Re: usleep

[Eric Blake]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

According to Ludovic Courtès on 11/18/2009 4:35 PM:
>> +    {
>> +      seconds -= limit;
>> +      unsigned int result = sleep (limit);
> 
> This declaration-after-statement is a C99 thing.  Isn’t it something
> usually avoided in Gnulib?

Yep.  I'll fix it in my respin.  Besides, I found out that rpl_nanosleep
needs the same bug fix, and not only that, but it has a logic bug - it
ignores EINTR, forcing the sleep to last until a fatal signal or the
entire time has elapsed, on systems where the remaining time properly
decreases.  Worse, it goes into an infinite 100% CPU busy loop on cygwin
1.5.x (where the result is always -1 for EINVAL on large enough time, and
the time remaining never decreases).  Note, however, that my claim that
rpl_nanosleep is broken is by inspection only, since I don't want to wait
7 weeks to prove that 'sleep 50d' will eventually exit ;)

- --
Don't work too hard, make some time for fun as well!

Eric Blake             address@hidden
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (Cygwin)
Comment: Public key at home.comcast.net/~ericblake/eblake.gpg
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/

iEYEARECAAYFAksEl70ACgkQ84KuGfSFAYBP6wCePS/MBrvwD7K0P0leKI1e6Hdn
RF8AoJdmyM09oVvPr7kq56p1ksTBcH3h
=CHGq
-----END PGP SIGNATURE-----