Ken Raeburn [Wed, 20 Oct 1993 15:51:24 +0000 (15:51 +0000)]
new test m68k/pic1.s - simple pic code generation
Ken Raeburn [Wed, 20 Oct 1993 15:50:04 +0000 (15:50 +0000)]
new test diff1.s - difference of two undefined symbols (should be rejected)
Ken Raeburn [Wed, 20 Oct 1993 15:47:59 +0000 (15:47 +0000)]
* Makefile.in (OBJDUMP_FOR_TARGET): Define similar to AS_FOR_TARGET.
(check): Don't pass ASFLAGS variable.
(site.exp): Put ASFLAGS, OBJDUMP, OBJDUMPFLAGS into site.exp.
Ken Raeburn [Wed, 20 Oct 1993 15:38:42 +0000 (15:38 +0000)]
Makefile.in (OBJDUMP_FOR_TARGET): Define similar to AS_FOR_TARGET.
(check): Don't pass ASFLAGS variable.
(site.exp): Put ASFLAGS, OBJDUMP, OBJDUMPFLAGS into site.exp.
From Jeff Law:
lib/gas-defs.exp (objdump_start, objdump_finish): New functions
so that tests can parse the output of objdump looking for errors
in relocation entires, file headers and the like.
Jim Kingdon [Wed, 20 Oct 1993 15:33:40 +0000 (15:33 +0000)]
* hosts/i386bsd.h [__bsdi__]: Define HOST_DATA_START_ADDR.
Jim Kingdon [Wed, 20 Oct 1993 15:01:15 +0000 (15:01 +0000)]
* configure.in: Add comment about HP compiler vs. emacs.
Jim Kingdon [Wed, 20 Oct 1993 03:58:18 +0000 (03:58 +0000)]
* mh-hp300: Don't define CFLAGS to empty. Why should hp300 be
different from anything else? ("gdb doesn't understand the native
debug format" isn't a good enough answer because we might be using
gcc).
Roland Pesch [Wed, 20 Oct 1993 02:03:50 +0000 (02:03 +0000)]
fix one bad date, credits
Fred Fish [Wed, 20 Oct 1993 00:30:09 +0000 (00:30 +0000)]
* Makefile.in (ALLCONFIG): Add config/i386/{i386m3.mh, i386m3.mt,
i386/i386mk.mh i386/i386mk.mt}, config/mips/{mipsm3.mh,
mipsm3.mt}, config/ns32k/{ns32km3.mh, ns32m3.mt}
* Makefile.in (remote_utils_h): Add remote-sim.h
* Makefile.in (NONSRC): Add i386-nlmstub.c
* Makefile.in (HFILES): Add coff-solib.h
David D. Zuhn [Tue, 19 Oct 1993 23:04:39 +0000 (23:04 +0000)]
don't build ld on solaris2
Ken Raeburn [Tue, 19 Oct 1993 22:24:16 +0000 (22:24 +0000)]
configure.in: Accept alpha-dec-osf1* instead of just -osf1, since config.guess
will include the full version number.
Ian Lance Taylor [Tue, 19 Oct 1993 22:18:15 +0000 (22:18 +0000)]
* config/tc-m68k.h (AOUT_MACHTYPE): Define to be external variable
m68k_aout_machtype.
* config/tc-m68k.c (omagic): Remove obsolete and unused variable.
(m68k_aout_machtype): New variable, if OBJ_AOUT.
(md_assemble): Initialize m68k_aout_machtype based on
current_architecture, if OBJ_AOUT.
(md_parse_option): Remove obsolete reference to omagic.
Kung Hsu [Tue, 19 Oct 1993 22:17:57 +0000 (22:17 +0000)]
Modified Files:
ChangeLog
* values.c (value_virtual_fn_field): Fix the offset calculation
when calling virtual functions. (gdb.t22/virtfunc.exp).
* eval.c (evaluate_subexp): same as above.
Jim Kingdon [Tue, 19 Oct 1993 21:40:09 +0000 (21:40 +0000)]
* gdb.t15/funcargs.exp: Don't put comments on same line as
setup_xfail (@#$*%& tcl braindamage!).
Roland Pesch [Tue, 19 Oct 1993 21:25:32 +0000 (21:25 +0000)]
Tue Oct 19 14:21:18 1993 Roland H. Pesch (pesch@fowanton.cygnus.com)
* a4rc.sed: update to work with Andreas Vogel papersize params
* refcard.tex: use Andreas Vogel simplifications of papersize
params; remove useless version info; update copyright date.
Kung Hsu [Tue, 19 Oct 1993 21:20:37 +0000 (21:20 +0000)]
Modified Files:
values.c eval.c
* values.c (value_virtual_fn_field): Fix the offset calculation
when calling virtual functions. (gdb.t22/virtfunc.exp).
* eval.c (evaluate_subexp): same as above.
Ian Lance Taylor [Tue, 19 Oct 1993 19:59:16 +0000 (19:59 +0000)]
* configure.in: Build linker and binutils for alpha-dec-osf1.
Ian Lance Taylor [Tue, 19 Oct 1993 19:55:52 +0000 (19:55 +0000)]
* configure.in (alpha-*-osf*): New target; use alpha.mt.
* Makefile.in (ALL_EMULATIONS): Added em_alpha.o.
(em_alpha.c): New target; use alpha.sh and alpha.sc.
* config/alphaosf.mh (NATIVE_LIB_DIRS, HOSTING_CRT0): Define.
* config/alpha.mt: New file.
* emulparams/alpha.sh: New file.
* scripttempl/alpha.sc: New file.
Ian Lance Taylor [Tue, 19 Oct 1993 19:52:33 +0000 (19:52 +0000)]
New files for Alpha linker.
Ian Lance Taylor [Tue, 19 Oct 1993 19:41:44 +0000 (19:41 +0000)]
* coff-alpha.c: Implemented reloc handling for the linker, made
various other related changes.
* libecoff.h (ecoff_backend_data): Added rdata_in_text field.
Renamed finish_reloc to adjust_reloc_in. Added adjust_reloc_out
field.
(ecoff_bfd_get_relocated_section_contents): Don't define.
* ecoff.c (ecoff_sec_to_styp_flags, ecoff_styp_to_sec_flags):
Handle .lita section.
(ecoff_slurp_reloc_table): Handle RELOC_SECTION_FINI.
finish_reloc backend field renamed to adjust_reloc_in.
(ecoff_bfd_seclet_link): Adjust number of aux entries to
debug_align boundary.
(ecoff_compute_section_file_positions): If rdata_in_text put
.rdata section in text segment, not data segment. Put .pdata
section in text segment.
(ecoff_write_object_contents): Likewise. Also, call
adjust_reloc_out, and handle .lita, .xdata, .pdata, .fini and
absolute sections when writing out relocs.
* coffswap.h (coff_swap_aouthdr_out): For ALPHAECOFF, output 2 for
bldrev field, as on Alpha OSF/1. Padding field is now 2 bytes.
* coff-mips.c (mips_ecoff_swap_reloc_out): Added assertion.
(mips_adjust_reloc_in): Renamed from mips_finish_reloc.
(mips_adjust_reloc_out): New function (does nothing).
(mips_ecoff_backend_data): Initialize new fields.
(ecoff_bfd_get_relocation_section_contents): Define to be
bfd_generic_get_relocated_section_contents.
* reloc.c (bfd_perform_relocation): Added casts to avoid Alpha
OSF/1 cc bug.
Fred Fish [Tue, 19 Oct 1993 19:38:05 +0000 (19:38 +0000)]
Clean out all references to snapshots.readme, a file that is not
distributed in gdb releases.
David D. Zuhn [Tue, 19 Oct 1993 17:44:57 +0000 (17:44 +0000)]
remove debugging echo
Jim Kingdon [Tue, 19 Oct 1993 17:20:53 +0000 (17:20 +0000)]
* config/rs6000/rs6000.mh (TERMCAP): Define to -lcurses.
* Makefile.in: Define CXXFLAGS.
Jim Kingdon [Tue, 19 Oct 1993 16:47:37 +0000 (16:47 +0000)]
* Makefile.in: Remove -O from CXXFLAGS for consistency with CFLAGS,
and gdb/testsuite/Makefile.in.
Stu Grossman [Tue, 19 Oct 1993 16:34:16 +0000 (16:34 +0000)]
* sparclite/Makefile.in: Fixup so that this works with Sun make
and VPATH.
Jim Kingdon [Tue, 19 Oct 1993 16:26:06 +0000 (16:26 +0000)]
* gdb.texinfo (Symbols): Add class NAME to doc for ptype.
Jim Kingdon [Tue, 19 Oct 1993 16:05:23 +0000 (16:05 +0000)]
* breakpoint.h (struct bpstat_what): Don't use bitfields.
Jim Kingdon [Tue, 19 Oct 1993 15:56:20 +0000 (15:56 +0000)]
* typeprint.c: Add "class CLASS-NAME" to docstring for ptype.
Fred Fish [Tue, 19 Oct 1993 13:27:55 +0000 (13:27 +0000)]
* Makefile.in (ALLPARAM): Add config/m88k/{nm-delta88v4.h,
tm-delta88v4.h, xm-dgux.h}.
* Makefile.in (ALLCONFIG): Add config/m88k/{delta88v4.mh,
delta88v4.mt}.
Fred Fish [Tue, 19 Oct 1993 13:19:25 +0000 (13:19 +0000)]
* README: Remove comment about SunOS 5.x programs leaving
coredumps. Info from Sun is that this was not in customer
releases.
Fred Fish [Tue, 19 Oct 1993 13:15:42 +0000 (13:15 +0000)]
* snapshots.readme: Reword paragraph about not publicising
location of gdb snapshots.
Jim Kingdon [Tue, 19 Oct 1993 02:56:08 +0000 (02:56 +0000)]
* gdb.disasm/hppa.exp: Use $objdir/$subdir/$binfile not just $binfile.
Stu Grossman [Tue, 19 Oct 1993 01:17:42 +0000 (01:17 +0000)]
* Makefile.in: Add dependancy for elf.o to make Sun make happy
when using VPATH.
Jim Kingdon [Mon, 18 Oct 1993 23:06:19 +0000 (23:06 +0000)]
* hppa-tdep.c (restore_pc_queue): Call target_terminal_ours after
done stepping the inferior.
David D. Zuhn [Mon, 18 Oct 1993 22:28:18 +0000 (22:28 +0000)]
handle unixware & hp emacs, textutils on rs6000
Jim Kingdon [Mon, 18 Oct 1993 18:01:51 +0000 (18:01 +0000)]
* c-exp.y: Remove never-used (because of shift/reduce conflicts)
rules for pointers to members.
* Makefile.in: Remove notice about expected shift/reduce conflicts.
Jim Kingdon [Mon, 18 Oct 1993 15:40:04 +0000 (15:40 +0000)]
* buildsym.c (finish_block): If we pop the context stack and it is
not empty, complain () instead of abort ().
Jim Kingdon [Mon, 18 Oct 1993 01:10:25 +0000 (01:10 +0000)]
* parse.c, parser-defs.h (follow_types): New function.
* c-exp.y (ptype : typebase abs_decl): Use it.
* c-exp.y (ptype): Add support for type qualifiers after the
typebase. The typebase rule already has support for them before
the typebase.
* Makefile.in: Change the expected number of shift/reduce
conflicts to 6. This is OK--the 2 new conflicts are basically the
same as one of the old ones.
Fred Fish [Sun, 17 Oct 1993 20:08:10 +0000 (20:08 +0000)]
Bump version to 4.10.3 in main trunk after branch for gdb 4.11
Jim Kingdon [Sun, 17 Oct 1993 14:30:20 +0000 (14:30 +0000)]
* infrun.c (wait_for_inferior): Clean up comments which were at
the top of the file, making them more concise and moving them with
the code (Sorry, Randy, but these stream-of-consciousness comments
really have to go). Switch the order of the "&&", which makes
things clearer and turns out to be an improvement with respect to
side effects and speed.
Stu Grossman [Sun, 17 Oct 1993 09:12:58 +0000 (09:12 +0000)]
* procfs.c: Handle process exits more elegantly by trapping on
entry to _exit. Also, cleanup procinfo list when process dies of
it's own accord (as opposed to being killed).
Jim Kingdon [Sun, 17 Oct 1993 02:21:12 +0000 (02:21 +0000)]
* infrun.c: Add comment about signals.
Jim Kingdon [Sun, 17 Oct 1993 02:19:59 +0000 (02:19 +0000)]
* config/m88k/xm-dgux.h: Define NO_PTRACE_H.
Jim Kingdon [Sun, 17 Oct 1993 02:09:07 +0000 (02:09 +0000)]
* gdb.disasm: New directory for GDB disassembler tests.
Jim Kingdon [Sun, 17 Oct 1993 02:09:06 +0000 (02:09 +0000)]
* gdb.disasm: New directory for GDB disassembler tests.
Jim Kingdon [Sun, 17 Oct 1993 02:05:58 +0000 (02:05 +0000)]
* corelow.c (add_to_thread_list): Need a cast to go from PTR to
asection *.
Jim Kingdon [Sun, 17 Oct 1993 01:56:01 +0000 (01:56 +0000)]
* infrun.c: Add comment about signals.
* fork-child.c (fork_inferior): Remove CREATE_INFERIOR_HOOK again.
Stu reinstated it (accidently I assume).
Stu Grossman [Sun, 17 Oct 1993 01:02:57 +0000 (01:02 +0000)]
* procfs.c (procfs_wait): Losing Unixware can't do poll on /proc
files. Use PIOCWSTOP instead.
* corelow.c (add_to_thread_list): Fix arg to match prototype.
Stu Grossman [Sat, 16 Oct 1993 22:32:04 +0000 (22:32 +0000)]
* procfs.c (procfs_set_sproc_trap): Don't use this if sproc
isn't available.
* (procfs_notice_signals): Fix prototype.
Stu Grossman [Sat, 16 Oct 1993 22:31:07 +0000 (22:31 +0000)]
*** empty log message ***
Stu Grossman [Sat, 16 Oct 1993 06:42:02 +0000 (06:42 +0000)]
* breakpoint.c (breakpoint_thread_match break_command_1):
Thread-specific breakpoint support.
* breakpoint.h (struct breakpoint): Add thread id field.
* fork-child.c (fork_inferior): Move call to init_thread_list()
back a bit so that init_trace_fun can do thread functions.
* hppa-tdep.c (restore_pc_queue): Add pid to call to target_wait.
* hppab-nat.c (child_resume): Handle default pid.
* hppah-nat.c (child_resume): Handle default pid.
* i386lynx-nat.c (child_wait): New arg pid.
* inflow.c (kill_command): Reset thread list.
* infptrace.c (child_resume): Handle default pid.
* infrun.c: Thread-specific breakpoint support.
* inftarg.c (child_wait): Add pid arg.
* osfsolib.c (solib_create_inferior_hook): Add pid to call to
target_resume.
* procfs.c: Multi-thread support.
* remote-bug.c (bug_wait): Add pid arg.
* remote-hms.c (hms_wait): Add pid arg.
* remote-mips.c (mips_wait): Add pid arg.
* remote-mon.c (monitor_wait): Add pid arg.
* remote-nindy.c (nindy_wait): Add pid arg.
* remote-sim.c (gdbsim_wait): Add pid arg.
* remote-udi.c (udi_wait): Add pid arg.
* remote-vx.c (vx_wait): Add pid arg.
* remote-z8k.c (sim_wait): Add pid arg.
* remote.c (remote_wait): Add pid arg.
* solib.c (solib_create_inferior_hook): Add pid to call to
target_resume.
* target.h (struct target_ops): Add pid arg to to_wait and
to_notice_signals.
* thread.c (valid_thread_id): New func to validate thread #s.
* (pid_to_thread_id): New func to do the obvious.
* thread.h: Prototypes for above.
* coff-solib.c (coff_solib_add): Use nameoffset field to locate
filename.
Jim Kingdon [Sat, 16 Oct 1993 05:07:31 +0000 (05:07 +0000)]
* h8300-tdep.c, h8500-tdep.c: Define sim_load only, but not
sim_kill, sim_open, or sim_set_args.
Jim Kingdon [Sat, 16 Oct 1993 04:59:47 +0000 (04:59 +0000)]
* h8300-tdep.c, h8500-tdep.c: Define sim_load only, but not
sim_kill, sim_open, or sim_set_args.
Jim Kingdon [Sat, 16 Oct 1993 04:58:20 +0000 (04:58 +0000)]
x
Jim Kingdon [Sat, 16 Oct 1993 04:37:45 +0000 (04:37 +0000)]
* stack.c (print_stack_frame): Put catch_errors around
print_frame_info so (for example) error printing source doesn't
cause auto-displays to get skipped in normal_stop.
Jim Kingdon [Sat, 16 Oct 1993 02:41:17 +0000 (02:41 +0000)]
* findvar.c (value_from_register): When preparing to cast a value
from REGISTER_VIRTUAL_TYPE to type, copy the REGISTER_VIRTUAL_SIZE;
the old code didn't copy the whole thing.
* valops.c (value_assign): Add comment.
Peter Schauer [Fri, 15 Oct 1993 19:58:39 +0000 (19:58 +0000)]
* mipsread.c (upgrade_type): Replace bitsize sanity checks and
complaint by a comment explaining why they were useless.
Jim Kingdon [Fri, 15 Oct 1993 19:37:29 +0000 (19:37 +0000)]
* Move comments on bypassing call dummy breakpoint from stack.c
to breakpoint.h.
Jim Kingdon [Fri, 15 Oct 1993 18:50:29 +0000 (18:50 +0000)]
* gdb.t06/break.exp: For "next over recursive call", fail on any
incorrect output, not just on "factorial (value=2)".
Jim Kingdon [Fri, 15 Oct 1993 17:13:27 +0000 (17:13 +0000)]
Fri Oct 15 11:52:56 1993 Peter Schauer (pes@regent.e-technik.tu-muenchen.de)
* gdb.t12/scope.exp: Remove xfails, filename::var syntax should be
working now. Add an xfail if accesing a bss variable causes
a memory error if the target is not yet run.
Jim Kingdon [Fri, 15 Oct 1993 17:11:11 +0000 (17:11 +0000)]
Fri Oct 15 11:52:56 1993 Peter Schauer (pes@regent.e-technik.tu-muenchen.de)
* symtab.c (lookup_partial_symtab): If filename is not found and
contains no slashes, try again and compare without leading path
components.
* symtab.c (lookup_symtab_1): Replace open coded version of
lookup_partial_symtab with a function call.
David Edelsohn [Fri, 15 Oct 1993 09:29:32 +0000 (09:29 +0000)]
* ldlang.c (lang_size_sections, lang_common): ALIGN_N can't handle
types of different sizes (eg: 64 and 32 bits), so coerce.
* ld.h (ALIGN_N): Add warning about usage.
Jim Kingdon [Fri, 15 Oct 1993 04:26:06 +0000 (04:26 +0000)]
* fork-child.c (fork_inferior), remote-eb.c (eb_create_inferior),
remote-mon.c (monitor_create_inferior), remote-nindy.c
(nindy_create_inferior), remote-st.c (st2000_create_inferior),
remote-vx.c (vx_create_inferior): Remove CREATE_INFERIOR_HOOK; it
is replaced by init_trace_fun.
* config/convex/xm-convex.h, convex-xdep.c: Add comments explaining
how to do without CREATE_INFERIOR_HOOK for whoever fixes the Convex
port.
* Makefile.in: Add Mach files to ALLDEPFILES, etc.
* m3-nat.c: Clean up more hair--message(), cprocs.
* configure.in: Recognize Mach targets and hosts.
* config/ns32k/tm-umax.h: Add some #ifndef's so tm-ns32km3.h can
include this file.
* Mach headers in config/*/tm-*.h: Fix includes to match correct
locations of files.
Rob Savoye [Fri, 15 Oct 1993 03:41:58 +0000 (03:41 +0000)]
* remote-mon.c (general_open): Set dev_name. Minor tweaking to get
it working again.
* config/m68k/tm-monitor.h: Remove floating point register names
as there aren't any on any of the monitors that use this code.
Jim Kingdon [Fri, 15 Oct 1993 00:36:26 +0000 (00:36 +0000)]
* gdb.t24/demangle.exp: Clean up spacing (2 strings), inclusion of
constructor name (2 string), name of _GLOBAL_ demangling (2 strings),
name of __st* demangling (2 strings). Did Kung really get the
more than 95 others right without testing them?
Kung Hsu [Thu, 14 Oct 1993 23:33:33 +0000 (23:33 +0000)]
Modified Files:
ChangeLog
* gdb.t24/demangle.exp: clean up XFAILS, more than 100 of them.
Jim Kingdon [Thu, 14 Oct 1993 21:58:58 +0000 (21:58 +0000)]
Thu Oct 14 11:40:30 1993 Jeffrey Law (law@cs.uah.edu)
* configure.in (configdirs): Add gdb.disasm
* gdb.disasm: New directory for GDB disassembler tests.
* gdb.disasm/hppa.{exp,s,mt}: Disassembly tests for the HPPA.
Ian Lance Taylor [Thu, 14 Oct 1993 21:06:52 +0000 (21:06 +0000)]
* config/tc-mips.c (md_pseudo_table): Ignore .livereg pseudo-op.
(s_option): Ignore .option O* and .option pic*.
(s_ent): Skip whitespace between symbol and optional digit.
Jim Kingdon [Thu, 14 Oct 1993 21:03:36 +0000 (21:03 +0000)]
Thu Oct 14 11:40:30 1993 Jeffrey Law (law@cs.uah.edu)
* configure.in (configdirs): Add gdb.disasm
* gdb.disasm: New directory for GDB disassembler tests.
* gdb.disasm/hppa.{exp,s,mt}: Disassembly tests for the HPPA.
Ian Lance Taylor [Thu, 14 Oct 1993 21:03:14 +0000 (21:03 +0000)]
* app.c (do_scrub_next_char): Always accept 'x' and 'X' as escape
characters in state 6.
* read.c (next_char_of_string): Accept \Xh* and \xh* where h* are
hexidecimal digits.
Ian Lance Taylor [Thu, 14 Oct 1993 20:54:09 +0000 (20:54 +0000)]
* config/tc-i386.c (md_apply_fix_1): Make cross segment calls work
for ELF by hacking around bizarre bfd_perform_relocation behaviour
that I don't dare change.
Jim Kingdon [Thu, 14 Oct 1993 18:53:31 +0000 (18:53 +0000)]
* gdb.t00/help.exp: Don't test "help set".
Jim Kingdon [Thu, 14 Oct 1993 18:40:20 +0000 (18:40 +0000)]
* gdb.t20/inherit.exp, gdb.t22/virtfunc.exp: Remove all xfails
from within expect statements. None of these happen anymore and
they cause regressions to be silently unnoticed.
* gdb.t20/inherit.exp, gdb.t22/virtfunc.exp: Add notes (in failure
messages) about known failures with gcc cygnus-2.4.5-930417.
Jim Kingdon [Thu, 14 Oct 1993 18:20:52 +0000 (18:20 +0000)]
* gdb.t20/inherit.exp, gdb.t22/virtfunc.exp: Remove all xfails
from within expect statements. None of these happen anymore and
they cause regressions to be silently unnoticed.
* gdb.t20/inherit.exp, gdb.t22/virtfunc.exp: Add notes (in failure
messages) about known failures with gcc cygnus-2.4.5-930417.
* gdb.t22/virtfunc.exp: Add notes (in comment and failure message)
about problem with gcc cygnus-2.4.5-930828.
Jim Kingdon [Thu, 14 Oct 1993 17:50:24 +0000 (17:50 +0000)]
* gdb.t20/inherit.exp: Remove all xfails from within expect statements.
None of these happen anymore and they cause regressions to be silently
unnoticed.
* gdb.t20/inherit.exp: Add note (in failure message) about known
failure with gcc cygnus-2.4.5-930417.
Jim Kingdon [Thu, 14 Oct 1993 17:29:01 +0000 (17:29 +0000)]
* gdb.t20/inherit.exp: Add note (in failure message) about known
failure with gcc cygnus-2.4.5-930417.
* gdb.t22/virtfunc.exp: Add notes (in comments and failure message)
about problem with gcc cygnus-2.4.5-930828.
David Edelsohn [Thu, 14 Oct 1993 17:13:29 +0000 (17:13 +0000)]
Remove references to shade.
Jim Kingdon [Thu, 14 Oct 1993 16:47:34 +0000 (16:47 +0000)]
* gdb.t02/Makefile.in, gdb.t08/Makefile.in: Make all depend on
comp-info.exp so it gets built if it doesn't exist.
* gdb.t13/{Makefile.in,bitfields.exp}: Remove comp-info.exp stuff;
it is no longer used.
* gdb.t13/comp-info.c: Removed.
Ian Lance Taylor [Thu, 14 Oct 1993 16:04:35 +0000 (16:04 +0000)]
* coffcode.h (styp_to_sec_flags): Don't set SEC_DEBUGGING until it
is made to work.
(coff_compute_section_file_positions): Remove check for
USE_DISCARDED_SECTIONS_COUNT.
* coff-i386.c: Don't define USE_DISCARDED_SECTIONS_COUNT. Basing
it on host defines is wrong.
Ian Lance Taylor [Thu, 14 Oct 1993 15:41:52 +0000 (15:41 +0000)]
Thu Oct 14 11:33:25 1993 Michael Meissner (meissner@osf.org)
* config/tc-i386.c: (md_begin): Do not zero static arrays. Don't
call strchr for each character to see if it is a special char,
instead add a second loop over special_chars. Set alignment
of text, data and bss sections to 4.
(pi, te, pt, pe, ps): Add declarations so that DEBUG386 can be
used again.
(reloc): Don't return 8 and 16 bit non-PC relative relocations on
ELF, since the ELF object format does not have these type of
relocations. Change the abort into as as_bad and return
BFD_RELOC_NONE to silence compiler warnings.
(md_assemble): Keep track of the instruction size. Allow white
space between the $ and the constant for compatibility with older
gases and other assemblers.
(i386_operand): Skip spaces between $ and expression.
(tc_gen_reloc): Don't allow anything but 32 bit relocations on
ELF. Convert abort into an as_bad and assert into as_fatal.
Jim Kingdon [Thu, 14 Oct 1993 06:03:23 +0000 (06:03 +0000)]
* gdb.t06/signals.exp: Add xfail for sun3.
Ian Lance Taylor [Thu, 14 Oct 1993 05:13:43 +0000 (05:13 +0000)]
* ecoff.c (ecoff_bfd_seclet_link): Don't link a .reginfo section.
(ecoff_write_object_contents): Don't require the .reginfo section
to be a particular size.
Jim Kingdon [Thu, 14 Oct 1993 02:48:33 +0000 (02:48 +0000)]
* inflow.c: Pass pointer to process group, not process group itself,
to TIOCSPGRP ioctl.
Ken Raeburn [Wed, 13 Oct 1993 23:09:08 +0000 (23:09 +0000)]
keep apollo.h
Ken Raeburn [Wed, 13 Oct 1993 23:08:10 +0000 (23:08 +0000)]
More "make dep" fixups.
Apollo 68k support from Troy Rollo <troy@cbme.unsw.edu.au>.
Ken Raeburn [Wed, 13 Oct 1993 22:33:48 +0000 (22:33 +0000)]
Oopsie.
Use "[^ ]*", not ".*", in matching filename patterns.
Updated dependencies in Makefile.in.
Ken Raeburn [Wed, 13 Oct 1993 21:42:42 +0000 (21:42 +0000)]
remove inclusion of strings.h
Ken Raeburn [Wed, 13 Oct 1993 20:54:54 +0000 (20:54 +0000)]
obj-coffbfd change for wellfleet pic
Ken Raeburn [Wed, 13 Oct 1993 20:54:03 +0000 (20:54 +0000)]
* config/obj-coffbfd.c (fixup_segment) [DIFF_EXPR_OK]: If
sub_symbolP is in the current segment, convert to a PC-relative
fixup and discard the symbol.
(Copied from write.c.)
Ian Lance Taylor [Wed, 13 Oct 1993 20:04:30 +0000 (20:04 +0000)]
* ldmain.c (enter_global_ref): Just ignore any weak symbol for
which we already have a definition, rather than checking in
several different places whether the symbol is weak.
Ken Raeburn [Wed, 13 Oct 1993 19:17:05 +0000 (19:17 +0000)]
coff-i386.c (i3coff_object_p): Now static.
Ken Raeburn [Wed, 13 Oct 1993 19:04:43 +0000 (19:04 +0000)]
68kcoff, makedep changes
Ken Raeburn [Wed, 13 Oct 1993 19:04:08 +0000 (19:04 +0000)]
Make "make dep" easier to work with. Updated lots of dependencies.
Don't try "make dep" alone; use "make dep DEP=mymkdep", where "mymkdep" has
been hacked to use "gcc -MM"...
Ken Raeburn [Wed, 13 Oct 1993 18:57:21 +0000 (18:57 +0000)]
Define ONLY_DECLARE_RELOCS, to avoid name conflicts when "--with-targets=all".
Ken Raeburn [Wed, 13 Oct 1993 18:56:39 +0000 (18:56 +0000)]
for m68klynx_coff_vec, include coff-m68k.o for now
Ian Lance Taylor [Wed, 13 Oct 1993 18:53:03 +0000 (18:53 +0000)]
* config/obj-ecoff.c (ecoff_build_symbols): Handle st_End symbol
for st_StaticProc just like st_Proc.
Jim Kingdon [Wed, 13 Oct 1993 18:29:09 +0000 (18:29 +0000)]
* inflow.c (terminal_ours_1): Don't print warning on failure to
set process group.
Jim Kingdon [Wed, 13 Oct 1993 18:17:49 +0000 (18:17 +0000)]
* printcmd.c (printf_command): Instead of using makeva* and
calling vprintf, just make the appropriate calls to printf.
* printcmd.c, config/pa/xm-pa.h, config/mips/xm-makeva.h,
config/alpha/xm-alpha.h, config/m88k/xm-m88k.h: Remove all
traces of makeva*. My apologies to everyone (including me!)
who spent so much time getting it to work on various machines,
but look at the bright side, at least you won't have to do it
again in the future.
* printcmd.c (printf_command): Make a cleanup for val_args (fixes
a memory leak).
Ian Lance Taylor [Wed, 13 Oct 1993 18:04:53 +0000 (18:04 +0000)]
* write.c (relax_and_size_all_segments): Moved #endif for OBJ_BOUT
so that OBJ_BOUT doesn't forget to adjust all the fragments in the
.bss section.
Fixes critical PR 3600.
Jim Kingdon [Wed, 13 Oct 1993 04:00:46 +0000 (04:00 +0000)]
* config/mips/xm-mips.h: Remove comment about HAVE_SGTTY vs. usleep.