- tz - lists.iana.org

If leap seconds go away, should TZif leap-second tables expire?
by Paul Eggert Nov. 29, 2023

Nov. 29, 2023

I have a question about leap-second table expiration in TZif files. Please bear with a longish explanation first; the actual question is at the end of this message. Since release 2020a, TZDB's zic command has supported "Expires" lines in the leapseconds file, and has propagated that info to TZif files using an extension to Internet RFC 8536. This extension is a bit of a hack: if a TZif file's last leap-second record has the same correction as the penultimate record, then the last record is the leap second expiry. For example, if the file 'leap-seconds.list' contains the line: #@ 3928521600 which stands for 28 June 2024; and if you run the command 'make EXPIRES_LINE=1 leapseconds', the file 'leapseconds' will contain: Expires 2024 Jun 28 00:00:00 and if you then run 'make install', the resulting TZif files with leap seconds will contain two leap transitions with correction 27, one transition for 1483228826 (representing 2016-12-31 23:59:60 UTC, the most recent leap second), and one transition for 1719532827 (representing 2017-06-28 00:00:00 UTC, the leap second table expiration). This Expires-line feature was introduced to TZDB in release 2020a, replacing an older hack using an "#expires" comment. However, in the interest of stability the feature has been disabled by default and you must use a non-default setting like "make EXPIRES_LINE=1" to enable it. Our current draft -09 revision to Internet RFC 8536 documents this new feature for the planned version 4 of the TZif file. See <https://datatracker.ietf.org/doc/draft-murchison-rfc8536bis/09/> section 3.2 item "corr(ection)", which says: "... in version 4 files with two or more leap-second records, the correction value of the last two records MAY be the same, with the occurrence of last record indicating the expiration time of the leap-second table." "The leap-second table expiration time is the time at which the table no longer records the presence or absence of future leap- second corrections, and post-expiration timestamps can not be accurately calculated. For example, a leap-second table published in January, which predicts the presence or absence of a leap second at June's end, might expire in mid-December because it is not known whether a leap second will occur at December's end. To complicate matters, a year ago the General Conference on Weights and Measures adopted a resolution calling for leap seconds to be phased out by 2035 <https://www.bipm.org/en/cgpm-2022/resolution-4>. If this is indeed done as late as possible there's a chance we could have our first-ever negative leap second before 2035 <https://insidegnss.com/will-we-have-a-negative-leap-second/> which could cause software havoc, and it's not clear whether the authorities would go through with this. One more thing: the World Radio Conference is currently meeting in Dubai and this meeting may take further action regarding leap seconds. Draft -09 currently has the following text about the possible impending demise of leap seconds: "If leap seconds become permanently discontinued, as requested by the General Conference on Weights and Measures [CGPM-2022-R4], leap-second tables published after the discontinuation time SHOULD NOT expire, since they will not be updated in the foreseeable future." With all the above in mind, here is my question. Should TZif leap-second tables have no expiration date if leap seconds are discontinued, as draft -09 suggests? Or should they have an expiration date of (say) January 2100, which is likely a lower bound on the date that the leap-second regime would be replaced by some other regime, such as leap minutes? Or should we do something else about TZif leap-second expiry after leap seconds are discontinued?

6 9

Re: [tz] If leap seconds go away, should TZif leap-second tables expire?
by Hal Murray Nov. 29, 2023

Nov. 29, 2023

Paul Eggert said: > Should TZif leap-second tables have no expiration date if leap seconds are > discontinued, as draft -09 suggests? Or should they have an expiration date > of (say) January 2100, which is likely a lower bound on the date that the > leap-second regime would be replaced by some other regime, such as leap > minutes? Or should we do something else about TZif leap-second expiry after > leap seconds are discontinued? I suggest waiting. Everybody else is going to have the same problem. As part of working out the no-leap details, somebody will have to pick a value so they can say something like "we won't have any leaps for the next N years". Then you can plug that number in. -- These are my opinions. I hate spam.

1 0

Re: [tz] localtime crash and fix
by Alois Treindl Nov. 29, 2023

Nov. 29, 2023

Thanks for fixing. I still find the code unnecessarily obscure due to the use of two unions stacked inside each other. > /* Input buffer for data read from a compiled tz file. */ > union input_buffer { > /* The first part of the buffer, interpreted as a header. */ > struct tzhead tzhead; > > /* The entire buffer. */ > char buf[2 * sizeof(struct tzhead) + 2 * sizeof(struct state) > + 4 * TZ_MAX_TIMES]; > }; > > /* TZDIR with a trailing '/' rather than a trailing '\0'. */ > static char const tzdirslash[sizeof TZDIR] = TZDIR "/"; > > /* Local storage needed for 'tzloadbody'. */ > union local_storage { > /* The results of analyzing the file's contents after it is opened. */ > struct file_analysis { > /* The input buffer. */ > union input_buffer u; > > /* A temporary state used for parsing a TZ string in the file. */ > struct state st; > } u; > > /* The file name to be opened. */ > char fullname[max(sizeof(struct file_analysis), sizeof tzdirslash + 1024)]; > }; The size of local_starage is in the range of 80 kb, not exactly something for a "small stack". The element buf is rather large. I think you save only 1kb extra spece for fullname, and some 30 bytes for tzhead. I do not think that these very minor saving justify the obscurity introduced in the code. It makes it hard to read and hard to debug. Please consider that others want to understand tzcode also. > Am 24.11.2023 um 23:14 schrieb Paul Eggert <eggert(a)cs.ucla.edu>: > > On 2023-11-24 08:22, Alois Treindl via tz wrote: >> The use of union to save memory is dangerous, in my opinion. There is no need in today's machines to save a few kilobytes of RAM for a process. >> Not even in embedded software for watches. > > When the memory is on the stack it still makes sense in some cases to save even a few kilobytes of RAM, as highly-threaded apps often have surprisingly small stacks. And anyway, the union has nothing to do with this particular bug.

2 1

[PROPOSED] Fix zic bug with -r @X where X is deduced from TZ
by Paul Eggert Nov. 27, 2023

Nov. 27, 2023

Problem reported by Ken Murchison for ‘zic -b slim -r @2145916800 asia’, which generated an Asia/Jerusalem TZif file with a single transition from -00 to IDT at 2145916800 (2038-01-01 00:00:00 UTC). The transition should be to IST. * NEWS: Mention this. * zic.c (main): Don’t let redundant_time be lower than lo_time. (outzone): Fix code for omitting trailing transitions when -R is specified, which was inadvertently messed up by the 2023-10-15 fix for Palestine after 2075. --- NEWS | 3 ++- zic.c | 15 +++++++++++---- 2 files changed, 13 insertions(+), 5 deletions(-) diff --git a/NEWS b/NEWS index 0d7fde80..c07cc7ad 100644 --- a/NEWS +++ b/NEWS @@ -37,7 +37,8 @@ Unreleased, experimental changes zic no longer mishandles data for Palestine after the year 2075. Previously, it incorrectly omitted post-2075 transitions that are - predicted for just before and just after Ramadan. + predicted for just before and just after Ramadan. (Thanks to Ken + Murchison for debugging help.) zic now works again on Linux 2.6.16 and 2.6.17 (2006). (Problem reported by Rune Torgersen.) diff --git a/zic.c b/zic.c index e10cb21b..b260851f 100644 --- a/zic.c +++ b/zic.c @@ -897,7 +897,8 @@ static zic_t const max_time = MAXVAL(zic_t, TIME_T_BITS_IN_FILE); static zic_t lo_time = MINVAL(zic_t, TIME_T_BITS_IN_FILE); static zic_t hi_time = MAXVAL(zic_t, TIME_T_BITS_IN_FILE); -/* The time specified by the -R option, defaulting to MIN_TIME. */ +/* The time specified by the -R option, defaulting to MIN_TIME; + or lo_time, whichever is greater. */ static zic_t redundant_time = MINVAL(zic_t, TIME_T_BITS_IN_FILE); /* The time specified by an Expires line, or negative if no such line. */ @@ -1109,6 +1110,8 @@ main(int argc, char **argv) fprintf(stderr, _("%s: -R time exceeds -r cutoff\n"), progname); return EXIT_FAILURE; } + if (redundant_time < lo_time) + redundant_time = lo_time; if (bloat == 0) { static char const bloat_default[] = ZIC_BLOAT_DEFAULT; if (strcmp(bloat_default, "slim") == 0) @@ -3442,6 +3445,9 @@ outzone(const struct zone *zpfirst, ptrdiff_t zonecount) if (defaulttype < 0) defaulttype = 0; if (!do_extend && !want_bloat()) { + /* Keep trailing transitions that are no greater than this. */ + zic_t keep_at_max; + /* The earliest transition into a time governed by the TZ string. */ zic_t TZstarttime = ZIC_MAX; for (i = 0; i < timecnt; i++) { @@ -3452,10 +3458,11 @@ outzone(const struct zone *zpfirst, ptrdiff_t zonecount) if (TZstarttime == ZIC_MAX) TZstarttime = nonTZlimtime; - /* Omit trailing transitions deducible from the TZ string. */ + /* Omit trailing transitions deducible from the TZ string, + and not needed for -r or -R. */ + keep_at_max = max(TZstarttime, redundant_time); for (i = j = 0; i < timecnt; i++) - if (redundant_time <= attypes[i].at - && attypes[i].at <= TZstarttime) { + if (attypes[i].at <= keep_at_max) { attypes[j].at = attypes[i].at; attypes[j].dontmerge = (attypes[i].at == TZstarttime && (nonTZlimtype != attypes[i].type -- 2.40.1

1 0

[PROPOSED 1/4] Fix bug with ‘make typecheck; make clean’
by Paul Eggert Nov. 25, 2023

Nov. 25, 2023

* Makefile (clean_misc): Also remove typecheck_*.dir. --- Makefile | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/Makefile b/Makefile index 8df3b621..edbc4d16 100644 --- a/Makefile +++ b/Makefile @@ -944,7 +944,7 @@ check_zishrink_posix check_zishrink_right: \ touch $@ clean_misc: - rm -fr check_*.dir + rm -fr check_*.dir typecheck_*.dir rm -f *.o *.out $(TIME_T_ALTERNATIVES) \ check_* core typecheck_* \ date tzselect version.h zdump zic libtz.a -- 2.40.1

1 3

Re: [tz] localtime crash and fix
by Alois Treindl Nov. 24, 2023

Nov. 24, 2023

That is effectively the same. But I would rather if Paul Eggert fixed the code. When I diff localtime.c from 2023c release with the current github version, there are changes which are obscure to me. The use of union to save memory is dangerous, in my opinion. There is no need in today's machines to save a few kilobytes of RAM for a process. Not even in embedded software for watches. It makes the code obscure. I have the gut feeling that the bug is in the line sp->goahead = ts->goahead; where ts has inherited old data from a previous call. On 24.11.23 16:54, Carlo wrote: > Dear Alois > > > Have you tried using this, letting the compiler perform the memset: > > #else > union local_storage ls = {0}; > return tzloadbody(name, sp, doextend, &ls); > #endif > > On Thu, Nov 23, 2023 at 7:32 PM Alois Treindl via tz <tz(a)iana.org> wrote: > > I have observed occasional crashes in localtime.c in the current > github version. > > In one of my applications, localtime() is called multiple times > for several zones. > > The condition when the segmentation fault appears are herd to > reproduce or demonstrate. > > This patch in function tzload() however solves the issue: > > --- localtime.c 2023-11-16 20:18:01.904577574 +0100 > +++ a/localtime.c 2023-11-23 19:22:21.062249899 +0100 > @@ -769,6 +769,7 @@ > } > #else > union local_storage ls; > + memset(&ls, 0, sizeof(ls)); > return tzloadbody(name, sp, doextend, &ls); > #endif > } >

2 1

localtime crash and fix
by Alois Treindl Nov. 24, 2023

Nov. 24, 2023

I have observed occasional crashes in localtime.c in the current github version. In one of my applications, localtime() is called multiple times for several zones. The condition when the segmentation fault appears are herd to reproduce or demonstrate. This patch in function tzload() however solves the issue: --- localtime.c 2023-11-16 20:18:01.904577574 +0100 +++ a/localtime.c 2023-11-23 19:22:21.062249899 +0100 @@ -769,6 +769,7 @@ } #else union local_storage ls; + memset(&ls, 0, sizeof(ls)); return tzloadbody(name, sp, doextend, &ls); #endif }

1 3

Did Greenland abolish daylight saving from 2024 on?
by Paul Eggert Nov. 20, 2023

Nov. 20, 2023

This week several English-language news sources reported that Greenland has moved its clocks forward for the last time. These reports disagree with current TZDB[1], which follows Thomas M. Steenholdt's December email[2] saying that although in November Greenlanders voted to not fall back this fall, they also voted to continue to observe daylight saving time (DST) next year, effectively moving their time zone one hour east instead of abolishing DST. There's no disagreement about timekeeping this year. The first point of disagreement is March 30, 2024 at 23:00, when TZDB says Nuuk, Greenland will spring forward to 24:00, whereas the recent news reports say clocks will neither spring forward nor fall back during 2024. So who's right: Steenholdt in December, or these several recent news reports? Currently TZDB goes with Steenholdt. After all, he cited the relevant Greenland legislation and his email address ends in ".gl". I suppose another possibility, though, is that Greenland's government has changed its mind since November. However, if so I'd like to see citations to the newer legislation. The earliest recent news story I found on this topic was a week ago by Lebawit Lily Girma[3], senior travel news reporter for Bloomberg Pursuits. I'm taking the liberty of bcc'ing Girma to see if she can help get to the bottom of the situation. [1]: https://iana.org/tz [2]: https://mm.icann.org/pipermail/tz/2022-December/032402.html [3]: Girma LL. Greenland Solves the Daylight Saving Time Debate. Bloomberg. 2023-03-24. https://www.bloomberg.com/news/articles/2023-03-24/greenland-keeps-daylight… Here are some of the other news reports I found. Greenland to stay in daylight saving time forever. AP. 2023-03-27. https://apnews.com/article/greenland-daylight-saving-summertime-utc-gmt-3b5… Martin M. Greenland is keeping daylight saving time permanently. NPR. 2023-03-28. https://www.wunc.org/2023-03-28/greenland-is-keeping-daylight-saving-time-p… Selcho M. Here’s why Greenland is permanently keeping daylight saving time. Deseret News. 2023-03-28. https://www.deseret.com/u-s-world/2023/3/28/23660379/why-greenland-keeping-… Jerusalem Post staff. Greenland changed their clocks to daylight savings for the last time. Jerusalem Post. 2023-03-29. https://www.jpost.com/international/article-735751 Carpineti A. Greenland Has Changed Its Clocks For The Last Time, Embracing Daylight Savings Forever. IFLScience. 2023-03-29. https://www.iflscience.com/greenland-has-changed-its-clocks-for-the-last-ti…

16 48

make error
by Alois Treindl Nov. 20, 2023

Nov. 20, 2023

I just made a new clone of tz. I work normally on RHEL 7 (redhat enterprise Linux) with gcc version 4.8.5 20150623 (Red Hat 4.8.5-44) (GCC) When I say make or make all, I get something I never observed before: cc -DTZDIR='"/usr/share/zoneinfo"' -c -o zic.o zic.c In file included from zic.c:16:0: private.h:791:60: error: conflicting types for ‘restrict’ struct tm *localtime_rz(timezone_t restrict, time_t const *restrict, ^ private.h:791:36: note: previous definition of ‘restrict’ was here struct tm *localtime_rz(timezone_t restrict, time_t const *restrict, ^ private.h:792:15: error: conflicting types for ‘restrict’ struct tm *restrict); ^ private.h:791:60: note: previous definition of ‘restrict’ was here struct tm *localtime_rz(timezone_t restrict, time_t const *restrict, ^ private.h:793:49: error: conflicting types for ‘restrict’ time_t mktime_z(timezone_t restrict, struct tm *restrict); ^ private.h:793:28: note: previous definition of ‘restrict’ was here time_t mktime_z(timezone_t restrict, struct tm *restrict); This does not happen when I compile on RHEL 9 gcc version 11.4.1 20230605 (Red Hat 11.4.1-2) (GCC) Can anyone point me to a fix?

7 17

[PROPOSED 1/5] * tz-link.html: Minor tweaks to citations.
by Paul Eggert Nov. 19, 2023

Nov. 19, 2023

--- tz-link.html | 10 +++++----- 1 file changed, 5 insertions(+), 5 deletions(-) diff --git a/tz-link.html b/tz-link.html index a29231d5..84bea935 100644 --- a/tz-link.html +++ b/tz-link.html @@ -994,9 +994,9 @@ are larger for countries farther away from the equator, while subtropical regions consume more electricity because of <abbr>DST</abbr>."</li> <li>Malow BA. It is time to abolish the clock change and adopt permanent standard time in the United States: a Sleep Research Society position statement. -Sleep. 2022;45(12):zsac236. +Sleep. 2022;45(12):zsac236. doi:<a href="https://doi.org/10.1093/sleep/zsac236">10.1093/sleep/zsac236</a>. -After reviewing of the scientific literature, the Sleep Research Society +After reviewing the scientific literature, the Sleep Research Society advocates permanent standard time due to its health benefits. <li>Rishi MA, Cheng JY, Strang AR et al. <a href="https://jcsm.aasm.org/doi/10.5664/jcsm.10898">Permanent standard time @@ -1004,15 +1004,15 @@ is the optimal choice for health and safety: an American Academy of Sleep Medicine position statement</a>. J Clin Sleep Med. 2023-10-31. doi:<a href="https://doi.org/10.5664/jcsm.10898">10.5664/jcsm.10898</a>. -This argues for permanent standard time due to health and safety risks +The AASM argues for permanent standard time due to health and safety risks and economic costs of both <abbr>DST</abbr> transitions and permanent <abbr>DST</abbr>.</li> <li>Roenneberg T, Wirz-Justice A, Skene DJ et al. <a href="https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7205184/">Why should we abolish Daylight Saving Time?</a> -J Biol Rhythms. 2019;34(3):227–230. +J Biol Rhythms. 2019;34(3):227–230. doi:<a href="https://doi.org/10.1177/0748730419854197">10.1177/0748730419854197</a>. -This position paper of the Society for Research on Biological Rhythms +The Society for Research on Biological Rhythms opposes DST changes and permanent DST, and advocates that governments adopt "permanent Standard Time for the health and safety of their citizens".</li> </ul> -- 2.40.1

1 4