[PATCH] rpc: fix garbage in printk in svc_tcp_accept()

[PATCH] rpc: fix garbage in printk in svc_tcp_accept()

[J. Bruce Fields]

we upgraded the kernel of a nfs-server from 2.6.17.11 to 2.6.22.6. Since
then we get the message

lockd: too many open TCP sockets, consider increasing the number of nfsd threads
lockd: last TCP connect from ^\\236^\É^D

These random characters in the second line are caused by a bug in
svc_tcp_accept.

(Note: there are two previous __svc_print_addr(sin, buf, sizeof(buf))
calls in this function, either of which would initialize buf correctly;
but both are inside "if"'s and are not necessarily executed.  This is
less obvious in the second case, which is inside a dprintk(), which is a
macro which expands to an if statement.)

Signed-off-by: Wolfgang Walter <wolfgang.walter@studentenwerk.mhn.de>
Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
---
 net/sunrpc/svcsock.c |    3 ++-
 1 files changed, 2 insertions(+), 1 deletions(-)

We're printk'ing random uninitialized memory here, which I assume makes
it appropriate for both 2.6.23 and 2.6.22.x.--b.

diff --git a/net/sunrpc/svcsock.c b/net/sunrpc/svcsock.c
index 1a89992..036ab52 100644
--- a/net/sunrpc/svcsock.c
+++ b/net/sunrpc/svcsock.c
@@ -1110,7 +1110,8 @@ svc_tcp_accept(struct svc_sock *svsk)
 						   serv->sv_name);
 				printk(KERN_NOTICE
 				       "%s: last TCP connect from %s\n",
-				       serv->sv_name, buf);
+				       serv->sv_name, __svc_print_addr(sin,
+							buf, sizeof(buf)));
 			}
 			/*
 			 * Always select the oldest socket. It's not fair,
-- 
1.5.3.1.139.g9346b

Re: [PATCH] rpc: fix garbage in printk in svc_tcp_accept()

[Linus Torvalds]

On Thu, 20 Sep 2007, J. Bruce Fields wrote:
>
> we upgraded the kernel of a nfs-server from 2.6.17.11 to 2.6.22.6. Since
> then we get the message
...

Bruce - you're losing authorship information.

Please don't do that.

Put a 

	From: Wolfgang Walter <wolfgang.walter@studentenwerk.mhn.de>

at the top of the email, so that the author gets properly attributed. I 
don't know how many of these I've missed - in this case I just ended up 
googling for the commit message because the "Signed-off-by:" lines implied 
that you weren't the original author, and see the real patch authorship 
that way.

But it's *not* the case that the first sign-off is always the author, and 
I don't want to need to google for each patch I get, so please make it a 
habit to make sure that there's proper attribution for the patches.

Some people (read: Andrew) go so far as to *always* do it (ie Andrew puts 
a "From: Andrew Morton .." at the heads of the emails he sends out), and I 
appreciate it. That way, even if the email then gets forwarded by somebody 
else (because Andrew sent it to some submaintainer instead of me), the 
original authorship doesn't get lost when somebody else then forwards the 
email.

So that "always explicitly say who the author is" is a good idea, but when 
forwarding somebody elses patch, it's *more* than a good idea - at that 
point it's absolutely required for proper attribution.

		Linus

Re: [PATCH] rpc: fix garbage in printk in svc_tcp_accept()

[J. Bruce Fields]

On Thu, Sep 20, 2007 at 01:15:05PM -0700, Linus Torvalds wrote:
> 
> 
> On Thu, 20 Sep 2007, J. Bruce Fields wrote:
> >
> > we upgraded the kernel of a nfs-server from 2.6.17.11 to 2.6.22.6. Since
> > then we get the message
> ...
> 
> Bruce - you're losing authorship information.

Whoops, yes, I usually rely on git-send-email to do that, but was
sending the patch by hand this time in a hurry, reading the commit into
my mail client, and forgot the From: line.  Thanks for catching that.

> So that "always explicitly say who the author is" is a good idea, but when 
> forwarding somebody elses patch, it's *more* than a good idea - at that 
> point it's absolutely required for proper attribution.

Yeah, I know.  Apologies.--b.