[linux-cifs-client] Re: [PATCH 0/7] cifs: clean up socket creation, connection and sending (try #2)

[Steve French]

After review of Jeff's 2 recent patch series, and discussion with Jeff
on IRC (as a result Jeff respun and made minor modifications to a
couple), I have just merged 10 of his 14 patches into cifs-2.6.git.
See below:

3 min ago 	Jeff Layton 	cifs: make ipv6_connect take a TCP_Server_Info
3 min ago 	Jeff Layton 	cifs: make ipv4_connect take a TCP_Server_Info
14 min ago 	Jeff Layton 	cifs: don't declare smb_vol info on the stack
15 min ago 	Jeff Layton 	cifs: move allocation of new TCP_Server_Info
into separ ...
7 min ago 	Jeff Layton 	cifs: account for IPv6 in ses->serverName and clean ...
21 min ago 	Jeff Layton 	cifs: make dnotify thread experimental code
5 hours ago 	Jeff Layton 	cifs: convert tcpSem to a mutex
5 hours ago 	Jeff Layton 	cifs: take module reference when starting cifsd
6 hours ago 	Jeff Layton 	cifs: display addr and prefixpath options in /proc ...
6 hours ago 	Jeff Layton 	cifs: remove unused SMB session pointer from
struct ...

On Sun, Nov 30, 2008 at 12:40 PM, Jeff Layton <jlayton at redhat.com> wrote:
> This patchset is intended as a cleanup of the code that creates and
> connects sockets. It's also intended to unify the smb_send and smb_send2
> functions, and simplify the sending code in the blocking I/O case. There
> is a lot of duplicate code in those functions, and this should make it
> easier to ensure that we fix bugs with it properly. This should also fix
> the case where we get a partial send in smb_send. Previously, we did not
> force a reconnect in that case.
>
> This is the second post of this patchset. The main differences from the
> original set are:
>
> 1) fix some symbol naming style issues that HCH pointed out
>
> 2) move the address handling into cifs_get_tcp_server and out of
>   cifs_mount altogether (also pointed out by HCH)
>
> 3) the addition of a patch to reduce stack consumption in cifs_mount by
>   dynamically allocating the volume_info
>
> This patchset is also intended for 2.6.29 and depends on the 5 patch set
> of cleanups that I sent last week.
>
> I've only given this set cursory testing. If it looks OK, I'll plan to
> do some more with it. The last set was more heavily tested, and showed
> no functional or performance regressions as best I could tell. Since the
> differences between that set and this one are fairly superficial, I
> don't expect much difference in performance here.
>
> Jeff Layton (7):
>  cifs: convert tcpSem to a mutex
>  cifs: move allocation of new TCP_Server_Info into separate function
>  cifs: don't declare smb_vol info on the stack
>  cifs: make ipv4_connect take a TCP_Server_Info arg
>  cifs: make ipv6_connect take a TCP_Server_Info arg
>  cifs: turn smb_send into a wrapper around smb_sendv
>  cifs: don't retry on blocking sends
>
>  fs/cifs/cifsglob.h  |    2 +-
>  fs/cifs/cifsproto.h |    4 +-
>  fs/cifs/connect.c   |  609 ++++++++++++++++++++++++++-------------------------
>  fs/cifs/transport.c |  147 ++++---------
>  4 files changed, 359 insertions(+), 403 deletions(-)
>
>



-- 
Thanks,

Steve