- 21 Aug, 2016 40 commits
-
-
wang di authored
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/9511Reviewed-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 authored
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/9100Reviewed-by:
James Simmons <uja.ornl@gmail.com> Reviewed-by:
Christopher J. Morrone <chris.morrone.llnl@gmail.com> Reviewed-by:
-
wang di authored
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:
Lai Siyao <lai.siyao@intel.com> Reviewed-by:
Fan Yong <fan.yong@intel.com> Reviewed-by:
-
John L. Hammond authored
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:
Robert Read <robert.read@intel.com> Signed-off-by:
-
Vitaly Fertman authored
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. Signed-off-by:
Vitaly Fertman <vitaly_fertman@xyratex.com> Signed-off-by:
Andriy Skulysh <andriy.skulysh@seagate.com> Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-2177Reviewed-by:
Alexey Lyashkov <alexey_lyashkov@xyratex.com> Reviewed-by:
Andriy Skulysh <andriy_skulysh@xyratex.com> Reviewed-by:
Bobi Jam <bobijam@gmail.com> Reviewed-by:
Mirza Arshad Mirza Hussain <arshad.hussain@seagate.com> Reviewed-by:
Alexey Leonidovich Lyashkov <alexey.lyashkov@seagate.com> Reviewed-on: http://review.whamcloud.com/17791Reviewed-by:
Niu Yawei <yawei.niu@intel.com> Reviewed-by:
Alex Zhuravlev <alexey.zhuravlev@intel.com> Reviewed-by:
-
Fan Yong authored
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 authored
Unify the flow control interfaces for MDC RPC and FLD RPC. We allow to adjust the maximum inflight RPCs count via /sys interface. Signed-off-by:
-
John L. Hammond authored
Rename request to preq. Signed-off-by:
Nathaniel Clark <nathaniel.l.clark@intel.com> Reviewed-by:
Bob Glossman <bob.glossman@intel.com> Reviewed-by:
-
John L. Hammond authored
In lmv_getattr_name() don't return a freed request in the error path. Signed-off-by:
-
Jinshan Xiong authored
Calculate the end of current stripe correctly when the stripe size is not power 2. Signed-off-by:
Jinshan Xiong <jinshan.xiong@intel.com> Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-4860 Reviewed-on: http://review.whamcloud.com/9882Reviewed-by:
-
Jinshan Xiong authored
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. Signed-off-by:
-
Nathaniel Clark authored
Ensure there aren't invalid pointers hanging around after ptlrpc_req_finished is called. Signed-off-by:
-
wang di authored
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:
-
John L. Hammond authored
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:
Dmitry Eremin <dmitry.eremin@intel.com> Reviewed-by:
Swapnil Pimpale <spimpale@ddn.com> Reviewed-by:
-
John L. Hammond authored
Several inline functions return a structure that was passed in. Their is no need for this so just make these function void. Signed-off-by:
-
John L. Hammond authored
Change the return type of several predicate functions from int to bool. Signed-off-by:
-
John L. Hammond authored
Add a const qualifier wherever possible to the pointer parameters of the inline helper functions in lustre_idl.h and lustre_fid.h. Signed-off-by:
-
John L. Hammond authored
Remove the const for the __u64 parameters for inline functions in lustre_idl.h. Signed-off-by:
-
John L. Hammond authored
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:
Faccini Bruno <bruno.faccini@intel.com> Reviewed-by:
-
wang di authored
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:
-
Jinshan Xiong authored
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. Signed-off-by:
Prakash Surya <surya1@llnl.gov> Reviewed-by:
-
Fan Yong authored
This enables the client to inject an error by altering the parent FID in order to test if the server side file system checker behaves properly. Signed-off-by:
-
Ryan Haasken authored
Since the log file name contains the current time in seconds, dumping the logs more than once per second causes EEXIST errors to be emitted. Add a static variable to libcfs_debug_dumplog_internal that records the time of the last Lustre log dump. If the current time in seconds is equal to the last time, do not dump logs again. Note that this is not thread-safe. However, in the rare case that two threads try to access last_dump_time simultaneously, the worst thing that could happen is that one of the threads will get an EEXIST error when trying to write the log file. This is no worse than the current situation, and it is not likely to happen. Signed-off-by:
Ryan Haasken <haasken@cray.com> Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-4129 Reviewed-on: http://review.whamcloud.com/8964Reviewed-by:
-
Dmitry Eremin authored
'plock.cookie' might be used uninitialized in this function. sscanf format specification '%d' expects type 'int *' for 'd', but parameter 3 has a different type '__u32*' Signed-off-by:
-
wang di authored
Add client support to migrate the individual inodes from one MDT to another MDT, and this functionality will only migrate inode layout on MDT but not touch data object on OST. The directory will be migrated from top to the bottom, i.e. migrating parent first, then migrating the child. Signed-off-by:
-
Fan Yong authored
There was no protection when inc/dec lu_device_type::ldt_device_nr, which may caused the ldt_device_nr to be wrong and trigger assert. This patch redefine lu_device_type::ldt_device_nr as atomic type. There was no protection when add/del lu_device_type::ldt_linkage into/from the global lu_device_types list, which may caused bad address accessing. This patch uses the existing obd_types_lock to protect related operations. We do NOT need lu_types_stop() any longer. Such function scans the global lu_device_types list, and for each type item on it which has zerod lu_device_type::ldt_device_nr, call its stop() method. In fact, the lu_device_type::ldt_device_nr only will be zero when the last lu_device_fini() is called, and at that time, inside the lu_device_fini(), its stop() method will be called. So it is unnecessary to call the stop() again via lu_types_stop(). Signed-off-by:
Jian Yu <jian.yu@intel.com> Reviewed-by:
-
Jian Yu authored
This patch handles the upgrade situation that old mountdata already contains comma-separated NIDs. The correct way to fix the original issue is to parse comma-separated NIDs in lmd_parse(). Signed-off-by:
Sebastien Buisson <sebastien.buisson@bull.net> Reviewed-by:
-
wang di authored
Add delete striped directory, it includes 1. enable sync log between MDTs, so slave objects will be delete by unlink log, which is similar as deleting ost object. 2. retrieve layout information of striped directory on MDT, then lock all of the slave objects before unlink. Signed-off-by:
-
wang di authored
Fix "lfs getdirstripe", so it can show layout information of striped directory [root@testnode tests]# ../utils/lfs getdirstripe /mnt/lustre/test1 /mnt/lustre/test1 lmv_stripe_count: 2 lmv_stripe_offset: 0 mdtidx FID[seq:oid:ver] 0 [0x280000400:0x1:0x0] 1 [0x2c0000400:0x1:0x0] Signed-off-by: -
wang di authored
1. client send create request to the master MDT, which will allocate FIDs and create slaves. for all of slaves. 2. Client needs to revalidate slaves during intent getattr and open request. 3. lmv_stripe_md will include attributes(size, nlink etc) from all of stripe, which will be protected by UPDATE lock. client needs to merge these attributes when update inode. 4. send create request to the MDT where the file is located, which can help creating master stripe of striped directory. Signed-off-by:
-
wang di authored
Use lmv_free_memmd for proper cleanup instead of the generic obd_free_memmd. Signed-off-by:
-
wang di authored
Completely replace lmv_get_easize with lmv_mds_md_size. With this change we can delete lmv_get_easize. Signed-off-by:
-
wang di authored
The lmv_[un]pack_md function are used to calculate the size of the data used to represent the LMV striping data. The original code was straight forward in its calculate with lmv_get_easize since only one type of data format could exist. We want to be able to support different version of this data in the future so this patch moves to generating the size of the data using the stripe count and which LMV_MAGIC_* version. Signed-off-by:
-
wang di authored
The MAX_HASH_* macros already exist in obd.h. Remove the duplicated defines in lustre_idl.h. Signed-off-by:
-
wang di authored
The function lmv_get_mea() is not used so remove it. Signed-off-by:
-
wang di authored
If the operation is IT_READDIR don't need to handle the mdc RPC lock. Signed-off-by:
-
wang di authored
Change the 3rd parameter remove to a bool for ll_release_page function. Signed-off-by:
-
wang di authored
Add a inode parameter to ll_release_page. This will be used in the future. Signed-off-by:
-
wang di authored
Group function prototypes together related to dir.c. Move ll_release_page to be with function declarations. Signed-off-by:
-
wang di authored
Remove comment about fixing swabbing that is not needed. Signed-off-by:
-
