source: trunk/third/top/Changes @ 9084

Revision 9084, 24.3 KB checked in by ghudson, 28 years ago (diff)
This commit was generated by cvs2svn to compensate for changes in r9083, which included commits to RCS files with non-trunk default branches.
Line 
1Thu Aug 29 1996 - wnl (3.4)
2        Replaced modules (from Tim Pugh): next 32, next40.
3        Fixed bug in username.c: hashing negative uids.
4
5Thu Aug 22 1996 - wnl (3.4beta3)
6        Patched modules:  ultrix4, sunos4, sunos5, utek, decosf1, irix5.
7        Added modules: next40, next32.
8        Fixed procstates update bug in display.c.
9        Fixed divide by zero bug in utils.c.
10        Fixed bad number in layout.h
11        Minor fixes to Configure.
12        Complete overhaul of FAQ.
13
14Tue Feb 13 1996 - wnl (3.4beta3)
15        Added convex module from Warren Vosper (originally written by
16        William Jones).
17
18Tue Feb 13 1996 - wnl (3.4beta2)
19        Fixed format_k in utils.c to calculate K and M values correctly.
20        Added check for gigabyte values ('G').  Changed sumamry_format
21        in display.c to use format_k where appropriate.
22        Changed creation of distribution tar file to place everything in
23        a top level directory.
24
25Tue Jan 30 1996 - wnl (3.4beta2)
26        Added m_aix41 module.  Added new tag type to module comments:
27        TERMCAP, which defined the library to use for a termcap library.
28        If no TERMCAP tag is found in the module's initial comment, then
29        Configure will default to "-ltermcap".  AIX needs this since it
30        put all the termcap routines in libcurses(!)
31
32        Added m_bsdos2 (found lingering in my mailbox).
33        Updated m_svr4 to include support for NCR multiprocessors.
34        Fixed small bug in utils.c
35
36Thu Jan 25 1996 - wnl (3.4beta1)
37        Fixed m_sunos5 invocation of gettimeofday to include "NULL" as
38        second argument.  This provides compatability with the Posix-
39        compliant template provided with SunOS 5.5, but doesn't hurt
40        previous versions since they do bother with a template for that
41        function.
42
43        Made changes (recommended by net users) to hpux10, ultrix4,
44        netbsd10, aux3 (replaced aux31).  Added module for linux.
45
46Fri Oct 10 1995 - wnl (3.4beta1)
47        Added user-contributed modules for SCO Unix, IRIX 5, HP/UX 10,
48        Pyramid DC/OSX.  Changed Configure so that it runs in environments
49        whose c-shells have no 'eval'(!).  Added support for multiple sort
50        ordering methods via the -o switch.  This option requires support
51        from the machine dependent module: such support was added to
52        sunos5 (thus sunos54) and sunos4.
53
54        display.c:  Changed CPU states display line to shorten the leading
55        tag if the data won't fit in the current width.  Fixed a divide-by-
56        zero bug that affected ultrasparc servers (and potentially other
57        systems).
58
59        m_sunos5.c: Now asks the system for the correct pagesize rather than
60        assuming it is 4K.
61
62Thu Mar  2 1995 - wnl (3.3 RELEASE)
63        Added module netbsd10 and renamed netbsd to netbsd08.  Changed
64        Configure so that it does not use an initial default module name.
65        Made other compatability fixes to Configure.  Added comments to
66        decosf1 concerning optimizer bug.  Other documentation changes.
67        Added use of "prime.c" to Configure script.
68
69Tue Feb  7 1995 - wnl (3.3beta6)
70        Still one more beta....
71        Fixes for sunos5 2.4 gcc core dump (it was an alignment problem).
72        Fixed and improvements for decosf1 (including use of format_k
73        for proper SIZE column formatting).  Added modules freebsd20 and
74        ncr3000.
75
76Thu Feb  2 1995 - wnl (3.3beta5)
77        One more beta....
78        Fixed a few bugs in the sunos5 port pertaining to casting and
79        very large memory counts.  Added "ifndef HAVE_GETOPT" to getopt.c
80        to provide for conditional compilation of the getopt function.
81        Those systems that have getopt in libc can add -DHAVE_GETOPT to
82        the CFLAGS line in the module to prevent the function from being
83        compiled.  Added sunos54 module to accomodate SunOS 5.4
84        peculiarities.  Added module for aux3.1.
85
86Wed Jan  4 1995 - wnl (3.3beta4)
87        This is really taking too long......sigh.
88        Fixed SIGWINCH handling once and for all.  It now remembers the
89        number of processes you want displayed even thru window resizes.
90        Fixed buffer conflict in utils.c (itoa and itoa7).
91        Lots of small improvements to the various modules were made over
92        the past month: too numberous to list here.  SunOS 5 module made
93        more secure thru use of seteuid calls (other SVR4 modules should
94        be modified similarly).  One final MP fix to sunos5, too.  Module
95        for decosf1 was modified to accomodate V3.0.
96
97Mon Apr 18 1994 - wnl (3.3beta3)
98        I think I finally got a sunos5 module that will work on MP
99        machines.  Fixed cpu states figure in osmp41a so that
100        percentages never exceed 100%.  Added shell script "install"
101        since Unix vendors can't seem to make up their minds on what
102        options they want to use for the one that comes with the OS.
103        Added netbsd modules from Christos.  Fixed lots of other little
104        things over the past few months that I have long since forgotten.
105
106Wed Dec 15 1993 - wnl (3.3beta2)
107        Added module patches from various users:  hpux9, sunos5.
108        Fixed bug with batch mode (screen_width wasn't getting set).
109        Changes to accomodate 64 bit machines.
110        Fixed some bugs in command parsing ("renice 19 " did something
111        unexpected).
112
113Mon Aug 30 1993 - wnl (3.3beta)
114        Added lots of little patches from various users.
115        Added routines to utils.c for intelligent formatting of kilobytes
116        and time.  These are intended to be used in the modules when
117        formatting a process line.  Added code to "summary_format" in
118        display.c to do intelligent formatting of memory quantities.
119        Redid display.c to allow for varying line widths and dynamic
120        reallocation of the screen buffer.
121        Added a SIGWINCH handler to top.c!
122        Added a constant, MAX_COLS, to top.h which defines the absolute
123        widest line we will ever allow.  Changed allocations of "char fmt"
124        in all machine modules to use this constant rather than an abitrary
125        number.
126
127Fri Aug 13 1993 - wnl (3.3)
128        Changed return value definition of time-related functions in top.c,
129        display.c, and m_ultrix4.c to time_t (stuart@coral.cs.jcu.edu.au).
130        Fixed bug in display.c: line_update when start != 0.
131
132Wed Aug  4 1993 - wnl (3.2 release)
133        Changes to Configure from Paul Vixie.  Added modules for hpux9 and
134        bsd386.
135
136Tue Jul 13 1993 - wnl (3.1 release)
137        More small changes and minor bug fixes.  Brought bsd44 up to date
138        and added a module for svr4.2.  Changed shar packaging to use Rich
139        Salz's cshar stuff.
140
141Wed Jul  7 1993 - wnl (3.1BETA)
142        More changes and bug fixes to Configure.  Applied some other
143        minor bug fixes and suggestions from the beta testers.  Added
144        the "metatop" shell script and the "installmeta" rule to the
145        Makefile to make handling multiple machine models and OS versions
146        easier.  Added INSTALL and FAQ files.
147
148Tue May 18 1993 - wnl (3.1BETA)
149        Changed Configure to be compatible with most SVR4 environments
150        (differing output from "ls -lg").  Also changed Configure,
151        Makefile.X, etc., to look for module files in the subdirectory
152        "machine" (thanks to Christos Zoulas).
153
154Tue Apr 20 1993 - wnl (3.1BETA)
155        Changed both occurences of "ls -1" in Configure to "ls".  This
156        SHOULD produce the same result, and has the advantage that it
157        doesn't produce an error on a system 5 machine.  Integrated other
158        changes recommended in the first round of beta testing.
159
160Wed Mar 10 1993 - wnl (3.1BETA)
161        MAJOR CHANGE:  I have added a required function to all machine
162        dependent modules, called proc_owner.  It takes a pid as an argument
163        and returns the uid of the process's owner.  Such capability is
164        necessary for top to run securely as a set-uid program, something
165        that is needed for SVR4 implementations to read /proc.  I have
166        retrofitted all modules except dgux with this function, but was
167        not able to test most of them.  Top should now run securely as
168        a setuid program.  Added 386bsd and sunos5 modules.  Added sunos4mp
169        module for MP Suns.
170
171Sat Feb 20 1993 - wnl (3.1ALPHA)
172        Modified top.c and commands.c to compile correctly on System V
173        derived Unixes (especially SVR4), but in a way that doesn't rely
174        on an oracle-like declaration (that is, I don't use "ifdef SYSV").
175        Fixed some bugs in "Configure" and "getans".  Added inspection of
176        env variable "TOP" for options, and made -I default to showing
177        idle processes.  Added "u" command to change username restriction
178        on the fly.  Created shell script "suntop" for poor multi-version
179        SunOS folks (like myself).
180
181Wed Jun  3 1992 - wnl (3.0)
182        "max_topn" wasn't being used everywhere it was supposed to be
183        in top.c.  Many cosmetic changes, including copyright notices in
184        all the .c files.  Version number is now handled by version.c and
185        reflects the current patchlevel (which is initially set to 0).
186        Changed Configure and Makefile to allow configurable variables for
187        certain commands:  shell, cc, awk, install.  Updated README and
188        Porting.  Ready to release to the world!
189
190Mon May 18 1992 - wnl (2.9BETA)
191        Added modules provided by Christos Zoulas.  Replaced screen.c
192        with one modified by Christos and that will appropriately select
193        and handle the sgtty, termio, or termios system.  Integrated many
194        other changes recommended by Christos.  Fixed (I hope) the "-b"
195        batch mode display bug.  Had to change loadavg to load_avg to avoid
196        a conflict with 4.4BSD. 
197
198Mon Apr 27 1992 - wnl (2.8BETA)
199        Added modules provided by Daniel Trinkle.  Added patchlevel.h,
200        but the patch level is not yet reflected in the version number.
201        Cleaned up m_sunos4.c a little.
202
203Wed Apr 22 1992 - wnl (2.8BETA)
204        Major internal reorganization.  All of the system dependent stuff
205        is now really and truly separated from everything else.  The
206        system dependent functions are contained in a separate .c file
207        called a "module".  The Configure script knows how to find and
208        set up these modules, but the human installer still needs to tell
209        Configure which module to use (no automagic determination of
210        machine type---sorry).  Added -U option to specify one user's
211        processes, but there is no corresponding command...yet.  Other
212        changes and improvements too numerous to mention here.  Currently
213        there are only two modules:  sunos4 and umax.  But after this beta
214        release is sent around, I expect more to be written.  I just hope
215        that the machine-dependent abstractions don't need to change in
216        the process.
217
218Thu Mar 26 1992 - wnl (2.7BETA)
219        Beta release with minimal architecture support.  Updated README
220        and added a first cut at a Porting guide.  Added ioctl TIOCGWINSZ
221        code from top2.5+ (courtesy of David MacKenzie).  I didn't even
222        try porting the Ultrix support since I don't have access to an
223        Ultrix machine.
224
225Fri Oct 11 1991 - wnl (2.6)
226        This version was not widely released.  It contained many changes.
227        Here are the major ones:
228
229        Put in Vixie's idle process hack.
230
231        Enhanced type field in new_message to handle delayed messages.
232
233        Changed u_process to automatically adjust for varying lines of
234        output.  Management of screenbuf should now be completely contained
235        in display.c.  Removed now extraneous code from CMD_number[12]
236        portion of command switch in top.c.  This was the stuff that dealt
237        with zeroing out lines in screenbuf.
238
239        Finally made it all work correctly on a 386i.  Problems I had to
240        overcome: kvm_nlist doesn't return 0 on success as advertised (it
241        returns 1 instead); the results of a kvm_nlist are different
242        (n_type can be zero even for a symbol that exists).
243
244        Serious rearrangement for processor dependent stuff.  All nlists
245        are now in separate files with the suffix ".nlist".  Most machine
246        specific code is in "machine.c" surrounded by appropriate ifdefs---
247        the goal is to eventually have all machine specific code in this
248        file.  Managed to find a way to detect SunOS 4.x at compile-time:
249        this is contained in the include file "sun.h".  Completely changed
250        the memory display line for SunOS 4.x---it now displays a far
251        more appropriate report.
252
253        Created the shell script "Configure" to aid in the configuration
254        step.
255
256        Fixed a bug in init_termcap:  it will now tolerate an environment
257        which does not have TERM defined (thanks to Sam Horrocks for
258        pointing this out).
259
260Tue Aug  9 1988 - wnl (2.5)
261        Added changes to make top work under version 4.0 of the Sun
262        operating system.  Changes were provided by Scott Alexander of the
263        University of Pennsylvania.  Thanks!  Compile with "-Dsunos4" to
264        get them.  Virtual memory statistics are not readily accessible
265        under 4.0, so they don't show up in the output.
266
267Thu Jul 31 1987 - wnl (2.4)
268        Fixed a problem with the 4.0 Pyramid code.  The label "cp_time"
269        doesn't exist in the 4.0 kernel anymore.  I think the code Carl
270        sent me wants "percpu" instead.  That is what I am using and it
271        appears to work.  375 code is still untested (at least by me).
272        Also picked a great deal of lint out of the source.  Lint now only
273        complains about a very few nitpicky things (there are far too many
274        calls to "printf" to put a "(void)" in front of!), at least under
275        SunOS.
276
277Tue Jul 28 1987 - wnl (2.4a)
278        Added changes for a Symmetrics Computer Systems s/375 machine.
279        Changes were provided by Paul Vixie.  Thanks!  According to Mr.
280        Vixie:  "These changes were not made at, by, or for SCS proper.
281        SCS would probably be interested in them, but so far only the
282        users' group has them.  They were made in February, 1987, to
283        version 2.1 of the program, by Paul Vixie
284        (dual!ptsfa!vixie!paul@ucbvax.Berkeley.EDU)."  His changes were
285        integrated into version 2.3 to make version 2.4.
286
287        The SCS peculiarities are summarized in Changes.scs.
288
289Tue Jun  9 1987 - wnl (2.3 for real)
290        Changed the includes for the extra code Carl sent me to only
291        compile on Version 4.0 Pyramid machines.  This makes top still
292        compilable on pre-4.0 Pyramids.  Specifically, this code is only
293        compiled when both "pyr" and "CPUFOUND" are defined.
294
295Wed Jun  3 1987 - wnl (2.3 with Pyramid additions)
296        It's been a month and I still haven't done anything about
297        distributing this version.  However, Carl Gutekunst from Pyramid
298        has sent me some extra patches for some of the Pyramid code.  I
299        just added those and will make them part of 2.3.  This fixes the
300        following Pyramid problems:  adds the inclusion of <sys/systm.h>,
301        uses the correct size for getting the kernel value _ccpu (this bug
302        affected the Vax version as well), sums the elements of the percpu
303        array to calculate a cp_time value (for OSx 4.0).
304
305Fri May  1 1987 - wnl (2.3)
306        I have finally finished all the changes for better support of
307        oddbal terminals.  Added the low-level routine "clear_eol" which
308        makes handling terminals without "ce" easy:  it uses spaces
309        instead.  All direct uses of "clear_line" outside of screen.c have
310        been changed to use this primitive.  A terminal with "os" is now
311        handled in such that all situations that need overwriting are
312        completely avoided (including several commands).  This required
313        some changes to the way commands are translated into action (in
314        "top.c").  Made several important changes to display.c to prevent
315        overflowing of any of the fields.  Specifically, more than 99
316        total processes and a cpu state that reaches 100%.  Had to make a
317        small change to two casts in top.c, because the Sun 3.2 compiler
318        was giving warnings on them.  Added the "-q" option which lets
319        root run top at a nice of -20 (in case he thinks he really needs it).
320
321Tue Dec 30 1986 - wnl (2.2)
322        I think I fixed a bug reported by Julian Onions at Nottingham.
323        Occasionally, top will core dump when the sprintf in either
324        i_process or u_process overflows due to an exceptionally
325        unrealistic time value.  I think it highly unlikely that top can
326        get a bad proc structure (although I suppose it is possible), but
327        the process time is read from the user structure, and that can
328        sometimes be part garbage.  So, "get_ucpu" checks the value it
329        returns to make sure its formatted form will not overflow the
330        sprintf.  If this doesn't fix the bug, then more drastic measures
331        will be necessary.  I plan to make this version the official
332        "top 2.2".  [[ This version was never distributed very widely. ]]
333
334Tue Dec  2 1986 - wnl (2.2c)
335        Added to top.c the notion of a "failed command".  When a command
336        produces a message (on the message line), an update does not
337        follow it.  Before, the message was written and a new display was
338        shown---purposefully not overwriting the message.  But the
339        improvements to handle overstriking terminals and terminals
340        without "ce" clear the screen before every display, which would
341        erase the message.  Now, the message is displayed and top waits
342        another full time interval before updating the display.  This
343        works much better all around.
344
345Mon Nov 24 1986 - wnl (2.2b)
346        Created a new file, utils.c, and made appropriate changes to
347        Makefile.  This new file holds all utility functions that can and
348        may be used by more than one "module".  Improved i_memory and
349        u_memory (display.c) so that screen updates for the values
350        displayed are only changed when necessary.  Also made the line
351        look better:  the last fixes made for a rather ugly display.
352        Added the locally defined constant "LoadMax" and added code to
353        top.c to send the cursor home after a space command is entered if
354        the load average is higher than "LoadMax".  This provides visual
355        feedback on loaded systems.
356
357Mon Nov  3 1986 - wnl (2.2a)
358        Widened the format for memory usage so that it can display 5
359        digits.  This makes that line look a little ugly---maybe I'll fix
360        that later.  Screen handling now understands "os" and a missing
361        "ce".  It treats them identically:  clear the screen between each
362        display.  Screen handling code now uses "cd" when appropriate
363        (i.e.:  when user has shortened the screen).  Made i_loadave clear
364        then screen and took out most of the explicit calls to "clear" in
365        top.c.  This method is cleaner, especially in conjunction with
366        "os" handling.  Added preprocessor variable "RANDOM_PW" for
367        systems that access the passwd file randomly (Sun's yp and 4.3).
368        With "RANDOM_PW" set, "getpwuid" is used instead of "getpwnam",
369        but uid->username mappings are still hashed internally (because
370        that is still faster than going to disk).
371
372Mon Oct  6 1986 - wnl (2.1)
373        A bug with the kill command was pointed out by "dciem!tim"---
374        specifying a signal by name did not work correctly.  This bug has
375        been fixed with a simple change to commands.c.  Another bug made
376        the cpu state percentages incorrect the first time they were
377        displayed.  This bug has also been fixed (changed top.c).
378
379Thu Sep  4 1986 - wnl (2.0, at last)
380        This is the version that will (hopefully) get released to the
381        world as top 2.0.
382        Added the "r" and "k" commands for renice and kill, respectively.
383        This required adding a way to handle system call errors, and the
384        addition of the "e" command.  Help screen and manual page were
385        changed to reflect this change.  Changed all "#ifdef SUN" directives
386        to "#ifdef sun", and changed all "#ifdef PYRAMID" directives to
387        "#ifdef pyr".  As much as I hate those choices of preprocessor
388        names (they too easily conflict with real variable names), it does
389        make automatic compilation possible---people don't have to change
390        the Makefile anymore for specific machines.  The manual page was
391        changed to automatically incorporate the defaults as set in the
392        Makefile (including an infinite value for TOPN) and the way the
393        manual page is generated by the Makefile was changed to make
394        maintenance of this information automatic.
395
396Mon Jul 28 1986 - wnl (still pre 2.0)
397        Real close now.  I put in a new definition for the macro "pagetok"
398        that does an explicit shift of a constant expression involving
399        PGSHIFT.  Appropriate checks are made if PGSHIFT is to small.
400        "pagetok" is now used exclusively everywhere to convert kernel
401        clicks to kilobytes.  I added a full blown interactive mode with
402        the ability to change some of the runtime parameters (how many to
403        display, time delay, etc.) while top is running.  I also
404        incorporated a few ideas from the net:  control characters in the
405        command name are replaced with '?'; the '-S' option makes the
406        swapper and pager visible; options have been added to control the
407        number of displays produced (this makes it easier to make
408        performance snapshots with top).  I have also added the notion of
409        "infinite" values for number of processes and number of displays.
410        I fixed a long-standing bug in the uid to username mapping code
411        that was only aggravated on the pyramids:  it was an ill-defined
412        expression (akin to i = i++).  I tweaked the proc_compar routine
413        for qsort slightly so that stopped processes were more likely to
414        show up.  Manual page was updated to reflect all changes
415        noticeable to the user.
416
417Tue Jul  1 1986 - wnl (pre 2.0 -- 1.9999?)
418        In the process of major revamping on the way to version 2.0.
419        I have completely done away with curses by adding my own screen
420        management routines in a separate file (screen.c).  The rationale
421        for this is that top knows a whole lot more about what is and is
422        not redundant on the screen and can compare simple integer values
423        where curses would have to compare strings.  This has turned out
424        to be a very big win speed-wise.  The proc_compar routine for
425        sorting has been rewritten to include several more keys.  I
426        decided this was necessary when I noticed that the "top" process
427        itself kept disappearing off the top 10 list on a Sun-3.  All the
428        processes had the same percentage (0%) and the sort wasn't really
429        doing anything worthwhile.  I changed the expression that computes
430        memory usage to use the ctob macro instead of just assuming that
431        pages were 512 bytes.  More work still needs to be done before
432        this version is usable.  I changed options-processing to use
433        getopt and added appropriate incantations to the Makefile.
434
435Wed Feb 20 1985 - wnl (still 1.8)
436        Put in the ifdef FOUR_ONE statements to make top still compilable
437        on a 4.1 system.  Apparently, there are some users out there that
438        need this functionality.  Oh well.  I don't guarantee any of it,
439        since I can't test it.  Made appropriate changes to README and
440        final installation related changes to Makefile.
441
442Sat Feb  2 1985 - wnl (1.8)
443        Removed all the ifdef FOUR_TWO statements and made "top" into a
444        4.2 only program.  If someone really wants to still run it on 4.1,
445        then they can do all the work.  We don't have a 4.1 machine
446        anymore, so I don't even know if the thing still works under 4.1.
447        Cleaned up the Makefile and the README.  Added installation rules
448        to the Makefile, as requested by several sites.  Fixed a very
449        obscure divide-by-zero bug.  Added a second "key" to the qsort
450        comparison function (proc_compar) so that comparisons are based on
451        cpu ticks if the percentages are equal (provided by Jonathon
452        Feiber at Sun).
453
454Tue Dec 11 1984 - wnl (1.7)
455        Added the virtual and real memory status line to the header area
456        (provided by Jonathon Feiber at Sun)
457
458Tue Nov 20 1984 - wnl (1.6)
459        Added an "exit" if sbrk's fail.  Added changes from Jonathon
460        Feiber at Sun:  ifdef SUN to make top work on Suns (they don't use
461        doubles in the proc structure), register declarations, check for
462        getting a user structure that has disappeared since the proc array
463        was read (it used to die, now it just shows the process as swapped).
464
465Tue Nov 13 1984 - wnl (1.5)
466        If the number of displayable processes ("active_procs") was less
467        than the number of requested processes ("topn"), top would
468        segmentation fault.  This bug has been fixed.  Thanks to Prentiss
469        Riddle at ut-sally for pointing out the existence of this bug.
470
471Tue Oct 23 1984 - wnl (1.4)
472        Finally fixed the hash table bug that caused processes owned by
473        root to sometimes appear with either no name or a different name
474        that had UID 0 (such as "operator").  Removed all the ifdef DEBUG
475        blocks to make top ready for distribution to the real world.
476
477Sun Apr  8 1984 - wnl (still 1.3)
478        Made some slight changes to the display format.  It now looks more
479        aesthetically pleasing.  Added some preprocessor constants so that
480        the two defaults (number of processes and seconds of delay) easier
481        to change.
482
483Thu Apr  5 1984 - wnl (1.3)
484        Changed the order in which things are done at initialization time.
485        This way, if an error occurs before starting the main loop, curses
486        will never get started.  Also changed other error handlers so that
487        endwin() is called before any flavor of exit.  Specifying a number
488        of processes that is more than the screen can handle is no longer
489        fatal.  It displays a warning message and pretends the user
490        specified the maximum for the screen.  Finally cured all the TSTP
491        blues (well, almost all).  I removed my TSTP handler and convinced
492        the system to always use the one that curses sets up.  Turns out
493        that "sleep" was stepping all over it during a pause.  So, I don't
494        use sleep anymore.  The only problem that remains with it now is
495        redrawing the old display before updating it after a pause.
496
497Tue Apr  3 1984 - wnl (from 1.0 to 1.2)
498        I changed the format of the TIME column from just "seconds" to
499        "minutes:seconds".  I also made pausing work correctly.  Screen
500        redraws with an up to date display.  For compatibility with 4.2, I
501        changed the name of the "zero" function to "bzero".  The makefile
502        has been altered to handle versions for 4.1 and 4.2, and README
503        has been updated to reflect these recent changes.
Note: See TracBrowser for help on using the repository browser.