RFC: All glibc machine maintainers: Is " RLIM_INFINITY as ((__rlim_t) -1)" OK?

[H.J. Lu]

On Fri, May 18, 2012 at 10:17 PM, Roland McGrath <roland@hack.frob.com> wrote:
>> and I notice that i386 has bit of code encapsulating the rlimit calls
>> which translates between RLIM_INFINITY and "RLIM_INFINITY >> 1" in
>> certain situations.
>
> It would be nice if someone could research the details and history of that
> (both in libc and on the kernel side) and follow up here for our archived
> edification.

See:

http://lkml.indiana.edu/hypermail/linux/kernel/0007.3/0582.html
http://lkml.indiana.edu/hypermail/linux/kernel/0203.2/0716.html
http://lists.gnu.org/archive/html/bug-glibc/2000-10/msg00077.html

I think RLIM_INFINITY got changed from signed to unsigned
or something like that.

-- 
H.J.