bug#40046: closed (28.0.50; infinite recursion futimens -> fdutimens -> futimens -> ...)

[GNU bug Tracking System]

Your message dated Fri, 13 Mar 2020 11:21:18 +0100
with message-id <address@hidden>
and subject line Re: 28.0.50; infinite recursion futimens -> fdutimens -> 
futimens -> ...
has caused the debbugs.gnu.org bug report #40046,
regarding 28.0.50; infinite recursion futimens -> fdutimens -> futimens -> ...
to be marked as done.

(If you believe you have received this mail in error, please contact
address@hidden.)


-- 
40046: http://debbugs.gnu.org/cgi/bugreport.cgi?bug=40046
GNU Bug Tracking System
Contact address@hidden with problems
> --- Begin Message ---
>
>
>
> Subject: 
>
> 28.0.50; infinite recursion futimens -> fdutimens -> futimens -> ...
>
>
>
>
> Date: 
>
> Thu, 12 Mar 2020 23:03:28 +0100
>
>
>
>
> On macOS 10.14, futimens in lib/futimens.c calls fdutimens in lib/utimens.c 
> which calls futimens again, recursing until the stack is full. This cause 
> Fcopy_file to crash, for instance.
>
> No doubt I've made a silly configuration error, but I've wiped the 
> (out-of-tree) build directory, regenerated configure and all. As expedience I 
> commented out the futimens definition. Not sure how this is supposed to work 
> really -- what am I missing? HAVE_FUTIMENS is set in config.h.
>
>
>
>
> --- End Message ---
> --- Begin Message ---
>
>
>
> Subject: 
>
> Re: 28.0.50; infinite recursion futimens -> fdutimens -> futimens -> ...
>
>
>
>
> Date: 
>
> Fri, 13 Mar 2020 11:21:18 +0100
>
>
>
>
> 13 mars 2020 kl. 00.47 skrev Paul Eggert <address@hidden>:
>
> > This feels like Bug#39948. I suggest cloning the source directory from git 
> > master, and starting afresh. Merely wiping out the build directory does not 
> > always suffice (I don't know why).
>
> Thanks, and you were right. After carefully removing all non-versioned files 
> from the source directory, the problems went away.
> Closed.
>
>
>
> --- End Message ---