of the PC being in a stack dummy when looking at a core file
without symbols. The other is the incorrect passing of char
arguments during expression evaluation (ie: p foo('a','b') would
mess up the passing of it's args because it wasn't coercing the
char's to ints).
* hppah-tdep.c: Rename global functions to have consistent hppa_
prefix. Make more functions static. Drop hp_ prefix from static
functions. (hppa_push_arguments): Call value_arg_coerce to cast
char to int args if necessary. (hppa_fix_call_dummy): Create
this routine from FIX_CALL_DUMMY macro in tm-hppa.h.
* inferior.h (PC_IN_CALL_DUMMY): Check for frame_address being
valid (ie: != 0) before doing comparison against PC.
* valops.c (call_function_by_hand): Adjust call to FIX_CALL_DUMMY
to reflect new arguments.
* config/pa/tm-hppa.h (POP_FRAME, PUSH_ARGUMENTS): Use new hppa_
prefix for func name. (FIX_CALL_DUMMY): Move code into
hppah-tdep.c.
* testsuite/gdb.t16/gdbme.c, testsuite/gdb.t17/gdbme.c: Add calls
to malloc() so that we can test GDB eval of dynamically created
arrays (like char strings in `print "foo"').
read_memory to get byte order right.
* hppah-tdep.c (find_unwind_info): Don't read in unwind info
anymore. This is done in paread.c now. We expect unwind info
to hang off of objfiles, and search all of the objfiles when until
we find a match.
* (skip_trampoline_code): Cast arg to target_read_memory.
* objfiles.h (struct objfile): Add new field obj_private to hold
per object file private data (unwind info in this case).
* paread.c (read_unwind_info): New routine to read unwind info
for the objfile. This data is hung off of obj_private.
* tm-hppa.h: Define struct obj_unwind_info, to hold pointers to
the unwind info for this objfile. Also define OBJ_UNWIND_INFO to
make this easier to access.
comment.
* gdbtypes.h, gdbtypes.c: Use const in decl of
cplus_struct_default, now that pa-gas assembler has been fixed.
* hppah-nat.c: Formatting.
* hppah-tdep.c: Remove lots of useless externs for variables we
don't use.
* (find_unwind_entry): Speed up by using binary search, and a one
entry cache.
* (rp_saved): New routine to see what unwind info says about RP
being saved on the stack frame.
* (frame_saved_pc): Look for prologue to see if we need to
examine the stack for the saved RP or not.
* (init_extra_frame_info): Check for prologue, instead of
framesize to determine if we are frameless or not.
* (frame_chain_valid): Stop backtraces when we run into _start.
* (push_dummy_frame): Reformat to make more readable.
* (find_dummy_frame_regs): ditto.
* (hp_pop_frame): ditto.
* (hp_restore_pc_queue): small cleanup.
* (hp_push_arguments): ditto.
* (pa_do_registers_info): ditto.
* (skip_prologue): New routine created from SKIP_PROLOGUE macro.
* tm-hppa.h: Move contents of SKIP_PROLOGUE into hppah-tdep.c.
* Define FRAME_CHAIN_VALID.
* Turn on BELIEVE_PCC_PROMOTION so that we can access char args
passed to functions.
* paread.c (pa_symtab_read): Use new bfd conventions for
accessing linker symbol table.
* (pa_symfile_init): Access embedded STAB info via BFD section
mechanism and related macros.
accessing minimal symbol data.
* infcmd.c (read_pc): Use #ifdef, not #if.
* symfile.c (syms_from_objfile): Add CONST to decl for targets.
* tm-hppa.h (FIX_CALL_DUMMY): Use new macros for accessing
minimal symbol data.
frame. Use find_return_regnum to find the caller.
* (find_unwind_entry): New routine to locate stack frame info
associated with a procedure. This looks in the $UNWIND_START$
section in the SOM file.
* (find_return_regnum): New routine. Uses find_unwind_entry() to
figure out where the caller's return address is stored.
* (find_proc_framesize): New routine. Uses find_unwind_entry()
to figure out the frame size for a procedure.
* (saved_pc_after_call): New routine, moved from tm-hppa.h.
* (init_extra_frame_info): New routine. Corrects PC and FP for
outermost frame if necessary.
* (frame_chain): New routine, moved from tm-hppa.h.
* (skip_trampoline_code): Handle computed function calls (ie:
calls from $$dyncall).
* (unwind_command): Temporary support function to allow user
to control/observe aspects of the unwind (stack frame) info.
* infcmd.c (read_pc): (Temporary), put a hack in to see if the PC
was in a system call, if so, then read the PC from r31.
* tm-hppah.h (SKIP_TRAMPOLINE_CODE, IN_SOLIB_TRAMPOLINE): Deal
with extra arg for skip_trampoline_code().
* (INIT_EXTRA_FRAME_INFO): Define to point at subr (see above).
* (FRAME_CHAIN, FRAME_CHAIN_VALID): Turn into real subroutines.
* tm-hppa.h (SAVED_PC_AFTER_CALL): Turn into real subroutine.
branch targets to compute target address using const from previous
instruction if necessary.
* Add `Q' operator to print out bit position field various
instructions.
* hppah-nat.c: #include sys/param.h, and sys/user.h. General
cleanups, use new code from Utah.
* (store_inferior_registers): Update to new code from Utah.
* (initialize_kernel_u_addr): Re-enable decl of struct user u.
* (fetch_register): Clear out priv level when reading PCs.
* hppah-tdep.c: Get rid of gobs of KERNELDEBUG stuff.
* Remove decl of errno, #include wait.h and target.h.
* (frame_saved_pc): Check `flags' pseudo-register to see if we
were inside of a kernel call. If so, then PC is in a different
register. Also, mask out bottom two bits of all PCs so as not to
confuse higher level code.
* (push_dummy_frame): Create from #define in tm-hppa.h.
* (find_dummy_frame_regs): Update from Utah.
* (hp_pop_frame): Create from #define in tm-hppa.h.
* (hp_restore_pc_queue): New, from Utah.
* (hp_push_arguments): Big fixes from Utah.
* (pa_do_registers_info, pa_print_registers): Only print out fp
regs upon request.
* (skip_trampoline_code): New routine to deal with stubs that
live in nowhereland between callers and callees.
* i860-tdep.c: Remove decl of attach_flag.
* infrun.c (wait_for_inferior): Add new macro
INSTRUCTION_NULLIFIED, which can tell if the instruction pointed
at by PC will be nullified. If so, then step the target once more
so as to avoid confusing the user.
* (just before step_over_function:): Use stop_func_start, not
stop_pc when checking for the existance of line number info.
stop_func_start will reflect the proper address of the target
routine, not of the stub that we may be traversing to get there.
* tm-hppa.h: define SKIP_TRAMPOLINE_CODE and IN_SOLIB_TRAMPOLINE
to deal with the stubs that PA compilers sometimes stick between
callers and callees. Also, define FLAGS_REGNUM for access to the
`flags' pseudo-reg.
* (REGISTER_CONVERT_TO_VIRTUAL, REGISTER_CONVERT_TO_RAW): Use
memcpy, not bcopy.
* (CANNOT_STORE_REGISTER): New from Utah. Says that we can't
write gr0, PC regs, and PSW!
* (FRAME_FIND_SAVED_REGS): Bug fixes from Utah.
* (PUSH_DUMMY_FRAME, POP_FRAME): Make into real routines in
hppah-nat.c.
* (CALL_DUMMY, FIX_CALL_DUMMY): Fixes from Utah.
* Define struct unwind_table_entry.
* valops.c (call_function_by_hand): Add another arg to
FIX_CALL_DUMMY (under #ifdef GDB_TARGET_IS_HPPA). Why is this
necessary?
decl for attach_flag, it now lives in inferior.h.
* hppa-pinsn.c: Reformat opcode tables. Add function prototypes.
Make most functions static.
* hppah-nat.c: General cleanups, remove BSD specific code (since
that all lives in hppab-nat.c).
* hppah-tdep.c (frame_chain_valid), tm-hppa.h (FRAME_CHAIN):
Change sense of test against inside_entry_file(). This fix is
from U. of Utah.
* tm-hppa.h (PUSH_DUMMY_FRAME, POP_FRAME): Use char * for 2nd arg
to read/write_register_bytes().
Stu Grossman
K. Richard Pixley
John Gilmore