David> It [times()] has the same (erroneous) behavior. [i.e. on RedHat
David> Linux/Alpha 2.1 than on the good old BLADE_0.3]
David> I ran across this while working on GNU libc and it's fixed in my
David> sources now, but I was hoping on not having to do another
David> libc-4.6.27-based release. If you can't wait to have this fixed,
David> let me know and I'll see about it.
times():
I've tracked what was interesting for me while porting GNU time-1.6
to Linux/Alpha (actually to BLADE_0.3). So there is no
emergency for me to get a new libc incorporating your fix.
pre-compiled updates:
I think that it would be a great benefit (for BLADE_0.3
users? if I am not the last) to have archived on the net,
in usual compressed-tar format, pre-compiled versions of the updates
you make in Linux/alpha commands and libraries (along with source and/or
patch files). You used to do that before, and you still do it for
kernel updates; of course it would be (time-, at least) expensive,
so it's up to you.
fred
PS (times()): Was it a missing 's' in the library sources you're working
on ? I think so, because (at least with BLADE_0.3 libs) times() and time()
return the same value!
PPS(pre-compiled updates): Could you tell me if there are known
incompatibilities beetween the BLADE_0.3 kernel+system header
files and the axp-libc.40 (.39, .38, .37, ...) you've put on AZStarNet ?
Sorry to insist, but I think it would be stupid to just `try and see'
if you know (even approximately) the answer... Thanks.