Commits · c0eae8cd77bc34b7e4c52037eeb53712f46fa05c · Kirill Smelkov / linux

12 Feb, 2016 3 commits

orangefs: get rid of handle_io_error() · c0eae8cd

Al Viro authored Feb 11, 2016

the second caller never needs to cancel, actually
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Mike Marshall <hubcap@omnibond.com>

c0eae8cd

orangefs: wait_for_direct_io(): restore the position in iter when restarting · 7b9761af
Al Viro authored Feb 07, 2016
```
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Mike Marshall <hubcap@omnibond.com>
```
7b9761af
orangefs: avoid freeing a slot twice in wait_for_direct_io() · e17be9fd
Al Viro authored Feb 06, 2016
```
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Mike Marshall <hubcap@omnibond.com>
```
e17be9fd

04 Feb, 2016 6 commits

Orangefs: added a couple of WARN_ONs, perhaps just temporarily. · 6ebcc3fc
Mike Marshall authored Feb 04, 2016
```
Signed-off-by: Mike Marshall <hubcap@omnibond.com>
```
6ebcc3fc
orangefs: Do not retrieve size from servers unless it it necessary. · 237f8282
Martin Brandenburg authored Feb 03, 2016
```
Signed-off-by: Martin Brandenburg <martin@omnibond.com>
Signed-off-by: Mike Marshall <hubcap@omnibond.com>
```
237f8282

orangefs: Implement inode_operations->permission(). · 933287da

Martin Brandenburg authored Jan 30, 2016

Thus d_revalidate is not obliged to check on as much, which will
eventually lead the way to hammering the filesystem servers much less.
Signed-off-by: Martin Brandenburg <martin@omnibond.com>
Signed-off-by: Mike Marshall <hubcap@omnibond.com>

933287da

orangefs: Only compare attributes specified in orangefs_inode_getattr. · fe88adc3
Martin Brandenburg authored Jan 30, 2016
```
Signed-off-by: Martin Brandenburg <martin@omnibond.com>
Signed-off-by: Mike Marshall <hubcap@omnibond.com>
```
fe88adc3

Orangefs: clean up slab allocation. · 2d4cae0d

Mike Marshall authored Feb 04, 2016

A couple of caches were no longer needed:

 - iov_iter improvements to orangefs_devreq_write_iter eliminated
   the need for the dev_req_cache.

 - removal (months ago) of the old AIO code eliminated the need
   for the kiocb_cache.

Also, deobfuscation of use of GFP_KERNEL when calling kmem_cache_(z)alloc
for remaining caches.
Signed-off-by: Mike Marshall <hubcap@omnibond.com>

2d4cae0d

Orangefs: improve gossip statement · 5090c967

Mike Marshall authored Feb 04, 2016

There were two just alike, making it hard maybe to tell which one
you were looking at in syslog... so I changed it a little by adding
some extra interesting tidbits to it...
Signed-off-by: Mike Marshall <hubcap@omnibond.com>

5090c967

28 Jan, 2016 2 commits

orangefs: Fix revalidate. · 99109822

Martin Brandenburg authored Jan 28, 2016

Previously, it would update a live inode. This was fixed, but it did not
ever check that the inode attributes in the dcache are correct. This
checks all inode attributes and rejects any that are not correct, which
causes a lookup and thus a new getattr.

Perhaps inode_operations->permission should replace or augment some of
this.

There is no actual caching, and this does a rather excessive amount of
network operations back to the filesystem server.
Signed-off-by: Martin Brandenburg <martin@omnibond.com>
Signed-off-by: Mike Marshall <hubcap@omnibond.com>

99109822

orangefs: Util functions shouldn't operate on inode where it can be avoided. · 394f647e
Martin Brandenburg authored Jan 25, 2016
```
Signed-off-by: Martin Brandenburg <martin@omnibond.com>
Signed-off-by: Mike Marshall <hubcap@omnibond.com>
```
394f647e

23 Jan, 2016 23 commits

orangefs: clean up op_alloc() · 115b93a8

Al Viro authored Jan 23, 2016

fold orangefs_op_initialize() in there, don't bother locking something
nobody else could've seen yet, use kmem_cache_zalloc() instead of
explicit memset()...
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Mike Marshall <hubcap@omnibond.com>

115b93a8

orangefs: move handle_io_error() to file.c · b0bc3a7b

Al Viro authored Jan 23, 2016

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Mike Marshall <hubcap@omnibond.com>

b0bc3a7b

orangefs: don't reinvent completion.h... · 2a9e5c22

Al Viro authored Jan 23, 2016

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Mike Marshall <hubcap@omnibond.com>

2a9e5c22

if ORANGEFS_VFS_OP_FILE_IO request had been given up, don't bother waiting · 4f55e397

Al Viro authored Jan 23, 2016

... we are not going to get woken up anyway, so it's just going to time out
and whine.
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Mike Marshall <hubcap@omnibond.com>

4f55e397

orangefs: get rid of MSECS_TO_JIFFIES · 727cbfea

Al Viro authored Jan 23, 2016

All timeouts are in _seconds_, so all calls are of form
MSECS_TO_JIFFIES(n * 1000), which is a convoluted way to
spell n * HZ.
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Mike Marshall <hubcap@omnibond.com>

727cbfea

orangefs_clean_up_interrupted_operation: call with op->lock held · eab9b389
Al Viro authored Jan 23, 2016
```
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Mike Marshall <hubcap@omnibond.com>
```
eab9b389

orangefs: reduce nesting in wait_for_matching_downcall() · 70c6ea26

Al Viro authored Jan 23, 2016

reorder if branches...
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Mike Marshall <hubcap@omnibond.com>

70c6ea26

orangefs: remove cargo-culting spin_lock_irqsave() in service_operation() · e1056a9c
Al Viro authored Jan 23, 2016
```
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Mike Marshall <hubcap@omnibond.com>
```
e1056a9c

orangefs: hopefully saner op refcounting and locking · ed42fe05

Al Viro authored Jan 22, 2016

* create with refcount 1
* make op_release() decrement and free if zero (i.e. old put_op()
  has become that).
* mark when submitter has given up waiting; from that point nobody
  else can move between the lists, change state, etc.
* have daemon read/write_iter grab a reference when picking op
  and *always* give it up in the end
* don't put into hash until we know it's been successfully passed to
  daemon

* move op->lock _lower_ than htab_in_progress_lock (and make sure
  to take it in purge_inprogress_ops())
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Mike Marshall <hubcap@omnibond.com>

ed42fe05

orangefs: make sure that reopening pvfs2-req won't overlap with the end of close · fee25ce1
Al Viro authored Jan 22, 2016
```
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Mike Marshall <hubcap@omnibond.com>
```
fee25ce1

orangefs: nothing should remain in request list and in hash · 96acf9d6

Al Viro authored Jan 22, 2016

... otherwise some thread is running in .text that is about to
be freed.
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Mike Marshall <hubcap@omnibond.com>

96acf9d6

orangefs: move wakeups into set_op_state_{serviced,purged}() · 60831949
Al Viro authored Jan 21, 2016
```
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Mike Marshall <hubcap@omnibond.com>
```
60831949

orangefs: make wait_for_...downcall() static · ade3d781

Al Viro authored Jan 21, 2016

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Mike Marshall <hubcap@omnibond.com>

ade3d781

orangefs: move wakeups into set_op_state_{serviced,purged}() · 831d0949
Al Viro authored Jan 21, 2016
```
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Mike Marshall <hubcap@omnibond.com>
```
831d0949

orangefs: make wait_for_...downcall() static · b7ae37b0

Al Viro authored Jan 21, 2016

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Mike Marshall <hubcap@omnibond.com>

b7ae37b0

make orangefs_clean_up_interrupted_operation() static · e07db0a2

Al Viro authored Jan 21, 2016

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Mike Marshall <hubcap@omnibond.com>

e07db0a2

orangefs: kill orangefs_inode_s ->list · 1264ddfd

Al Viro authored Jan 21, 2016

no users...
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Mike Marshall <hubcap@omnibond.com>

1264ddfd

orangefs: get rid of <censored> macros · fc916da5

Al Viro authored Jan 19, 2016

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Mike Marshall <hubcap@omnibond.com>

fc916da5

orangefs: ->poll() doesn't need spinlock · 90e54e36

Al Viro authored Jan 19, 2016

not just for list_empty()...
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Mike Marshall <hubcap@omnibond.com>

90e54e36

orangefs: kill ioctl32 rudiments · 8016387c

Al Viro authored Jan 19, 2016

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Mike Marshall <hubcap@omnibond.com>

8016387c

orangefs: ->poll() is only called between successful ->open() and ->release() · 83595db0
Al Viro authored Jan 19, 2016
```
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Mike Marshall <hubcap@omnibond.com>
```
83595db0
orangefs: generic_file_open() is pointless for character devices · fb6d2526
Al Viro authored Jan 19, 2016
```
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Mike Marshall <hubcap@omnibond.com>
```
fb6d2526
orangefs: use DEFINE_MUTEX (and mutex_init() had been too late) · 3e1dd9aa
Al Viro authored Jan 19, 2016
```
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Mike Marshall <hubcap@omnibond.com>
```
3e1dd9aa

19 Jan, 2016 1 commit
- Orangefs: make gossip statement more palatable to xtensa · cf0c2771
  Mike Marshall authored Jan 19, 2016
```
Thanks to Intel's kbuild test robot
Signed-off-by: Mike Marshall <hubcap@omnibond.com>
```
  cf0c2771
15 Jan, 2016 2 commits

Orangefs: merge with V4.4 · 5e1f3938

Mike Marshall authored Jan 15, 2016

Merge tag 'v4.4' of git://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux into current

Linux 4.4

5e1f3938

Orangefs: add verification to decode_dirents · 1808f8cc

Mike Marshall authored Jan 15, 2016

Also add comments to decode_dirents and make it more readable.
Signed-off-by: Mike Marshall <hubcap@omnibond.com>

1808f8cc

13 Jan, 2016 3 commits
- Orangefs: add protocol information to Documentation/filesystems/orangefs.txt · fcac9d57
  Mike Marshall authored Jan 13, 2016
```
Signed-off-by: Mike Marshall <hubcap@omnibond.com>
```
  fcac9d57
- Orangefs: make .statfs gossip_debug more complete. · be57366e
  Mike Marshall authored Jan 13, 2016
```
Signed-off-by: Mike Marshall <hubcap@omnibond.com>
```
  be57366e
- Orangefs: define a minimum compatible userspace version. · 569dbfc6
  Mike Marshall authored Jan 13, 2016
```
Signed-off-by: Mike Marshall <hubcap@omnibond.com>
```
  569dbfc6