1.9758 [2018-03-21]
- fix build in Win32 with Visual C by introducing a fake struct timezone
[rt.cpan.org #124844]
- in utime.t detect better being run in a noatime filesystem,
and if so, skip the test (for the HAMMER filesystem of DragonflyBSD)
- also for the HAMMER fs (if not in noatime), lower the expected subsecond
timestamp granularity to microseconds
- fix the version number typo in Changes: 1.9577 -> 1.9757
1.9577 [2018-03-15]
- fix win32 (and cygwin?) builds which have been broken since 1.9755:
problem was that gettimeofday() is a macro and needs to stay such
- regenerate ppport.h with Perl 5.26.1 (was previously generated with
5.25.6) and Devel::PPPort 3.35 (previously 3.36) -- this doesn't
change the ppport.h, though
1.9756 [2018-03-14]
- division by zero in new test tv_interval.t [rt.cpan.org #124775]:
made the test pass even if the difference is zero, but also
made it practically impossible to ever be zero by adding
a sub-second sleep.
- remove comments from inside qw [rt.cpan.org #124777],
only seemed to be caught by Perl 5.18.4
1.9755 [2018-03-14]
- adjust the error messages of tv_interval()
- the NV_1EX constants now cast to be NV so that
tv_interval() with long double builds does not
produce ugly results [rt.cpan.org #106456]
- add tests for tv_interval()
- centralize the mygettimeofday() logic
- make the mygettimeofday() a function instead of macro
- add the OS X 10.12+ clock_gettime() constants
CLOCK_MONOTONIC_RAW_APPROX CLOCK_UPTIME_RAW_APPROX
- regenerate the fallback files
Upstream changes:
1.9754 [2018-02-16]
- unify the gettimeofday() and time() forked code
that had near identical code paths for MacOS Classic
(which has two unusual features: unsigned time_t,
and a special tz struct member), and for non-MacOS Classic
- tv_interval should be implemented in XS
[rt.cpan.org #106456]
thanks to Sergey Aleynikov (suggested implementation simplified
by the above-mentioned MacOS Classic simplification)
1.9753 [2018-01-11]
- in t/clock.t in a fast system we need to burn more CPU,
reported and fix suggested by Joel C. Maslak <jmaslak@antelope.net>
[rt.cpan.org #124025]
- for t/utime.t in netbsd we need /sbin in PATH to find the mount command,
this is needed for the fix in 1.9751 to actually help netbsd with noatime
mounts.
(thanks to Nigel Horne)
Upstream changes:
1.9752 [2018-01-04]
- fix an error in the error message of utimensat() not available:
it said futimens() not available
- add --force alias for Makefile.PL --configure
1.9751 [2018-01-02]
- in macOS/OSX/Darwin, use __has_builtin() check also for utimensat(),
can cause errors like
"HiRes.xs:1474:16: error: unrecognized platform name macOS"
[rt.cpan.org #123994]
(oversight from 1.9749)
- do not define TIME_HIRES_STAT/d_hires_stat if none was found, instead
of defining it to be zero, which case has no implementation in hrstatns()
(thanks to Nigel Horne)
- in t/utime.t try to divine if the filesystem of the tempfiles has been
mounted with the 'noatime' option, which can prohibit updating the
access time timestamp. Also document this in HiRes.pm.
(thanks to Nigel Horne, original analysis by Slaven Rezic)
- synchronize the constant lists in HiRes.pm:@EXPORT_OK
and Makefile.PL:doConstants and regenerate fallback/const-c.inc
and fallback/const-xs.inc, this fixes Perl 5.6.2 issue with
d_futimens not allegedly being a valid macro in t/utime.t
(using Perl 5.26.1 for the regenerating, not 5.6.2)
(thanks to Nigel Horne)
- in t/utime.t define a nop sub done_testing for ancient Perls
(like Perl 5.6.2)
- in Perl 5.6.2 a bogus warning
"Use of uninitialized value in subroutine entry"
is issued from t/alarm.t: add a comment documenting that
1.9746 [2017-08-17]
- Unreliable t/usleep.t and t/utime.t tests [rt.cpan.org #122819]
Avoid testing for $dt = $t2 - $t1 and assuming $dt is less than
some value since a heavily loaded machine can delay the $t2.
Testing for greater than is fine.
- Tweak the configuring messages about subsecond stat/utime.
1.9745 [2017-08-14]
- Properly check for futimens/utimensat actually doing something.
This should handle 'gnukfreebsd' (which has only stubs, so far
we have been skipping the test) and as a new thing 'gnu' (Hurd)
(also only stubs). Thanks to Nigel Horne.
- Scan in t/utime.t whether the filesystem (wherever File::Temp
puts it tempfiles) supports subsecond timestamps. This removes
the fragile Linux/ext3 specific hack. As a side effect, the
setting of subsecond timestamps is tested only if reading of
them is supported. Thanks to Carsten Gaebler for the test idea,
and Ryan Voots for testing.
1.9744 [2017-07-27]
- add more potential clock constants, like CLOCK_MONOTONIC_FAST
(available in FreeBSD), and not all potentially found clock
constants were properly exported to be available from Perl,
see your system's clock_gettime() documentation for the available ones
1.9743 [2017-07-20]
- correct declared minimum Perl version (should be 5.6, was declared
as 5.8 since 1.9727_03): blead af94b3ac
- fix the fix for 'do file' to load hints in Makefile.PL: blead 3172fdbc
Upstream changes:
1.9739 [2016-06-28]
- the upcoming macOS 10.12 (Sierra, the operating system formerly
known as OS X, or Darwin) has implemented the clock_gettime()
and clock_getres() interfaces but not the clock_nanosleep()
1.9738 [2016-06-26]
- fix the dates in the Changes, it's not July yet
[rt.cpan.org #115553]
- in cygwin, drop the utime.t expected timestamp resolution
- documented for utime that the actual timestamp resolution depends
on the operating system and the filesystem, just like for stat
- looks like in gnukfreebsd 10.1 neither futimens or utimensat work
even though they exist, skipping utime tests there for now
- demodernize the Test::More use in utime.t (no subtest) to
better support older Perl installations
- for all the tests, drop the Test::More 0.82 requirement,
the only "modern" thing used is the note routine which
can be replaced with print("# ...\n") (or printf),
with this works back at least with Test::More 0.54 (perl 5.8.7)
1.9737 [2016-06-22]
- utime in HiRes.xs had dXSARGS which was unnecessary, leading
to two markstack pops, panic, and two-minute wait for the watchdog
(thanks to Dave Mitchell for the eagle-eye)
- even though 5.10.x should have the Perl_ck_warner, many CPAN
smokes proved otherwise, so requiring now at least 5.12 for
Perl_ck_warner use in IS_SAFE_PATHNAME()
1.9736 [2016-06-22]
- the IS_SAFE_PATHNAME() emulation was too modern for Perl 5.10
- make it so that only one value is set for -DTIME_HIRES_STAT
even on systems that support many options
- the 1.9734 (only released in blead, and as part of the development
release 5.25.2) fix for Darwin's clock_gettime was wrong under
threaded builds, leading into mutex panics in clock.t
- the d_hires_utime determination was made more precise
- the utime.t was failing in platforms where only one of
futimens/utimensat was present (neither or both was okay)
1.9735 [2016-06-20]
- Time::HiRes should override `utime` to allow setting hires
(futimens and utimensat support)
[rt.perl.org #114809]
- the utime patch used IS_SAFE_PATHNAME() which isn't available in
too old Perls, so emulate (in case the Devel::PPPort is too old)
1.9734 [2016-06-17]
- fix Darwins with clock_gettime: blead 2d41a263
[rt.perl.org #128427]
1.9733 [2016-04-23]
- C90 declaration-after-statement error with darwin threads: blead de1003b4
[rt.cpan.org #113856]
Upstream changes:
1.9732 [2016-03-13]
- MUTEX_LOCK, not PERL_MUTEX_LOCK: blead e5b02b5d
- also hrt_ualarm_itimer() is unused: 1cb6cce3
- the mutex needs init: 2d639e20
- version bump to 1.9732: 730d7fdc
1.9731 [2016-03-13]
- mark unused variable as such: blead a914236c
- OS X emulation mutex accidentally unused: da7a6455
- remove hrt_ualarm: 6da77c36
- pod error fixes: 919ca095
- nanosleep and clock_nanosleep cleanups: c8ea02b8..e3ff671b
- static funcs in HiRes.xs: 4e922c26
- Remove unused variable: 52ffb1b5
1.9730 [2016-02-17]
- TIME_HIRES_DONT_RUN_PROBES=1 to build the probes but not run them
[rt.cpan.org #111391]
1.9729 [2016-02-13]
- new constants
[rt.cpan.org #111170]
- clock_getres(), clock_gettime(), clock_nanosleep() emulation for OS X
[rt.cpan.org #78566]
1.9728 [2015-11-13]
- No changes since 1.9727_03
1.9727_03 [2015-11-06]
- use PERL_UNUSED_ARG() in the not-here-croak branches
- Makefile.PL/metadata improvements
1.9727_02 [2015-08-29]
- correct documentation to note that Time::NVtime is NV, not double
[perl #123879]
- avoid a warning about clock_t on OS X / clang 3.6 - 3.6
[rt.cpan.org #102718]
1.9727_01 [2015-08-17]
- Remove usage of 5.005 thread APIs in favor of 5.6 and up thread APIs
[rt.cpan.org #102799].
- Silence a C++ compiler warning [rt.cpan.org #101497].
- respect $Config{run} for cross-compiling [rt.cpan.org #92607]
Problems found with mismatching existing digests for:
distfiles/asclock-classic-1.0.tar.gz
distfiles/asclock-gtk-2.1.10beta.tar.gz
distfiles/asclock-xlib-2.0.11.tar.gz
distfiles/emiclock-2.0.2.tar.gz
Otherwise, existing SHA1 digests verified and found to be the same on
the machine holding the existing distfiles (morden). All existing
SHA1 digests retained for now as an audit trail.
Do it for all packages that
* mention perl, or
* have a directory name starting with p5-*, or
* depend on a package starting with p5-
like last time, for 5.18, where this didn't lead to complaints.
Let me know if you have any this time.
Upstream changes:
1.9726 [2013-08-17]
- Correct s/us splitting of usleep(1000000) [rt.cpan.org #78266].
- Avoid integer overflow in itimer-based alarm() with large
argument [rt.cpan.org #87160].
- Define PERL_NO_INLINE_FUNCTIONS during probe compilations, to
avoid false negatives caused by not linking with the perl core.
- Be more careful about context in stat().
- Install into "site" directories by default on perl 5.11+
[rt.cpan.org #79797].
- Fix a couple of doc typos [rt.cpan.org #85365].
- Fix function name in a doc example [rt.cpan.org #86318].
- Provide lstat() that yields high-res timestamps, alongside
the existing high-res stat() [rt.cpan.org #78732].
a) refer 'perl' in their Makefile, or
b) have a directory name of p5-*, or
c) have any dependency on any p5-* package
Like last time, where this caused no complaints.
1.9724nb1 to 1.9725.
upstream changes:
1.9725 [2012-02-01]
- Correct stack discipline in stat(), which was screwing up list
operations in expressions containing calls to it [rt.cpan.org
#72926].
- Add missing OUTPUT sections to the XS code [rt.cpan.org #70930].
- Skip itimer tests on GNU/Hurd, which has the API but lacks
the implementation [rt.cpan.org #72754].
- Fix a doubled word in the documentation [rt.cpan.org #72763].
time/p5-Time-HiRes from 1.9721nb2 to 1.9724.
Upstream changes:
1.9724 [2011-06-09]
- Correct XS parameter list, and therefore prototype, for
unimplemented-on-this-platform version of clock_nanosleep()
[rt.cpan.org #68700].
- Declare package variables with "our" rather than "use vars".
- Corresponding to "our" usage, check for minimum Perl version
5.006.
- Declare module dependencies.
1.9723 [2011-06-07]
- Remove $ENV{PERL_CORE} logic from test suite, which is no
longer desired in the core.
- Convert test suite to use Test::More.
- Factor out watchdog code from test suite.
- In test suite, be consistent about using fully-qualified form
of function names.
- Divide test suite into feature-specific scripts.
- Make ualarm timing test less vulnerable to delay-induced false
failure, from Dave Mitchell.
1.9722 [2011-05-18]
- Update for changes in build process in the core, patches
from BinGOs [rt.cpan.org #58858] and Craig Berry [rt.cpan.org
#63363].
- Fix broken linkage on Windows with gcc 3.4 seen with ActivePerl,
report from Christian Walde [rt.cpan.org #61648], fix derived
from Vincent Pit.
- Jump through hoops to avoid compiler warnings.
to trigger/signal a rebuild for the transition 5.10.1 -> 5.12.1.
The list of packages is computed by finding all packages which end
up having either of PERL5_USE_PACKLIST, BUILDLINK_API_DEPENDS.perl,
or PERL5_PACKLIST defined in their make setup (tested via
"make show-vars VARNAMES=..."), minus the packages updated after
the perl package update.
sno@ was right after all, obache@ kindly asked and he@ led the
way. Thanks!
Upstream changes:
1.9721 [2010-03-17]
- Address [rt.cpan.org #54196] alarm and ularm return values are bogus,
additional fix from Gisle Aas
- Address [rt.cpan.org #55665] "Bad plan" on Windows,
report and fix from Jan Dubois
1.9719 [2009-01-04]
- As with QNX, Haiku has the API of interval timers but not
the implementation (bleadperl change #34630), hence skip
the tests, via David Mitchell.
1.9718 [2008-12-31]
- .xs code cleanup from Albert Dvornik
- in the #39 and #40 do not do us I did, mixing alarm() and
sleep(). Now instead spin until enough time has passed.
1.9717 [2008-12-30]
- Skip the tests added in 1.9716 (#39, #40) if there's no subsecond
alarm capability, like with the older subsecond alarm tests
1.9716 [2008-12-26]
- Change documentation to agree with reality: there are
no interval timers in Win32.
- Address [rt.cpan.org #35899] (problem in subsecond sleeps),
add two tests to guard against this problem
- Address [rt.cpan.org #36600] 'Division by zero' failure in test suite
- Address [rt.cpan.org #37340] [PATCH] Address timer process in test
- Address [rt.cpan.org#40311 ] bad implementation of hrt_usleep
with TIME_HIRES_NANOSLEEP
to trigger/signal a rebuild for the transition 5.8.8 -> 5.10.0.
The list of packages is computed by finding all packages which end
up having either of PERL5_USE_PACKLIST, BUILDLINK_API_DEPENDS.perl,
or PERL5_PACKLIST defined in their make setup (tested via
"make show-vars VARNAMES=...").
1.9715 [2008-04-08]
- Silly me: Makefile.PL does need to accept arguments other than mine.
Some testing frameworks obviously do this.
- Add retrying for tests 34..37, which are the most commonly
failing tests. If this helps, consider extending the retry
framework to all the tests. [Inspired by Slaven Rezic,
[rt.cpan.org #34711] Occasional failures of test 35 or 36 (FreeBSD)]
1.9714 [2008-04-07]
- Under Perl 5.6.* NVgf needs to be "g", reported by Zefram,
it seems that ppport.h 3.13 gets this wrong.
- remove the check in Makefile.PL for 5.7.2, shouldn't be
(a) necessary (b) relevant
- add logic to Makefile.PL to skip configure/write Makefile
step if the "xdefine" file already exists, indicating that
the configure step has already been done, one can still
force (re)configure by "perl Makefile.PL configure",
or of course by "make clean && perl Makefile.PL".
1.9713 [2008-04-04]
- for alarm() and ualarm() [Perl] prefer setitimer() [C]
instead of ualarm() [C] since ualarm() [C] cannot portably
(and standards-compliantly) be used for more than 999_999
microseconds (rt.cpan.org #34655)
- it seems that HP-UX has started (at least in 11.31 ia64)
#defining the CLOCK_REALTIME et alia (instead of having
them just as enums)
- document all the diagnostics
1.9712 [2008-02-09]
- move the sub tick in the test file back to where it used to be
- in the "consider upgrading" message recommend at least Perl 5.8.8
and make the message to appear only for 5.8.0 since 5.8.1 and
later have the problem fixed
- VOS tweak for Makefile (core perl change #33259)
- since the test #17 seems to fail often, relax its limits a bit
1.9711 [2007-11-29]
- lost VMS test skippage from Craig Berry
- reformat the test code a little
1.9710 [2007-11-29]
- I got the sense of the QNX test the wrong way in an attempt
to generalize it for future
1.9709 [2007-11-28]
- casting fixes from Robin Barker for g++ and 64bitint
- in QNX skip the itimer tests because though the API
is there, the implementation isn't, from Matt Kraai
- raise the dead man timer to 180 seconds for really
slow/busy systems
- elaborate the UTF-8 locale warning from Makefile.PL
1.9708 [2007-10-05]
- [rt.cpan.org #29747]: Build failure with perl 5.005_05
Fixed by regenerating the ppport.h using Devel::PPPort 3.13.
1.9707 [2007-02-27]
- t/HiRes.t failed in Perl 5.6.2,
"action is not of type POSIX::SigAction at t/HiRes.t line 318",
reported and fixed by Anton Berezin, the reason was faulty
use of sigaction() when restoring the old action.
1.9706 [2007-02-25]
- with bleadperl in VMS the HiRes.t overrun the maximum number
of deferred signals because the libc SIGALRM was not strong
enough to interrupt select(), and select() got restarted every
time, solution is to use POSIX::SigAction if available.
A fix from Craig Berry (not 100% there, but helps).
- allow for more measuring noise for ualarm() tests 35..37
1.9705 [2007-02-06]
- nanosleep() and clock_nanosleep() detection and use were
quite broken; in Linux -lrt needed; fixes from Zefram
- [internal] slightly cleaner building of $DEFINE in Makefile.PL,
should avoid double/conflicting -D flags
1.9704 [2007-01-01]
- allow 10% of slop in test #14 (testing difference between
CORE::time() and Time::HiRes::time()), there seem to be often
transient failures from Perl smoke builds on this test
- small pod tweaks
1.9703 [2006-12-08]
- use int main(int argc, char **argv) consistently in Makefile.PL,
should help with
[rt.cpan.org #23868] nanosleep not detected under Mac OS 10.3.9 starting with Time::HiRes 1.96
- if someone still has the locale-broken Perl 5.8.0,
suggest that they upgrade their Perl
1.9702 [2006-12-06]
- restore the -DATLEASTFIVEOHOHFIVE, Win32 needed it still
1.9701 [2006-12-04]
- upgrade to ppport.h 3.10_02
- remove the -DATLEASTFIVEOHOHFIVE
- use the ppport.h PL_ppaddr, PL_statcache, PL_laststatval
- use the ppport.h aTHXR for calling Perl stat()
- switch into four-digit version since 2.0 is coming up
awfully fast but not feeling like a major rewrite
1.97 [2006-11-30]
- 1.95 broke building in Win32 (since pp_stat is not exported),
figured out how to call an op directly in 5.005 (use Perl_ppaddr
instead of PL_ppaddr)
- backport to Perl 5.004_05 (requires using statcache
and laststatval instead of PL_statcache and PL_laststatval)
(also checked to work in 5.005_04, 5.6.1, and 5.8.8 with threads)
1.96 [2006-11-30]
- 1.95 broke builds for threaded Perls, rt.cpan.org tickets:
[rt.cpan.org #23694] Time::HiRes fails tests on Solaris and Perl 5.6.1
[rt.cpan.org #23712] Time-HiRes 1.95 Fails make on AIX 5.2 with Perl 5.8.8
[rt.cpan.org #23730] Time::HiRes 1.95 fails make on MacOS X 10.3.9/perl 5.8.8
- use main() prototype consistently in Makefile.PL
1.95 [2006-11-29]
- integrate core change #29180: Silence VC++ compiler warnings
from Steve Hay
- do not use PL_ppaddr in stat() because that is not available
in Perl 5.005_04
- regenerate fallback/*.inc for older Perls without
ExtUtils::Constant because of d_hires_stat, resolves
[rt.cpan.org #23694] Time::HiRes fails tests on Solaris and Perl 5.6.1
- Make Makefile.PL more defensive against false PERL_CORE
1.94 [2006-10-16]
- file timestamps oddities seen: the atime and mtime
can be out of sync (modify first and read second can leave
atime < mtime) and mtime can be subsecond while atime is not.
So make the test more forgiving.
1.93 [2006-10-15]
- the ualarm() tests (34-37) assumed that ualarm(N)
could never alarm in less than N seconds, widened
the acceptable relative range to 0.9..1.5. Addresses
[rt.cpan.org #22090] and [rt.cpan.org #22091].
- skip the stat() tests in cygwin and win32, because
if run on FAT the timestamp granularity is only 2 seconds.
Any good way to detect (cygwin or win32) whether we are
being run on NTFS or anywhere with better timestamps?
Addresses [rt.cpan.org #22089] and [rt.cpan.org #22098].
1.92 [2006-10-13]
- scan for subsecond resolution timestamps in struct stat,
some known possibilities:
(1) struct timespec st_atimespec;
st_atimespec.tv_nsec;
(2) time_t st_atime;
long st_atimensec;
(3) time_t st_atime;
int st_atime_n;
(4) timestruc_t st_atim;
st_atim.tv_nsec
(5) time_t st_atime;
int st_uatime;
If something like this is found, one can do
use Time::HiRes;
my @stat = Time::HiRes::stat();
or even override the standard stat():
use Time::HiRes qw(stat);
to get the stat() timestamps
my ($atime, $mtime, $ctime) = @stat[8, 9, 10];
with subsecond resolution (assuming both the operating
system and the filesystem support that kind of thing).
Contributions for more systems (especially non-UNIX,
e.g. but not limited to: Win32, VMS, OS/2) gladly accepted.
(also more UNIX variants welcome: HP-UX? IRIX?)
Thanks to H.Merijn Brand, John Peacock, and Craig
Berry for brave beta testing.
1.91 [2006-09-29]
- ualarm() in SuSE 10.1 was overflowing after ~4.2 seconds,
possibly due to a glibc bug/feature (suspected overflow at
2**32 microseconds?), workaround by using the setitimer()
implementation of ualarm() if either useconds or
interval > 999_999 (this case seems to vary between systems:
are useconds more than 999_999 for ualarm() defined or not)
Added more ualarm() tests to catch various overflow points,
hopefully no problems in various platforms.
(The problem report by Mark Seger and Jon Paul Sullivan of HP.)
1.90 [2006-08-22]
- tweak still needed for Const64(), from Jerry Hedden
- get a freshly generated ppport.h
- update Copyright years
1.89 [2006-08-22]
- Const64() already appends an 'LL' (or i64), so provide LL and i64
forms for the IV_1E[679] (effects Win32 and Cygwin), reported by
Jerry Hedden.
- the Changes entry for 1.88 talked about [IN]V_1[679],
missing the 'E'.
1.88 [2006-08-21]
- clean up the g++ warnings in HiRes.xs, all of them
about mixing integer and floating point, introduce
constants IV_1E[679] and NV_1E[679]
1.87 [2006-02-13]
- [rt.cpan.org #17442] 'make test' frequently fails under
Cygwin Perl v5.8.8, reported and patched by J. R. Hedden
(two race condition bugs in the END block in the case the
main process dies before the timer process, unearthed
by a bug in Cygwin ualarm)
1.86 [2005-12-17]
- HiRes.t:s/ok 32/ok 33/, from Dominic Dunlop
- tighten up the clock() test marginally by requiring non-negative
- clock_nanosleep() and clock() doc tweaks
1.85 [2005-12-16]
- the interface to clock_nanosleep() is more natural
when it is like (hires) time() (instead of like nanosleep),
and the .xs implementation of clock_nanosleep() in 1.84
was broken anyway
- the semantics of clock() are not quite so silly as I thought,
but still somewhat odd, documented as such
- additional enhancements to the clock() documentation
- add test for clock_nanosleep() (I cannot test this
since none of my systems have the function)
- add test for clock()
1.84 [2005-12-16]
- add clock() which returns the processor time in
(floating point) seconds since an arbitrary era
- add clock_nanosleep() which suspends the current
thread until either absolute time or for relative time
- [rt.cpan.org #16486] printf missing value in HiRes.t
- add constants CLOCKS_PER_SEC, CLOCK_SOFTTIME, TIMER_ABSTIME
- tiny typo fixes
1.83 [2005-11-19]
- has_symbol() was wrong since e.g. ITIMER_VIRTUAL is exported
via @EXPORT_OK even when it is not available. This is heinous.
@EXPORT_OK should be determined at Makefile.PL time.
- be more lenient is testing clock_gettime(): allow more slop,
and retry up to three times, sleeping a random nap between
the retries
- human months are one-based (noticed by Anton Berezin)
1.82 [2005-10-06]
- CLOCK_REALTIME is an enum value (of the clockid_t enum)
in HP-UX (and might be so elsewhere, too), debugged by
H. Merijn Brand
- include const-c.inc as late as possible (from Randy Kobes,
[rt.cpan.org #15552] to avoid undefined usleep() on Win32
1.81 [2005-11-05]
- try to be more robust and consistent in the detection of
CLOCK_REALTIME and ITIMER_VIRTUAL in HiRes.t: the proper
way is
sub has_symbol {
my $symbol = shift;
eval 'import Time::HiRes qw($symbol)';
return 0 unless $@ eq '';
return exists ${"Time::HiRes::$symbol"};
}
and then use
&FOO_BAR
in the test. All these moves are needed because
1) one cannot directly do eval 'Time::HiRes::FOO_BAR'
because FOO_BAR might have a true value of zero
(or in the general case an empty string or even undef)
2) In case FOO_BAR is not available in this platform,
&FOO_BAR avoids the bareword warning
- wait more (1.5 seconds instead of 0.1) for the CLOCK_REALTIME test
but expect the 'customary' slop of 0.20 instead of 0.25
- fixed inside a comment HAS_POLL -> TIME_HIRES_NANOSLEEP
- at the end of HiRest.t tell how close we were to termination
1.80 [2005-11-04]
- Gisle noticed a mistake (using HAS_NANOSLEEP) in 1.79
1.79 [2005-11-03]
- try nanosleep for emulating usleep -- may help in some weird
embedded realtime places which have nanosleep but neither usleep
nor select nor poll (doesn't have to be weird embedded realtime
place, though -- in many places usleep is nanosleep anyway)
- try poll for emulating usleep -- this may help some obscure/old
SVR4 places that have neither usleep nor select
- a redundant test guard in HiRes.t
1.78 [2005-11-03]
- ITIMER_VIRTUAL detection in HiRes.t had problems (that we cannot
in the general case fail already at 'use' phase is suboptimal)
- fixes to the documentation of clock_gettime() and clock_getres()
1.77 [2005-11-03]
- add support for the POSIX clock_gettime() and clock_getres(),
if available, either as library calls or as syscalls
- be more defensive about missing functionality: break out
early (during 'use') if no e.g. clock_getres() is available,
and protect our back by trapping those cases also in HiRes.xs
- the test added in 1.76 could cause an endless loop e.g. in Solaris,
due to mixing of sleep() and alarm() (bad programmer, no cookie!)
1.76 [2005-10-22]
- testing for nanosleep had wrong logic which caused nanosleep
to become undefined for e.g. Mac OS X
- added a test for a core dump that was introduced by Perl 5.8.0
safe signals and was fixed for the time of 5.8.1 (one report of
the core dump was [perl #20920]), the test skipped pre-5.8.1.
- *cough* s/unanosleep/nanosleep/g; *cough*
1.75 [2005-10-18]
- installation patch from Gisle Aas: in Perls 5.8.x and later
use MakeMaker INSTALLDIRS value of 'perl' instead of 'site'.
1.74 [2005-09-19]
- [cpan #14608] Solaris 8 perl 5.005_03 File::Spec module does not have method rel2abs
(the workaround is not to use rel2abs, should not be necessary)
- [cpan #14642] U2time wrongly exported on the C API
(patch supplied by the reporter, SALVA@cpan.org)
- add release dates to Changes
1.73 [2005-08-16]
- Time::HiRes::nanosleep support for Solaris [PATCH]
(POSIX::uname() not available if building with core perl,
from Gisle Aas, via perl5-porters, perl change #25295)
1.72 [2005-07-01]
- going back to the 1.68 loader setup (using DynaLoader)
since too many weird things starting breaking
- fix a typo in José Auguste-Etienne's name
1.71 [2005-06-28]
- a thinko in the nanosleep() detection
- move more changes stuff from the README to Changes
- add -w to the Makefile.PL
1.70 [2005-06-26]
- oops in 1.69 about @ISA (not affecting anything but silly)
- add copyright 2005 to HiRes.pm
- add copyright and license to HiRes.xs
- add copyrights 2003, 2004, 2005 to README
1.69 [2005-06-25]
- actually run a test for nanosleep
(if there is no $Config{d_nanosleep}) since e.g. in AIX 4.2
it seems that one can link in nanosleep() but then calling
it fails instantly and sets errno to ENOSYS (Not implemented).
This may be fixable in the AIX case by figuring out the right
(realtime POSIX?) libs and whatnot, but in the general case
running a real test case is better. (Of course, this change
will no doubt run into portability problems because of the
execution step...) Note that because of hysterical raisins
most Perls do NOT have $Config{d_nanosleep} (scanning for
it by Configure would in many platforms require linking in
things like -lrt, which would in many platforms be a bad idea
for Perl itself).
(from José Auguste-Etienne)
- support XSLoader also since it's much faster
(from Alexey Tourbin)
- add SEE ALSO (BSD::Resource and Time::TAI64)
1.68 [2005-05-14]
- somehow 1.67 had a lot of doubled lines (a major cut-and-paste
error suspected), but miraculously it still worked since the
doubling took place below the __END__ token
- undef Pause() before defining it to avoid redefinition warnings
during compilation in case perl.h had already defined Pause()
(part of perl change #24271)
- minor doc tweaks
1.67 [2005-05-04]
- (internal) don't ignore the return value of gettimeofday()
- (external) return undef or an empty if the C gettimeofday() fails
(affects Time::HiRes gettimeofday() and the hires time())
I got few private comments that one should not use buildlink3.mk but the
standard DEPENDS statement because we do not link against libraries
in this case.
Changes since 1.59:
1.66
- add nanosleep()
- fix the 'hierachy' typo in Makefile.PL [rt.cpan.org #8492]
- should now build in Solaris [rt.cpan.org #7165] (since 1.64)
- should now build in Cygwin [rt.cpan.org #7535] (since 1.64)
- close also [rt.cpan.org #5933] "Time::HiRes::time does not pick up time adjustments like ntp" since ever reproducing it in the same environment
has become rather unlikely
1.65
- one should not mix u?alarm and sleep (the tests modified
by 1.65, #12 and #13, hung in Solaris), now we just busy
loop executing an empty block
- in the documentation underline the unspecificity of mixing
sleeps and alarms
- small spelling fixes
1.64
- regenerate ppport.h with Devel::PPPort 3.03,
now the MY_CXT_CLONE is defined in ppport.h,
we no more need to do that.
- the test #12 would often hang in sigsuspend() (at least that's
where Mac OS X' ktrace shows it hanging). With the sleep()s
changed to sleep(1)s, the tests still pass but no hang after
a few hundred repeats.
1.63
- Win32 and any ithread build: ppport.h didn't define
MY_CXT_CLONE, which seems to be a Time-HiResism.
1.62
- Skip testing if under PERL_CORE and Time::HiRes has not
been Configured (from Marcus Holland-Moritz, core change
#23246)
- Use ppport.h generated by Devel::PPPort 3.01,
allowing cutting away our own portability code.
- Don't use $ENV{PERL_CORE} for < 5.6.0.
- Don't use "for my $i" for <= 5.003.
- Don't use Pause() for <= 5.003.
- Can't use newSVpvf for <= 5.003.
(most of the changes from Marcus)
1.61
- Win32: reset reading from the performance counters every
five minutes to better track wall clock time (thanks to
PC timers being often quite bad), should help long-running
programs.
1.60
- Win32: Patch from Steve Hay
[PATCH] Re: [perl #30755] [Win32] Different results from Time::HiRes::gettimeofdayunder the debugger
to [perl #30755] reported by Nigel Sandever
- Cygwin: Use the Win32 recalibration code also in Cygwin if the
<w32api/windows.h> APIs are available. Cygwin testing by
Yitzchak Scott-Thoennes.
- Solaris: use -lposix4 to get nanosleep for Solaris 2.6,
after that keep using -lrt, patch from Alan Burlison,
bug reported in [cpan #7165]
module directory has changed (eg. "darwin-2level" vs.
"darwin-thread-multi-2level").
binary packages of perl modules need to be distinguishable between
being built against threaded perl and unthreaded perl, so bump the
PKGREVISION of all perl module packages and introduce
BUILDLINK_RECOMMENDED for perl as perl>=5.8.5nb5 so the correct
dependencies are registered and the binary packages are distinct.
addresses PR pkg/28619 from H. Todd Fujinaka.
