[Support for trace debugging: registers that were not collected.]
* remote.c (remote_fetch_registers): accept 'xxxx' in the register
packet, with the meaning "register value is not available".
Set register_valid to -1, which will connote "no value available".
* findvar.c (read_relative_register_raw_bytes): return failure if
register_valid == -1. (value_of_register): return failure if
register_valid == -1. (read_var_value): return error if
value_of_register fails for a register variable.
(value_from_register): return failure if register_valid == -1.
* eval.c (evaluate_subexp_standard): return error if
value_of_register fails for a register used in an expression.
* infcmd.c (do_registers_info): display "value not available"
for registers for which register_valid == -1.
* tracepoint.c (set_raw_tracepoint): just save the filename as is
from the symbol table, rather than trying to prepend the dir name.
Also save the bfd section. (tracepoints_info): use the section
when looking up the function name.
* tracepoint.h: add section field to tracepoint struct.
* patch approved by shebs
Wed Oct 28 12:33:52 EST 1998 Frank Ch. Eigler <fche@cygnus.com>
* configure.in: Accept both -skyb- and -sky- for
--with-sim-funit/gpu2 options.
* configure: Rebuilt.
* configure.in (AM_EXEEXT): Use AC_EXEEXT instead.
* configure: Regenerated.
Geoff just removed the AM_EXEEXT definition from ../binutils/aclocal.m4
so we would have been hosed in the near future.
* command.c copying.c copying.awk core-aout.c core-regset.c
corelow.c dcache.c i386-tdep.c i386v4-nat.c i387-tdep.c
infcmd.c infptrace.c infrun.c remote.c solib.c symfile.c
symmisc.c valarith.c: Add prototypes.
* defs.h: Add prototype for utils.c::do_run_cleanups.
* gdbtypes.c: Add prototypes.
(make_pointer_type): Add braces to remove nested if-else ambiguity.
(make_reference_type): Ditto.
* printcmd.c (printf_command): Initialize 'f' and 'string' at
function startup to suppress possibly-used-before-initialized warning.
* remote-utils.c: Add prototypes.
(sr_pollchar): Add braces to remove nested if-else ambiguity.
* ser-tcp.c: Add prototypes.
(wait_for): Add braces to remove nested if-else ambiguity.
(tcp_readchar): Ditto.
* ser-unix.c: Add prototypes.
(get_tty_state): Don't define errno here.
(get_tty_state): Don't define errno here.
(hardwire_readchar): Only define 't' if we are compiling in a Cygwin
environment.
* symtab.c: Add prototypes.
(find_methods): Add braces to remove nested if-else ambiguity.
(search_symbols): Set 'i' to an initial value to suppress a
possibly-used-before-initialized warning.
* valops.c: Add prototypes.
(value_cast): Set 'eltype2' to an initial value to suppress a
possibly-used-before-initialized warning.
(value_of_variable): Add braces to remove nested if-else ambiguity.
(value_of_this): Ditto.
* valprint.c: Add prototypes.
(print_floating): Add braces to remove nested if-else ambiguity.
Looks like a big change, but it is really just a lot of small stuff.
In the cases where GCC was flagging a possible use-before-initialized
warning on variables, it turned out that these were being used properly
but GCC couldn't see that.
"wait.h" was defining all WIF* macro's instead of filling in those
that <wait.h> missed. Stops heaps of warnings from <wait.h>
re-defining WIF*s defined in "wait.h".
* defs.h: Move _initialize_printcmd, _initialize_stack,
_initialize_blockframe out of here and in to their respective .c files.
* blockframe.c: Move _initialize_blockframe prototype to here.
* printcmd.c: Move _initialize_printcmd prototype to here.
* stack.c: Move _initialize_stack prototype to here.
* source.c, symtab.h: Move _initialize_source prototype to the .c file.
* values.c, value.h: Move _initialize_values prototype to the .c file.
* gdbthread.h, thread.c: Move _initialize_thread prototype to the .c
file.
* breakpoint.c, breakpoint.h: Move _initialize_breakpoint prototype
to the .c file.
* abug-rom.c alpha-nat.c alpha-tdep.c annotate.c ax-gdb.c bcache.c:
Standardize comments for the prototype section of these files.
* configure.in: Look in libc for wctype before looking for it in libc.
The last one is to fix the GNU ld (~2.9.1) + Solaris 2.6 interaction problem
where an empty stub library (libw) causes a core dump when we call vasprintf
(e.g. `info br') in the final linked gdb.
version didn't work for sol2.6; pushed it to autoconf.
* configure.in (gdb_cv_proc_service_is_old): new test.
* acconfig.h (PROC_SERVICE_IS_OLD): new define.
* configure, config.in: regenerate.
* blockframe.c (find_pc_sect_partial_function): Add braces to avoid
possible nested-if confusion.
* breakpoint.c (breakpoint_here_p): Ditto.
(breakpoint_inserted_here_p): Ditto.
(breakpoint_thread_match): Ditto.
* gnu-regex.c: Define _REGEX_RE_COMP only if it isn't already defined.
* gnu-regex.h: Define _REGEX_RE_COMP to pick up old compatability
prototypes.
* symtab.h: Add prototype for _initialize_source.
* value.h: Add prototype for _initialize_value.
* defs.h: Include sys/types.h or stddef.h to get size_t.
(make_cleanup): Add make_cleanup_func typedef and switch to using
a prototype for this function.
(mfree): Add prototypes for mmalloc, mrealloc, mfree if we aren't
using mmalloc.
* ax-gdb.c breakpoint.c coffread.c corelow.c dbxread.c dwarf2read.c
dwarfread.c elfread.c eval.c exec.c gdbtk-cmds.c gdbtk.c infcmd.c
infrun.c mipsread.c nlmread.c os9kread.c parse.c printcmd.c symfile.c
symmisc.c symtab.c thread.c top.c tracepoint.c typeprint.c valops.c:
Cast parameters passed to make_cleanup to use the new
make_cleanup_func typedef.
More warning cleanups. There are still a bunch of places where the first
argument to make_cleanup is not cast to (make_cleanup_func); these are
either due to the function fitting the make_cleanup_func specification
already (e.g. free()) or they are in files that weren't compiled when
I did my make on a Linux native system. Bwahahaha. You can see them
like this:
grep make_cleanup\ * | grep -v make_cleanup_func
I'll surely go back and clean up the remaining suspicious calls in
GDB one of these days. :-)
* configure.in: Check for sys/debugreg.h, asm/debugreg.h.
* i386v-nat.c: Include asm/debugreg.h, sys/debugreg.h if it is not
present.
This is to work around a conflict where the Linux 2.1.x kernel and glibc
2.0.x are not in sync; including <sys/debugreg.h> will result in an error.
With luck, these losers will get their act together and we can trash
this hack in the near future.
address range of a compilation unit without children.
* mdebugread.c (parse_partial_symbols): Fix handling of stabs
continuations, use xmalloc and xrealloc.
There should be sufficient information/hooks now to eliminate
this hack.
* exec.c (file_command): Add a new hook here to inform ui's
when the exec file has changed. Adding it here allows the
ui to be informed after symbol reading.
* gdbcore.h: Add declaration of file_changed_hook.
* stabsread.c (rs6000_builtin_type): Create a complex float instead
of an error.
(read_sun_floating_type): Similarly.
(read_range_type): Create a complex float if self_subrange is
true.
* gdbtk-cmds.c (gdb_set_bp_addr): New command. Sets a
breakpoint at an address. Use this instead of gdb_cmd "break"
because the syntax of the break command is broken and doesn't
allow you to create a thread-specific BP at an address. Also
this is faster.
* gdbtk-cmds.c (gdb_set_bp): Add an optional thread number.
(gdb_find_bp_at_line): New function. Returns a list of bpnums
at the specified line number.
(gdb_find_bp_at_addr): New function. Returns a list of bpnums
at an address..
exec_file changes.
(gdbtk_add_hooks): Define exec_file_display_hook (to gdbtk_exec_file_changed)
* gdbtk-cmds.c (gdb_stop): target_stop is ALWAYS defined, so
compare against something a little more meaningful (target_ignore).
(remote_interrupt_twice): Remove. remote_stop now handles it.
(remote_stop): New function which handles interrupting the
remote target so that CLUI and GUI use the same core functions
to achieve the same goal.
(remote_wait): Change to handle remote_stop properly.
[interrupted_already]: New static global to help remote_stop.
[remote_ops, extended_remote_ops]: Add remote_stop for to_stop member.
* target.c: Rename static function "ignore" to "target_ignore" and
export it so that gdb can determin if some target vector member is
actually not defined. Replace all occurances of ignore.
* target.h: Export target_ignore.
Thu Oct 1 15:39:27 EDT 1998 Frank Ch. Eigler <fche@cygnus.com>
* breakpoint.c (bpstat_stop_status): Do not consider an
untripped watchpoint as a "hit".
* rdi-share/unixcomm.c: If using cygwin32, also use the SERPORT and
PARPORT defines for win32.
(Unix_MatchValidSerialDevice): For cygwin32, valid serial port names
start with "com", not "/dev/tty".
(Unix_OpenSerial): Do not use O_NONBLOCK on cygwin32.
* rdi-share/devsw.c (DevSW_Close): Free the device's state (SwitcherState)
so that the device may be reopened.
* remote-rdi.c (mywritec): Send all output through gdb's *_unfiltered
functions, ignoring non-ASCII chars, so that non-tty UI's can snarf
the output from fputs_hook.
(mywrite): Ditto.
(arm_rdi_open): Set inferior_pid.
(arm_rdi_detach): Pop the target off the target stack so that
users can attach and detach multiple times.
(arm_rdi_close): Close the opened device and reset inferior_pid, too.
* v850ice.c (WM_ADDR_TO_SYM): New message.
(v850ice_wndproc): Add handler for WM_SOURCE.
(v850ice_wait): Call the ui_loop_hook occasionally.
(ice_cont): Acknowledge message before doing anything.
(ice_stepi): Ack message and let gdbtk do stepping.
(ice_nexti): Ack message and let gdbtk do stepping.
(view_source): New function ICE calls to display source code.
* configure.in: Add --enable-warnings.
Adjust whitespace of other --with and --enable options so that
configure --help lines up correctly.
* aclocal.m4: Ditto.
* Makefile.in (WARN_CFLAGS): Add. Set by configure.
* configure: Regenerated.
* txvu-tdep.c (check_overlap): new function to guarantee that
overlays are either mapped or unmapped in their entirety.
* txvu-tdep.c (compress_blockvector): compress blockvectors correctly
* Fix formatting of live range splitting code.
* (resolve_symbol_reference define_symbol resolve_live_range): Change
errors to complaints so that bad live range symbols won't abort the
entire symbol table. Handle errors by aborting just the current
symbol.
* (ref_init): Goes away. Folded into ref_add().
* (REF_MAP_SIZE): Put parens around parameter so that args like
`1 + 2' get handled correctly (yes, this was a real bug).
* (ref_add): Remove check for allocation failures. Not necessary
when using xrealloc(). Fix pointer arithmetic problem when clearing
memory. This and the previous patch prevent random SEGV's when there
are lots of live range symbols.
* gdb.c++/classes.exp: Change all regular expressions to match
arbitrary combinations of newline/carriage-return, so that they
will work equally well on Unix and Windows.
* gdb.c++/inherit.exp: ditto.
* gdb.c++/virtfunc.exp: ditto.
* blockframe.c(find_pc_sect_partial_function): look for min syms in
the same section when trying to guess the end of a function.
* symfile.c(list_overlays_command): use print_address_numeric
* remote-sim.c: export simulator_command
* tm-r5900.h: add COP0 registers
* txvu-tdep.c: printvector and printvector-order commands
* tm-txvu.h: add COP0 registers
* mips-tdep.c: use NUM_CORE_REGS
* breakpoint.c (bpstat_stop_status): Declare a bp match if the
current fp matches the bp->fp OR if the current fp is less than
the bp->fp if we're looking at a bp_step_resume breakpoint.
PR 15992.
With a bp_step_resume type breakpoint, the frame pointer check exists in
case we recurse and execute the same text section--we want to not stop
there because it would confuse the user. So gdb checks to see that the
frame pointer matches the one recorded in the breakpoint.
If the current frame pointer is less than the frame pointer recorded in
the breakpoint, then something odd is happened and we should declare that
we're at the breakpoint so that we clear it out.
This happens in Purify instrumented executables on PA systems, at least.
They insert some calls out to Purify code in the function epilogue and
it confuses GDB (which assumes that the frame pointer does not move
around through the entire execution of a function).
* symtab.h (struct symtab): Remove EXTRA_SYMTAB_INFO hook,
not currently used.
* symfile.c (allocate_symtab): Deprecate use of
INIT_EXTRA_SYMTAB_INFO here.
pseudo-register, not the same as a3.
(D2_REGNUM, D3_REGNUM, A2_REGNUM, A3_REGNUM): Define.
* mn10300-tdep.c (fix_frame_pointer): New function.
(set_movm_offsets): Use register number macros instead of
hard-coded constants.
(mn10300_analyze_prologue): Fix to handle redefinition of FP_REGNUM.
(mn10300_frame_chain): Fix to handle redefinition of FP_REGNUM;
use register number macros instead of hard-coded constants;
add missing parameter to call of mn10300_analyze_prologue.
(mn10300_frame_saved_pc): Use register number macros instead of
hard-coded constants.
and "global constructors", do not append any elements to the result.
(gdb_get_vars_command): Install patch from Jun 9 1998 which appears
to have been lost.
Clean up some tabbing mess.
