* d10v-tdep.c (d10v_frame_chain_valid): Remove unnecessary tests.
* hppa-tdep.c (hppa_frame_chain_valid): Remove unnecessary test.
* blockframe.c: Include "gdbcmd.h" and "command.h".
(backtrace_below_main): New variable.
(file_frame_chain_valid, func_frame_chain_valid)
(nonnull_frame_chain_valid, generic_file_frame_chain_valid)
(generic_func_frame_chain_valid): Remove functions.
(frame_chain_valid, do_flush_frames_sfunc): New functions.
(_initialize_blockframe): New function.
* Makefile.in (blockframe.o): Update dependencies.
* frame.c (frame_saved_regs_id_unwind, get_prev_frame): Remove FIXME
comment. Call frame_chain_valid ().
* frame.h: Remove old prototypes. Add prototype for
frame_chain_valid and update comments to match.
* gdbarch.sh: Change FRAME_CHAIN_VALID into a predicated function.
Remove old comment.
* gdbarch.h: Regenerated.
* gdbarch.c: Regenerated.
Plus updates to all other targets not to provide a FRAME_CHAIN_VALID.
GDB_OSABI_UNINITIALIZED.
* gdbarch.sh: Add osabi to struct gdbarch and to struct
gdbarch_info. Include "osabi.h" in gdbarch.c. Check osabi
in gdbarch_list_lookup_by_info and in gdbarch_update_p.
* gdbarch.c: Regenerated.
* gdbarch.h: Regenerated.
* osabi.c (gdbarch_lookup_osabi): Return GDB_OSABI_UNINITIALIZED if
there's no BFD.
(gdbarch_init_osabi): Remove osabi argument; use info.osabi.
* osabi.h (enum gdb_osabi): Move to defs.h.
(gdbarch_init_osabi): Update prototype.
* defs.h (enum gdb_osabi): Moved here.
* Makefile.in: Update dependencies.
Plus updates to alpha, arm, hppa, i386, mips, ns32k, ppc, sh, sparc, and vax
ports to match.
* sparc-tdep.c: Include osabi.h.
(gdbarch_tdep): Add osabi member.
(_initialize_sparc_tdep): Use gdbarch_register.
(sparc_gdbarch_init): Use generic OS ABI framework.
(sparc_dump_tdep): New function.
* sparc-tdep.c (sparc_gdbarch_skip_prologue): Kill, duplicates
sparc_skip_prologue.
(sparc_skip_prologue): Kill frameless_p arg, and use line number
information to find prologue when possible.
(sparc_prologue_frameless_p): Call examine_prologue directly.
(sparc_gdbarch_init): Update set_gdbarch_skip_prologue call.
* config/sparc/tm-sparc.h (sparc_skip_prologue): Update for killed
second argument.
(SKIP_PROLOGUE): Likewise.
* sparc-tdep.h: New file.
* sparc-tdep.c (elf-bfd.h): Include.
(SPARC_HAS_FPU, FP_REGISTER_BYTES, FP_MAX_REGNUM,
SPARC_INTREG_SIZE, DUMMY_REG_SAVE_OFFSET, gdbarch_tdep,
CALL_DUMMY_CALL_OFFSET): Move out to sparc-tdep.h
(sparc_dump_tdep): New function.
(_initialize_sparc_tdep): Use gdbarch_register instead of
register_gdbarch_init which is deprecated.
(process_note_abi_tag_sections, sparc_abi_handler,
sparc_abi_handler_list, sparc_gdbarch_register_os_abi): New.
(sparc_gdbarch_init): Determine elf os ident key based upon
either the elf header or by using process_note_abi_tag_sections.
Fix 'gdbarch already exists' test by comparing os_ident as well.
As the last action, invoke OS abi_handler if installed.
instead of ->prev.
* z8k-tdep.c (z8k_frame_chain): Do not use ->prev.
* s390-tdep.c (s390_frame_chain): Do not use ->prev.
* rs6000-tdep.c (frame_get_saved_regs): Use rs6000_frame_chain()
instead of ->prev.
Adjust call_dummy_breakpoint_offset, so that `finish' after a stop
in a call dummy works.
(sparc_gdbarch_init): Fix setting of pc_in_call_dummy, it depends
on SPARC32/64_CALL_DUMMY_ON_STACK, not DO_CALL_DUMMY_ON_STACK.
* gregset.h: New file. Typedefs for gdb_gregset_t and
gdb_fpregset_t, prototypes for supply_gregset and friends.
* procfs.c: Include gregset.h. Delete local prototypes for
supply_gregset etc., and local typedef gdb_gregset_t etc.
* sol-thread.c: Include gregset.h, delete local prototypes,
add appropriate casts to gdb_gregset_t.
* uw-thread.c, lin-thread.c, core-sol2.c, core-regset.c,
sparc-tdep.c, ptx4-nat.c, ppc-linux-nat.c, mipsv4-nat.c,
m88k-nat.c, m68klinux-nat.c, m68k-tdep.c, irix5-nat.c,
irix4-nat.c, ia64-linux-nat.c, i386v4-nat.c, cxux-nat.c,
arm-linux-nat.c, alpha-nat.c: Include gregset.h.
* config/nm-linux.h: Define GDB_GREGSET_T, GDB_FPREGET_T.
* config/sparc/tm-sun4sol2.h: Ditto.
Make Sparc a Multi-Arch target. Discard PARAMS macro (require ANSI).
* sparc-tdep.c: include arch-utils.h.
(SPARC_HAS_FPU, FP_REGISTER_BYTES, FP_MAX_REG NUM, SPARC_INTREG_SIZE,
DUMMY_REG_SAVE_OFFSET): provide multi-arch-compatible definitions.
(GDB_TARGET_IS_SPARC64): make into a runtime test.
(struct frame_extra_info): Define, use instead of the macro.
(Many places): Use alloca instead of statically allocated buffers
that depend on a multi-arch variable such as MAX_REGISTER_RAW_SIZE.
(sparc_extract_struct_value_address): Accept a pointer arg instead
of an array sized by REGISTER_BYTES.
(examine_prologue): Accept a pointer to an array of CORE_ADDR,
instead of the defunct struct frame_saved_regs. Recognize new
Sparc64 store instructions as part of the prologue. Ignore the
destination of a frame store when parsing the prologue (so long
as it's on the stack).
(sparc_push_dummy_frame): Fix incorrect buffer offset for PSTATE.
(sparc_frame_find_saved_regs): Accept a ptr to an array of CORE_ADDR
instead of the defunct struct frame_saved_regs.
(supply_gregset): Discard unnecessary 'zerobuf': just send NULL to
supply_register. Provide 4-byte offset to compensate for diff
between size of the prgreg_t elements on a 64-bit host and size
of the registers for a 32-bit target. Fill all inaccessible regs
with zero so they won't keep being requested again and again.
(fill_gregset): Handle 32/64 size difference between registers
and prgreg_t. Handle as many new 64-bit regs as possible.
(supply_fpregset, fill_fpregset): Attempt to handle 64-bit world.
(sparc_push_arguments): Rename to sparc32_push_arguments.
Copy arguments into registers as well as onto stack, so that the
CALL_DUMMY (code pushed onto the target stack) is not required.
(sparc_extract_return_value): Rename to sparc32_extract_return_value.
(sparc_store_return_value): Use memset instead of bzero.
Use write_register_gen instead of write_register_bytes.
(sparclet_store_return_value): New function.
(_initialize_sparc_tdep): Call register_gdbarch_init to activate
the gdbarch multi-architecture system.
(sp64_push_arguments): Rename to sparc64_push_arguments.
Extend to store arguments in general registers as well as on stack.
(sparc64_extract_return_value): Rename to sp64_extract_return_value.
Use as a private function, to be called by the new external function
sparc64_extract_return_value.
(sparclet_extract_return_value): New function.
(sparc32_stack_align, sparc64_stack_align, sparc32_register_name,
sparc64_register_name, sparc_print_extra_frame_info,
sparclite_register_name, sparclet_register_name,
sparc_push_return_address, sparc64_use_struct_convention,
sparc32_store_struct_return, sparc64_store_struct_return,
sparc32_register_virtual_type, sparc64_register_virtual_type,
sparc32_register_size, sparc64_register_size,
sparc32_register_byte, sparc64_register_byte,
sparc_gdbarch_skip_prologue, sparc_convert_to_virtual,
sparc_convert_to_raw, sparc_frame_init_saved_regs,
sparc_frame_address, sparc_gdbarch_fix_call_dummy,
sparc_coerce_float_to_double, sparc_call_dummy_address,
sparc_y_regnum, sparc_reg_struct_has_addr, sparc_intreg_size,
sparc_return_value_on_stack): New functions supporting multi-arch.
(sparc_gdbarch_init): New function; initialize multi-arch.
(struct gdbarch_tdep): Define, use for private multi-arch data.
* config/sparc/tm-sparc.h: Move definitions around, enclose with
#ifdef GDB_MULTI_ARCH tests, provide some multi-arch alternate
definitions. Add enums for register names, to help debugging gdb.
This header file must work for non-multi-arch and for multi-arch.
* config/sparc/tm-sp64.h: Add GDB_MULTI_ARCH configuration. Also add
AT_ENTRY_POINT definitions for CALL_DUMMY, for non-multi-arch case.
Define GDB_MULTI_ARCH.
* config/sparc/tm-sparclet.h: Add GDB_MULTI_ARCH configuration.
Do not define GDB_MULTI_ARCH (bfd does not correctly identify target).
* config/sparc/tm-sparclite.h: Ditto.
* config/sparc/tm-sun4sol2.h: Define GDB_MULTI_ARCH.
* sparclet-rom.c (sparclet_regnames): Initialize explicitly, to
avoid using deprecated REGISTER_NAMES macro.
* Makefile.in: Let sparc-tdep.c depend on arch-utils.h.
* gdbarch.sh: Add FP0_REGNUM to list of multi-arched register numbers.
* gdbarch.c, gdbarch.h: Regenerate.
* core-regset.c: Change FP0_REGNUM to a runtime test.
* procfs.c: Ditto.
* sun3-nat.c: Ditto.
* sparc-tdep.c: Ditto.
* i386mach-nat.c: Remove unnecessary ifdef for FP0_REGNUM.
* ptx4-nat.c: Ditto.
* sol-thread.c (sol_thread_create_inferior): only add the thread
to the thread list if it is not already in there.
(prototypes for thread_db func ptrs): pretty up formatting.
set endianness based on machine type.
(_initialize_sparc_tdep): Initialize target_architecture_hook.
(sparc_print_register_hook): Print PSR and FPSR in fancy format
on 32-bit machines.
* config/sparc/tm-sparc.h (PRINT_REGISTER_HOOK): Redefine to
call sparc_print_register_hook instead of using inline code.
* config/sparc/tm-sp64.h (PRINT_REGISTER_HOOK): Remove.
work correctly on little-endian hosts.
(sparc_push_arguments): New function.
(gdb_print_insn_sparc): New function.
(_initialize_sparc_tdep): Make gdb_print_insn_sparc the default
disassembler, so that SPARClite-specific instructions will
be recognized.
* sparcl-tdep.c (readchar): Print debugging information.
(debug_serial_write): New function, a replacement for SERIAL_WRITE
that prints debugging information.
* config/sparc/tm-sparc.h (PUSH_ARGUMENTS): Define.
(sparc_push_arguments): Declare.
tm_print_insn_info.
* gdbtk.c (gdb_disassemble): Setup di.mach from
tm_print_insn_info.mach, and set endian from TARGET_BYTE_ORDER.
* i386-tdep.c (set_assembly_language_command): set
tm_print_insn_info.mach to the appropriate value for 386 or 8086
disassembly.
* printcmd.c (print_insn): Move init of disassembler_info to
_initialize_printcmd. Set endian for disassembler here.
* sparc-tdep.c: Set tm_print_insn_info.mach as appropriate to
select sparc/sparclite.
* config/sparc/{tm-sparc.h tm-sparclite.h}: Get rid of
TM_PRINT_INSN. Set TM_PRINT_INSN_MACH to
bfd_mach_sparc/bfd_mach_sparc_sparclite.
config/sparc/tm-sparclite.h: Initialize tm_print_insn from
TM_PRINT_INSN, which comes from the tm file.
* Makefile.in (INTERNAL_LDFLAGS): Add in flags from configure.
* configure configure.in: Only make sol-thread.o for native.
Also, switch to dlopened libthread_db.so.1.
* sol-thread.c: Switch to using dlopen to get the thread_db
library.
* configure, configure.in: Change test for libthread_db to only
work for configs where build/host/target are the same.
* monitor.h,monitor.c:
Added monitor flags MO_NO_ECHO_ON_SETMEM (don't expect echo on
setmem command), MO_RUN_FIRST_TIME (if command to start process
running on target is different from one to continue execution),
MO_HEX_PREFIX (if addresses from monitor have a "0x" prefix).
* monitor.c,parse.c,sparc-tdep.c:
Don't require strings in the registers array. This is to allow
NULLs to be place holders in the tm-*.h file so that only minor
changes are needed when a new processor is introduced (eg, one
without floating point).
* sparc-tdep.c: Conditionally remove dependancies on floating point.
* sparclet-rom.c,config/sparc/sparclet.mt,config/sparc/tm-sparclet.h:
New files for target sparclet.
* symfile.c: Add option for 2nd parameter on load command : a load
offset added to the vma of each section.
read_next_frame_reg to alpha_find_saved_regs, handle saved
floating point registers.
* mips-tdep.c: Move sigtramp handling of saved registers from
read_next_frame_reg to mips_find_saved_regs, handle saved
floating point registers.
* config/mips/tm-irix3.h, config/mips/tm-irix5.h,
config/mips/tm-mipsv4.h (SIGFRAME_FPREGSAVE_OFF): Define.
* sparc-tdep.c (sparc_pc_adjust): Fix check for `unimp'
instruction to handle functions returning structures with
large sizes properly.
with the real source file name.
* dbxread.c (end_psymtab): Handle static functions in the
SOFUN_ADDRESS_MAYBE_MISSING case by passing pst->filename
to lookup_minimal_symbol.
(process_one_symbol): Ignore extra outermost context from
SunPRO cc and acc.
* stabsread.c (define_symbol): Do not complain for SunPRO
static variable encoding if STATIC_TRANSFORM_NAME is defined.
* sparc-tdep.c, config/sparc/tm-sun4sol2.h
(sunpro_static_transform_name): Renamed from
solaris_static_transform_name.
* config/sparc/tm-sun4os4.h (STATIC_TRANSFORM_NAME):
Define to sunpro_static_transform_name for acc 3.0 compiled
executables.
* procfs.c, config/alpha/nm-osf2.h (PROCFS_DONT_TRACE_FAULTS):
Renamed from PROCFS_DONT_TRACE_IFAULT, don't trace any faults
if defined.
* procfs.c (info_proc_siginfo): Cast sip->si_addr to
`unsigned long' and use `lx' format for printing it.
* sparc-tdep.c (sparc_init_extra_frame_info): Set it.
(examine_prologue, sparc_init_extra_frame_info): Use ->frame plus
->sp_offset to compute the address something is saved at, not
->bottom.
* sparc-tdep.c (get_saved_register): New function.
* tm-sparc.h: Define GET_SAVED_REGISTER; don't define
FRAME_FIND_SAVED_REGS, HAVE_REGISTER_WINDOWS or REGISTER_IN_WINDOW_P.
* stack.c (frame_info): Add comment about what to do if
FRAME_FIND_SAVED_REGS is not defined.
* sparc-tdep.c (sparc_init_extra_frame_info): Set ->frame field
here. Get it right for flat frames.
* sparc-tdep.c (sparc_frame_chain): Instead of returning
meaningful value for ->frame field, just return dummy value.
This change is needed because the old code didn't deal with mixed
flat and non-flat frames.
* sparc-tdep.c (sparc_pop_frame): Write SP_REGNUM from
frame->frame, don't go through saved regs for this.
* sparc-tdep.c: Move guts of skip_prologue to new function
examine_prologue. Check for flat prologue and set is_flat.
Provide the caller with the information about what is saved where
if desired.
(skip_prologue, sparc_frame_find_saved_regs): Call examine_prologue.
* sparc-tdep.c: Replace union sparc_insn_layout and anonymous
union in isannulled, which won't work on a little-endian host,
with X_* macros.
* sparc-tdep.c (sparc_frame_saved_pc): If addr == 0, the saved PC
is still in %o7.
* config/sparc/tm-sparc.h: Define INIT_FRAME_PC and
INIT_FRAME_PC_FIRST.
* blockframe.c (get_prev_frame_info): Modify comments regarding
INIT_FRAME_PC_FIRST and the sparc.
* sparc-tdep.c (single_step): Use 4 not sizeof (long) for size of
instruction.
the return type of methods to avoid infinite loops with anonymous
types.
* valops.c (search_struct_field): Handle anonymous unions.
* sparc-tdep.c (sunos4_skip_trampoline_code): New function
to correctly handle steps into -g compiled PIC objects in the
main executable.
* config/sparc/tm-sun4os4.h (SKIP_TRAMPOLINE_CODE):
Redefine to use sunos4_skip_trampoline_code.
* dwarfread.c (DWARF_REG_TO_REGNUM): Provide a default mapping
from DWARF to GDB register numbering.
* dwarfread.c (locval): Use DWARF_REG_TO_REGNUM to map the
register value.
* config/mips/tm-mipsv4.h (DWARF_REG_TO_REGNUM): Define.
* symfile.h (ADD_PSYMBOL_VT_TO_LIST),
symfile.c (add_psymbol_to_list, add_psymbol_addr_to_list):
No longer demangle partial symbols.
* symtab.c (lookup_symbol, list_symbols): Handle mangled
variables, e.g. C++ static members, via the minimal symbols.
Handle reordered functions in an objfile, for Irix 5.2 shared
libraries.
* objfiles.h (OBJF_REORDERED): New bit in the objfile flags,
set if the functions in an objfile are reordered.
* mdebugread.c (parse_partial_symbols): Detect reordered
functions in an objfile.
* symtab.c (find_pc_psymtab, find_pc_symtab): Use expensive
lookup algorithm if the functions in the objfile are reordered.
* xcoffexec.c (exec_close): If the current target has a copy
of the exec_ops sections, reflect the freeing of the sections
in current_target.
* valops.c (call_function_by_hand): Use `sizeof dummy1', not
`sizeof dummy', for constructing the call dummy code.
* config/sparc/tm-sparc.h: Add PARAMS declarations to all
function declarations.
* sparc-tdep.c (sparc_pop_frame): Cast result of
read_memory_integer to CORE_ADDR when passing it to PC_ADJUST.
* irix5-nat.c (enable_break): Set breakpoint at the entry point
of the executable, to handle the case where main resides in a
shared library.
* irix5-nat.c (solib_create_inferior_hook): Reset stop_soon_quietly
after shared library symbol reading, to get rid of a warning from
heuristic_proc_start if the startup code has no symbolic debug info.
* breakpoint.h (struct breakpoint): Add new fields language
and input_radix, to enable breakpoint resetting with the
proper language and radix.
* breakpoint.c (set_raw_breakpoint): Initialize them.
(breakpoint_re_set_one): Use them when resetting the breakpoint.
(breakpoint_re_set): Preserve current language and input_radix
across breakpoint_re_set_one calls.
* symtab.c (decode_line_1): Do not build a canonical line
specification for `*expr' line specifications.
* breakpoint.h (bpstat_stop_status): Fix prototype declaration.
and stack pointer during stepping, to speed things up.
A. Changes to not select a frame until we need a selected frame:
* blockframe.c (flush_cached_frames): Call select_frame (NULL, -1).
* infrun.c (wait_for_inferior): Move call to select_frame back to
normal_stop. This reverts a change of 13 Apr 94 (it says Jeff
Law, but the change was my idea); the only reason for that change
was so we could save and restore the selected frame in
wait_for_inferior, and now that flush_cached frames clears the
selected frame, that should work OK now.
B. Changes to not create a current_frame until we need one:
* blockframe.c (get_current_frame): If current_frame is NULL, try
to create an innermost frame.
* sparc-tdep.c (sparc_pop_frame), infcmd.c (run-stack_dummy),
infrun.c (wait_for_inferior), thread.c (thread_switch),
convex-tdep.c (set_thread_command), a29k-tdep.c (pop_frame),
alpha-tdep.c (alpha_pop_frame), convex-xdep.c (core_file_command),
h8300-tdep.c (h8300_pop_frame), h8500-tdep.c (h8300_pop_frame),
hppa-tdep.c (hppa_pop_frame), i386-tdep.c (i386_pop_frame),
i960-tdep.c (pop_frame), m68k-tdep.c
(m68k_pop_frame), mips-tdep.c (mips_pop_frame), rs6000-tdep.c
(push_dummy_frame, pop_dummy_frame, pop_frame), sh-tdep.c
(pop_frame), config/arm/tm-arm.h (POP_FRAME),
config/convex/tm-convex.h (POP_FRAME), config/gould/tm-pn.h
(POP_FRAME), config/ns32k/tm-merlin.h (POP_FRAME),
config/ns32k/tm-umax.h (POP_FRAME), config/tahoe/tm-tahoe.h
(POP_FRAME), config/vax/tm-vax.h (POP_FRAME): Don't
call create_new_frame.
* corelow.c (core_open), altos-xdep.c (core_file_command),
arm-xdep.c (core_file_command), gould-xdep.c (core_file_command),
m3-nat.c (select_thread), sun386-nat.c (core_file_command),
umax-xdep.c (core_file_command): Don't call create_new_frame; do
call flush_cached_frames.
* blockframe.c (reinit_frame_cache): Don't call create_new_frame
or select_frame.
C. Changes to get rid of stop_frame_address and instead only
fetch the frame pointer when we need it.
* breakpoint.c (bpstat_stop_status): Remove argument
frame_address; use FRAME_FP (get_current_frame ()).
* infrun.c (wait_for_inferior): Don't pass frame pointer to
bpstat_stop_status.
* infrun.c (wait_for_inferior): Use FRAME_FP (get_current_frame
()) instead of stop_frame_address.
* infrun.c (save_inferior_status, restore_inferior_status),
inferior.h (struct inferior_status): Don't save and restore
stop_frame_address.
* inferior.h, infcmd.c, thread.c (thread_switch), m3-nat.c
(select_thread): Remove stop_frame_address and uses thereof.
D. Same thing for the stack pointer.
* infrun.c (wait_for_inferior): Remove stop_sp and replace
uses thereof with read_sp ().
E. Change to eliminate one nasty little spot where we were
wanting to know the frame pointer from before the current step
(idea from GDB 3.5, which saved my ass, because my other ideas of
how to fix it were very baroque).
* infrun.c: Remove prev_frame_address.
* infrun.c (wait_for_inferior, step_over_function): Use
step_frame_address instead of prev_frame_address.
F. Same basic idea for the stack pointer.
* inferior.h, infcmd.c: New variable step_sp.
* infcmd.c (step_1, until_next_command): Set it.
* infrun.c: Remove prev_sp and replace uses by step_sp.
* infrun.c (wait_for_inferior): If we get out of the step
range, then set step_sp to the current stack pointer before we
start going again.
find, to locate TM_FILE, XM_FILE, and NAT_FILE. This is faster
and means that these filenames no longer need be unique across all
the config/* directories.
* configure.in: Put the config/*/ into TM_FILE, etc.
* m68k-stub.c (computeSignal): Return SIGFPE, not SIGURG, for chk
and trapv exceptions.
* target.h (struct section_table), objfiles.h (struct obj_section):
Change name of field sec_ptr to the_bfd_section. More mnemonic
and avoids the (sort of, for the ptx compiler) name clash with
the name of the typedef.
* exec.c, xcoffexec.c, sparc-tdep.c, rs6000-nat.c, osfsolib.c,
solib.c, irix5-nat.c, objfiles.c, remote.c: Change users.
* utils.c: Include readline.h.
* Makefile.in (utils.o): Add dependency.
* remote.c (getpkt): Add support for run-length encoding.
in case execution was stopped in the called function.
* stack.c (print_frame_info, frame_info): If backtracing through
a call dummy, handle the starting source line number on a line
boundary like backtracing through sigtramp.
* sparc-tdep.c (sparc_frame_find_saved_regs): Get frame address
for call dummy frame right. Remove old test for dummy frame,
it has been unused at least since gdb-3.5.
* sparc-tdep.c (sparc_push_dummy_frame): Set return address register
of the dummy frame.
({extract,store}_{signed_integer,unsigned_integer,address}):
New routines to replace SWAP_TARGET_AND_HOST.
All over: All uses of SWAP_TARGET_AND_HOST on integers replaced.
* solib.[hc] (find_pc_section_from_so_list): new function and
prototype.
* sparc-tdep.c (in_solib_trampoline): new function.
* symfile.[hc] (find_pc_section): new function and prototypes.
* target.[hc] (find_pc_section_from_targets): new function and
prototypes.
* config/sparc/tm-sun4sol2.h (IN_SOLIB_TRAMPOLINE): redefine to
in_solib_trampoline.
