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

b=16247
i=sheng.yang
i=johann

Allow liblustre sanity to exclude tests

b=15552
i=adilger
i=nikita.danilov

- mapping ino for FID into [2B, 4B]

fix KEY_IS macro

Branch HEAD
b=14032
i=umka
i=deen

b=15797
i=huanghua
i=vitaly

Misc fix for updating file {a/m/c}time as following policy:
1) Lustre file "{a/m/c}time" should base on client-side clock, including all
the file operations.
2) All the client-side clocks should be synchronized. It is no matter whether
the server-side clocks are synchronized with client-side ones or not.
3) File "ctime" is increased only.
4) File "{a,m}time" are increased only, except "utime()" operation can change
them backwards.

Quiet compiler warnings.

r=tappro,fanyong
- fixes unaligned access on 64 bit platforms. Thanks to Nikita for these patches.

b=15969
i=isaac
i=adilger

Move libcfs to its own module.

This removes libcfs from lnet and changes the build system
to use the new libcfs module.

this allows us to quickly find what pages to write and remove
once lock callback is received.

b=10718
i=green
i=johann

b=15851
i=zam

b=14929
i=zam
i=nikita

r=Nathan Rutman, Alexey Lyashkov

Ported the patch 'to organize the stats on per-nid basis instead of the existing per-export basis' to HEAD.
This patch also includes all its related bug fixes(bug 14016, bug 14370, bug 14873).

i=johann
i=alexey.lyashkov

fix error building lustre/liblustre/tests for Cray XT3 catamount

i=adilger
i=rread

fixes error linking lustre/liblustre/tests/echo_test during catamount
build of b1_6 for Cray XT3 systems

b=14929
i=zam

b=15517
i=johan
i=deen

sys/syscall.h exist at all os'es instead of <syscall.h> which exist only
at linux.
b=14929
i=nikita

i=yury.umanets
i=adilger

this patch extends lu_dirent->lde_hash element, which is required for dmu based mdd.
but I am keeping MEA_MAGIC_HASH_SEGMENT hash within 32 bit.

obd objects and this also block reconnect with -EALREADY.
This isn't fix all lprocfs bugs - but make it rare.
b=14533
i=adilger
i=nathan

b=13805
i=ericm
i=bobijam

add support for several checksum algorithms. Currently, CRC32 and
Adler-32 are supported. The checksum type can be changed on the fly
through /proc/fs/lustre/osc/*/checksum_type.

support other encrypt/checksum algorithms for bulk i/o, default in gss
mode is sha1/aes128, for others default enable bulk checksum by adler32.
b=13667
r=yong.fan
r=h.huang
r=adilger

b=13821
i=wangdi
i=umka

r=adilger

wait for async rpc completion in liblustre

b=14149
a=15319
i=yong.fan
i=rahul.deshmukh

add OBD_CONNECT_FID support for liblustre connection.

i=yong.fan
i=rahul.deshmukh
i=nikita.danilov

- use req_capsule interface for client.
- add some interoperability support on server side.

o=Bob Glossman(bogl@cray.com)
i=adilger

fix building errors for xt3 in SLES10.

b=9977
i=umka
i=deen

b=13375
i=adilger
i=johann
i=nathan

Descriptoin: make lov_create() will not stuck in obd_statfs_rqset()
Details    : If an OST is down the MDS will hang indefinitely in
             obd_statfs_rqset() waiting for the statfs data. While for
             MDS QOS usage of statfs, it should not stuck in waiting.

b=13317
r=shadow, johann

enable liblustre to connect to mds with acl

b=14064

- readdir() for liblustre: i=nikita,fanyong
- flock for liblustre and various fixes for liblustre: i=umka,fanyong

b=11791
i=adilger, nathan.rutman

Description: Inconsistent usage of lustre_pack_reply()
Details    : Standardize the usage of lustre_pack_reply() such that it
             always generate a CERROR on failure.

b=11622
i=adilger
i=johann

1. add page allocation statistics for lustre;
2. use percpu counters for both memory and page accounting;
3. update leak-finder to understand the new format of debug info.

i=adilger, deen

change the error message in llu_file_prwv()

b=13622
i=tappro
i=green

1) make class_handle_hash_back() to work;
2) ll_update_inode() checks UPDATE ibit lock still presents before setting LLIF_MDS_SIZE_LOCK;
3) ldlm_lock_match() returns the matched mode instead of 1; pass bitwise sets of modes into;

b=13497
i=deen
i=johann

Description: LASSERT_{REQ,REP}SWAB macros are buggy
Details    : If SWAB_PARANOIA is disabled, the LASSERT_REQSWAB and
             LASSERT_REPSWAB macros become no-ops, which is incorrect. Drop
             these macros and replace them with their difinitions instead.

  b:13816
  i:shadow,yury

b=13732
i=johann, liuy

change order of libsysio includes

i=adilger
i=green

b=13581
i=tappro
i=huanghua

Replay Size-on-MDS related RPCs properly.