1 Revision history for Perl extension Time::HiRes.
4 - [rt.cpan.org #17442] 'make test' frequently fails under
5 Cygwin Perl v5.8.8, reported and patched by J. R. Hedden
6 (two race condition bugs in the END block in the case the
7 main process dies before the timer process, unearthed
8 by a bug in Cygwin ualarm)
11 - HiRes.t:s/ok 32/ok 33/, from Dominic Dunlop
12 - tighten up the clock() test marginally by requiring non-negative
13 - clock_nanosleep() and clock() doc tweaks
16 - the interface to clock_nanosleep() is more natural
17 when it is like (hires) time() (instead of like nanosleep),
18 and the .xs implementation of clock_nanosleep() in 1.84
20 - the semantics of clock() are not quite so silly as I thought,
21 but still somewhat odd, documented as such
22 - additional enhancements to the clock() documentation
23 - add test for clock_nanosleep() (I cannot test this
24 since none of my systems have the function)
25 - add test for clock()
28 - add clock() which returns the processor time in
29 (floating point) seconds since an arbitrary era
30 - add clock_nanosleep() which suspends the current
31 thread until either absolute time or for relative time
32 - [rt.cpan.org #16486] printf missing value in HiRes.t
33 - add constants CLOCKS_PER_SEC, CLOCK_SOFTTIME, TIMER_ABSTIME
37 - has_symbol() was wrong since e.g. ITIMER_VIRTUAL is exported
38 via @EXPORT_OK even when it is not available. This is heinous.
39 @EXPORT_OK should be determined at Makefile.PL time.
40 - be more lenient is testing clock_gettime(): allow more slop,
41 and retry up to three times, sleeping a random nap between
43 - human months are one-based (noticed by Anton Berezin)
46 - CLOCK_REALTIME is an enum value (of the clockid_t enum)
47 in HP-UX (and might be so elsewhere, too), debugged by
49 - include const-c.inc as late as possible (from Randy Kobes,
50 [rt.cpan.org #15552] to avoid undefined usleep() on Win32
53 - try to be more robust and consistent in the detection of
54 CLOCK_REALTIME and ITIMER_VIRTUAL in HiRes.t: the proper
59 eval 'import Time::HiRes qw($symbol)';
60 return 0 unless $@ eq '';
61 return exists ${"Time::HiRes::$symbol"};
68 in the test. All these moves are needed because
70 1) one cannot directly do eval 'Time::HiRes::FOO_BAR'
71 because FOO_BAR might have a true value of zero
72 (or in the general case an empty string or even undef)
74 2) In case FOO_BAR is not available in this platform,
75 &FOO_BAR avoids the bareword warning
77 - wait more (1.5 seconds instead of 0.1) for the CLOCK_REALTIME test
78 but expect the 'customary' slop of 0.20 instead of 0.25
79 - fixed inside a comment HAS_POLL -> TIME_HIRES_NANOSLEEP
80 - at the end of HiRest.t tell how close we were to termination
83 - Gisle noticed a mistake (using HAS_NANOSLEEP) in 1.79
86 - try nanosleep for emulating usleep -- may help in some weird
87 embedded realtime places which have nanosleep but neither usleep
88 nor select nor poll (doesn't have to be weird embedded realtime
89 place, though -- in many places usleep is nanosleep anyway)
90 - try poll for emulating usleep -- this may help some obscure/old
91 SVR4 places that have neither usleep nor select
92 - a redundant test guard in HiRes.t
95 - ITIMER_VIRTUAL detection in HiRes.t had problems (that we cannot
96 in the general case fail already at 'use' phase is suboptimal)
97 - fixes to the documentation of clock_gettime() and clock_getres()
100 - add support for the POSIX clock_gettime() and clock_getres(),
101 if available, either as library calls or as syscalls
102 - be more defensive about missing functionality: break out
103 early (during 'use') if no e.g. clock_getres() is available,
104 and protect our back by trapping those cases also in HiRes.xs
105 - the test added in 1.76 could cause an endless loop e.g. in Solaris,
106 due to mixing of sleep() and alarm() (bad programmer, no cookie!)
109 - testing for nanosleep had wrong logic which caused nanosleep
110 to become undefined for e.g. Mac OS X
111 - added a test for a core dump that was introduced by Perl 5.8.0
112 safe signals and was fixed for the time of 5.8.1 (one report of
113 the core dump was [perl #20920]), the test skipped pre-5.8.1.
114 - *cough* s/unanosleep/nanosleep/g; *cough*
117 - installation patch from Gisle Aas: in Perls 5.8.x and later
118 use MakeMaker INSTALLDIRS value of 'perl' instead of 'site'.
121 - [cpan #14608] Solaris 8 perl 5.005_03 File::Spec module does not have method rel2abs
122 (the workaround is not to use rel2abs, should not be necessary)
123 - [cpan #14642] U2time wrongly exported on the C API
124 (patch supplied by the reporter, SALVA@cpan.org)
125 - add release dates to Changes
128 - Time::HiRes::nanosleep support for Solaris [PATCH]
129 (POSIX::uname() not available if building with core perl,
130 from Gisle Aas, via perl5-porters, perl change #25295)
133 - going back to the 1.68 loader setup (using DynaLoader)
134 since too many weird things starting breaking
135 - fix a typo in José Auguste-Etienne's name
138 - a thinko in the nanosleep() detection
139 - move more changes stuff from the README to Changes
140 - add -w to the Makefile.PL
143 - oops in 1.69 about @ISA (not affecting anything but silly)
144 - add copyright 2005 to HiRes.pm
145 - add copyright and license to HiRes.xs
146 - add copyrights 2003, 2004, 2005 to README
149 - actually run a test for nanosleep
150 (if there is no $Config{d_nanosleep}) since e.g. in AIX 4.2
151 it seems that one can link in nanosleep() but then calling
152 it fails instantly and sets errno to ENOSYS (Not implemented).
153 This may be fixable in the AIX case by figuring out the right
154 (realtime POSIX?) libs and whatnot, but in the general case
155 running a real test case is better. (Of course, this change
156 will no doubt run into portability problems because of the
157 execution step...) Note that because of hysterical raisins
158 most Perls do NOT have $Config{d_nanosleep} (scanning for
159 it by Configure would in many platforms require linking in
160 things like -lrt, which would in many platforms be a bad idea
162 (from José Auguste-Etienne)
163 - support XSLoader also since it's much faster
164 (from Alexey Tourbin)
165 - add SEE ALSO (BSD::Resource and Time::TAI64)
168 - somehow 1.67 had a lot of doubled lines (a major cut-and-paste
169 error suspected), but miraculously it still worked since the
170 doubling took place below the __END__ token
171 - undef Pause() before defining it to avoid redefinition warnings
172 during compilation in case perl.h had already defined Pause()
173 (part of perl change #24271)
177 - (internal) don't ignore the return value of gettimeofday()
178 - (external) return undef or an empty if the C gettimeofday() fails
179 (affects Time::HiRes gettimeofday() and the hires time())
183 - fix the 'hierachy' typo in Makefile.PL [rt.cpan.org #8492]
184 - should now build in Solaris [rt.cpan.org #7165] (since 1.64)
185 - should now build in Cygwin [rt.cpan.org #7535] (since 1.64)
186 - close also [rt.cpan.org #5933] "Time::HiRes::time does not
187 pick up time adjustments like ntp" since ever reproducing it
188 (and therefore verifying a possible fix) in the same environment
189 has become rather unlikely
192 - one should not mix u?alarm and sleep (the tests modified
193 by 1.65, #12 and #13, hung in Solaris), now we just busy
194 loop executing an empty block
195 - in the documentation underline the unspecificity of mixing
197 - small spelling fixes
200 - regenerate ppport.h with Devel::PPPort 3.03,
201 now the MY_CXT_CLONE is defined in ppport.h,
202 we no more need to do that.
204 - the test #12 would often hang in sigsuspend() (at least that's
205 where Mac OS X' ktrace shows it hanging). With the sleep()s
206 changed to sleep(1)s, the tests still pass but no hang after
207 a few hundred repeats.
210 - Win32 and any ithread build: ppport.h didn't define
211 MY_CXT_CLONE, which seems to be a Time-HiRes-ism.
214 - Skip testing if under PERL_CORE and Time::HiRes has not
215 been Configured (from Marcus Holland-Moritz, core change
217 - Use ppport.h generated by Devel::PPPort 3.01,
218 allowing cutting away our own portability code.
219 - Don't use $ENV{PERL_CORE} for < 5.6.0.
220 - Don't use "for my $i" for <= 5.003.
221 - Don't use Pause() for <= 5.003.
222 - Can't use newSVpvf for <= 5.003.
223 (most of the changes from Marcus)
226 - Win32: reset reading from the performance counters every
227 five minutes to better track wall clock time (thanks to
228 PC timers being often quite bad), should help long-running
232 - Win32: Patch from Steve Hay
233 [PATCH] Re: [perl #30755] [Win32] Different results from Time::HiRes::gettimeofdayunder the debugger
234 to [perl #30755] reported by Nigel Sandever
236 - Cygwin: Use the Win32 recalibration code also in Cygwin if the
237 <w32api/windows.h> APIs are available. Cygwin testing by
238 Yitzchak Scott-Thoennes.
240 - Solaris: use -lposix4 to get nanosleep for Solaris 2.6,
241 after that keep using -lrt, patch from Alan Burlison,
242 bug reported in [cpan #7165]
245 - Change the Win32 recalibration limit to 0.5 seconds and tweak
246 the documentation to blather less about the gory details of the
247 Win32 implementation and more about the complications in general
248 of meddling with the system clock.
251 - Document the 1.57 change better.
254 - Win32/Cygwin/MinGW: if the performance counter drifts by more
255 than two seconds from the system clock (due to ntp adjustments,
256 for example), recalibrate our internal counter: from Jan Dubois,
257 based on [cpan #5933] by Jerry D. Hedden.
260 - Give a clearer message if the tests timeout (perl change #22253)
261 - Don't use /tmp or its moral equivalents (perl bug #15036,
265 - Windows: mingw32 patch from Mike Pomraning (use Perl's Const64()
266 instead of VC-specific i64 suffix)
269 - Solaris: like Tru64 (dec_osf) also Solaris need -lrt for nanosleep
272 - Windows: higher resolution time() by using the Windows
273 performance counter API, from Jan Dubois and Anton Shcherbinin.
274 The exact new higher resolution depends on the hardware,
275 but it should be quite a bit better than using the basic
279 - In AIX (v?) with perl 5.6.1 the HiRes.t can hang after
280 the subtest 18. No known analysis nor fix, but added
281 an alarm (that requires fork() and alarm()) to the test.
284 - doc tweaks from mjd (perl change #20456)
285 - NCR MP-RAS hints file added (svr4.pl) (perl change #21249)
288 - add a message (for non-core builds) to Makefile.PL about
289 the LC_ALL=C workaround
290 - &Time::HiRes::d_nanosleep was broken (perl change #20131)
291 - the nanosleep() probe was broken (perl change #20061)
292 - use existence instead of definedness for feature probes
294 - MPE/iX tweak (perl change #20042)
295 - do not use HAS_NANOSLEEP (perl change #19898)
298 - UVuf for non-IVSIZE platforms (from Keiichiro Nagano)
299 - OS/2 can always mix subsecond sleeps with signals
300 (part of perl change #19789)
303 - workaround for buggy gcc 2.95.3 in openbsd/sparc64
307 - do not use -lrt in Linux (from March Lehmann, perl change #19449)
308 - unnecessary (nanosleep is in libc anyway)
309 - harmful (-lrt slows down execution)
310 - incompatible (with many distributions' pthreads)
313 - do not create files in blib directories under core
314 (perl change #19160, from rgs)
315 - detypo s/VTLARM/VTARLM/ (perl change #19328, from mjd)
318 - guarantee that $xdefine in HiRes.t is always defined
319 (perl change #19109, from IlyaZ)
320 - a cleaner way to detect PERL_CORE (perl change #19111,
324 - add hints/irix.pl to turn off overly POSIX flags that
325 cause hide struct timespec to be hidden (and compilation
326 to fail) (bleadperl change #19085)
327 - documentation tweaks
330 - add c:/temp to the list of temp directories to probe
331 so that cygwin (and win*?) builds are happy. This was
332 needed at least in my cygwin 1.3.20/w2k setup.
335 - modernize the constants code (from Nicholas Clark)
338 - At some point the ability to figure our the correct incdir
339 for EXTERN.h (either a core perl build, or an installed perl)
340 had broken (which lead into all test compiles failing with
341 a core perl build, but thanks to the robustness of Makefile.PL
342 nothing of this was visible). The brokenness seemed to be
343 caused by $ENV{PERL_CORE} not being on for core builds?
344 Now stole a trick from the Encode that sets $ENV{PERL_CORE}
345 right, and both styles of build should work again.
348 - Nicholas Clark noticed that the my_catdir() emulation function
349 was broken (which means that we didn't really work for Perls
351 - inspired by fixing the above made the whole Makefile.PL -w
353 - tightened up the Makefile.PL output, less whitespace
356 - fix from Craig Berry for better building in VMS with PERL_CORE
359 - no functional changes
360 - move lib/Time/HiRes.pm as Hires.pm
361 - libraries scanning was slightly broken (always scanned
362 for a library even when $Config{libs} already had it)
365 - Ray Zimmerman ran into a race condition in Mac OS X.
366 A 0.01-second alarm fired before the test expected.
367 The test first slept indefinitely (blocking for signals)
368 and only after that tested for the signal having been sent.
369 Since the signal had already been sent, the test #12 never
370 completed. The solution: test first, then block.
371 - default to being silent on all probing attempts, set the
372 environment variable VERBOSE to a true value to see the
373 details (the probing command and the possible errors)
376 - do not clear MAN3PODS in Makefile.PL (Radoslaw Zielinski)
377 - INSTALLDIRS => 'perl' missing which means that Time::HiRes
378 cannot be upgraded from CPAN to override the 5.8.0 version
380 - Time::HiRes 1.35 could not be dropped as-is to bleadperl
381 because the include directories did not adjust themselves
382 if $ENV{PERL_CORE} (Hugo van der Sanden)
383 - add documentation about the restart of select() under alarm()
386 - small documentation tweaks
390 - better VMS operation (Craig Berry)
393 - our time machine is accelerating: now works with Perl 5.004_01
394 (tried with 5.003_07 and 5.002 but I get segmentation faults
395 from running the Makefile.PL with those in Tru64 4.0D)
398 - backward compatibility (pre-5.6.0) tweaks:
399 - no XSLoader in 5.00503, use DynaLoader instead
400 - no SvPV_nolen, either
401 - no PerlProc_pause(), either
402 - now tested with 5.00404 and 5.00503
403 - Makefile.PL requires 5.00404 (no more 5.002)
404 - use nanosleep instead of usleep, if it is available (Wilson Snyder)
405 (this means that one can mix subsecond sleeps with alarms)
406 - because of nanosleep we probe for -lrt and -lposix4
407 - the existence of getitimer/nanosleep/setitimer/ualarm/usleep
408 is available by exportable constants Time::HiRes::d_func
409 (since older Perl do not have them in %Config, and even
410 5.8.0 does not probe for nanosleep)
413 - backward compatibility (pre-5.6.1) tweaks:
414 - define NV if no NVTYPE
415 - define IVdf if needed (note: the Devel::PPPort
416 in 5.8.0 does not try hard hard enough since
417 the IVSIZE might not be defined)
418 - define NVgf if needed
419 - grab the typemap from 5.8.0 for the NV stuff
421 1.31 and 1.32 add more backward compatibility (now all the way
422 back to Perl 5.00404), and using nanosleep() (if available) for
427 - release 1.29_02 as 1.30
429 1.30 adds all the changes made during the Perl 5.6->5.7->5.8
430 development cycle. Most notably portability across platforms has been
431 enhanced, and the interval timers (setitimer, getitimer) have been
432 added. Note that the version of Time::HiRes that is included in Perl
433 5.8.0 calls itself 1.20_00, but it is equivalent to this Time::HiRes
434 version. Note also that in 1.30 Wegscheid turns over the maintenance
435 to Jarkko Hietaniemi.
439 - fix a silly unclosed comment typo in HiRes.xs
440 - document and export REALTIME_REALPROF (Solaris)
444 - only getitimer(ITIMER_REAL) available in Cygwin and Win32
445 (need to patch this also in Perl 5.[89])
446 - remove CVS revision log from HiRes.xs
450 The following numbered patches refer to the Perl 5.7 changes,
451 you can browse them at http://public.activestate.com/cgi-bin/perlbrowse
453 - 17558: Add #!./perl to the .t
454 - 17201: linux + usemorebits fix, from Rafael Garcia-Suarez
455 - 16198: political correctness, from Simon Cozens
456 - 15857: doc tweaks, from Jarkko Hietaniemi
457 - 15593: optimization in .xs, from Paul Green
458 - 14892: pod fixes, from Robin Barker
459 - 14100: VOS fixes, from Paul Green
460 - 13422: XS segfault, from Marc Lehmann
461 - 13378: whether select() gets restarted on signals, depends
462 - 13354: timing constraints, again, from Andy Dougherty
463 - 13278: can't do subsecond alarms with ualarm;
464 break out early if alarms do not seem to be working
465 - 13266: test relaxation (cygwin gets lower hires
466 times than lores ones)
467 - 12846: protect against high load, from Jarkko Hietaniemi
468 - 12837: HiRes.t VMS tweak, from Craig A. Berry
469 - 12797: HiRes.t VMS tweak, from Charles Lane
470 - 12769: HiRes.t VMS tweak, from Craig A. Berry
471 - 12744: gcc vs MS 64-bit constant syntax, from Nick Ing-Simmons
472 - 12722: VMS ualarm for VMS without ualarm, from Charles Lane
473 - 12692: alarm() ain't gonna work if ualarm() ain't,
474 from Gurusamy Sarathy
475 - 12680: minor VMS tweak, from Charles Lane
476 - 12617: don't try to print ints as IVs, from Jarkko Hietaniemi
477 - 12609: croak on negative time, from Jarkko Hietaniemi
478 - 12595: Cygwin rounds up for time(), from Jarkko Hietaniemi
479 - 12594: MacOS Classic timeofday, from Chris Nandor
480 - 12473: allow for more than one second for sleep() and usleep()
481 - 12458: test tuning, relax timing constraints,
482 from Jarkko Hietaniemi
483 - 12449: make sleep() and usleep() to return the number
484 of seconds and microseconds actually slept (analogously
485 with the builtin sleep()), also make usleep() croak if
486 asked for more than 1_000_000 useconds, from Jarkko Hietaniemi
487 - 12366: Time::HiRes for VMS pre-7.0, from Charles Lane
488 - 12199: do not use ftime on Win32, from Gurusamy Sarathy
489 - 12196: use ftime() on Win32, from Artur Bergman
490 - 12184: fix Time::HiRes gettimeofday() on Win32, from Gurusamy Sarathy
491 - 12105: use GetSystemTime() on Win32, from Artur Bergman
492 - 12060: explain the 1e9 seconds problem, from Jarkko Hietaniemi
493 - 11901: UNICOS sloppy division, from Jarkko Hietaniemi
494 - 11797: problem in HiRes.t, from John P. Linderman
495 - 11414: prototype from Time::HiRes::sleep(), from Abhijit Menon-Sen
496 - 11409: Time::HiRes qw(sleep) failed, from Abhijit Menon-Sen
497 - 11270: dynix/ptx 4.5.2 hints fix, from Peter Prymmer
498 - 11032: VAX VMS s/div/lib\$ediv/ fix, from Peter Prymmer
499 - 11011: VAX VMS s/qdiv/div/ fix, from Peter Prymmer
500 - 10953: SCO OpenServer 5.0.5 requires an explicit -lc for usleep(),
502 - 10942: MPE/IX test tweaks, from Mark Bixby
503 - 10784: unnecessary pod2man calls, from Andy Dougherty
504 - 10354: ext/ + -Wall, from Doug MacEachern
505 - 10320: fix the BOOT section to call myU2time correctly
506 - 10317: correct casting for AIX< from H. Merijn Brand
507 - 10119: document that the core time() may be rounding, not truncating
508 - 10118: test fix, from John Peacock
509 - 9988: long =item, from Robin Barker
510 - 9714: correct test output
511 - 9708: test also the scalar aspect of getitimer()
512 - 9705: Add interval timers (setitimer, getitimer)
513 - 9692: do not require at least 5.005 using XS
515 The following changes were made on top of the changes
516 made for Time::HiRes during the Perl 5.7 development
517 cycle that culminated in the release of Perl 5.8.0.
519 - add "require 5.005" to the Makefile.PL
520 - remove the REVISION section (CVS log) from HiRes.pm
521 - add jhi's copyright alongside Douglas'
522 - move HiRes.pm to lib/Time/
524 - modify HiRes.t to use $ENV{PERL_CORE}
525 - modify the original Time::HiRes version 1.20 Makefile.PL
526 to work both with Perl 5.8.0 and the new code with pre-5.8.0
527 Perls (tried with 5.6.1)
528 - tiny tweaks and updates in README and TODO
529 - bump the VERSION to 1.29
531 1.20 Wed Feb 24 21:30 1999
532 - make our usleep and ualarm substitutes into hrt_usleep
533 and hrt_ualarm. This helps static links of Perl with other
534 packages that also have usleep, etc. From
535 Ilya Zakharevich <ilya@math.ohio-state.edu>
536 - add C API stuff. From Joshua Pritikin
537 <joshua.pritikin@db.com>
538 - VMS Makefile.PL fun. From pvhp@forte.com (Peter Prymmer)
539 - hopefully correct "-lc" fix for SCO.
542 1.20 adds a platform neutral set of C accessible routines if you are
543 running 5.005+. All other changes are packaging changes and build
544 fixes(?) for statically linked Perl, SCO, and VMS.
546 1.19 Tue Sep 29 22:30 1998
547 - put VMS gettimeofday() in. Patch is from Sebastian Bazley
548 <seb@stian.demon.co.uk>
549 - change GIMME_V to GIMME to help people with older versions of
551 - fix Win32 version of gettimeofday(). It didn't affect anything,
552 but it confuses people reading the code when the return value
553 is backwards (0 is success).
554 - fix Makefile.PL (more) so that detection of gettimeofday is
557 1.19 has better VMS support.
559 1.18 Mon Jul 6 22:40 1998
560 - add usleep() for Win32.
561 - fix Makefile.PL to fix reported HP/UX feature where unresolved
562 externals still cause an executable to be generated (though no
563 x bit set). Thanks to David Kozinn for report and explanation.
564 Problems with the fix are mine :)
566 1.18 has limited Win32 support (no ualarm). Added usleep for Win32.
567 Probably buggy. I'm sure I'll hear.
569 1.17 Wed Jul 1 20:10 1998
570 - fix setitimer calls so microseconds is not more than 1000000.
571 Hp/UX 9 doesn't like that. Provided by Roland B Robert, PhD.
572 - make Win32. We only get gettimeofday (the select hack doesn't
573 seem to work on my Win95 system).
574 - fix test 4 on 01test.t. add test to see if time() and
575 Time::HiRes::time() are close.
577 1.16 Wed Nov 12 21:05 1997
578 - add missing EXTEND in new gettimeofday scalar code.
580 1.16+ should be closer to building out of the box on Linux. Thanks
581 to Gisle Aas for patches, and the ualarm equivalent using setitimer.
583 If your underlying operating system doesn't implement ualarm(), then
584 a fake using setitimer() will be made. If the OS is missing usleep(),
585 a fake one using select() will be made. If a fake can't be made for
586 either ualarm() or usleep(), then the corresponding Perl function will
587 not be available. If the OS is missing gettimeofday(), you will get
588 unresolved externals, either at link- or run-time.
590 This is an improvement; the package used to not even build if
591 you were missing any of these bits. Roderick Schertler
593 <roderick@argon.org> did all the conditional compilation stuff,
594 look at HiRes.pm and the test suites; it's good educational reading.
596 1.15 Mon Nov 10 21:30 1997
597 - HiRes.pm: update pod. Provided by Gisle Aas.
598 - HiRes.xs: if gettimeofday() called in scalar context, do
599 something more useful than before. Provided by Gisle Aas.
600 - README: tell of xsubpp '-nolinenumber' woes. thanks to
601 Edward Henigin <ed@texas.net> for pointing out the problem.
603 1.14 Wed Nov 5 9:40 1997
604 - Makefile.PL: look for setitimer
605 - HiRes.xs: if missing ualarm, but we have setitimer, make up
606 our own setitimer. These were provided by Gisle Aas.
608 1.13 Tue Nov 4 23:30 1997
609 - Makefile.PL: fix autodetect mechanism to do try linking in addition
610 to just compiling; should fix Linux build problem. Fix was provided
613 1.12 Sun Oct 12 12:00:00 1997
614 - Makefile.PL: set XSOPT to '-nolinenumbers' to work around xsubpp bug;
615 you may need to comment this back out if you have an older xsubpp.
616 - HiRes.xs: set PROTOTYPES: DISABLE
618 1.11 Fri Sep 05 16:00:00 1997
620 Had some line commented out that shouldn't have been (testing
623 Previous version was corrupted.
625 1.10 Thu May 22 20:20:00 1997
626 - HiRes.xs, HiRes.pm, t/*:
627 - only compile what we have OS support for (or can
629 - only test what we compiled
630 - gross improvement to the test suite
632 This work was all done by Roderick Schertler
633 <roderick@argon.org>. If you run Linux or
634 one of the other ualarm-less platforms, and you like this
635 module, let Roderick know; without him, it still wouldn't
636 be working on those boxes...
637 - Makefile.PL: figure out what routines the OS has and
638 only build what we need. These bits were written by Jarkko
639 Hietaniemi <jhi@iki.fi>. Again, gratitude is due...
641 1.02 Mon Dec 30 08:00:00 1996
642 - HiRes.pm: update documentation to say what to do when missing
643 ualarm() and friends.
644 - README: update to warn that ualarm() and friends need to exist
646 1.01 Fri Oct 17 08:00:00 1996
647 - Makefile.PL: make XSPROTOARGS => '-noprototyopes'
648 - HiRes.pm: put blank line between __END__ and =head1 so that
651 1.00 Tue Sep 03 13:00:00 1996
652 - original version; created by h2xs 1.16