3 Fix stack unwinding through _sigtramp on Irix. These patches are from
5 * mipsread.c (fixup_sigtramp): Find _sigtramp on Irix even when the
6 executable uses sigvec.
7 * mips-tdep.c (read_next_frame_reg): Allow tm-file to override
9 * config/mips/tm-irix3.h: Add sigcontext offsets for Irix.
13 * infrun.c (wait_for_inferior): Clear stop_signal if it should not
14 be passed to the inferior to make "handle <signal> nopass nostop" work.
18 * config/mips/tm-irix3.h: Clean up, use tm-bigmips.h and redefine
20 * findvar.c (value_from_register): Fix uninitialized first_addr
21 which caused problems with assignment of doubles to register variables
23 * mipsread.c: Remove TM_FILE_OVERRIDE, include tm.h and provide the
24 missing mips definitions if necessary.
26 Fix handling of double register variables for mips targets and big
28 * config/mips/tm-mips.h: Increase MAX_REGISTER_{RAW,VIRTUAL}_SIZE to
30 * config/mips/tm-mips.h (REGISTER_CONVERT_TO_TYPE): New macro for
31 conversion of type held in multiple registers to host format.
32 * config/mips/tm-mips.h (REGISTER_CONVERT_FROM_TYPE): New macro,
33 companion to REGISTER_CONVERT_TO_TYPE.
34 * config/mips/tm-mips.h (EXTRACT_RETURN_VALUE, STORE_RETURN_VALUE):
35 Convert to function calls.
36 * config/mips/tm-mips.h (FIX_CALL_DUMMY): New code for big endian
38 * mips-tdep.c (mips_print_register): Raw buffer now needs just
39 MAX_REGISTER_RAW_SIZE bytes.
40 * mips-tdep.c (mips_print_register): Use REGISTER_CONVERT_TO_TYPE
41 (if defined) for doubles.
42 * mips-tdep.c: (mips_extract_return_value, mips_store_return_value):
43 New functions, take care of REGISTER_CONVERT_TO/FROM_TYPE.
44 * valops.c (value_assign): Use REGISTER_CONVERT_TO_TYPE if
46 * findvar.c (value_from_register): Use REGISTER_CONVERT_TO_TYPE if
51 * configure.in: Add i[34]86-*-isc*.
53 * stabsread.c: Make sure all complain() pass the address of the struct.
55 * xcoffread.c: Make sure all struct complaints are static not auto.
57 * Makefile.in: Add rule for xcoffexec.o like that for paread.o.
59 * xcoffread.c (process_xcoff_symbol, case C_LSYM): Use define_symbol.
63 * config/i386/linux.mh: Re-enable coredumps now that they should work.
67 * config/m68k/tm-m68k.h (FRAME_CHAIN): add missing close paren.
71 * config/pa/nm-hppab.h: Comment PTRACE_ARG3_TYPE.
75 * Makefile.in (INSTALLED_LIBS): New variable.
79 * main.c (quit_command): In the "quit anyway?" message, tell the user
80 whether we are planning to detach or kill the program.
82 * config/vax/vaxbsd.mh: Add "NAT_FILE= nm-vax.h".
83 * config/vax/xm-vaxbsd.h: Use <sys/param.h> not <machine/limits.h>
85 * infcmd.c (read_pc): Doc fix.
87 * printcmd.c (print_address_symbolic): Use %u not %d for offset.
89 * blockframe.c (get_prev_frame_info): If pc in sigtramp, set
90 signal_handler_caller.
91 * tm-68k.h (FRAME_{CHAIN,SAVED_PC}): Deal with sigtramp.
92 * tm-hp300bsd.h: Define SIGTRAMP_{START,END} not IN_SIGTRAMP.
93 * inferior.h (IN_SIGTRAMP): Definition moved from infrun.c.
94 Use SIGTRAMP_START if defined.
95 * infcmd.c (step_1): Use SIGTRAMP_{START,END} if needed.
96 * infrun.c (wait_for_inferior): Check IN_SIGTRAMP before SKIP_PROLOGUE.
98 * infptrace.c: Remove unused KERNEL_U_ADDR_HPUX code.
100 * infcmd.c (step_1): Fix poorly worded error message.
102 * config/{i386/linux.mh,m68k/isi.mh} (NATDEPFILES):
103 Comment out corelow.c because core dumps are broken on these machines.
105 * Makefile.in (depend): Put "${srcdir}" in generated dependencies
106 if srcdir is not ".".
107 Also put in -I${BFD_DIR} or -I${READLINE_DIR} for files which need it.
108 (INCLUDE_CFLAGS): Remove BFD_DIR and READLINE_DIR.
109 * depend: Update to latest automatically built version.
113 * ChangeLog, ChangeLog-92: Split ChangeLog at 1993.
114 * Makefile.in (NONSRC): Add ChangeLog-92
118 * findvar.c ({read,write}_register): Use REGISTER_RAW_SIZE
119 not typo RAW_REGISTER_SIZE.
121 * frame.h, inferior.h: Doc fixes.
125 * findvar.c (write_register): Add sanity check for register size.
126 (read_register): Fixup sanity check for register size to be
127 consistent with write_register().
131 * sparclite/Makefile.in: Add dummy info, install and install-info
136 * remote-nindy.c: Removed declaration of coffstrip.
137 * nindy-share/nindy.c: #if 0 coffstrip routine; no longer used.
141 * Makefile.in (VERSION): Bump to 4.9.1 after release and cvs
144 * Makefile.in (VERSION): GDB 4.9 release.
148 * README: Update known bugs to include the Solaris bug that
149 leaves core dumps in the current directory when restarting the
150 inferior with "run". Expand on the testsuite information.
152 * Makefile.in (VERSION): Bump to 4.8.96 for what should hopefully
153 be the last 4.9 prerelease test archive.
157 * config/m68k/xm-hp300bsd.h: Include <sys/param.h> to avoid INT_MAX
162 * README, NEWS: Update for gdb 4.9 release.
166 * ch-exp.y (MAX, MIN): Rename to MAX_TOKEN, MIN_TOKEN.
167 * target.c (MIN): #undef before defining.
171 Patch from Jeffrey Law:
172 * gdb/config/pa/nm-hppab.h (PTRACE_ARG3_TYPE): Define as caddr_t.
176 * hppa-tdep.c (hppa_push_arguments): Allocate correct amount of
181 * ch-exp.y (start): Apply work-around to avoid bison warning.
185 * Makefile.in (BISON): Remove double quotes around BISON
186 definition when bison is used.
188 * configure.in (hppa*-hp-bsd): Change to hppa*-hp-bsd*
189 * configure.in (hppa*-hp-hpux): Change to hppa*-hp-hpux*
190 * configure.in (m68*-hp-bsd): Change to m68*-hp-bsd*
191 * configure.in (m68*-hp-hpux): Change to m68*-hp-hpux*
192 * configure.in (hppa*-*-bsd): Change to hppa*-*-bsd*
193 * configure.in (hppa*-*-hpux): Change to hppa*-*-hpux*
194 * configure.in (m68*-hp-bsd): Change to m68*-hp-bsd*
195 * configure.in (m68*-hp-hpux): Change to m68*-hp-hpux*
197 * Makefile.in (VERSION): Bump to 4.8.6.
201 * config/pa/xm-hppah.h (MALLOC_INCOMPATIBLE): Define it, and
202 include declarations for malloc/realloc/free. Both malloc and
203 realloc return 'void *' for non-ANSI compilations.
207 * coffread.c (read_coff_symtab): Don't fclose stream as it is no
212 * solib.c (clear_solib): Don't close bfd if it is NULL.
216 * core.c (dis_asm_read_memory): Cast second arg of
217 target_read_memory to "char *".
218 * breakpoint.c (watchpoint_check): Change arg type from PTR to
219 "char *", to match other functions called by catch_errors().
224 * gdb/config/nm-hppab.h (PTRACE_ARG3_TYPE): Define as caddr_t.
225 * gdb/config/pa/tm-hppah.h (millicode_start, millicode_end):
226 Delete unnecessary declarations.
230 * ser-unix.c (wait_for): Use VTIME to do timeouts instead of
231 poll() for termio{s}.
235 * i386-tdep.c (i386_frame_num_args): Always return -1.
240 * gdb/hppa-tdep.c: Declare frame_saved_pc.
241 (frameless_function_invocation): New function.
242 (frame_saved_pc, init_extra_frame_info): Use
243 frameless_function_invocation.
244 * gdb/config/pa/tm-hppa.h (SAVED_PC_AFTER_CALL): Use saved_pc_after
245 call instead of just grabbing the value currently in %r2.
246 (FRAMELESS_FUNCTION_INVOCATION): Use frameless_function_invocation.
247 * gdb/config/pa/tm-hppah.h (SAVED_PC_AFTER_CALL): Delete private
248 definition and use the common one in tm-hppa.h.
249 * gdb/hppa-tdep.c (frame_chain_valid): If "use_unwind" is true, then
250 use unwind descriptors to determine if the frame chain is valid.
251 * gdb/hppa-tdep.c (find_dummy_frame_regs): Rework so that
252 it does not assume %r4 is the frame pointer.
253 * gdb/hppa-pinsn.c (print_insn): Handle 'r' and 'R' for break, rsm,
254 and ssm instructions.
255 * gdb/hppa-tdep.c (extract_5r_store, extract_5R_store): New
256 helper functions for print_insn.
257 * gdb/hppa-tdep.c (gcc_p, hpux_cc_p): Delete unused functions.
258 * gdb/config/pa/tm-hppa.h (ABOUT_TO_RETURN): Handle a return
259 which nullifies the following instruction.
263 * infptrace.c [FIVE_ARG_PTRACE]: Define ptrace to call_ptrace and
264 pass the 5th arg there, rather than using an ANSI C-specific macro.
266 * Makefile.in (depend): Don't include ${CC} command for *.tab.c.
270 * Makefile.in (VERSION): Bump to 4.8.5
271 * Makefile.in (INCLUDE_CFLAGS): Add BFD_DIR and READLINE_DIR
272 directories to include search path.
273 * Makefile.in (CLIBS, CDEPS, ADD_FILES, ADD_DEPS): Clean up
275 * Makefile.in (depend): For gcc -MM line, use INTERNAL_CFLAGS
276 * Makefile.in (main.o, dbxread.o, coffread.o, mipsread.o,
277 elfread.o, dwarfread.o, stabsread.o, xcoffread.o, xcoffexec.o,
278 xdr_ld.o, xdr_rdb.o, nindy.o, Onindy.o, ttybreak.o, ttyflush.o,
279 udr.o, udip2soc.o): Remove explicit rules, use the ones that
280 are automatically generated in "depend".
281 * Makefile.in (paread.o): Document why a dependency doesn't get
282 automatically generated in "depend" and leave this explicit rule
284 * depend: Update to latest automatically generated version.
288 * xcoffread.c: Doc fix.
290 * Makefile.in (depend): Include $(CC) command in generated output.
294 * Makefile.in (NONSRC): Remove ${srcdir}/putenv.c.
295 * Makefile.in (SFILES): Add ${srcdir}/putenv.c.
296 * depend: Update to latest automatically built version.
300 * sparclite/Makefile.in: Create default target that does nothing
301 in order to force user to build by hand.
303 * sparclite/Makefile: Remove. It's not necessary anymore.
305 * ser-unix.c (wait_for): New routine to handle read timeouts,
306 etc. Uses poll() if HAVE_TERMIO[S] is defined, select() otherwise.
310 * mips-pinsn.c (print_insn): Return value.
314 * Makefile.in (SFILES): Remove ser-hardwire.c; it is a link made
315 at configuration time and doesn't belong in the distribution archive.
317 * Makefile.in (NONSRC): Add 29k-share/README.
318 * Makefile.in (HFILES): Add 29k-share/udi/udiids.h.
320 * defs.h (UINT_MAX, LONG_MAX, INT_MAX, INT_MIN): Replace hex
321 constants with slightly more portable definitions (still depends
322 on 2's complement arithmetic though).
323 * config/i386/nm-linux.h: Define NO_SYS_REG_H for no <sys/reg.h>.
324 * i386v-nat.c (sys/reg.h): Conditionalize include on
325 NO_SYS_REG_H. Linux doesn't have <sys/reg.h>.
326 * ser-unix.c (termio.h): Include <termio.h> like other files that
327 include termio.h, not <sys/termio.h> which may not exist (on
332 * valprint.c (print_longest): Change format parameter from a
333 'char' to an 'int'. We can't have 'char' parameters with the
334 current coding style, where we mix prototypes with pre-ANSI
336 * value.h (print_longest): Change format parameter in prototype
337 from a 'char' to an 'int'.
341 * config/mips/tm-mips.h (STAB_REG_TO_REGNUM): Match it with the gcc
343 * config/mips/tm-irix3.h (STAB_REG_TO_REGNUM): Add.
344 * irix4-nat.c (fill_fpregset): Fix bug with indexing into fpregsetp.
349 * config/pa/hppabsd.mh: Add more files to NATDEPFILES.
350 * config/pa/xm-hppa[bh].h: Define FIVE_ARG_PTRACE.
351 * hppab-nat.c: Delete WANT_NATIVE_TARGET ifdefs.
352 ptrace needs 5 arguments, #define ptrace to always
353 pass zero as the 5th argument.
357 * configure.in: Match z8k-*-sim for z8000.
358 * config/h8500/tm-h8500.h, h8500-tdep.c: Lint.
359 * remote-hms.c: Update to use new serial protocol.
363 * mips-tdep.c: remove include of sys/dir.h. Doesn't seem
364 necessary and Solaris doesn't have it.
366 * Makefile.in (clean-info, install, install-info, info, dvi,
367 check, all): do not echo recursion lines.
369 * 29k-share/udip2soc.c (UDIConnect): replace union wait with int.
371 * config/sparc/sun4sol2.mh (XM_CLIBS): add -lsocket which is
372 required target ports which use sockets (like a29k-udi).
374 * remote-udi.c (udi_wait): Use SIGURG, as Solaris doesn't have SIGLOST.
378 * ser-unix.c [USE_{TERMIO,ALARM}_TIMEOUT]: New code to deal with
379 systems lacking select().
381 * Makefile.in (TAGS): Doc fix. Deal with empty DEPFILES.
385 * alldeps.mak, depend: Update with latest automatically built
390 * Makefile.in (SFILES): Add ser-unix.c and ser-go32.c.
392 * Makefile.in (make-proto-testsuite.dir): New target to make
393 prototype testsuite tree.
395 * Makefile.in (VERSION): Bump to 4.8.4.
399 * stabsread.c (define_symbol): If unrecognized constant type,
400 complain() not error().
404 * infptrace.c: Add missing close paren to test for
405 FIVE_ARG_PTRACE defined.
407 * defs.h (CC_HAS_LONG_LONG): Set up to define CC_HAS_LONG_LONG
408 when compiling with gcc, but disable it for now. See comment.
409 * defs.h (LONGEST): Define as either "long" or "long long"
410 based on CC_HAS_LONG_LONG.
411 * defs.h (longest_to_int): Use CC_HAS_LONG_LONG to control
412 how longest_to_int is defined.
413 * c-valprint.c (c_val_print): Call print_longest.
414 * expprint.c (dump_expression): Use PRINTF_HAS_LONG_LONG
415 instead of LONG_LONG.
416 * {printcmd.c, gdbtypes.h} (LONG_LONG): Replace usages with
418 * printcmd.c (print_scalar_formatted): Call print_longest
419 and let it figure out what to do for PRINTF_HAS_LONG_LONG.
420 * typeprint.c (print_type_scalar): Call print_longest and let
421 it figure out what to do for PRINTF_HAS_LONG_LONG.
422 * valprint.c (val_print_type_code_int): Call print_longest
423 and let it figure out what to do for PRINTF_HAS_LONG_LONG.
424 * stabsread.c (LONG_LONG): Replace usages with CC_HAS_LONG_LONG.
425 * value.h (struct value): Replace usage of LONG_LONG with
427 * value.h (print_longest): Add prototype.
428 * values.c (LONG_LONG): Replace usages with CC_HAS_LONG_LONG.
429 * values.c (unpack_double): Collapse code that was unnecessarily
430 dependent on CC_HAS_LONG_LONG. Use LONGEST instead of direct types.
431 * values.c (value_from_longest): Remove dependency on
432 CC_HAS_LONG_LONG and just use LONGEST.
433 * solib.c (solib_map_sections): Use bfd_get_filename
434 to access filename field.
435 * solib.c (clear_solib): Save filename and free it later, after
436 bfd_close, since bfd_close may reference it. Use bfd_get_filename
438 * config/convex/xm-convex.h (LONG_LONG): Replace with
439 CC_HAS_LONG_LONG. Add define for PRINTF_HAS_LONG_LONG.
440 * doc/gdbint.texinfo (LONG_LONG): Replace with CC_HAS_LONG_LONG.
441 Add PRINTF_HAS_LONG_LONG references.
445 * inflow.c (kill_command), infcmd.c (attach_command),
446 remote.c (remote_interrupt_twice): In messages for the user, call it
447 "the program" or "the program being debugged" not "the inferior".
449 * hp300ux-nat.c: Cast second arg to supply_register calls.
450 (_initialize_kernel_u_addr, getpagesize): New functions.
451 (store_inferior_register_1): Change arg name from value to val.
452 (fetch_core_registers): Make arg core_reg_size unsigned.
453 Pass 5 args to ptrace.
454 * config/m68k/xm-hp300hpux.h: Define FIVE_ARG_PTRACE.
455 Remove KERNEL_U_ADDR stuff.
456 * infptrace.c [FIVE_ARG_PTRACE]: Pass 5th arg to ptrace.
457 * config/m68k/hp300hpux.m{t,h}:
458 Move exec.o from NATDEPFILES to TDEPFILES
459 * config/m68k/hp300hpux.mt: Mention GAS requirement. Remove
460 hp-include stuff. Add m68k-tdep.o to TDEPFILES.
464 * ch-exp.y (yylex): Don't STREQ with simplename if it is NULL.
468 * config/sparc/xm-sun4os4.h [__STDC__]: Don't use MALLOC_INCOMPATIBLE.
472 * doc/gdb.texinfo: make node "Shell Commands" unconditional;
473 describe `set demangle-style arm' (not cfront);
474 mention can type `q' to discard output, when gdb pages
478 * valops.c (search_struct_field): Fix gdb core dump with incomplete
483 * remote.c: Change timeout to 2.
484 (remote_open): Use unpush_target not remote_close.
485 (remote_resume): If siggnal != 0, give warning not error().
486 (remote_wait, remote_interrupt, remote_interrupt_twice):
487 If we get two interrupts, let the user get out if they want.
488 (remote_{kill,mourn}): New functions.
489 i386-stub.c (handle_exception, case 'k'): Don't BREAKPOINT.
493 * config/sparc/sun4sol2.mh (XM_CLIBS): Define to be -lnsl.
497 * Remote targets (mourn): Call unpush_target.
499 * config/sparc/xm-sun4os4.h: Declare free() to return int.
500 Remove twisted use of PARAMS.
502 * config/rs6000/xm-rs6000.h: Don't define MALLOC_INCOMPATIBLE now
503 that ansidecl.h assumes ANSI on AIX.
507 * README: Move most stuff about hacking GDB to doc/gdbint.texinfo.
508 (Known bugs): Remove AIX bugs, revise SPARC struct bug description.
512 * expprint.c (print_subexp): Fix bug with OP_SCOPE operator output.
516 * remote-vx.c (net_connect): Allow numeric IP address for host.
520 * config/sh/sh.mt, config/sh/tm-sh.h, sh-tdep.c: New files.
524 * rs6000-tdep.c (branch_dest): Deal with stepping through system call.
526 * symtab.h, xcoffread.c: Revise linetable sorting comments.
530 * valops.c (value_cast): A cast might also change the object
531 representation in C++.
532 * dbxread.c (end_psymtab): Copy subpst read_symtab function from pst
533 to get the proper read_symtab function when called from mipsread.c.
534 * mipsread.c (mipscoff_psymtab_to_symtab, psymtab_to_symtab_1):
535 Set cur_bfd in psymtab_to_symtab_1 as CURBFD(pst) is invalid
536 for dummy psymtabs, inhibit processing of dummy psymtabs.
540 * Changes from (or inspired by) AMD:
541 * remote-udi.c (udi_attach): Assignments to Space and Offset were
543 (udi_wait): Make error message (UDIGetStdout) match error.
544 (udi_wait): Handle UDIStdinNeeded.
545 * command.c [CANT_FORK]: Use system().
546 * utils.c (prompt_for_continue): Allow quit with 'q'.
548 * solib.c (solib_add): Don't call special_symbol_handling if there
549 were errors in symbol_add_stub. Also set so->from_tty before
550 calling symbol_add_stub.
554 * Merge in HPPA/BSD patches from Utah:
555 * defs.h: Add const to 2nd arg of psignal prototype.
556 * hppah-tdep.c: Renamed to hppa-tdep.c 'cuz it's common code with
558 * hppab-core.c: Deleted. No longer useful.
559 * hppab-nat.c: #include more files. Use PT_WUREGS, not
561 * hppab-tdep.c: Deleted. Supplanted by hppa-tdep.c.
562 * config/pa/hppabsd.mh (NATDEPFILES): Remove hppab-core.o.
563 * config/pa/hppabsd.mt (TDEPFILES): hppab-tdep.o => hppa-tdep.o
564 * config/pa/hppahpux.mt (TDEPFILES): hppab-tdep.o => hppa-tdep.o
565 * config/pa/xm-hppab.h: #define SET_STACK_LIMIT_HUGE.
569 * Fix two bugs found by deja-gnu. One is the incorrect reporting
570 of the PC being in a stack dummy when looking at a core file
571 without symbols. The other is the incorrect passing of char
572 arguments during expression evaluation (ie: p foo('a','b') would
573 mess up the passing of it's args because it wasn't coercing the
575 * hppah-tdep.c: Rename global functions to have consistent hppa_
576 prefix. Make more functions static. Drop hp_ prefix from static
577 functions. (hppa_push_arguments): Call value_arg_coerce to cast
578 char to int args if necessary. (hppa_fix_call_dummy): Create
579 this routine from FIX_CALL_DUMMY macro in tm-hppa.h.
580 * inferior.h (PC_IN_CALL_DUMMY): Check for frame_address being
581 valid (ie: != 0) before doing comparison against PC.
582 * valops.c (call_function_by_hand): Adjust call to FIX_CALL_DUMMY
583 to reflect new arguments.
584 * config/pa/tm-hppa.h (POP_FRAME, PUSH_ARGUMENTS): Use new hppa_
585 prefix for func name. (FIX_CALL_DUMMY): Move code into
588 * testsuite/gdb.t16/gdbme.c, testsuite/gdb.t17/gdbme.c: Add calls
589 to malloc() so that we can test GDB eval of dynamically created
590 arrays (like char strings in `print "foo"').
594 * printcmd.c (print_address_symbolic): Search symtabs as well as the
595 minimal symbols for a nearby symbol.
599 * coffread.c: Comment changes around minimal symbol recording.
603 * command.c: comment changes only.
605 * mips-tdep.c (heuristic_fence_post): new static variable.
606 (heuristic_proc_start): use heuristic_fence_post, print better
607 warnings, but only if not stop_soon_quietly.
608 (_initialize_mips_tdep): add_set_cmd for heuristic-fence-post.
612 * symtab.h: Fix LOC_REF_ARG comment.
617 * stabsread.c (define_symbol): Combine a 'p', 'r' arg pair to a
619 * config/sparc/tm-sparc.h (REG_STRUCT_HAS_ADDR): Revise comments.
620 symfile.c (compare_symbols): Don't check first character; STRCMP
623 * stabsread.c (define_symbol): Generate a LOC_REGPARM_ADDR for
624 structures that are passed by address in a register.
625 * symtab.h (enum address_class): Add LOC_REGPARM_ADDR.
626 * findvar.c (read_var_value),
627 printcmd.c (address_info, print_frame_args),
628 stack.c (print_frame_arg_vars), symmisc.c (print_{,partial_}symbol),
629 * symtab.c (lookup_block_symbol): Deal with it.
633 * objfiles.h (obj_section), objfiles.c (build_objfile_section_table):
635 * objfiles.c (find_pc_section): Return a struct obj_section *.
636 * sparc-tdep.c (in_solib_trampoline): Deal with find_pc_section return.
637 * symfile.c (syms_from_objfile) [IBM6000_TARGET]:
638 Don't use obj_section hack.
639 * xcoffexec (vmap_symtab): Relocate obj_sections.
640 * printcmd.c (containing_function_bounds): Use find_pc_section.
642 * symtab.h: Clean up SYMBOL_VALUE comments.
646 * stack.c (print_frame_arg_vars), printcmd.c (print_frame_args):
647 Expand comments about LOC_ARG/LOC_LOCAL pairs.
649 * coffread.c (read_coff_symtab): Use rewind before fseek.
653 * ch-exp.y: Removed unused structure_primitive_value and FIXME_23.
654 * Makefile.in: Add $(YFLAGS) when using $(YACC).
655 * Makefile.in: Remove message to expect conflicts and unused
656 rules in ch-exp.y, since there no longer are any such.
658 Wed Apr 21 13:27:50 1993 david d `zoo' zuhn (zoo at cirdan.cygnus.com)
660 * stabs.texinfo: fixed bad xrefs (un-initialized statics)
664 * xcoffexec.c (xcoff_relocate_core): If no exec file, error()
665 rather than dumping core.
667 * Makefile.in: Add ${srcdir}/ to all source files.
668 (alldeps.mak): Add "${srcdir}/" to files when generating alldeps.mak.
669 (TAGS): Deal with srcdir and new config directory scheme.
671 Makefile.in (NONSRC): Remove createtags.
672 alldeps.mak: Updated.
674 * rs6000-tdep.c: Delete unused function print_frame.
676 * frame.h (struct frame_info): Doc fix for next_frame.
677 New field signal_handler_caller.
678 blockframe.c (create_new_frame, get_prev_frame_info),
679 config/rs6000/tm-rs6000.h (INIT_EXTRA_FRAME_INFO): Set it (needs
680 INIT_FRAME_PC_FIRST).
681 stack.c (print_frame_info), rs6000-tdep.c (rs6000_frame_chain):
686 * irix4-nat.c (fetch_core_registers): Special version of this for
687 Irix 4.x, which stores regs a bit differently from other /proc
689 * procfs.c, core-svr4.c: Move fetch_core_registers from procfs.c
690 to new file core-svr4.c.
691 * config/i386/i386sol2.mh, config/i386/i386v4.mh, config/m68k/amix.mh,
692 config/i386/ncr3000.mh, config/sparc/sun4sol2.mh: Add core-svr4.o
694 * config/mips/irix4.mh: Add corelow.o to NATDEPFILES.
698 * i387-tdep.c: Remove unused #includes.
700 * configure.in: Match i[34]86-*-sysv3.2 not i[34]86-*-sysv32.
702 * config/i386/nm-i386v.h: Define NO_PTRACE_H.
706 * xcoffread.c: Nuke NO_DEFINE_SYMBOL code. There is no going back.
708 * stabsread.c (define_symbol): 'R' is synonym for 'P', not 'r'.
709 xcoffread.c (process_xcoff_symbol, case C_RPSYM):
710 Don't muck with SYMBOL_CLASS.
714 * munch: Don't use head command. It doesn't exist everywhere.
718 * inflow.c (new_tty): Remove spurious 'o' character at end
723 * mips-tdep.c (mips_skip_prologue): Always skip the typical prologue
724 instructions and nothing more.
725 * mipsread.c (add_line): Add comment why we have to combine line number
726 entries for the same line number.
730 * symtab.{c,h}: Doc fixes (remove symseg references, last relevant
735 * depend, alldeps.mak: Update, now that gcc -MM bug is fixed.
739 * source.c (select_source_symtab): Clean up comment. Also, if
740 we have a current_source_symtab, and s is NULL, return without
742 xcoffread.c (xcoff_symfile_read): Don't call select_source_symtab.
743 breakpoint.c (breakpoint_re_set): Don't call select_source_symtab.
747 * dbxread.c (unknown_symchar_complaint): Add new complaint.
748 * stabsread.h: Declare it.
749 * partial-stab.h: Use it.
751 * utils.c (malloc_botch): Don't forward-declare if NO_MMALLOC.
755 * stack.c (print_frame_info): Print specially if dummy frame.
757 * breakpoint.c: Add comments regarding within_scope future direction.
761 * xcoffread.c (record_include_{begin,end}): Change fatal to complain.
765 * ch-exp.y: Fix thinko that broke parsing of FALSE.
769 * stabsread.c (read_member_functions): Initialize domain for stubbed
770 member functions to avoid gdb core dumps when printing pointers
772 * cp-valprint.c (cp_print_class_method): Check for stubbed member
777 * expprint.c (print_subexp): If opcode not found in op_print_tab,
778 stop with an error().
779 eval.c (evaluate_subexp): Change error message.
781 * objfiles.c (build_objfile_section_table): Cast return value
786 * config/rs6000/tm-rs6000.h, rs6000-tdep.c: Move FRAME_CHAIN
787 to rs6000_frame_chain and deal with it if we're in a signal handler.
788 (FRAME_SAVED_PC): Use rs6000_frame_chain.
790 * breakpoint.c (within_scope): New function.
791 (enable_breakpoint, watchpoint_check): Use it.
793 * source.c (openp): Handle "exec-file ./ls" correctly.
795 * breakpoint.c (breakpoint_1): Use wrap_here before "at".
799 * ch-exp.y: Clean up lexing of identifiers and
800 reserved words. (E.g. don't recognize FALSEXXX as the
801 keyword FALSE followed by the identifier XXX.)
802 Also, convert identifiers to lower case.
806 * remote-mips.c, remote-monitor.c, remote-st2000.c: Convert to
807 new serial interface.
811 * remote.c (remote_open): Use SERIAL_OPEN instead of serial_open.
812 (putpkt, getpkt): Use new return codes for SERIAL_READCHAR.
813 * ser-go32.c: Return -1 on most failures, 0 on most successes,
814 and use new return codes for go32_readchar().
815 * ser-unix.c: Ditto. Also, move error handling up to caller for
816 SERIAL_SETBAUDRATE().
817 * serial.c (serial_open): Internal call, not SERIAL_OPEN to get
819 (serial_close): New routine to wrap around device close routine.
820 serial.h: Clean & document return values more clearly.
824 * rs6000-pinsn.c (print_operand): Deal with no operand instructions.
826 * rs6000-pinsn.c (print_operand, case LI): Print condition register
827 operand in decimal rather than wrong textual versions.
829 * printcmd.c (_initialize_printcmd): Clean up docstring for "x"
830 (mention 't', remove false thing about 'g' only good with 'f').
832 * breakpoint.h: move "struct breakpoint" and friends to top of
833 file so that bpstat_find_breakpoint prototype works.
835 * solib.c (struct so_list): Add bfd field.
836 (solib_map_sections): Leave bfd open and scratch_pathname allocated.
837 Put the bfd in bfd field of the so_list.
838 (clear_solib): Free bfd name and close_bfd on the bfd.
842 * valarith.c (value_subscript): Add COERCE_REF.
843 * ch-exp.y (operand_5): We can generalize the 2nd operand
844 of a string repetition ot 'literal' without ambiguity.
848 * breakpoint.h (struct bpstat): Remove momentary field.
849 Remove bpstat_momentary_breakpoint. This was always kludgy
850 and is no longer used.
852 * breakpoint.h: Add enum bpstat_what.
853 breakpoint.h (struct bpstat), breakpoint.c (bpstat_stop_status):
854 stop and print fields of bpstat now per-breakpoint, not just
855 one for the whole chain.
856 breakpoint.{c,h} (bpstat_what): New function.
857 breakpoint.h: Remove bpstat_stop and bpstat_should_print.
858 infrun.c: Replace switch (stop_bpstat->breakpoint_at->type)
859 with call to bpstat_what.
860 README: Remove watchpoint/breakpoint bug from known bugs.
862 * breakpoint.h: Prototype bpstat_find_breakpoint.
866 * symtab.c (find_methods, gdb_mangle_name): Note that functions
868 * symtab.h (VTBL_FNADDR_OFFSET, OPNAME_PREFIX_P, VTBL_PREFIX_P,
869 DESTRUCTOR_PREFIX_P): Note that macros are g++ specific.
873 * i960-pinsn.c (tabent): Copied struct definition from
878 * symtab.h (DESTRUCTOR_PREFIX_P): New macro to check if physname
880 * symtab.c (gdb_mangle_name): Use it.
881 * symtab.c (find_methods): Do not add destructors to choice list
883 * symtab.c (decode_line_1): Make breakpoints on destructors work
888 * ser-go32.c: Make it use serial port name.
889 * go32-xdep.c: Put in def for strlwr, needed by dir.o in go32 libc.
891 * infcmd.c (read_pc): Make sure that we read PC_REGNUM when not
896 * configure.in: Only configure sparclite subdir when target_cpu
901 * xcoffread.c (struct coff_symbol): Change c_sclass to unsigned char.
902 Remove FIXME comment regarding this.
904 * symfile.h: Change NULL->'\0' in comment (that wasn't a typo).
906 * xcoffread.c (read_xcoff_symtab): Use E_SYMNMLEN.
910 Add section table to objfile struct. Use it for find_pc_section.
911 * objfiles.c (add_to_objfile_sections,
912 build_objfile_section_table, find_pc_section): new functions.
913 (allocate_objfile): build section table.
914 * objfiles.h (struct obj_section): new structure.
915 (struct objfile): add section table.
916 (find_pc_section): new prototype.
917 * solib.[ch] (find_pc_section_from_so_list): removed.
918 * sparc-tdep.c: include objfiles.h for find_pc_section. include
919 symfile.h for objfiles.h.
920 (in_solib_trampoline): adjusted for new find_pc_section
921 prototype. Removed BAD_RICH_HACK ifdefs.
922 * symfile.c (syms_from_objfile): offset objfile sections.
923 (find_pc_section): removed. Also removed BAD_RICH_HACK ifdefs.
924 * symfile.h (find_pc_section): prototype removed. Also fixed
925 comment typo NUL -> NULL.
926 * target.[ch] (find_pc_section_from_targets): removed.
927 * config/sparc/tm-sun4sol2.h (BAD_RICHH_HACK): removed.
931 * ser-go32.c: Format. (go32_open): Use proper return value.
933 * configure.in: Undo conditional configdirs hack for sparclite.
937 * symtab.c (list_symbols): When call break_command, pass both
938 filename and function name not just function name.
942 (Changes and new files to make "none" a full fledged configuration)
943 * config/none/{nm-none.h, tm-none.h, xm-none.h}: New files.
944 Currently only tm-none.h has any meaningful contents.
945 * config/none/none.mh (NAT_FILE): Use nm-none.h
946 * config/none/none.mh (XM_FILE): Use xm-none.h
947 * config/none/none.mt (TM_FILE): Use tm-none.h
948 * Makefile.in (depend): Remove comment about parse errors in
949 valops.c, it now parses correctly and generates a correct depend
950 line. Remove line that touches xm.h, tm.h, and nm.h; they are
951 now linked to config/none/{xm-none.h, tm-none.h, nm-none.h}.
955 * values.c (USE_STRUCT_RETURN): Only use gcc wierdness for gcc1.
957 * xcoffread.c (read_xcoff_symtab): Deal correctly with symbols of
958 exactly 8 characters.
962 * configure.in: Sparclite uses sparc config dir. Also has it's
963 own tm- & .mt files now. Also add sparclite to configdirs.
964 * go32-xdep.c: Dummy routines for sigsetmask & strlwr.
965 * config/i386/go32.mh: Nullify def of TERMCAP.
966 * config/i386/xm-go32.h: Get rid of redef of EIO.
967 * config/sparc/{sparclite.mh tm-sparclite.h}: New sparclite
968 specific configs. Very similar to sun4os4, but without solib.
969 * sparclite/{Makefile.in configure.in}: First cut at making this
974 * ser-go32.c: First cut at adapting to new serial interface.
978 * Makefile.in (SFILES OBS): Add serial.[co] & ser-hardwire.[co].
979 These implement a new serial line interface for talking to remote
981 * configure.in: Link ser-hardwire.c to ser-unix.c for all hosts,
982 EXCEPT go32, which gets ser-go32.c.
983 * remote.c: Use new serial interface. More remote-xxx's to be
985 * ser-bsd.c, ser-termios.c: Removed.
986 * serial.c: New. Implements common operations for all serial
988 * ser-unix.c: New. Unix specific serial operations for various
989 flavors of Unix (Posix, SysV, BSD).
990 * serial.h: Generic serial interface defs.
991 * config/i386/go32.mh, config/i386/i386bsd.h,
992 config/m68k/apollo68b.mh, config/sparc/sun4os4.mh: Remove
993 ser-bsd.o from XDEPFILES. All the magic is now handled in
998 * config/h8500/tm-h8500.h: Clean up brain damage found by GCC.
1002 * xcoffread.c (xcoff_symfile_offsets): Use 0 not addr for offsets.
1004 * rs6000-tdep.c (frameless_function_invocation): Don't even think
1005 about framelessness except on the innermost frame.
1007 * xcoffexec.c: Call fatal() not abort().
1009 * stabsread.c (patch_block_stabs): If stab & no symbol, make
1010 a LOC_OPTIMIZED_OUT symbol.
1011 symtab.h (enum address_class): Add LOC_OPTIMIZED_OUT.
1012 findvar.c (read_var_value), printcmd.c (address_info),
1013 symmisc.c (print_{,partial_}symbol), c-exp.y (variable),
1014 m2-exp.y (yylex): Deal with it.
1015 ch-exp.y (yylex): Deal with it.
1019 * findvar.c (value_from_register): H8500 specific, check to see
1020 if we are looking at short pointer. If so, skip crock.
1021 * h8500-tdep.c (h8500_frame_chain): Mask down value from
1022 read_memory_integer() to avoid getting messed up by sign extension.
1026 * sparc-tdep.c (in_solib_trampoline), symfile.c (find_pc_section):
1027 ifdef protect using BAD_RICH_HACK. This should be removed soon.
1028 * config/sparc/tm-sun4sol2.h (BAD_RICH_HACK): define.
1032 * i960-pinsn.c, a29k-pinsn.c: Much abridged, just use libopcodes.a.
1034 * core.c (dis_asm_print_address): New function.
1036 * core.c (dis_asm_read_memory): Reinstate 4th arg. The prototype
1041 * breakpoint.c (bpstat_print, bpstat_stop_status): Change to walk the
1042 entire breakpoint chain and print only the first entry that needs to
1043 be printed and needs to be stopped for. Fixes problems with printing
1044 of multiple breakpoints with different conditions.
1045 * breakpoint.c (print_it_done): Renamed from print_it_noop as it
1046 effectively stops printing of the breakpoint chain.
1047 * breakpoint.c (print_it_noop): New routine to print nothing
1048 for this breakpoint entry and dont stop printing.
1049 * breakpoint.c (breakpoint_re_set_one): mention the reevaluated
1050 watchpoint only if it is enabled.
1051 * mipsread.c (parse_procedure): Correct incorrect setjmp procedure
1052 descriptor from the library to make backtraces through setjmp work.
1053 * mipsread.c (fixup_sigtramp): Correct pcreg and fregoffset for
1055 * mips-tdep.c (read_next_frame_reg): Provide correct values for
1056 all registers saved within sigtramp, cleanup.
1060 * sparc-pinsn.c: Much abridged, just calls version in libopcodes.a.
1064 * core.c (dis_asm_read_memory): drop fourth arg which conflicts
1065 with prototype in ../include/dis-asm.h.
1069 * core.c (dis_asm_{read_memory,memory_error}): New functions.
1070 m68k-pinsn.c, h8500-tdep.c, i386-pinsn.c, mips-pinsn.c, z8k-tdep.c:
1071 Use read_memory_func interface to disassembler.
1075 Teach sparc solaris to next over shared library functions.
1076 * solib.[hc] (find_pc_section_from_so_list): new function and
1078 * sparc-tdep.c (in_solib_trampoline): new function.
1079 * symfile.[hc] (find_pc_section): new function and prototypes.
1080 * target.[hc] (find_pc_section_from_targets): new function and
1082 * config/sparc/tm-sun4sol2.h (IN_SOLIB_TRAMPOLINE): redefine to
1083 in_solib_trampoline.
1087 * infrun.c (wait_for_inferior): Revise comment.
1089 * command.c (do_setshow_command): Use %u with var_{u,z}integer.
1091 * command.{c,h}: New var_type var_integer.
1092 main.c: Use it for history_size.
1094 * rs6000-tdep.c, xcoffexec.c, config/rs6000/xm-rs6000.h, breakpoint.c:
1095 Lint and byte-order fixups.
1097 * breakpoint.c (print_it_normal): Return 0 after hitting watchpoint.
1099 * breakpoint.h (bpstat): New field print_it.
1100 breakpoint.c (bpstat_print): Use it.
1101 (print_it_normal): New function (from old bpstat_print code).
1102 (bpstat_{alloc,stop_status}): Set print_it field.
1104 * breakpoint.c (bpstat_stop_status): Use catch_errors when
1105 evaluating watchpoint condition, via new function watchpoint_check.
1106 Also stop if watchpoint disabled due to leaving its block.
1108 * findvar.c [REG_STRUCT_HAS_ADDR]: Add comment.
1112 * mips-pinsn.c: Add missing include of dis-asm.h.
1116 * Makefile.in (clean, distclean, realclean): Recursively apply
1117 to subdirs first, rather than last. This avoids, for example,
1118 Makefile being removed in a parent directory before the recursive
1121 * alldeps.mak, depend: Update for below changes.
1123 * config/m68k/tm-m68k.h: Renamed from config/m68k/tm-68k.h.
1124 * m68k/{tm-3b1.h, tm-altos.h, tm-amix.h, tm-es1800.h,
1125 tm-hp300bsd.h, tm-hp300hpux.h, tm-isi.h, tm-news.h, tm-os68k.h,
1126 tm-st2000.h, tm-sun2.h, tm-sun3.h, tm-vx68.h}: Include tm-m68k.h
1127 instead of tm-68k.h.
1128 * Makefile.in (HFILES): tm-68k.h renamed to tm-m68k.h.
1129 * README, a29k-pinsn.c, m68k-pinsn.c, m68k-stub.c, remote-vx.c,
1130 m68k/{altos.mh, altos.mt, apollo68b.mh, nm-apollo68b.h,
1131 nm-hp300bsd.h, config/m68k/xm-apollo68b.h}: Map '68k' to 'm68k'.
1132 * a29k/tm-a29k.h, doc/gdbint.texinfo: Account for renaming of
1133 tm-68k.h to tm-m68k.h.
1134 * m68k/m68k-fp.mt (TM_FILE): tm-68k-fp.h renamed to tm-m68k-fp.h.
1135 * m68k/m68k-nofp.mt (TM_FILE): tm-68k-nofp.h renamed to
1138 * config/a29k/tm-a29k.h: Renamed from config/a29k/tm-29k.h.
1139 * a29k-pinsn.c: Renamed from am29k-pinsn.c.
1140 * a29k-tdep.c: Renamed from am29k-tdep.c.
1141 * remote-eb.c, config/a29k/tm-ultra3.h: Include renamed tm-a29k.h.
1142 * remote-monitor.c, remote-st2000.c, config/a29k/{nm-ultra3.h,
1143 tm-a29k.h, xm-ultra3.h}, config/romp/rtbsd.mh, doc/gdbinv-s.texi,
1144 testsuite/gdb.t15/funcargs.exp, testsuite/gdb.t17/callfuncs.exp:
1145 Map '29k' to 'a29k'.
1146 * config/a29k/{a29k-kern.mt, a29k-udi.mt, a29k.mt, ultra3.mt}
1147 (TDEPFILES): Use renamed a29k-pinsn.o and a29k-tdep.o.
1148 * config/a29k/{a29k-udi.mt, a29k.mt} (TM_FILE): Use renamed
1150 * config/a29k/a29k-udi.mt (MT_CFLAGS): Remove TARGET_AM29K
1151 define that does not appear anywhere else in the gdb source tree.
1152 * doc/gdbinit.texinfo: Document renaming of tm-29k.h to tm-a29k.h.
1156 * breakpoint.c: Add comments regarding breakpoint_re_set.
1158 * xcoffread.c (sort_syms, compare_symbols): Remove.
1159 (xcoff_symfile_read): Use sort_all_symtab_syms from symfile.c
1160 not our own sort_syms (it is identical).
1162 * xcoffread.c: Nuke NAMES_HAVE_DOT define (not used).
1166 * breakpoint.c (breakpoint_re_set_one): Fix storage leak.
1167 * breakpoint.c (enable_breakpoint): Don't enable watchpoint if it
1169 * exec.c (exec_close): Fix storage leak.
1170 * exec.c (exec_file_command): Make sure that bfd doesn't realign the
1171 output sections when patching an executable.
1172 * mips-nat.c (store_inferior_registers): Use REGISTER_PTRACE_ADDR
1173 when writing all registers.
1174 * mips-tdep.c (mips_push_dummy_frame): Save floating point registers
1175 at the right offset in the dummy frame.
1176 * mipsread.c (psymtab_to_symtab_1): Do not complain for stProc,
1177 stStaticProc and stEnd symbols as they are generated by gcc-2.x.
1178 * mipsread.c (mipscoff_new_init): Initialize stabsread and buildsym.
1182 * Makefile.in (TARFILES): Avoid trailing backslash.
1186 * breakpoint.{c,h}: Add exp_string to struct breakpoint and use
1187 it in breakpoint_re_set.
1188 * breakpoint.c (watch_command, enable_breakpoint): Fetch lazy values.
1190 * rs6000-tdep.c (single_step): Misc cleanups (CORE_ADDR not int,
1191 don't use sizeof(int) for target stuff, etc).
1195 * alldeps.mak, configure.in, i860-break.h, i860-opcode.h,
1196 i860-pinsn.c, i860-tdep.c, config/i860/*: Remove incomplete i860
1197 support that can't be integrated anyway due to lack of clear
1202 * findvar.c (read_register, write_register): Make these capable
1203 of reading/writing registers that are shorter than REGISTER_TYPE.
1204 * (value_from_register): Install H8500 specific code to return
1205 proper value when register is being used as a pointer.
1206 * h8500-tdep.c: Remove extra defines of NUM_REGS.
1207 (h8500_skip_prologue): Use correct lengths for LINK instructions.
1208 (FRAME_CHAIN): Change name to h8500_frame_chain. Rewrite code to
1209 chain frames properly by combining frame pointer with T reg.
1210 (init_extra_frame_info): Delete. It's now a macro.
1211 (frame_args_address): Don't add PTR_SIZE. Stack args are already
1212 offset by the correct amount off of the frame pointer.
1213 (register_byte): Delete. It's now a macro.
1214 (register_raw_size, register_virtual_size): Delete. Replaced by
1215 common routine h8500_register_size, cuz there's no difference
1216 between the raw & virtual sizes on this machine.
1217 (register_convert_to_raw, register_convert_to_virtual): Delete,
1218 cuz there's no difference between the raw & virtual forms.
1219 Replaced by memcpy in tm file.
1220 (register_virtual_type): Rename to h8500_register_virtual_type.
1221 Get rid of pointer pseudo-regs, use _REGNUM with all reg names.
1222 (_initialize_h8500_tdep): Get rid of crock to ensure that GDB &
1223 emulator have same reg offsets. This is all handled in the
1225 (h8500_trapped_internalvar): New routine to detect references to
1226 convenience vars acting as pointer pseudo-regs.
1227 (h8500_value_trapped_internalvar): Conjure up value of pointer
1229 (h8500_set_trapped_internalvar): Convert set value in real
1230 register references.
1231 infcmd.c (read_pc, write_pc): Add h8500 specific code to handle
1232 code segment register.
1233 infrun.c (proceed): Simplify. Call write_pc instead of doing it
1235 (wait_for_inferior): Add h8500 specific code to add stack segment
1236 when reading SP register.
1237 remote-sim.c (fetch_register): Spacing.
1238 tm-h8500.h: #define GDB_TARGET_IS_H8500 to make it easier to
1239 detect cruft. Redo all register manipulation stuff. Get rid of
1240 pointer pseudo-regs. (INIT_EXTRA_FRAME_INFO): Adds stack segment
1241 to frame pointer. (IS_TRAPPED_INTERNALVAL,
1242 VALUE_OF_TRAPPED_INTERNALVAR, SET_TRAPPED_INTERNALVAR): Use these
1243 to create internal vars for pointer pseudo-regs.
1247 * Makefile.in: Numerous small changes to macro definitions
1248 and rules for building gdb distribution tree. Many macros
1249 eliminated or merged, and rules simplified.
1250 * alldeps.mak: Update.
1253 Wed Mar 24 13:52:29 1993 david d `zoo' zuhn (zoo at poseidon.cygnus.com)
1255 * Makefile.in: recurse through SUBDIRS for dvi target too
1259 * Clean up xcoff relocation.
1260 objfiles.h (struct objfiles): Add section_offsets, num_sections.
1261 symfile.c (syms_from_objfile), xcoffread.c (xcoff_symfile_offsets):
1263 symtab.h (struct general_symbol_info): Add section field.
1264 minsyms.c (prim_record_minimal_symbol{,_and_info}): Set it.
1265 xcoffread.c: Set section for symbols and msymbols.
1266 (struct symtab): Add block_line_section field.
1267 buildsym.c (end_symtab): Set it.
1268 (end_symtab and callers): Add section parameter.
1269 objfiles.c (objfile_relocate): New funciton.
1270 xcoffexec.c (vmap_symtab): Use it.
1271 xcoffsolib.h (struct vmap): Remove unused fields.
1272 config/rs6000/tm-rs6000.h, stack.c, xcoffexec.c: Remove
1273 CORE_NEEDS_RELOCATION, symtab_relocated.
1274 config/rs6000/tm-rs6000.h: Remove use of loadinfotext.
1275 rs6000-tdep.c: Make loadinfotext static.
1276 breakpoint.c (fixup_breakpoints): Doc fix.
1277 symtab.h (struct symtab), config/rs6000/tm-rs6000.h, buildsym.c
1278 (end_symtab): primary field replaces nonreloc.
1282 * symtab.h (struct linetable_entry): Remove confusing comment.
1284 Tue Mar 23 00:01:23 1993 david d `zoo' zuhn (zoo at cirdan.cygnus.com)
1286 * Makefile.in: add installcheck target
1290 * config/{a29k, arm, convex, gould, h8300, i386, i860, i960, m68k,
1291 m88k, mips, none, ns32k, pa, pyr, romp, rs6000, sparc, tahoe, vax,
1292 z8k}: New directories to hold cpu specific configuration files.
1293 Naming follows gcc convention.
1294 * config/{*.mt, *.mh}: All target and host makefile fragment
1295 config files moved to an appropriate config/<cpu> subdirectory.
1296 * nm-*, xm-*, tm-*: All native, host, and target files, which
1297 get linked to nm.h, xm.h, and tm.h respectively by configure,
1298 moved to appropriate config/<cpu> subdirectory.
1299 * nm-sysv4.h, xm-sysv4.h, tm-sysv4.h, tm-sunos.h, nm-trash.h:
1300 Native, host, and target files that are common across more than
1301 one cpu architecture and included by one of the configured
1302 native, host, or target files, get moved to config directory.
1303 * Makefile.in (INCLUDE_CFLAGS): Add -I${srcdir}/config to
1304 pick up native, host, or target include files moved to one of
1305 the config subdirectories, and that are included by other files.
1306 * Makefile.in (alldeps.mak): Modify to account for new config
1307 directory structure.
1308 * alldeps.mak, depend: Update for new config directory structure.
1309 * config/*/[ntx]m-*.h: Modify all files that include other
1310 [ntx]m-*.h files to use path relative to gdb/config. I.E.
1311 "a29k/tm-ultra3.h" includes "a29k/tm-29k.h" rather than just
1313 * remote-eb.c (tm-29k.h): Include a29k/tm-29k.h.
1314 * mipsread.c (tm-mips.h): Include mips/tm-mips.h.
1315 * i860-pinsn.c (tm-i860.h): Include i860/tm-i860.h.
1316 * configure.in: Default gdb_host_cpu to host_cpu, and remap
1317 the ones where the default is not unique or different than the
1318 config subdirectory name. Similarly, handle gdb_target_cpu.
1319 Modify configure.in as appropriate to make use of gdb_host_cpu
1320 and gdb_target_cpu to find makefile fragments and make links.
1324 * mipsread.c (compare_blocks): Sort blocks with the same start
1325 address by decreasing ending address.
1329 * mipsread.c (parse_procedure): Save cur_fdr accross call to
1330 lookup_symbol as it might get clobbered by the call.
1332 * mipsread.c (parse_partial_symbols): Use ADD_PSYMBOL_ADDR_TO_LIST.
1333 The previous code did not initialize the language field for the psymtab
1338 * c-exp.y (parse_number): Avoid shift warning.
1339 * serial.h (struct ttystate): Declare empty one on DOS.
1343 * xm-sun4os4.h: Return type of free() should be void, not int.
1345 * vx-share/vxWorks.h: Remove #def of NULL.
1349 * tm-rs6000.h: Nuke no-op STAB_REG_TO_REGNUM.
1353 * z8k-tdep.c (print_insn): Include the new dis-asm header file.
1357 * ieee-float.c: Moved to ../libiberty.
1358 * ieee-float.h: Moved to ../include.
1359 * Makefile.in: Update accordingly.
1360 * i386-pinsn.c (print_insn), m68k-pinsn.c (print_insn):
1361 Convert to stubs that call disassemblers in ../opcodes/*-dis.c.
1362 * m68k-tdep.c: Removed definition of ext_format ext_format_68881;
1363 it is now in ../opcodes/m68881-ext.c.
1364 * mips-tdep.c (mips_skip_prologue): Try to skip more of the
1365 prologue (some callers _do_ care).
1366 * mips-pinsn.c (print_insn), z8k-tdep.c (print_insn): Convert to
1367 new interface of ../opcodes/*-dis.c.
1368 * ch-exp.y: Add #include <ctype.h>.
1372 * xcoffexec.c (exec_close): Don't close exec_bfd twice.
1374 * xcoffread.c (enter_line_range): endaddr is exclusive, not inclusive.
1378 * xcoffread.c (arrange_linetable): Use x{m,re}alloc not {m,re}alloc.
1382 * z8k-tdep.c (extract_return_value, write_return_value,
1383 store_struct_return): New functions from macros in tm-z8k.h.
1387 * valops.c (value_arg_coerce): Apply temporary patch to
1388 fix problem with coercion of array and function types when
1389 passed as arguments to C functions, pending a more complete
1390 review of when and how coercion should be done, depending
1391 upon context and language.
1395 * xcoffread.c (MIN_TBTABSIZ): Change to 12.
1397 * xcoffread.c (xcoff_symfile_read): Only read stringtab and
1398 debugsec if there are a non-zero number of symbols.
1402 * command.c (show_user): Avoid fprintf_filtered botch (AGAIN!).
1406 * xcoffexec.c (add_vmap): Wrap symbol read in catch_errors.
1408 * xcoffread.c (read_symbol_lineno): Look to end of symbols for .bf,
1409 not just 50 symbols.
1410 (symtbl_num_syms): New variable.
1411 (read_xcoff_symtab): Set it.
1412 (read_symbol_nvalue): Check for bad symno.
1413 (read_symbol_{lineno,nvalue}, callers): Don't pass symtable; it's
1418 * config/rs6000.mh: Get rid of -Dfd_set=int crock.
1419 This is defined in defs.h if necessary.
1420 * vx-share/vxWorks.h: Remove #defs of min and max.
1421 * vx-share/xdr_ld.c, vx-share/xdr_ptrace.c,
1422 vx-share/xdr_rdb.c: include defs.h.
1426 * xcoffread.c (retrieve_tracebackinfo): Move assignment out
1429 * xcoffread.c (enter_line_range): complain() on bad endoffset.
1430 xcoffread.c: Doc fixes.
1434 * tm-rs6000.h (CORE_NEEDS_RELOCATION): Just call xcoff_relocate_core.
1435 xcoffexec.c (xcoff_relocate_core): New function.
1436 (text_adjustment): Removed.
1437 (add_vmap): Return the vmap.
1438 rs6000-tdep.c (add_text_to_loadinfo): No longer static.
1442 * xcoffsolib.h: Add objfile member to struct vmap.
1443 xcoff{exec,solib}.c: Use it, not lookup_objfile_bfd.
1444 xcoffexec.c (add_vmap): Allocate objfiles here.
1448 Support 68000 series without floating point.
1450 * configure.in (m68000-*-{aout,elf,coff}): New configs.
1451 * tm-68k-nofp.h: New file, lacks 68881 support.
1452 * config/m68k-nofp.mt: New file.
1456 Remove a few remaining underscore/no-underscore remnants from
1459 * config/{m68k-un.mt, sparc-un.mt}: Remove.
1460 * config/m68k-noun.mt: Rename to m68k-fp.mt.
1461 * config/sparc-noun.mt: Rename to sparc-em.mt.
1462 * tm-68k-noun.h, tm-spc-noun.h: Remove.
1463 * tm-68k-un.h: Rename to tm-68k-fp.h.
1464 * tm-spc-un.h: Rename to tm-spc-em.h.
1465 * tm-sun4sol2.h: Cleanup.
1466 * configure.in (m68k-*, sparc-* targets): Corresponding changes.
1470 * symmisc.c (std_in, std_out, std_err): Move initializations
1471 to runtime code, in case they aren't constant.
1475 * symtab.c (find_pc_symtab): some object file formats, notably
1476 mips, have holes in the address ranges of symtabs. Change
1477 this algorithm from first hit to tightest fit.
1479 * mips-tdep.c (heuristic_proc_start): if we walk the pc into the
1480 fence post without finding the enclosing function, then print a
1485 * utils.c (fputs_demangled, fprint_symbol): Remove.
1486 * utils.c (fprintf_symbol_filtered): New function which combines
1487 the functionality of fputs_demangled and fprint_symbol. Uses a
1488 caller provided language parameter to select the appropriate
1489 demangler, and caller provided args to pass to the demangler.
1490 * defs.h (enum language): Move further up in file so enum can
1491 be used in prototypes.
1492 * defs.h (fputs_demangled, fprint_symbol): Remove prototypes.
1493 * defs.h (fprintf_symbol_filtered): Add prototype.
1494 * c-typeprint.c (cp_type_print_method_args): Replace calls to
1495 fputs_demangled with call to fprintf_symbol_filtered.
1496 * cp-valprint.c (demangle.h): Include
1497 * cp-valprint.c (cp_print_value_fields): Replace calls to
1498 fprint_symbol with calls to fprintf_symbol_filtered.
1499 * printcmd.c (print_frame_args): Replace call to fprint_symbol
1500 with call to fprintf_symbol_filtered.
1501 * stack.c (print_frame_info): Remove obsolete code so we don't
1502 have to update fputs_demangled usage in it.
1503 * stack.c (print_frame_info, frame_info): Add language variable
1504 to pass to fprintf_symbol_demangled and initialize it from the
1505 symbol's language. Replace calls to fputs_demangled with calls
1506 to fprintf_symbol_filtered.
1507 * symtab.c (find_methods): Replace call to fputs_demangled with
1508 call to fprintf_symbol_filtered.
1509 * ch-valprint.c (demangle.h): Include.
1510 * ch-valprint.c (chill_print_value_fields): Replace call to
1511 fprint_symbol with call to new fprintf_symbol_filtered.
1515 * Makefile.in (VERSION): Bump version to 4.8.2.
1517 * main.c (source_command): Require an explicit pathname of file
1518 to source, since previous behavior of defaulting to gdb init file
1519 was troublesome and undocumented.
1520 * printcmd.c (disassemble_command): Add missing '{}' pair to
1521 else with two statements. Bug reported by Stephane Tsacas
1523 * symtab.c (find_pc_line): Don't complain about zero length or
1524 negative length line numbers for the moment, since we may not own
1525 the terminal when called, such as when single stepping. (FIXME)
1526 * language.h (CAST_IS_CONVERSION): True if current language is
1527 C++ as well as C. Fix from Peter Schauer.
1528 * environ.c (get_in_environ, set_in_environ, unset_in_environ):
1529 Use STREQN macro rather than bare '!strncmp()'.
1530 * environ.c (unset_in_environ): Avoid use of memcpy on
1531 overlapping memory regions, as suggested by Paul Eggert
1533 * c-exp.y (%union struct): Remove unused ulval as suggested
1538 * main.c (gdbinit): Make static.
1539 * main.c (inhibit_gdbinit): Move to file scope.
1540 * main.c (main): Remove local inhibit_gdbinit.
1541 * main.c (source_command): Don't source '.gdbinit' file by
1542 default if gdb has been told to ignore it.
1546 * Makefile.in (MAKEOVERRIDES): Define to be empty for GNU Make
1551 * printcmd.c (print_address_symbolic): Only print if offset
1552 is shorter than max_symbolic_offset.
1553 (initialize_printcmd): `set print max-symbolic-offset'.
1555 * am29k-tdep.c (TAGWORD_ZERO_MASK): New #define.
1556 (examine_tag): Use it.
1557 (read_register_stack): Only look in the local registers for a
1558 memory address if it's between rfb and rsp; go to memory otherwise.
1559 (initialize_29k): Fix call_scratch_address doc. Remove reginv_com.
1560 (reginv_com): Remove ancient kludge command.
1564 * tm-irix3.h (ZERO_REGNUM): copy this macro from tm-mips.h so that
1565 irix4 will again compile.
1567 * tm-mips.h (GDB_TARGET_IS_MIPS): no longer used, now removed.
1569 * configure.in: accept mips-sgi-irix4* for irix4.
1573 * z8k-tdep.c (print_register_hook): Lint.
1579 * command.c (do_setshow_command): var_uintegers are unsigned.
1580 * sparc-tdep.c (save_insn_opcodes, restore_insn_opcodes):
1581 unsigned, since they use hex values with the high bit set.
1586 * m88k-pinsn.c (sprint_address): Use SYMBOL_NAME macro to
1588 * m88k-nat-c (SXIP_OFFSET, SNIP_OFFSET, SFIP_OFFSET): Enclose
1589 macro definitions in parenthesis.
1591 * dbxread.c (dbx_symfile_init): Catch the case where there is
1592 no string table, but the only way we find out is by reading zero
1597 * dbxread.c (dbx_symfile_init): Make size of the string table
1598 size field a define (DBX_STRINGTAB_SIZE_SIZE). Ensure that the
1599 offset to the string table is nonzero and handle the nonexistant
1600 string table case, should it occur. Ensure that the string table
1601 size read from the file is reasonable, with a minimum lower bound
1602 of DBX_STRINGTAB_SIZE_SIZE instead of zero.
1606 * Makefile.in: Changes to build testsuite correctly.
1607 (FLAGS_TO_PASS): Added CXX and CXXFLAGS.
1608 (CC_FOR_TARGET, CXX, CXX_FOR_TARGET): New variables.
1609 (TARGET_FLAGS_TO_PASS): New variable.
1610 (SUBDIRS): Added testsuite.
1611 (all): Build testsuite using TARGET_FLAGS_TO_PASS, so that
1612 testsuite is compiled with CC_FOR_TARGET rather than CC.
1616 * dbxread.c (dbx_symfile_init): Fix for nonexistant string table,
1620 * configure.in (vax-*-ultrix2*): New triplet.
1621 * config/vaxult2.mh: New file.
1622 * xm-vaxult2.h: New file.
1624 * c-exp.y (parse_number): Change high_bit to unsigned.
1625 * demangle.c: Change all references to cfront to ARM, since the
1626 actual algorithm is the one specified in the Annotated Reference
1627 Manual. This was confusing users into thinking that full cfront
1628 support was implemented.
1629 * dwarfread.c (CFRONT_PRODUCER): Remove, was never really used.
1630 * eval.c (evaluate_subexp): For STRUCTOP_PTR pass the arg type
1631 directly to lookup_struct_elt_type, which will do the
1632 dereferencing itself.
1633 * gdbtypes.c (lookup_struct_elt_type): Expand comments. Fix
1634 NULL dereferencing bug for unnamed structs, comment out
1639 * coffread.c (process_coff_symbol): Change PCC argument correction
1640 so that it only happens on big-endian targets; so that it only
1641 happens if the short or char argument is aligned on an int
1642 boundary; and so that it changes the location, rather than the
1643 type, of the argument. These changes tend to parallel similar
1644 (old) changes in stabsread.c.
1646 * coffread.c (coff_read_enum_type): Use the specified size for
1647 enums, don't assume that they are int-sized.
1649 * c-valprint.c (c_val_print): Don't assume enums are the same as
1652 * coredep.c: Handle NO_PTRACE_H in coredep.c. Fix by Michael
1657 * language.h (local_decimal_format_custom): Add prototype.
1658 * language.c (local_decimal_format_custom): Add function, bug
1663 * xcoffexec.c (vmap_ldinfo): Fix "/" for '/' typo, reported
1668 * symfile.c (syms_from_objfile), tm-29k.h, tm-3b1.h, tm-68k-un.h,
1669 tm-altos.h, tm-arm.h, tm-convex.h, tm-es1800.h, tm-h8300.h,
1670 tm-hp300bsd.h, tm-hp300hpux.h, tm-hppa.h, tm-i386bsd.h,
1671 tm-i386v.h, tm-i960.h, tm-irix3.h, tm-isi.h, tm-linux.h,
1672 tm-m88k.h, tm-merlin.h, tm-mips.h, tm-news.h, tm-np1.h, tm-pn.h,
1673 tm-pyr.h, tm-rs6000.h, tm-spc-un.h, tm-sun386.h, tm-sunos.h,
1674 tm-symmetry.h, tm-sysv4.h, tm-tahoe.h, tm-umax.h, tm-vax.h,
1675 tm-vx68.h, tm-z8k.h: Remove remnants of NAMES_HAVE_UNDERSCORE.
1679 * symtab.h (SYMBOL_INIT_DEMANGLED_NAME): Initialize contents
1680 of demangled name fields to NULL if no demangling exists for
1681 a symbol. SYMBOL_INIT_LANGUAGE_SPECIFIC does this for new
1682 symbols if their language is known at the time they are created,
1683 but sometimes the language is not known until later.
1685 * ch-typeprint.c (chill_print_type_base): Name changed to
1686 chill_type_print_base to match pattern for C and C++ names.
1687 * ch-typeprint.c (chill_print_type): Change "char" to "CHAR"
1688 to be consistent with other usages.
1689 * ch-typeprint.c (chill_type_print_base): Add support for
1690 printing Chill STRUCT types.
1691 * ch-valprint.c: Include values.h.
1692 * ch-valprint.c (chill_print_value_fields): New function and
1693 prototype for printing Chill STRUCT values.
1694 * ch-valprint.c (chill_val_print): Fix call to val_print_string
1695 that was being called with two args instead of three.
1696 * ch-valprint.c (chill_val_print): Call chill_print_value_fields
1697 to print Chill STRUCT values.
1701 * configure.in: added testsuite to configdirs.
1705 * doc/stabs.texi: The `this' pointer is now known by the name
1706 `this' instead of `$t'.
1710 * dwarfread.c (read_tag_string_type): Rewrite to allow forward
1711 references of TAG_string_type DIEs in user defined types.
1712 * ch-lang.c (chill_create_fundamental_type): Track compiler
1713 change that now emits debugging info with the type long for Chill
1718 * remote-mips.c: New file; implements MIPS remote debugging
1720 * config/idt.mt: New file; uses remote-mips.c
1721 * configure.in (mips-idt-ecoff): New target; uses idt.mt.
1723 * mips-tdep.c (mips_fpu): New variable; controls use of MIPS
1724 floating point coprocessor.
1725 (mips_push_dummy_frame): If not mips_fpu, don't save floating
1727 (mips_pop_frame): If not mips_fpu, don't restore floating point
1729 (_initialize_mips_tdep): New function; let the user reset mips_fpu
1731 * tm-mips.h (EXTRACT_RETURN_VALUE, STORE_RETURN_VALUE): If not
1732 mips_fpu, don't use fp0 as floating point return register.
1733 (FIX_CALL_DUMMY): If not mips_fpu, don't save floating point
1738 * gdb/testsuite: made modifications to testcases, etc., to allow
1739 them to work properly given the reorganization of deja-gnu and the
1740 relocation of the testcases from deja-gnu to a "tool" subdirectory.
1744 * gdb/testsuite: Initial creation of gdb/testsuite.
1745 Migrated dejagnu testcases and support files for testing nm to
1746 gdb/testsuite from deja-gnu. These files were moved "as is"
1747 with no modifications. This migration is part of a major overhaul
1748 of dejagnu. The modifications to these testcases, etc., which
1749 will allow them to work with the new version of dejagnu will be
1750 made in a future update.
1754 * NEWS: Add reminders for next release.
1758 * mipsread.c (parse_lines): Correct check for files compiled with
1763 * Makefile.in (VERSION): 4.8.1 to distinguish local versions.
1767 * Makefile.in (VERSION): GDB-4.8 release!
1768 * README, NEWS: Update for release.
1772 * am29k-pinsn.c (print_insn): Minor nits with const.
1773 * am29k-tdep.c: More minor nits with arg types for
1774 supply_register, NULL vs. 0, read_register_gen, & reginv_com.
1778 * gcc.patch: Update for a different GCC (G++) bug.
1779 * main.c (print_gdb_version): Update copyright year to 1993.
1780 * nm-hp300bsd.h: Decide whether this is BSD 4.3 or 4.4,
1781 conditionalize this file on it. FIXME, right way is to split
1782 these into two config files.
1783 (ATTACH_DETACH): Define for BSD 4.4
1784 (PTRACE_ARG_TYPE): caddr_t for BSD 4.4, unset for 4.3.
1785 (U_REGS_OFFSET): Revise for 4.4.
1786 (REGISTER_U_ADDR): Separate for 4.4, but it doesn't work yet.
1787 * xm-hp300bsd.h: Move definitions of UINT_MAX, INT_MAX, INT_MIN,
1788 LONG_MAX into this file to avoid cpp "redefinition" warnings.
1792 * nm-hp300bsd.h (PTRACE_ARG3_TYPE): FSF's hp300's have int* not
1797 * c-lang.c (c_printstr): Bugfix for length==0 case.
1799 * c-lang.c (c_printstr): If a C string ends in a null, don't
1802 Thu Feb 18 02:39:21 1993 Stu Grossman (grossman at cygnus.com)
1804 * defs.h (STRCMP): Make it work for unsigned chars.
1808 * nm-hp300bsd.h (ATTACH_DETACH, PTRACE_ATTACH, PTRACE_DETACH): define.
1809 * config/hp300bsd.mh (REGEX, REGEX1): Define.
1810 * m68k-pinsn.c (BREAK_UP_BIG_DECL, AND_OTHER_PART): #if __GNUC__,
1811 define to kludge the large opcode table into two smaller tables,
1812 since GCC take exponential space to build the table. Lint.
1813 (NOPCODES): Remove, use "numopcodes" from opcode/m68k.h instead.
1817 * Makefile.in (VERSION): Roll to 4.7.9.
1818 * xm-hp300bsd.h: Define PSIGNAL_IN_SIGNAL_H and put a compatible
1819 definition here, to handle both BSD 4.3 and 4.4 systems.
1820 * mipsread.c (ZMAGIC): #undef to avoid duplicate define.
1821 * remote.c (alarm): Move declaration to global level, before
1822 first reference to it.
1823 * tm-i386bsd.h (NUM_REGS): There are only eleven, not twelve.
1824 * dbxread.c (process_one_symbol): Cast to unsigned char, not int.
1828 * remote.c (readchar): forward declare alarm which otherwise looks
1829 like an undeclared variable to gcc.
1831 * dbxread.c (process_one_symbol): cast enum value N_SO into int
1832 when comparing against an int. Avoids superfluous warning from
1835 * inflow.c (set_sigint_trap): add cast to assignment from signal.
1836 Avoids superfluous warnings from some systems and/or compilers
1837 (like vax ultrix 4.2.)
1839 * language.c (struct op_print unk_op_print_tab): use the enum
1840 values rather naked zeros as initializers. Avoids warnings from
1841 ultrix type compilers.
1845 * Makefile.in (VERSION): Roll to 4.7.6.
1846 (SFILES_SUBDIR): Add 29k-share/udi_soc.
1847 (SFILES_SUBSUBDIR): Move 29k-share/udi files to this macro.
1848 (alldeps.mak): Make ALLDEPFILES_SUBSUBDIR for files in sub sub dirs.
1849 (ALLDEPFILES_SUBSUBDIR): Depend on this for deeper dep files.
1850 (HFILES): Remove all nm-* except nm-trash.h. Add ns32k-opcode.h.
1851 (depend): Fix bug where nm-files in config files weren't noticed.
1852 (make-proto-gdb-1): Avoid changing directories while building new
1853 prototype. Build SFILES_SUBSUBDIR with longer symlinks.
1857 * remote.c: Improve error recovery. Allow user to break out
1858 of initial connection attempt with INTERRUPT. Treat a timeout
1859 while waiting for remote packet like a retry, unless the remote
1860 side is actively running user code. Fix a few long printf_filtered's.
1862 * xcoffread.c (read_xcoff_symtab): Don't use null symbol name for
1865 * buildsym.c (start_subfile): Allow null file name.
1869 * xcoffread.c (process_xcoff_symbol, read_symbol_lineno): complain
1870 expects a pointer to complaint rather than a complaint
1872 (process_linenos): free the previously allocated subfile name,
1873 then allocate the new one from the heap.
1877 * h8300-tdep.c, tm-h8300.h: turn off some experimental features
1881 * stabsread.c (dbx_lookup_type): Handle negative type numbers.
1882 Previously, would bogusly index off the bottom of type_vector.
1883 (rs6000_builtin_type): Accept type number as argument.
1884 (read_type, case '-'): Handle negatives like any other type number.
1886 * symfile.c (deduce_language_from_filename): Handle null name.
1888 * mips-tdep.c (isa_NAN): Fix byte order dependency.
1892 * xcoffread.c (parmsym): Don't use an initializer to set up
1893 this struct symbol. Set it up in initialize_xcoffread.
1894 (read_xcoff_symtab, xcoff_symfile_read): Surround code that only
1895 works on real rs/6000 target with #ifndef FAKING_RS6000.
1899 * stabsread.c (rs6000_builtin_type): Move function from
1900 xcoffread.c:builtin_type.
1901 * xcoffread.c (builtin_type): Move to stabsread. Remove
1902 IBM6000_HOST dependency. Move misplaced comments.
1903 (various): Change printf's to complaints.
1904 (patch_block_stabs, process_xcoff_symbol case C_DECL): Add
1905 objfile argument to read_type calls under #if 0.
1906 (process_xcoff_symbol case C_RSYM): Fix typo in #ifdef.
1907 * xcoffexec.c (map_vmap): Don't allocate an objfile for the exec_file.
1908 * Makefile.in: xcoffread.o is not built by default.
1909 * xm-rs6000.h (IBM6000_HOST): Remove.
1910 * config/rs6000.mh (NATDEPFILES): xcoffread.o is native only.
1911 * doc/gdbint.texinfo: Eliminate IBM6000_HOST, document
1914 Wed Feb 10 18:31:20 1993 Stu Grossman (grossman at cygnus.com)
1916 * findvar.c (read_var_value): If REG_STRUCT_HAS_ADDR, then set
1917 VALUE_LVAL to be lval_memory so that we don't try to modify wild
1918 register numbers when user tries to modify elements in structs
1919 passed as arguments.
1920 * inflow.c (child_terminal_info): Move banner outside of system
1922 * tm-hppa.h (REG_STRUCT_HAS_ADDR): Define this for HPPA, which
1923 passes struct/union arguments by address.
1928 * nm-i386sco4.h: New file. Like nm-i386sco.h, but define
1929 ATTACH_DETACH, PTRACE_ATTACH and PTRACE_DETACH.
1930 * config/i386sco4.mh (NAT_FILE): Use nm-i386sco4.h.
1934 * remote-udi.c (FREEZE_MODE): Fix && for & typo. Found and
1939 * config/i386sco4.mh (MUNCH_DEFINE): Pass -p to nm to avoid bug in
1940 cc debugging output.
1944 * stabsread.c (define_symbol): Complain about unrecognized names
1945 that begin with CPLUS_MARKER (often '$'), but don't die. Fix
1947 (read_cpp_abbrev): Don't use the class name as part of the
1948 vtable pointer member name (_vptr$) in $vf abbrevs or unrecognized
1949 abbrevs. Inspired by Mike Tiemann.
1950 (read_tilde_fields): Comment. Remove ancient dead code.
1951 Remove erroneous but non-dead code. Simplify. Add complaints.
1952 (in general): Remove extraneous (parentheses) in return
1957 * coffread.c (coff_lookup_type): Fix fencepost error reported
1960 Fix long file name bug reported on SCO Open Desktop 2.0 by Ulf Lunde
1963 * coffread.c (getfilename): Eliminate COFF_NO_LONG_FILE_NAMES
1964 test, which is apparently left over from when we used native
1965 include files and couldn't depend on the member names being there.
1966 * tm-3b1.h, tm-altos.h, tm-i386v.h: Don't set it.
1970 * mipsread.c: Major overhaul to use new BFD symbol table reading
1971 routines. Now swaps information as it is needed, rather than
1972 swapping everything when the file is read.
1976 * Makefile.in (TARDIRS): Add sparclite demo dir.
1977 (*.tab.c): Change dependency on Makefile to depend on
1978 Makefile.in, otherwise it always rebuilds after configuring.
1979 Force output *.tab.c file into current directory even in "make"
1980 versions that rewrite dependent file names used in command lines.
1982 * TODO: Remove some things we did.
1983 * am29k-opcode.h, convx-opcode: Remove; now in ../include/opcode.
1984 * os68k-xdep.c: Remove; useless file (os68k is a target only).
1985 * convex-pinsn.c: Use ../include/opcode/convex.h. Add CONST.
1986 * symtab.h: Eliminate unnamed unions and structs.
1990 * Makefile.in (VERSION): Roll to 4.7.5.
1994 * breakpoint.c (breakpoint_re_set_one): Handle watchpoints when
1995 re-evaluating symbol pointers.
1999 * c-exp.y (lcurly, rcurly): New nonterminals.
2000 * c-exp.y (exp): Use lcurly and rcurly for arrays and UNOP_MEMVAL
2002 * parse.c (free_funcalls): Moved prototype from parser-defs.h,
2003 made function static.
2004 * parse.c (struct funcall): Moved struct def from parser-defs.h.
2005 * parse.c (funcall_chain): Moved from parser-defs.h, made static.
2006 * parse.c (start_arglist):
2007 * parser-defs.h (free_funcalls): Moved prototype to parse.c.
2008 * parser-defs.h (struct funcall): Moved struct def to parse.c.
2009 * parser-defs.h (funcall_chain): Moved to parse.c.
2010 * printcmd.c (print_frame_nameless_args): Fix prototype.
2011 * tm-mips.h (setup_arbitrary_frame): Fix prototype.
2012 * tm-sparc.h (setup_arbitrary_frame): Fix prototype.
2013 * valops.c (typecmp): Moved prototype from values.h.
2014 * value.h (typecmp): Moved prototype to valops.c, made static.
2015 * ch-exp.y (yylex): Change way control sequences are disabled.
2019 * tm-mips.h, tm-sparc.h: Fix thinko in SETUP_ARBITRARY_FRAME.
2023 * mipsread.c (upgrade_type): Build array types correctly, using
2024 create_range_type and create_array_type.
2028 * remote-nindy.c: Cleanup.
2030 * infrun.c (wait_for_inferior): When rolling back the PC after
2031 a breakpoint, call write_pc so that NPC gets rolled back as well
2034 * blockframe.c (inside_entry_file, inside_main_func,
2035 inside_entry_func): PC of zero is always "bottom of stack".
2037 * printcmd.c (print_frame_args, print_frame_nameless_args):
2038 Let print_frame_nameless_args decide whether there are any,
2039 laying groundwork for possibly later printing 29K args for
2040 functions where we have tag words but no symbols.
2044 * Makefile.in: fix GDB doc targets for new doc subdir structure
2048 * stack.c (parse_frame_specification): Parse as many arguments
2049 as there are (up to MAXARGS). Pass all of them in argc, argv
2050 format to SETUP_ARBITRARY_FRAME. Put the burden of checking how
2051 many there were, onto SETUP_ARBITRARY_FRAME.
2052 * tm-mips.h, tm-sparc.h: Corresponding changes.
2053 * mips-tdep.c, sparc-tdep.c: Ditto.
2057 * hp300ux-nat.c: Update copyrights.
2058 * mipsread.c (parse_partial_symbols): Complain about block
2059 indexes that go backwards. Fix from Peter Schauer.
2060 * symfile.c (syms_from_objfile, symbol_file_add): Allow a
2061 symbol-file that has no linkage symbols to be read.
2062 * tm-rs6000.h, xm-rs6000.h: (SIGWINCH_HANDLER and friends): Move
2063 from tm- file to xm-file, since they're host dependent.
2064 * valarith.c (value_binop): Typo.
2066 Mon Feb 1 16:16:59 1993 Stu Grossman (grossman at cygnus.com)
2068 * sparclite/aload.c: Add copyleft.
2069 * sparclite/crt0.s: Add comment at beginning.
2073 * remote-z8k.c, z8k-tdep.c: support for the Z8001 and Z8002.
2074 * parse.c (std_regs): Only declare if NO_STD_REGS is defined.
2078 * values.c (value_headof): Fix typo in which VTBL and ARG were
2079 being confused for one another.
2081 * valops.c (typecmp): Now static.
2083 * gdbtypes.c (fill_in_vptr_fieldno): Don't ignore the first
2084 baseclass--we don't always inherit its virtual function table
2087 * eval.c (evaluate_subexp): In OP_FUNCALL case, adjust `this'
2088 pointer correctly in case value_struct_elt moves it around.
2090 * valops.c (typecmp): Now static. Also, now groks references
2093 * gdbtypes.c (lookup_struct_elt_type): Pass NOERR instead of
2094 zero on recursive call. If NAME is the name of TYPE, return TYPE.
2098 * hppah-nat.c: Eliminate <sys/user.h> and other unnecessary stuff,
2099 to avoid "too much defining" error from native C compiler (!).
2101 * Makefile.in (HFILES): Add typeprint.h.
2102 * typeprint.[ch]: Update copyrights.
2106 * Makefile.in: Update to match doc/ subdir changes.
2108 * config/hp300hpux.mh: No cross-host file needed, just native.
2109 * config/go32.mh: Remove nonexistent "native" support.
2112 * tm-m88k.h (REGISTER_CONVERT_TO_RAW): Fix typo.
2113 * m88k-tdep.c (next_insn): Lint, cleanup.
2114 (store_parm_word): Lint.
2120 * expression.h (BINOP_CONCAT): Document use for self concatenation
2121 an integral number of times.
2122 * language.c (binop_type_check): Extend BINOP_CONCAT for self
2124 * valarith.c (value_concat): Rewrite to support self
2125 concatenation an integral number of times.
2126 * Makefile.in (ch-exp.tab.c): Change "expect" message.
2127 * ch-exp.y (FIXME's): Make all FIXME tokens distinct, to
2128 eliminate hundreds of spurious shift/reduce and reduce/reduce
2129 conflicts that mask the 5 real ones.
2130 * ch-exp.y (STRING, CONSTANT, SC): Remove unused tokens.
2131 * ch-exp.y (integer_literal_expression): Remove production,
2136 * eval.c (evaluate_subexp): Fix OP_ARRAY, remove code that
2137 implied that "no side effects" was nonfunctional.
2138 * eval.c (evaluate_subexp): Add BINOP_CONCAT case to deal with
2139 character string and bitstring concatenation.
2140 * expprint.c (dump_expression): Add case for BINOP_CONCAT.
2141 * expression.h (exp_opcode): Add BINOP_CONCAT.
2142 * gdbtypes.h (type_code): Add TYPE_CODE_BITSTRING.
2143 * language.c (string_type): Add function to determine if a type
2145 * language.c (binop_type_check): Add case for BINOP_CONCAT.
2146 * valarith.c (value_concat): New function to concatenate two
2147 values, such as character strings or bitstrings.
2148 * valops.c (value_string): Remove error stub and implement
2150 * value.h (value_concat): Add prototype.
2151 * ch-exp.y (operand_3): Add actions for SLASH_SLASH (//).
2152 * ch-exp.y (yylex): Recognize SLASH_SLASH.
2153 * ch-lang.c (chill_op_print_tab): Add SLASH_SLASH (//) as
2158 * c-exp.y (exp): Add production to support direct creation
2159 of array constants using the obvious syntax.
2160 * c-valprint.c (c_val_print): Set printed string length.
2161 * dwarfread.c (read_tag_string_type): New prototype and
2162 function that handles TAG_string_type DIEs.
2163 * dwarfread.c (process_dies): Add case for TAG_string_type
2164 that calls new read_tag_string_type function.
2165 * expprint.c (print_subexp): Add support for OP_ARRAY.
2166 * gdbtypes.c (create_range_type, create_array_type): Inherit
2167 objfile from the index type.
2168 * ch-typeprint.c (chill_print_type): Add case for
2170 * ch-valprint.c (chill_val_print): Fix case for
2175 * mipsread.c (CODE_MASK, MIPS_IS_STAB, MIPS_MARK_STAB,
2176 MIPS_UNMARK_STAB, STABS_SYMBOLS): Removed; now in
2177 include/coff/mips.h.
2181 * c-exp.y (exp:STRING): Convert C strings into array-of-char
2182 constants with an explicit null byte terminator. OP_STRING is
2183 now used for real string types.
2184 * c-lang.c (builtin_type_*): Move declarations to lang.c since
2185 they are used by all languages.
2186 * c-lang.c (_initialize_c_language): Move initializations of
2187 builtin_type_* to lang.c.
2188 * c-typeprint.c (c_type_print_varspec_prefix,
2189 c_type_print_varspec_suffix): TYPE_CODE_PASCAL_ARRAY renamed
2190 to TYPE_CODE_STRING.
2191 * c-valprint.c (c_val_print): Change the way character arrays
2192 are printed as strings to be consistent with the way strings
2193 are printed when pointer-to-char types are dereferenced.
2194 Remove test of print_max before calling val_print_string, which
2195 now does it's own test.
2196 * eval.c (evaluate_subexp): Add case for OP_ARRAY.
2197 * expprint.c (print_subexp, dump_expression): Add case for OP_ARRAY.
2198 * expression.h (enum exp_opcode): Add OP_ARRAY and document.
2199 * gdbtypes.c (builtin_type_*): Add declarations moved from
2201 * gdbtypes.c (create_string_type): New function to create real
2203 * gdbtypes.c (recursive_dump_type): TYPE_CODE_PASCAL_ARRAY
2204 renamed to TYPE_CODE_STRING.
2205 * gdbtypes.c (_initialize_gdbtypes): Add initializations of
2206 builtin_type_* types moved from c-lang.c.
2207 * gdbtypes.h (enum type_code): TYPE_CODE_PASCAL_ARRAY renamed
2208 to TYPE_CODE_STRING.
2209 * gdbtypes.h (builtin_type_string): Add extern declaration.
2210 * gdbtypes.h (create_string_type): Add prototype.
2211 * m2-lang.c (m2_create_fundamental_type): TYPE_CODE_PASCAL_ARRAY
2212 renamed to TYPE_CODE_STRING.
2213 * m88k-tdep.c (pushed_size): TYPE_CODE_PASCAL_ARRAY renamed to
2215 * mipsread.c (_initialize_mipsread): TYPE_CODE_PASCAL_ARRAY
2216 renamed to TYPE_CODE_STRING.
2217 * parse.c (length_of_subexp, prefixify_subexp): Add case for
2219 * printcmd.c (print_formatted): Recognize TYPE_CODE_STRING.
2220 * typeprint.c (print_type_scalar): TYPE_CODE_PASCAL_ARRAY renamed
2221 to TYPE_CODE_STRING.
2222 * valops.c (allocate_space_in_inferior): New function and
2223 prototype, using code ripped out of value_string.
2224 * valops.c (value_string): Rewritten to use new function
2225 allocate_space_in_inferior, but temporarily disabled until some
2226 other support is in place.
2227 * valops.c (value_array): New function to create array constants.
2228 * valprint.c (val_print_string): Add comment to document use,
2229 complete rewrite to fix several small buglets.
2230 * value.h (value_array): Add prototype.
2231 * value.h (val_print_string): Change prototype to match rewrite.
2232 * ch-valprint.c (chill_val_print): Add case for TYPE_CODE_STRING.
2233 * ch-exp.y (match_character_literal): Disable recognition of
2234 control sequence form of character literals and document why.
2236 Thu Jan 14 15:48:12 1993 Stu Grossman (grossman at cygnus.com)
2238 * nindy-share/nindy.c: Add comments to #endif's to clarify
2241 * hppa-pinsn.c (print_insn): Use read_memory_integer, instead of
2242 read_memory to get byte order right.
2243 * hppah-tdep.c (find_unwind_info): Don't read in unwind info
2244 anymore. This is done in paread.c now. We expect unwind info
2245 to hang off of objfiles, and search all of the objfiles when until
2247 * (skip_trampoline_code): Cast arg to target_read_memory.
2248 * objfiles.h (struct objfile): Add new field obj_private to hold
2249 per object file private data (unwind info in this case).
2250 * paread.c (read_unwind_info): New routine to read unwind info
2251 for the objfile. This data is hung off of obj_private.
2252 * tm-hppa.h: Define struct obj_unwind_info, to hold pointers to
2253 the unwind info for this objfile. Also define OBJ_UNWIND_INFO to
2254 make this easier to access.
2258 * c-valprint.c (cp_print_class_member): Add extern decl.
2259 * c-valprint.c (c_val_print): Extract code for printing methods
2260 and move it to cp_print_class_method in cp-valprint.c.
2261 * c-valprint.c (c_val_print): Extract code to print strings and
2262 move it to val_print_string in valprint.c.
2263 * cp-valprint.c (cp_print_class_method): New function using
2264 code extracted from c_val_print.
2265 * valprint.c (val_print_string): New function using code
2266 extracted from c_val_print.
2267 * value.h (val_print_string): Add prototype.
2268 * ch-exp.y (CHARACTER_STRING_LITERAL): Set correct token type.
2269 * ch-exp.y (literal): Add action for CHARACTER_STRING_LITERAL.
2270 * ch-exp.y (tempbuf, tempbufsize, tempbufindex, GROWBY_MIN_SIZE,
2271 CHECKBUF, growbuf_by_size): New variables, macros, and support
2272 functions for implementing a dynamically expandable temp buffer.
2273 * ch-exp.y (match_string_literal): New lexer function.
2274 * ch-exp.y (match_bitstring_literal): Dynamic buffer code
2275 removed and replaced with new CHECKBUF macro.
2276 * ch-exp.y (yylex): Call match_string_literal when appropriate.
2277 * ch-valprint.c (ch_val_print): Add code for TYPE_CODE_PTR.
2279 Sat Jan 9 19:59:33 1993 Stu Grossman (grossman at cygnus.com)
2281 * Makefile.in: Add info for paread.o.
2282 * config/hppahpux.mh: Add paread.o to NATDEPFILES.
2284 * blockframe.c (frameless_look_for_prologue): Correct the
2286 * gdbtypes.h, gdbtypes.c: Use const in decl of
2287 cplus_struct_default, now that pa-gas assembler has been fixed.
2288 * hppah-nat.c: Formatting.
2289 * hppah-tdep.c: Remove lots of useless externs for variables we
2291 * (find_unwind_entry): Speed up by using binary search, and a one
2293 * (rp_saved): New routine to see what unwind info says about RP
2294 being saved on the stack frame.
2295 * (frame_saved_pc): Look for prologue to see if we need to
2296 examine the stack for the saved RP or not.
2297 * (init_extra_frame_info): Check for prologue, instead of
2298 framesize to determine if we are frameless or not.
2299 * (frame_chain_valid): Stop backtraces when we run into _start.
2300 * (push_dummy_frame): Reformat to make more readable.
2301 * (find_dummy_frame_regs): ditto.
2302 * (hp_pop_frame): ditto.
2303 * (hp_restore_pc_queue): small cleanup.
2304 * (hp_push_arguments): ditto.
2305 * (pa_do_registers_info): ditto.
2306 * (skip_prologue): New routine created from SKIP_PROLOGUE macro.
2307 * tm-hppa.h: Move contents of SKIP_PROLOGUE into hppah-tdep.c.
2308 * Define FRAME_CHAIN_VALID.
2309 * Turn on BELIEVE_PCC_PROMOTION so that we can access char args
2310 passed to functions.
2312 * paread.c (pa_symtab_read): Use new bfd conventions for
2313 accessing linker symbol table.
2314 * (pa_symfile_init): Access embedded STAB info via BFD section
2315 mechanism and related macros.
2318 Sat Jan 9 19:31:43 1993 Stu Grossman (grossman at cygnus.com)
2320 * sparc-stub.c: Use a seperate stack for our traps.
2321 * Handle recursive traps.
2322 * Remove all trap init code. This needs to be done by the
2324 * (set_mem_fault_trap): Call exceptionHandler() to setup this
2326 * (handle_exception): See if we are at breakinst, if so, then
2327 advance PC sp that users can just step out of breakpoint().
2328 * (case 'G'): Don't let GDB hack CWP. Also, copy saved regs to
2329 new place if SP has changed.
2330 * (case 's'): Get rid of this, we can't do it yet.
2331 * (case 't'): New command to test any old random feature.
2332 * (case 'r'): New command to reset the system.
2333 * (breakpoint): Add label to breakpoint trap instruction so that
2334 handle_exception() can detect where we are and get past the
2335 breakpoint trivially.
2339 * mips-pinsn.c: Actual work now done by opcodes/mips-dis.c.
2341 Thu Jan 7 09:21:51 1993 david d `zoo' zuhn (zoo at cirdan.cygnus.com)
2343 * configure.in: recognise all sparclite variants
2347 * symfile.c: If O_BINARY isn't defined, set it to 0, call openp for
2348 binary files oring in the right bit.
2350 * main.c, source.c, state.c, symmisc.c: use macros defined in
2351 fopen-{bin|both} when fopening files.
2355 * defs.h (HOST_CHAR_BIT): New macro, defaults to either CHAR_BIT
2356 from a configuration file (typically including <limits.h>), or to
2357 TARGET_CHAR_BIT if CHAR_BIT is not defined.
2358 * eval.c (evaluate_subexp): Use new BYTES_TO_EXP_ELEM macro.
2359 * eval.c (evaluate_subexp): Add case for OP_BITSTRING.
2360 * expprint.c (print_subexp): Use new BYTES_TO_EXP_ELEM macro.
2361 * exppritn.c (print_subexp, dump_expression): Add case for
2363 * expression.h (OP_BITSTRING): New expression element type for
2365 * expression.h (EXP_ELEM_TO_BYTES, BYTES_TO_EXP_ELEM): New
2366 macros to convert between number of expression elements and bytes
2367 to store that many elements.
2368 * i960-tdep.c (leafproc_return): Use new macros to access
2369 minimal symbol name and address fields.
2370 * m88k-pinsn.c (sprint_address): Use new macros to access
2371 minimal symbol name and address fields.
2372 * nindy-tdep.c (nindy_frame_chain_valid): Use new macro to access
2373 minimal symbol address field.
2374 * parse.c (write_exp_elt, write_exp_string, prefixify_expression,
2375 parse_exp_1): Use new EXP_ELEM_TO_BYTES macro.
2376 * parse.c (write_exp_string, length_of_subexp, prefixify_expression):
2377 Use new BYTES_TO_EXP_ELEM macro.
2378 * parse.c (write_exp_bitstring): New function to write packed
2379 bitstrings into the expression element vector.
2380 * parse.c (length_of_subexp, prefixify_subexp): Add case for
2382 * parser-defs.h (struct stoken): Document that it is used for
2383 OP_BITSTRING as well as OP_STRING.
2384 * parser-defs.h (write_exp_bitstring): Add prototype.
2385 * ch-exp.y (BIT_STRING_LITERAL): Change token type to sval.
2386 * ch-exp.y (NUM, PRED, SUCC, ABS, CARD, MAX, MIN, SIZE, UPPER,
2387 LOWER, LENGTH): New tokens for keywords.
2388 * ch-exp.y (chill_value_built_in_routine_call, mode_argument,
2389 upper_lower_argument, length_argument, array_mode_name,
2390 string_mode_name, variant_structure_mode_name): New non-terminals
2392 * ch-exp.y (literal): Useful production for BIT_STRING_LITERAL.
2393 * ch-exp.y (match_bitstring_literal): New lexer support function
2394 to recognize bitstring literals.
2395 * ch-exp.y (tokentab6): New token table for 6 character keywords.
2396 * ch-exp.y (tokentab5): Add LOWER, UPPER.
2397 * ch-exp.y (tokentab4): Add PRED, SUCC, CARD, SIZE.
2398 * ch-exp.y (tokentab3): Add NUM, ABS, MIN, MAX.
2399 * ch-exp.y (yylex): Check tokentab6.
2400 * ch-exp.y (yylex): Call match_bitstring_literal.
2404 * xcoffexec.c (vmap_symtab): Use new macros to access minimal
2405 symbol name and value fields.
2407 * c-exp.y (yylex): Make static, to match prototype and other
2410 * expression.h (exp_opcode): Add BINOP_MOD.
2411 * eval.c (evaluate_subexp): Handle new BINOP_MOD.
2412 * expprint.c (dump_expression): Handle new BINOP_MOD.
2413 * language.c (binop_type_check): Handle new BINOP_MOD.
2414 * main.c (float_handler): Re-enable float handler when hit.
2415 * valarith.c (language.h): Include, need current_language.
2416 * valarith.c (TRUNCATION_TOWARDS_ZERO): Define default macro
2417 for integer divide truncates towards zero for negative results.
2418 * valarith.c (value_x_binop): Handle BINOP_MOD if seen.
2419 * valarith.c (value_binop): Allow arithmetic operations on
2420 TYPE_CODE_CHAR variables. Add case to handle new BINOP_MOD.
2421 * ch-exp.y (operand_4): Add useful actions for MOD and REM.
2422 * ch-exp.y (tokentab3): Add MOD and REM.
2423 * ch-exp.y (yylex): Set innermost_block for symbols found
2424 in local scopes. Return LOCATION_NAME for local symbols.
2425 * ch-lang.c (chill_op_print_tab): Fix MOD entry to use
2426 BINOP_MOD instead of BINOP_REM. Add REM entry, using BINOP_REM.
2430 * command.c (shell_escape, make_command, _initialize_command):
2431 don't create or use fork if CANT_FORK is defined.
2432 * serial.h, ser-go32.c: now compiles, but "the obvious problems of
2433 code written for the IBM PC" remain.
2434 * xm-go32.h: define CANT_FORK
2438 * remote-sim.c: first attempt at general simulator interface
2439 * remote-hms.c: whitespace
2440 * h8300-tdep.c: (h8300_skip_prologue, examine_prologue):
2441 understand new stack layout. (print_register_hook): print ccr
2442 register in a fancy way.
2446 * eval.c (language.h): Include.
2447 * eval.c (evaluate_subexp_with_coercion): Only coerce arrays
2448 to pointer types when the current language is C. It loses for
2449 other languages when the lower index bound is nonzero.
2450 * valarith.c (value_subscript): Take array lower bounds into
2451 account when performing subscripting operations.
2452 * valops.c (value_coerce_array): Add comment describing why
2453 arrays with nonzero lower bounds are dealt with in value_subscript,
2454 rather than in value_coerce_array.
2458 * ch-exp.y (FLOAT_LITERAL): Add token.
2459 * ch-exp.y (literal): Add FLOAT_LITERAL.
2460 * ch-exp.y (match_float_literal): New lexer routine.
2461 * ch-exp.y (convert_float): Remove.
2462 * ch-exp.y (yylex): Call match_float_literal.
2463 * ch-exp.y (yylex): Match single '.' after trying
2464 to match floating point literals.
2466 * eval.c (evaluate_subexp): Add case MULTI_SUBSCRIPT.
2467 * expprint.c (print_subexp): Rename BINOP_MULTI_SUBSCRIPT to
2469 * expprint.c (dump_expression): New function for dumping
2470 expression vectors during gdb debugging.
2471 * expression.h (BINOP_MULTI_SUBSCRIPT): Name changed to
2472 MULTI_SUBSCRIPT and moved out of BINOP range.
2473 * expression.h (DUMP_EXPRESSION): New macro that calls
2474 dump_expression if DEBUG_EXPRESSIONS is defined.
2475 * m2-exp.y (BINOP_MULTI_SUBSCRIPT): Changed to MULTI_SUBSCRIPT.
2476 * parse.c (length_of_subexp, prefixify_subexp): Change
2477 BINOP_MULTI_SUBSCRIPT to MULTI_SUBSCRIPT.
2478 * parse.c (parse_exp_1): Call DUMP_EXPRESSION before and after
2479 prefixify'ing the expression.
2480 * printcmd.c (print_command_1): Add comment.
2481 * ch-exp.y (expression_list): Add useful actions.
2482 * ch-exp.y (value_array_element): Add useful actions.
2483 * ch-exp.y (array_primitive_value): Add production.
2484 * ch-exp.y (yylex): Recognize ',' as a token.
2486 Fri Jan 1 18:22:02 1993 david d `zoo' zuhn (zoo at cirdan.cygnus.com)
2488 * Makefile.in: pass prefix and exec_prefix via FLAGS_TO_PASS,
2489 POSIXize the recursive makes (make [variable assignments] target{s})
2493 * tm-sun4sol2.h (CPLUS_MARKER): Remove, now set in tm-sysv4.h.
2494 * tm-sysv4.h (CPLUS_MARKER): By default, g++ uses '.' as the
2495 CPLUS_MARKER for all SVR4 systems, so follow suit.
2496 * defs.h (strdup_demangled): Remove prototype.
2497 * dwarfread.c (enum_type, synthesize_typedef): Use new macro
2498 SYMBOL_INIT_LANGUAGE_SPECIFIC.
2499 * dwarfread.c (new_symbol): Use SYMBOL_INIT_DEMANGLED_NAME.
2500 * minsyms.c (install_minimal_symbols, prim_record_minimal_symbol,
2501 prim_record_minimal_symbol_and_info): Use new macro
2502 SYMBOL_INIT_LANGUAGE_SPECIFIC.
2503 * minsyms.c (install_minimal_symbols): Use new macro
2504 SYMBOL_INIT_DEMANGLED_NAME.
2505 * stabsread.c (define_symbol): Use new macro
2506 SYMBOL_INIT_DEMANGLED_NAME.
2507 * symfile.c (add_psymbol_to_list, add_psymbol_addr_to_list):
2508 Use new macro SYMBOL_INIT_DEMANGLED_NAME.
2509 * symfile.h (ADD_PSYMBOL_VT_TO_LIST): Use new macro
2510 SYMBOL_INIT_DEMANGLED_NAME.
2511 * symmisc.c (dump_msymbols, dump_symtab, print_partial_symbol):
2512 SYMBOL_DEMANGLED_NAME now tests language itself.
2513 * symtab.c (COMPLETION_LIST_ADD_SYMBOL): SYMBOL_DEMANGLED_NAME
2514 now tests language itself.
2515 * symtab.h (SYMBOL_CPLUS_DEMANGLED_NAME): New macro that does
2516 what SYMBOL_DEMANGLED_NAME used to do, directly access the C++
2517 mangled name member in the language dependent portion of a symbol.
2518 * symtab.h (SYMBOL_DEMANGLED_NAME): New macro that returns the
2519 mangled name member appropriate for a symbol's language.
2520 * symtab.h (SYMBOL_SOURCE_NAME, SYMBOL_LINKAGE_NAME,
2521 SYMBOL_MATCHES_NAME, SYMBOL_MATCHES_REGEXP):
2522 SYMBOL_DEMANGLED_NAME now tests language itself.
2523 * symtab.h (SYMBOL_INIT_LANGUAGE_SPECIFIC): New macro that
2524 initializes language dependent portion of symbol.
2525 * symtab.h (SYMBOL_INIT_DEMANGLED_NAME): New macro that
2526 demangles and caches the demangled form of symbol names.
2527 * utils.c (fputs_demangled, fprint_symbol): Use current language
2528 to select an appropriate demangling algorithm.
2529 * utils.c (strdup_demangled): Remove, no longer used.
2530 * symtab.h (SYMBOL_CHILL_DEMANGLED_NAME): New macro that directly
2531 access the Chill mangled name member in the language dependent
2532 portion of a symbol.
2533 * ch-lang.c (chill_demangle): New function, simple demangler.
2534 * defs.h (chill_demangle): Add prototype.
2535 * symtab.h (language_dependent_info): Add struct for Chill.
2537 For older changes see ChangeLog-92
2543 version-control: never