Hi,
2017-10-12 10:07 Ferenc Wágner:
Helmut Grohne <[email protected]> writes:
On Wed, Oct 11, 2017 at 03:24:50PM +0200, Ferenc Wágner wrote:
#hurd confirmed that this is a Hurd bug in glibc, and promised a fix it
in the next upload. So I guess an unconditional check for
_POSIX_MONOTONIC_CLOCK being positive would be a good general solution.
Do you agree?
I do agree that hurd should fix this. Yet, it may not be the thing to
rely on here.
It was the explicit recommendation I got on #hurd: "just use the POSIX
interface." If that breaks on Hurd right now, that's fine. The bug
will be fixed shortly.
Testing for a working CLOCK_MONOTONIC is something that many projects
do.
Aren't these all manifestations of the same misguided attempt to work
around the above bug in Hurd? In what other cases is this check
necessary?
If upstream Hurd people were so sure about it I'd strongly consider to
follow their advice.
From what you talk about in this bug report, if the extra check is
mostly to satisfy a bug in Hurd and they fix it, you could as well
remove it.
But on the other hand, as Helmut says, if many other projects do it and
since Hurd is not very popular, perhaps it's a wider problem than Hurd.
If the way that Helmut proposes fixes cross-compilation and still works
fine for "native" builds, maybe it will help to detect the same problem
as Hurd in new architectures in the future.
Cheers.
--
Manuel A. Fernandez Montecelo <[email protected]>
