Linus Torvalds wrote:
>
> On 28 Jul 2000, Ulrich Drepper wrote:
>
> > Linus Torvalds <torvalds@transmeta.com> writes:
> >
> > > There is only a stupid push to try to be more generic than anybody
> > > actually needs! It makes glibc slower and bigger than it needs to be, and
> > > quite frankly I don't see the point of having a system call that returns
> > > the value "32".
> >
> > It will not return 32 in all cases, that's the whole point. I will
> > save this mail and send it to all the people who complain about this.
> > There are lots of people who need larger numbers of groups. If you
> > don't need such a large number doesnot mean nobody needs it.
>
> PLEASE save the mail.
>
> But even more, please READ the mail too. Look at my suggestion. And when
> you send the mail out to somebody who complains about NRGROUPS, then send
> out the mail in ENTIRETY, and explain to those people why you continually
> ignore my suggestions even though they obviously fix the problem.
>
> I'm damn sick and tired of you blaming the kernel for your own stupidity
> and incompetence.
>
Now, the maximum number of groups is something that the kernel could
potentially change and that *would* affect user space. Similar to the
maximum number of files. Now... we have a perfectly good way to export
constants like that to user space, i.e. the sysctl (/proc/sys)
mechanism. This gives a couple of options:
a) The kernel implements such a sysctl, readonly at least for now, glibc
uses it, if available;
b) glibc doesn't do anything for now, if we change NGROUPS we export it
via sysctl and glibc will need to be updated then;
c) we reserve a namespace point, say /proc/sys/kernel/ngroups, and glibc
uses it, if available. If we ever change NGROUPS *then* we add it to
the kernel;
d) glibc may have a /etc/sysconf file, which can look like:
__SC_NGROUPS 32
... in the future, it may look like ...
__SC_NGROUPS /proc/sys/kernel/ngroups
-hpa
-- <hpa@transmeta.com> at work, <hpa@zytor.com> in private! "Unix gives you enough rope to shoot yourself in the foot." http://www.zytor.com/~hpa/puzzle.txt- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.rutgers.edu Please read the FAQ at http://www.tux.org/lkml/
This archive was generated by hypermail 2b29 : Mon Jul 31 2000 - 21:00:29 EST