On Aug 13, 2009, at 5:13 PM, Joe Perches wrote:

On Thu, 2009-08-13 at 17:02 -0400, Chuck Lever wrote:

quoted

On Aug 13, 2009, at 4:34 PM, Joe Perches wrote:

quoted

net/sunrpc/svcauth_unix.c:		seq_printf(m, "%s %pI6 %s\n", im-

quoted

m_class, &addr, dom);

This one might be a bad example.

There are 9 of them in net

$ grep -rP --include=*.[ch] "%pI6" net | grep seq_
net/sctp/ipv6.c:	seq_printf(seq, "%pI6 ", &addr->v6.sin6_addr);
net/ipv6/netfilter/nf_conntrack_l3proto_ipv6.c:	return seq_printf(s,  
"src=%pI6 dst=%pI6 ",
net/ipv6/ip6mr.c:		seq_printf(seq, "%pI6 %pI6 %-3hd",
net/netfilter/xt_hashlimit.c:		return seq_printf(s, "%ld %pI6:%u-> 
%pI6:%u %u %u %u\n",
net/netfilter/xt_recent.c:		seq_printf(seq, "src=%pI6 ttl: %u  
last_seen: %lu oldest_pkt: %u",
net/netfilter/ipvs/ip_vs_ctl.c:				seq_printf(seq, "%s  [%pI6]:%04X  
%s ",
net/netfilter/ipvs/ip_vs_conn.c:			seq_printf(seq, "%-3s %pI6 %04X  
%pI6 %04X %pI6 %04X %-11s %7lu\n",
net/netfilter/ipvs/ip_vs_conn.c:			seq_printf(seq, "%-3s %pI6 %04X  
%pI6 %04X %pI6 %04X %-11s %-6s %7lu\n",
net/sunrpc/svcauth_unix.c:		seq_printf(m, "%s %pI6 %s\n", im- 

quoted

m_class, &addr, dom);

I checked with the NFSD maintainer.  He thinks this last one is for  
debugging.  It's hard to tell just by looking whether a seq_printf()  
call site actually has a strong user space dependence.

quoted

[ I would think user space in general should be using inet_pton(3)
everywhere for such interfaces, so the format of these addresses
wouldn't matter so much.  Probably impossible at this point. ]

David Miller is authoritative here.

I've seen enough to agree that a new formatter is a reasonable  
approach.  Thanks for checking.

quoted

I'm not arguing one way or the other, but it would be useful if
someone could check exactly what the dependencies are right now.  It
seems like we're speculating a bit.

cheers, Joe

--
Chuck Lever
chuck[dot]lever[at]oracle[dot]com