bug-gnulib
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: futimens broken on GNU Hurd (no support for UTIME_NOW/UTIME_OMIT)


From: Eric Blake
Subject: Re: futimens broken on GNU Hurd (no support for UTIME_NOW/UTIME_OMIT)
Date: Wed, 24 Sep 2014 08:54:19 -0600
User-agent: Mozilla/5.0 (X11; Linux x86_64; rv:31.0) Gecko/20100101 Thunderbird/31.1.0

On 09/24/2014 06:49 AM, Colin Watson wrote:
> On Wed, Sep 24, 2014 at 06:25:18AM -0600, Eric Blake wrote:
>> I'd rather we make this conditional on a feature probe (such as #ifdef
>> UTIME_NOW_WORKS, suitably defined via a configure-time test) than a
>> platform probe, so that when the platform is eventually fixed, we quit
>> using the fallback.
> 
> OK.  The existing gl_cv_func_futimens_works test already spots the
> problem, but it's in m4/futimens.m4, which is one module layer up from
> lib/utimens.c.  Duplicating that code would seem suboptimal.  What would
> you advise here?

Without looking at the .m4 files (just quickly typing this off my head),
is there any way to refactor the m4 modules to write the test only once
and then have both futimens and utimens depend on the test running first
and looking at that cache variable result?  I can probably help with
this when I have a longer chunk of time to look at it, but it won't be
today.  You've already done a decent job in pointing me to the details
that need tweaking.

-- 
Eric Blake   eblake redhat com    +1-919-301-3266
Libvirt virtualization library http://libvirt.org

Attachment: signature.asc
Description: OpenPGP digital signature


reply via email to

[Prev in Thread] Current Thread [Next in Thread]