leap-second-enabled FreeBSD doesn't work right with R12B4 erts/emulator/beam/erl_time_sup.c; correction patch included

[Kenji Rikitake]

A patch to correct erlang:universaltime_to_localtime/1
for FreeBSD running leap-second-enabled timezone
by Kenji Rikitake <kenji.rikitake@...>
14-SEP-2008

* Summary

This patch fixes the time calculation problem of
FreeBSD 6.x and 7.x, which has the internal leap-second
correction enabled.
This patch is tested with Erlang/OTP R12B-4 source distribution.

* Symptom

Without this patch,
erlang:localtime_to_universaltime/1
and
erlang:universaltime_to_localtime/1
are not symmetric and will break
calendar:local_time_to_universal_time_dst_1/1
and
httpd_util:rfc1123_date/1

* Example of symptom:

(under where local time is GMT + 9 hours)

1> erlang:localtime_to_universaltime({{2008,9,1},{12,0,0}}).
{{2008,9,1},{3,0,0}}
2> erlang:universaltime_to_localtime({{2008,9,1},{3,0,0}}).
{{2008,9,1},{11,59,37}}

(Note that as of September 1, 2008, TAI - UTC = 33 seconds.
UNIX time_t with TAI correction is 10 seconds ahead of UTC.
So the 23-second difference occurs when the leap-second
correction is NOT performed, as in the C function of
univ_to_local() in erts/emulator/beam/erl_time_sup.c)

* Workaround in this patch

This patch changes the operation of
erlang:universaltime_to_localtime/1
so that the "universaltime" is handled properly
with leap-year correction.
(Note: OS time_t is in TAI)

See FreeBSD man time2posix(3) and
/usr/src/lib/libc/stdtime/localtime.c
for the further details.

This patch will NOT affect a FreeBSD machine without
leap-year correction; posix2time() will do nothing
in such a situation. (NOT tested though)

* Caveats, TODO and suggestions

There is no portable way to do this among UNIX-derived OSes

HAVE_POSIX2TIME should be set by configure

A Linux patch for leap-second supported systems is highly desired
(I found posix2time() is also available on Linux, but I'm not sure)

Should I rather do this using timegm(3) and throw away
the rather naive computation algorithm in univ_to_local()?
(Note that this is not necessarily portable either)

* How to apply this patch

Apply this at Erlang R12B4 source tree's directory under:
erts/emulator/beam

--- erl_time_sup.c.FCS 2008-04-07 22:57:50.000000000 +0900
+++ erl_time_sup.c 2008-09-14 09:56:10.000000000 +0900
@@ -71,6 +71,10 @@
 **
 */
 
+/* FreeBSD internal leap year correction function */
+/* define this for FreeBSD 6.x and 7.x */
+#define HAVE_POSIX2TIME
+
 #ifdef HAVE_CONFIG_H
 #  include "config.h"
 #endif
@@ -686,6 +690,18 @@
     
     the_clock = *second + 60 * (*minute + 60 * (*hour + 24 *
                                             gregday(*year, *month, *day)));
+#ifdef HAVE_POSIX2TIME
+    /*
+     * leap-second correction performed
+     * if system is configured so;
+     * do nothing if not
+     * See FreeBSD 6.x and 7.x
+     * /usr/src/lib/libc/stdtime/localtime.c
+     * for the details
+     */
+    the_clock = posix2time(the_clock);
+#endif
+
 #ifdef HAVE_LOCALTIME_R
     localtime_r(&the_clock, (tm = &tmbuf));
 #else


_______________________________________________
erlang-patches mailing list
erlang-patches@...
http://www.erlang.org/mailman/listinfo/erlang-patches

[Kenji Rikitake]

In the message <cf683a060809132003j724d7ef5m9cbae52f483e3043@...>
dated Sat, Sep 13, 2008 at 08:03:30PM -0700,
Matthew Dempsky <matthew.dempsky@...> writes:
> > Should I rather do this using timegm(3) and throw away
> > the rather naive computation algorithm in univ_to_local()?
> > (Note that this is not necessarily portable either)
>
> mktime(3) is defined by POSIX.

mktime() certainly is, but timegm() isn't, AFAIK.

An example implementation in the following URL will not be thread-safe,
due to the usage of getenv() and setenv().
http://developer.apple.com/documentation/Darwin/Reference/ManPages/man3/timegm.3.html

BTW local_to_univ() in erl_time_sup.c uses mktime(), and I see no
problem with it.

Kenji Rikitake



_______________________________________________
erlang-patches mailing list
erlang-patches@...
http://www.erlang.org/mailman/listinfo/erlang-patches

[Kenji Rikitake]

I also think using time2posix(3) and posix2time(3) will be a better way
than timegm(), as Matthew describes.

I've also found out that the following three functions

calendar:now_to_datetime/1
calendar:now_to_universal_time/1 (equivalent to calendar:now_to_datetime/1)
calendar:now_to_local_time/1

do not work in the leap-second-enabled environment, due to the fact that
erlang:now/0 shows the internal clock value as is, with gettimeofday(2).
The converted results of the functions include the offset of
((TAI-UTC) - 10)
seconds.

Modifying erlang:now/0, defined as get_now() in erl_time_sup.c to
include the offset of time2posix(3) is a possible solution, though I
don't feel like to doing it because it will surely break the assumption
of continuous monotonous increasing of erlang:now/0.

Fixing only the calendar module functions by adding a time2posix(3)
calculation routine written in C somewhere in the BEAM BIFs looks better
to me, though I need to investigate further.

Regards,
Kenji Rikitake

In the message <cf683a060809132100g6527575bi4a50e31293479d45@...>
dated Sat, Sep 13, 2008 at 09:00:20PM -0700,
Matthew Dempsky <matthew.dempsky@...> writes:

> On Sat, Sep 13, 2008 at 8:12 PM, Kenji Rikitake <kenji.rikitake@...> wrote:
> > mktime() certainly is, but timegm() isn't, AFAIK.
>
> My mistake.  I confused the local-vs-UTC time functions.
>
> time2posix(3) and posix2time(3) come from Olson's tz library, which
> according to Wikipedia[1], is pretty widely used, so they seem like
> they should be reasonably portable.  Anywhere that doesn't have them
> probably isn't leap second aware anyways.  They're also present in
> FreeBSD, NetBSD, and OpenBSD's cvs trees from over 10 years ago, so
> they're hardly recent additions either.
>
> [1] http://en.wikipedia.org/wiki/Zoneinfo#Use_in_software_systems
>

_______________________________________________
erlang-patches mailing list
erlang-patches@...
http://www.erlang.org/mailman/listinfo/erlang-patches

[Kenji Rikitake]

The attached patch includes an additional BIF
erlang:now_utc/0
to solve the problem, as well as the previous fix for
erlang:universaltime_to_localtime/1.

Adding a BIF is surely experimental, but I think this is the easiest way
to solve this issue.  Changing erlang:now/0 semantics is quite risky.

I'm still wondering if this is the best way to deal with the systems
with leap-second-enabled wall clocks (and OSes).  I'd appreciate if I
can read any comment from Erlang/OTP Team on this issue.

Regards,
Kenji Rikitake

In the message <20080914052418.GA10612@...>
dated Sun, Sep 14, 2008 at 02:23:55PM +0900,
Kenji Rikitake <kenji.rikitake@...> writes:

> I've also found out that the following three functions
>
> calendar:now_to_datetime/1
> calendar:now_to_universal_time/1 (equivalent to calendar:now_to_datetime/1)
> calendar:now_to_local_time/1
>
> do not work in the leap-second-enabled environment, due to the fact that
> erlang:now/0 shows the internal clock value as is, with gettimeofday(2).
> The converted results of the functions include the offset of
> ((TAI-UTC) - 10)
> seconds.
>
> Modifying erlang:now/0, defined as get_now() in erl_time_sup.c to
> include the offset of time2posix(3) is a possible solution, though I
> don't feel like to doing it because it will surely break the assumption
> of continuous monotonous increasing of erlang:now/0.
>
> Fixing only the calendar module functions by adding a time2posix(3)
> calculation routine written in C somewhere in the BEAM BIFs looks better
> to me, though I need to investigate further.

---

A patch to correct erlang:universaltime_to_localtime/1
and adding erlang:now_utc/0 new BIF for obtaining UTC
for FreeBSD running leap-second-enabled timezone
by Kenji Rikitake <kenji.rikitake@...>
18-SEP-2008

* Summary

This patch fixes the time calculation problem of
FreeBSD 6.x and 7.x, which has the internal leap-second
correction enabled.
This patch is tested with Erlang/OTP R12B-4 source distribution
on FreeBSD 6.3-RELEASE and 7.0-RELEASE port (lang/erlang).

* Symptom

Without this patch,
erlang:localtime_to_universaltime/1
and
erlang:universaltime_to_localtime/1
are not symmetric and will break
calendar:local_time_to_universal_time_dst_1/1
and
httpd_util:rfc1123_date/1.

Without this patch,
erlang:now/0
returns uncorrected time (of gettimeofday(2))
and represents TAI on leap-second-enabled system.
This will affect computations of
calendar:now_to_datetime/1
calendar:now_to_universal_time/1 (equivalent to calendar:now_to_datetime/1)
calendar:now_to_local_time/1
because the above three functions assume that
the argument (type Now) value is based on non-leap-second value.
This will break calendar:now_to_local_time/1.

* Example of symptom:

(under where local time is GMT + 9 hours)

1> erlang:localtime_to_universaltime({{2008,9,1},{12,0,0}}).
{{2008,9,1},{3,0,0}}
2> erlang:universaltime_to_localtime({{2008,9,1},{3,0,0}}).
{{2008,9,1},{11,59,37}}

(Note that as of September 1, 2008, TAI - UTC = 33 seconds.
UNIX time_t with TAI correction is 10 seconds ahead of UTC.
So the 23-second difference occurs when the leap-second
correction is NOT performed, as in the C function of
univ_to_local() in erts/emulator/beam/erl_time_sup.c)

* Workaround in this patch

This patch changes the operation of
erlang:universaltime_to_localtime/1
so that the "universaltime" is handled properly
with leap-year correction.
(Note: OS time_t is in TAI)

This patch adds a new BIF of
erlang:now_utc/0
which has the same semantics of erlang:now/0
while the return value corresponds to non-leap-second
value (of UTC) in leap-second-enabled system.
Using erlang:now_utc/0 instead of erlang:now/0
for
calendar:now_to_datetime/1
calendar:now_to_universal_time/1 (equivalent to calendar:now_to_datetime/1)
calendar:now_to_local_time/1
will return the correct values based on UTC.

See FreeBSD man time2posix(3), posix2time(3) and
/usr/src/lib/libc/stdtime/localtime.c
for the further details.

This patch will NOT affect a FreeBSD machine
without leap-year correction;
time2posix() and posix2time() will do nothing
in such a situation. (NOT tested though)

* Caveats, TODO and suggestions

There is no portable way to do this among UNIX-derived OSes.

HAVE_POSIX2TIME and HAVE_TIME2POSIX should be set by configure.

Linux, OpenBSD, NetBSD also have time2posix() and posix2time(),
though in other UNIX-derived OSes the availability is unconfirmed.

* How to apply this patch

Apply this at Erlang R12B4 source tree's directory under:
erts/emulator/beam
then recompile the distribution.

* Acknowledgement

Matthew Dempsky for his analysis of time2posix() and
posix2time() availability in various operating systems.

[End of patch document]

--- erl_time_sup.c.FCS 2008-04-07 22:57:50.000000000 +0900
+++ erl_time_sup.c 2008-09-18 10:27:03.000000000 +0900
@@ -71,6 +71,12 @@
 **
 */
 
+/* FreeBSD internal leap year correction function */
+/* those macros should be automatically configured */
+/* define this for FreeBSD 6.x and 7.x */
+#define HAVE_POSIX2TIME
+#define HAVE_TIME2POSIX
+
 #ifdef HAVE_CONFIG_H
 #  include "config.h"
 #endif
@@ -686,6 +692,18 @@
     
     the_clock = *second + 60 * (*minute + 60 * (*hour + 24 *
                                             gregday(*year, *month, *day)));
+#ifdef HAVE_POSIX2TIME
+    /*
+     * leap-second correction performed
+     * if system is configured so;
+     * do nothing if not
+     * See FreeBSD 6.x and 7.x
+     * /usr/src/lib/libc/stdtime/localtime.c
+     * for the details
+     */
+    the_clock = posix2time(the_clock);
+#endif
+
 #ifdef HAVE_LOCALTIME_R
     localtime_r(&the_clock, (tm = &tmbuf));
 #else
@@ -866,3 +884,53 @@
     
     erts_smp_mtx_unlock(&erts_timeofday_mtx);
 }
+
+/* beginning of extension by Kenji Rikitake */
+
+/* get a timestamp with time2posix correction */
+/* (subtract TAI-UTC offset from the get_now() value) */
+void
+get_now_utc(Uint* megasec, Uint* sec, Uint* microsec)
+{
+    SysTimeval now;
+    time_t now_utc_tvsec;
+    
+    erts_smp_mtx_lock(&erts_timeofday_mtx);
+    
+    get_tolerant_timeofday(&now);
+    do_erts_deliver_time(&now);
+
+    /* Make sure time is later than last */
+    if (then.tv_sec > now.tv_sec ||
+ (then.tv_sec == now.tv_sec && then.tv_usec >= now.tv_usec)) {
+ now = then;
+ now.tv_usec++;
+    }
+    /* Check for carry from above + general reasonability */
+    if (now.tv_usec >= 1000000) {
+ now.tv_usec = 0;
+ now.tv_sec++;
+    }
+    then = now;
+    
+    erts_smp_mtx_unlock(&erts_timeofday_mtx);
+
+#ifdef HAVE_TIME2POSIX
+    /* see gettimeofday(2) for the type difference issue */
+    /*
+     * leap-second correction performed
+     * if system is configured so;
+     * do nothing if not
+     * See FreeBSD 6.x and 7.x
+     * /usr/src/lib/libc/stdtime/localtime.c
+     * for the details
+     */
+    now_utc_tvsec = time2posix((time_t) now.tv_sec);
+#endif
+
+    *megasec = (Uint) (now.tv_sec / 1000000);
+    *sec = (Uint) (now_utc_tvsec % 1000000);
+    *microsec = (Uint) (now.tv_usec);
+}
+
+/* end of extension */
--- bif.c.FCS 2008-09-02 22:21:30.000000000 +0900
+++ bif.c 2008-09-18 10:28:55.000000000 +0900
@@ -3850,3 +3850,23 @@
     }
     BIF_RET(ret);
 }
+
+/* Kenji Rikitake mods */
+
+/**********************************************************************/
+
+
+/* return a timestamp with time2posix offset correction */
+BIF_RETTYPE now_utc_0(BIF_ALIST_0)
+{
+    Uint megasec, sec, microsec;
+    Eterm* hp;
+
+    get_now_utc(&megasec, &sec, µsec);
+    hp = HAlloc(BIF_P, 4);
+    BIF_RET(TUPLE3(hp, make_small(megasec), make_small(sec),
+   make_small(microsec)));
+}
+
+/**********************************************************************/
+
--- bif.tab.FCS 2008-09-01 21:50:45.000000000 +0900
+++ bif.tab 2008-09-18 10:30:18.000000000 +0900
@@ -724,3 +724,9 @@
 #
 
 bif erlang:hash/2
+
+### Kenji Rikitake extension
+
+bif erlang:now_utc/0
+
+### end of extension

---

_______________________________________________
erlang-patches mailing list
erlang-patches@...
http://www.erlang.org/mailman/listinfo/erlang-patches

[Björn Gustavsson]

On Sun, Sep 14, 2008 at 3:41 AM, Kenji Rikitake <kenji.rikitake@...> wrote:

A patch to correct erlang:universaltime_to_localtime/1
for FreeBSD running leap-second-enabled timezone
by Kenji Rikitake <kenji.rikitake@...>
14-SEP-2008

* Summary

This patch fixes the time calculation problem of
FreeBSD 6.x and 7.x, which has the internal leap-second
correction enabled.
This patch is tested with Erlang/OTP R12B-4 source distribution.

We will probably add the patch to R12B-5 with slight modifications.

/Bjorn

--
Björn Gustavsson, Erlang/OTP, Ericsson AB

_______________________________________________
erlang-patches mailing list
erlang-patches@...
http://www.erlang.org/mailman/listinfo/erlang-patches

[Kenji Rikitake]

The previous patch I posted on September 18, 2008 had a bug in the new
BIF erlang:now_utc/0 which I added.  This patch is a revised one for
R12B4, now also tested on a non-leap-second system (Ubuntu 8.04 LTS
Desktop).

Kenji Rikitake

---

A patch to correct erlang:universaltime_to_localtime/1
and adding erlang:now_utc/0 new BIF for obtaining UTC
for FreeBSD running leap-second-enabled timezone
by Kenji Rikitake <kenji.rikitake@...>
First version 18-SEP-2008
Revised version 31-OCT-2008

* Changelog

31-OCT-2008: revised the erl_time_sup.c patch for a non-leap-second
system, tested under Ubuntu 8.04 LTS Desktop edition, which did not
handle the variable megasec properly in get_now_utc() of
erlang:now_utc/0.

* Summary

This patch fixes the time calculation problem of
FreeBSD 6.x and 7.x, which has the internal leap-second
correction enabled.
This patch is tested with Erlang/OTP R12B-4 source distribution
on FreeBSD 6.3-RELEASE and 7.0-RELEASE port (lang/erlang).

* Symptom

Without this patch,
erlang:localtime_to_universaltime/1
and
erlang:universaltime_to_localtime/1
are not symmetric and will break
calendar:local_time_to_universal_time_dst_1/1
and
httpd_util:rfc1123_date/1.

Without this patch,
erlang:now/0
returns uncorrected time (of gettimeofday(2))
and represents TAI on leap-second-enabled system.
This will affect computations of
calendar:now_to_datetime/1
calendar:now_to_universal_time/1 (equivalent to calendar:now_to_datetime/1)
calendar:now_to_local_time/1
because the above three functions assume that
the argument (type Now) value is based on non-leap-second value.
This will break calendar:now_to_local_time/1.

* Example of symptom:

(under where local time is GMT + 9 hours)

1> erlang:localtime_to_universaltime({{2008,9,1},{12,0,0}}).
{{2008,9,1},{3,0,0}}
2> erlang:universaltime_to_localtime({{2008,9,1},{3,0,0}}).
{{2008,9,1},{11,59,37}}

(Note that as of September 1, 2008, TAI - UTC = 33 seconds.
UNIX time_t with TAI correction is 10 seconds ahead of UTC.
So the 23-second difference occurs when the leap-second
correction is NOT performed, as in the C function of
univ_to_local() in erts/emulator/beam/erl_time_sup.c)

* Workaround in this patch

This patch changes the operation of
erlang:universaltime_to_localtime/1
so that the "universaltime" is handled properly
with leap-year correction.
(Note: OS time_t is in TAI)

This patch adds a new BIF of
erlang:now_utc/0
which has the same semantics of erlang:now/0
while the return value corresponds to non-leap-second
value (of UTC) in leap-second-enabled system.
Using erlang:now_utc/0 instead of erlang:now/0
for
calendar:now_to_datetime/1
calendar:now_to_universal_time/1 (equivalent to calendar:now_to_datetime/1)
calendar:now_to_local_time/1
will return the correct values based on UTC.

See FreeBSD man time2posix(3), posix2time(3) and
/usr/src/lib/libc/stdtime/localtime.c
for the further details.

This patch will NOT affect a FreeBSD machine
without leap-year correction;
time2posix() and posix2time() will do nothing
in such a situation. (NOT tested though)

* Caveats, TODO and suggestions

There is no portable way to do this among UNIX-derived OSes.

HAVE_POSIX2TIME and HAVE_TIME2POSIX should be set by configure.

Linux, OpenBSD, NetBSD also have time2posix() and posix2time(),
though in other UNIX-derived OSes the availability is unconfirmed.

* How to apply this patch

Apply this at Erlang R12B4 source tree's directory under:
erts/emulator/beam
then recompile the distribution.

* Acknowledgement

Matthew Dempsky for his analysis of time2posix() and
posix2time() availability in various operating systems.

[End of patch document]

--- bif.tab.orig 2008-09-01 21:50:45.000000000 +0900
+++ bif.tab 2008-09-18 11:17:25.000000000 +0900
@@ -724,3 +724,9 @@
 #
 
 bif erlang:hash/2
+
+### Kenji Rikitake extension
+
+bif erlang:now_utc/0
+
+### end of extension
--- bif.c.orig 2008-09-02 22:21:30.000000000 +0900
+++ bif.c 2008-09-18 11:17:25.000000000 +0900
@@ -3850,3 +3850,23 @@
     }
     BIF_RET(ret);
 }
+
+/* Kenji Rikitake mods */
+
+/**********************************************************************/
+
+
+/* return a timestamp with time2posix offset correction */
+BIF_RETTYPE now_utc_0(BIF_ALIST_0)
+{
+    Uint megasec, sec, microsec;
+    Eterm* hp;
+
+    get_now_utc(&megasec, &sec, µsec);
+    hp = HAlloc(BIF_P, 4);
+    BIF_RET(TUPLE3(hp, make_small(megasec), make_small(sec),
+   make_small(microsec)));
+}
+
+/**********************************************************************/
+
--- erl_time_sup.c.orig 2008-04-07 22:57:50.000000000 +0900
+++ erl_time_sup.c 2008-10-31 08:17:00.000000000 +0900
@@ -71,6 +71,12 @@
 **
 */
 
+/* FreeBSD internal leap year correction function */
+/* those macros should be automatically configured */
+/* define this for FreeBSD 6.x and 7.x */
+#define HAVE_POSIX2TIME
+#define HAVE_TIME2POSIX
+
 #ifdef HAVE_CONFIG_H
 #  include "config.h"
 #endif
@@ -686,6 +692,18 @@
     
     the_clock = *second + 60 * (*minute + 60 * (*hour + 24 *
                                             gregday(*year, *month, *day)));
+#ifdef HAVE_POSIX2TIME
+    /*
+     * leap-second correction performed
+     * if system is configured so;
+     * do nothing if not
+     * See FreeBSD 6.x and 7.x
+     * /usr/src/lib/libc/stdtime/localtime.c
+     * for the details
+     */
+    the_clock = posix2time(the_clock);
+#endif
+
 #ifdef HAVE_LOCALTIME_R
     localtime_r(&the_clock, (tm = &tmbuf));
 #else
@@ -866,3 +884,56 @@
     
     erts_smp_mtx_unlock(&erts_timeofday_mtx);
 }
+
+/* beginning of extension by Kenji Rikitake */
+
+/* get a timestamp with time2posix correction */
+/* (subtract TAI-UTC offset from the get_now() value) */
+void
+get_now_utc(Uint* megasec, Uint* sec, Uint* microsec)
+{
+    SysTimeval now;
+    time_t now_utc_tvsec;
+    
+    erts_smp_mtx_lock(&erts_timeofday_mtx);
+    
+    get_tolerant_timeofday(&now);
+    do_erts_deliver_time(&now);
+
+    /* Make sure time is later than last */
+    if (then.tv_sec > now.tv_sec ||
+ (then.tv_sec == now.tv_sec && then.tv_usec >= now.tv_usec)) {
+ now = then;
+ now.tv_usec++;
+    }
+    /* Check for carry from above + general reasonability */
+    if (now.tv_usec >= 1000000) {
+ now.tv_usec = 0;
+ now.tv_sec++;
+    }
+    then = now;
+    
+    erts_smp_mtx_unlock(&erts_timeofday_mtx);
+
+#ifdef HAVE_TIME2POSIX
+    /* see gettimeofday(2) for the type difference issue */
+    /*
+     * leap-second correction performed
+     * if system is configured so;
+     * do nothing if not
+     * See FreeBSD 6.x and 7.x
+     * /usr/src/lib/libc/stdtime/localtime.c
+     * for the details
+     */
+    now_utc_tvsec = time2posix((time_t) now.tv_sec);
+#else /* HAVE_TIME2POSIX */
+    /* copy the original time value */
+    now_utc_tvsec = now.tv_sec;
+#endif /* HAVE_TIME2POSIX */
+
+    *megasec = (Uint) (now_utc_tvsec / 1000000);
+    *sec = (Uint) (now_utc_tvsec % 1000000);
+    *microsec = (Uint) (now.tv_usec);
+}
+
+/* end of extension */

---

_______________________________________________
erlang-patches mailing list
erlang-patches@...
http://www.erlang.org/mailman/listinfo/erlang-patches

[Kenji Rikitake]

I found out a part of my patch was incorporated in
R13A erts/emulator/beam/erl_time_sup.c, and I saw
a question in the comment that there was no test case
for checking whether the TAI (leap second) support
really works.

Here is a piece of code for testing on FreeBSD.
(see /usr/src/share/zoneinfo/Makefile for the details)
I think this will work on 7.x as well as 6.x.

For testing a TAI-based clock on FreeBSD
(with leap second support)
perform the following script on root
then the zone file will include the leap seconds:

# -- begin --
cd /usr/src/share/zoneinfo
make -DLEAPSECONDS clean install
# CAUTION: set to your local timezone
cp /usr/share/zoneinfo/Europe/Stockholm /etc/localtime
# -- end --

For putting back into non-leap-second clock on FreeBSD,
perform the following script on root:

# -- begin --
cd /usr/src/share/zoneinfo
make clean install
# CAUTION: set to your local timezone
cp /usr/share/zoneinfo/Europe/Stockholm /etc/localtime
# -- end --

These scripts will change the timezone definition *system-wide*.

I hope this helps and I appreciate you put in the code into R13A.

Regards,
Kenji Rikitake

In the message <6672d0160810170154h4f74cc8do7a79828363112729@...>
dated Fri, Oct 17, 2008 at 10:53:37AM +0200,
Bjorn Gustavsson <bgustavsson@...> writes:

> > A patch to correct erlang:universaltime_to_localtime/1
> > for FreeBSD running leap-second-enabled timezone
> > by Kenji Rikitake <kenji.rikitake@...>
> > 14-SEP-2008
> >
> > * Summary
> >
> > This patch fixes the time calculation problem of
> > FreeBSD 6.x and 7.x, which has the internal leap-second
> > correction enabled.
> > This patch is tested with Erlang/OTP R12B-4 source distribution.
> >
>
> We will probably add the patch to R12B-5 with slight modifications.
>
> /Bjorn

_______________________________________________
erlang-patches mailing list
erlang-patches@...
http://www.erlang.org/mailman/listinfo/erlang-patches