Commit Graph

5916 Commits

Author SHA1 Message Date
Sage Weil
a69f197e44 todos 2009-02-25 13:07:23 -08:00
Sage Weil
e9f470e204 kclient: add module author 2009-02-25 12:49:48 -08:00
Sage Weil
6adae97e87 kclient: dup kobj_sysfs_ops from kobject.c
They're not exported, bleh.
2009-02-25 12:49:18 -08:00
Sage Weil
662afd5ea5 journal: initiate full commit when journal hits 50% full
We want to avoid filling up the journal, particularly since
_start_sync is a no-op with a journal enabled.
2009-02-25 12:38:32 -08:00
Sage Weil
9643b96035 filestore: don't initiate sync on _start_sync if there's a journal
Journal will commit immediately anyway.. don't slow the rest
down with a full commit to disk.
2009-02-25 12:37:28 -08:00
Sage Weil
34ccad598f osd: don't clobber log output stream
Don't print a null char pointer.
2009-02-25 12:20:43 -08:00
Sage Weil
96b3f36789 mon: only do big sync if we're flushing lots of states
The rest of the time, fsync them individually.
2009-02-25 12:19:54 -08:00
Sage Weil
dd7ff9dd4c kclient: include sysfs (and not proc) in kernel client patchset 2009-02-25 11:10:43 -08:00
Sage Weil
f5f3688830 osd: take journal on command line
./cosd dev -j journaldev ...
2009-02-25 11:05:55 -08:00
Sage Weil
c23235ea67 kclient: remove /proc/fs/ceph
/sys/fs/ceph now does everything it used to.
2009-02-25 10:12:06 -08:00
Sage Weil
fb0183da12 kclient: add other maps to sysfs 2009-02-25 10:12:05 -08:00
Sage Weil
783215b06a kclient: add debug levels, mask to /sys/fs/ceph 2009-02-25 10:12:05 -08:00
Sage Weil
1a0b99d5d4 kclient: some sysfs cleanup 2009-02-25 10:12:05 -08:00
Sage Weil
290b7d326d dstop: remove cosd hostname prefix 2009-02-24 16:58:29 -08:00
Sage Weil
34db3a1161 todos 2009-02-24 16:51:09 -08:00
Sage Weil
7d1af707e7 dstart: remove cosd hostname prefix 2009-02-24 16:51:04 -08:00
Sage Weil
909781dd2e kclient: consider issued AND implemented AND dirty when updating inode
We do not want to update the local inode field(s) if we
 * still hold EXCL
 * are having EXCL revoked (not issued, but still implemented)
 * are writing back EXCL (not issued or implemented, but dirty)
2009-02-24 16:37:55 -08:00
Sage Weil
7c60c042fb mds: flush cap update if affected lock is unstable 2009-02-24 16:33:46 -08:00
Sage Weil
df453b0e12 debian: include new binaries, scripts in ceph-common 2009-02-24 14:30:12 -08:00
Sage Weil
f825647e48 Makefile: mount.ceph goes in usr/sbin 2009-02-24 14:30:00 -08:00
Sage Weil
0324487023 debian: pull version from configure.ac 2009-02-24 14:21:25 -08:00
Sage Weil
e75b4fa269 debian: specify arch on command line 2009-02-24 14:13:37 -08:00
Sage Weil
e235eda1dd makefile: add missing files for kernel module 2009-02-24 14:08:19 -08:00
Sage Weil
a6282628f8 debian: script to build and upload debs to repo 2009-02-24 13:57:16 -08:00
Sage Weil
bc39a2d693 make unstable v0.7 2009-02-24 13:15:53 -08:00
Sage Weil
a8dbeb0fd5 common: uninit var 2009-02-24 12:33:21 -08:00
Sage Weil
b82705add1 debian: cleanup, changelog fix 2009-02-24 12:29:28 -08:00
Sage Weil
31fa2f81fc kclient: typo 2009-02-24 12:22:16 -08:00
Sage Weil
a5d747a874 mds: xlock filelock during truncate
This allows client to see projected truncation before it commits.
2009-02-24 12:21:45 -08:00
Sage Weil
04cb3f779e mds: look at active (not projected) inode when issuing truncation
We are interested in applying the truncation that just committed, not one
that is projected.
2009-02-24 11:55:58 -08:00
Sage Weil
39cd8b0ee3 mds: only do truncation process when file size goes down
For file size increase, no special action is required...
2009-02-24 11:55:23 -08:00
Sage Weil
4c3a7aa1a7 mds: limit propagation of full auth_pin count up hierarchy
We only care if there are _some_ auth pins, not the exact count.  So only
propagation whether there are any pings from dir -> inode, to avoid walking
all the way up the hierarchy for each pin/unpin event.
2009-02-24 11:52:56 -08:00
Sage Weil
8be929d451 kclient: don't remove from clients list if we're not on it 2009-02-23 13:46:46 -08:00
Sage Weil
151d5f550e Makefile: link libcommon, lockdep last
Link libcommon.a last for everything.

Include lockdep last in libcommon so to initialize first, shutdown
last, so that g_lockdep is initialized to zero for all Mutexes.

Put config.cc next to last so that the atomic_t thread counter is
initialized before all threads are created, shutdown after all
threads are destroyed.
2009-02-23 10:19:15 -08:00
Yehuda Sadeh
940a78ffc4 objecter: shouldn't swap bufferlist buffers 2009-02-20 14:49:44 -08:00
Yehuda Sadeh
e1fd643ba1 common: remove warning 2009-02-19 15:42:56 -08:00
Yehuda Sadeh
b5860ee361 common: virtual destructor for class Timer 2009-02-19 15:43:16 -08:00
Sage Weil
52b90a4c3f todo 2009-02-19 15:25:21 -08:00
Yehuda Sadeh
740da72fbd SafeTimer: getting rid of global timer 2009-02-19 15:25:11 -08:00
Sage Weil
06e396139b kclient: only unregister after both acks
Partially reverts behavior of b732c0ddb8

We need the request to stay in the request_tree so that when the
safe arrives we can look it up (duh).  Also, we need to keep a
reference to the request during that period...
2009-02-19 15:22:12 -08:00
Sage Weil
440a84fd87 mds: reverse rename argument order
Order change introduced by 3dc43aa8e6
2009-02-19 14:47:03 -08:00
Sage Weil
d083f8ea1d client: reverse rename argument order 2009-02-19 14:45:37 -08:00
Sage Weil
e92f0a60c1 kclient: add missing r_wait list_head init 2009-02-19 14:38:33 -08:00
Sage Weil
15b2d60d49 kclient: initialize request list_heads 2009-02-19 14:30:16 -08:00
Sage Weil
76ec1b6fbb kclient: cast path
path[12] will either point to freshly allocated paths (that we
later kfree) or a provided const path.
2009-02-19 14:23:38 -08:00
Sage Weil
b732c0ddb8 kclient: unregister mds requests on unsafe reply
Fresh vs unsafe mds request replay is handled separately.  So,
unregister requests on first reply.  This avoids a stray dentry
reference if we unmount before getting a final safe (although
eventually those should be explicitly flushed as well).
2009-02-19 13:59:15 -08:00
Sage Weil
d7cb96bd12 kclient: clean up some mdsc locking
Do not drop mds mutex in __do_request().
2009-02-19 13:11:07 -08:00
Sage Weil
919b21429b kclient: fix path memory leak 2009-02-19 13:11:07 -08:00
Sage Weil
c3ac056a41 kclient: async mds requestion completion callbacks 2009-02-19 13:11:07 -08:00
Sage Weil
0990d83f83 kclient: fix fh_to_dentry; move build_path to mds_client.c 2009-02-19 11:58:25 -08:00