1 If you find inaccuracies in this list, please send mail to
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.
7 Known problems in GDB 5.0
8 =========================
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
14 (The names in paren indicate people that posted the original problem.)
18 GDB doesn't build under IRIX6.4
22 Has anyone successfully built the latest (from cvs) gdb on IRIX6.4 or
23 later? The first problem I hit is that proc-api.c includes
24 sys/user.h, which no longer exists under IRIX6.4. If I comment out
25 that include, the next problem I hit is that PIOCGETPR and PIOCGETU
26 are no longer defined in IRIX6.4 (presumably related to the
27 disappearance of user.h).
31 The BFD directory requires bug-fixed AUTOMAKE et.al.
33 AUTOMAKE 1.4 incorrectly set the TEXINPUTS environment variable. It
34 contained the full path to texinfo.tex when it should have only
35 contained the directory. The bug has been fixed in the current
36 AUTOMAKE sources. Automake snapshots can be found in:
37 ftp://sourceware.cygnus.com/pub/gdb/snapshots
38 and ftp://sourceware.cygnus.com/pub/binutils
42 gdb-cvs fails to build on freebsd-elf
43 http://sourceware.cygnus.com/ml/gdb/2000-04/msg00004.html
45 Either the FreeBSD group need to contribute their local GDB changes
46 back to the master sources or someone needs to provides a new
47 (clean-room) implementation. Since the former involves a fairly
48 complicated assignment the latter may be easier. [cagney]
52 Generic: lin-thread cannot handle thread exit (Mark Kettenis, Michael
53 Snyder) http://sourceware.cygnus.com/ml/gdb/2000-q1/msg00525.html
55 The thread_db assisted debugging code doesn't handle exiting threads
56 properly, at least in combination with glibc 2.1.3 (the framework is
57 there, just not the actual code). There are at least two problems
58 that prevent this from working.
60 As an additional reference point, the pre thread_db code did not work
65 Java (Anthony Green, David Taylor)
67 Anthony Green has a number of Java patches that did not make it into
71 http://sourceware.cygnus.com/ml/gdb-patches/2000-q1/msg00512.html
74 http://sourceware.cygnus.com/ml/gdb-patches/2000-q1/msg00515.html
76 Patch: handle N_MAIN stab
77 http://sourceware.cygnus.com/ml/gdb-patches/2000-q1/msg00527.html
81 Pascal (Pierre Muller, David Taylor)
83 Pierre Muller has contributed patches for adding Pascal Language
86 2 pascal language patches inserted in database
87 http://sourceware.cygnus.com/ml/gdb/2000-q1/msg00521.html
90 http://sourceware.cygnus.com/ml/gdb/2000-q1/msg00496.html
94 GNU/Linux/x86 and random thread signals (and Solaris/SPARC but not
96 http://sourceware.cygnus.com/ml/gdb/2000-q1/msg00336.html
98 Christopher Blizzard writes:
100 So, I've done some more digging into this and it looks like Jim
101 Kingdon has reported this problem in the past:
103 http://sourceware.cygnus.com/ml/bug-gdb/1999-10/msg00058.html
105 I can reproduce this problem both with and without Tom's patch. Has
106 anyone seen this before? Maybe have a solution for it hanging around?
109 There's a test case for this documented at:
111 when debugging threaded applications you get extra SIGTRAPs
112 http://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=9565
114 [There should be a GDB testcase - cagney]
118 Possible regressions with some devel GCCs.
119 http://sourceware.cygnus.com/ml/gdb/2000-q1/msg00475.html
121 gcc-2.95.2 outputs a line note *before* the prologue (and one for the
122 closing brace after the epilogue, instead of before it, as it used to
123 be). By disabling the RTL-style prologue generating mechanism
124 (undocumented GCC option -mno-schedule-prologue), you get back the
125 traditional behaviour.
126 http://sourceware.cygnus.com/ml/gdb/2000-q1/msg00510.html
128 This should now be fixed.
132 RFD: infrun.c: No bpstat_stop_status call after proceed over break?
134 http://sourceware.cygnus.com/ml/gdb-patches/2000-q1/msg00665.html
136 GDB misses watchpoint triggers after proceeding over a breakpoint on
141 x86 linux GDB and SIGALRM (???)
142 http://sourceware.cygnus.com/ml/gdb/2000-q1/msg00803.html
144 I know there are problems with single stepping through signal
145 handlers. These problems were present in 4.18. They were just masked
146 because 4.18 failed to recognize signal handlers. Fixing it is not
147 easy, and will require changes to handle_inferior_event(), that I
148 prefer not to make before the 5.0 release.
154 Revised UDP support (was: Re: [Fwd: [patch] UDP transport support])
155 http://sourceware.cygnus.com/ml/gdb-patches/2000-04/msg00000.html
157 (Broken) support for GDB's remote protocol across UDP is to be
158 included in the follow-on release.
162 Can't build IRIX -> arm GDB.
163 http://sourceware.cygnus.com/ml/gdb-patches/2000-04/msg00356.html
166 > Now I'm building for an embedded arm target. If there is a way of turning
167 > remote-rdi off, I couldn't find it. It looks like it gets built by default
168 > in gdb/configure.tgt(line 58) Anyway, the build dies in
169 > gdb/rdi-share/unixcomm.c. SERPORT1 et. al. never get defined because we
170 > aren't one of the architectures supported.
174 Problem with weak functions
175 http://sourceware.cygnus.com/ml/gdb/2000-05/msg00060.html
177 Dan Nicolaescu writes:
178 > It seems that gdb-4.95.1 does not display correctly the function when
179 > stoping in weak functions.
181 > It stops in a function that is defined as weak, not in the function
182 > that is actualy run...
186 GDB5 TOT on unixware 7
187 http://sourceware.cygnus.com/ml/gdb/2000-04/msg00119.html
190 > I just spun the top of tree of the GDB5 branch on UnixWare 7. As a
191 > practical matter, the current thread support is somewhat more annoying
192 > than when GDB was thread-unaware.
199 The following code cleanups are planned for the follow-on release to
206 The need for this as almost been eliminated. The next version of GCC
207 (assuming cagney gets the relevant patch committed) will be able to
208 supress unused parameter warnings.
212 Delete macro TARGET_BYTE_ORDER_SELECTABLE.
214 Patches in the database.
220 Readline 4.? is out. A merge wouldn't hurt.
226 Eliminate all uses of PARAMS in GDB's source code.
230 Elimination of make_cleanup_func. (Andrew Cagney)
232 make_cleanup_func elimination
233 http://sourceware.cygnus.com/ml/gdb-patches/2000-q1/msg00791.html
234 http://sourceware.cygnus.com/ml/gdb-patches/2000-q1/msg00814.html
238 ChangeLog.mi vs ChangeLog-mi (Andrew Cagney)
239 Needs further debate.
241 Re: [PATCH] Add change-log variables to more MI files
242 http://sourceware.cygnus.com/ml/gdb-patches/2000-q1/msg00811.html
246 Re: Various C++ things
248 value_headof/value_from_vtable_info are worthless, and should be removed.
249 The one place in printcmd.c that uses it should use the RTTI functions.
251 RTTI for g++ should be using the typeinfo functions rather than the vtables.
252 The typeinfo functions are always at offset 4 from the beginning of the vtable,
253 and are always right. The vtables will have weird names like E::VB sometimes.
254 The typeinfo function will always be "E type_info function", or somesuch.
256 value_virtual_fn_field needs to be fixed so there are no failures for virtual
257 functions for C++ using g++.
259 Testsuite cases are the major priority right now for C++ support, since i have
260 to make a lot of changes that could potentially break each other.
264 GDBARCH cleanup (Andrew Cagney)
266 The non-generated parts of gdbarch.{sh,h,c} should be separated out
267 into arch-utils.[hc].
269 The ``info architecture'' command should be replaced with a fixed
270 ``set architecture'' (implemented using the command.c enum code).
272 Document that gdbarch_init_ftype could easily fail because it didn't
273 identify an architecture.
277 Migrate qfThreadInfo packet -> qThreadInfo. (Andrew Cagney)
279 Add support for packet enable/disable commands with these thread
280 packets. General cleanup.
282 [PATCH] Document the ThreadInfo remote protocol queries
283 http://sourceware.cygnus.com/ml/gdb-patches/2000-q1/msg00832.html
285 [PATCH] "info threads" queries for remote.c
286 http://sourceware.cygnus.com/ml/gdb-patches/2000-q1/msg00831.html
290 Eliminate gdb/tui/Makefile.in.
291 Cleanup configury support for optional sub-directories.
293 Check how GCC handles multiple front ends for an example of how things
294 could work. A tentative first step is to rationalize things so that
295 all sub directories are handled in a fashion similar to gdb/mi.
299 [PATCH/5] src/intl/Makefile.in:distclean additions
300 http://sourceware.cygnus.com/ml/gdb-patches/2000-04/msg00363.html
302 Do not forget to merge the patch back into the trunk.
306 Re: [RFC] Change configure.in so -W arnings match reality
307 http://sourceware.cygnus.com/ml/gdb-patches/2000-04/msg00350.html
309 Some GCC compilers do not like -Wreturn-type. (Going forward there
310 may be more problems like that). Need to check which of the warning
311 options are valid. Need to probably disable warnings by default.
320 Register Cache Cleanup (below from Andrew Cagney)
322 I would depict the current register architecture as something like:
329 register + REGISTER_BYTE(reg_nr)
332 -------------------------
333 | extern register[] |
334 -------------------------
336 where neither the high (valops.c et.al.) or low gdb (*-tdep.c) are
337 really clear on what mechanisms they should be using to manipulate that
338 buffer. Further, much code assumes, dangerously, that registers are
339 contigious. Having got mips-tdep.c to support multiple ABIs, believe
340 me, that is a bad assumption. Finally, that register cache layout is
341 determined by the current remote/local target and _not_ the less
342 specific target ISA. In fact, in many cases it is determined by the
343 somewhat arbitrary layout of the [gG] packets!
346 How I would like the register file to work is more like:
374 The main objectives being:
376 o a clear separation between the low
377 level target and the high level GDB
379 o a mechanism that solves the general
380 problem of register aliases, overlaps
381 etc instead of treating them as optional
382 extras that can be wedged in as an after
383 thought (that is a reasonable description
384 of the current code).
386 Identify then solve the hard case and the
387 rest just falls out. GDB solved the easy
388 case and then tried to ignore the real
391 o a removal of the assumption that the
392 mapping between the register cache
393 and virtual registers is largely static.
394 If you flip the USR/SSR stack register
395 select bit in the status-register then
396 the corresponding stack registers should
399 o a mechanism that clearly separates the
400 gdb internal register cache from any
401 target (not architecture) dependant
402 specifics such as [gG] packets.
404 Of course, like anything, it sounds good in theory. In reality, it
405 would have to contend with many<->many relationships at both the
406 virt<->cache and cache<->target level. For instance:
409 Modifying an mmx register may involve
410 scattering values across both FP and
411 mmpx specific parts of a buffer
414 When writing back a SP it may need to
415 both be written to both SP and USP.
420 Rather than let this like the last time it was discussed, just slip, I'm
421 first going to add this e-mail (+ references) to TODO. I'd then like to
422 sketch out a broad strategy I think could get us there.
425 First thing I'd suggest is separating out the ``extern registers[]''
426 code so that we can at least identify what is using it. At present
427 things are scattered across many files. That way we can at least
428 pretend that there is a cache instead of a global array :-)
430 I'd then suggest someone putting up a proposal for the pseudo-reg /
431 high-level side interface so that code can be adopted to it. For old
432 code, initially a blanket rename of write_register_bytes() to
433 deprecated_write_register_bytes() would help.
435 Following that would, finaly be the corresponding changes to the target.
439 Check that GDB can handle all BFD architectures (Andrew Cagney)
441 There should be a test that checks that BFD/GDB are in sync with
442 regard to architecture changes. Something like a test that first
443 queries GDB for all supported architectures and then feeds each back
444 to GDB.. Anyone interested in learning how to write tests? :-)
448 Add support for Modula3
450 Get DEC/Compaq to contribute their Modula-3 support.
457 This list is not up to date, and opinions vary about the importance or
458 even desirability of some of the items. If you do fix something, it
459 always pays to check the below.
463 Document trace machinery.
465 Document overlay machinery.
467 Extend .gdbinit mechanism to specify name on command line, allow for
468 lists of files to load, include function of --tclcommand.
470 @c This does not work (yet if ever). FIXME.
471 @c @item --parse=@var{lang} @dots{}
472 @c Configure the @value{GDBN} expression parser to parse the listed languages.
473 @c @samp{all} configures @value{GDBN} for all supported languages. To get a
474 @c list of all supported languages, omit the argument. Without this
475 @c option, @value{GDBN} is configured to parse all supported languages.
477 Add an "info bfd" command that displays supported object formats,
478 similarly to objdump -i.
480 START_INFERIOR_TRAPS_EXPECTED need never be defined to 2, since that
481 is its default value. Clean this up.
483 It should be possible to use symbols from shared libraries before we know
484 exactly where the libraries will be loaded. E.g. "b perror" before running
485 the program. This could maybe be done as an extension of the "breakpoint
486 re-evaluation" after new symbols are loaded.
488 Make single_step() insert and remove breakpoints in one operation.
490 Speed up single stepping by avoiding extraneous ptrace calls.
492 Speed up single stepping by not inserting and removing breakpoints
493 each time the inferior starts and stops.
495 Breakpoints should not be inserted and deleted all the time. Only the
496 one(s) there should be removed when we have to step over one. Support
497 breakpoints that don't have to be removed to step over them.
499 Update gdbint.texinfo to include doc on the directory structure and
500 the various tricks of building gdb.
502 Do a tutorial in gdb.texinfo on how to do simple things in gdb.
503 E.g. how to set a breakpoint that just prints something and continues.
504 How to break on aborts. Etc.
506 Provide "voodoo" debugging of core files. This creates a zombie
507 process as a child of the debugger, and loads it up with the data,
508 stack, and regs of the core file. This allows you to call functions
509 in the executable, to manipulate the data in the core file.
511 GDB reopens the source file on every line, as you "next" through it.
513 Referencing the vtbl member of a struct doesn't work. It prints OK
514 if you print the struct, but it gets 0 if you try to deref it.
516 Persistent command history: A feature where you could save off a list
517 of the commands you did, so you can edit it into something that will bring
518 the target to the same place every time you source it.
519 This would also be useful for automated fast watchpointing; if you go
520 past the place where it watchpoints, you just start it over again and
521 do it more carefully.
523 Deal with the SunOS 4.0 and 4.1.1 ptrace bug that loses the registers if
524 the stack is paged out.
526 Finish the C++ exception handling stub routines. Lint points them out
527 as unused statics functions.
529 Perhaps "i source" should take an argument like that of "list".
531 See if core-aout.c's fetch_core_registers can be used on more machines.
532 E.g. MIPS (mips-xdep.c).
534 unpack_double() does not handle IEEE float on the target unless the host
535 is also IEEE. Death on a vax.
537 Set up interface between GDB and INFO so that you can hop into interactive
538 INFO and back out again. When running under Emacs, should use Emacs
539 info, else fork the info program. Installation of GDB should install
540 its texinfo files into the info tree automagically, including the readline
543 "help address" ought to find the "help set print address" entry.
545 Remove the VTBL internal guts from printouts of C++ structs, unless
548 Remove "at 0xnnnn" from the "b foo" response, if `print address off' and if
549 it matches the source line indicated.
551 The prompt at end of screen should accept space as well as CR.
553 Check STORE_RETURN_VALUE on all architectures. Check near it in tm-sparc.h
554 for other bogosities.
556 Check for storage leaks in GDB, I'm sure there are a lot!
558 vtblprint of a vtbl should demangle the names it's printing.
560 Backtrace should point out what the currently selected frame is, in
561 its display, perhaps showing "@3 foo (bar, ...)" or ">3 foo (bar,
562 ...)" rather than "#3 foo (bar, ...)".
564 "i program" should work for core files, and display more info, like what
565 actually caused it to die.
567 "x/10i" should shorten the long name, if any, on subsequent lines.
569 Check through the code for FIXME comments and fix them. dbxread.c,
570 blockframe.c, and plenty more. (I count 634 as of 940621 - sts)
572 "next" over a function that longjumps, never stops until next time you happen
573 to get to that spot by accident. E.g. "n" over execute_command which has
576 "set zeroprint off", don't bother printing members of structs which
577 are entirely zero. Useful for those big structs with few useful
580 GDB does four ioctl's for every command, probably switching terminal modes
581 to/from inferior or for readline or something.
583 terminal_ours versus terminal_inferior: cache state. Switch should be a noop
584 if the state is the same, too.
588 Clean up invalid_float handling so gdb doesn't coredump when it tries to
589 access a NaN. While this might work on SPARC, other machines are not
592 "b value_at ; commands ; continue ; end" stops EVERY OTHER TIME!
593 Then once you enter a command, it does the command, runs two more
594 times, and then stops again! Bizarre... (This behaviour has been
595 modified, but it is not yet 100% predictable when e.g. the commands
596 call functions in the child, and while there, the child is interrupted
597 with a signal, or hits a breakpoint.)
599 help completion, help history should work.
601 Check that we can handle stack trace through varargs AND alloca in same
604 wait_for_inferior loops forever if wait() gives it an error.
606 "i frame" shows wrong "arglist at" location, doesn't show where the args
607 should be found, only their actual values.
609 There should be a way for "set" commands to validate the new setting
610 before it takes effect.
612 A mess of floating point opcodes are missing from sparc-opcode.h.
613 Also, a little program should test the table for bits that are
614 overspecified or underspecified. E.g. if the must-be-ones bits
615 and the must-be-zeroes bits leave some fields unexamined, and the format
616 string leaves them unprinted, then point this out. If multiple
617 non-alias patterns match, point this out too. Finally, there should
618 be a sparc-optest.s file that tries each pattern out. This file
619 should end up coming back the same (modulo transformation comments)
620 if fed to "gas" then the .o is fed to gdb for disassembly.
622 Eliminate all the core_file_command's in all the xdep files.
623 Eliminate separate declarations of registers[] everywhere.
625 "ena d" is ambiguous, why? "ena delete" seems to think it is a command!
627 Perhaps move the tdep, xdep, and nat files, into the config
628 subdirectories. If not, at least straighten out their names so that
629 they all start with the machine name.
631 inferior_status should include stop_print_frame. It won't need to be
632 reset in wait_for_inferior after bpstat_stop_status call, then.
634 i line VAR produces "Line number not known for symbol ``var''.". I
635 thought we were stashing that info now!
637 We should be able to write to random files at hex offsets like adb.
639 Make "target xxx" command interruptible.
641 Handle add_file with separate text, data, and bss addresses. Maybe
642 handle separate addresses for each segment in the object file?
644 Handle free_named_symtab to cope with multiply-loaded object files
645 in a dynamic linking environment. Should remember the last copy loaded,
646 but not get too snowed if it finds references to the older copy.
648 Generalize and Standardize the RPC interface to a target program,
649 improve it beyond the "ptrace" interface, and see if it can become a
650 standard for remote debugging. (This is talking about the vxworks
651 interface. Seems unlikely to me that there will be "a standard" for
652 remote debugging anytime soon --kingdon, 8 Nov 1994).
654 Remove all references to:
661 now that we have BFD. All remaining are in machine dependent files.
663 When quitting with a running program, if a core file was previously
664 examined, you get "Couldn't read float regs from core file"...if
665 indeed it can't. generic_mourn_inferior...
667 Have remote targets give a warning on a signal argument to
668 target_resume. Or better yet, extend the protocols so that it works
669 like it does on the Unix-like systems.
671 Sort help and info output.
673 Re-organize help categories into things that tend to fit on a screen
676 renote-nindy.c handles interrupts poorly; it error()s out of badly
677 chosen places, e.g. leaving current_frame zero, which causes core dumps
680 Add in commands like ADB's for searching for patterns, etc. We should
681 be able to examine and patch raw unsymboled binaries as well in gdb as
682 we can in adb. (E.g. increase the timeout in /bin/login without source).
684 Those xdep files that call register_addr without defining it are
685 probably simply broken. When reconfiguring this part of gdb, I could
686 only make guesses about how to redo some of those files, and I
687 probably guessed wrong, or left them "for later" when I have a
688 machine that can attempt to build them.
690 When doing "step" or "next", if a few lines of source are skipped between
691 the previous line and the current one, print those lines, not just the
692 last line of a multiline statement.
694 When searching for C++ superclasses in value_cast in valops.c, we must
695 not search the "fields", only the "superclasses". There might be a
696 struct with a field name that matches the superclass name. This can
697 happen when the struct was defined before the superclass (before the
698 name became a typedef).
700 Handling of "&" address-of operator needs some serious overhaul
701 for ANSI C and consistency on arrays and functions.
702 For "float point[15];":
703 ptype &point[4] ==> Attempt to take address of non-lvalue.
704 For "char *malloc();":
705 ptype malloc ==> "char *()"; should be same as
706 ptype &malloc ==> "char *(*)()"
707 call printf ("%x\n", malloc) ==> weird value, should be same as
708 call printf ("%x\n", &malloc) ==> correct value
710 Fix dbxread.c symbol reading in the presence of interrupts. It
711 currently leaves a cleanup to blow away the entire symbol table when a
712 QUIT occurs. (What's wrong with that? -kingdon, 28 Oct 1993).
714 Mipsread.c reads include files depth-first, because the dependencies
715 in the psymtabs are way too inclusive (it seems to me). Figure out what
716 really depends on what, to avoid recursing 20 or 30 times while reading
719 value_add() should be subtracting the lower bound of arrays, if known,
720 and possibly checking against the upper bound for error reporting.
722 mipsread.c symbol table allocation and deallocation should be checked.
723 My suspicion is that it's full of memory leaks.
725 SunOS should have a target_lookup_symbol() for common'd things allocated
726 by the shared library linker ld.so.
728 When listing source lines, check for a preceding \n, to verify that
729 the file hasn't changed out from under us.
731 When listing source lines, eat leading whitespace corresponding to the
732 line-number prefix we print. This avoids long lines wrapping.
734 mipsread.c needs to check for old symtabs and psymtabs for the same
735 files, the way it happens for dbxread.c and coffread.c, for VxWorks
736 incremental symbol table reloading.
738 Get all the remote systems (where the protocol allows it) to be able to
739 stop the remote system when the GDB user types ^C (like remote.c
740 does). For ebmon, use ^Ak.
742 Possible feature: A version of the "disassemble" command which shows
743 both source and assembly code ("set symbol-filename on" is a partial
746 investigate "x/s 0" (right now stops early) (I think maybe GDB is
747 using a 0 address for bad purposes internally).
749 Make "info path" and path_command work again (but independent of the
750 environment either of gdb or that we'll pass to the inferior).
752 Make GDB understand the GCC feature for putting octal constants in
753 enums. Make it so overflow on an enum constant does not error_type
754 the whole type. Allow arbitrarily large enums with type attributes.
755 Put all this stuff in the testsuite.
757 Make TYPE_CODE_ERROR with a non-zero TYPE_LENGTH more useful (print
758 the value in hex; process type attributes). Add this to the
759 testsuite. This way future compilers can add new types and old
760 versions of GDB can do something halfway reasonable.
762 Clean up formatting of "info registers" on MIPS and 88k. See if it
763 is possible to do this generically across all target architectures.
765 GDB gets bfd/corefile.c and gdb/corefile.c confused (this should be easy to
766 repeat even with something more recent than GDB 4.9).
768 Check that unmatched RBRAC doesn't abort().
770 Fix mdebugread.c:parse_type to do fundamental types right (see
771 rs6000_builtin_type in stabsread.c for what "right" is--the point is
772 that the debug format fixes the sizes of these things and it shouldn't
773 depend on stuff like TARGET_PTR_BIT and so on. For mdebug, there seem
774 to be separate bt* codes for 64 bit and 32 bit things, and GDB should
775 be aware of that). Also use a switch statement for clarity and speed.
777 Investigate adding symbols in target_load--some targets do, some
780 Put dirname in psymtabs and change lookup*symtab to use dirname (so
781 /foo/bar.c works whether compiled by cc /foo/bar.c, or cd /foo; cc
784 Merge xcoffread.c and coffread.c. Use breakpoint_re_set instead of
787 Fix byte order and int size sins in tm-a29k.h
788 (EXTRACT_RETURN_VALUE). Perhaps should reproduce bug and verify fix
789 (or perhaps should just fix it...).
791 Make a watchpoint on a constant expression an error (or warning
794 Make a watchpoint which contains a function call an error (it is
795 broken now, making it work is probably not worth the effort).
797 Re-do calls to signal() in remote.c, and inflow.c (set_sigint_trap and
798 so on) to be independent of the debugging target, using target_stop to
799 stop the inferior. Probably the part which is now handled by
800 interrupt_query in remote.c can be done without any new features in
801 the debugging target.
803 New test case based on weird.exp but in which type numbers are not
804 renumbered (thus multiply defining a type). This currently causes an
805 infinite loop on "p v_comb".
811 "source file more recent" loses on re-read
813 Fix 386 floating point so that floating point registers are real
814 registers (but code can deal at run-time if they are missing, like
815 mips and 68k). This would clean up "info float" and related stuff.
817 Look at Solaris bug in interrupt.exp. Can get out of syscall with
818 PRSABORT (syscall will return EINTR) but merely doing that leads to a
819 "can't read memory" error.
821 gcc -g -c enummask.c then gdb enummask.o, then "p v". GDB complains
822 about not being able to access memory location 0.
824 -------------------- enummask.c
843 If try to modify value in file with "set write off" should give
844 appropriate error not "cannot access memory at address 0x65e0".
846 Why do we allow a target to omit standard register names (NO_STD_REGS
847 in tm-z8k.h)? I thought the standard register names were supposed to
848 be just that, standard.
850 Allow core file without exec file on RS/6000.
852 Make sure "shell" with no arguments works right on DOS.
854 Make gdb.ini (as well as .gdbinit) be checked on all platforms, so
855 the same directory can be NFS-mounted on unix or DOS, and work the
858 cd ~/tmp/<M-?> causes infinite loop (where ~/tmp is a directory).
860 Get SECT_OFF_TEXT stuff out of objfile_relocate (might be needed to
861 get RS/6000 to work right, might not be immediately relevant).
863 Clean up add_toc_to_loadinfo
865 Think about attached processes and sharing terminal.
867 John sez in reference to ignoring errors from tcsegpgrp if attach_flag:
868 set_tty_state should not have any trouble with attached processes.
869 Instead, the tty handling should leave the pgrp of the tty alone when
870 attaching to processes (perhaps pass terminal_init_inferior a flag
871 saying whether we're attaching).
873 PAGE_SIZE redefined warnings on AIX. Probably should be using
874 BFD_PAGE_SIZE throughout BFD.
876 Rewrite proceed, wait_for_inferior, and normal_stop to clean them up.
879 1) Make each test in wait_for_inferior a seperate subroutine
881 2) Combine wait_for_inferior and normal_stop to clean up
882 communication via global variables.
883 3) See if you can find some way to clean up the global
884 variables that are used; possibly group them by data flow
885 and information content?
887 Work out some kind of way to allow running the inferior to be done as
888 a sub-execution of, eg. breakpoint command lists. Currently running
889 the inferior interupts any command list execution. This would require
890 some rewriting of wait_for_inferior & friends, and hence should
891 probably be done in concert with the above.
893 Add function arguments to gdb user defined functions.
895 Add convenience variables that refer to exec file, symbol file,
896 selected frame source file, selected frame function, selected frame
899 Add a "suspend" subcommand of the "continue" command to suspend gdb
900 while continuing execution of the subprocess. Useful when you are
901 debugging servers and you want to dodge out and initiate a connection
902 to a server running under gdb.
904 Add stab information to allow reasonable debugging of inline functions
905 (possibly they should show up on a stack backtrace? With a note
906 indicating that they weren't "real"?).
908 Modify the naked "until" command to step until past the current source
909 line, rather than past the current pc value. This is tricky simply
910 because the low level routines have no way of specifying a multi-line
911 step range, and there is no way of saying "don't print stuff when we
912 stop" from above (otherwise could just call step many times).
914 Modify the handling of symbols grouped through BINCL/EINCL stabs to
915 allocate a partial symtab for each BINCL/EINCL grouping. This will
916 seriously decrease the size of inter-psymtab dependencies and hence
917 lessen the amount that needs to be read in when a new source file is
920 Do an "x/i $pc" after each stepi or nexti.
922 Modify all of the disassemblers to use printf_filtered to get correct
925 Modify gdb to work correctly with Pascal.
927 Add a command for searching memory, a la adb. It specifies size,
928 mask, value, start address. ADB searches until it finds it or hits
929 an error (or is interrupted).
931 Remove the range and type checking code and documentation, if not