1 Revision history for Perl extension Time::HiRes.
4 - clean up the g++ warnings in HiRes.xs, all of them
5 about mixing integer and floating point, introduce
6 constants IV_1[679] and NV_1[679]
9 - [rt.cpan.org #17442] 'make test' frequently fails under
10 Cygwin Perl v5.8.8, reported and patched by J. R. Hedden
11 (two race condition bugs in the END block in the case the
12 main process dies before the timer process, unearthed
13 by a bug in Cygwin ualarm)
16 - HiRes.t:s/ok 32/ok 33/, from Dominic Dunlop
17 - tighten up the clock() test marginally by requiring non-negative
18 - clock_nanosleep() and clock() doc tweaks
21 - the interface to clock_nanosleep() is more natural
22 when it is like (hires) time() (instead of like nanosleep),
23 and the .xs implementation of clock_nanosleep() in 1.84
25 - the semantics of clock() are not quite so silly as I thought,
26 but still somewhat odd, documented as such
27 - additional enhancements to the clock() documentation
28 - add test for clock_nanosleep() (I cannot test this
29 since none of my systems have the function)
30 - add test for clock()
33 - add clock() which returns the processor time in
34 (floating point) seconds since an arbitrary era
35 - add clock_nanosleep() which suspends the current
36 thread until either absolute time or for relative time
37 - [rt.cpan.org #16486] printf missing value in HiRes.t
38 - add constants CLOCKS_PER_SEC, CLOCK_SOFTTIME, TIMER_ABSTIME
42 - has_symbol() was wrong since e.g. ITIMER_VIRTUAL is exported
43 via @EXPORT_OK even when it is not available. This is heinous.
44 @EXPORT_OK should be determined at Makefile.PL time.
45 - be more lenient is testing clock_gettime(): allow more slop,
46 and retry up to three times, sleeping a random nap between
48 - human months are one-based (noticed by Anton Berezin)
51 - CLOCK_REALTIME is an enum value (of the clockid_t enum)
52 in HP-UX (and might be so elsewhere, too), debugged by
54 - include const-c.inc as late as possible (from Randy Kobes,
55 [rt.cpan.org #15552] to avoid undefined usleep() on Win32
58 - try to be more robust and consistent in the detection of
59 CLOCK_REALTIME and ITIMER_VIRTUAL in HiRes.t: the proper
64 eval 'import Time::HiRes qw($symbol)';
65 return 0 unless $@ eq '';
66 return exists ${"Time::HiRes::$symbol"};
73 in the test. All these moves are needed because
75 1) one cannot directly do eval 'Time::HiRes::FOO_BAR'
76 because FOO_BAR might have a true value of zero
77 (or in the general case an empty string or even undef)
79 2) In case FOO_BAR is not available in this platform,
80 &FOO_BAR avoids the bareword warning
82 - wait more (1.5 seconds instead of 0.1) for the CLOCK_REALTIME test
83 but expect the 'customary' slop of 0.20 instead of 0.25
84 - fixed inside a comment HAS_POLL -> TIME_HIRES_NANOSLEEP
85 - at the end of HiRest.t tell how close we were to termination
88 - Gisle noticed a mistake (using HAS_NANOSLEEP) in 1.79
91 - try nanosleep for emulating usleep -- may help in some weird
92 embedded realtime places which have nanosleep but neither usleep
93 nor select nor poll (doesn't have to be weird embedded realtime
94 place, though -- in many places usleep is nanosleep anyway)
95 - try poll for emulating usleep -- this may help some obscure/old
96 SVR4 places that have neither usleep nor select
97 - a redundant test guard in HiRes.t
100 - ITIMER_VIRTUAL detection in HiRes.t had problems (that we cannot
101 in the general case fail already at 'use' phase is suboptimal)
102 - fixes to the documentation of clock_gettime() and clock_getres()
105 - add support for the POSIX clock_gettime() and clock_getres(),
106 if available, either as library calls or as syscalls
107 - be more defensive about missing functionality: break out
108 early (during 'use') if no e.g. clock_getres() is available,
109 and protect our back by trapping those cases also in HiRes.xs
110 - the test added in 1.76 could cause an endless loop e.g. in Solaris,
111 due to mixing of sleep() and alarm() (bad programmer, no cookie!)
114 - testing for nanosleep had wrong logic which caused nanosleep
115 to become undefined for e.g. Mac OS X
116 - added a test for a core dump that was introduced by Perl 5.8.0
117 safe signals and was fixed for the time of 5.8.1 (one report of
118 the core dump was [perl #20920]), the test skipped pre-5.8.1.
119 - *cough* s/unanosleep/nanosleep/g; *cough*
122 - installation patch from Gisle Aas: in Perls 5.8.x and later
123 use MakeMaker INSTALLDIRS value of 'perl' instead of 'site'.
126 - [cpan #14608] Solaris 8 perl 5.005_03 File::Spec module does not have method rel2abs
127 (the workaround is not to use rel2abs, should not be necessary)
128 - [cpan #14642] U2time wrongly exported on the C API
129 (patch supplied by the reporter, SALVA@cpan.org)
130 - add release dates to Changes
133 - Time::HiRes::nanosleep support for Solaris [PATCH]
134 (POSIX::uname() not available if building with core perl,
135 from Gisle Aas, via perl5-porters, perl change #25295)
138 - going back to the 1.68 loader setup (using DynaLoader)
139 since too many weird things starting breaking
140 - fix a typo in José Auguste-Etienne's name
143 - a thinko in the nanosleep() detection
144 - move more changes stuff from the README to Changes
145 - add -w to the Makefile.PL
148 - oops in 1.69 about @ISA (not affecting anything but silly)
149 - add copyright 2005 to HiRes.pm
150 - add copyright and license to HiRes.xs
151 - add copyrights 2003, 2004, 2005 to README
154 - actually run a test for nanosleep
155 (if there is no $Config{d_nanosleep}) since e.g. in AIX 4.2
156 it seems that one can link in nanosleep() but then calling
157 it fails instantly and sets errno to ENOSYS (Not implemented).
158 This may be fixable in the AIX case by figuring out the right
159 (realtime POSIX?) libs and whatnot, but in the general case
160 running a real test case is better. (Of course, this change
161 will no doubt run into portability problems because of the
162 execution step...) Note that because of hysterical raisins
163 most Perls do NOT have $Config{d_nanosleep} (scanning for
164 it by Configure would in many platforms require linking in
165 things like -lrt, which would in many platforms be a bad idea
167 (from José Auguste-Etienne)
168 - support XSLoader also since it's much faster
169 (from Alexey Tourbin)
170 - add SEE ALSO (BSD::Resource and Time::TAI64)
173 - somehow 1.67 had a lot of doubled lines (a major cut-and-paste
174 error suspected), but miraculously it still worked since the
175 doubling took place below the __END__ token
176 - undef Pause() before defining it to avoid redefinition warnings
177 during compilation in case perl.h had already defined Pause()
178 (part of perl change #24271)
182 - (internal) don't ignore the return value of gettimeofday()
183 - (external) return undef or an empty if the C gettimeofday() fails
184 (affects Time::HiRes gettimeofday() and the hires time())
188 - fix the 'hierachy' typo in Makefile.PL [rt.cpan.org #8492]
189 - should now build in Solaris [rt.cpan.org #7165] (since 1.64)
190 - should now build in Cygwin [rt.cpan.org #7535] (since 1.64)
191 - close also [rt.cpan.org #5933] "Time::HiRes::time does not
192 pick up time adjustments like ntp" since ever reproducing it
193 (and therefore verifying a possible fix) in the same environment
194 has become rather unlikely
197 - one should not mix u?alarm and sleep (the tests modified
198 by 1.65, #12 and #13, hung in Solaris), now we just busy
199 loop executing an empty block
200 - in the documentation underline the unspecificity of mixing
202 - small spelling fixes
205 - regenerate ppport.h with Devel::PPPort 3.03,
206 now the MY_CXT_CLONE is defined in ppport.h,
207 we no more need to do that.
209 - the test #12 would often hang in sigsuspend() (at least that's
210 where Mac OS X' ktrace shows it hanging). With the sleep()s
211 changed to sleep(1)s, the tests still pass but no hang after
212 a few hundred repeats.
215 - Win32 and any ithread build: ppport.h didn't define
216 MY_CXT_CLONE, which seems to be a Time-HiRes-ism.
219 - Skip testing if under PERL_CORE and Time::HiRes has not
220 been Configured (from Marcus Holland-Moritz, core change
222 - Use ppport.h generated by Devel::PPPort 3.01,
223 allowing cutting away our own portability code.
224 - Don't use $ENV{PERL_CORE} for < 5.6.0.
225 - Don't use "for my $i" for <= 5.003.
226 - Don't use Pause() for <= 5.003.
227 - Can't use newSVpvf for <= 5.003.
228 (most of the changes from Marcus)
231 - Win32: reset reading from the performance counters every
232 five minutes to better track wall clock time (thanks to
233 PC timers being often quite bad), should help long-running
237 - Win32: Patch from Steve Hay
238 [PATCH] Re: [perl #30755] [Win32] Different results from Time::HiRes::gettimeofdayunder the debugger
239 to [perl #30755] reported by Nigel Sandever
241 - Cygwin: Use the Win32 recalibration code also in Cygwin if the
242 <w32api/windows.h> APIs are available. Cygwin testing by
243 Yitzchak Scott-Thoennes.
245 - Solaris: use -lposix4 to get nanosleep for Solaris 2.6,
246 after that keep using -lrt, patch from Alan Burlison,
247 bug reported in [cpan #7165]
250 - Change the Win32 recalibration limit to 0.5 seconds and tweak
251 the documentation to blather less about the gory details of the
252 Win32 implementation and more about the complications in general
253 of meddling with the system clock.
256 - Document the 1.57 change better.
259 - Win32/Cygwin/MinGW: if the performance counter drifts by more
260 than two seconds from the system clock (due to ntp adjustments,
261 for example), recalibrate our internal counter: from Jan Dubois,
262 based on [cpan #5933] by Jerry D. Hedden.
265 - Give a clearer message if the tests timeout (perl change #22253)
266 - Don't use /tmp or its moral equivalents (perl bug #15036,
270 - Windows: mingw32 patch from Mike Pomraning (use Perl's Const64()
271 instead of VC-specific i64 suffix)
274 - Solaris: like Tru64 (dec_osf) also Solaris need -lrt for nanosleep
277 - Windows: higher resolution time() by using the Windows
278 performance counter API, from Jan Dubois and Anton Shcherbinin.
279 The exact new higher resolution depends on the hardware,
280 but it should be quite a bit better than using the basic
284 - In AIX (v?) with perl 5.6.1 the HiRes.t can hang after
285 the subtest 18. No known analysis nor fix, but added
286 an alarm (that requires fork() and alarm()) to the test.
289 - doc tweaks from mjd (perl change #20456)
290 - NCR MP-RAS hints file added (svr4.pl) (perl change #21249)
293 - add a message (for non-core builds) to Makefile.PL about
294 the LC_ALL=C workaround
295 - &Time::HiRes::d_nanosleep was broken (perl change #20131)
296 - the nanosleep() probe was broken (perl change #20061)
297 - use existence instead of definedness for feature probes
299 - MPE/iX tweak (perl change #20042)
300 - do not use HAS_NANOSLEEP (perl change #19898)
303 - UVuf for non-IVSIZE platforms (from Keiichiro Nagano)
304 - OS/2 can always mix subsecond sleeps with signals
305 (part of perl change #19789)
308 - workaround for buggy gcc 2.95.3 in openbsd/sparc64
312 - do not use -lrt in Linux (from March Lehmann, perl change #19449)
313 - unnecessary (nanosleep is in libc anyway)
314 - harmful (-lrt slows down execution)
315 - incompatible (with many distributions' pthreads)
318 - do not create files in blib directories under core
319 (perl change #19160, from rgs)
320 - detypo s/VTLARM/VTARLM/ (perl change #19328, from mjd)
323 - guarantee that $xdefine in HiRes.t is always defined
324 (perl change #19109, from IlyaZ)
325 - a cleaner way to detect PERL_CORE (perl change #19111,
329 - add hints/irix.pl to turn off overly POSIX flags that
330 cause hide struct timespec to be hidden (and compilation
331 to fail) (bleadperl change #19085)
332 - documentation tweaks
335 - add c:/temp to the list of temp directories to probe
336 so that cygwin (and win*?) builds are happy. This was
337 needed at least in my cygwin 1.3.20/w2k setup.
340 - modernize the constants code (from Nicholas Clark)
343 - At some point the ability to figure our the correct incdir
344 for EXTERN.h (either a core perl build, or an installed perl)
345 had broken (which lead into all test compiles failing with
346 a core perl build, but thanks to the robustness of Makefile.PL
347 nothing of this was visible). The brokenness seemed to be
348 caused by $ENV{PERL_CORE} not being on for core builds?
349 Now stole a trick from the Encode that sets $ENV{PERL_CORE}
350 right, and both styles of build should work again.
353 - Nicholas Clark noticed that the my_catdir() emulation function
354 was broken (which means that we didn't really work for Perls
356 - inspired by fixing the above made the whole Makefile.PL -w
358 - tightened up the Makefile.PL output, less whitespace
361 - fix from Craig Berry for better building in VMS with PERL_CORE
364 - no functional changes
365 - move lib/Time/HiRes.pm as Hires.pm
366 - libraries scanning was slightly broken (always scanned
367 for a library even when $Config{libs} already had it)
370 - Ray Zimmerman ran into a race condition in Mac OS X.
371 A 0.01-second alarm fired before the test expected.
372 The test first slept indefinitely (blocking for signals)
373 and only after that tested for the signal having been sent.
374 Since the signal had already been sent, the test #12 never
375 completed. The solution: test first, then block.
376 - default to being silent on all probing attempts, set the
377 environment variable VERBOSE to a true value to see the
378 details (the probing command and the possible errors)
381 - do not clear MAN3PODS in Makefile.PL (Radoslaw Zielinski)
382 - INSTALLDIRS => 'perl' missing which means that Time::HiRes
383 cannot be upgraded from CPAN to override the 5.8.0 version
385 - Time::HiRes 1.35 could not be dropped as-is to bleadperl
386 because the include directories did not adjust themselves
387 if $ENV{PERL_CORE} (Hugo van der Sanden)
388 - add documentation about the restart of select() under alarm()
391 - small documentation tweaks
395 - better VMS operation (Craig Berry)
398 - our time machine is accelerating: now works with Perl 5.004_01
399 (tried with 5.003_07 and 5.002 but I get segmentation faults
400 from running the Makefile.PL with those in Tru64 4.0D)
403 - backward compatibility (pre-5.6.0) tweaks:
404 - no XSLoader in 5.00503, use DynaLoader instead
405 - no SvPV_nolen, either
406 - no PerlProc_pause(), either
407 - now tested with 5.00404 and 5.00503
408 - Makefile.PL requires 5.00404 (no more 5.002)
409 - use nanosleep instead of usleep, if it is available (Wilson Snyder)
410 (this means that one can mix subsecond sleeps with alarms)
411 - because of nanosleep we probe for -lrt and -lposix4
412 - the existence of getitimer/nanosleep/setitimer/ualarm/usleep
413 is available by exportable constants Time::HiRes::d_func
414 (since older Perl do not have them in %Config, and even
415 5.8.0 does not probe for nanosleep)
418 - backward compatibility (pre-5.6.1) tweaks:
419 - define NV if no NVTYPE
420 - define IVdf if needed (note: the Devel::PPPort
421 in 5.8.0 does not try hard hard enough since
422 the IVSIZE might not be defined)
423 - define NVgf if needed
424 - grab the typemap from 5.8.0 for the NV stuff
426 1.31 and 1.32 add more backward compatibility (now all the way
427 back to Perl 5.00404), and using nanosleep() (if available) for
432 - release 1.29_02 as 1.30
434 1.30 adds all the changes made during the Perl 5.6->5.7->5.8
435 development cycle. Most notably portability across platforms has been
436 enhanced, and the interval timers (setitimer, getitimer) have been
437 added. Note that the version of Time::HiRes that is included in Perl
438 5.8.0 calls itself 1.20_00, but it is equivalent to this Time::HiRes
439 version. Note also that in 1.30 Wegscheid turns over the maintenance
440 to Jarkko Hietaniemi.
444 - fix a silly unclosed comment typo in HiRes.xs
445 - document and export REALTIME_REALPROF (Solaris)
449 - only getitimer(ITIMER_REAL) available in Cygwin and Win32
450 (need to patch this also in Perl 5.[89])
451 - remove CVS revision log from HiRes.xs
455 The following numbered patches refer to the Perl 5.7 changes,
456 you can browse them at http://public.activestate.com/cgi-bin/perlbrowse
458 - 17558: Add #!./perl to the .t
459 - 17201: linux + usemorebits fix, from Rafael Garcia-Suarez
460 - 16198: political correctness, from Simon Cozens
461 - 15857: doc tweaks, from Jarkko Hietaniemi
462 - 15593: optimization in .xs, from Paul Green
463 - 14892: pod fixes, from Robin Barker
464 - 14100: VOS fixes, from Paul Green
465 - 13422: XS segfault, from Marc Lehmann
466 - 13378: whether select() gets restarted on signals, depends
467 - 13354: timing constraints, again, from Andy Dougherty
468 - 13278: can't do subsecond alarms with ualarm;
469 break out early if alarms do not seem to be working
470 - 13266: test relaxation (cygwin gets lower hires
471 times than lores ones)
472 - 12846: protect against high load, from Jarkko Hietaniemi
473 - 12837: HiRes.t VMS tweak, from Craig A. Berry
474 - 12797: HiRes.t VMS tweak, from Charles Lane
475 - 12769: HiRes.t VMS tweak, from Craig A. Berry
476 - 12744: gcc vs MS 64-bit constant syntax, from Nick Ing-Simmons
477 - 12722: VMS ualarm for VMS without ualarm, from Charles Lane
478 - 12692: alarm() ain't gonna work if ualarm() ain't,
479 from Gurusamy Sarathy
480 - 12680: minor VMS tweak, from Charles Lane
481 - 12617: don't try to print ints as IVs, from Jarkko Hietaniemi
482 - 12609: croak on negative time, from Jarkko Hietaniemi
483 - 12595: Cygwin rounds up for time(), from Jarkko Hietaniemi
484 - 12594: MacOS Classic timeofday, from Chris Nandor
485 - 12473: allow for more than one second for sleep() and usleep()
486 - 12458: test tuning, relax timing constraints,
487 from Jarkko Hietaniemi
488 - 12449: make sleep() and usleep() to return the number
489 of seconds and microseconds actually slept (analogously
490 with the builtin sleep()), also make usleep() croak if
491 asked for more than 1_000_000 useconds, from Jarkko Hietaniemi
492 - 12366: Time::HiRes for VMS pre-7.0, from Charles Lane
493 - 12199: do not use ftime on Win32, from Gurusamy Sarathy
494 - 12196: use ftime() on Win32, from Artur Bergman
495 - 12184: fix Time::HiRes gettimeofday() on Win32, from Gurusamy Sarathy
496 - 12105: use GetSystemTime() on Win32, from Artur Bergman
497 - 12060: explain the 1e9 seconds problem, from Jarkko Hietaniemi
498 - 11901: UNICOS sloppy division, from Jarkko Hietaniemi
499 - 11797: problem in HiRes.t, from John P. Linderman
500 - 11414: prototype from Time::HiRes::sleep(), from Abhijit Menon-Sen
501 - 11409: Time::HiRes qw(sleep) failed, from Abhijit Menon-Sen
502 - 11270: dynix/ptx 4.5.2 hints fix, from Peter Prymmer
503 - 11032: VAX VMS s/div/lib\$ediv/ fix, from Peter Prymmer
504 - 11011: VAX VMS s/qdiv/div/ fix, from Peter Prymmer
505 - 10953: SCO OpenServer 5.0.5 requires an explicit -lc for usleep(),
507 - 10942: MPE/IX test tweaks, from Mark Bixby
508 - 10784: unnecessary pod2man calls, from Andy Dougherty
509 - 10354: ext/ + -Wall, from Doug MacEachern
510 - 10320: fix the BOOT section to call myU2time correctly
511 - 10317: correct casting for AIX< from H. Merijn Brand
512 - 10119: document that the core time() may be rounding, not truncating
513 - 10118: test fix, from John Peacock
514 - 9988: long =item, from Robin Barker
515 - 9714: correct test output
516 - 9708: test also the scalar aspect of getitimer()
517 - 9705: Add interval timers (setitimer, getitimer)
518 - 9692: do not require at least 5.005 using XS
520 The following changes were made on top of the changes
521 made for Time::HiRes during the Perl 5.7 development
522 cycle that culminated in the release of Perl 5.8.0.
524 - add "require 5.005" to the Makefile.PL
525 - remove the REVISION section (CVS log) from HiRes.pm
526 - add jhi's copyright alongside Douglas'
527 - move HiRes.pm to lib/Time/
529 - modify HiRes.t to use $ENV{PERL_CORE}
530 - modify the original Time::HiRes version 1.20 Makefile.PL
531 to work both with Perl 5.8.0 and the new code with pre-5.8.0
532 Perls (tried with 5.6.1)
533 - tiny tweaks and updates in README and TODO
534 - bump the VERSION to 1.29
536 1.20 Wed Feb 24 21:30 1999
537 - make our usleep and ualarm substitutes into hrt_usleep
538 and hrt_ualarm. This helps static links of Perl with other
539 packages that also have usleep, etc. From
540 Ilya Zakharevich <ilya@math.ohio-state.edu>
541 - add C API stuff. From Joshua Pritikin
542 <joshua.pritikin@db.com>
543 - VMS Makefile.PL fun. From pvhp@forte.com (Peter Prymmer)
544 - hopefully correct "-lc" fix for SCO.
547 1.20 adds a platform neutral set of C accessible routines if you are
548 running 5.005+. All other changes are packaging changes and build
549 fixes(?) for statically linked Perl, SCO, and VMS.
551 1.19 Tue Sep 29 22:30 1998
552 - put VMS gettimeofday() in. Patch is from Sebastian Bazley
553 <seb@stian.demon.co.uk>
554 - change GIMME_V to GIMME to help people with older versions of
556 - fix Win32 version of gettimeofday(). It didn't affect anything,
557 but it confuses people reading the code when the return value
558 is backwards (0 is success).
559 - fix Makefile.PL (more) so that detection of gettimeofday is
562 1.19 has better VMS support.
564 1.18 Mon Jul 6 22:40 1998
565 - add usleep() for Win32.
566 - fix Makefile.PL to fix reported HP/UX feature where unresolved
567 externals still cause an executable to be generated (though no
568 x bit set). Thanks to David Kozinn for report and explanation.
569 Problems with the fix are mine :)
571 1.18 has limited Win32 support (no ualarm). Added usleep for Win32.
572 Probably buggy. I'm sure I'll hear.
574 1.17 Wed Jul 1 20:10 1998
575 - fix setitimer calls so microseconds is not more than 1000000.
576 Hp/UX 9 doesn't like that. Provided by Roland B Robert, PhD.
577 - make Win32. We only get gettimeofday (the select hack doesn't
578 seem to work on my Win95 system).
579 - fix test 4 on 01test.t. add test to see if time() and
580 Time::HiRes::time() are close.
582 1.16 Wed Nov 12 21:05 1997
583 - add missing EXTEND in new gettimeofday scalar code.
585 1.16+ should be closer to building out of the box on Linux. Thanks
586 to Gisle Aas for patches, and the ualarm equivalent using setitimer.
588 If your underlying operating system doesn't implement ualarm(), then
589 a fake using setitimer() will be made. If the OS is missing usleep(),
590 a fake one using select() will be made. If a fake can't be made for
591 either ualarm() or usleep(), then the corresponding Perl function will
592 not be available. If the OS is missing gettimeofday(), you will get
593 unresolved externals, either at link- or run-time.
595 This is an improvement; the package used to not even build if
596 you were missing any of these bits. Roderick Schertler
598 <roderick@argon.org> did all the conditional compilation stuff,
599 look at HiRes.pm and the test suites; it's good educational reading.
601 1.15 Mon Nov 10 21:30 1997
602 - HiRes.pm: update pod. Provided by Gisle Aas.
603 - HiRes.xs: if gettimeofday() called in scalar context, do
604 something more useful than before. Provided by Gisle Aas.
605 - README: tell of xsubpp '-nolinenumber' woes. thanks to
606 Edward Henigin <ed@texas.net> for pointing out the problem.
608 1.14 Wed Nov 5 9:40 1997
609 - Makefile.PL: look for setitimer
610 - HiRes.xs: if missing ualarm, but we have setitimer, make up
611 our own setitimer. These were provided by Gisle Aas.
613 1.13 Tue Nov 4 23:30 1997
614 - Makefile.PL: fix autodetect mechanism to do try linking in addition
615 to just compiling; should fix Linux build problem. Fix was provided
618 1.12 Sun Oct 12 12:00:00 1997
619 - Makefile.PL: set XSOPT to '-nolinenumbers' to work around xsubpp bug;
620 you may need to comment this back out if you have an older xsubpp.
621 - HiRes.xs: set PROTOTYPES: DISABLE
623 1.11 Fri Sep 05 16:00:00 1997
625 Had some line commented out that shouldn't have been (testing
628 Previous version was corrupted.
630 1.10 Thu May 22 20:20:00 1997
631 - HiRes.xs, HiRes.pm, t/*:
632 - only compile what we have OS support for (or can
634 - only test what we compiled
635 - gross improvement to the test suite
637 This work was all done by Roderick Schertler
638 <roderick@argon.org>. If you run Linux or
639 one of the other ualarm-less platforms, and you like this
640 module, let Roderick know; without him, it still wouldn't
641 be working on those boxes...
642 - Makefile.PL: figure out what routines the OS has and
643 only build what we need. These bits were written by Jarkko
644 Hietaniemi <jhi@iki.fi>. Again, gratitude is due...
646 1.02 Mon Dec 30 08:00:00 1996
647 - HiRes.pm: update documentation to say what to do when missing
648 ualarm() and friends.
649 - README: update to warn that ualarm() and friends need to exist
651 1.01 Fri Oct 17 08:00:00 1996
652 - Makefile.PL: make XSPROTOARGS => '-noprototyopes'
653 - HiRes.pm: put blank line between __END__ and =head1 so that
656 1.00 Tue Sep 03 13:00:00 1996
657 - original version; created by h2xs 1.16