[Libtirpc-devel] Fwd: Re: proposed patch to rpcbind to provide finer-grained security controls than offered by the -i option
Steve Dickson
SteveD@redhat.com
Tue Dec 14 20:53:00 GMT 2010
On 12/14/2010 02:14 PM, Ulrich Drepper wrote:
> On Tue, Dec 14, 2010 at 13:57, Steve Dickson <SteveD@redhat.com> wrote:
>> We do install (at least in Fedora) two .x files in the
>> header directories /usr/include/tirpc/rpc/ and
>> /usr/include/tirpc/rpcsvc. But not clear as to why...
>
> Are they tested?
No they are not.
For fun, I just ran rpcgen on one of them. Two .c files and
a .h file were generated without any errors. But comparing
the header file that was generated with the one we distribute
in /usr/include/tirpc/rpc, they were very different...
>
>
>> It probably makes sense to move the command from
>> glibc-common into libtirpc...
>
> Until there is a really good reason we probably shouldn't do it.
>
> If and when you make changes to rpcgen (for IPv6, perhaps) then we can
> think about moving it. But until then, just for stability and to not
> introduce surprises, it should probably stay where it is. It's not
> creating any problems.
Cool... As long as it will not stop you from deprecating the glibc
header files, I'm all for leaving it where it is...
What is the next step?
steved.
More information about the Libc-alpha
mailing list