: I2095cb1df05ff45c2671edaed5a61bcdd4a56266
Reviewed-on: https://review.whamcloud.com/24811
Tested-by: Jenkins
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: Andreas Dilger <andreas.dilger@intel.com>
Reviewed-by: Jian Yu <jian.yu@intel.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>
3 years agoLU-8966 tests: rename sanity 39, 43, 44, 241 tests 85/24485/2
Elena Gryaznova [Wed, 21 Dec 2016 19:17:44 +0000 (21:17 +0200)]
LU-8966 tests: rename sanity 39, 43, 44, 241 tests

The following tests are renamed to be run separately
from other tests in these groups:
test_39() to test_39a()
test_43() to test_43A()
test_44() to test_44A()
test_241() to test_241a()

Test-Parameters: trivial testlist=sanity
Signed-off-by: Elena Gryaznova <elena.gryaznova@seagate.com>
Change-Id: I7b841da0dc2d307e3174dddccc033c8dd37c27fd
Reviewed-on: https://review.whamcloud.com/24485
Reviewed-by: James Nunez <james.a.nunez@intel.com>
Reviewed-by: Jian Yu <jian.yu@intel.com>
Tested-by: Jenkins
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>
3 years agoLU-8969 llite: sanity test_56x read failed: Invalid argument 52/24552/2
Yang Sheng [Fri, 30 Dec 2016 21:09:31 +0000 (05:09 +0800)]
LU-8969 llite: sanity test_56x read failed: Invalid argument

In generic_file_read_iter() does not check EOF before invoke
direct_IO callback. So we have to check it by ourselves.

Signed-off-by: Yang Sheng <yang.sheng@intel.com>
Change-Id: Ia30d2a42da81beb6c5217f081c18ce4d93553540
Reviewed-on: https://review.whamcloud.com/24552
Reviewed-by: Bob Glossman <bob.glossman@intel.com>
Tested-by: Jenkins
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: Bobi Jam <bobijam@hotmail.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>
3 years agoLU-8973 tests: in sanity 103a use $tdir instead of $MOUNT 18/24518/3
Emoly Liu [Mon, 26 Dec 2016 03:43:42 +0000 (11:43 +0800)]
LU-8973 tests: in sanity 103a use $tdir instead of $MOUNT

In sanity.sh test_103a we should use $tdir instead of $MOUNT
in case of any unclean testing directory.

Test-Parameters: trivial
Signed-off-by: Emoly Liu <emoly.liu@intel.com>
Change-Id: I891bbc0aab28bb58589b66405e0ad5c40fdad25f
Reviewed-on: https://review.whamcloud.com/24518
Tested-by: Jenkins
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: Yang Sheng <yang.sheng@intel.com>
Reviewed-by: Bob Glossman <bob.glossman@intel.com>
Reviewed-by: Jian Yu <jian.yu@intel.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>
3 years agoLU-8957 obdclass: fix exp_nid_hash race 34/24434/5
Yang Sheng [Mon, 19 Dec 2016 19:07:30 +0000 (03:07 +0800)]
LU-8957 obdclass: fix exp_nid_hash race

Move exp_nid_hash unhash operation under exp_lock. To avoid
parallel export disconnect may race in there.

Change-Id: I02a9cf550b4edee7c42ba90d3f3926cd2584c24b
Signed-off-by: Yang Sheng <yang.sheng@intel.com>
Reviewed-on: https://review.whamcloud.com/24434
Tested-by: Jenkins
Reviewed-by: John L. Hammond <john.hammond@intel.com>
Reviewed-by: Niu Yawei <yawei.niu@intel.com>
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>
3 years agoLU-8378 obd: move s3 in lmd_parse to inner loop 20/23820/4
James Simmons [Sat, 17 Dec 2016 17:18:28 +0000 (12:18 -0500)]
LU-8378 obd: move s3 in lmd_parse to inner loop

The upstream client complains with:

obdclass/obd_mount.c: In function lmd_parse:
obdclass/obd_mount.c:880: warning: variable set but not used

The solution is to move s3 to the inner loop
where it is only used.

Test-Parameters: trivial

Change-Id: I95280d08e5f00cf5fdae62acb21e2c08805a9948
Signed-off-by: James Simmons <uja.ornl@yahoo.com>
Reviewed-on: https://review.whamcloud.com/23820
Tested-by: Jenkins
Reviewed-by: Andreas Dilger <andreas.dilger@intel.com>
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: Jinshan Xiong <jinshan.xiong@intel.com>
3 years agoLU-8635 tests: Use wait_update to expire jobstats 99/22699/2
Emoly Liu [Fri, 23 Sep 2016 08:39:18 +0000 (16:39 +0800)]
LU-8635 tests: Use wait_update to expire jobstats

Instead of arbitarily adjusting sleep count, use wait_update()
with maximum of (left+5) seconds to expire old jobstats.

Test-Parameters: trivial
Seagate-bug-id: MRP-2781
Signed-off-by: Ashish Purkar <ashish.purkar@seagate.com>
Signed-off-by: Emoly Liu <emoly.liu@intel.com>
Change-Id: I75e017fc75d35faf1754faddfed168c0fbf432fe
Reviewed-on: https://review.whamcloud.com/22699
Tested-by: Jenkins
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: James Nunez <james.a.nunez@intel.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>
3 years agoLU-8753 llog: remove lgh_write_offset 08/24008/3
Di Wang [Sat, 5 Nov 2016 00:40:07 +0000 (20:40 -0400)]
LU-8753 llog: remove lgh_write_offset

Remove lgh_write_offset for remote llog object,
since it is not easy to be refreshed when remote
transaction fails, which might cause llog file
corruption. Instead let's rely on the osp object
cache to invalidate and update the llog object
size.

Signed-off-by: Di Wang <di.wang@intel.com>
Change-Id: I38aa9ce2c5ac6485bf0cf6be023b8edde986b5d3
Reviewed-on: https://review.whamcloud.com/24008
Tested-by: Jenkins
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: Fan Yong <fan.yong@intel.com>
Reviewed-by: Lai Siyao <lai.siyao@intel.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>
3 years agoLU-6854 llite: Do not set writeback for sync write pages 10/15610/6
Jinshan Xiong [Wed, 15 Jul 2015 21:06:53 +0000 (14:06 -0700)]
LU-6854 llite: Do not set writeback for sync write pages

Sync-write pages don't have Dirty bit set, it will confuse newer
kernels by having writeback bit set.

Signed-off-by: Jinshan Xiong <jinshan.xiong@intel.com>
Signed-off-by: Oleg Drokin <oleg.drokin@intel.com>
Change-Id: I2613919a9e942a7ee3dc48b250f7f0e5e4594e48
Reviewed-on: https://review.whamcloud.com/15610
Tested-by: Jenkins
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: Bob Glossman <bob.glossman@intel.com>
Reviewed-by: Lai Siyao <lai.siyao@intel.com>
Reviewed-by: Yang Sheng <yang.sheng@intel.com>
3 years agoRevert "LU-8922 lod: check master stripes properly" 50/24750/5
James Nunez [Fri, 6 Jan 2017 18:29:55 +0000 (18:29 +0000)]
Revert "LU-8922 lod: check master stripes properly"

It causes frequent testing failures tracked in LU-8985

This reverts commit 1be03a1088d2aa583c2651435634ba336b48f939.

Test-Parameters: testlist=conf-sanity,conf-sanity,conf-sanity,conf-sanity,conf-sanity,conf-sanity

Change-Id: I444ce9e65c72db7db3f53692d9fe4f823943fade
Signed-off-by: James Nunez <james.a.nunez@intel.com>
Reviewed-on: https://review.whamcloud.com/24750
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>
3 years agoLU-4423 lnet: use 64-bit time for selftest 52/23352/2
Arnd Bergmann [Tue, 25 Oct 2016 02:00:56 +0000 (22:00 -0400)]
LU-4423 lnet: use 64-bit time for selftest

The lustre selftest code has multiple time stamps that are kept
as 'time_t' or 'unsigned long' and can therefore overflow on
32-bit systems.

This changes the code to use time64_t instead.

Linux-commit: d9f79e6bf02c109f117132163239bfffa6475ccb

Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Change-Id: I74d704d104e7398af411e73180fac57c0b943eb7
Signed-off-by: Oleg Drokin <green@linuxhacker.ru>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Signed-off-by: James Simmons <uja.ornl@yahoo.com>
Reviewed-on: https://review.whamcloud.com/23352
Tested-by: Jenkins
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: Frank Zago <fzago@cray.com>
Reviewed-by: Dmitry Eremin <dmitry.eremin@intel.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>
3 years agoLU-8965 kernel: kernel update [SLES11 SP4 3.0.101-91] 77/24477/2
Bob Glossman [Wed, 21 Dec 2016 17:06:31 +0000 (09:06 -0800)]
LU-8965 kernel: kernel update [SLES11 SP4 3.0.101-91]

Update SLES11 SP4 kernel to 3.0.101-91

Test-Parameters: mdsdistro=sles11sp4 ossdistro=sles11sp4 \
  clientdistro=sles11sp4 mdsfilesystemtype=ldiskfs \
  mdtfilesystemtype=ldiskfs ostfilesystemtype=ldiskfs \
  testgroup=review-ldiskfs

Signed-off-by: Bob Glossman <bob.glossman@intel.com>
Change-Id: I8b7aa39f60626a511b58d1a7a9a31189c17b4f1a
Reviewed-on: https://review.whamcloud.com/24477
Tested-by: Jenkins
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: Trent Geerdes <trent.geerdes@intel.com>
Reviewed-by: James Nunez <james.a.nunez@intel.com>
Reviewed-by: Dmitry Eremin <dmitry.eremin@intel.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>
3 years agoLU-6401 header: remove libcfs.h in interval_tree.h 23/24323/3
James Simmons [Tue, 13 Dec 2016 18:28:31 +0000 (13:28 -0500)]
LU-6401 header: remove libcfs.h in interval_tree.h

The header libcfs.h for interval_tree.h is not needed
so we can just use the standard linux kernel headers
instead.h

Test-Parameters: trivial

Change-Id: I9c2c4f7e9bc227876818ff7213f392bdeb1112e0
Signed-off-by: James Simmons <uja.ornl@yahoo.com>
Reviewed-on: https://review.whamcloud.com/24323
Tested-by: Jenkins
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: Frank Zago <fzago@cray.com>
Reviewed-by: John L. Hammond <john.hammond@intel.com>
Reviewed-by: Dmitry Eremin <dmitry.eremin@intel.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>
3 years agoLU-8889 llite: replace memory allocation with ll_env usage 64/24064/2
Dmitry Eremin [Wed, 9 Nov 2016 12:52:30 +0000 (15:52 +0300)]
LU-8889 llite: replace memory allocation with ll_env usage

Replace memory allocation with ll_env usage in ll_file_read
like in other functions.

Change-Id: I44445ab80cc856f45271e0888ea2538d2a44cdeb
Signed-off-by: Dmitry Eremin <dmitry.eremin@intel.com>
Reviewed-on: https://review.whamcloud.com/24064
Tested-by: Jenkins
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: Bob Glossman <bob.glossman@intel.com>
Reviewed-by: Andreas Dilger <andreas.dilger@intel.com>
Reviewed-by: John L. Hammond <john.hammond@intel.com>
3 years agoLU-6210 mgs: Change positional struct initializers to C99 94/23794/2
Steve Guminski [Wed, 16 Nov 2016 15:30:45 +0000 (10:30 -0500)]
LU-6210 mgs: Change positional struct initializers to C99

This patch makes no functional changes.  Struct initializers in the
mgs directory that use C89 or GCC-only syntax are updated to C99
syntax.  Whitespace is corrected to match the coding style guidelines.

The C99 syntax prevents incorrect initialization if values are
accidently placed in the wrong position, allows changes in the struct
definition, and clears any members that are not given an explicit
value.

The following struct initializers have been updated:

lustre/mgs/mgs_handler.c:
struct lustre_handle lockh
struct lustre_handle conn
lustre/mgs/mgs_llog.c:
struct mgs_fsdb_handler_data d

Test-Parameters: trivial
Signed-off-by: Steve Guminski <stephenx.guminski@intel.com>
Change-Id: I36cbeaae50982b6f72d271997b229c46d085a6ee
Reviewed-on: https://review.whamcloud.com/23794
Tested-by: Jenkins
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: James Simmons <uja.ornl@yahoo.com>
Reviewed-by: Nathaniel Clark <nathaniel.l.clark@intel.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>
3 years agoLU-6210 fid: Change positional struct initializers to C99 89/23789/2
Steve Guminski [Wed, 16 Nov 2016 14:24:34 +0000 (09:24 -0500)]
LU-6210 fid: Change positional struct initializers to C99

This patch makes no functional changes.  Struct initializers in the
fid directory that use C89 or GCC-only syntax are updated to C99
syntax.

The C99 syntax prevents incorrect initialization if values are
accidently placed in the wrong position, allows changes in the struct
definition, and clears any members that are not given an explicit
value.

The following struct initializers have been updated:

lustre/fid/fid_lib.c:
const struct lu_seq_range LUSTRE_SEQ_SPACE_RANGE
const struct lu_seq_range LUSTRE_SEQ_ZERO_RANGE
lustre/fid/lproc_fid.c:
struct lu_seq_range tmp

Test-Parameters: trivial
Signed-off-by: Steve Guminski <stephenx.guminski@intel.com>
Change-Id: I73b6a37a0fcac29ba48410e7befb98f13bdb17cc
Reviewed-on: https://review.whamcloud.com/23789
Tested-by: Jenkins
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: Nathaniel Clark <nathaniel.l.clark@intel.com>
Reviewed-by: James Simmons <uja.ornl@yahoo.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>
3 years agoLU-6210 osd: Change positional struct initializers to C99 70/23770/2
Steve Guminski [Mon, 14 Nov 2016 20:42:44 +0000 (15:42 -0500)]
LU-6210 osd: Change positional struct initializers to C99

This patch makes no functional changes.  Struct initializers in the
osd directories that use C89 or GCC-only syntax are updated to C99
syntax.

The C99 syntax prevents incorrect initialization if values are
accidently placed in the wrong position, allows changes in the struct
definition, and clears any members that are not given an explicit
value.

The following struct initializers have been updated:

lustre/osd-ldiskfs/osd_scrub.c:
static const struct osd_lf_map osd_lf_maps[]
static const struct osd_lf_map osd_dl_maps[]
lustre/osd-zfs/osd_oi.c:
static const struct named_oid oids[]

Test-Parameters: trivial
Signed-off-by: Steve Guminski <stephenx.guminski@intel.com>
Change-Id: I1aa2bf64f0efcbd9c34f21f60c462f7a95c41f98
Reviewed-on: https://review.whamcloud.com/23770
Tested-by: Jenkins
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: James Simmons <uja.ornl@yahoo.com>
Reviewed-by: Nathaniel Clark <nathaniel.l.clark@intel.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>
3 years agoLU-6210 ldlm: Change positional struct initializers to C99 43/23743/2
Steve Guminski [Mon, 14 Nov 2016 14:02:13 +0000 (09:02 -0500)]
LU-6210 ldlm: Change positional struct initializers to C99

This patch makes no functional changes.  Struct initializers in the
ldlm directory that use C89 or GCC-only syntax are updated to C99
syntax.

The C99 syntax prevents incorrect initialization if values are
accidently placed in the wrong position, allows changes in the struct
definition, and clears any members that are not given an explicit
value.

The following struct initializers have been updated:

lustre/ldlm/ldlm_extent.c:
struct interval_node_extent limiter
struct interval_node_extent ext

Test-Parameters: trivial
Signed-off-by: Steve Guminski <stephenx.guminski@intel.com>
Change-Id: I219f5b2bd4b96d92eaf120f7080e762d72eb3105
Reviewed-on: https://review.whamcloud.com/23743
Tested-by: Jenkins
Reviewed-by: Andreas Dilger <andreas.dilger@intel.com>
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: James Simmons <uja.ornl@yahoo.com>
Reviewed-by: Dmitry Eremin <dmitry.eremin@intel.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>
3 years agoLU-6210 mdt: Change positional struct initializers to C99 02/23702/2
Steve Guminski [Thu, 10 Nov 2016 18:21:23 +0000 (13:21 -0500)]
LU-6210 mdt: Change positional struct initializers to C99

This patch makes no functional changes.  Struct initializers in the
mdt directory that use C89 or GCC-only syntax are updated to C99
syntax.

The C99 syntax prevents incorrect initialization if values are
accidently placed in the wrong position, allows changes in the struct
definition, and clears any members that are not given an explicit
value.

The following struct initializers have been updated:

lustre/mdt/mdt_lvb.c:
struct ldlm_valblock_ops mdt_lvbo
lustre/mdt/mdt_mds.c:
static struct lprocfs_vars lprocfs_mds_obd_vars[]

Test-Parameters: trivial
Signed-off-by: Steve Guminski <stephenx.guminski@intel.com>
Change-Id: If8b3b6233e34f6e8406ab6b8e4e70845bf94bfc8
Reviewed-on: https://review.whamcloud.com/23702
Tested-by: Jenkins
Reviewed-by: Andreas Dilger <andreas.dilger@intel.com>
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: James Simmons <uja.ornl@yahoo.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>
3 years agoLU-6210 obd: Change positional struct initializers to C99 97/23697/3
Steve Guminski [Mon, 7 Nov 2016 18:52:20 +0000 (13:52 -0500)]
LU-6210 obd: Change positional struct initializers to C99

This patch makes no functional changes.  Struct initializers in the
obd directory that use C89 or GCC-only syntax are updated to C99
syntax.  Whitespace is changed to match the code style guidelines.

The C99 syntax prevents incorrect initialization if values are
accidently placed in the wrong position, allows changes in the struct
definition, and clears any members that are not given an explicit
value.

The following struct initializers have been updated:

lustre/obdclass/lprocfs_jobstats.c:
static const struct seq_operations lprocfs_jobstats_seq_sops
lustre/obdclass/obd_config.c:
struct llog_process_cat_data
lustre/obdclass/obd_mount.c:
struct lustre_mount_data2 lmd2 (2 occurrences)

Test-Parameters: trivial
Signed-off-by: Steve Guminski <stephenx.guminski@intel.com>
Change-Id: I74a7d6ebb250e8df0f24f60ce87933fddf325c30
Reviewed-on: https://review.whamcloud.com/23697
Tested-by: Jenkins
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: James Simmons <uja.ornl@yahoo.com>
Reviewed-by: John L. Hammond <john.hammond@intel.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>
3 years agoLU-6210 tests: Change positional struct initializers to C99 68/23568/3
Steve Guminski [Thu, 3 Nov 2016 15:44:04 +0000 (11:44 -0400)]
LU-6210 tests: Change positional struct initializers to C99

This patch makes no functional changes.  Struct initializers in the
tests directory that use C89 or GCC-only syntax are updated to C99
syntax.

The C99 syntax prevents incorrect initialization if values are
accidently placed in the wrong position, allows changes in the struct
definition, and clears any members that are not given an explicit
value.

The following struct initializers have been updated:

lustre/tests/checkfiemap.c:
struct option long_opts[]
lustre/tests/llapi_layout_test.c:
static struct test_tbl_entry test_tbl[]
lustre/tests/mcreate.c:
struct option opts[]
lustre/tests/mmap_sanity.c:
struct test_case tests[]
lustre/tests/mpi/mdsrate.c:
struct option longOpts[]
lustre/tests/statmany.c:
struct option longopts[]

Test-Parameters: trivial
Signed-off-by: Steve Guminski <stephenx.guminski@intel.com>
Change-Id: Iadc141507664c97dcc3112c382d4e379a1bf8041
Reviewed-on: https://review.whamcloud.com/23568
Tested-by: Jenkins
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: Frank Zago <fzago@cray.com>
Reviewed-by: James Simmons <uja.ornl@yahoo.com>
Reviewed-by: Dmitry Eremin <dmitry.eremin@intel.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>
3 years agoLU-6210 snmp: Change positional struct initializers to C99 63/23463/2
Steve Guminski [Thu, 27 Oct 2016 19:18:43 +0000 (15:18 -0400)]
LU-6210 snmp: Change positional struct initializers to C99

This patch makes no functional changes.  Struct initializers in the
snmp directory that use C89 or GCC-only syntax are updated to C99
syntax.

The C99 syntax prevents incorrect initialization if values are
accidently placed in the wrong position, allows changes in the struct
definition, and clears any members that are not given an explicit
value.

The following struct initializers have been updated:

snmp/lustre-snmp.c:
struct variable7 clusterFileSystems_variables[]
struct oid_table osd_table[]
struct oid_table osc_table[]
struct oid_table mds_table[]
struct oid_table mdc_table[]
struct oid_table cli_table[]
struct oid_table lov_table[]
struct oid_table ldlm_table[]

Test-Parameters: trivial
Signed-off-by: Steve Guminski <stephenx.guminski@intel.com>
Change-Id: Iaedd0165bca2a2af2ce6a0f535602a85c6217da8
Reviewed-on: https://review.whamcloud.com/23463
Tested-by: Jenkins
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: Frank Zago <fzago@cray.com>
Reviewed-by: Dmitry Eremin <dmitry.eremin@intel.com>
Reviewed-by: James Simmons <uja.ornl@yahoo.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>
3 years agoLU-8066 obd: Move /proc/fs/lustre root level files to sysfs 25/23425/2
Oleg Drokin [Thu, 27 Oct 2016 14:59:50 +0000 (10:59 -0400)]
LU-8066 obd: Move /proc/fs/lustre root level files to sysfs

except devices, for now.

Linux-commit : 8b82844505699f10f80c7e554364e3d0618cd77d

Signed-off-by: Oleg Drokin <green@linuxhacker.ru>
Change-Id: Iacdba1e0f703eb90688c1b93a2ef81c0c8ad7a91
Signed-off-by: Dmitry Eremin <dmitry.eremin@intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Signed-off-by: James Simmons <uja.ornl@yahoo.com>
Reviewed-on: https://review.whamcloud.com/23425
Tested-by: Jenkins
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: Emoly Liu <emoly.liu@intel.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>
3 years agoLU-8066 obd: Generic helpers for sysfs 09/22209/3
Oleg Drokin [Thu, 27 Oct 2016 14:43:49 +0000 (10:43 -0400)]
LU-8066 obd: Generic helpers for sysfs

Add generic helpers to allow displaying of lustre-specific values
in /sys/fs/lustre

except devices, for now.

Linux-commit : 5e66f70e963e4ba2266071343f5a153c7db4327b

Change-Id: Idfbf419d8218546463b993a8031ef00609f8dce9
Signed-off-by: Oleg Drokin <green@linuxhacker.ru>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Signed-off-by: James Simmons <uja.ornl@yahoo.com>
Reviewed-on: https://review.whamcloud.com/22209
Tested-by: Jenkins
Reviewed-by: Dmitry Eremin <dmitry.eremin@intel.com>
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: Niu Yawei <yawei.niu@intel.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>
3 years agoLU-6401 headers: move lu_fid, ost_id funcs out of lustre_idl.h 12/22712/10
Ben Evans [Mon, 24 Oct 2016 16:48:26 +0000 (12:48 -0400)]
LU-6401 headers: move lu_fid, ost_id funcs out of lustre_idl.h

Move lu_fid functions into lustre/lustre_fid.h
Move ost_id functions into lustre/lustre_ostid.h
Fix indenting, include new headers as needed

Signed-off-by: Ben Evans <bevans@cray.com>
Change-Id: Idcbb95f614829b434c48a476172eedb421fe488f
Reviewed-on: https://review.whamcloud.com/22712
Tested-by: Jenkins
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: James Simmons <uja.ornl@yahoo.com>
Reviewed-by: Frank Zago <fzago@cray.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>
3 years agoLU-8765 ptlrpc: update replay cursor when close during replay 18/23418/4
Niu Yawei [Thu, 27 Oct 2016 06:38:58 +0000 (02:38 -0400)]
LU-8765 ptlrpc: update replay cursor when close during replay

The replay cursor should be updated properly when close happened
during replay, otherwise, ptlrpc_replay_next() could run into a
dead loop due to an invalid replay cursor:

- replay cursor is moved to an open request during replay;
- application close that open file, so the rq_replay of the open
  request is cleared;
- ptlrpc_replay_next() calls ptlrpc_free_committed() to free
  committed/closed requests, the open request is removed from
  the committed list, so the replay cursor is changed to an
  empty list_head now. The open request won't be freed now since
  it's still held by the pending close request;
- ptlrpc_replay_next() continue to move the replay cursor to
  next and run into a dead loop at the end;

Another change in this patch is to remove the out of date comments
in ptlrpc_replay_next() and cover the whole process of finding
replay request within imp_lock, because:

1. With two separated replay lists and replay cursor introduced,
   finding replay request won't take much time as before, it's
   not necessary to do this "lock -> unlock -> lock -> unlock"
   trick anymore;

2. Nowadays there are various kind of non-replay requests are
   allowed during recovery, so ptlrpc_free_committed() may run in
   parallel to remove an open request while ptlrpc_replay_next()
   is iterating the open requests list;

Signed-off-by: Niu Yawei <yawei.niu@intel.com>
Change-Id: I358c4f55dbf38b0c9122c3aff40cf9ec96619d6a
Reviewed-on: https://review.whamcloud.com/23418
Tested-by: Jenkins
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: Yang Sheng <yang.sheng@intel.com>
Reviewed-by: John L. Hammond <john.hammond@intel.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>
3 years agoLU-4423 libcfs: use 64-bit times for cfs_srand seed 51/23351/2
Arnd Bergmann [Tue, 25 Oct 2016 01:58:26 +0000 (21:58 -0400)]
LU-4423 libcfs: use 64-bit times for cfs_srand seed

Several functions in Lustre call cfs_srand with do_gettimeofday
as the seed to get a pseudo-random number.
There is no bug here, but changing it to use ktime_get_ts64()
gets us closer to deprecating do_gettimeofday() and makes it slightly
more random.

Affected functions are:
lnet_shuffle_seed, init_lustre_lite and class_handle_init

Linux-commit: 1f4fc343c008981d3a91351c030e2c29f5cd1b1c

Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Change-Id: I86d156afd71456d7d5412ebb8a2e519367d4f81d
Signed-off-by: Oleg Drokin <green@linuxhacker.ru>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Signed-off-by: James Simmons <uja.ornl@yahoo.com>
Reviewed-on: https://review.whamcloud.com/23351
Tested-by: Jenkins
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: Frank Zago <fzago@cray.com>
Reviewed-by: Dmitry Eremin <dmitry.eremin@intel.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>
3 years agoLU-4423 ptlrpc: use 64-bit time for adaptive timeout 70/23270/2
Arnd Bergmann [Thu, 20 Oct 2016 01:24:02 +0000 (21:24 -0400)]
LU-4423 ptlrpc: use 64-bit time for adaptive timeout

The adaptive timeout handling stores absolute times in 32-bit time_t
quantities, which will overflow in 2038.

This changes it to use time64_t.

Linux-commit: 0ac0478b65de1e5b8b62f608bdf1d8ba631acc4d

Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Change-Id: I50d1090865af8970276d039281fc2ca0b87fa8af
Signed-off-by: Oleg Drokin <green@linuxhacker.ru>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Signed-off-by: James Simmons <uja.ornl@yahoo.com>
Reviewed-on: https://review.whamcloud.com/23270
Tested-by: Jenkins
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: Andreas Dilger <andreas.dilger@intel.com>
Reviewed-by: Dmitry Eremin <dmitry.eremin@intel.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>
3 years agoLU-4423 ldlm: use time64_t for l_last_activity 41/23141/2
Arnd Bergmann [Thu, 13 Oct 2016 18:50:00 +0000 (14:50 -0400)]
LU-4423 ldlm: use time64_t for l_last_activity

The l_last_activity struct member is used to keep track lock hold
times, and it is printed for debugging purposes. For the elapsed
time, we can use 'long' here, but it's better to use time64_t
for storing the real time to avoid an overflow in 2038.

Update server side code as well that is not present in upstream
client.

Linux-commit: bf6d21539d58f64812d71c956ecec3d67650b2ee

Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Change-Id: Idbbc64afbb9307e6133c671a7aa2122318c62679
Signed-off-by: Oleg Drokin <green@linuxhacker.ru>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Signed-off-by: James Simmons <uja.ornl@yahoo.com>
Reviewed-on: https://review.whamcloud.com/23141
Reviewed-by: Frank Zago <fzago@cray.com>
Tested-by: Jenkins
Reviewed-by: Dmitry Eremin <dmitry.eremin@intel.com>
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: Andreas Dilger <andreas.dilger@intel.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>
3 years agoLU-4423 libcfs: tracefile document seconds overflow 37/23137/2
Arnd Bergmann [Thu, 13 Oct 2016 18:04:39 +0000 (14:04 -0400)]
LU-4423 libcfs: tracefile document seconds overflow

The lustre tracefile has a timestamp defined as

__u32 ph_sec;
__u64 ph_usec;

which seems completely backwards, as the microsecond portion of
a time stamp will always fit into a __u32 value, while the second
portion will overflow in 2038 or 2106 (in case of unsigned seconds).

Changing this would unfortunately change the format in an incompatible
way, breaking all existing user space tools that access the data.

This uses ktime_get_real_ts64() to replace the insufficient
do_gettimeofday() and then truncates the seconds portion to
an u32 type, along with comments to explain the result.

A possible alternative would be the use of ktime_get_ts64() to
read a monotonic timestamp that never overflows, but this would
trigger a check in user space 'hdr->ph_sec < (1 << 30)' that
attempts to ensure that the values are within a reasonable range.

Linux-commit: 65f28840e4c653b7d78feabcb65280c09d3d0d23

Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Change-Id: I5fdf6d09d39fbb4da8ccba51a745b00996f116f7
Signed-off-by: Oleg Drokin <green@linuxhacker.ru>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Signed-off-by: James Simmons <uja.ornl@yahoo.com>
Reviewed-on: https://review.whamcloud.com/23137
Reviewed-by: Andreas Dilger <andreas.dilger@intel.com>
Tested-by: Jenkins
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: Frank Zago <fzago@cray.com>
Reviewed-by: Dmitry Eremin <dmitry.eremin@intel.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>
3 years agoLU-4423 libcfs: use 64-bit inode timestamps internally 36/23136/2
Arnd Bergmann [Thu, 13 Oct 2016 17:03:03 +0000 (13:03 -0400)]
LU-4423 libcfs: use 64-bit inode timestamps internally

Lustre has 64-bit timestamps in its network data structures, but
on 32 bit systems, it converts them directly into time_t, which is
32 bit wide.

This changes the code to use 64-bit time stamps for files. The Linux
VFS code still uses time_t though, and will be changed in a separate
patch series.

Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Change-Id: I6d3e986fa4eac2fe76a6d66a55f2cdbfe8fdae55
Signed-off-by: Oleg Drokin <green@linuxhacker.ru>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Signed-off-by: James Simmons <uja.ornl@yahoo.com>
Reviewed-on: https://review.whamcloud.com/23136
Tested-by: Jenkins
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: Jinshan Xiong <jinshan.xiong@intel.com>
Reviewed-by: Dmitry Eremin <dmitry.eremin@intel.com>
Reviewed-by: Andreas Dilger <andreas.dilger@intel.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>
3 years agoLU-6245 libcfs: add ktime_get_real_seconds support 66/22866/10
James Simmons [Tue, 25 Oct 2016 01:43:12 +0000 (21:43 -0400)]
LU-6245 libcfs: add ktime_get_real_seconds support

In newer kernels time support is moving to u64 to avoid
the 2038 limitation. The function get_seconds() which
only returns a 32 bit value is being replaced by a new
function ktime_get_real_seconds(). This patch adds a
wrapper for older platforms that lack this function.
Only the libcfs/LNet layer has been ported in this
patch since it is a huge change.

Signed-off-by: James Simmons <uja.ornl@yahoo.com>
Change-Id: Ic16859f1b052bfb72299cc85fe50364a9e70d660
Reviewed-on: https://review.whamcloud.com/22866
Tested-by: Jenkins
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: Frank Zago <fzago@cray.com>
Reviewed-by: Andreas Dilger <andreas.dilger@intel.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>
3 years agoLU-8314 lfs: improve getdirstripe interface 19/24319/4
Lai Siyao [Tue, 13 Dec 2016 15:57:40 +0000 (23:57 +0800)]
LU-8314 lfs: improve getdirstripe interface

* remove unused options '-v' '-R' '-q' for "lfs getdirstripe".
* add "[--mdt-hash|-t]" option for "lfs getdirstripe", and print
  hash_type by default.

Signed-off-by: Lai Siyao <lai.siyao@intel.com>
Change-Id: Ib7eaf00744d0cd5dec13e683c39e2ecf1d54808b
Reviewed-on: https://review.whamcloud.com/24319
Tested-by: Jenkins
Reviewed-by: Andreas Dilger <andreas.dilger@intel.com>
Reviewed-by: Niu Yawei <yawei.niu@intel.com>
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>
3 years agoLU-8922 lod: check master stripes properly 47/24247/2
Di Wang [Thu, 8 Dec 2016 09:55:26 +0000 (04:55 -0500)]
LU-8922 lod: check master stripes properly

When creating striped directory, it should
first check if the stripe has been created
on the current MDT, otherwise it might create
duplicate stripes on the master MDT, especially
when one MDT is deactived, and specified stripes
is larger than the active MDTs.

Signed-off-by: Di Wang <di.wang@intel.com>
Change-Id: I089f5cf517c71816151e8ece1b36b9064ee3b7f2
Reviewed-on: https://review.whamcloud.com/24247
Tested-by: Jenkins
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: Andreas Dilger <andreas.dilger@intel.com>
Reviewed-by: Fan Yong <fan.yong@intel.com>
Reviewed-by: Lai Siyao <lai.siyao@intel.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>
3 years agoLU-8569 linkea: linkEA size limitation 00/23500/9
Fan Yong [Sun, 28 Aug 2016 10:15:37 +0000 (18:15 +0800)]
LU-8569 linkea: linkEA size limitation

Under DNE mode, if we do not restrict the linkEA size, and if there
are too many cross-MDTs hard links to the same object, then it will
casue the llog overflow. On the other hand, too many linkEA entries
in the linkEA will serious affect the linkEA performance because we
only support to locate linkEA entry consecutively.

So we need to restrict the linkEA size. Currently, it is 4096 bytes,
that is independent from the backend. If too many hard links caused
the linkEA overflowed, we will add overflow timestamp in the linkEA
header. Such overflow timestamp has some functionalities:

1. It will prevent the object being migrated to other MDT, because
   some name entries may be not in the linkEA, so we cannot update
   these name entries for the migration.

2. It will tell the namespace LFSCK that the 'nlink' attribute may
   be more trustable than the linkEA, then avoid misguiding the
   namespace LFSCK to repair 'nlink' attribute based on linkEA.

There will be subsequent patch(es) for namespace LFSCK to handle the
linkEA size limitation and overflow cases.

Signed-off-by: Fan Yong <fan.yong@intel.com>
Change-Id: I2d6c2be04305c1d7e3af160d8b80e73b66a36483
Reviewed-on: https://review.whamcloud.com/23500
Tested-by: Jenkins
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: Andreas Dilger <andreas.dilger@intel.com>
Reviewed-by: wangdi <di.wang@intel.com>
Reviewed-by: Lai Siyao <lai.siyao@intel.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>
3 years agoNew tag 2.9.51 2.9.51 v2_9_51 v2_9_51_0
Oleg Drokin [Tue, 27 Dec 2016 23:06:27 +0000 (18:06 -0500)]
New tag 2.9.51

Change-Id: Ie0e5df746be2336ad77da4ebe50faf7cf8337813
Signed-off-by: Oleg Drokin <oleg.drokin@intel.com>
3 years agoLU-8972 tests: Disable conf-sanity test 101 17/24517/2
Oleg Drokin [Sun, 25 Dec 2016 21:09:07 +0000 (16:09 -0500)]
LU-8972 tests: Disable conf-sanity test 101

It causes a lot of failures for apparently no good reason.

Change-Id: I0e29c3274b18eece1249e79bed2b58edc493836f
Signed-off-by: Oleg Drokin <oleg.drokin@intel.com>
Reviewed-on: https://review.whamcloud.com/24517
Tested-by: Jenkins
Tested-by: Maloo <hpdd-maloo@intel.com>
3 years agoLU-8925 kernel: kernel update RHEL7.3 [3.10.0-514.2.2.el7] 72/24372/3
Bob Glossman [Mon, 12 Dec 2016 19:20:23 +0000 (11:20 -0800)]
LU-8925 kernel: kernel update RHEL7.3 [3.10.0-514.2.2.el7]

update RHEL 7.3 kernel to 3.10.0-514.2.2.el7

Signed-off-by: Bob Glossman <bob.glossman@intel.com>
Change-Id: I684e25eede267c96d41c37b1c3295301c38b5573
Reviewed-on: https://review.whamcloud.com/24372
Tested-by: Jenkins
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: Yang Sheng <yang.sheng@intel.com>
Reviewed-by: Minh Diep <minh.diep@intel.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>
3 years agoLU-3425 utils: rename l_getgroups.8 to l_getidentity.8 19/24219/4
Andreas Dilger [Wed, 7 Dec 2016 20:56:40 +0000 (13:56 -0700)]
LU-3425 utils: rename l_getgroups.8 to l_getidentity.8

Rename the man page for l_getgroups.8 to l_getidentity.8 to match
the "new" name of the utility, and update the content accordingly.

Test-Parameters: trivial
Signed-off-by: Andreas Dilger <andreas.dilger@intel.com>
Change-Id: I698c8170cb0623ff692bfe76962b5c71d75bc18b
Reviewed-on: https://review.whamcloud.com/24219
Tested-by: Jenkins
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: Steve Guminski <stephenx.guminski@intel.com>
Reviewed-by: Bob Glossman <bob.glossman@intel.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>
3 years agoLU-8890 ptlrpc: fix module_param in ptlrpcd.c 65/24065/3
Dmitry Eremin [Fri, 25 Nov 2016 18:00:14 +0000 (21:00 +0300)]
LU-8890 ptlrpc: fix module_param in ptlrpcd.c

Missing module parameter description for ptlrpcd_per_cpt_max

Change-Id: Id00d6bd73cf2779e0f12db5184292ebdb0cfa8ae
Signed-off-by: Dmitry Eremin <dmitry.eremin@intel.com>
Reviewed-on: https://review.whamcloud.com/24065
Reviewed-by: John L. Hammond <john.hammond@intel.com>
Tested-by: Jenkins
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: Andreas Dilger <andreas.dilger@intel.com>
3 years agoLU-8887 ptlrpc: remove unused pc->pc_env 61/24061/3
Dmitry Eremin [Thu, 1 Dec 2016 09:08:59 +0000 (12:08 +0300)]
LU-8887 ptlrpc: remove unused pc->pc_env

Environment for request interpreters is not used any more.

Change-Id: Id218fa251bb022f2fff11c79adeb83da65fe3088
Signed-off-by: Dmitry Eremin <dmitry.eremin@intel.com>
Reviewed-on: https://review.whamcloud.com/24061
Tested-by: Jenkins
Reviewed-by: John L. Hammond <john.hammond@intel.com>
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: Bob Glossman <bob.glossman@intel.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>
3 years agoLU-8886 lfsck: handle -ENODATA for the end of iteration 56/24056/2
Fan Yong [Sun, 28 Aug 2016 09:41:14 +0000 (17:41 +0800)]
LU-8886 lfsck: handle -ENODATA for the end of iteration

When the LFSCK engine makes checkpoint, it will record the current
position (hash) of the iteration that can be used to resume the
LFSCK scanning from the latest checkpoint. But if someone removed
the target corresponding to the LFSCK checkpoint position, then
the LFSCK will start scanning from the 'next' position when try to
resume from the latest checkpoint. But if the 'next' hits the end
of the iteration, the lower layer (ldiskfs) will return "-ENODATA"
for dt_it_ops::load() call. The LFSCK engine needs to handle such
case properly to avoid reguarding it as failure.

Signed-off-by: Fan Yong <fan.yong@intel.com>
Change-Id: I5a6a46cf3eefbdc680780abcc3402860736d6e0a
Reviewed-on: https://review.whamcloud.com/24056
Tested-by: Jenkins
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: Lai Siyao <lai.siyao@intel.com>
Reviewed-by: Andreas Dilger <andreas.dilger@intel.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>
3 years agoLU-8869 build: Fix building of DEB packages 99/23999/6
Thomas Stibor [Tue, 29 Nov 2016 14:13:34 +0000 (15:13 +0100)]
LU-8869 build: Fix building of DEB packages

Building DEB packages failed with errors:
dh_install: lustre-dev missing files (debian/tmp/usr/lib/*.so.*)
and dpkg-checkbuilddeps: Unmet build dependencies.
Moreover, signing the DEB packages failed due signing problems
of the author listed in debian/changelog.
Additionally, latest DEB based distributions ship Lustre modules
from kernel staging which can conflict with modules build from
upstream. To fix the problem, build modules with
configure --with-kmp-moddir=updates and add search order
of "search updates built-in" in /etc/depmod.d/lustre.conf

Signed-off-by: Thomas Stibor <t.stibor@gsi.de>
Change-Id: I02c58103bfaa83b97f5cd207af823fb477e0f0d3
Reviewed-on: https://review.whamcloud.com/23999
Tested-by: Jenkins
Reviewed-by: Dmitry Eremin <dmitry.eremin@intel.com>
Reviewed-by: Bob Glossman <bob.glossman@intel.com>
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>
3 years agoLU-8724 utils: Use open() instead of opendir() with dirfd() 81/23981/6
Steve Guminski [Mon, 5 Dec 2016 19:39:16 +0000 (14:39 -0500)]
LU-8724 utils: Use open() instead of opendir() with dirfd()

Several llapi functions use opendir() and dirfd() to obtain a file
descriptor.  These calls are replaced with a single call to open().
This makes the code cleaner and more efficient.

This patch also corrects whitespace to match the Coding Style
Guidelines.

Test-Parameters: trivial
Signed-off-by: Steve Guminski <stephenx.guminski@intel.com>
Change-Id: I798165960fb500920b82adabe389975135ed3ac9
Reviewed-on: https://review.whamcloud.com/23981
Tested-by: Jenkins
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: John L. Hammond <john.hammond@intel.com>
Reviewed-by: Nathaniel Clark <nathaniel.l.clark@intel.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>
3 years agoLU-8776 build: Fix weird inline definitions 79/23979/2
Steve Guminski [Tue, 22 Nov 2016 18:43:52 +0000 (13:43 -0500)]
LU-8776 build: Fix weird inline definitions

This patch removes unnecessary inline keywords from functions
that do not have to be inlined.  It also corrects the ordering
of the keywords in several functions from "inline static" to
"static inline".

Test-Parameters: trivial
Signed-off-by: Steve Guminski <stephenx.guminski@intel.com>
Change-Id: I13dde114a803513b72d6497f42be460c4df06883
Reviewed-on: https://review.whamcloud.com/23979
Tested-by: Jenkins
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: John L. Hammond <john.hammond@intel.com>
Reviewed-by: Nathaniel Clark <nathaniel.l.clark@intel.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>
3 years agoLU-7952 osp: Move strlen() call below LASSERT() 19/20919/5
Arshad Hussain [Tue, 21 Jun 2016 03:21:29 +0000 (08:51 +0530)]
LU-7952 osp: Move strlen() call below LASSERT()

This patch moves the strlen() call below LASSERT(). This
avoids a case when parameter to strlen() is NULL and
dereferencing the NULL pointer would cause fault before
it reaches LASSERT().

Signed-off-by: Arshad Hussain <arshad.hussain@seagate.com>
Change-Id: Ie7a683751cf44d7dfad6de1a9024c5fb7a35c35d
Reviewed-on: https://review.whamcloud.com/20919
Tested-by: Jenkins
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: Andreas Dilger <andreas.dilger@intel.com>
Reviewed-by: Dmitry Eremin <dmitry.eremin@intel.com>
3 years agoLU-7988 hsm: don't use a ptlrpc_thread struct for the cdt 61/22761/8
Frank Zago [Wed, 28 Sep 2016 15:08:07 +0000 (11:08 -0400)]
LU-7988 hsm: don't use a ptlrpc_thread struct for the cdt

Since only the waitq and the event flag from the ptlrpc_thread
structure are used, declare these directly and do not use the
ptlrpc_thread structure at all.

Test-Parameters: trivial testlist=sanity-hsm
Signed-off-by: frank zago <fzago@cray.com>
Change-Id: Iba63ac70ad381f7e40976ac423594eae46c5249b
Reviewed-on: https://review.whamcloud.com/22761
Tested-by: Jenkins
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: Henri Doreau <henri.doreau@cea.fr>
Reviewed-by: John L. Hammond <john.hammond@intel.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>
3 years agoLU-6142 socklnd: change UAPI typedefs to proper structure 06/18506/5
James Simmons [Mon, 12 Dec 2016 18:15:48 +0000 (13:15 -0500)]
LU-6142 socklnd: change UAPI typedefs to proper structure

The upstream kernel requires proper structures so
convert all the UAPI typedefs in socklnd.h.

Remove the inline function socklnd_init_msg.
Its only used by the kernel code and functions
are not allowed in UAPI headers.

Change-Id: I4120c5abc4dda456289146452ed61e0162062cea
Signed-off-by: James Simmons <uja.ornl@yahoo.com>
Reviewed-on: https://review.whamcloud.com/18506
Reviewed-by: Dmitry Eremin <dmitry.eremin@intel.com>
Tested-by: Jenkins
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: Doug Oucharek <doug.s.oucharek@intel.com>
Reviewed-by: John L. Hammond <john.hammond@intel.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>
3 years agoLU-8562 osp: fix precreate_cleanup_orphans/precreate_reserve race 11/22211/7
Sergey Cheremencev [Thu, 10 Nov 2016 15:30:53 +0000 (23:30 +0800)]
LU-8562 osp: fix precreate_cleanup_orphans/precreate_reserve race

osp_statfs_interpret can clear error in opd_pre_status despite of the
fact that osp_precreate_cleanup_orphans got error and doesn't know
exactly OST object last_id. Example:
1. mdt sends req "create objects x..y"
2. objects created. mdt gets OK
3. MDT->OST reconnection
4. MDT sends cleanup_orphans last_used_fid=x
5. OST removes x..y and sends reply OK and last_id=x
6. MDT->OST connection aborted. cleanup_orphans exits with EIO
7. osp_statfs_interpret changes opd_pre_status from EIO to 0
8. osp_precreate_reserve reserves object and changes last_used_id from x to x+1
9. connection restored. MDT sends cleanup_orphans last_id=x+1
In fine OST has a gap - object x was removed by cleanup_orphans.

So don't clear opd_pre_recovering until we get last_id from OST.

Change-Id: I1647053fdab9a0c9bf59a048b0814e7b2dec52f2
Signed-off-by: Sergey Cheremencev <sergey.cheremencev@seagate.com>
Seagate-bug-id: MRP-3693
Reviewed-on: https://review.whamcloud.com/22211
Tested-by: Jenkins
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: Alex Zhuravlev <alexey.zhuravlev@intel.com>
Reviewed-by: John L. Hammond <john.hammond@intel.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>
3 years agoLU-8367 osp: do not block orphan cleanup 68/23168/17
Alex Zhuravlev [Fri, 14 Oct 2016 17:48:50 +0000 (20:48 +0300)]
LU-8367 osp: do not block orphan cleanup

do not block orphan cleanup till all reserved objects are consumed,
otherwise we risk to get into a livelock when one blocked thread
holding reservation can block recovery. instead ask OST to recreate
possible missing objects in our precreate window.

Change-Id: I066b0783cce54d7ecd25a08da5c76f211b7244f5
Signed-off-by: Alex Zhuravlev <alexey.zhuravlev@intel.com>
Reviewed-on: https://review.whamcloud.com/23168
Reviewed-by: Sergey Cheremencev <sergey.cheremencev@seagate.com>
Tested-by: Jenkins
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: Niu Yawei <yawei.niu@intel.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>
3 years agoLU-8807 llite: check reply status in ll_migrate() 66/23666/2
Niu Yawei [Wed, 9 Nov 2016 02:43:36 +0000 (21:43 -0500)]
LU-8807 llite: check reply status in ll_migrate()

ll_migrate() should check reply status before trying to read
reply buffer, checking if request is NULL doesn't make sense.

Signed-off-by: Niu Yawei <yawei.niu@intel.com>
Change-Id: I585844e0c1118fad23f19791dc95e34bc0d6e38a
Reviewed-on: https://review.whamcloud.com/23666
Tested-by: Jenkins
Reviewed-by: Andreas Dilger <andreas.dilger@intel.com>
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: Lai Siyao <lai.siyao@intel.com>
Reviewed-by: John L. Hammond <john.hammond@intel.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>
3 years agoLU-6374 tests: wait for zfs commit 88/23688/2
Niu Yawei [Thu, 10 Nov 2016 04:15:16 +0000 (23:15 -0500)]
LU-6374 tests: wait for zfs commit

wait_delete_completed_mds() should always wait for zfs commit to free
used space.

Signed-off-by: Niu Yawei <yawei.niu@intel.com>
Change-Id: If57e67841060d9f13ec3ba13e44f7b88a82301c5
Reviewed-on: https://review.whamcloud.com/23688
Tested-by: Jenkins
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: Jian Yu <jian.yu@intel.com>
Reviewed-by: Hongchao Zhang <hongchao.zhang@intel.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>
3 years agoLU-8901 misc: update Intel copyright messages for 2016 33/24233/4
Andreas Dilger [Thu, 8 Dec 2016 08:35:18 +0000 (01:35 -0700)]
LU-8901 misc: update Intel copyright messages for 2016

Update copyright messages for files updated in 2016, excluding
trivial patches.

Add trivial patches to updatecw.sh script exclude list.  Improve
the output of patches causing the most changes to include the
patch summary instead of just the hash.

Test-Parameters: trivial
Signed-off-by: Andreas Dilger <andreas.dilger@intel.com>
Change-Id: Ifb098e8126f698ea975e10ba13f5f47c449ae4ad
Reviewed-on: https://review.whamcloud.com/24233
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>
Tested-by: Oleg Drokin <oleg.drokin@intel.com>
3 years agoLU-7765 tests: Fix replay-dual test_26 stderr redirection 65/18365/6
Oleg Drokin [Tue, 9 Feb 2016 17:05:30 +0000 (12:05 -0500)]
LU-7765 tests: Fix replay-dual test_26 stderr redirection

Using 2&>1 just redirects to a file named 1, while we need
to use 2>&1 to redirect to descriptor one.  In fact the
recommended form is &>/dev/null that means >/dev/null 2>&1.

Test-Parameters: trivial
Change-Id: I008e212cebe2f3d6f6c7b7ce000b508aaf63da38
Signed-off-by: Oleg Drokin <oleg.drokin@intel.com>
Reviewed-on: https://review.whamcloud.com/18365
Reviewed-by: Andreas Dilger <andreas.dilger@intel.com>
Tested-by: Jenkins
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: James Nunez <james.a.nunez@intel.com>
3 years agoLU-7735 utils: Allow revert of NO_PRIMNODE 04/18304/5
Sebastien Piechurski [Thu, 4 Feb 2016 16:45:16 +0000 (11:45 -0500)]
LU-7735 utils: Allow revert of NO_PRIMNODE

When formatting a target or modifying an existing one with the
parameter "--servicenode", the flag no_primnode is set on the target,
modifying the way the target is then registered on the MGS to avoid
adding the NID of the node actually registering and let only the nids
given with --servicenode parameters.

However, if someone wants to revert to the older behavior with the
"--failnode" parameter, the no_primnode flag is never unset, and the
primary node NID is never added to the mountdata.

Test-Parameters: trivial testlist=conf-sanity
Signed-off-by: Nathaniel Clark <nathaniel.l.clark@intel.com>
Change-Id: I2251f66236c91365c03e7d138cd8e6508956d73d
Reviewed-on: https://review.whamcloud.com/18304
Reviewed-by: Bob Glossman <bob.glossman@intel.com>
Tested-by: Jenkins
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: Andreas Dilger <andreas.dilger@intel.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>