GNU bug report logs - #76063
[PATCH 0/3] Switch to the Shepherd's system log

Previous Next

Full log

Message #29 received at 76063 <at> debbugs.gnu.org (full text, mbox):

From: Ludovic Courtès <ludo <at> gnu.org>
To: Maxim Cournoyer <maxim.cournoyer <at> gmail.com>
Cc: 76063 <at> debbugs.gnu.org
Subject: Re: [bug#76063] [PATCH 1/3] services: Add ‘system-log’ Shepherd service.
Date: Sun, 09 Feb 2025 12:05:20 +0100

Maxim Cournoyer <maxim.cournoyer <at> gmail.com> skribis:

> Not really relevant here, but that makes me recall that Guile currently
> lacks monotonic time support (time that can't move backward the
> adjusting the system clock).  It could make funny log time stamps.
>
> See in srfi-19.scm:
>
> ;; -- we define it to be the same as TAI.
> ;;    A different implementation of current-time-monotonic
> ;;    will require rewriting all of the time-monotonic converters,
> ;;    of course.
>
> (define (current-time-monotonic)
>   ;; Guile monotonic and TAI times are the same.
>   (let ((tai (current-time-tai)))
>     (make-time time-monotonic
>                (time-nanosecond tai)
>                (time-second tai))))
>
>
> which as far as I can see uses CLOCK_REALTIME and not CLOCK_MONOTONIC as
> it relies on gettimeofday(2), which has this note in its documentation:
>
>        The time returned by gettimeofday() is affected by discontinuous  jumps
>        in  the system time (e.g., if the system administrator manually changes
>        the system time).  If you need a monotonically  increasing  clock,  see
>        clock_gettime(2).
>
> Perhaps we should open an issue in Guile to remember to tacle this
> deficiency?

Yes indeed, that’s a problem!

(Monotonic time support is available by other means like
‘get-internal-run-time’ but it would be good to fix SRFI-19.)

Ludo’.

Previous Next

GNU bug tracking system
Copyright (C) 1999 Darren O. Benham, 1997,2003 nCipher Corporation Ltd, 1994-97 Ian Jackson.