This is the mail archive of the libc-alpha@sourceware.org mailing list for the glibc project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

Re: [PATCH v5 2/2] Y2038: make __tz_convert compatible with 64-bit-time


Hi Paul,

On Tue, 19 Jun 2018 16:35:53 -0700, Paul Eggert <eggert@cs.ucla.edu>
wrote :

> On 06/18/2018 12:14 PM, Albert ARIBAUD (3ADEV) wrote:
> > +#include <errno.h>  
> 
> Why does time/ctime.h need this addition? I don't see where the rest of 
> the patch is using anything that errno.h defines. Similarly for 
> time/ctime_r.c, time/gmtime.c, time/localtime.c. I suspect this is a 
> leftover of the older version of this patch that had __set_errno 
> (EINVAL), and I suggest not including <errno.h> in these files.

Indeed.

> >   /* Figure out the correct timezone for TM and set `__tzname',
> > -   `__timezone', and `__daylight' accordingly.  */
> > +   `__timezone', and `__daylight' accordingly.
> > +   NOTE: this takes a __time64_t value, so passing a time_t value is OK. */
> >   void
> > -__tz_compute (time_t timer, struct tm *tm, int use_localtime)
> > +__tz_compute (__time64_t timer, struct tm *tm, int use_localtime)  
> 
> Why is that "NOTE:" comment line important or needed? The proposed set 
> of patches never pass a time_t value to __tz_compute. I suspect this 
> line is a leftover from an older version of the patch, and suggest 
> removing that line.

You're right.

> Otherwise, the patch looks OK to me.

Thanks!

Cordialement,
Albert ARIBAUD
3ADEV


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]