i=Scjody
i=Adilger
multiop race fix

only ignore error code of 67a, not 67b.
b=13721
r=nathan

i=Adilger

rsh do_node fix

Increase build version to 1.6.4.52 for new tag.

Fix a relatively harmless buffer overlow in "lfs df".
b=15183
i=komal.nawandar
i=green

Make the Lustre ioctl.h header conditional upon _IOWR (which is defined
in every arch ioctl.h) instead of the arch-specific _ASMI386_IOCTL_H
symbol.

That is still #defined to avoid re-inclusion of the kernel header.
b=14746
i=bowen.zhou

b=15202

Remove test_27o and test_27q from ALWAYS_EXCEPT list.

add some do_facet in sanity-quota.sh
i=johann
i=andrew

b=12743

since we don't support diskfs blocksize != 4K right now, so change the test case to
make all the disk with 4K blocksize.

b=15010
o=green
i=vitaly
i=adilger

Description: mdc_set_open_replay_data LBUG
Details    : Set replay data for requests that are eligible for replay.

i=Adilger

test_78: limit filesize to 1/2 of memtotal -256MB

b=14763
r=bwzhou

dump the memory debugging after all modules are unloaded

b=13285

minor change of error message

- fix wrong patch in last commit according to Andreas's suggestion.

b=14538
i=eric.mei
i=yong.fan

- use lustre_msg_v2 for CONNECT

b=13285
r=johann, nathan

cid-290: fixes some oti_init to return earlier if NULL value

b=13285
r=johann, deen

Uses ldd to figure out which libs need to be copied and works for 64 bit ppc SLES9 systems

Fix "make rpms" due to missing file.
b=14321

improve test_18 and test_18a in sanity-quota.sh
b=14840
i=adilger
i=johann

b=14321
o=Brian Behlendorf(behlendorf1@llnl.gov)
i=bobijam
i=shadow

missed this part of the patch.

b=14850
i=eric.mei
i=adilger

set timeout value before mounting MDT/OST

b=14321
o=Brian Behlendorf(behlendorf1@llnl.gov)
i=bobijam
i=shadow

add MGS handler statistics.

b=14321
o=Brian Behlendorf(behlendorf1@llnl.gov)
i=bobijam
i=shadow

Description: lustre_mgs: operation 101 on unconnected MGS
Details    : When MGC is disconnected from MGS long enough, MGS will evict the
             MGC, and late on MGC cannot successfully connect to MGS and a lot
             of the error messages complaining that MGS is not connected.

r=Adilger
Changed recovery-small.sh to use 'lctl {get,set}_param'

Make description of bug 14775 more clear.

b=15036
i=johann
i=tianzy

update local vars types to handle 64-bit data

b=15036
i=johann
i=tianzy

update local vars types to handle with 64-bit limits

A few corrections in earlier patch to change sanity.sh to use 'lctl {get,set}_param'

Forget to initialize the li_group.

print the stripecount as a signed value

i=Adilger

test_23a fix

b=14483
detect stride IO mode in read-ahead.
i=adilger, nikita

r=Adilger
to solve RHEL5/ia64 build error related to 'lctl {get,set}_param' patch

b=14775
i=green
i=nathan
i=johann

b=11300
r=oleg,adilger

Fix interval tree issues at the customer's side.

b=13715
i=johann
i=nathan
i=adilger

protect accessing imp->generation w/ imp_lock.

b=12584
i=adilger

There is a issue with getfacl which just literally find the symbol link
target with `pwd`+[symbol content], regardless where the link file is
located.

This fix avoids the issue by chdir to the symbol link's dir before getting it's
acl list.

to have two identical connections in imp_conn_list. We must
compare not conn's pointers but NIDs, otherwise we can defeat
connection throttling.

b=14774
i=adilger
i=johann

b=14873
i=johann
i=shadow

protect competitive adding to obd->obd_nid_stats.

r=eeb,adilger

wait for async rpc completion in liblustre