1 2006-10-12 Nick Garnett <nickg@ecoscentric.com>
3 * cdl/uitron.cdl: Add reqirement for UITRON conformant mailbox
4 implementation to CYGIMP_UITRON_STRICT_CONFORMANCE.
6 2005-08-02 Andrew Lunn <andrew.lunn@ascom.ch>
8 * tests/test2.c (task1): Cast to fix compiler warning.
10 2003-02-24 Jonathan Larmour <jifl@eCosCentric.com>
12 * cdl/uitron.cdl: Update doc link.
14 2002-05-27 Jesper Skov <jskov@redhat.com>
16 * cdl/uitron.cdl: Don't build C tests when C++ options selected.
18 2002-05-22 Nick Garnett <nickg@redhat.com>
20 * doc/uitron.sgml: Fixed dangling xref.
22 2002-02-19 Mark Salter <msalter@redhat.com>
24 * doc/uitron.sgml: Fixed typos and such.
26 2002-02-14 Hugo Tyson <hmt@redhat.com>
28 * doc/uitron.sgml: General polishing... Tidied up broken emPHAsis
29 on closing parentheses of function lists. Reduced line length of
30 same to fit real paper, I hope. Added LITERAL tags all over
31 config option mentions.
33 New stuff... Added a whole new section "[u]ITRON Configuration
34 FAQ" based on some email answers I have given in the past in
37 2002-02-13 Hugo Tyson <hmt@redhat.com>
39 * doc/uitron.sgml: NEW FILE: Make id tags qualified with
40 compat-uitron- and add a couple of tags; tidy up formatting of
41 some function listings; correct comment on using functions in your
44 2002-01-24 Jesper Skov <jskov@redhat.com>
46 * include/uit_func.inl (ena_int, dis_int): Wrap
47 HAL_MASK_INTERRUPT/HAL_UNMASK_INTERRUPT in interrupt
48 disable/restore pairs.
49 * tests/testintr.cxx (detach_isr, attach_isr): Change order of
50 mask/unmask and disable/restore.
52 2001-04-30 Jonathan Larmour <jlarmour@redhat.com>
54 * src/uit_objs.cxx: Workaround xscale tools preprocessor bug
55 by avoiding determining the stack size with the preprocessor.
56 * cdl/tasks.cdl (CYGDAT_UITRON_TASK_EXTERNS and
57 CYGDAT_UITRON_TASK_INITIALISERS): Ditto
59 2001-04-21 Bart Veer <bartv@redhat.com>
61 * tests/testintr.cxx (task1):
62 Disable one of the tests on the synthetic target, it requires
63 consistent cpu timing that cannot be guaranteed.
65 2000-11-01 Jesper Skov <jskov@redhat.com>
67 * include/uit_func.inl: CYG_SCHED_UNIQUE_PRIORITIES changed to
68 CYGINT_KERNEL_SCHEDULER_UNIQUE_PRIORITIES
70 2000-09-13 Jonathan Larmour <jlarmour@redhat.com>
72 * include/uit_objs.hxx (CYG_UITRON_OBJS_INIT_PRIORITY): Define
73 with correct priority now.
75 2000-08-03 Jonathan Larmour <jlarmour@redhat.co.uk>
77 * include/uit_func.inl (ref_mpf): Compute used blocks from
78 bytes used / blocks size rather than now obsolete blockcount member
80 2000-07-04 Jonathan Larmour <jlarmour@redhat.co.uk>
82 * include/uit_func.inl: Update all memory pool related functions to
83 reflect the new interface to memory pools, and that they come from
84 the new CYGPKG_MEMALLOC package
86 * cdl/uitron.cdl: CYGPKG_UITRON_MEMPOOLFIXED and
87 CYGPKG_UITRON_MEMPOOLVAR require the CYGPKG_MEMALLOC package
89 * include/uit_objs.hxx: Update includes - memory pool implementations
90 are now in CYGPKG_MEMALLOC
92 2000-06-15 Gary Thomas <gthomas@redhat.com>
94 * include/uit_ifnc.h (CYGPRI_UITRON_SET_RETCODE): Always return
95 'HANDLED'. This fixes some confusion when interrupt chaining.
97 2000-05-15 Hugo Tyson <hmt@cygnus.co.uk>
99 * include/uit_ifnc.h (cyg_uitron_dsr): make this be extern "C"
100 always; it went wrong when uitron funcs are inline or C++.
102 * src/uit_ifnc.cxx (CYGIMP_UITRON_INLINE_FUNCS): When [re]defining
103 this, match what CDL defined. Warnings--;
105 2000-05-08 Jesper Skov <jskov@redhat.com>
107 * tests/testintr.cxx (attach_isr, detach_isr): Don't check for
108 clean vector after a detach when chaining is enabled.
110 2000-03-28 Jesper Skov <jskov@redhat.com>
112 * tests/testintr.cxx: Synchronize with clock before making timer
115 2000-03-28 John Dallaway <jld@cygnus.co.uk>
119 Adjust documentation URLs.
121 2000-03-22 Jesper Skov <jskov@redhat.com>
123 * include/uit_ifnc.h (ret_int): Fix compiler warning.
125 2000-03-13 Jesper Skov <jskov@redhat.com>
127 * tests/testintr.cxx: Don't try to disable clock on PowerPC.
129 2000-03-03 Hugo Tyson <hmt@cygnus.co.uk>
131 * tests/testintr.cxx (task1): Only test ena_int() and dis_int()
132 for E_PAR if CYGSEM_UITRON_BAD_PARAMS_RETURN_ERRORS.
134 2000-02-28 Hugo Tyson <hmt@cygnus.co.uk>
136 * include/uit_func.h: add ena_int() and dis_int().
138 * include/uit_func.inl (dis_int): New function added.
139 (ena_int): New function added.
141 * tests/testintr.cxx (task1): Add a test of ena_int() and
142 dis_int(); easiest done here because we have an interrupt number -
143 for the clock - in our hands here already for the rest of the
146 2000-02-02 Jesper Skov <jskov@redhat.com>
148 * tests/testintr.cxx: Don't output DELAYLOCKSCHED quite so often
151 2000-01-31 Hugo Tyson <hmt@cygnus.co.uk>
153 * tests/testintr.cxx (CHECK_TID): Test that get_tid() returns 0 in
154 ISR context ie. non-task portion.
156 * include/uit_func.inl (get_tid): Check the sched lock for
157 non-task portions, ie. ISR/DSR will have the scheduler locked.
159 2000-01-31 Simon FitzMaurice <sdf@cygnus.co.uk>
160 * compat/uitron/current/cdl/uitron.cdl
162 Adjust help URLs in line with new doc layout.
164 2000-01-28 Simon FitzMaurice <sdf@cygnus.co.uk>
165 * compat/uitron/current/cdl/uitron.cdl
167 Adjust help URLs in line with new doc layout.
169 2000-01-28 Jonathan Larmour <jlarmour@redhat.co.uk>
171 * tests/testintr.cxx (DELAYLOCKSCHED): When in long testing loops,
172 occasionally output a message to show the test is still alive
174 2000-01-26 Hugo Tyson <hmt@cygnus.co.uk>
176 * include/pkgconf/uitron.h: Add new option
177 CYGSEM_UITRON_TIME_IS_MILLISECONDS to control just what it says.
179 * include/uit_func.inl (CYG_UITRON_TIME_UIT_TO_SYS32): et al; new
180 macros optionally to convert to/from milliSeconds if option
181 CYGSEM_UITRON_TIME_IS_MILLISECONDS is set.
183 * src/uit_func.cxx: Initialize Cyg_Clock::converter type objects
184 if the time unit is set to milliSeconds.
186 * cdl/uitron.cdl: Add new option
187 CYGSEM_UITRON_TIME_IS_MILLISECONDS to control just what it says.
189 Add an include of the kernel config file to the generated
190 pkgconf/uitron.h: tests at the very least need it! This is a
191 correct thing to do because the kernel config defines a lot of the
192 semantics of the uITRON layer, as well as being needed for
193 backward compatibility.
195 Also add uit_ifnc.cxx to the files that need compiling. Due to
196 the previous screwup, no testing has been applied to the uITRON
197 system under CDL, because it thought the kernel was absent, so
198 this had not been detected. Doh.
200 2000-01-25 Jesper Skov <jskov@cygnus.co.uk>
202 * tests/testintr.cxx (cyg_start): externC addded.
204 2000-01-19 Hugo Tyson <hmt@cygnus.co.uk>
206 * cdl/*.cdl: Add descriptions to a number of options &c which were
207 lacking same, also tidied up other typos as noticed en passant.
209 1999-11-09 Jonathan Larmour <jlarmour@cygnus.co.uk>
211 * src/uit_objs.cxx: Make the stack size failsafe be tested against
212 the minimum stack size, not the typical one.
214 1999-11-03 John Dallaway <jld@cygnus.co.uk>
216 * cdl/uitron.cdl: Define tests.
218 1999-10-07 John Dallaway <jld@cygnus.co.uk>
220 * cdl/uitron.cdl: Specify radio buttons using CDL interfaces.
222 1999-09-06 Jonathan Larmour <jlarmour@cygnus.co.uk>
224 * include/uit_objs.hxx (CYG_UITRON_OBJS_INIT_PRIORITY): Revert the
225 below change - we may still be shipping antiquated compilers to
228 1999-09-02 Hugo Tyson <hmt@cygnus.co.uk>
230 * include/uit_objs.hxx (CYG_UITRON_OBJS_INIT_PRIORITY): Enable
231 using init priority on uitron objects, now that the compilers all
232 support it fully. AFAI can tell from a quite broad experiment.
234 1999-09-01 Hugo Tyson <hmt@cygnus.co.uk>
236 * tests/testintr.cxx (attach_isr): Make it all work: my chosen
237 default interrupt "level" of 0 means make no interrupts on the
238 tx39 ;-( oh well. Made the whole test simulator-friendly ie. much
239 shorter, made the output more friendly to me.
241 1999-08-26 Hugo Tyson <hmt@cygnus.co.uk>
243 * cdl/uitron.cdl (CYGNUM_UITRON_ISR_ACTION_QUEUESIZE): add a
244 "default_value 32" - ooops.
246 1999-08-26 Hugo Tyson <hmt@cygnus.co.uk>
248 * cdl/uitron.cdl: Add configury for lovely new interrupt-safe
249 functions: just "try immediate execution" [is the sched lock one?]
252 * include/pkgconf/uitron.h: Add configury for lovely new
253 interrupt-safe functions: just "try immediate execution" [ie. if
254 the sched lock is one] and queue size.
256 * include/uit_ifnc.inl: Respond to general configury wrt the
257 presence of semas, flags, mboxes: only define the ixxx_yyy()
258 function if the corresponding xxx_yyy() func exists. Remove the
259 static definitions of the configuration symbols.
261 * src/uit_ifnc.cxx: Respond to general configury wrt the presence
262 of semas, flags, mboxes: the functions called may not exist!
264 1999-08-25 Hugo Tyson <hmt@cygnus.co.uk>
266 * include/uit_ifnc.h: New file. Prototypes of interrupt-safe
267 ixxx_yyy() style functions as a porting/backwards-compatibility
268 aid. Prototype of a suitable DSR to mate with an ISR that uses
271 * include/uit_ifnc.inl: New file. Bodies for possible inlining of
272 the ixxx_yyy() style functions; organized much like the standard
275 * src/uit_ifnc.cxx: New file. Implementation of the DSR provided,
276 plus concrete instantiations of the ixxx_yyy() style functions.
278 * tests/testintr.cxx: New file. Test for the ixxx_yyy() functions
279 newly provided above. Rather unpleasant: it is necessary to break
280 type-safety to get access to the kernel's interrupt object that
281 drives the realtime clock. Perhaps a neater way will follow.
283 * tests/PKGconf.mak (TESTS): Add new testcase testintr.cxx
284 * src/PKGconf.mak (COMPILE): Add new source file uit_ifnc.cxx
286 1999-08-17 John Dallaway <jld@cygnus.co.uk>
288 * cdl/uitron.cdl: Implement radio buttons using
289 "FIXME radio" hack in description field for now.
291 1999-07-30 John Dallaway <jld@cygnus.co.uk>
293 * cdl/uitron.cdl: Tidy display string capitalisation.
295 1999-05-26 Jesper Skov <jskov@cygnus.co.uk>
297 * tests/testcx7.cxx: Include new stackmon header.
299 1999-05-20 Hugo Tyson <hmt@cygnus.co.uk>
301 * tests/testcx7.cxx (task1): Add some statistics dumping about
302 stack usage; since this is quite a long and arduous test. This
303 just helps en passant with testing the interrupt stack work that's
304 been happening recently. C++ only, so not for test7.
306 1999-05-17 Hugo Tyson <hmt@masala.cygnus.co.uk>
308 * include/uit_func.inl (ter_tsk): Only up the priority of the
309 killee if priorities are not unique or we are not already
310 elevated. This is to prevent asserts with the bitmap scheduler.
311 (chg_pri): Support a level X feature, pri zero => reset to the
312 thread's initial priority.
314 1999-05-14 Hugo Tyson <hmt@masala.cygnus.co.uk>
316 * include/uit_objs.hxx: object array declaration macro now
317 contains constructor priority ordering rune; if 0'd out pending
318 99r1 compiler cutover.
320 1999-04-23 Hugo Tyson <hmt@cygnus.co.uk>
322 * tests/testcx7.cxx: T2_MALLOC definition; only make it smaller
323 than the main consumer malloc if coaleasing is enabled, otherwise
324 the mempool cannot allocate blocks for the consumer for next time.
325 Fixes PR 18817 - only a random perm disable coalescence.
327 * tests/test7.c: Ditto.
329 1999-04-15 John Dallaway <jld@cygnus.co.uk>
331 * include/pkgconf/uitron.h: Tidy display string capitalization.
333 1999-04-09 Hugo Tyson <hmt@cygnus.co.uk>
335 * include/uit_func.inl (dis_dsp): Remove bad old code which used
336 to lock the scheduler to implement dis_dsp() and ena_dsp() - which
337 also stopped the clock, for example. Instead change the calling
338 thread's priority to 0 internally so that it cannot be preempted
339 by uITRON threads at least, whose priorities are 1--N. Same
340 applies to loc/unl_cpu(). Save the "real" priority in a static,
341 and take notice of it everywhere relevant, such as thread and
342 system state inquiries, changing (our own) priority, checking for
343 dispatch enabled in sleeping calls, and so on.
345 * src/uit_func.cxx: New variable cyg_uitron_dis_dsp_old_priority
346 for holding the "real" priority of a thread whilst we change our
347 priority to zero (the highest) to prevent dispatching in dis_dsp()
350 * include/uit_func.inl (get_tid): Return 0 for the task id if not
351 in a uITRON thread, this is better conformance to the spec.
353 1999-03-22 Hugo Tyson <hmt@cygnus.co.uk>
356 Use CYGNUM_HAL_STACK_SIZE_TYPICAL for the stack size instead of
357 CYGNUM_HAL_MINIMUM_STACK_SIZE.
359 1999-02-23 Jonathan Larmour <jlarmour@cygnus.co.uk>
361 * tests/test1.c, tests/test2.c, tests/test3.c, tests/test4.c,
362 tests/test5.c, tests/test6.c, tests/test7.c, tests/test8.c,
363 tests/test9.c, tests/testcxx.cxx, tests/testcx2.cxx,
364 tests/testcx3.cxx, tests/testcx4.cxx, tests/testcx5.cxx,
365 tests/testcx6.cxx, tests/testcx7.cxx, tests/testcx8.cxx,
367 Change all non-applicable cases to use CYG_TEST_NA rather than
368 CYG_TEST_PASS/CYG_TEST_EXIT
370 1999-02-22 Hugo Tyson <hmt@cygnus.co.uk>
372 * src/uit_objs.cxx: Eeeek! Previous fix was bogus: the override
373 was too late in the file to take effect. The declaration of the
374 objects (including stacks) is moved to below the override CPP
377 1999-02-12 Hugo Tyson <hmt@cygnus.co.uk>
379 * src/uit_objs.cxx (CYG_UITRON_DECL):
380 Override CYGNUM_UITRON_STACK_SIZE if CYGNUM_HAL_MINIMUM_STACK_SIZE
383 * include/pkgconf/uitron.h (CYGNUM_UITRON_STACK_SIZE):
384 Document that this option can be overridden by HALs demands.
386 1999-02-02 Jesper Skov <jskov@cygnus.co.uk>
388 * tests/test2.c (task1):
389 * tests/testcx2.cxx (task1):
390 Reduce run time on SIM.
392 1999-01-26 Jesper Skov <jskov@cygnus.co.uk>
394 * tests/test4.c (task1):
395 * tests/testcx4.cxx (task1):
396 Extend timeouts when running on HW to avoid failures due to
397 overhead of GDB interaction.
399 1999-01-25 Jesper Skov <jskov@cygnus.co.uk>
401 * include/pkgconf/uitron.h (CYGDAT_UITRON_SEMA_INITIALIZERS):
402 Let default initializers have values that don't cause tests to
405 1998-11-25 Hugo Tyson <hmt@cygnus.co.uk>
407 * include/uit_func.inl (del_xxx &c):
408 Make error returns more consistent; E_NOEXS rather than E_OBJ if
409 the object disappeared during a race.
411 1998-11-19 Hugo Tyson <hmt@cygnus.co.uk>
413 * src/uit_func.cxx (SET_UP_PTRS):
414 PR 17999; work around codegen bug on tx39 with -Os.
415 The code is better with the fix regardless, so it's got no
416 downside. Explicitly condition whether to run a follow-on loop
417 rather than letting the for-loop sort it out itself.
419 1998-11-16 Hugo Tyson <hmt@cygnus.co.uk>
421 * tests/testcx7.cxx (check_waitstate):
422 Add a routine to check the reported waiting state of the synch
423 object in question, and call it as necessary. This only added to
424 the C++ version of the test for enhanced coverage, in case the
425 enquiry call perturbs the object state; having both versions
426 present is better. This checks a very recent fix to
427 kernel...thread.cxx (1998-10-27).
429 1998-10-21 Hugo Tyson <hmt@cygnus.co.uk>
432 Provide weakly named dummies for task[1-4], which are the defaults
433 referred to by the task array, so that the uITRON system can be
434 initialized even when no uITRON tasks have been provided (so long
435 as no other changes to the task init configury have been made)
436 This does not pollute the namespace, for these are weak symbols.
437 Changing them to cyg_uitron_task... would be possible, but it
438 carries the false implication that those are cygnus names, when
439 they are really entirely up to the user.
441 * tests/test[123456789].c:
443 * tests/testcx[23456789].cxx:
444 Undo the previous change: remove dummies for task[1-4] when we
445 decide not to test anything, they will be provided with weak names
446 regardles in src/uit_func.cxx.
448 1998-10-20 Hugo Tyson <hmt@cygnus.co.uk>
450 * tests/test[123456789].c:
452 * tests/testcx[23456789].cxx:
453 Provide dummies for task[1-4] when we decide not to test anything,
454 to encourage correct linking.
456 1998-10-16 Hugo Tyson <hmt@cygnus.co.uk>
458 * tests/testcx*.cxx (cyg_start):
460 changed cyg_start to externC so that it truly does override the
461 default one; this is needed for handling the "I cannot test in
462 this configuration" failures. Of course the C ones were C to
465 1998-10-16 Hugo Tyson <hmt@cygnus.co.uk>
467 * tests/test[123456789].c:
469 * tests/testcx[23456789].cxx:
470 All tests are now much more self-configuring wrt turning off
471 if there are no uITRON objects of the right type to manipulate.
472 The huge (50-line) #if statement is arranged the way it is so that
473 clauses checking on each feature can be pasted in and out easily.
474 So yes, the "&& 1" at the end has a purpose.
476 Here for the record is a table of feature versus test-used-in:
477 tsk :: 1 2 3 4 5 6 7 8 9
487 1998-10-16 Hugo Tyson <hmt@cygnus.co.uk>
491 New tests, test create and delete of memory pools
493 1998-10-15 Hugo Tyson <hmt@cygnus.co.uk>
495 * include/pkgconf/uitron.h:
496 Add configury for create/delete of memory pools.
497 Tidy up naming of initializer macros to use for non-existent
498 (initially) uITRON objects.
499 Improve comments in CDL-- info fields.
501 * include/uit_func.h:
502 Add prototypes for cre/del_mpf/mpl().
504 * include/uit_func.inl:
505 Implement cre/del_mpf/mpl().
508 Support CYG_UIT_MEMPOOLFIXED_NOEXS/CYG_UIT_MEMPOOLVAR_NOEXS(...)
509 mempool construction macros for non-existent (initially) mempools.
510 Rename _NONE construction macros to _NOEXS which will make a lot
511 more sense to uITRON folks.
513 1998-10-14 Hugo Tyson <hmt@masala.cygnus.co.uk>
515 * include/pkgconf/uitron.h:
516 Add configury for create/delete of tasks.
518 * include/uit_func.h:
519 Add prototypes for cre/del_tsk(); correct comment in exd_tsk().
521 * include/uit_func.inl:
522 Implement cre/del_tsk() and complete exd_tsk().
523 Improve NULL/NADR checking on other cre/del calls.
525 * src/uit_func.cxx (cyg_uitron_start):
526 Deal better with start tasks stuff, for task create/delete.
529 Support CYG_UIT_TASK_NONE(...) task constructor for those tasks
530 which do not initially exist
533 Add new tests test8.c, testcx[678].cxx to the build.
536 Tidy comments, untabify.
539 Tidy comments, untabify, test creation with NADR record pointer.
540 No impact from the new cre/del_tsk() functionality on test7, since
541 a task may only be deleted when it is dormant, so deletion is NOT
542 another way out of a waiting state.
545 New test, tests cre_tsk() and del_tsk() and exd_tsk() ie. the
546 creation and deletion of tasks.
551 New tests, C++ versions of tests 6, 7, and 8.
553 1998-10-09 Hugo Tyson <hmt@cygnus.co.uk>
556 Test interactions with deleting objects (as applicable) and
557 killing, signalling and the like. Also test killing the task
558 after other treatments more generically.
560 * tests/test6.c (task1):
561 Change "N/A:" to "N/A" in the exit message when we didn't actually
562 do any testing due to create/delete being disabled for all object
565 1998-10-08 Hugo Tyson <hmt@masala.cygnus.co.uk>
567 In general, these changes are to support create/delete of uITRON
568 objects; this requires that an object can be destroyed whilst
569 there are threads waiting on it, and that they shall be awoken
570 with a specific return code E_DLT. The implementation uses an
571 array of pointers in addition to the array of objects: ptrs[N]
572 should either be NULL or &objs[N] in this version. That
573 restriction is not checked, in order that in future, true dynamic
574 allocation can be used, and ptrs[N] could point anywhere. The
575 implications of this are mainly in the area of getting the
576 addresses of objects, via the ptr array or from the object array?
577 The config of whether there is a ptr array is per object type.
579 * include/pkgconf/uitron.h:
580 Configury of the existence or not of create/delete for object
581 type: initially only SEMAS, MBOXES, FLAGS.
582 Configury of the number of objects created initially, if cre/del
588 Two new test programs (duh!): test6 tests create and delete of
589 SEMAS, MBOXES, FLAGS explicitly, for bad params handling, ability
590 to use after deletion and recreation &c &c. test7 tests the
591 semantics of sleeping on some synchronization primitive, permed
592 with suspension, signalling, killing, releasing and so forth.
593 Both will migrate into C++ versions also in future.
595 * include/uit_func.h:
596 Enable cre/del_{mbx,flg,sem} prototypes.
598 * include/uit_func.inl:
599 Implement cre/del_{mbx,flg,sem} depending on configury, and handle
600 the pointer arrays generally in other client routines.
602 * include/uit_objs.hxx:
603 Define the pointer arrays for access to created objects.
605 * include/uit_type.h:
606 Enable definitions of the structures used in create calls.
609 Initialization code for the pointer arrays, akin to the task
613 Instantiate the pointer arrays themselves.
616 1998-09-25 Bart Veer <bartv@cygnus.co.uk>
618 * include/pkgconf/uitron.h:
619 PR 17482: added #ifdef protection. If CYGPKG_UITRON_MEMPOOLFIXED
620 is not defined then the contained option
621 CYGNUM_UITRON_MEMPOOLFIXED may not be defined either. Ditto for
622 CYGNUM_UITRON_MEMPOOLVAR which suffers from the same problem.
623 And ditto for CYGPKG_UITRON_SEMAS
625 1998-09-20 Mark Galassi <rosalia@cygnus.com>
627 * include/pkgconf/uitron.h: added one or two CDL doc strings.
629 Tue Sep 15 19:12:28 1998 David Moore <dsm@keema.cygnus.co.uk>
631 * include/pkgconf/uitron.h: Cleaned up comments.
633 1998-09-12 Mark Galassi <rosalia@cygnus.com>
634 small changes to the descriptive information in the uITRON CDL
636 1998-09-12 Mark Galassi <rosalia@cygnus.com>
638 * include/pkgconf/uitron.h: small editing of the description
639 fields: renamed things like <name> to NAME, so that it does not
640 confuse SGML when incorporated into the documentation.
641 Also added various "doc" fields to the CFG_DATA comments.
643 Wed Sep 9 18:36:15 1998 Hugo Tyson <hmt@cygnus.co.uk>
645 * include/uit_func.inl:
646 * include/uit_objs.hxx:
648 Condition features on CYGPKG_UITRON_SEMAS [for example] as well as
649 0 < CYG_UITRON_NUM( SEMAS ) so that we get warning free
650 compilation with no semas enabled. Ditto for flags, mboxes,
651 mempools fixed and variable, alarms and cyclics.
653 * include/pkgconf/uitron.h:
654 Add a couple of requires statements from cyclics and alarms on the
656 Also change some grammar after review.
657 Add an option (actually a pair, makes it better documentation IMO)
658 to require strict semantics from the kernel setup.
659 Add an option to control the optional sema initializer; it was
660 omitted since the editor way of configury used a #if 0.
662 Fri Sep 4 17:54:29 1998 Hugo Tyson <hmt@cygnus.co.uk>
664 * include/pkgconf/uitron.h:
665 Remove redundant and now unnecessary defs of CYGPKG_UITRON.
667 Wed Sep 2 19:06:17 1998 Hugo Tyson <hmt@cygnus.co.uk>
669 * include/uit_func.inl:
670 Memory pools and message boxes and flags
671 and cnt_sem2-type semaphores now all have absolute timeouts
672 instead of relative ones in the timely wait functions.
673 We add in the current time here first.
675 Wed Sep 2 16:45:40 1998 Hugo Tyson <hmt@cygnus.co.uk>
677 * include/pkgconf/uitron.h (CYGDAT_UITRON_TASK_INITIALIZERS):
678 Use the new kernel thread initializer with priority and thread
681 * include/uit_func.inl:
682 * include/uit_objs.hxx:
683 * src/uit_func.cxx (cyg_uitron_start):
685 Our tasks are now Cyg_Threads rather than a derived class.
686 Original priorities of tasks are held in an array. An accessor
687 macro is used for the off-by-one nature of uitron IDs. The array
688 of original priorities is set up in cyg_uitron_start(), and
689 created in uit_objs.cxx.
691 Wed Sep 2 15:15:26 1998 Hugo Tyson <hmt@cygnus.co.uk>
693 * include/pkgconf/uitron.h:
694 Comment more strongly in the 'description' that the number of
695 initializers better match the number of things specified.
697 Tue Sep 1 19:00:02 1998 Hugo Tyson <hmt@masala.cygnus.co.uk>
699 * include/uit_objs.hxx:
700 Do not include <cyg/kernel/kernel.h>; it is deprecated.
701 Instead, include only kernel headers that are needed.
703 * include/uit_func.inl (ref_sys):
704 Use new kernel function Cyg_Interrupt::interrupts_enabled()
705 instead of a state variable. Tidy up associated code.
706 Include kernel headers that are needed because they were not
707 included in uit_objs.hxx for tidiness.
710 Elide state variable that used to be used for tracking interrupt
713 * tests/testcx5.cxx (task1):
714 * tests/test5.c (task1):
715 Test interrupt enable and disable stuff, and status reporting.
717 Tue Sep 1 15:12:29 1998 Hugo Tyson <hmt@cygnus.co.uk>
719 * include/uit_func.inl (ref_mpf):
720 Change comment "ECC" to "eCos".
722 1998-08-28 Bart Veer <bartv@cygnus.co.uk>
724 * include/uit_func.inl:
725 * tests/test1.c, tests/test2.c, tests/test3.c, tests/test4.c,
726 tests/test5.c, tests/testcxx.cxx, tests/testcx2.cxx,
727 tests/testcx3.cxx, tests/testcx4.cxx, tests/testcx5.cxx:
729 Updated for new kernel configuration option symbol names
732 Fri Aug 28 09:21:55 1998 Jonathan Larmour <jlarmour@cygnus.co.uk>
734 * src/uit_func.cxx (cyg_uitron_start):
735 Remove call to Cyg_Scheduler::start() and adjust comments to
736 explain where to call it from
738 * tests/test1.c, tests/test2.c, tests/test3.c, tests/test4.c,
739 tests/test5.c, tests/testcxx.cxx, tests/testcx2.cxx,
740 tests/testcx3.cxx, tests/testcx4.cxx, tests/testcx5.cxx:
741 Change the normal entry point at the top to be cyg_package_start()
742 under the new startup scheme. This overrides the default package
743 configuration and ensure cyg_uitron_start() gets called.
744 Change the "default" entry point at the bottom (when the test is
745 N/A for some reason) to cyg_start() under the new startup scheme.
747 Wed Aug 26 18:20:47 1998 Hugo Tyson <hmt@masala.cygnus.co.uk>
749 * include/pkgconf/uitron.h:
750 Initial version of cdl control statements for config options.
751 All very groovy, and looking good IMO.
752 A few FIX-MEs remain, no big deal.
754 Mon Aug 24 19:05:52 1998 Hugo Tyson <hmt@masala.cygnus.co.uk>
756 * include/pkgconf/uitron.h:
757 Change the names of config options to helpful, positive sense, and
758 tidy some of the names for inclusion in the GUI tool. Clean up
761 * include/uit_func.h:
762 * include/uit_func.inl:
765 Configure according to new option names and object construction
766 macros. Simple changes but widespread.
768 * tests/test[12345].c:
769 * tests/testcx[x2345].cxx:
770 Configure according to new option names.
771 Make it main( void ) to avoid a warning.
773 Fri Aug 21 18:45:16 1998 Hugo Tyson <hmt@cygnus.co.uk>
775 * tests/testcx4.cxx (task1):
776 * tests/test4.c (task1):
777 * include/uit_func.inl (def_alm):
778 Tidy the alarm and timer funcs in the course of fixing a kernel
779 bug actually. Test the functionality affected by the kernel
782 Thu Aug 20 14:33:36 1998 Hugo Tyson <hmt@masala.cygnus.co.uk>
784 * include/uit_func.inl (ref_alm):
785 Elide an unused variable following the previous change.
787 Mon Aug 17 15:45:07 1998 Hugo Tyson <hmt@masala.cygnus.co.uk>
789 * include/pkgconf/uitron.h:
790 Add the config option CYGSEM_UITRON_PARAMS_CHECK_NADR_ONLY.
795 Avoid testing with NULL as a bad parameter when only NADR is
796 checked. (NADR does not cast to any-pointer in C++ so we don't
797 test it in these programs) Check for NADR various places as well
800 * tests/test1.c (task1):
801 * tests/test2.c (task1):
802 * tests/test4.c (task1):
803 Avoid testing with NULL as a bad parameter when only NADR is
804 checked. Check for NADR various places as well as NULL, and test
805 NADR as a bad param as well as NULL.
808 Mon Aug 17 14:45:30 1998 Hugo Tyson <hmt@cygnus.co.uk>
810 * include/uit_func.inl:
811 PR# 16536: Check for dormant tasks various places; E_OBJ
812 must be returned. All this error checking rather bloats the code,
817 Test the new checks above.
819 Fri Aug 14 17:41:35 1998 Hugo Tyson <hmt@cygnus.co.uk>
821 * include/uit_func.inl ([t]wai/pol_flg):
822 Check for zero in a wait-for-flag operation; such a wait can never
823 be awakened, and so should return E_PAR.
825 * tests/testcx2.cxx (task1):
826 * tests/test2.c (task1):
827 Test the above new checking.
829 Fri Jul 24 17:12:54 1998 Hugo Tyson <hmt@masala.cygnus.co.uk>
832 * include/uit_func.inl: return a bool (strictly 1 or 0) for wtsk
833 rather than -1 or zero as the spec implies.
834 ref_flg() ref_sem() only affected.
836 Fri Jul 24 13:26:51 1998 Hugo Tyson <hmt@cygnus.co.uk>
839 * include/uit_func.inl: return E_OBJ when (force)resuming a
841 * tests/test1.c: test for that error code.
842 * tests/testcxx.cxx: test for that error code.
844 Fri Jul 24 13:02:46 1998 Hugo Tyson <hmt@cygnus.co.uk>
846 * Changelog: Initial ChangeLog entry.
849 //===========================================================================
850 //####ECOSGPLCOPYRIGHTBEGIN####
851 // -------------------------------------------
852 // This file is part of eCos, the Embedded Configurable Operating System.
853 // Copyright (C) 1998, 1999, 2000, 2001, 2002 Red Hat, Inc.
854 // Copyright (C) 2006 eCosCentric Ltd.
856 // eCos is free software; you can redistribute it and/or modify it under
857 // the terms of the GNU General Public License as published by the Free
858 // Software Foundation; either version 2 or (at your option) any later version.
860 // eCos is distributed in the hope that it will be useful, but WITHOUT ANY
861 // WARRANTY; without even the implied warranty of MERCHANTABILITY or
862 // FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License
865 // You should have received a copy of the GNU General Public License along
866 // with eCos; if not, write to the Free Software Foundation, Inc.,
867 // 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA.
869 // As a special exception, if other files instantiate templates or use macros
870 // or inline functions from this file, or you compile this file and link it
871 // with other works to produce a work based on this file, this file does not
872 // by itself cause the resulting work to be covered by the GNU General Public
873 // License. However the source code for this file must still be made available
874 // in accordance with section (3) of the GNU General Public License.
876 // This exception does not invalidate any other reasons why a work based on
877 // this file might be covered by the GNU General Public License.
878 // -------------------------------------------
879 //####ECOSGPLCOPYRIGHTEND####
880 //===========================================================================