[PULL] v9fs bug fixes and documentation updates for 2.6.26

[PULL] v9fs bug fixes and documentation updates for 2.6.26

[Eric Van Hensbergen]

Linus, please pull the following bug-fixes and documentation updates
from my kernel.org for-linus branch.

The following changes since commit dd286422fefdcff784e8d336deeb88ce817e14db:
  Linus Torvalds (1):
        Merge branch 'for-linus' of git://git.kernel.org/.../roland/infiniband

are available in the git repository at:

  ssh://master.kernel.org/pub/scm/linux/kernel/git/ericvh/v9fs.git for-linus

Eric Van Hensbergen (4):
      9p: Documentation updates
      9p: propagate parse_option changes to client and transports
      9p: make cryptic unknown error from server less scary
      9p: fix error path during early mount

Jim Meyering (1):
      fs/9p/v9fs.c (v9fs_parse_options): Handle kstrdup and
match_strdup failure. Now that this function can fail, return an int,
diagnose other option-parsing failures, and adjust the sole caller:
(v9fs_session_init): Handle kstrdup failure. Propagate any new
v9fs_parse_options failure "up".

Josef 'Jeff' Sipek (2):
      9p: use struct mutex instead of struct semaphore
      9p: Correct fidpool creation failure in p9_client_create

Markus Armbruster (1):
      add match_strlcpy() us it to make v9fs make uname and remotename
parsing more robust

Steven Rostedt (1):
      9p: fix flags length in net

 fs/9p/fid.h                |   15 +++
 fs/9p/v9fs.c               |   57 ++++++++---
 fs/9p/v9fs.h               |   85 +++++++++++-----
 fs/9p/vfs_addr.c           |    2 +-
 fs/9p/vfs_dir.c            |    2 +-
 fs/9p/vfs_file.c           |   11 +-
 fs/9p/vfs_inode.c          |   50 +++++++---
 fs/9p/vfs_super.c          |   35 +++----
 include/linux/parser.h     |    2 +-
 include/net/9p/9p.h        |  240 ++++++++++++++++++++++++++++++++++++++------
 include/net/9p/client.h    |   35 +++++++
 include/net/9p/transport.h |   44 ++++++++-
 lib/parser.c               |   32 ++++--
 net/9p/Kconfig             |   10 --
 net/9p/Makefile            |    3 -
 net/9p/client.c            |   32 +++++--
 net/9p/conv.c              |  128 +++++++++++++++++++++++-
 net/9p/error.c             |   15 ++-
 net/9p/fcprint.c           |    8 ++
 net/9p/mod.c               |    8 +-
 net/9p/trans_fd.c          |  204 +++++++++++++++++++++++++++++++-------
 net/9p/trans_virtio.c      |  175 ++++++++++++++++++++++++++++----
 net/9p/util.c              |   36 +++++--
 23 files changed, 1014 insertions(+), 215 deletions(-)

Re: [PULL] v9fs bug fixes and documentation updates for 2.6.26

[Linus Torvalds]

On Wed, 14 May 2008, Eric Van Hensbergen wrote:
> 
> Jim Meyering (1):
>       fs/9p/v9fs.c (v9fs_parse_options): Handle kstrdup and
> match_strdup failure. Now that this function can fail, return an int,
> diagnose other option-parsing failures, and adjust the sole caller:
> (v9fs_session_init): Handle kstrdup failure. Propagate any new
> v9fs_parse_options failure "up".

Grr. Somebody isn't following the nice rules we have and that git 
encourages: make a commit message be a nice "one-line header" with the 
more complete explanation separated by an empty line (and with nice 
line-breaks etc).

IOW, womebody has bad habits from SCM systems that didn't have good 
visualization tools, so likely nobody actually looked at the messages 
anyway.

In fact, looking closer, I don't even know how you did that. The thing 
must have been forwarded as email, because the author is Jim Meyering, but 
it's been forwarded through Andrew etc, so I bet the message was fine at 
some point. And all the git email-application tools do the rigth thing. 

Anyway, I don't *quite* care enough to ask you to re-do this, and I merged 
it, but I do actually enjoy the fact that our commit messages are 
generally pretty readable, and the one-liner header rule means that gitk 
and various shortlog tools all give nice summaries too.

So please look out for this in the future.

There are other commits that don't follow the rules, btw: you've applied 
patches from others that have the rigth Sign-off's from the authors, but 
you should sign off yourself, not just ack them. By actually taking that 
thing and committing it, you're doing more than ack'ing somebody elses 
work.

I know it isn't really a big deal, especially with those things mostly 
being pretty trivial, but I'd rather nip these issues in the bud than have 
it come back when it actually might matter.

			Linus

Re: [PULL] v9fs bug fixes and documentation updates for 2.6.26

[Eric Van Hensbergen]

On Wed, May 14, 2008 at 9:40 PM, Linus Torvalds
<torvalds@linux-foundation.org> wrote:
>
>
> On Wed, 14 May 2008, Eric Van Hensbergen wrote:
>>
>> Jim Meyering (1):
>>       fs/9p/v9fs.c (v9fs_parse_options): Handle kstrdup and
>> match_strdup failure. Now that this function can fail, return an int,
>> diagnose other option-parsing failures, and adjust the sole caller:
>> (v9fs_session_init): Handle kstrdup failure. Propagate any new
>> v9fs_parse_options failure "up".
>
> Grr. Somebody isn't following the nice rules we have and that git
> encourages: make a commit message be a nice "one-line header" with the
> more complete explanation separated by an empty line (and with nice
> line-breaks etc).
>

Sorry boss.  I had a nagging feeling I was doing the wrong thing about
not diving in and fixing that comment.  Not sure where I messed up
applying the email, but if I see that sort of thing in the future I'll
edit it by hand.

>
> There are other commits that don't follow the rules, btw: you've applied
> patches from others that have the rigth Sign-off's from the authors, but
> you should sign off yourself, not just ack them. By actually taking that
> thing and committing it, you're doing more than ack'ing somebody elses
> work.
>

Sure, long term misunderstand on my part.  Thanks for clearing this
up, I'll be sure to properly sign off in the future.

          -eric