[Devel] Re: [PATCH 4/9] namespaces: utsname: switch to using uts namespaces

[Eric W. Biederman]

"Randy.Dunlap" <rdunlap at xenotime.net> writes:

> On Thu, 18 May 2006 10:49:36 -0500 Serge E. Hallyn wrote:
>
>> Replace references to system_utsname to the per-process uts namespace
>> where appropriate.  This includes things like uname.
>> 
>> Changes: Per Eric Biederman's comments, use the per-process uts namespace
>> 	for ELF_PLATFORM, sunrpc, and parts of net/ipv4/ipconfig.c
>> 
>> Signed-off-by: Serge E. Hallyn <serue at us.ibm.com>

>
> OK, here's my big comment/question.  I want to see <nodename> increased to
> 256 bytes (per current POSIX), so each field of struct <variant>_utsname
> needs be copied individually (I think) instead of doing a single
> struct copy.

Where is it specified?  Looking at the spec as SUSV3 I don't see a size
specified for nodename.

> I've been working on this for the past few weeks (among other
> things).  Sorry about the timing.
> I could send patches for this against mainline in a few days,
> but I'll be glad to listen to how it would be easiest for all of us
> to handle.
>
> I'm probably a little over half done with my patches.
> They will end up adding a lib/utsname.c that has functions for:
>   put_oldold_uname()	// to user
>   put_old_uname()	// to user
>   put_new_uname()	// to user
>   put_posix_uname()	// to user

Sounds reasonable, if we really need a 256 byte nodename.

As long as they take a pointer to the appropriate utsname
structure these patches should not fundamentally conflict.

Eric