32bit UID support in 2.3.39

32bit UID support in 2.3.39

[Andreas Jaeger]

Hi,

just a short note: Linux 2.3.39 contains the 32bit UID support from
Chris Wing.  I'll try to add his patches to glibc asap into glibc
2.1.3 and 2.2 and send them here when they're ready.

Andreas
-- 
 Andreas Jaeger
  SuSE Labs aj@suse.de
   private aj@arthur.rhein-neckar.de

Re: 32bit UID support in 2.3.39

[Mark Kettenis]

   From: Andreas Jaeger <aj@suse.de>
   Date: 11 Jan 2000 08:34:17 +0100

   Hi,

   just a short note: Linux 2.3.39 contains the 32bit UID support from
   Chris Wing.  I'll try to add his patches to glibc asap into glibc
   2.1.3 and 2.2 and send them here when they're ready.

I think it would be wise to release 2.1.3 without 32-bit UID support.

Mark

Re: 32bit UID support in 2.3.39

[Andreas Jaeger]

>>>>> Mark Kettenis writes:

Mark>    From: Andreas Jaeger <aj@suse.de>
Mark>    Date: 11 Jan 2000 08:34:17 +0100

Mark>    Hi,

Mark>    just a short note: Linux 2.3.39 contains the 32bit UID support from
Mark>    Chris Wing.  I'll try to add his patches to glibc asap into glibc
Mark>    2.1.3 and 2.2 and send them here when they're ready.

Mark> I think it would be wise to release 2.1.3 without 32-bit UID support.

This depends on the time frame we have for glibc 2.1.3 and for 2.2.  I
don't think we should wait another 6 months or more until the 32bit
UID support goes in.

What you imply is that the patches are not sufficiently tested.  AFAIK
the have been tested by Chris who developed them but haven't had
widespread testing.  If they go in, we should consider releasing a
2.1.3pre2.

Andreas
-- 
 Andreas Jaeger
  SuSE Labs aj@suse.de
   private aj@arthur.rhein-neckar.de

Re: 32bit UID support in 2.3.39

[Roland McGrath]

I would agree with Mark here, that 2.1.3 (which is already on the doorstep)
should go out without these changes.  If they are in fact so stable, then
2.1.4 can go out with them in not too long.  But we need 2.1.3 out asap
without any new ghost of a chance of incompatibility or instability.
This doesn't need to be such a big deal if we just push the bug-fix releases
out more often when we have stable bug fixes, and keep them really "bug-fix".

To my mind, any interface changes at all (even with versioning) going into
2.1.x at all need to be very strongly justified; I'd prefer to see this
wait for 2.2.