b=17644
i=green
i=adilger

send 1 extra rpc in flight if this is a high priority request

b=17153
i=johann, adilger

Remove 2.4 compatibility.

r=tappro,wangdi

- fixes ost cleanup issue due to missed llcd_put() in the case ost does not receive disconnect from mds;

  - do not sleep on hanging llcd. Instead assert on it _after_ stopping recov_thread's ptlrpcd which should kill any remeining llcds;

  - fixes and cleanups, comments.

r=tappro,johann

- implements proper locking for rq pool freeing

r=adilger,deen

- missed lustre_put_lsi() in couple of places;

- do not call deregister_mount() in mount error path, this makes it impossible for MDT to do put_mount() and thus, its lsi left not released;

- fixes error handling with llog_setup/llog_cleanup in couple of places;
- fixes error handling after hash_init errors in class_setup();
- cleanups.

Backport warning fixes from b1_8.
i=girish (original patch)
i=robert

-   fixed a typo in error message.

-   fixed a typo in error message.

Branch b1_6
b=17059
i=ZhiYong Tian
i=Johann Lombardi

Always build client with quota support

i=nathan
i=adilger
OST pools on HEAD, comprehensive patch including 17054:19007;
16935:18918,19012,19089,19128; 16978:18872

b=15949
i=adilger
i=shadow

mds_body.capability fix

Quiet compiler warnings about mismatched signed/unsigned comparisons.

Branch HEAD

Branch b1_6

request will resend before finished ost llog init and this produce
access to non init llog context.

Branch HEAD
b=16839
i=adilger
i=rread

request will resend before finished ost llog init and this produce
access to non init llog context.

Branch b1_6
b=16839
i=adilger
i=rread

or write request because the client is already doing a lock match
itself so there isn't any reason the OST should have to re-do that
search.

b=15198
i=johann
i=shadow

b=16573
o=adilger
i=johann, nathan

Description: Export bytes_read/bytes_write count on OSC/OST.

b=16573
o=adilger
i=johann, nathan

Description: Export bytes_read/bytes_write count on OSC/OST.

b=10555
i=adilger
i=green
i=nikita
i=girish

add ldiskfs patches and lustre support for FIEMAP ioctl

b=10555
i=adilger
i=green
i=girish

add ldiskfs patches and lustre support for FIEMAP ioctl

b=16098

Add URL for GPLv2 license in copyright headers

b=16098

Add GPLv2 URL to copyright headers

b=16317
o=Herb Wartens(hwartens@llnl.gov)
i=wangdi, johann

Description: exports in /proc are broken
Details    : recreate /proc entries for clients when they reconnect.

b=16227

i=adilger
i=shadow

add lsm argument to obd_get_info. For some get_info calls the lsm was being sent as part of the key which was a hack. Now lsm can be sent as an argument.

b=16227

i=adilger
i=shadow

add lsm argument to obd_get_info. For some get_info calls the lsm was being sent as part of the key which was a hack. Now lsm can be sent as an argument.

b=15981
i=adilger
i=robert.read

b=16098

change CFS license headers to Sun headers

 b=16098

  Replace CFS license headers with Sun headers

b=16368
i=adilger
i=alex.zhuravlev

- put object group in ldlm_res_id.name[1]

b=15981
i=adilger
i=rread

CONFIG_PAGE_ALLOC enabled

Branch HEAD

b=12975
i=green
i=adilger

b=14010
i=Mike Pershin (tappro)
i=Andreas Dilger (adilger)

Doxygen comments (the missing ones)

b=15516
i=tom.wang
i=yury.umanets

This patch is first step towards shifting mds-lov work to new
lustre mds stack.
        -       kill OBD_CLEANUP_SELF_EXP
        -       added rw_sem to serialize mds_notify and mds_precleanup
        -       kill OBD_CLEANUP_OBD

port Adaptive Timeouts from b1_6 to HEAD.
b=14071
r=nathan
r=adilger

sync import destroy with llog thread/lprocfs access
b=14629
i=robert,jay

sync procfs/llog_thread with import destory
b=15684
i=Robert, Jay

1.Filter group llog cleanup (patch from shadow).
2.Separate filter group llog create and find. then filter group llog will
  only be created in the mds/ost syncing process.
3.Some minor fixes.
b=14629
i=robert,yury

b=15940
i=isaac
i=oleg

Convert kp30.h to an internal libcfs include and change users of kp30.h to
include libcfs.h if they were not already.

b=15950
i=wangdi
i=shadow

The direct IO path doesn't call check_rpcs to submit a new RPC once
one is completed. As a result, some RPCs are stuck in the queue
and are never sent.
Merge brw_interpret() and brw_interpret_oap().