1 If you find inaccuracies in this list, please send mail to
3 of these, you should consider sending mail to the same address, to
4 find out whether anyone else is working on it.
10 Below is a list of problems identified during the GDB 5.0 release
11 cycle. People hope to have these problems fixed in 5.1.
15 Wow, three bug reports for the same problem in one day! We should
16 probably make fixing this a real priority :-).
18 Anyway, thanks for reporting.
20 The following patch will fix the problems with setting breakpoints in
21 dynamically loaded objects:
23 http://sources.redhat.com/ml/gdb-patches/2000-05/msg00230.html
25 This patch isn't checked in yet (ping Michael/JimB), but I hope this
26 will be in the next GDB release.
28 There should really be a test in the testsuite for this problem, since
29 it keeps coming up :-(. Any volunteers?
35 GDB 5.1 - New features
36 ======================
38 The following new features should be included in 5.1.
45 The following code cleanups will hopefully be applied to GDB 5.1.
49 GDB 5.1 - Known Problems
50 ========================
56 The z8k has suffered bit rot and is known to not build. The problem
57 was occuring in the opcodes directory.
61 The BFD directory requires bug-fixed AUTOMAKE et.al.
63 AUTOMAKE 1.4 incorrectly set the TEXINPUTS environment variable. It
64 contained the full path to texinfo.tex when it should have only
65 contained the directory. The bug has been fixed in the current
66 AUTOMAKE sources. Automake snapshots can be found in:
67 ftp://sources.redhat.com/pub/gdb/infrastructure
68 and ftp://sources.redhat.com/pub/binutils
72 Solaris 8 x86 CURSES_H problem
73 http://sources.redhat.com/ml/gdb/2000-07/msg00038.html
75 The original problem was worked around with:
79 * configure.in: Enable autoconf to find curses.h on Solaris 2.8.
80 * configure: Regenerate.
82 When building both GDB and SID using the same source tree the problem
83 will still occure. sid/component/configure.in mis-configures
84 <curses.h> and leaves wrong information in the config cache.
93 GDB 5.2 - New features
94 ======================
98 GCC 3.0 ABI support (but hopefully sooner...).
102 Objective C/C++ support (but hopefully sooner...).
106 Import of readline 4.2
113 The following cleanups have been identified as part of GDB 5.2.
119 Eliminate warnings for all targets on at least one host for one of the
120 -W flags. Flags up for debate include: -Wswitch -Wcomment -trigraphs
121 -Wtrigraphs -Wunused-function -Wunused-label -Wunused-variable
122 -Wunused-value -Wchar-subscripts -Wtraditional -Wshadow -Wcast-qual
123 -Wcast-align -Wwrite-strings -Wconversion -Wstrict-prototypes
124 -Wmissing-prototypes -Wmissing-declarations -Wredundant-decls
125 -Woverloaded-virtual -Winline
129 Deprecate, if not delete, the following:
134 Replaced by, on the target side
136 and on core-gdb side:
137 {read,write}_register_gen()
138 Remote.c will need to use something
139 other than REGISTER_BYTE() and
140 REGISTER_RAW_SIZE() when unpacking
143 STORE_PSEUDO_REGISTER
144 FETCH_PSEUDO_REGISTER
145 Now handed by the methods
146 gdbarch_{read,write}_register()
147 which sits between core GDB and
151 REGISTER_CONVERT_TO_RAW
152 REGISTER_CONVERT_TO_VIRTUAL
153 I think these three are redundant.
154 gdbarch_register_{read,write} can
155 do any conversion it likes.
157 REGISTER_VIRTUAL_SIZE
158 MAX_REGISTER_VIRTUAL_SIZE
159 REGISTER_VIRTUAL_TYPE
160 I think these can be replaced by
162 FRAME_REGISTER_TYPE(frame, regnum)
163 REGISTER_TYPE(regnum)
167 FRAME_REGISTER_INFO (frame, ...)
170 If nothing else rename this so that
171 how it relates to rawreg and the
175 The size of the cache can be computed
178 IS_TRAPPED_INTERNALVAR
179 The pseudo registers should eventually make
184 Obsolete the targets:
192 Obsolete the protocols:
196 ``As of version 5.3, WindRiver has removed the RDB server (RDB
197 protocol support is built into gdb).'' -- Till.
201 Restructure gdb directory tree so that it avoids any 8.3 and 14
206 Convert GDB build process to AUTOMAKE.
208 See also sub-directory configure below.
210 The current convention is (kind of) to use $(<header>_h) in all
211 dependency lists. It isn't done in a consistent way.
215 GDB 5.2 - Known Problems
216 ========================
220 Code Cleanups: General
221 ======================
223 The following are more general cleanups and fixes. They are not tied
224 to any specific release.
227 New Features and Fixes
228 ======================
230 These are harder than cleanups but easier than work involving
231 fundamental architectural change.
238 New languages come onto the scene all the time.
242 Re: Various C++ things
244 RTTI for g++ should be using the typeinfo functions rather than the
245 vtables. The typeinfo functions are always at offset 4 from the
246 beginning of the vtable, and are always right. The vtables will have
247 weird names like E::VB sometimes. The typeinfo function will always
248 be "E type_info function", or somesuch.
250 value_virtual_fn_field needs to be fixed so there are no failures for
251 virtual functions for C++ using g++.
253 Testsuite cases are the major priority right now for C++ support,
254 since i have to make a lot of changes that could potentially break
265 Investiagate ways of reducing memory.
269 Investigate ways of improving load time.
276 There are never to many testcases.
280 Better thread testsuite.
284 Better C++ testsuite.
288 Architectural Changes: General
289 ==============================
291 These are harder than simple cleanups / fixes and, consequently
292 involve more work. Typically an Architectural Change will be broken
293 down into a more digestible set of cleanups and fixes.
297 Architectural Change: Multi-arch et al.
298 =======================================
300 The long term objective is to remove all assumptions that there is a
301 single target with a single address space with a single instruction
302 set architecture and single application binary interface.
304 This is an ongoing effort. The first milestone is to enable
305 ``multi-arch'' where by all architectural decisions are made at
308 It should be noted that ``gdbarch'' is really ``gdbabi'' and
309 ``gdbisa''. Once things are multi-arched breaking that down correctly
310 will become much easier.
314 Architectural Change: MI, LIBGDB and scripting languages
315 ========================================================
317 See also architectural changes related to the event loop. LIBGDB
318 can't be finished until there is a generic event loop being used by
321 The long term objective is it to be possible to integrate GDB into
326 Architectural Change: Async
327 ===========================
329 While GDB uses an event loop when prompting the user for input. That
330 event loop is not exploited by targets when they allow the target
331 program to continue. Typically targets still block in (target_wait())
332 until the program again halts.
334 The closest a target comes to supporting full asynchronous mode are
335 the remote targets ``async'' and ``extended-async''.