glusterfs.git/api/src/glfs-resolve.c, branch v3.4.5beta2

gfapi: backport handles to 3.4 for nfs-ganesha

2013-12-04T11:34:13+00:00

nfs-ganesha-2.0 will ship shortly. It depends on the gfapi handle support,
which won't otherwise be released until 3.5 sometime in 2014.

Change-Id: I104d6fb275bb2c710790340fdc7d998446403026
Signed-off-by: Kaleb S. KEITHLEY 
Reviewed-on: http://review.gluster.org/6394
Tested-by: Gluster Build System 
Reviewed-by: Shyamsundar Ranganathan 
Reviewed-by: Vijay Bellur

gfapi: remove unnecessary call to glfs_resolve_base()

2013-11-26T19:02:17+00:00

Calling glfs_resolve_base() on the root inode for every resolver
invocation is unnecessary and wasteful.

Here are the results from running a test program which performs
path based operations (creates and deletes 1000 files):

Without patch:
[root@blackbox ~]# sync
[root@blackbox ~]# time ./a.out 1

real    0m4.314s
user    0m1.923s
sys     0m1.144s
[root@blackbox ~]# sync
[root@blackbox ~]# time ./a.out 1

real    0m4.383s
user    0m1.940s
sys     0m1.177s
[root@blackbox ~]# sync
[root@blackbox ~]# time ./a.out 1

real    0m4.339s
user    0m1.863s
sys     0m1.129s

With patch:

[root@blackbox ~]# sync
[root@blackbox ~]# time ./a.out 1

real    0m3.005s
user    0m1.162s
sys     0m0.816s
[root@blackbox ~]# sync
[root@blackbox ~]# time ./a.out 1

real    0m3.188s
user    0m1.222s
sys     0m0.867s
[root@blackbox ~]# sync
[root@blackbox ~]# time ./a.out 1

real    0m2.999s
user    0m1.131s
sys     0m0.832s

Change-Id: Id160a24f44b4dccfcfce99a6f69ddb8938523cd5
BUG: 953694
Signed-off-by: Anand Avati 
Reviewed-on: http://review.gluster.org/6321
Tested-by: Gluster Build System

gfapi: wake migration waiters after migration

2013-11-26T18:38:26+00:00

Wake call threads which are waiting on migration to complete with
a cond_broadcast. Else if any call which arrives right when migration
is attempted will end up hanging indefinitely.

Change-Id: I7df5298f93998d9a54fb12c16654e62333018ece
BUG: 953694
Signed-off-by: Anand Avati 
Reviewed-on: http://review.gluster.org/6323
Tested-by: Gluster Build System 
Reviewed-by: Vijay Bellur

api: Fill loc.path and loc.gfid before syncop_open.

2013-09-10T08:20:05+00:00

syncop_open was crashing because of NULL dereference.
This fixes that.

Change-Id: I4bc48fac2a6c0b15c806cbbb4ae45c67891ab7ed
BUG: 1002577
Signed-off-by: Raghavendra Talur 
Reviewed-on: http://review.gluster.org/5877
Tested-by: Gluster Build System 
Reviewed-by: Vijay Bellur

gfapi: apply an upper bound on nested symlink resolution

2013-09-10T08:16:09+00:00

In case of nested symlink resolution, implement an upper
bound on the number of such nested levels the resolver will
descend. This limit is arbitrary, and set to 2048 nested
levels.

Change-Id: I264e5bd60d317eda97f4e6f49bd5d8694f8664a9
BUG: 1004100
Signed-off-by: Anand Avati 
Reviewed-on: http://review.gluster.org/5874
Tested-by: Gluster Build System 
Reviewed-by: Vijay Bellur

gfapi: store the open/create flags in fd_t object

2013-09-10T08:15:07+00:00

The flags passed on to open and create calls were not being saved
in the fd_t object, hence the fd migration was failing.

Change-Id: I486bb818477fe4c393d64a711534a082162a0e53
BUG: 1005159
Signed-off-by: Poornima 
Reviewed-on: http://review.gluster.org/5873
Tested-by: Gluster Build System 
Reviewed-by: Vijay Bellur

gfapi: handle failures in graph switches

2013-07-31T05:17:36+00:00

A graph switch failure can potentially result in NULL returned
for glfs_active_subvol(). Handle this gracefully in glfs_subvol_done()

Change-Id: I6adbd878c354b9ed2314fc71cea4671792efa0a4
BUG: 953694
Signed-off-by: Anand Avati 
Reviewed-on: http://review.gluster.org/5427
Reviewed-by: Vijay Bellur 
Tested-by: Gluster Build System

gfapi: handle graph switch (cwd, fds, locks)

2013-06-08T21:52:52+00:00

- Migrate open fds to new graph
- Migrate locks held in open fd to new fd
- Refresh CWD, so resolution of relative paths happens on refreshed
  inode (on new graph).

Change-Id: I4b18083b9b290107ebda1f917fc85b635ab72fb4
BUG: 953694
Signed-off-by: Anand Avati 
Reviewed-on: http://review.gluster.org/5167
Tested-by: Gluster Build System 
Reviewed-by: Vijay Bellur

gfapi: optimistic resolution with ESTALE detection

2013-06-08T21:50:20+00:00

Enhance the path resolver to be optimistic. Instead of performing
a wasteful ->lookup() fop for every component in every API call,
build in speculation logic to consider the previous resolution
result temporarily, and if it results in ESTALE either later in
the path resolution or in the FOP, then retry resolution with
->lookup() at every level and retry the FOP.

This is the same optimistic resolution/retry logic Linux kernel
resolver is proposing to use.

Change-Id: Iecbc25248754caf0915cd7205910563ff88ce5e7
BUG: 953694
Signed-off-by: Anand Avati 
Reviewed-on: http://review.gluster.org/5159
Tested-by: Gluster Build System 
Reviewed-by: Vijay Bellur

gfapi: handle root (/) resolution gracefully for realpath()

2013-06-08T21:38:53+00:00

Resolving path of / is crucial for realpath() to work properly.

Change-Id: I8a1b482beba37f01afbd20446d9e6f6125f2ec34
BUG: 953694
Signed-off-by: Anand Avati 
Reviewed-on: http://review.gluster.org/5152
Tested-by: Gluster Build System 
Reviewed-by: Vijay Bellur