Commit | Line | Data |
---|---|---|
c906108c | 1 | If you find inaccuracies in this list, please send mail to |
2a00c9ce AC |
2 | gdb-patches@sourceware.cygnus.com. If you would like to work on any |
3 | of these, you should consider sending mail to the same address, to | |
4 | find out whether anyone else is working on it. | |
c906108c | 5 | |
138f88c0 | 6 | |
2a00c9ce AC |
7 | Known problems in GDB 5.0 |
8 | ========================= | |
138f88c0 | 9 | |
bc9e5bbf AC |
10 | Below is a list of problems identified during the GDB 5.0 release |
11 | cycle. People hope to have these problems fixed in a follow-on | |
12 | release. | |
138f88c0 | 13 | |
bc9e5bbf | 14 | (The names in paren indicate people that posted the original problem.) |
138f88c0 AC |
15 | |
16 | -- | |
17 | ||
6e8cb14a | 18 | GDB requires GCC to build under IRIX |
c906108c | 19 | |
6e8cb14a AC |
20 | IRIX, being more pedantic than GCC reports as errors certain |
21 | assignments that GCC treats as warnings. | |
bc9e5bbf | 22 | |
6e8cb14a | 23 | This can be worked around by building GDB with the GCC compiler. |
138f88c0 AC |
24 | |
25 | -- | |
26 | ||
4fd99b5a AC |
27 | The BFD directory requires bug-fixed AUTOMAKE et.al. |
28 | ||
29 | AUTOMAKE 1.4 incorrectly set the TEXINPUTS environment variable. It | |
30 | contained the full path to texinfo.tex when it should have only | |
31 | contained the directory. The bug has been fixed in the current | |
32 | AUTOMAKE sources. Automake snapshots can be found in: | |
33 | ftp://sourceware.cygnus.com/pub/gdb/snapshots | |
34 | and ftp://sourceware.cygnus.com/pub/binutils | |
35 | ||
36 | -- | |
37 | ||
bc9e5bbf AC |
38 | gdb-cvs fails to build on freebsd-elf |
39 | http://sourceware.cygnus.com/ml/gdb/2000-04/msg00004.html | |
138f88c0 | 40 | |
bc9e5bbf AC |
41 | Either the FreeBSD group need to contribute their local GDB changes |
42 | back to the master sources or someone needs to provides a new | |
43 | (clean-room) implementation. Since the former involves a fairly | |
44 | complicated assignment the latter may be easier. [cagney] | |
138f88c0 AC |
45 | |
46 | -- | |
47 | ||
bc9e5bbf AC |
48 | Generic: lin-thread cannot handle thread exit (Mark Kettenis, Michael |
49 | Snyder) http://sourceware.cygnus.com/ml/gdb/2000-q1/msg00525.html | |
138f88c0 AC |
50 | |
51 | The thread_db assisted debugging code doesn't handle exiting threads | |
52 | properly, at least in combination with glibc 2.1.3 (the framework is | |
53 | there, just not the actual code). There are at least two problems | |
54 | that prevent this from working. | |
55 | ||
bc9e5bbf | 56 | As an additional reference point, the pre thread_db code did not work |
138f88c0 AC |
57 | either. |
58 | ||
59 | -- | |
60 | ||
61 | Java (Anthony Green, David Taylor) | |
62 | ||
bc9e5bbf AC |
63 | Anthony Green has a number of Java patches that did not make it into |
64 | the 5.0 release. | |
138f88c0 AC |
65 | |
66 | Patch: java tests | |
67 | http://sourceware.cygnus.com/ml/gdb-patches/2000-q1/msg00512.html | |
68 | ||
69 | Patch: java booleans | |
70 | http://sourceware.cygnus.com/ml/gdb-patches/2000-q1/msg00515.html | |
71 | ||
72 | Patch: handle N_MAIN stab | |
73 | http://sourceware.cygnus.com/ml/gdb-patches/2000-q1/msg00527.html | |
74 | ||
138f88c0 AC |
75 | -- |
76 | ||
77 | Pascal (Pierre Muller, David Taylor) | |
78 | ||
bc9e5bbf AC |
79 | Pierre Muller has contributed patches for adding Pascal Language |
80 | support to GDB. | |
67edb2c6 AC |
81 | |
82 | 2 pascal language patches inserted in database | |
83 | http://sourceware.cygnus.com/ml/gdb/2000-q1/msg00521.html | |
84 | ||
bc9e5bbf AC |
85 | Indent -gnu ? |
86 | http://sourceware.cygnus.com/ml/gdb/2000-q1/msg00496.html | |
138f88c0 AC |
87 | |
88 | -- | |
89 | ||
90 | GNU/Linux/x86 and random thread signals (and Solaris/SPARC but not | |
bc9e5bbf AC |
91 | Solaris/x86). |
92 | http://sourceware.cygnus.com/ml/gdb/2000-q1/msg00336.html | |
138f88c0 AC |
93 | |
94 | Christopher Blizzard writes: | |
95 | ||
96 | So, I've done some more digging into this and it looks like Jim | |
97 | Kingdon has reported this problem in the past: | |
98 | ||
99 | http://sourceware.cygnus.com/ml/bug-gdb/1999-10/msg00058.html | |
100 | ||
101 | I can reproduce this problem both with and without Tom's patch. Has | |
102 | anyone seen this before? Maybe have a solution for it hanging around? | |
103 | :) | |
104 | ||
105 | There's a test case for this documented at: | |
106 | ||
107 | when debugging threaded applications you get extra SIGTRAPs | |
108 | http://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=9565 | |
109 | ||
110 | [There should be a GDB testcase - cagney] | |
111 | ||
112 | -- | |
113 | ||
bc9e5bbf | 114 | Possible regressions with some devel GCCs. |
138f88c0 AC |
115 | http://sourceware.cygnus.com/ml/gdb/2000-q1/msg00475.html |
116 | ||
bc9e5bbf AC |
117 | gcc-2.95.2 outputs a line note *before* the prologue (and one for the |
118 | closing brace after the epilogue, instead of before it, as it used to | |
119 | be). By disabling the RTL-style prologue generating mechanism | |
120 | (undocumented GCC option -mno-schedule-prologue), you get back the | |
121 | traditional behaviour. | |
122 | http://sourceware.cygnus.com/ml/gdb/2000-q1/msg00510.html | |
138f88c0 | 123 | |
bc9e5bbf | 124 | This should now be fixed. |
138f88c0 AC |
125 | |
126 | -- | |
127 | ||
bc9e5bbf AC |
128 | RFD: infrun.c: No bpstat_stop_status call after proceed over break? |
129 | (Peter Schauer) | |
138f88c0 AC |
130 | http://sourceware.cygnus.com/ml/gdb-patches/2000-q1/msg00665.html |
131 | ||
bc9e5bbf AC |
132 | GDB misses watchpoint triggers after proceeding over a breakpoint on |
133 | x86 targets. | |
134 | ||
138f88c0 AC |
135 | -- |
136 | ||
67edb2c6 AC |
137 | x86 linux GDB and SIGALRM (???) |
138 | http://sourceware.cygnus.com/ml/gdb/2000-q1/msg00803.html | |
139 | ||
bc9e5bbf AC |
140 | I know there are problems with single stepping through signal |
141 | handlers. These problems were present in 4.18. They were just masked | |
142 | because 4.18 failed to recognize signal handlers. Fixing it is not | |
143 | easy, and will require changes to handle_inferior_event(), that I | |
144 | prefer not to make before the 5.0 release. | |
67edb2c6 | 145 | |
bc9e5bbf | 146 | Mark |
67edb2c6 AC |
147 | |
148 | -- | |
149 | ||
9d6d78f2 AC |
150 | Revised UDP support (was: Re: [Fwd: [patch] UDP transport support]) |
151 | http://sourceware.cygnus.com/ml/gdb-patches/2000-04/msg00000.html | |
152 | ||
bc9e5bbf AC |
153 | (Broken) support for GDB's remote protocol across UDP is to be |
154 | included in the follow-on release. | |
138f88c0 | 155 | |
b2f4b24d AC |
156 | -- |
157 | ||
158 | Can't build IRIX -> arm GDB. | |
159 | http://sourceware.cygnus.com/ml/gdb-patches/2000-04/msg00356.html | |
160 | ||
161 | David Whedon writes: | |
162 | > Now I'm building for an embedded arm target. If there is a way of turning | |
163 | > remote-rdi off, I couldn't find it. It looks like it gets built by default | |
164 | > in gdb/configure.tgt(line 58) Anyway, the build dies in | |
165 | > gdb/rdi-share/unixcomm.c. SERPORT1 et. al. never get defined because we | |
166 | > aren't one of the architectures supported. | |
167 | ||
6bc37a96 AC |
168 | -- |
169 | ||
170 | Problem with weak functions | |
171 | http://sourceware.cygnus.com/ml/gdb/2000-05/msg00060.html | |
172 | ||
173 | Dan Nicolaescu writes: | |
174 | > It seems that gdb-4.95.1 does not display correctly the function when | |
175 | > stoping in weak functions. | |
176 | > | |
177 | > It stops in a function that is defined as weak, not in the function | |
178 | > that is actualy run... | |
b2f4b24d | 179 | |
3fffcb5e AC |
180 | -- |
181 | ||
182 | GDB5 TOT on unixware 7 | |
183 | http://sourceware.cygnus.com/ml/gdb/2000-04/msg00119.html | |
184 | ||
185 | Robert Lipe writes: | |
186 | > I just spun the top of tree of the GDB5 branch on UnixWare 7. As a | |
187 | > practical matter, the current thread support is somewhat more annoying | |
188 | > than when GDB was thread-unaware. | |
189 | ||
138f88c0 AC |
190 | -- |
191 | ||
7ae38352 | 192 | Code Cleanups |
2a00c9ce | 193 | ============= |
bc9e5bbf | 194 | |
7ae38352 AC |
195 | The following are small cleanups that will hopefully be completed by |
196 | the follow on to 5.0. | |
bc9e5bbf AC |
197 | |
198 | -- | |
199 | ||
200 | ATTRIBUTE_UNUSED | |
201 | ||
202 | The need for this as almost been eliminated. The next version of GCC | |
203 | (assuming cagney gets the relevant patch committed) will be able to | |
204 | supress unused parameter warnings. | |
205 | ||
206 | -- | |
207 | ||
d8038014 AC |
208 | Eliminate more compiler warnings. |
209 | ||
210 | Of course there also needs to be the usual debate over which warnings | |
211 | are valid and how to best go about this. | |
212 | ||
213 | One method: choose a single option; get agreement that it is | |
214 | reasonable; try it out to see if there isn't anything silly about it | |
215 | (-Wunused-parameters is an example of that) then incrementally hack | |
216 | away. | |
217 | ||
218 | The other method is to enable all warnings and eliminate them from one | |
219 | file at a time. | |
220 | ||
221 | -- | |
222 | ||
bc9e5bbf AC |
223 | Delete macro TARGET_BYTE_ORDER_SELECTABLE. |
224 | ||
225 | Patches in the database. | |
226 | ||
227 | -- | |
228 | ||
229 | Updated readline | |
230 | ||
231 | Readline 4.? is out. A merge wouldn't hurt. | |
232 | ||
233 | -- | |
234 | ||
235 | Purge PARAMS | |
236 | ||
237 | Eliminate all uses of PARAMS in GDB's source code. | |
238 | ||
239 | -- | |
240 | ||
241 | Elimination of make_cleanup_func. (Andrew Cagney) | |
242 | ||
243 | make_cleanup_func elimination | |
244 | http://sourceware.cygnus.com/ml/gdb-patches/2000-q1/msg00791.html | |
245 | http://sourceware.cygnus.com/ml/gdb-patches/2000-q1/msg00814.html | |
246 | ||
247 | -- | |
248 | ||
5683e87a AC |
249 | Elimination of ``(catch_errors_ftype *) func''. |
250 | ||
251 | Like make_cleanup_func it isn't portable. | |
252 | ||
253 | -- | |
254 | ||
bc9e5bbf AC |
255 | Re: Various C++ things |
256 | ||
257 | value_headof/value_from_vtable_info are worthless, and should be removed. | |
258 | The one place in printcmd.c that uses it should use the RTTI functions. | |
259 | ||
260 | RTTI for g++ should be using the typeinfo functions rather than the vtables. | |
261 | The typeinfo functions are always at offset 4 from the beginning of the vtable, | |
262 | and are always right. The vtables will have weird names like E::VB sometimes. | |
263 | The typeinfo function will always be "E type_info function", or somesuch. | |
264 | ||
265 | value_virtual_fn_field needs to be fixed so there are no failures for virtual | |
266 | functions for C++ using g++. | |
267 | ||
268 | Testsuite cases are the major priority right now for C++ support, since i have | |
269 | to make a lot of changes that could potentially break each other. | |
138f88c0 | 270 | |
67edb2c6 AC |
271 | -- |
272 | ||
7ae38352 AC |
273 | Fix ``set architecture <tab>'' |
274 | ||
275 | This command should expand to a list of all supported architectures. | |
276 | At present ``info architecture'' needs to be used. That is simply | |
277 | wrong. It involves the use of add_set_enum_cmd(). | |
278 | ||
279 | -- | |
280 | ||
67edb2c6 AC |
281 | GDBARCH cleanup (Andrew Cagney) |
282 | ||
283 | The non-generated parts of gdbarch.{sh,h,c} should be separated out | |
bc9e5bbf | 284 | into arch-utils.[hc]. |
67edb2c6 | 285 | |
67edb2c6 AC |
286 | Document that gdbarch_init_ftype could easily fail because it didn't |
287 | identify an architecture. | |
288 | ||
289 | -- | |
290 | ||
bc9e5bbf AC |
291 | Migrate qfThreadInfo packet -> qThreadInfo. (Andrew Cagney) |
292 | ||
293 | Add support for packet enable/disable commands with these thread | |
294 | packets. General cleanup. | |
295 | ||
296 | [PATCH] Document the ThreadInfo remote protocol queries | |
297 | http://sourceware.cygnus.com/ml/gdb-patches/2000-q1/msg00832.html | |
298 | ||
299 | [PATCH] "info threads" queries for remote.c | |
300 | http://sourceware.cygnus.com/ml/gdb-patches/2000-q1/msg00831.html | |
301 | ||
7ae38352 AC |
302 | -- |
303 | ||
304 | General Cleanups / Fixes | |
305 | ======================== | |
306 | ||
307 | The following are more general cleanups and fixes. They are not tied | |
308 | to any specific release. | |
309 | ||
310 | -- | |
311 | ||
312 | Nuke USG define. | |
313 | ||
bc9e5bbf AC |
314 | -- |
315 | ||
747d1ccb AC |
316 | Eliminate gdb/tui/Makefile.in. |
317 | Cleanup configury support for optional sub-directories. | |
318 | ||
319 | Check how GCC handles multiple front ends for an example of how things | |
320 | could work. A tentative first step is to rationalize things so that | |
321 | all sub directories are handled in a fashion similar to gdb/mi. | |
322 | ||
323 | -- | |
324 | ||
9debab2f AC |
325 | [PATCH/5] src/intl/Makefile.in:distclean additions |
326 | http://sourceware.cygnus.com/ml/gdb-patches/2000-04/msg00363.html | |
327 | ||
328 | Do not forget to merge the patch back into the trunk. | |
329 | ||
330 | -- | |
331 | ||
7ae38352 AC |
332 | Update ALPHA so that it uses ``struct frame_extra_info'' instead of |
333 | EXTRA_FRAME_INFO. | |
e55e8cee | 334 | |
7ae38352 AC |
335 | This is a barrier to replacing mips_extra_func_info with something |
336 | that works with multi-arch. | |
e55e8cee AC |
337 | |
338 | -- | |
339 | ||
7ae38352 AC |
340 | Multi-arch mips_extra_func_info. |
341 | ||
342 | This first needs the alpha to be updated so that it uses ``struct | |
343 | frame_extra_info''. | |
344 | ||
345 | -- | |
346 | ||
347 | Send normal output to gdb_stdout. | |
348 | Send error messages to gdb_stderror. | |
349 | Send debug and log output log gdb_stdlog. | |
350 | ||
351 | GDB still contains many cases where (f)printf or printf_filtered () is | |
352 | used when it should be sending the messages to gdb_stderror or | |
5683e87a AC |
353 | gdb_stdlog. The thought of #defining printf to something has crossed |
354 | peoples minds ;-) | |
7ae38352 AC |
355 | |
356 | -- | |
357 | ||
358 | Rationalize the host-endian code (grep for HOST_BYTE_ORDER). | |
359 | ||
5683e87a | 360 | At present defs.h includes <endian.h> (which is linux specific) yet |
7ae38352 AC |
361 | almost nothing depends on it. Suggest "gdb_endian.h" which can also |
362 | handle <machine/endian.h> and only include that where it is really | |
363 | needed. | |
364 | ||
365 | -- | |
366 | ||
367 | Replace asprintf() calls with xasprintf() calls. | |
368 | ||
369 | As with things like strdup() most calls to asprintf() don't check the | |
370 | return value. | |
371 | ||
372 | -- | |
373 | ||
374 | Rationaize savestring(), msavestring() and mstrsave(). | |
375 | ||
376 | In general libiberty's xstrdup () can be used. | |
377 | ||
378 | -- | |
379 | ||
380 | Eliminate mmalloc() from GDB. | |
381 | ||
382 | Also eliminate it from defs.h. | |
383 | ||
384 | -- | |
385 | ||
386 | Check/cleanup MI documentation. | |
387 | ||
388 | The list of commands specified in the documentation needs to be | |
389 | checked against the mi-cmds.c table in a mechanical way (so that they | |
390 | two can be kept up-to-date). | |
391 | ||
392 | -- | |
393 | ||
394 | Eliminate error_begin(). | |
395 | ||
396 | With ui_file, there is no need for the statefull error_begin () | |
397 | function. | |
398 | ||
399 | -- | |
400 | ||
401 | Add built-by, build-date, tm, xm, nm and anything else into gdb binary | |
402 | so that you can see how the GDB was created. | |
403 | ||
404 | Some of these (*m.h) would be added to the generated config.h. That | |
405 | in turn would fix a long standing bug where by the build process many | |
406 | not notice a changed tm.h file. Since everything depends on config.h, | |
407 | a change to *m.h forces a change to config.h and, consequently forces | |
408 | a rebuild. | |
409 | ||
410 | -- | |
411 | ||
412 | Replace gdb_stdtarg with gdb_targout (and possibly gdb_targerr). | |
413 | ||
414 | gdb_stdtarg is easily confused with gdb_stdarg. | |
415 | ||
416 | -- | |
417 | ||
418 | Remote protocol doco feedback. | |
419 | ||
420 | Too much feedback to mention needs to be merged in (901660). Search | |
421 | for the word ``remote''. | |
422 | ||
423 | -- | |
424 | ||
425 | set/show remote X-packet ... | |
426 | ||
427 | ``(gdb) help set remote X-packet'' doesn't list the applicable | |
428 | responses. The help message needs to be expanded. | |
429 | ||
430 | -- | |
431 | ||
432 | Extra ui_file methods - dump. | |
433 | ||
434 | These are for debugging / testing. An aside is to set up a whitebox | |
435 | testsuite for key internals such as ui_file. | |
436 | ||
437 | -- | |
438 | ||
439 | Add an "info bfd" command that displays supported object formats, | |
440 | similarly to objdump -i. | |
441 | ||
442 | Is there a command already? | |
443 | ||
d8038014 AC |
444 | -- |
445 | ||
446 | Eliminate PTR. ISO-C allows ``void *''. | |
447 | ||
448 | -- | |
449 | ||
450 | Eliminate abort (). | |
451 | ||
452 | GDB should never abort. GDB should either throw ``error ()'' or | |
453 | ``internal_error ()''. Better still GDB should naturally unwind with | |
454 | an error status. | |
455 | ||
7ae38352 AC |
456 | -- |
457 | ||
458 | Architectural Changes | |
459 | ===================== | |
460 | ||
461 | These are harder than simple cleanups / fixes and, consequently | |
462 | involve more work. Typically an Architectural Change will be broken | |
463 | down into a more digestible set of cleanups and fixes. | |
464 | ||
465 | -- | |
466 | ||
467 | Replace READ_FP() with FRAME_HANDLE(). | |
468 | ||
469 | READ_FP() is a hangover from the days of the vax when the ABI really | |
470 | did have a frame pointer register. Modern architectures typically | |
471 | construct a virtual frame-handle from the stack pointer and various | |
472 | other bits of string. | |
473 | ||
474 | Unfortunatly GDB still treats this synthetic FP register as though it | |
475 | is real. That in turn really confuses users (arm and ``print $fp'' VS | |
476 | ``info registers fp''). The synthetic FP should be separated out of | |
477 | the true register set presented to the user. | |
478 | ||
479 | -- | |
480 | ||
481 | MI's input does not use buffering. | |
482 | ||
483 | At present the MI interface reads raw characters of from an unbuffered | |
484 | FD. This is to avoid several nasty buffer/race conditions. That code | |
485 | should be changed so that it registers its self with the event loop | |
486 | (on the input FD) and then push commands up to MI as they arrive. | |
487 | ||
488 | The serial code already does this. | |
2a00c9ce AC |
489 | |
490 | -- | |
491 | ||
492 | Register Cache Cleanup (below from Andrew Cagney) | |
493 | ||
494 | I would depict the current register architecture as something like: | |
495 | ||
496 | High GDB --> Low GDB | |
497 | | | | |
498 | \|/ \|/ | |
499 | --- REG NR ----- | |
500 | | | |
501 | register + REGISTER_BYTE(reg_nr) | |
502 | | | |
503 | \|/ | |
504 | ------------------------- | |
505 | | extern register[] | | |
506 | ------------------------- | |
507 | ||
508 | where neither the high (valops.c et.al.) or low gdb (*-tdep.c) are | |
509 | really clear on what mechanisms they should be using to manipulate that | |
510 | buffer. Further, much code assumes, dangerously, that registers are | |
511 | contigious. Having got mips-tdep.c to support multiple ABIs, believe | |
512 | me, that is a bad assumption. Finally, that register cache layout is | |
513 | determined by the current remote/local target and _not_ the less | |
514 | specific target ISA. In fact, in many cases it is determined by the | |
515 | somewhat arbitrary layout of the [gG] packets! | |
516 | ||
517 | ||
518 | How I would like the register file to work is more like: | |
519 | ||
520 | ||
521 | High GDB | |
522 | | | |
523 | \|/ | |
524 | pseudo reg-nr | |
525 | | | |
526 | map pseudo <-> | |
527 | random cache | |
528 | bytes | |
529 | | | |
530 | \|/ | |
531 | ------------ | |
532 | | register | | |
533 | | cache | | |
534 | ------------ | |
535 | /|\ | |
536 | | | |
537 | map random cache | |
538 | bytes to target | |
539 | dependant i-face | |
540 | /|\ | |
541 | | | |
542 | target dependant | |
543 | such as [gG] packet | |
544 | or ptrace buffer | |
545 | ||
546 | The main objectives being: | |
547 | ||
548 | o a clear separation between the low | |
549 | level target and the high level GDB | |
550 | ||
551 | o a mechanism that solves the general | |
552 | problem of register aliases, overlaps | |
553 | etc instead of treating them as optional | |
554 | extras that can be wedged in as an after | |
555 | thought (that is a reasonable description | |
556 | of the current code). | |
557 | ||
558 | Identify then solve the hard case and the | |
559 | rest just falls out. GDB solved the easy | |
560 | case and then tried to ignore the real | |
561 | world :-) | |
562 | ||
563 | o a removal of the assumption that the | |
564 | mapping between the register cache | |
565 | and virtual registers is largely static. | |
566 | If you flip the USR/SSR stack register | |
567 | select bit in the status-register then | |
568 | the corresponding stack registers should | |
569 | reflect the change. | |
570 | ||
571 | o a mechanism that clearly separates the | |
572 | gdb internal register cache from any | |
573 | target (not architecture) dependant | |
574 | specifics such as [gG] packets. | |
575 | ||
576 | Of course, like anything, it sounds good in theory. In reality, it | |
577 | would have to contend with many<->many relationships at both the | |
578 | virt<->cache and cache<->target level. For instance: | |
579 | ||
580 | virt<->cache | |
581 | Modifying an mmx register may involve | |
582 | scattering values across both FP and | |
583 | mmpx specific parts of a buffer | |
584 | ||
585 | cache<->target | |
586 | When writing back a SP it may need to | |
587 | both be written to both SP and USP. | |
588 | ||
589 | ||
590 | Hmm, | |
591 | ||
592 | Rather than let this like the last time it was discussed, just slip, I'm | |
593 | first going to add this e-mail (+ references) to TODO. I'd then like to | |
594 | sketch out a broad strategy I think could get us there. | |
595 | ||
596 | ||
597 | First thing I'd suggest is separating out the ``extern registers[]'' | |
598 | code so that we can at least identify what is using it. At present | |
599 | things are scattered across many files. That way we can at least | |
600 | pretend that there is a cache instead of a global array :-) | |
601 | ||
602 | I'd then suggest someone putting up a proposal for the pseudo-reg / | |
603 | high-level side interface so that code can be adopted to it. For old | |
604 | code, initially a blanket rename of write_register_bytes() to | |
605 | deprecated_write_register_bytes() would help. | |
606 | ||
607 | Following that would, finaly be the corresponding changes to the target. | |
bc9e5bbf AC |
608 | |
609 | -- | |
610 | ||
d8038014 AC |
611 | Fix ``I'm sorry, Dave, I can't do that.'' from symfile.c. |
612 | ||
613 | This requires internationalization. | |
614 | ||
615 | -- | |
616 | ||
67edb2c6 AC |
617 | Check that GDB can handle all BFD architectures (Andrew Cagney) |
618 | ||
619 | There should be a test that checks that BFD/GDB are in sync with | |
620 | regard to architecture changes. Something like a test that first | |
621 | queries GDB for all supported architectures and then feeds each back | |
622 | to GDB.. Anyone interested in learning how to write tests? :-) | |
623 | ||
624 | -- | |
625 | ||
2a00c9ce AC |
626 | Add support for Modula3 |
627 | ||
628 | Get DEC/Compaq to contribute their Modula-3 support. | |
629 | ||
7ae38352 AC |
630 | -- |
631 | ||
632 | Convert ALL architectures to MULTI-ARCH. | |
633 | ||
634 | -- | |
635 | ||
636 | Convert GDB build process to AUTOMAKE. | |
637 | ||
638 | -- | |
639 | ||
640 | Restructure gdb directory tree so that it avoids any 8.3 and 14 | |
641 | filename problems. | |
642 | ||
643 | -- | |
644 | ||
645 | Can the xdep files be replaced by autoconf? | |
646 | Can the tm.h and nm.h files be eliminated by multi-arch. | |
647 | ||
648 | -- | |
649 | ||
650 | Add a transcript mechanism to GDB. | |
651 | ||
652 | Such a mechanism might log all gdb input and output to a file in a | |
653 | form that would allow it to be replayed. It could involve ``gdb | |
654 | --transcript=FILE'' or it could involve ``(gdb) transcript file''. | |
655 | ||
656 | -- | |
657 | ||
658 | Make MI interface accessable from existing CLI. | |
659 | ||
660 | -- | |
661 | ||
662 | Select the initial multi-arch ISA / ABI based on --target or similar. | |
663 | ||
664 | At present the default is based on what ever is first in the BFD | |
665 | archures table. It should be determined based on the ``--target=...'' | |
666 | name. | |
667 | ||
668 | -- | |
669 | ||
670 | Truly multi-arch. | |
671 | ||
672 | Enable the code to recognize --enable-targets=.... like BINUTILS does. | |
673 | ||
674 | -- | |
675 | ||
676 | Add a breakpoint-edit command to MI. | |
677 | ||
678 | It would be similar to MI's breakpoint create but would apply to an | |
679 | existing breakpoint. It saves the need to delete/create breakpoints | |
680 | when ever they are changed. | |
681 | ||
682 | -- | |
683 | ||
684 | Add directory path to MI breakpoint. | |
685 | ||
686 | That way the GUI's task of finding the file within which the | |
687 | breakpoint was set is simplified. | |
688 | ||
689 | -- | |
690 | ||
691 | Re-do GDB's output pager. | |
692 | ||
693 | GDB's output pager still relies on people correctly using *_filtered | |
694 | for gdb_stdout and *_unfiltered for gdb_stdlog / gdb_stderr. | |
695 | Hopefully, with all normal output going to gdb_stdout, the pager can | |
696 | just look at the ui_file that the output is on and then use that to | |
697 | decide what to do about paging. Sounds good in theory. | |
698 | ||
699 | -- | |
700 | ||
701 | Add mechanism to reject expression classes to MI | |
702 | ||
703 | There are situtations where you don't want GDB's expression | |
704 | parser/evaluator to perform inferior function calls or variable | |
705 | assignments. | |
706 | ||
707 | -- | |
708 | ||
709 | Remove sideffects from libgdb breakpoint create function. | |
710 | ||
711 | The user can use the CLI to create a breakpoint with partial | |
712 | information - no file (gdb would use the file from the last | |
713 | breakpoint). | |
714 | ||
715 | The libgdb interface currently affects that environment which can lead | |
716 | to confusion when a user is setting breakpoints via both the MI and | |
717 | the CLI. | |
718 | ||
719 | This is also a good example of how getting the CLI ``right'' will be | |
720 | hard. | |
721 | ||
722 | -- | |
723 | ||
724 | GDB doesn't recover gracefully from remote protocol errors. | |
725 | ||
726 | GDB wasn't checking for NAKs from the remote target. Instead a NAK is | |
727 | ignored and a timeout is required before GDB retries. A pre-cursor to | |
728 | fixing this this is making GDB's remote protocol packet more robust. | |
729 | ||
730 | While downloading to a remote protocol target, gdb ignores packet | |
731 | errors in so far as it will continue to edownload with chunk N+1 even | |
732 | if chunk N was not correctly sent. This causes gdb.base/remote.exp to | |
733 | take a painfully long time to run. As a PS that test needs to be | |
734 | fixed so that it builds on 16 bit machines. | |
735 | ||
736 | -- | |
737 | ||
738 | Move gdb_lasterr to ui_out? | |
739 | ||
740 | The way GDB throws errors and records them needs a re-think. ui_out | |
741 | handles the correct output well. It doesn't resolve what to do with | |
742 | output / error-messages when things go wrong. | |
743 | ||
744 | -- | |
745 | ||
746 | Fix implementation of ``target xxx''. | |
747 | ||
748 | At present when the user specifies ``target xxxx'', the CLI maps that | |
749 | directly onto a target open method. It is then assumed that the | |
750 | target open method should do all sorts of complicated things as this | |
751 | is the only chance it has. Check how the various remote targets | |
752 | duplicate the target operations. Check also how the various targets | |
753 | behave differently for purely arbitrary reasons. | |
754 | ||
755 | What should happen is that ``target xxxx'' should call a generic | |
756 | ``target'' function and that should then co-ordinate the opening of | |
757 | ``xxxx''. This becomes especially important when you're trying to | |
758 | open an asynchronous target that may need to perform background tasks | |
759 | as part of the ``attach'' phase. | |
760 | ||
761 | Unfortunatly, due to limitations in the old/creaking command.h | |
762 | interface, that isn't possible. The function being called isn't told | |
763 | of the ``xxx'' or any other context information. | |
764 | ||
765 | Consequently a precursor to fixing ``target xxxx'' is to clean up the | |
766 | CLI code so that it passes to the callback function (attatched to a | |
767 | command) useful information such as the actual command and a context | |
768 | for that command. Other changes such as making ``struct command'' | |
769 | opaque may also help. | |
770 | ||
771 | -- | |
772 | ||
773 | Document trace machinery | |
774 | ||
775 | -- | |
776 | ||
777 | Document overlay machinery. | |
778 | ||
2a00c9ce AC |
779 | -- |
780 | ||
781 | Legacy Wish List | |
782 | ================ | |
783 | ||
784 | This list is not up to date, and opinions vary about the importance or | |
785 | even desirability of some of the items. If you do fix something, it | |
786 | always pays to check the below. | |
787 | ||
788 | -- | |
c906108c | 789 | |
b83266a0 SS |
790 | @c This does not work (yet if ever). FIXME. |
791 | @c @item --parse=@var{lang} @dots{} | |
792 | @c Configure the @value{GDBN} expression parser to parse the listed languages. | |
793 | @c @samp{all} configures @value{GDBN} for all supported languages. To get a | |
794 | @c list of all supported languages, omit the argument. Without this | |
795 | @c option, @value{GDBN} is configured to parse all supported languages. | |
796 | ||
7ae38352 | 797 | -- |
c906108c SS |
798 | |
799 | START_INFERIOR_TRAPS_EXPECTED need never be defined to 2, since that | |
800 | is its default value. Clean this up. | |
801 | ||
7ae38352 AC |
802 | -- |
803 | ||
c906108c SS |
804 | It should be possible to use symbols from shared libraries before we know |
805 | exactly where the libraries will be loaded. E.g. "b perror" before running | |
806 | the program. This could maybe be done as an extension of the "breakpoint | |
807 | re-evaluation" after new symbols are loaded. | |
808 | ||
7ae38352 AC |
809 | -- |
810 | ||
c906108c SS |
811 | Make single_step() insert and remove breakpoints in one operation. |
812 | ||
7ae38352 AC |
813 | -- |
814 | ||
c906108c SS |
815 | Speed up single stepping by avoiding extraneous ptrace calls. |
816 | ||
7ae38352 AC |
817 | -- |
818 | ||
c906108c SS |
819 | Speed up single stepping by not inserting and removing breakpoints |
820 | each time the inferior starts and stops. | |
821 | ||
822 | Breakpoints should not be inserted and deleted all the time. Only the | |
823 | one(s) there should be removed when we have to step over one. Support | |
824 | breakpoints that don't have to be removed to step over them. | |
825 | ||
7ae38352 | 826 | [this has resulted in numerous debates. The issue isn't clear cut] |
c906108c | 827 | |
7ae38352 | 828 | -- |
c906108c SS |
829 | |
830 | Provide "voodoo" debugging of core files. This creates a zombie | |
831 | process as a child of the debugger, and loads it up with the data, | |
832 | stack, and regs of the core file. This allows you to call functions | |
833 | in the executable, to manipulate the data in the core file. | |
834 | ||
7ae38352 | 835 | [you wish] |
c906108c | 836 | |
7ae38352 | 837 | -- |
c906108c | 838 | |
7ae38352 | 839 | GDB reopens the source file on every line, as you "next" through it. |
c906108c | 840 | |
7ae38352 | 841 | [still true? I've a memory of this being fixed] |
c906108c | 842 | |
7ae38352 | 843 | -- |
c906108c SS |
844 | |
845 | Perhaps "i source" should take an argument like that of "list". | |
846 | ||
7ae38352 | 847 | -- |
c906108c SS |
848 | |
849 | Remove "at 0xnnnn" from the "b foo" response, if `print address off' and if | |
850 | it matches the source line indicated. | |
851 | ||
7ae38352 | 852 | -- |
c906108c | 853 | |
7ae38352 | 854 | The prompt at end of screen should accept space as well as CR. |
c906108c | 855 | |
7ae38352 | 856 | -- |
c906108c SS |
857 | |
858 | Backtrace should point out what the currently selected frame is, in | |
859 | its display, perhaps showing "@3 foo (bar, ...)" or ">3 foo (bar, | |
860 | ...)" rather than "#3 foo (bar, ...)". | |
861 | ||
7ae38352 AC |
862 | -- |
863 | ||
c906108c SS |
864 | "i program" should work for core files, and display more info, like what |
865 | actually caused it to die. | |
866 | ||
7ae38352 AC |
867 | -- |
868 | ||
c906108c SS |
869 | "x/10i" should shorten the long name, if any, on subsequent lines. |
870 | ||
7ae38352 | 871 | -- |
c906108c SS |
872 | |
873 | "next" over a function that longjumps, never stops until next time you happen | |
874 | to get to that spot by accident. E.g. "n" over execute_command which has | |
875 | an error. | |
876 | ||
7ae38352 AC |
877 | -- |
878 | ||
c906108c SS |
879 | "set zeroprint off", don't bother printing members of structs which |
880 | are entirely zero. Useful for those big structs with few useful | |
881 | members. | |
882 | ||
7ae38352 AC |
883 | -- |
884 | ||
c906108c SS |
885 | GDB does four ioctl's for every command, probably switching terminal modes |
886 | to/from inferior or for readline or something. | |
887 | ||
7ae38352 AC |
888 | -- |
889 | ||
c906108c SS |
890 | terminal_ours versus terminal_inferior: cache state. Switch should be a noop |
891 | if the state is the same, too. | |
892 | ||
7ae38352 | 893 | -- |
c906108c SS |
894 | |
895 | "i frame" shows wrong "arglist at" location, doesn't show where the args | |
896 | should be found, only their actual values. | |
897 | ||
7ae38352 AC |
898 | -- |
899 | ||
c906108c SS |
900 | There should be a way for "set" commands to validate the new setting |
901 | before it takes effect. | |
902 | ||
7ae38352 | 903 | -- |
c906108c SS |
904 | |
905 | "ena d" is ambiguous, why? "ena delete" seems to think it is a command! | |
906 | ||
7ae38352 | 907 | -- |
c906108c SS |
908 | |
909 | i line VAR produces "Line number not known for symbol ``var''.". I | |
910 | thought we were stashing that info now! | |
911 | ||
7ae38352 AC |
912 | -- |
913 | ||
c906108c SS |
914 | We should be able to write to random files at hex offsets like adb. |
915 | ||
7ae38352 AC |
916 | -- |
917 | ||
c906108c SS |
918 | Make "target xxx" command interruptible. |
919 | ||
7ae38352 AC |
920 | -- |
921 | ||
922 | [elena - delete this] | |
923 | ||
c906108c SS |
924 | Handle add_file with separate text, data, and bss addresses. Maybe |
925 | handle separate addresses for each segment in the object file? | |
926 | ||
7ae38352 AC |
927 | -- |
928 | ||
929 | [Jimb/Elena delete this one] | |
930 | ||
c906108c SS |
931 | Handle free_named_symtab to cope with multiply-loaded object files |
932 | in a dynamic linking environment. Should remember the last copy loaded, | |
933 | but not get too snowed if it finds references to the older copy. | |
934 | ||
7ae38352 AC |
935 | -- |
936 | ||
937 | [elena delete this also] | |
c906108c SS |
938 | |
939 | Remove all references to: | |
940 | text_offset | |
941 | data_offset | |
942 | text_data_start | |
943 | text_end | |
944 | exec_data_offset | |
945 | ... | |
946 | now that we have BFD. All remaining are in machine dependent files. | |
947 | ||
7ae38352 | 948 | -- |
c906108c SS |
949 | |
950 | Re-organize help categories into things that tend to fit on a screen | |
951 | and hang together. | |
952 | ||
7ae38352 | 953 | -- |
c906108c SS |
954 | |
955 | Add in commands like ADB's for searching for patterns, etc. We should | |
956 | be able to examine and patch raw unsymboled binaries as well in gdb as | |
957 | we can in adb. (E.g. increase the timeout in /bin/login without source). | |
958 | ||
7ae38352 AC |
959 | [actually, add ADB interface :-] |
960 | ||
961 | -- | |
c906108c SS |
962 | |
963 | When doing "step" or "next", if a few lines of source are skipped between | |
964 | the previous line and the current one, print those lines, not just the | |
965 | last line of a multiline statement. | |
966 | ||
7ae38352 | 967 | -- |
c906108c SS |
968 | |
969 | Handling of "&" address-of operator needs some serious overhaul | |
970 | for ANSI C and consistency on arrays and functions. | |
971 | For "float point[15];": | |
972 | ptype &point[4] ==> Attempt to take address of non-lvalue. | |
973 | For "char *malloc();": | |
974 | ptype malloc ==> "char *()"; should be same as | |
975 | ptype &malloc ==> "char *(*)()" | |
976 | call printf ("%x\n", malloc) ==> weird value, should be same as | |
977 | call printf ("%x\n", &malloc) ==> correct value | |
978 | ||
7ae38352 AC |
979 | -- |
980 | ||
c906108c SS |
981 | Fix dbxread.c symbol reading in the presence of interrupts. It |
982 | currently leaves a cleanup to blow away the entire symbol table when a | |
983 | QUIT occurs. (What's wrong with that? -kingdon, 28 Oct 1993). | |
984 | ||
7ae38352 AC |
985 | [I suspect that the grype was that, on a slow system, you might want |
986 | to cntrl-c and get just half the symbols and then load the rest later | |
987 | - scary to be honest] | |
988 | ||
989 | -- | |
990 | ||
c906108c SS |
991 | Mipsread.c reads include files depth-first, because the dependencies |
992 | in the psymtabs are way too inclusive (it seems to me). Figure out what | |
993 | really depends on what, to avoid recursing 20 or 30 times while reading | |
994 | real symtabs. | |
995 | ||
7ae38352 AC |
996 | -- |
997 | ||
c906108c SS |
998 | value_add() should be subtracting the lower bound of arrays, if known, |
999 | and possibly checking against the upper bound for error reporting. | |
1000 | ||
7ae38352 | 1001 | -- |
c906108c SS |
1002 | |
1003 | When listing source lines, check for a preceding \n, to verify that | |
1004 | the file hasn't changed out from under us. | |
1005 | ||
7ae38352 AC |
1006 | [fixed by some other means I think. That hack wouldn't actually work |
1007 | reliably - the file might move such that another \n appears. ] | |
c906108c | 1008 | |
7ae38352 | 1009 | -- |
c906108c SS |
1010 | |
1011 | Get all the remote systems (where the protocol allows it) to be able to | |
1012 | stop the remote system when the GDB user types ^C (like remote.c | |
1013 | does). For ebmon, use ^Ak. | |
1014 | ||
7ae38352 AC |
1015 | -- |
1016 | ||
c906108c SS |
1017 | Possible feature: A version of the "disassemble" command which shows |
1018 | both source and assembly code ("set symbol-filename on" is a partial | |
1019 | solution). | |
1020 | ||
7ae38352 AC |
1021 | [has this been done? It was certainly done for MI and GDBtk] |
1022 | ||
1023 | -- | |
1024 | ||
c906108c SS |
1025 | investigate "x/s 0" (right now stops early) (I think maybe GDB is |
1026 | using a 0 address for bad purposes internally). | |
1027 | ||
7ae38352 AC |
1028 | -- |
1029 | ||
c906108c SS |
1030 | Make "info path" and path_command work again (but independent of the |
1031 | environment either of gdb or that we'll pass to the inferior). | |
1032 | ||
7ae38352 AC |
1033 | -- |
1034 | ||
c906108c SS |
1035 | Make GDB understand the GCC feature for putting octal constants in |
1036 | enums. Make it so overflow on an enum constant does not error_type | |
1037 | the whole type. Allow arbitrarily large enums with type attributes. | |
1038 | Put all this stuff in the testsuite. | |
1039 | ||
7ae38352 AC |
1040 | -- |
1041 | ||
c906108c SS |
1042 | Make TYPE_CODE_ERROR with a non-zero TYPE_LENGTH more useful (print |
1043 | the value in hex; process type attributes). Add this to the | |
1044 | testsuite. This way future compilers can add new types and old | |
1045 | versions of GDB can do something halfway reasonable. | |
1046 | ||
7ae38352 | 1047 | -- |
c906108c SS |
1048 | |
1049 | Fix mdebugread.c:parse_type to do fundamental types right (see | |
1050 | rs6000_builtin_type in stabsread.c for what "right" is--the point is | |
1051 | that the debug format fixes the sizes of these things and it shouldn't | |
1052 | depend on stuff like TARGET_PTR_BIT and so on. For mdebug, there seem | |
1053 | to be separate bt* codes for 64 bit and 32 bit things, and GDB should | |
1054 | be aware of that). Also use a switch statement for clarity and speed. | |
1055 | ||
7ae38352 AC |
1056 | -- |
1057 | ||
c906108c SS |
1058 | Investigate adding symbols in target_load--some targets do, some |
1059 | don't. | |
1060 | ||
7ae38352 AC |
1061 | -- |
1062 | ||
c906108c SS |
1063 | Put dirname in psymtabs and change lookup*symtab to use dirname (so |
1064 | /foo/bar.c works whether compiled by cc /foo/bar.c, or cd /foo; cc | |
1065 | bar.c). | |
1066 | ||
7ae38352 AC |
1067 | -- |
1068 | ||
c906108c SS |
1069 | Merge xcoffread.c and coffread.c. Use breakpoint_re_set instead of |
1070 | fixup_breakpoints. | |
1071 | ||
7ae38352 | 1072 | -- |
c906108c SS |
1073 | |
1074 | Make a watchpoint which contains a function call an error (it is | |
1075 | broken now, making it work is probably not worth the effort). | |
1076 | ||
7ae38352 | 1077 | -- |
c906108c SS |
1078 | |
1079 | New test case based on weird.exp but in which type numbers are not | |
1080 | renumbered (thus multiply defining a type). This currently causes an | |
1081 | infinite loop on "p v_comb". | |
1082 | ||
7ae38352 | 1083 | -- |
c906108c | 1084 | |
7ae38352 | 1085 | [Hey! Hint Hint Delete Delete!!!] |
c906108c SS |
1086 | |
1087 | Fix 386 floating point so that floating point registers are real | |
1088 | registers (but code can deal at run-time if they are missing, like | |
1089 | mips and 68k). This would clean up "info float" and related stuff. | |
1090 | ||
7ae38352 | 1091 | -- |
c906108c SS |
1092 | |
1093 | gcc -g -c enummask.c then gdb enummask.o, then "p v". GDB complains | |
1094 | about not being able to access memory location 0. | |
1095 | ||
1096 | -------------------- enummask.c | |
1097 | enum mask | |
1098 | { | |
1099 | ANIMAL = 0, | |
1100 | VEGETABLE = 1, | |
1101 | MINERAL = 2, | |
1102 | BASIC_CATEGORY = 3, | |
1103 | ||
1104 | WHITE = 0, | |
1105 | BLUE = 4, | |
1106 | GREEN = 8, | |
1107 | BLACK = 0xc, | |
1108 | COLOR = 0xc, | |
1109 | ||
1110 | ALIVE = 0x10, | |
1111 | ||
1112 | LARGE = 0x20 | |
1113 | } v; | |
1114 | ||
7ae38352 AC |
1115 | -- |
1116 | ||
c906108c SS |
1117 | If try to modify value in file with "set write off" should give |
1118 | appropriate error not "cannot access memory at address 0x65e0". | |
1119 | ||
7ae38352 | 1120 | -- |
c906108c | 1121 | |
c906108c SS |
1122 | Allow core file without exec file on RS/6000. |
1123 | ||
7ae38352 AC |
1124 | -- |
1125 | ||
c906108c SS |
1126 | Make sure "shell" with no arguments works right on DOS. |
1127 | ||
7ae38352 AC |
1128 | -- |
1129 | ||
c906108c SS |
1130 | Make gdb.ini (as well as .gdbinit) be checked on all platforms, so |
1131 | the same directory can be NFS-mounted on unix or DOS, and work the | |
1132 | same way. | |
1133 | ||
7ae38352 AC |
1134 | -- |
1135 | ||
1136 | [Is this another delete???] | |
c906108c SS |
1137 | |
1138 | Get SECT_OFF_TEXT stuff out of objfile_relocate (might be needed to | |
1139 | get RS/6000 to work right, might not be immediately relevant). | |
1140 | ||
7ae38352 | 1141 | -- |
c906108c SS |
1142 | |
1143 | Work out some kind of way to allow running the inferior to be done as | |
1144 | a sub-execution of, eg. breakpoint command lists. Currently running | |
1145 | the inferior interupts any command list execution. This would require | |
1146 | some rewriting of wait_for_inferior & friends, and hence should | |
1147 | probably be done in concert with the above. | |
1148 | ||
7ae38352 AC |
1149 | -- |
1150 | ||
c906108c SS |
1151 | Add function arguments to gdb user defined functions. |
1152 | ||
7ae38352 AC |
1153 | -- |
1154 | ||
c906108c SS |
1155 | Add convenience variables that refer to exec file, symbol file, |
1156 | selected frame source file, selected frame function, selected frame | |
1157 | line number, etc. | |
1158 | ||
7ae38352 AC |
1159 | -- |
1160 | ||
c906108c SS |
1161 | Add a "suspend" subcommand of the "continue" command to suspend gdb |
1162 | while continuing execution of the subprocess. Useful when you are | |
1163 | debugging servers and you want to dodge out and initiate a connection | |
1164 | to a server running under gdb. | |
1165 | ||
7ae38352 | 1166 | [hey async!!] |
c906108c | 1167 | |
7ae38352 | 1168 | -- |
c906108c SS |
1169 | |
1170 | Modify the handling of symbols grouped through BINCL/EINCL stabs to | |
1171 | allocate a partial symtab for each BINCL/EINCL grouping. This will | |
1172 | seriously decrease the size of inter-psymtab dependencies and hence | |
1173 | lessen the amount that needs to be read in when a new source file is | |
1174 | accessed. | |
1175 | ||
7ae38352 | 1176 | -- |
c906108c | 1177 | |
7ae38352 | 1178 | [Comming...] |
c906108c SS |
1179 | |
1180 | Modify gdb to work correctly with Pascal. | |
1181 | ||
7ae38352 AC |
1182 | -- |
1183 | ||
c906108c SS |
1184 | Add a command for searching memory, a la adb. It specifies size, |
1185 | mask, value, start address. ADB searches until it finds it or hits | |
1186 | an error (or is interrupted). | |
1187 | ||
7ae38352 AC |
1188 | -- |
1189 | ||
b83266a0 SS |
1190 | Remove the range and type checking code and documentation, if not |
1191 | going to implement. | |
1192 | ||
c906108c SS |
1193 | # Local Variables: |
1194 | # mode: text | |
1195 | # End: |