Jim Kingdon [Fri, 24 Mar 1995 16:38:06 +0000 (16:38 +0000)]
* gdb.base/a1-selftest.exp: Don't check for # followed by a digit
somewhere between `read' and `main.c'. I'm pretty sure the
pattern ".*#\[0-9\].*" was slowing down pattern matching a lot,
and it isn't particularly useful.
Jeff Law [Fri, 24 Mar 1995 06:21:44 +0000 (06:21 +0000)]
* somsolib.c (som_solib_add): Handle case where a shared library
referenced by a core file has sections without the SEC_ALLOC bit
set (eg stabs sections).
Kung Hsu [Wed, 22 Mar 1995 01:13:08 +0000 (01:13 +0000)]
* lib/gdb.exp: add proc runto_main, for targets that use stubs, this
will not runto main but do a 'step' to step out of breakpoint().
* config/mt-slite: add -Dusestubs.
Jim Kingdon [Tue, 21 Mar 1995 22:02:06 +0000 (22:02 +0000)]
* gdb.base/signals.exp: Add test for "handle all print".
* TODO: Remove "handle all print". Also remove item about
checking copyright date (I don't like the idea of a spurious FAIL
based on when we run the tests).
Ian Lance Taylor [Tue, 21 Mar 1995 20:17:07 +0000 (20:17 +0000)]
* ldlang.c (lang_common): Change power to int. Pass larger values
before smaller values.
(lang_one_common): Treat info as int *. Don't bother to check for
last value, since it is now zero.
Stu Grossman [Mon, 20 Mar 1995 23:40:33 +0000 (23:40 +0000)]
* findvar.c (read_register_bytes write_register_bytes): Make
these routines much smarter about updating registers from the
target, only doing so when absolutely necessary. This really
speeds up register modification on some remote targets.
* monitor.c: More cleanups. Get rid of monitor_load_ascii_srec.
BFD makes this unnecessary. Lots of debugging speedups.
* (expect): NULL terminate return string.
* (monitor_open monitor_supply_register parse_register_dump
monitor_wait monitor_fetch_register): Switch to using GNU regexp
library to parse multi-register displays.
* (monitor_read_memory): Read multiple bytes (up to 16) at once.
* (monitor_create_inferior): Call clear_proceed_status to make run
command notice first breakpoint.
* (monitor_load): Clean up. Reset inferior_pid, set pc to start
address and reset symbol table stuff to make loads put things into
a fresh state.
* (monitor_load_srec): Lower sleep time to 1 second.
* monitor.h (struct monitor_ops): Add register_pattern and
supply_register to monitor_ops.
* rom68k-rom.c: Add new support for handling register dumps.
* config/m68k/tm-m68k.h: Define D0_REGNUM and A0_REGNUM for register
dump handling.
Stu Grossman [Mon, 20 Mar 1995 15:38:56 +0000 (15:38 +0000)]
* hppah-nat.c (store_inferior_registers): Move check for
CANNOT_STORE_REGISTER to a better place. Fixes ptrace I/O errors
found by test suite during function calls, which attempts to write
unwritable registers.
Peter Schauer [Sat, 18 Mar 1995 13:16:51 +0000 (13:16 +0000)]
* mdebugread.c (parse_symbol): If finishing a function without
known parameter type info, set that from parameter symbols.
Remove commented-out add_param_to_type support.
Stan Shebs [Sat, 18 Mar 1995 00:50:51 +0000 (00:50 +0000)]
* mpw-config.in (TDEFINES): Put empty definition into
makefile fragment.
* mpw-make.in (xmalloc.c.o): Remove.
(as.new): Depend on Version.r.
(Version.r): Generate from version info.
Stan Shebs [Sat, 18 Mar 1995 00:42:17 +0000 (00:42 +0000)]
* mpw-configure: Revert previous move of target_cpu
calculation, mpw-config.in needs it.
* mpw-config.in: Add generic include forwards for cpu-specific
include files in aout and elf directories.
Stan Shebs [Sat, 18 Mar 1995 00:13:26 +0000 (00:13 +0000)]
* mpw-config.in (i386-unknown-aout): Change to i386-unknown-go32.
(i386-unknown-coff): Remove.
(sh-hitachi-hms): New target.
(emulation_ofiles): Set correctly for each target.
(version, TDEFINES): Add to makefile fragment.
* mpw-make.in (BISON): Use byacc instead of bison.
(em_*.c): Replace with e*.c everywhere.
(ldgram.h): Separate action from ldgram.c generation.
(LD_PROG): Depend on Version.r.
(Version.r): generate from version info.
* mpw-em.c: Remove.
* mpw-emipsidt.c: New file, modified for MPW from emipsidt.c.
Jeff Law [Fri, 17 Mar 1995 23:45:36 +0000 (23:45 +0000)]
* config/hppa.sed: Sed script to transform bintest.s into proper
PA assembly code.
* binutils-all/nm.exp: Enable these tests on the PA.
* binutils-all/objcopy.exp: Enable these tests on the PA. Expect
simple copy to fail.
* binutils-all/objdump.exp: Enable these tests on the PA. Handle
"CODE" as a section name.
* binutils-all/size.exp: Enable these tests on the PA.
* lib/utils-lib.exp (default_binutils_assemble): For "hppa*-*-*",
run the assembly through a sed script before passing it to the
assembler.
Stan Shebs [Fri, 17 Mar 1995 23:21:12 +0000 (23:21 +0000)]
* mpw-config.in (i386-unknown-go32): Change from aout.
(sh-hitachi-hms): New target.
* mpw-make.in (BISON): Use byacc instead of bison.
(VERSION): Bump to 4.14.1.
(*-pinsn.*): Remove mentions everywhere.
(all): Don't build MacGDB.
(gdb, SiowGDB): Depend on Version.r.
(Version.r): Generate from version info.
* main.c (main): Pass program name to START_PROGRESS, END_PROGRESS.
(main) [MPW]: Remove debugging hook.
* mac-xdep.c (debug_openp): New flag.
(mac_init): Set flag if env variable defined.
* ser-mac.c (mac_open): Clarify error message.
(mac_readchar): Call PROGRESS while looping.
(mac_write): Call sleep instead of sec_sleep.
(sec_sleep): Remove.
* source.c (openp) [MPW]: Only print debugging info if debugging.
* utils.c (query) [MPW]: Clarify that behavior is a bug.
Jim Kingdon [Fri, 17 Mar 1995 00:41:40 +0000 (00:41 +0000)]
* xcoffread.c (process_linenos): Make sure filename we pass to
start_subfile will cause deduce_language_from_filename to return
the correct thing. Reindent function to GNU standards.
Ian Lance Taylor [Thu, 16 Mar 1995 19:06:41 +0000 (19:06 +0000)]
* ldlang.c (load_symbols): If whole_archive is true and ENTRY
describes an archive, call bfd_link_add_symbols on each member.
* ldmain.c (whole_archive): New variable.
(main): Initialize it to false.
* ldmain.h: Declare whole_archive.
* lexsup.c (parse_args): Grok --whole-archive switch.
Stu Grossman [Thu, 16 Mar 1995 00:06:51 +0000 (00:06 +0000)]
* config/rom68k.exp: New file to support Motorola IDP board.
* gdb.base/a2-run.exp: Skip this file if noargs is set.
* gdb.base/break.c: Change things around so that this program
doesn't depend upon args. This is necessary to make remote
targets work (in general, they can't take args).
* gdb.base/break.exp: Don't try to send args to program. Don't
expect output. Also, replace lots of code with gdb_test.
Stu Grossman [Wed, 15 Mar 1995 23:31:14 +0000 (23:31 +0000)]
* findvar.c (read_register_bytes write_register_bytes): Make
these routines much smarter about updating registers from the
target, only doing so when absolutely necessary. This really
speeds up register modification on some remote targets.
* monitor.c: More cleanups. Get rid of monitor_load_ascii_srec.
BFD makes this unnecessary. Lots of debugging speedups.
* (expect): NULL terminate return string.
* (monitor_open monitor_supply_register parse_register_dump
monitor_wait monitor_fetch_register): Switch to using GNU regexp
library to parse multi-register displays.
* (monitor_read_memory): Read multiple bytes (up to 16) at once.
* (monitor_create_inferior): Call clear_proceed_status to make run
command notice first breakpoint.
* (monitor_load): Clean up. Reset inferior_pid, set pc to start
address and reset symbol table stuff to make loads put things into
a fresh state.
* (monitor_load_srec): Lower sleep time to 1 second.
* monitor.h (struct monitor_ops): Add register_pattern and
supply_register to monitor_ops.
* rom68k-rom.c: Add new support for handling register dumps.
Jim Kingdon [Wed, 15 Mar 1995 23:24:24 +0000 (23:24 +0000)]
* utils.c, defs.h (putchar_unfiltered, fputc_unfiltered): Make
argument be an int, not a char. Using a prototype followed by an
old-style function definition in a case where an argument is
widened is a GCC-ism not supported by the native AIX compiler.
Jim Kingdon [Wed, 15 Mar 1995 21:23:14 +0000 (21:23 +0000)]
* gdb.base/printcmds.exp (test_character_literals_accepted): Test
printing of '\'' (which is what the comment says we are testing,
even though we were not), not '''.
(test_integer_literals_rejected): Test that printing ''' is an error.
* gdb.fortran/exprs.exp, gdb.fortran/types.exp,
gdb.chill/chexp.exp, gdb.base/printcmds.exp,
gdb.c++/cplusfuncs.exp, gdb.chill/callch.exp, gdb.chill/misc.exp,
gdb.chill/pr-6292.exp, gdb.chill/string.exp, gdb.chill/tuples.exp:
Use gdb_test not test_print_accept.
* lib/gdb.exp: Remove test_print_accept.
Ian Lance Taylor [Wed, 15 Mar 1995 02:39:39 +0000 (02:39 +0000)]
* config/tc-mips.c (load_register): Don't use addiu for dli of an
unsigned seemingly negative number. Don't bother shifting a zero
value.
(mips_ip): For case 'j', if there are more alternatives, and the
ISA level is at least 3, don't accept an unsigned seemingly
negative number.
Jim Kingdon [Tue, 14 Mar 1995 21:25:19 +0000 (21:25 +0000)]
* lib/gdb.exp (gdb_test): Between $pattern and $prompt, expect
only \r\n, not .*. The test can pass .* as the last thing in
$pattern if that is what it wants. In addition to providing this
flexibility, this change should speed up pattern matching in cases
where the pattern already ended with .* (there were a number of
them). This change also helps catch bad patterns--in the old
scheme the typo "char \*" instead of "char \\*" would pass. Now
it is caught.
* Many .exp files: Update callers.
Stan Shebs [Tue, 14 Mar 1995 21:07:19 +0000 (21:07 +0000)]
* mpw-README: Filled in more details about how to install,
use, and rebuild.
* mpw-configure: Added copyright.
(target_cpu, target_vendor, target_os): Move after reading
of mpw-config.in, paste into generated Makefiles.
(TDEFINES): Remove.
* mpw-config.in: Check for presence of required build tools.
(target_libs): Add newlib.
(target_tools): Add examples.
(Read Me): Generate as "Read Me for MPW" instead.
* mpw-build.in: Base sub-builds on all-foo instead of do-foo.
(all-byacc, do-byacc, all-flex, do-flex, do-newlib): New actions.
(do-gas, do-gcc, do-gdb, do-ld): Build Version.r first.
Jim Kingdon [Tue, 14 Mar 1995 16:47:38 +0000 (16:47 +0000)]
* gdb.base/printcmds.exp (test_integer_literals_rejected): Change
"p '\'", which is the same as "p ''" once tcl gets done with
quoting, to "p '\\'", which I suspect is what is intended (one
backslash gets sent to GDB).
* gdb.base/printcmds.exp (test_artificial_arrays): New tests.
Ken Raeburn [Tue, 14 Mar 1995 02:58:16 +0000 (02:58 +0000)]
* __bb_exit_func.c: New file, from David Mosberger-Tang.
Thu Feb 9 16:56:07 1995 David Mosberger-Tang <davidm@piston.cs.arizona.edu>
* All *.c: More cleanup towards GNU format.
* gmon_out.h (struct gmon_hist_hdr, struct gmon_cg_arc_record): replaced sizeof
(bfd_vma) by size (char*) because Ken tells me that bfd_vma is only guaranteed
to be at least as big as a pointer.
(GMON_Record_tag): added explicit enumeration values to ensure compatibility
across compilers.
* gmon_io.c (get_vma, put_vma): replaced sizeof(bfd_vma) by sizeof(char*).
Ian Lance Taylor [Mon, 13 Mar 1995 21:55:44 +0000 (21:55 +0000)]
* elfcode.h (write_relocs): Make static. Use PTR argument as a
pointer to a boolean variable set to true if something fails.
Cast to PTR rather than void *.
(sym_is_global): Rewrite for clarity.
(map_program_segments): If the program header size is too small,
call _bfd_error_handler and return failure rather than aborting.
(NAME(bfd_elf,write_object_contents)): Pass boolean variable when
calling write_relocs.
(elf_symbol_from_bfd_symbol): Use BFD_ASSERT rather than abort.
(struct elf_info_failed): Define.
(NAME(bfd_elf,size_dynamic_sections)): Declare variables in inner
blocks. Pass elf_info_failed structure when calling
elf_export_symbol and elf_adjust_dynamic_symbol.
(elf_export_symbol): Treat data argument as elf_info_failed rather
than bfd_link_info. Set failed field on error.
(elf_adjust_dynamic_symbol): Likewise.
(struct elf_finfo_failed): Define.
(elf_bfd_final_link): Pass elf_finfo_failed structure when calling
elf_link_output_extsym. Use BFD_ASSERT rather than abort.
(elf_link_output_extsym): Treat data argument as elf_finfo_failed
rather than elf_final_link_info. Set failed field on error.
(elf_link_input_bfd): Use BFD_ASSERT rather than abort.
(elf_reloc_link_order): Likewise.