Re: [PATCH] mktime & nanosleep: make sure SIGALRM is not blocked

[Mike Frysinger]

On Tuesday 19 May 2009 07:08:37 Bruno Haible wrote:
> Mike Frysinger wrote:
> > we're of course trying to
> > figure out who is blocking the signal in the first place, but it isnt
> > exactly a trivial affair when it crops up on some users systems and they
> > cant always reproduce it.
>
> Yes, I know that on Gentoo systems, many reported bugs can never be closed
> because they cannot be reproduced on other Gentoo systems. Still, the rule
> is: file a bug report with the program that is buggy.
>
> > i'm not talking about all programs.  i'm talking about the common ones i
> > keep fielding bug reports for (gnulib).  in the process,
>
> I understand that it is easier to file a bug report about 'configure' files
> than about other programs like 'inetd', because 'configure' operates in a
> more transparent way (logging output, config.log, config.status). But
> that's not a reason for claiming that 'configure' is buggy when the bug is
> elsewhere. Instead, you should make the other program more transparent
> (through logging or tracing features).

quoting the same spec you did indicates that the gnulib tests themselves are 
buggy for assuming a specific signal state.
-mike

signature.asc
Description: This is a digitally signed message part.