Re: [PATCH 112/124] staging: lustre: ptlrpc: remove unnecessary EXPORT_SYMBOL

Hi frank, [auto build test ERROR on staging/staging-testing] [cannot apply to v4.8-rc7 next-20160916] [if your patch is applied to the wrong git tree, please drop us a note to help improve the system] [Suggest to use git(>=2.9.0) format-patch --base= (or --base=auto for convenience) to record

Re: [PATCH v2] staging: lustre: lustre/ldlm: Fixed sparse warnings

> On Sep 18, 2016, at 23:18, Nayeemahmed Badebade > wrote: > > Added __acquires / __releases sparse locking annotations > to lock_res_and_lock() and unlock_res_and_lock() functions > in l_lock.c, to fix below sparse warnings: > > l_lock.c:47:22: warning: context

[PATCH v2] staging: lustre: lustre/ldlm: Fixed sparse warnings

Added __acquires / __releases sparse locking annotations to lock_res_and_lock() and unlock_res_and_lock() functions in l_lock.c, to fix below sparse warnings: l_lock.c:47:22: warning: context imbalance in 'lock_res_and_lock' - wrong count at exit l_lock.c:61:6: warning: context imbalance in

[PATCH 103/124] staging: lustre: llite: pack suppgid to MDS correctly

From: Fan Yong The ll_lookup_it() may trigger IT_OPEN RPC to open a file by name. But at that time, the client does not know the target file's GID, so it cannot pack the necessary supplementary group ID in the RPC. Because of missing the supplementary group ID, the RPC maybe

[PATCH 048/124] staging: lustre: Flexible changelog format.

From: Henri Doreau Added jobid fields to Changelog records (and extended records). The CLF_JOBID flags allows to check if the field is present or not (old format) when reading an entry. Jobids are expressed as 32 chars long, zero-terminated strings. Updated test_205 in

[PATCH 043/124] staging: lustre: lmv: change type of lmv_obd->tgts_size to u32

From: Dmitry Eremin tgts_size is used as unsigned. Signed-off-by: Dmitry Eremin Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-5577 Reviewed-on: http://review.whamcloud.com/11881 Reviewed-by: Fan Yong Reviewed-by: John

[PATCH 035/124] staging: lustre: llite: style cleanup for ll_mkdir

From: John L. Hammond Style cleanup to make the code readable. Signed-off-by: John L. Hammond Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-2675 Reviewed-on: http://review.whamcloud.com/10769 Reviewed-by: Yang Sheng

[PATCH 020/124] staging: lustre: lmv: release request in lmv_revalidate_slaves()

From: John L. Hammond In lmv_revalidate_slaves() ensure that the request returned by md_intent_lock() is properly released on all paths. Signed-off-by: John L. Hammond Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-5452 Reviewed-on:

[PATCH 010/124] staging: lustre: llite: add LL_LEASE_{RD,WR,UN}LCK

From: John L. Hammond Define new constants LL_LEASE_{RD,WR,UN}LCK for use as the argument to and return value from the LL_IOC_{GET,SET}_LEASE ioctls. As arguments, these contants replace the use of F_{RD,WR,UN}LCK from fcntl.h. As return values they replace the use of

[PATCH 017/124] staging: lustre: lov: adjust page bufsize after layout change

From: Jinshan Xiong Otherwise, the coh_page_bufsize keeps increasing when the file's layout keeps changing in lov_init_raid0(). Signed-off-by: Jinshan Xiong Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-5459 Reviewed-on:

[PATCH 022/124] staging: lustre: llog: add newly opened llog at tail of handle list

From: Li Xi Add newly opened llog handle at the tail of handle list to increase lookup speed, especially for cancel operation, because the canceled log will be removed from the list, and lookup is from the beginning. Signed-off-by: Lai Siyao Signed-off-by: Li

[PATCH 008/124] staging: lustre: obd: add rnb_ prefix to struct niobuf_remote members

From: John L. Hammond Add the prefix rnb_ to the members of struct niobuf_remote. Delete the relevant compat macros from ofd_internal.h. Signed-off-by: John L. Hammond Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-5061 Reviewed-on:

[PATCH 023/124] staging: lustre: mdc: Report D_CHANGELOG messages as D_HSM

From: Henri Doreau Removed the D_CHANGELOG pseudo-debug flag that wasn't actually defined as a usable one. Report the D_CHANGELOG messages as D_HSM ones instead since this is the primary user of these messages. Signed-off-by: Henri Doreau Intel-bug-id:

[PATCH 116/124] staging: lustre: llite: make default_easize writeable in /sysfs

From: Ned Bass Allow default_easize to be tuned via /sysfs. A system administrator might want this if a rare access to widely striped files drives up the value on a filesystem where narrowly striped files are the more common case. In practice, however, this is wanted primarily to

[PATCH 014/124] staging: lustre: lmv: release locks if lmv_intent_lock() fails

From: John L. Hammond In lmv_intent_lock() if we will return an error then first release any locks referenced by the intent. Signed-off-by: John L. Hammond Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-5431 Reviewed-on:

[PATCH 122/124] staging: lustre: lnet: Enable setting per NI peer_credits

From: Doug Oucharek The code to allow peer_credits to be set per NI was originally "left inactive" because there were concerns about peer_credits interfering with the ability for IB nodes to connect to each other when peer_credits are not the same (peer_credits

[PATCH 123/124] staging: lustre: o2iblnd: Put back work queue check previously removed

From: Doug Oucharek The previous patch, http://review.whamcloud.com/21304/, removed a check needed until LU-5718 is properly addressed. With the check, LU-5718 results in an error message and a lost RDMA operation. Without it, we have memory corruption and a crash

[PATCH 001/124] staging: lustre: llite: fix ll_statahead_thread() problems on failure

From: Li Xi When ll_prep_md_op_data() fails, ll_statahead_thread() does not release reference of sai and parent dentry. It does not wake up parent thread either. This patch fixes these problems. Signed-off-by: Li Xi Intel-bug-id:

[PATCH 005/124] staging: lustre: obdclass: optimize busy loop wait

From: Alexander Zarochentsev two optimizations in the busy loop wait in class_cleanup(): (1) remove unnecessary spinlocks protecting obd_conn_inprogress access (2) replace cond_resched by more suitable yield to unconditionally yield the CPU if the

Re: [PATCH] staging: lustre: lmv: add missing function declaration

On Sunday, September 18, 2016 8:43:18 PM CEST James Simmons wrote: > > Nak. I'm not seeing this error with W=1. Also if you look > lmv_pack_md() is only used in lmv_obd.c and the function > appears early in the file before it is used, so no prototype > missing errors should happen. So it should

[PATCH 119/124] staging: lustre: lnet: potential deadlock in lnet

From: Quentin Bouget Fixes potential deadlock in LNetMDAttach Signed-off-by: Quentin Bouget Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-8249 Reviewed-on: http://review.whamcloud.com/20676 Reviewed-by: Doug Oucharek

[PATCH 108/124] staging: lustre: grant: quiet message on grant waiting timeout

From: Johann Lombardi Use at_max in osc_enter_cache() to bound how long we wait for grant space before switching to synchronous I/Os. Do not print a message on the console when the timeout is hit since such long wait can be legitimate with flaky network (i.e. BRW is

[PATCH 120/124] staging: lustre: lnet: check if ni is in current net namespace

From: Sebastien Buisson Add new 'ni_net_ns' field to struct lnet_ni to hold a reference to original net namespace in which ni is created. In LNetDist(), check if ni was created in same net namespace as current's one. If not, assign order above 0x, to make this ni not a

[PATCH 121/124] staging: lustre: lnet: Ensure routing is turned on first time

From: Doug Oucharek In lnet_rtrpools_enable(), a mistake was made and routing was not being turned on when the rtrpools are being allocated for the first time. This patch fixes that routine so we remember to turn on routing after allocating the rtrpools.

[PATCH 110/124] staging: lustre: obdclass: remove unnecessary EXPORT_SYMBOL

From: frank zago A lot of symbols don't need to be exported at all because they are only used in the module they belong to. Signed-off-by: frank zago Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-5829 Reviewed-on: http://review.whamcloud.com/13323

[PATCH 118/124] staging: lustre: lmv: fix parent FID for migration

From: wang di If the migrating directory is under striped directory, it needs to set right stripe FID for its parent. Signed-off-by: wang di Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-6263 Reviewed-on: http://review.whamcloud.com/13817

[PATCH 114/124] staging: lustre: ptlrpc: prevent request timeout grow due to recovery

From: Mikhail Pershin Patch fixes the issue seen on the client with growing request timeout which occurred after the server side patch landed for LU-5079. While commit itself is correct, it reveals another issue. If request is being processed for a long time on server

[PATCH 112/124] staging: lustre: ptlrpc: remove unnecessary EXPORT_SYMBOL

From: frank zago A lot of symbols don't need to be exported at all because they are only used in the module they belong to. Signed-off-by: frank zago Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-5829 Reviewed-on: http://review.whamcloud.com/12510

[PATCH 124/124] staging: lustre: update version to 2.6.99

With all of the the missing patches from the lustre 2.7 version merged upstream its time to update the upstream clients version. Signed-off-by: James Simmons --- drivers/staging/lustre/lustre/include/lustre_ver.h |6 +++--- 1 files changed, 3 insertions(+), 3

[PATCH 115/124] staging: lustre: mdt: add indexing option to default dir stripe

From: wang di Add indexing option to default dirstripe EA. If MDT find out the client send the create req to the wrong MDT because of default stripeEA, it will return -EREMOTE, then client will retrieve default stripeEA through xattr cache, and re-create the object. Also

[PATCH 096/124] staging: lustre: osc: osc_object_ast_clear() LBUG

From: Bobi Jam An OSC object could be destroyed with AGL locks waiting for granted, so we'd get rid of the osc_object_ast_clear() assertion that its dlm locks all getting granted. Signed-off-by: Bobi Jam Intel-bug-id:

[PATCH 106/124] staging: lustre: client: Fix mkdir -i 1 from DNE2 client to DNE1 server

From: Artem Blagodarenko After DNE phase 2 has been added to client it sends create request to slave MDT. DNT1-only server doesn't expect request to slave MDT from client. It expects only cross-mdt request from master MDT. Thus if DNE2 client tries to "mkdir -i

[PATCH 039/124] staging: lustre: remove lustre_lite.h

From: John L. Hammond Move several definition only used in lustre/llite/ to lustre/llite/llite_internal.h. Remove lustre/include/{,linux/}lustre_lite.h and fixup the missing includes in other headers that this exposes. Signed-off-by: John L. Hammond

[PATCH 061/124] staging: lustre: obd: cleanup struct md_op_data and uses

From: John L. Hammond Make the following changes in or around struct md_op_data: * rename enum op_cli_flags to enum md_cli_flags. * Change to type of the op_flags member from __u32 to enum md_op_flags. * Remove the used but never set member op_npages. * Remove

[PATCH 091/124] staging: lustre: clio: add coo_getstripe interface

From: Bobi Jam Use cl_object_operations::coo_getstripe() to handle LL_IOC_LOV_GETSTRIPE ops. Signed-off-by: Bobi Jam Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-5823 Reviewed-on: http://review.whamcloud.com/12452 Reviewed-by: John L. Hammond

[PATCH 117/124] staging: lustre: mdc: cl_default_mds_easize not refreshed

From: Ned Bass The client_obd::cl_default_mds_easize field should track the largest observed EA size advertised by the MDT, subject to a reasonable upper bound. The MDC uses cl_default_mds_easize to calculate the initial size of request buffers. The default value should be small

[PATCH 098/124] staging: lustre: lprocfs: cleanup stats locking code

From: Andreas Dilger Add comment blocks on lprocfs_stats_lock() and lprocfs_stats_unlock(). Move common NOPERCPU code out of the switch() statements to reduce code size and complexity, since it doesn't depend on the opc at all. Replace switch() in

[PATCH 097/124] staging: lustre: osc: change cl_extent_tax and *grants to unsigned

From: Dmitry Eremin Change the type accordant usage and remove warnings. Signed-off-by: Dmitry Eremin Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-5577 Reviewed-on: http://review.whamcloud.com/12386 Reviewed-by: James Simmons

[PATCH 101/124] staging: lustre: libcfs: check mask returned by cpumask_of_node

From: Liang Zhen cpumask_of_node can return NULL if NUMA node is unavailable, in this case cfs_node_to_cpumask will try to copy from NULL and cause kernel panic. Signed-off-by: Liang Zhen Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-5751

[PATCH 113/124] staging: lustre: obd: use proper flags for call_usermodehelper

When a parameter is permanently changed on the MGS the MGS send a changelog packet to the proper nodes that are affected by the change. Once the nodes receive the change they then call the userland utility lctl to change its local value. When calling a userland application from the kernel you

[PATCH 105/124] staging: lustre: clio: pass fid for OST setattr

From: Bobi Jam Store inode's fid in cl_setattr_ost() and OSC packs this info on the wire (via lustre_set_wire_obdo) so that OST can use. NOTE: currently lu_fid::f_ver and obdo::o_parent_ver are not used on OFD device, and we use obdo::o_stripe_idx as

[PATCH 045/124] staging: lustre: lmv: remove dead code

From: Dmitry Eremin The member lmv_obd->server_timeout and function lmv_set_timeouts() are not used. Signed-off-by: Dmitry Eremin Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-991 Reviewed-on: http://review.whamcloud.com/11880

[PATCH 111/124] staging: lustre: llite: lock the inode to be migrated

From: wang di Because the inode and its connected dentries will be cleared out of the cache after migration, the inode needs to be locked during the migration. Signed-off-by: wang di Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-4712 Reviewed-on:

[PATCH 100/124] staging: lustre: obd: change type of cl_conn_count to size_t

From: Dmitry Eremin Change type of cl_conn_count to size_t. Signed-off-by: Dmitry Eremin Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-5577 Reviewed-on: http://review.whamcloud.com/13125 Reviewed-by: James Simmons

[PATCH 102/124] staging: lustre: remove lustre/include/linux/

From: John L. Hammond Merge the contents of lustre/include/linux/lvfs.h into lustre/include/lvfs.h. Merge lustre/include/linux/lustre_user.h into lustre/include/lustre/lustre_user.h. Move lustre_compat25.h and lustre_patchless_compat.h from lustre/include/linux/ to

[PATCH 109/124] staging: lustre: misc: remove unnecessary EXPORT_SYMBOL

From: frank zago A lot of symbols don't need to be exported at all because they are only used in the module they belong to. Signed-off-by: frank zago Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-5829 Reviewed-on: http://review.whamcloud.com/13321

[PATCH 099/124] staging: lustre: llite: unlock inode size in ll_lov_setstripe_ea_info()

From: John L. Hammond In ll_lov_setstripe_ea_info() release the inode size lock on all appropriate exit paths. Signed-off-by: John L. Hammond Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-6059 Reviewed-on: http://review.whamcloud.com/13167

[PATCH 076/124] staging: lustre: ldlm: revert the changes for lock canceling policy

From: Jinshan Xiong The changes for LRU lock policy was introduced by commit bfae5a4e, where I was trying to revise the policy to pick locks for canceling. However, this caused two problems as mentioned in LU-5727. The first problem is that the lock can only be picked

[PATCH 026/124] staging: lustre: statahead: statahead thread wait for RPCs to finish

From: Lai Siyao Statahead thread should wait for inflight stat RPCs to finish in case statahead RPC callback may access data allocated in statahead thread context. ll_sa_entry_fini() should keep old entry if stat RPC is not finished yet. Simplify sai refcounting: * newly

[PATCH 087/124] staging: lustre: ptlrpc: Suppress error message when imp_sec is freed

From: Amir Shehata There is a race condition on client reconnect when the import is being destroyed. Some outstanding client bound requests are being processed when the imp_sec has alread been freed. Ensure to suppress the error message in import_sec_validate_get() in

[PATCH 094/124] staging: lustre: lov: flatten struct lov_stripe_md

From: John L. Hammond Flatten out the lsm_wire struct from the middle of struct lov_stripe_md and remove the member name macros. Signed-off-by: John L. Hammond Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-5814 Reviewed-on:

[PATCH 090/124] staging: lustre: obdclass: change cl_fault_io->ft_nob to size_t

From: Dmitry Eremin Change the type accordant usage. Signed-off-by: Dmitry Eremin Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-5577 Reviewed-on: http://review.whamcloud.com/12380 Reviewed-by: John L. Hammond

[PATCH 081/124] staging: lustre: obdclass: lu_htable_order() return type to long

From: Dmitry Eremin Change the type accordant usage. Signed-off-by: Dmitry Eremin Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-5577 Reviewed-on: http://review.whamcloud.com/12385 Reviewed-by: Bob Glossman

[PATCH 088/124] staging: lustre: ldlm: Recalculate interval in ldlm_pool_recalc()

From: Nathaniel Clark Instead of rechecking a static value, recalculate to see if pool stats need to be updated. Add newline so message will print instead of warning about missing newline. Signed-off-by: Nathaniel Clark Intel-bug-id:

[PATCH 095/124] staging: lustre: ptlrpc: fix race between connect vs resend

From: Alexander Boyko Buggy code at ptlrpc_connect_interpret() finish: rc = ptlrpc_import_recovery_state_machine(imp); ... Set import connection flags When import has FULL state ptlrpc_import_recovery_state_machine() wakeup all waiters on import and all

[PATCH 047/124] staging: lustre: llite: enforce pool name length limit

From: Li Xi The pool related codes have some inconsistency about the length of pool name. Creating and setting a pool name of length 16 to a directory will succeed. However, creating a file under that directory will fail. This patch disables any pool name which is longer or equal

[PATCH 092/124] staging: lustre: ptlrpc: fix comparison between signed and unsigned

From: Dmitry Eremin Change return type and size argiments of lustre_msg_hdr_size(), lustre_msg_buf{len,count}() and req_capsule_*_size() to __u32. Change type of req_format->rf_idx and req_format->rf_fields.nr to size_t. Also return zero for incorrect message magic

[PATCH 031/124] staging: lustre: vvp: Use lockless __generic_file_aio_write

From: Prakash Surya Testing multi-threaded single shard file write performance has shown the inode mutex to be a limiting factor when using the generic_file_write_iter function. To work around this bottle neck, this change replaces the locked version of that call with the lock

[PATCH 072/124] staging: lustre: echo: replace lov_stripe_md with lov_oinfo

From: John L. Hammond In echo_client replace uses of struct lov_stripe_md with struct lov_oinfo (since the instances of the former really only contained a single instance of the latter). Remove the then unneccessary functions echo_alloc_memmd(), echo_free_memmd(),

[PATCH 085/124] staging: lustre: obdclass: change loop indexes to unsigned

From: Dmitry Eremin Cleanup warnings about comparison between signed and unsigned. Signed-off-by: Dmitry Eremin Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-5577 Reviewed-on: http://review.whamcloud.com/12387 Reviewed-by: Bob Glossman

[PATCH 089/124] staging: lustre: obd: change brw_page->count to unsigned

From: Dmitry Eremin Pages count is unsigned. So, change the type accordant usage. Signed-off-by: Dmitry Eremin Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-5577 Reviewed-on: http://review.whamcloud.com/12378 Reviewed-by: John L. Hammond

[PATCH 059/124] staging: lustre: at: net AT after connect

From: Alexander Boyko Once connected, the previously gathered AT statistics is not valid anymore because may reflect other routing, etc. The connect by itself could take a long time due to different reasons (e.g. server was not ready) and net latency got very high

[PATCH 044/124] staging: lustre: misc: Reduce exposure to overflow on page counters.

From: Stephen Champion When the number of an object in use or circulation is tied to memory size of the system, very large memory systems can overflow 32 bit counters. This patch addresses overflow on page counters in the osc LRU and obd accounting. Signed-off-by: Stephen

[PATCH 084/124] staging: lustre: fiemap: set FIEMAP_EXTENT_LAST correctly

From: Bobi Jam When we've collected enough extents as user requested, we'd check one further to decide whether we've reached the last extent of the file. Signed-off-by: Bobi Jam Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-5933 Reviewed-on:

[PATCH 093/124] staging: lustre: ldlm: move LDLM_GID_ANY to lustre_dlm.h

From: Jinshan Xiong lustre_idl.h only includes wire data; lustre_dlm.h is the right place for LDLM_GID_ANY. Signed-off-by: Jinshan Xiong Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-6028 Reviewed-on: http://review.whamcloud.com/13074

[PATCH 078/124] staging: lustre: llog: prevent out-of-bound index

From: frank zago llog_process_thread() can be called from llog_cat_process_cb with an index already out of bound, leading to the following crash: LustreError: 3773:0:(llog.c:310:llog_process_thread()) ASSERTION(index <= last_index + 1 ) failed: LustreError:

[PATCH 083/124] staging: lustre: ldlm: evict clients returning errors on ASTs

From: Alexey Lyashkov To test proper behavior of clients returning errors on ASTs we can induce a failure with setting OBD_FAIL_LDLM_BL_CALLBACK_NET. Handle the new additonal case of cfs_fail_err being set as well so that the cfs_fail_err can be sent back in a reply.

[PATCH 075/124] staging: lustre: recovery: don't replay closed open

From: Niu Yawei To avoid scanning the replay open list every time in the ptlrpc_free_committed(), the fix of LU-2613 (4322e0f9) changed the ptlrpc_free_committed() to skip the open list unless the import generation is changed. That introduced a race which could make a closed

[PATCH 030/124] staging: lustre: llite: Replace write mutex with range lock

From: Prakash Surya Testing has shown the ll_inode_inode's lli_write_mutex to be a limiting factor with single shared file write performance, when using many writing threads on a single machine. Even if each thread is writing to a unique portion of the file, the lli_write_mutex

[PATCH 086/124] staging: lustre: obdclass: eliminate NULL error return

From: Bob Glossman Always return an ERR_PTR() on errors, never return a NULL, in lu_object_find_slice(). Also clean up callers who no longer need special case handling of NULL returns. Signed-off-by: Bob Glossman Intel-bug-id:

[PATCH 049/124] staging: lustre: lmv: move some inline functions to lustre_lmv.h

From: Fan Yong Move some inline code out of lmv core into lustre_lmv.h. This is to prepare for use outside of the lmv layer in the future of these functions. Change from passing in struct lmv_stripe_md to just int for lmv_is_known_hash_type. Signed-off-by: Fan Yong

[PATCH 054/124] staging: lustre: llite: Add ioctl to get parent fids from link EA.

From: Henri Doreau Added LL_IOC_GETPARENT to retrieve the /name(s) of a given entry, based on its link EA. This saves multiple calls to path2fid/fid2path. Merged with second later patch that does various cleanups. Avoid unneeded allocation. Get read-only attributes from the

[PATCH 080/124] staging: lustre: llite: fix dup flags names

From: Bob Glossman The name 'xattr' is used for two different ll_flags bits. Change the names to be distinct and different, reflecting the names of the bits used in LL_SBI_xbitnamex #defines. Signed-off-by: Bob Glossman Intel-bug-id:

[PATCH 079/124] staging: lustre: mgc: add nid iteration

From: Alexander Boyko mgc_apply_recover_logs use only first nid from entry, this could be the problem for a cluster with several network address for a one node. Signed-off-by: Alexander Boyko Intel-bug-id:

[PATCH 077/124] staging: lustre: ptlrpc: quiet errors on initial connection

From: Andreas Dilger It may be that a client or MDS is trying to connect to a target (OST or peer MDT) before that target is finished setup. Rather than spamming the console logs during initial connection, only print a console error message if there are repeated

[PATCH 046/124] staging: lustre: llite: handle concurrent use of cob_transient_pages

From: Stephen Champion With the lockless __generic_file_aio_write introduced in LU-1669, ll_direct_IO_26 is no longer protected by the inode i_isem. This renders obsoltete checks that all transient pages have been handled before and after entry, and requires atomic access to

[PATCH 073/124] staging: lustre: llite: remove ll_objects_destroy()

From: John L. Hammond Remove ll_objects_destroy(). This function is not needed for interoperability with servers of version 2.4 or higher. Remove the then unused function lov_destroy() and its supporting functions. Remove the lsm_destroy method of struct lsm_operations.

[PATCH 051/124] staging: lustre: llite: ensure all data flush out when umount

From: Yang Sheng Write out all extents when clear inode. Otherwise we may lose data while umount. Signed-off-by: Yang Sheng Signed-off-by: Jinshan Xiong Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-5584 Reviewed-on:

[PATCH 040/124] staging: lustre: obd: rename LUSTRE_STRIPE_MAXBYTES

From: John L. Hammond Rename LUSTRE_STRIPE_MAXBYTES to LUSTRE_EXT3_STRIPE_MAXBYTES and correct the comment describing its use. Signed-off-by: John L. Hammond Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-2675 Reviewed-on:

[PATCH 082/124] staging: lustre: mdc: Proper accessing struct lov_user_md

From: Yoshifumi Uemura In mdc_setattr_pack() access the members of struct lov_user_md by little endian byte order. Signed-off-by: Yoshifumi Uemura Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-5889 Reviewed-on: http://review.whamcloud.com/12683

[PATCH 034/124] staging: lustre: llite: turn mode to umode_t for ll_new_inode()

From: John L. Hammond Change int mode to umode_t. Signed-off-by: John L. Hammond Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-2675 Reviewed-on: http://review.whamcloud.com/10769 Reviewed-by: Yang Sheng Reviewed-by:

[PATCH 066/124] staging: lustre: lov: remove LL_IOC_RECREATE_{FID, OBJ}

From: John L. Hammond Remove the obsolete ioctls LL_IOC_RECREATE_FID and LL_IOC_RECREATE_OBJ along with their handlers in llite. Remove the then unused OBD method lov_create(). Remove OBD_FL_RECREATE_OBJS handling from osc_create(). Signed-off-by: John L. Hammond

[PATCH 033/124] staging: lustre: llite: remove mode from ll_create_it()

From: John L. Hammond Remove the unused mode parameter from ll_create_it(). Signed-off-by: John L. Hammond Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-2675 Reviewed-on: http://review.whamcloud.com/10769 Reviewed-by: Yang Sheng

[PATCH 018/124] staging: lustre: obdclass: fix comparison between signed and unsigned

From: Dmitry Eremin Make lu_buf->lb_len unsigned. Signed-off-by: Dmitry Eremin Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-5417 Reviewed-on: http://review.whamcloud.com/11281 Reviewed-by: Fan Yong Reviewed-by: John

[PATCH 000/124] missing patches from Lustre 2.7 release

This batch of patches includes various cleanups and fixes from the Lustre 2.7 release. Most of these patches are removal of cruft and unsigned to signed comparsion fixes. Also added in the latest LNet fixes as well. Alexander Boyko (3): staging: lustre: at: net AT after connect staging:

[PATCH 070/124] staging: lustre: statahead: small fixes and cleanup

From: Lai Siyao small fixes: * when 'unplug' is set for ll_statahead(), sa_put() shouldn't kill the entry found, because its inflight RPC may not finish yet. * remove 'sai_generation', add 'lli_sa_generation' because the former one is not safe to access without lock.

[PATCH 032/124] staging: lustre: llite: remove lookup_flags from ll_lookup_it()

From: John L. Hammond Remove the effectively unused lookup_flags parameter from ll_lookup_it(). Signed-off-by: John L. Hammond Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-2675 Reviewed-on: http://review.whamcloud.com/10769 Reviewed-by:

[PATCH 041/124] staging: lustre: llite: don't call make_bad_inode() on an old inode

From: John L. Hammond In ll_iget() if ll_update_inode() fails then do not call make_bad_inode() on the inode since it may still be in use. Signed-off-by: John L. Hammond Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-5468 Reviewed-on:

[PATCH 058/124] staging: lustre: statahead: race in start/stop statahead

From: Lai Siyao When starting statahead thread, it should check whether current lli_opendir_key was deauthorized in the mean time by another process. Signed-off-by: Lai Siyao Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-3270 Reviewed-on:

[PATCH 067/124] staging: lustre: changelog: fix comparison between signed and unsigned

From: Dmitry Eremin Change type of changelog_*{namelen,size}() to size_t. Fixed string specifier for unsigned types. Signed-off-by: Dmitry Eremin Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-5577 Reviewed-on:

[PATCH 068/124] staging: lustre: lov: remove unused {get, set}_info handlers

From: John L. Hammond In LOV and OSC remove handlers for the obsolete get and set info keys: KEY_CAPA_KEY, KEY_CONNECT_FLAG, KEY_EVICT_BY_NID, KEY_LAST_ID, KEY_LOCK_TO_STRIPE, KEY_MDS_CONN, KEY_NEXT_ID. Signed-off-by: John L. Hammond

[PATCH 063/124] staging: lustre: ldlm: count of pools is unsigned long

From: Dmitry Eremin Function ldlm_pools_count() return unsigned long but counter is int. Use ldlm_pool_granted() everywhere. Signed-off-by: Dmitry Eremin Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-5577 Reviewed-on:

[PATCH 062/124] staging: lustre: replace direct HZ access with kernel APIs

From: Jian Yu On some customer's systems, kernel was compiled with HZ defined to 100, instead of 1000. This improves performance for HPC applications. However, to use these systems with Lustre, customers have to re-build Lustre for the kernel because Lustre directly uses the

[PATCH 069/124] staging: lustre: fix messages with missing newlines

From: John L. Hammond Restore the trailing newline in the definition of OSC_DUMP_GRANT(). Remove an unnecessary CDEBUG() from ldlm_pool_recalc(). Signed-off-by: John L. Hammond Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-5551 Reviewed-on:

[PATCH 056/124] staging: lustre: statahead: use dcache-like interface for sa entry

From: Lai Siyao Rename ll_sa_entry to sa_entry, and manage sa_entry cache with dcache-like interfaces. sa_entry is not needed to be refcounted, because only scanner can free it, so after it's put in stat list, statahead thread shouldn't access it any longer.

[PATCH 060/124] staging: lustre: mdc: fix comparison between signed and unsigned

From: Dmitry Eremin Change type of client_obd->*_mds_*size from int to __u32 and argumanets of related create/rename/setattr functions. Change type of op_data->op_namelen to size_t. Change type of argument size for all mdc_*_pack() to size_t. Signed-off-by: Dmitry

[PATCH 016/124] staging: lustre: fix comparison between signed and unsigned

From: Dmitry Eremin Cleanup in general headers. * use size_t in cfs_size_round*() * make unsigned index and len in lustre_cfg_*() * make iteration variable the same type as comparing value * make unsigned pages counters Signed-off-by: Dmitry Eremin

[PATCH 064/124] staging: lustre: lu_dirent_calc_size() return type to size_t

From: Dmitry Eremin Change the type accordant usage. Signed-off-by: Dmitry Eremin Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-5577 Reviewed-on: http://review.whamcloud.com/12383 Reviewed-by: John L. Hammond

[PATCH 050/124] staging: lustre: ldlm: per-export lock callback timeout

From: Vitaly Fertman The lock callback timeout is calculated as an average per namespace. This does not reflect individual client behavior. Instead, we should calculate it on a per-export basis. This is the client side changes for upstream client. Signed-off-by:

[PATCH 055/124] staging: lustre: llite: allow setting stripes to specify OSTs

From: Jinshan Xiong Extend the llite layer to support specifying individual target OSTs. Support specifying OSTs for regular files only. Directory support will be implemented later in a separate project. With this a file could have for example a OST index layout of

  1   2   >