1. Adding iput(cache_inode) in smfs_clear_inode for clearing cache inode according to peter's advice. 2. other minor fixes in smfs_clear_inode

Andreas: I tested that OST will never grant more space to liblustre than
the initial 4M.

writepage.
patch=zab

Adds support for client-side write cache accounting via OST write credits
(grants).  There is not yet a "correction" algorithm for grants if they
get out of sync between the client and OST (OST can handle clients who think
they have grant but do not).  Peter had previously suggested that grants
should decay over time (faster on the clients than on the OSTs) so idle
nodes do not consume space they aren't using.

Use /proc/fs/lustre/osc/OSC*/cur_grant_bytes (and existing cur_dirty_bytes)
to check client-side cache values.

Includes lustre-side changes for lfsck support (create files with
specific objids, create specific objids on OSTs).

Also includes some changes to the testing scripts (local.sh, lov.sh, acc-sm)
so that they include a "client nid '*'" client and can be used for sanityN.sh
testing.  This has been put into the main acceptance-small.sh loop so that
sanityN.sh will be run on both single and multi-OST setups.

b=974 b=2349
r=shaver

r=shaver

get lconf to panic() when no host entry is found. using panic() will prevent
lconf from exiting cleanly when this happens.

It is really the fault of K&R that there are 2 different open prototypes.

during OSC failure).

Make sure this request was replied before unpacking, and if not return
an error.

that my root filesystem doesn't fill up with all the different files.

r=adilger

have lconf create large journals

r=shaver

replace OBD_IOC_LIST ioctl with reading /proc/fs/lustre/devices

- record all the test commands that are run

passing "> /dev/null" as an extra argument to the functions, and this
is not really what we want.  Since I don't think it's all that bad to
see the output of lconf, anyway, I just deleted the >/dev/null
entirely.

While looking into this, I also noticed that mount -f was still
attempting to fetch the log from the MDS. That is fixed, too.