Various macros in "plx9080.h" take the form `PLX_<REG>_<FIELD>(x)`,
where `<REG>` is a register name, `<FIELD>` is a field within the
register, and `x` is a value for the field specified by the caller. The
macros construct a partial register register with the specified field
value placed in the appropriate bits of the register value, and other
bits of the register value zeroed. Add corresponding macros of the form
`PLX_<REG>_TO_<FIELD>(r)` that extract a field value from a specified
register values. Remove macros of the form `PLX_<REG>_<FIELD>_SHIFT`
that specified the bit position of a field within a register value as
they are no longer useful.
Signed-off-by: Ian Abbott <abbotti@mev.co.uk> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Christian Gromm [Fri, 19 Aug 2016 09:13:01 +0000 (11:13 +0200)]
staging: most: hdm-usb: add support for new USB gadget
This patch is needed to make the driver support Microchip's OS81210 USB
MOST network interface controller. It simply adds the gadget's product
ID to the driver's ID table.
Signed-off-by: Christian Gromm <christian.gromm@microchip.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Christian Gromm [Fri, 19 Aug 2016 09:13:00 +0000 (11:13 +0200)]
staging: most: hdm-usb: remove unnecessary status assignment
The USB completion callbacks set the status field of an MBO object before
scheduling the clear_work. This patch removes this redundant assignment as
the work_struct does the same for all MBOs.
Christian Gromm [Fri, 19 Aug 2016 09:12:56 +0000 (11:12 +0200)]
staging: most: hdm-usb: synchronize release of struct buf_anchor
In case a channel that is going to be destroyed has been tagged as not
"healthy" by the function hdm_poison_channel() while the functions
hdm_write_completion() or hdm_read_completion() are being executed, they
race for destruction of buf_anchor.
Christian Gromm [Fri, 19 Aug 2016 09:12:54 +0000 (11:12 +0200)]
staging: most: hdm-usb: fix race between enqueue and most_stop_enqueue
The "broken in pipe" handler of the USB-HDM calls most_stop_enqueue() to
stop the MBO traffic before returning all MBOs back to the Mostcore. As
the enqueue() call from the Mostcore may run in parallel with the
most_stop_enqueue(), the HDM may run into the inconsistent state and
crash the kernel.
This patch synchronizes enqueue(), most_stop_enqueue() and
most_resume_enqueue() with a mutex, hence avoiding the race condition.
Christian Gromm [Fri, 19 Aug 2016 09:12:51 +0000 (11:12 +0200)]
staging: most: hdm-usb: make use of is_channel_healthy flag
This patch makes the write completion handler use the is_channel_healthy
flag to prevent the hdm from scheduling a second clear_halt workqueue in
case an endpoint reported a STALL condition.
Signed-off-by: Christian Gromm <christian.gromm@microchip.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Christian Gromm [Fri, 19 Aug 2016 09:12:47 +0000 (11:12 +0200)]
staging: most: hdm-usb: fix clear halt processing
This patch is needed to ensure that submitted URBs get unlinked before
the driver calls usb_clear_halt(). Since the halt condition of an USB
endpoint is channel related, the work_struct is moved from a buffer
basis to a channel basis.
Signed-off-by: Christian Gromm <christian.gromm@microchip.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Christian Gromm [Fri, 19 Aug 2016 11:09:34 +0000 (13:09 +0200)]
staging: most: aim-cdev: report error returned by alloc_chrdev_region
This patch forwards the error code returned by function
alloc_chrdev_region(). It is needed to stop the module
from hiding the actual cause of failure.
Signed-off-by: Christian Gromm <christian.gromm@microchip.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Christian Gromm [Fri, 19 Aug 2016 11:09:33 +0000 (13:09 +0200)]
staging: most: aim-cdev: fix reported error codes
Currently, the aim-cdev is returning different error codes for the same
root cause. This patch is needed to get rid of the module's inconsistency
when reporting errors.
Signed-off-by: Christian Gromm <christian.gromm@microchip.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Oleg Drokin [Sat, 20 Aug 2016 21:34:30 +0000 (17:34 -0400)]
Add James Simmons as another Lustre maintainer
James Simmons is also spendign a lot of efforts on
cleaning up staging tree Lustre code and also
helps to sync up all the missing changes from the other tree.
John L. Hammond [Fri, 19 Aug 2016 18:07:30 +0000 (14:07 -0400)]
staging: lustre: obd: decruft md_enqueue() and md_intent_lock()
Remove the lmm and lmmsize parameters from both functions, storing
that data in md_op_data when needed. Remove the unused lookup_flags
parameter from md_intent_lock(), and the unused reqp parameter from
md_enqueue(). Add a union ldlm_policy_data * parameter to
md_enqueue(). Remove the unused function lmv_enqueue_remote().
Signed-off-by: John L. Hammond <john.hammond@intel.com>
Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-2675
Reviewed-on: http://review.whamcloud.com/10205 Reviewed-by: wangdi <di.wang@intel.com> Reviewed-by: Jinshan Xiong <jinshan.xiong@intel.com> Reviewed-by: Oleg Drokin <oleg.drokin@intel.com> Signed-off-by: James Simmons <jsimmons@infradead.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
wang di [Fri, 19 Aug 2016 18:07:28 +0000 (14:07 -0400)]
staging: lustre: mdt: add OBD_CONNECT_DIR_STRIPE flag
Add OBD_CONNECT_DIR_STRIPE to tell if the client supports
striped dir, so only new client (>= 2.6) can access striped
directory.
Signed-off-by: wang di <di.wang@intel.com>
Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-4843
Reviewed-on: http://review.whamcloud.com/10773 Reviewed-by: Andreas Dilger <andreas.dilger@intel.com> Reviewed-by: John L. Hammond <john.hammond@intel.com> Reviewed-by: Bob Glossman <bob.glossman@intel.com> Signed-off-by: James Simmons <jsimmons@infradead.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
This patch fixes the following compile error by removing the
dead codes: "error: 'xxx_readpages' defined but not used".
Now that we have md_read_page functionality we can remove
all the *_readpage implementations.
Signed-off-by: Jian Yu <jian.yu@intel.com>
Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-4669
Reviewed-on: http://review.whamcloud.com/9810 Reviewed-by: John L. Hammond <john.hammond@intel.com> Reviewed-by: Andreas Dilger <andreas.dilger@intel.com> Signed-off-by: James Simmons <jsimmons@infradead.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
wang di [Fri, 19 Aug 2016 18:07:26 +0000 (14:07 -0400)]
staging: lustre: llite: move dir cache to MDC layer
Move directory entries cache from llite to MDC, so client
side dir stripe will use independent hash function(in LMV),
which does not need to be tightly coupled with the backend
storage dir-entry hash function. With striped directory, it
will be 2-tier hash, LMV calculate hash value according to the
name and hash-type in layout, then each MDT will store these
entry in disk by its own hash.
Signed-off-by: wang di <di.wang@intel.com>
Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-3531
Reviewed-on: http://review.whamcloud.com/7043 Reviewed-by: John L. Hammond <john.hammond@intel.com> Reviewed-by: Jinshan Xiong <jinshan.xiong@intel.com> Reviewed-by: Andreas Dilger <andreas.dilger@intel.com> Reviewed-by: Oleg Drokin <oleg.drokin@intel.com> Signed-off-by: James Simmons <jsimmons@infradead.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
wang di [Fri, 19 Aug 2016 18:07:25 +0000 (14:07 -0400)]
staging: lustre: lmv: implement lmv version of read_page
All the code needed to implement read_page. This
will eventually replace lmv_readpage.
Signed-off-by: wang di <di.wang@intel.com>
Reviewed-on: http://review.whamcloud.com/10761
Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-4906 Reviewed-by: John L. Hammond <john.hammond@intel.com> Reviewed-by: Andreas Dilger <andreas.dilger@intel.com> Reviewed-by: Oleg Drokin <oleg.drokin@intel.com> Signed-off-by: James Simmons <jsimmons@infradead.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Oleg Drokin [Sat, 20 Aug 2016 21:34:25 +0000 (17:34 -0400)]
staging/lustre: Always return EEXIST on mkdir for existing names
if the name already exists, but we don't have write permissions
in the parent, force talking to the MDS to determine what
more sensical error code to return.
This also happens to fix matlab and other such programs that
assume that EEXIST is the only valid error code for mkdir of
an existing directory.
Doug Oucharek [Tue, 16 Aug 2016 20:19:33 +0000 (16:19 -0400)]
staging: lustre: lnet: Stop Infinite CON RACE Condition
In current code, when a CON RACE occurs, the passive side will
let the node with the higher NID value win the race.
We have a field case where a node can have a "stuck"
connection which never goes away and is the trigger of a
never-ending loop of re-connections.
This patch introduces a counter to how many times a
connection in a connecting state has been the cause of a CON RACE
rejection. After 20 times (constant MAX_CONN_RACES_BEFORE_ABORT),
we assume the connection is stuck and let the other side (with
lower NID) win.
Signed-off-by: Doug Oucharek <doug.s.oucharek@intel.com>
Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-7646
Reviewed-on: http://review.whamcloud.com/19430 Reviewed-by: Amir Shehata <amir.shehata@intel.com> Reviewed-by: Andreas Dilger <andreas.dilger@intel.com> Reviewed-by: Oleg Drokin <oleg.drokin@intel.com> Signed-off-by: James Simmons <jsimmons@infradead.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Liang Zhen [Tue, 16 Aug 2016 20:19:32 +0000 (16:19 -0400)]
staging: lustre: lnet: lock improvement for ko2iblnd
kiblnd_check_sends() takes conn::ibc_lock at the begin and release
this lock at the end, this is inefficient because most use-case
needs to explicitly release ibc_lock before caling this function.
This patches changes it to kiblnd_check_sends_locked() and avoid
unnecessary lock dances.
Alexander Boyko [Tue, 16 Aug 2016 20:19:31 +0000 (16:19 -0400)]
staging: lustre: lnet: make connection more stable with packet loss
IB network may lose last connection handshake packet.
This problem isn't Lustre specific and described at
https://oss.oracle.com/pipermail/rds-devel/2007-December/000271.html
for example. Solution is to make conection established if any packet
is received for it.
Doug Oucharek [Tue, 16 Aug 2016 20:19:30 +0000 (16:19 -0400)]
staging: lustre: lnet: Correct position of lnet_ni_decref()
In fix http://review.whamcloud.com/#/c/19614/, the call
to lnet_ni_decref() should have followed the routines
which are using the NI. This patch correct that.
Signed-off-by: Doug Oucharek <doug.s.oucharek@intel.com>
Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-8022
Reviewed-on: http://review.whamcloud.com/21001 Reviewed-by: James Simmons <uja.ornl@yahoo.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> Signed-off-by: James Simmons <jsimmons@infradead.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Doug Oucharek [Tue, 16 Aug 2016 20:19:29 +0000 (16:19 -0400)]
staging: lustre: lnet: Do not drop message when shutting down LNet
There is a case in lnet_parse() where we discover that LNet
is shutting down but we continue to use the NI when we drop the
message and end up calling ko2iblnd_check_send_locked() which tries to
allocate from the Tx pool which has been cleaned up already.
This triggers a NULL pointer dereference.
This fix just returns from lnet_parse() when we disover LNet is
shutting down.
wang di [Tue, 16 Aug 2016 20:19:27 +0000 (16:19 -0400)]
staging: lustre: obd: implement md_read_page
This patch adds md_read_page which is a new more
flexiable api that will replace md_readpage.
Signed-off-by: wang di <di.wang@intel.com>
Reviewed-on: http://review.whamcloud.com/10761
Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-4906 Reviewed-by: John L. Hammond <john.hammond@intel.com> Reviewed-by: Andreas Dilger <andreas.dilger@intel.com> Reviewed-by: Oleg Drokin <oleg.drokin@intel.com> Signed-off-by: James Simmons <jsimmons@infradead.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
John L. Hammond [Tue, 16 Aug 2016 20:19:25 +0000 (16:19 -0400)]
staging: lustre: obd: validate open handle cookies
Add a const void *h_owner member to struct portals_handle. Add a const
void *owner parameter to class_handle2object() which must be matched
by the h_owner member of the handle in addition to the cookie.
Signed-off-by: John L. Hammond <john.hammond@intel.com>
Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-3233
Reviewed-on: http://review.whamcloud.com/6938 Reviewed-by: Andreas Dilger <andreas.dilger@intel.com> Reviewed-by: Fan Yong <fan.yong@intel.com> Reviewed-by: Mike Pershin <mike.pershin@intel.com> Signed-off-by: James Simmons <jsimmons@infradead.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
James Simmons [Tue, 16 Aug 2016 20:19:24 +0000 (16:19 -0400)]
staging: lustre: llite: remove assert for acl refcount
The purpose of this asssert to was to ensure lustre
was properly managing its posix_acl access. This test
is invalid due to the VFS layer also taking references
on the posix_acl. In reality their is no simple way to
detect this class of mistakes.
Signed-off-by: James Simmons <jsimmons@infradead.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Hongchao Zhang [Tue, 16 Aug 2016 20:19:22 +0000 (16:19 -0400)]
staging: lustre: osc: Automatically increase the max_dirty_mb
When RPC size or the max RPCs in flight is increased, the actual
limit might be max_dirty_mb. This patch automatically increases
the max_dirty_mb value at connection time and when the related
values are tuned manually by proc file system.
this patch also changes the unit of "cl_dirty" and "cl_dirty_max"
in client_obd from byte to page.
Signed-off-by: Li Xi <lixi@ddn.com> Signed-off-by: Hongchao Zhang <hongchao.zhang@intel.com>
Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-4933
Reviewed-on: http://review.whamcloud.com/10446 Reviewed-by: Jinshan Xiong <jinshan.xiong@intel.com> Reviewed-by: Andreas Dilger <andreas.dilger@intel.com> Signed-off-by: James Simmons <jsimmons@infradead.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Fan Yong [Tue, 16 Aug 2016 20:19:21 +0000 (16:19 -0400)]
staging: lustre: lmv: build master LMV EA dynamically build via readdir
When creating a striped directory, the master object saves the slave
objects (or shards) as internal sub-directories. The sub-directory's
name is composed of ${shard_FID}:${shard_idx}. With the name, we can
easily to know what the shard is and where it should be.
On the other hand, we need to store some information related with the
striped directory, such as magic, hash type, shards count, and so on.
That is the LMV EA (header). We do NOT store the FID of each shard in
the LMV EA. Instead, when we need the shards' FIDs (such as readdir()
on client-side), we can build the entrie LMV EA on the MDT (in RAM) by
iterating the sub-directory entries that are contained in the master
object of the striped directroy.
Above mechanism can simplify the striped directory create operation.
For very large striped directory, logging the FIDs array in the LMV
EA will be trouble. It also simplify the LFSCK for verifying striped
directory, because it reduces the inconsistency sources.
Another fixing is about the lmv_master_fid in master LMV EA header,
it is redundant information, and may become one of the inconsistency
sources. So replace it with two __u64 padding fields.
Signed-off-by: Fan Yong <fan.yong@intel.com>
Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-5223
Reviewed-on: http://review.whamcloud.com/10751 Reviewed-by: Andreas Dilger <andreas.dilger@intel.com> Reviewed-by: Alex Zhuravlev <alexey.zhuravlev@intel.com> Reviewed-by: Oleg Drokin <oleg.drokin@intel.com> Signed-off-by: James Simmons <jsimmons@infradead.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
wang di [Tue, 16 Aug 2016 20:19:19 +0000 (16:19 -0400)]
staging: lustre: lmv: try all stripes for unknown hash functions
For unknown hash type, LMV should try all stripes to locate
the name entry. But it will only for lookup and unlink, i.e.
we can only list and unlink entries under striped dir with
unknown hash type.
Signed-off-by: wang di <di.wang@intel.com>
Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-4921
Reviewed-on: http://review.whamcloud.com/10041 Reviewed-by: John L. Hammond <john.hammond@intel.com> Reviewed-by: Andreas Dilger <andreas.dilger@intel.com> Signed-off-by: James Simmons <jsimmons@infradead.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Mikhail Pershin [Tue, 16 Aug 2016 20:19:18 +0000 (16:19 -0400)]
staging: lustre: llog: keep llog ctxt indices constant
The llog context id table cannot be shrunk easily because that
will cause index shifting and incompatibility between old client
and new server and vice versa.
Patch moves llog_ctxt_id table to the lustre_idl.h because this is
wire protocol data, these values are added to the wirecheck.
Lai Siyao [Tue, 16 Aug 2016 20:19:16 +0000 (16:19 -0400)]
staging: lustre: fid: do open-by-fid by default
Currently client open-by-fid often packs name into the request,
but the name may be invalid, eg. NFS export, and even if it's
valid, it may cause inconsistency because this operation is done
on this fid, which is globally unique, but name not.
Since open-by-fid doesn't pack name, for striped dir we can't know
parent stripe fid on client, so we set parent fid the same as
child fid, and MDT has to find its parent fid from linkea (this is
already supported by MDT).
M_CHECK_STALE becomes obsolete.
Unset MDS_OPEN_FL_INTERNAL from open syscall flags, because these
flags are internally used, and should not be set from user space.
It's not necessary to store parent fid in lli_pfid, because MDT
can get it's parent fid from linkea, and now that DNE stripe
directory stores master inode fid in lli_pfid, stop storing parent
fid to avoid conflict.
Signed-off-by: Lai Siyao <lai.siyao@intel.com>
Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-3544
Reviewed-on: http://review.whamcloud.com/7476
Reviewed-on: http://review.whamcloud.com/10692 Reviewed-by: Fan Yong <fan.yong@intel.com> Reviewed-by: Nathaniel Clark <nathaniel.l.clark@intel.com> Reviewed-by: wangdi <di.wang@intel.com> Reviewed-by: John L. Hammond <john.hammond@intel.com> Reviewed-by: Oleg Drokin <oleg.drokin@intel.com> Signed-off-by: James Simmons <jsimmons@infradead.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Brian Behlendorf [Tue, 16 Aug 2016 20:19:15 +0000 (16:19 -0400)]
staging: lustre: obd: limit lu_object cache
As the LU cache grows it can consume large enough chunks of
memory that ends up preventing buffers for other objects,
such as the OIs, from being cached and severely impacting
the performance for FID lookups. Limit the lu_object cache
to a maximum of lu_cache_nr objects.
NOTES:
* In order to be able to quickly determine the number of objects in
the hash table the CFS_HASH_COUNTER flag is added. This adds an
atomic_inc/dec to the hash insert/remove paths but is not expected
to have any measurable impact of performance.
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-5164
Reviewed-on: http://review.whamcloud.com/10237 Reviewed-by: Andreas Dilger <andreas.dilger@intel.com> Reviewed-by: Alex Zhuravlev <alexey.zhuravlev@intel.com> Reviewed-by: Isaac Huang <he.huang@intel.com> Reviewed-by: Oleg Drokin <oleg.drokin@intel.com> Signed-off-by: James Simmons <jsimmons@infradead.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Emoly Liu [Tue, 16 Aug 2016 20:19:13 +0000 (16:19 -0400)]
staging: lustre: ldlm: improve ldlm_lock_create() return value
ldlm_lock_create() and ldlm_resource_get() always return NULL as
error reporting and "NULL" is interpretted as ENOMEM incorrectly
sometimes. This patch fixes this problem by using ERR_PTR() rather
than NULL.
Signed-off-by: Emoly Liu <emoly.liu@intel.com>
Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-4524
Reviewed-on: http://review.whamcloud.com/9004 Reviewed-by: Bobi Jam <bobijam@gmail.com> Reviewed-by: Andreas Dilger <andreas.dilger@intel.com> Reviewed-by: John L. Hammond <john.hammond@intel.com> Signed-off-by: James Simmons <jsimmons@infradead.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
wang di [Tue, 16 Aug 2016 20:19:11 +0000 (16:19 -0400)]
staging: lustre: mdc: always use D_INFO for debug info when mdc_put_rpc_lock fails
Also use D_INFO no matter what the error returned from
mdc_put_rpc_lock.
Signed-off-by: Andreas Dilger <andreas.dilger@intel.com> Signed-off-by: wang di <di.wang@intel.com>
Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-4973
Reviewed-on: http://review.whamcloud.com/10150 Reviewed-by: Mike Pershin <mike.pershin@intel.com> Signed-off-by: James Simmons <jsimmons@infradead.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
A few changes are made in this patch for unstable pages tracking:
1. Remove kernel NFS unstable pages tracking because it killed
performance
2. Track unstable pages as part of LRU cache. Otherwise Lustre
can use much more memory than max_cached_mb
3. Remove obd_unstable_pages tracking to avoid using global
atomic counter
4. Make unstable pages track optional. Tracking unstable pages is
turned off by default, and can be controlled by
llite.*.unstable_stats.
Signed-off-by: Jinshan Xiong <jinshan.xiong@intel.com>
Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-4841
Reviewed-on: http://review.whamcloud.com/10003 Reviewed-by: Andreas Dilger <andreas.dilger@intel.com> Reviewed-by: Lai Siyao <lai.siyao@intel.com> Reviewed-by: Oleg Drokin <oleg.drokin@intel.com> Signed-off-by: James Simmons <jsimmons@infradead.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Jinshan Xiong [Tue, 16 Aug 2016 20:19:09 +0000 (16:19 -0400)]
staging: lustre: clio: Reduce memory overhead of per-page allocation
A page in clio used to occupy 584 bytes, which will use size-1024
slab cache. This patch reduces the per-page overhead to 512 bytes
so it can use size-512 instead.
Signed-off-by: Jinshan Xiong <jinshan.xiong@intel.com>
Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-4793
Reviewed-on: http://review.whamcloud.com/10070 Reviewed-by: Andreas Dilger <andreas.dilger@intel.com> Reviewed-by: Bobi Jam <bobijam@gmail.com> Signed-off-by: James Simmons <jsimmons@infradead.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
John L. Hammond [Tue, 16 Aug 2016 20:19:08 +0000 (16:19 -0400)]
staging: lustre: mdt: add mbo_ prefix to members of struct mdt_body
Rename each member of struct mdt_body, adding the prefix mbo_.
Signed-off-by: John L. Hammond <john.hammond@intel.com>
Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-2675
Reviewed-on: http://review.whamcloud.com/10202 Reviewed-by: Bob Glossman <bob.glossman@intel.com> Reviewed-by: Andreas Dilger <andreas.dilger@intel.com> Reviewed-by: Mike Pershin <mike.pershin@intel.com> Signed-off-by: James Simmons <jsimmons@infradead.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Hongchao Zhang [Tue, 16 Aug 2016 20:19:07 +0000 (16:19 -0400)]
staging: lustre: llite: set dir LOV xattr length variable
the LOV xattr of directory could be either lov_user_md_v1
(size is 32) or lov_user_md_v3 (size is 48), then the actual
size of the LOV xattr should be return.
wang di [Tue, 16 Aug 2016 20:19:05 +0000 (16:19 -0400)]
staging: lustre: llite: use the correct mode for striped directory
Create striped directory with correct mode, which should be
handling same as mkdir.
Signed-off-by: wang di <di.wang@intel.com>
Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-4929
Reviewed-on: http://review.whamcloud.com/10028 Reviewed-by: Andreas Dilger <andreas.dilger@intel.com> Reviewed-by: John L. Hammond <john.hammond@intel.com> Signed-off-by: James Simmons <jsimmons@infradead.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
wang di [Tue, 16 Aug 2016 20:19:04 +0000 (16:19 -0400)]
staging: lustre: lmv: Match MDT where the FID locates first
With DNE every object can have two locks in different namespaces:
lookup lock in space of MDT storing direntry and update/open lock
in space of MDT storing inode. In lmv_find_cbdata/lmv_lock_lock,
it should try the MDT that the FID maps to first, since this can
be easily found, and only try others if that fails.
In the error handler of lmv_add_targets, it should check whether
ld_tgt_count is being increased before ld_tgt_count is being -1.
Signed-off-by: wang di <di.wang@intel.com>
Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-4098
Reviewed-on: http://review.whamcloud.com/8019 Reviewed-by: Andreas Dilger <andreas.dilger@intel.com> Reviewed-by: Fan Yong <fan.yong@intel.com> Signed-off-by: James Simmons <jsimmons@infradead.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Fan Yong [Tue, 16 Aug 2016 20:19:03 +0000 (16:19 -0400)]
staging: lustre: lov: new pattern flag for partially repaired file
When the layout LFSCK repairs orphan OST-object, if the parent
MDT-object was lost, then it will re-create the MDT-object and
regenerate the LOV EA and fill the target LOV EA slot with the
orphan information, and fill other slots with zero (LOV hole);
if related LOV EA slot is invalid or hole, then it will refill
the target LOV EA slot; if the target slot exceeds current LOV
EA tail, then extend the LOV EA, and fill the gaps as zero.
Some of the LOV EA holes may cannot be re-filled finally because
of lost some OST-objects. And even if they can be re-filled, but
there are still some possible race accessings from client before
the re-filling. If the client access the LOV EA with hole(s), it
may cause some strange behaviour, such as trigger LBUG()/LASSERT()
on the client.
So we will make the client to be aware of the LOV EA is incomplete.
We introduce a new LOV EA pattern flag LOV_PATTERN_F_HOLE for that:
any time when the LFSCK repairs the LOV EA with hole(s), the LOV EA
will be marked as LOV_PATTERN_F_HOLE; when all the holes in the LOV
EA are refilled, the LOV_PATTERN_F_HOLE will be dropped.
For a new client, it recongizes the pattern flag LOV_PATTERN_F_HOLE,
then it can permit/forbid some opertions on the file with LOV holes:
1) Normal read/write the file with LOV EA hole is permitted, but the
application will get EIO error when read data from the dummy slot
or write data to the dummy slot.
2) The users can dump the recovered data via some common read tools,
such as "dd conv=sync,noerror".
3) Append data to the file which has LOV EA hole will get EIO failure.
4) Other operations will skip the LOV EA hole(s), and will not get
failures, such as {s,g}etattr, {s,g}getxattr, stat, chown/chgrp,
chmod, touch, unlink, and so on.
For an old client, since it will not recognize the new pattern flag
LOV_PATTERN_F_HOLE. So the LOV EA with hole will be dicarded with
failure, but it will not cause the client to be crashed.
Signed-off-by: Fan Yong <fan.yong@intel.com>
Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-4675
Reviewed-on: http://review.whamcloud.com/10042 Reviewed-by: Andreas Dilger <andreas.dilger@intel.com> Reviewed-by: Alex Zhuravlev <alexey.zhuravlev@intel.com> Reviewed-by: Oleg Drokin <oleg.drokin@intel.com> Signed-off-by: James Simmons <jsimmons@infradead.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
wang di [Tue, 16 Aug 2016 20:19:02 +0000 (16:19 -0400)]
staging: lustre: lmv: validate lock with correct stripe FID
In ll_lookup_it_finish, we need use the real parent(stripe)
FID to validate the parent UPDATE lock.
Signed-off-by: wang di <di.wang@intel.com>
Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-4925
Reviewed-on: http://review.whamcloud.com/10026 Reviewed-by: John L. Hammond <john.hammond@intel.com> Reviewed-by: Andreas Dilger <andreas.dilger@intel.com> Reviewed-by: Oleg Drokin <oleg.drokin@intel.com> Signed-off-by: James Simmons <jsimmons@infradead.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
wang di [Tue, 16 Aug 2016 20:19:01 +0000 (16:19 -0400)]
staging: lustre: llite: a few fixes about readdir of striped dir.
Normally we know the value of op_mea1 when ll_readdir is called.
In the case of '.' or '..' op_mea1 is unknown so for that case
fetch the real parents FID.
Signed-off-by: wang di <di.wang@intel.com> Signed-off-by: Li Xi <lixi@ddn.com>
Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-4603
Reviewed-on: http://review.whamcloud.com/9191 Reviewed-by: John L. Hammond <john.hammond@intel.com> Reviewed-by: Andreas Dilger <andreas.dilger@intel.com> Reviewed-by: Fan Yong <fan.yong@intel.com> Reviewed-by: Li Xi <pkuelelixi@gmail.com> Reviewed-by: Oleg Drokin <oleg.drokin@intel.com> Signed-off-by: James Simmons <jsimmons@infradead.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
John L. Hammond [Tue, 16 Aug 2016 20:19:00 +0000 (16:19 -0400)]
staging: lustre: uapi: reduce scope of lustre_idl.h
Move the definition of OBD_OCD_VERSION() and similar macros from
lustre_idl.h to lustre_ver.h. These macros are primarily used in
comparisons to LUSTRE_VERSION_CODE which is defined in lustre_ver.h
and so should be defined there as well. Move a few definitions
(related to FIDs, quota and striping) from lustre_idl.h to
lustre_user.h.
Signed-off-by: John L. Hammond <john.hammond@intel.com>
Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-5065
Reviewed-on: http://review.whamcloud.com/10336 Reviewed-by: Andreas Dilger <andreas.dilger@intel.com> Reviewed-by: Bob Glossman <bob.glossman@intel.com> Signed-off-by: James Simmons <jsimmons@infradead.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Jinshan Xiong [Tue, 16 Aug 2016 20:18:58 +0000 (16:18 -0400)]
staging: lustre: llite: Fix the deadlock in balance_dirty_pages()
If the page is already dirtied in ll_write_end() and kernel tries
to call balance_dirty_pages() to write back dirty pages in the same
thread, this is deadlock case if the page is already held by clio.
This can also fix the issue of LU-4873.
Signed-off-by: Jinshan Xiong <jinshan.xiong@intel.com>
Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-4977
Reviewed-on: http://review.whamcloud.com/10149 Reviewed-by: Bobi Jam <bobijam@gmail.com> Reviewed-by: Lai Siyao <lai.siyao@intel.com> Reviewed-by: Oleg Drokin <oleg.drokin@intel.com> Signed-off-by: James Simmons <jsimmons@infradead.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
staging: lustre: Remove static declaration in anonymous union
It is not permitted in C++ to have a static declaration inside
of an anonymous union. The g++ compiler will complaine with an
error like this:
error: struct ost_id::<anonymous union>::ostid invalid; an
anonymous union can only have non-static data members [-fpermissive]
This patch changes the code to use an unnamed struct in place of
"struct ostid" inside of the anonymous union. That name declaration
was completely unnecessary anyway, since it was not used anywhere else.
Signed-off-by: Christopher J. Morrone <morrone2@llnl.gov>
Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-4987
Reviewed-on: http://review.whamcloud.com/10176 Reviewed-by: Robert Read <robert.read@intel.com> Reviewed-by: Andreas Dilger <andreas.dilger@intel.com> Signed-off-by: James Simmons <jsimmons@infradead.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Gregoire Pichon [Tue, 16 Aug 2016 20:18:56 +0000 (16:18 -0400)]
staging: lustre: llite: fix inconsistencies of root squash feature
Root squash exhibits inconsistent behaviour on a client when
enabled. If a file is not cached on the client, then root will get
a permission denied error when accessing the file. When
the file has recently been accessed by a regular user and is
still in cache, root will be able to access the file without error
because the permission check is only done by the client that
isn't aware of root squash.
While the only real security benefit from root squash is to deny
clients access to files owned by root itself, it also makes sense
to treat file access on the client in a consistent manner
regardless of whether the file is in cache or not.
This patch adds root squash settings to llite so that client is able
to apply root squashing when it is relevant.
Configuration of MDT root squash settings will automatically be
applied to llite config log as well.
Update cfs_str2num_check() routine by removing any modification
of the specified string parameter. Since string can come from ls_str
field of a lstr structure, this avoids inconsistent ls_len field.
John L. Hammond [Tue, 16 Aug 2016 20:18:55 +0000 (16:18 -0400)]
staging: lustre: llite: validate names
In ll_prep_md_op_data() validate names according to the same formula
used in mdd_name_check(). Add mdc_pack_name() to validate the name
actually packed in the request.
Signed-off-by: John L. Hammond <john.hammond@intel.com>
Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-4992
Reviewed-on: http://review.whamcloud.com/10198 Reviewed-by: wangdi <di.wang@intel.com> Reviewed-by: Andreas Dilger <andreas.dilger@intel.com> Signed-off-by: James Simmons <jsimmons@infradead.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
wang di [Tue, 16 Aug 2016 20:18:54 +0000 (16:18 -0400)]
staging: lustre: lmv: separate master object with master stripe
Separate master stripe with master object, so
1. stripeEA only exists on master object.
2. sub-stripe object will be inserted into master object
as sub-directory, and it can get the master object by "..".
By this, it will remove those specilities for stripe0 in
LMV and LOD. And also simplify LFSCK, i.e. consistency check
would be easier.
When then master object becomes an orphan, we should
mark all of its sub-stripes as dead object as well,
otherwise client might still be able to create files
under these stripes.
A few fixes for striped directory layout lock:
1. stripe 0 should be locked as EX, same as other stripes.
2. Acquire the layout for directory, when it is being unliked.
Signed-off-by: wang di <di.wang@intel.com>
Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-4690
Reviewed-on: http://review.whamcloud.com/9511 Reviewed-by: Andreas Dilger <andreas.dilger@intel.com> Reviewed-by: John L. Hammond <john.hammond@intel.com> Reviewed-by: Oleg Drokin <oleg.drokin@intel.com> Signed-off-by: James Simmons <jsimmons@infradead.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Chris Horn [Tue, 16 Aug 2016 20:18:53 +0000 (16:18 -0400)]
staging: lustre: ptlrpc: Early replies need to honor at_max
When determining whether an early reply can be sent the server will
calculate the new deadline based on an offset from the request
arrival time. However, when actually setting the new deadline
the server offsets the current time. This can result in deadlines
being extended more than at_max seconds past the request arrival
time. Instead, the server should offset the arrival time when updating
its request timeout.
When a client receives an early reply it doesn't know the server side
arrival time so we use the original sent time as an approximation.
Signed-off-by: Chris Horn <hornc@cray.com>
Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-4578
Reviewed-on: http://review.whamcloud.com/9100 Reviewed-by: James Simmons <uja.ornl@gmail.com> Reviewed-by: Christopher J. Morrone <chris.morrone.llnl@gmail.com> Reviewed-by: Oleg Drokin <oleg.drokin@intel.com> Signed-off-by: James Simmons <jsimmons@infradead.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
wang di [Tue, 16 Aug 2016 20:18:52 +0000 (16:18 -0400)]
staging: lustre: llite: add error handler in inode prepare phase
Add error handler during inode inialization, so inode will
become bad inode if something bad happens during inode prepare
phase, otherwise the striped directory will not get its layout
and being mis-regarded as normal directory.
Signed-off-by: wang di <di.wang@intel.com>
Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-4930
Reviewed-on: http://review.whamcloud.com/10170 Reviewed-by: Lai Siyao <lai.siyao@intel.com> Reviewed-by: Fan Yong <fan.yong@intel.com> Reviewed-by: John L. Hammond <john.hammond@intel.com> Reviewed-by: Oleg Drokin <oleg.drokin@intel.com> Signed-off-by: James Simmons <jsimmons@infradead.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
John L. Hammond [Tue, 16 Aug 2016 20:18:51 +0000 (16:18 -0400)]
staging: lustre: move ioctls to lustre_ioctl.h
Move ioctl definitions and related functions from lustre_dlm.h,
lustre_lib.h, obd.h, to lustre_ioctl.h. Replace the definitions of
retired ioctls with comment.
Signed-off-by: John L. Hammond <john.hammond@intel.com>
Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-4961
Reviewed-on: http://review.whamcloud.com/10139 Reviewed-by: Andreas Dilger <andreas.dilger@intel.com> Reviewed-by: Robert Read <robert.read@intel.com> Signed-off-by: James Simmons <jsimmons@infradead.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Vitaly Fertman [Tue, 16 Aug 2016 20:18:50 +0000 (16:18 -0400)]
staging: lustre: ldlm: flock completion fixes.
Move checks for FAILED, DESTROYED flags under ldlm spinlock,
destroy flock atomically with the check it is not destroyed yet.
Do not put the granted flock into the resource if this is
UNLOCK, TEST, or DEADLOCK'ed flock.
Later a regression for this patch was reported under LU-7626.
The refcount nonzero (1) after lock cleanup errors was reported.
The reason is that the case LCK_NL was not handled for obdecho.
Patch 17791 resolved this issue which has been combined into
this upstream patch.
Fan Yong [Tue, 16 Aug 2016 20:18:49 +0000 (16:18 -0400)]
staging: lustre: reorder LOV_MAGIC_* definition
Since all the LOV_MAGIC_* definitions have the same
postfix values break that value out into its own
definition. With this we can check whether the magic's
postfix match the LOV_MAGIC_MAGIC or not: if yes,
then it is quite possible that the clients has
encountered an newer LOV magic. This extra information
can let us handle those cases more gracefully.
Signed-off-by: Fan Yong <fan.yong@intel.com>
Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-4941
Reviewed-on: http://review.whamcloud.com/10045 Reviewed-by: Andreas Dilger <andreas.dilger@intel.com> Reviewed-by: Oleg Drokin <oleg.drokin@intel.com> Signed-off-by: James Simmons <jsimmons@infradead.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Jinshan Xiong [Tue, 16 Aug 2016 20:18:44 +0000 (16:18 -0400)]
staging: lustre: llite: avoid a deadlock in page write
For a partial page write, it will have to issue a READ RPC firstly
to get a full uptodate page. If another page is already locked by
this thread it can easily cause deadlock.
wang di [Tue, 16 Aug 2016 20:18:42 +0000 (16:18 -0400)]
staging: lustre: lmv: lookup remote migrating object in LMV
If remote object is being found in a migrating directory,
it should continue to lookup the object in remote MDT,
instead of return.
Signed-off-by: wang di <di.wang@intel.com>
Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-4805
Reviewed-on: http://review.whamcloud.com/9806 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> Signed-off-by: James Simmons <jsimmons@infradead.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
John L. Hammond [Tue, 16 Aug 2016 20:18:41 +0000 (16:18 -0400)]
staging: lustre: lmv: access lum_stripe_offset as little endian
By the time that a struct lmv_user_md reaches lmv_placement_policy()
it has already been converted to little endian. Therefore use the
appropriate macros around accesses to this this field. This issue was
found by rewriting the definition of struct lmv_user_md to use the
__leXX typedefs and running sparse.
Signed-off-by: John L. Hammond <john.hammond@intel.com>
Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-4738
Reviewed-on: http://review.whamcloud.com/9671 Reviewed-by: Dmitry Eremin <dmitry.eremin@intel.com> Reviewed-by: Swapnil Pimpale <spimpale@ddn.com> Reviewed-by: Oleg Drokin <oleg.drokin@intel.com> Signed-off-by: James Simmons <jsimmons@infradead.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
In mdc_ioc_swap_layouts() cancel *any* unused locks with LAYOUT or
XATTR IBITS set on the two files. (This matches the locks acquired in
mdt_swap_layouts(). Previously only locks that conflicted with a CR
LAYOUT lock were cancelled.)
Signed-off-by: John L. Hammond <john.hammond@intel.com>
Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-4320
Reviewed-on: http://review.whamcloud.com/9329 Reviewed-by: Andreas Dilger <andreas.dilger@intel.com> Reviewed-by: Faccini Bruno <bruno.faccini@intel.com> Reviewed-by: Oleg Drokin <oleg.drokin@intel.com> Signed-off-by: James Simmons <jsimmons@infradead.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
wang di [Tue, 16 Aug 2016 20:18:35 +0000 (16:18 -0400)]
staging: lustre: llite: a few fixes for migration.
1. Clear the client dentry cache before migrating file/directory
to the remote MDT.
2. Do not return stripe information to client, if it did not get
the layout lock.
Signed-off-by: wang di <di.wang@intel.com>
Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-4682
Reviewed-on: http://review.whamcloud.com/9522 Reviewed-by: Andreas Dilger <andreas.dilger@intel.com> Reviewed-by: John L. Hammond <john.hammond@intel.com> Signed-off-by: James Simmons <jsimmons@infradead.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Jinshan Xiong [Tue, 16 Aug 2016 20:18:34 +0000 (16:18 -0400)]
staging: lustre: osc: allow to call brw_commit() multiple times
Sometimes the rq_commit_cb of BRW RPC can be called twice if that RPC
has already committed at reply time. This will cause inaccuracy of
unstable pages accounting and then assertion.
projects / mirror_ubuntu-jammy-kernel.git / log