* resrc.c (define_icon): Fix storage of color attributes 'planes' and 'bit
count' in icon groups.

 Correct sym def order & pick right sym for .idata$7 reloc.

* sh-tdep.c: Running thru gdb_indent.sh.

* sh-tdep.c (sh_justify_value_in_reg): New function.
(sh_stack_allocsize): Ditto.
(flt_argreg_array): New array used for floating point argument
passing.
(sh_init_flt_argreg): New function.
(sh_next_flt_argreg): Ditto.
(sh_push_dummy_call_fpu): Simplify. Rename "odd_sized_struct" to
"pass_on_stack". Use new helper functions.  Accomodate Renesas ABI.
Fix argument passing strategy.
(sh_push_dummy_call_nofpu): Ditto.

2003-10-01  Phil Edwards  <pme@gcc.gnu.org>

* config-ml.in:  Use ac_configure_args directly instead of
ml_arguments.  Only set ml_norecursion if --no[-]recursion is
actually seen.

2003-10-01  Eric Botcazou  <ebotcazou@libertysurf.fr>

* config-ml.in: Propagate INSTALL variables.

2003-10-01  Andrew Cagney  <cagney@redhat.com>

* value.h (register_value_being_returned): Declare.  Replace
"value_being_returned".
* infcall.c (call_function_by_hand): Use
register_value_being_returned.
* infcmd.c (print_return_value): Call
"register_value_being_returned", handle struct return locally.
* values.c (register_value_being_returned): New function.  Replace
"value_being_returned".

daily update

*** empty log message ***

* strerror.c: Don't provide or reference sys_errlist if
strerror is available.

2003-09-24  Elena Zannoni  <ezannoni@redhat.com>

* linux-proc.c (linux_do_registers): New function.
(linux_make_note_section): Use linux_do_registers in case of
single threaded inferior programs.

* config/tc-mips.c (macro_build_ldst_constoffset,load_register,macro):
Unbreak overflow checks.

2003-10-01  Andrew Cagney  <cagney@redhat.com>

* infcall.c (call_function_by_hand): When STRUCT_RETURN, always
use STRUCT_ADDR.  When not using "struct return convention", pass
"0" to "value_being_returned".  Add FIXMEs.
* infcmd.c (print_return_value): Pass an explicit 0/1 to
value_being_returned.  Add comments.
* values.c (value_being_returned): Add fixme.
* hppa-tdep.c (hppa_extract_struct_value_address): Add FIXME.
(hppa_value_returned_from_stack): Add FIXME.

2003-10-01  H.J. Lu  <hongjiu.lu@intel.com>

* configure.in: Check if $MAKEINFO is missing.
* configure: Regenerated.

2003-10-01  Chris Demetriou  <cgd@broadcom.com>

        * config/tc-mips.c (s_cpreturn): Correct errors in comment.

* config/tc-mips.c (macro_build_ldst_constoffset): Fix sign extension
tests.
(load_register): Likewise.
(macro): Likewise.

* gas/mips/ldstla-32-shared.d: New file.
* gas/mips/ldstla-32.d: New file.
* gas/mips/ldstla-32.s: New file.
* gas/mips/ldstla-n32-shared.d: New file.
* gas/mips/ldstla-n32.d: New file.
* gas/mips/ldstla-n32.s: New file.
* gas/mips/ldstla-n64-shared.d: New file.
* gas/mips/ldstla-n64.d: New file.
* gas/mips/ldstla-n64.s: New file.
* gas/mips/mips.exp: Add ld-st-la tests.

    * xtensa-dis.c (fetch_data): Remove numBytes parameter.
        (print_insn_xtensa): Fix call to fetch_data.

daily update

*** empty log message ***

2003-09-30  David Carlton  <carlton@kealia.com>

* dwarf2read.c (struct die_info): Add 'parent' field; replace
'has_children' and 'next' by 'child' and 'sibling'.
(read_comp_unit): Rework algorithm, breaking body into
read_die_and_children and read_die_and_siblings.
(read_die_and_children, read_die_and_siblings): New.
(read_full_die): Add 'has_children' argument; set it instead of
the die's 'has_children' field.  Minor formatting cleanup.
(free_die_list): Use die->child and die->sibling instead of
die->next.
(dump_die_list): Ditto.
(sibling_die): Use die->sibling.
(psymtab_to_symtab_1): Use die's 'child' field in place of its
'has_children' and 'next' fields.
(process_die, read_file_scope, read_func_scope)
(read_lexical_block_scope, read_structure_scope)
(read_enumeration, read_array_type, read_common_block)
(read_namespace, read_subroutine_type, dump_die): Ditto.

* elfxx-ia64.c (ia64_howto_table): Fix size of R_IA64_TPREL64[LM]SB,
R_IA64_DTPREL{32,64}[LM]SB and R_IA64_DTPMOD64[LM]SB.

2003-09-29  Andrew Cagney  <cagney@redhat.com>

* rs6000-tdep.c (rs6000_gdbarch_init): Set the PowerOpen red zone
to 224, not 220.

[ gas/ChangeLog ]
2003-09-30  Chris Demetriou  <cgd@broadcom.com>

        * config/tc-mips.c (mips_ip): Capitalize first word of
        existing condition code warning, and add condition code
        warnings for .ps instructions, and for bc1any[24][tf].

[ gas/testsuite/ChangeLog ]
2003-09-30  Chris Demetriou  <cgd@broadcom.com>

        * lib/gas-defs.exp (run_dump_test): If stderr file is specified
        and there was no stderr output, compare anyway (i.e., cause a
        test failure).
        * gas/mips/mips64-mips3d.s: Add some new instructions to test warnings.
        * gas/mips/mips64-mips3d.l: New file.
        * gas/mips/mips64-mips3d.d: Use mips64-mips3d.l, and update for
        changes to mips64-mips3d.s.
        * gas/mips/mips64-mips3d-incl.d: Likewise.
        * gas/mips/set-arch.l: New file.
        * gas/mips/set-arch.d: Specify set-arch.l as stderr output to check.
        * gas/mips/mips5.l: Make error messages match source.

2003-09-30  H.J. Lu  <hongjiu.lu@intel.com>

* ld-checks/checks.exp (section_check): Remove ia64-*-elf*.

2003-09-30  H.J. Lu  <hongjiu.lu@intel.com>

* ld-checks/checks.exp (section_check): Skip ia64-*-* instead
of ia64-*-linux*.

2003-09-30  Andrew Cagney  <cagney@redhat.com>

* mi-main.c: Rename REGISTER_VIRTUAL_SIZE to
DEPRECATED_REGISTER_VIRTUAL_SIZE.

2003-09-30  Andrew Cagney  <cagney@redhat.com>

* gdbarch.sh (DEPRECATED_REGISTER_VIRTUAL_SIZE): Rename
REGISTER_VIRTUAL_SIZE.
* gdbarch.h, gdbarch.c: Regenerate.
* vax-tdep.h, sparc-tdep.c, regcache.h: Update.
* regcache.c, mn10300-tdep.c, mips-tdep.c: Update.
* infcmd.c, frame.c, findvar.c, cris-tdep.c: Update.

2003-09-30  Andrew Cagney  <cagney@redhat.com>

* gdbint.texinfo (Target Architecture Definition): Rename
REGISTER_VIRTUAL_SIZE to DEPRECATED_REGISTER_VIRTUAL_SIZE.
(Target Architecture Definition):

2003-09-30  H.J. Lu  <hongjiu.lu@intel.com>

* ld-ia64/ia64.exp: Remove -melf64_ia64.

* ld-ia64/tlsbin.dd: Remove -melf64_ia64. Match elf..-ia64-.*
instead of elf64-ia64-little.
* ld-ia64/tlsbin.rd: Likewise.
* ld-ia64/tlsbin.sd: Likewise.
* ld-ia64/tlsbin.td: Likewise.
* ld-ia64/tlsg.sd: Likewise.
* ld-ia64/tlspic.dd: Likewise.
* ld-ia64/tlspic.rd: Likewise.
* ld-ia64/tlspic.sd: Likewise.
* ld-ia64/tlspic.td: Likewise.

2003-09-30  Chris Demetriou  <cgd@broadcom.com>

        * NEWS: Mention support for MIPS64 Release 2.

[ binutils/ChangeLog ]
2003-09-30  Chris Demetriou  <cgd@broadcom.com>

        * NEWS: Add an indication of the cutoff for 2.14.

[ gas/ChangeLog ]
2003-09-30  Chris Demetriou  <cgd@broadcom.com>

        * NEWS: Add an indication of the cutoff for 2.14.

[ ld/ChangeLog ]
2003-09-30  Chris Demetriou  <cgd@broadcom.com>

        * NEWS: Add an indication of the cutoff for 2.14.

[ bfd/ChangeLog ]
2003-09-30  Chris Demetriou  <cgd@broadcom.com>

* archures.c (bfd_mach_mipsisa64r2): New define.
* bfd-in2.h: Regenerate.
* aoutx.h (NAME(aout,machine_type)): Handle bfd_mach_mipsisa64r2.
* cpu-mips.c (I_mipsisa64r2): New enum value.
(arch_info_struct): Add entry for I_mipsisa64r2.
* elfxx-mips.c (_bfd_elf_mips_mach)
(_bfd_mips_elf_print_private_bfd_data): Handle E_MIPS_ARCH_64R2.
(mips_set_isa_flags): Add bfd_mach_mipsisa64r2 case.
(mips_mach_extensions): Add entry for bfd_mach_mipsisa64r2.

[ binutils/ChangeLog ]
2003-09-30  Chris Demetriou  <cgd@broadcom.com>

* readelf.c (get_machine_flags): Handle E_MIPS_ARCH_64R2.

[ gas/Changelog ]
2003-09-30  Chris Demetriou  <cgd@broadcom.com>

* configure.in (mipsisa64r2, mipsisa64r2el, mipsisa64r2*): New CPUs.
* configure: Regenerate.
* config/tc-mips.c (imm2_expr): New variable.
(md_assemble, mips16_ip): Initialize imm2_expr.
(ISA_HAS_64BIT_REGS, ISA_HAS_DROR, ISA_HAS_ROR): Add ISA_MIPS64R2.
(macro_build): Handle +A, +B, +C, +E, +F, +G, and +H format operands.
(macro): Handle M_DEXT and M_DINS.
(validate_mips_insn): Handle +E, +F, +G, +H, and +I format operands.
(mips_ip): Likewise.
(OPTION_MIPS64R2): New define.
(md_longopts): New entry for -mips64r2 (OPTION_MIPS64R2).
OPTION_ASE_BASE): Increase to compensate for OPTION_MIPS64R2.
(md_parse_option): Handle OPTION_MIPS64R2.
(s_mipsset): Handle setting "mips64r2" ISA.
(mips_cpu_info_table): Add mips64r2.
(md_show_usage): Document -mips64r2 option.
* doc/as.texinfo: Docuemnt -mips64r2 option.
* doc/c-mips.texi: Likewise.

[ gas/testsuite/ChangeLog ]
2003-09-30  Chris Demetriou  <cgd@broadcom.com>

* gas/mips/cp0-names-mips64r2.d: New file.
* gas/mips/cp0sel-names-mips64r2.d: New file.
* gas/mips/elf_arch_mips64r2.d: New file.
* gas/mips/hwr-names-mips64r2.d: New file.
* gas/mips/mips32r2-ill-fp64.l: New file.
* gas/mips/mips32r2-ill-fp64.s: New file.
* gas/mips/mips64r2-ill.l: New file.
* gas/mips/mips64r2-ill.s: New file.
* gas/mips/mips64r2.d: New file.
* gas/mips/mips64r2.s: New file.
* gas/mips/mips.exp: Define "mips64r2" arch, and run new tests.

[ include/elf/ChangeLog ]
2003-09-30  Chris Demetriou  <cgd@broadcom.com>

* mips.h (E_MIPS_ARCH_64R2): New define.

[ include/opcode/ChangeLog ]
2003-09-30  Chris Demetriou  <cgd@broadcom.com>

* mips.h: Document +E, +F, +G, +H, and +I operand types.
Update documentation of I, +B and +C operand types.
(INSN_ISA64R2, ISA_MIPS64R2, CPU_MIPS64R2): New defines.
(M_DEXT, M_DINS): New enum values.

[ ld/ChangeLog ]
2003-09-30  Chris Demetriou  <cgd@broadcom.com>

* ldmain.c (get_emulation): Ignore "-mips64r2".

[ ld/testsuite/ChangeLog ]
2003-09-30  Chris Demetriou  <cgd@broadcom.com>

* ld-mips-elf/mips-elf-flags.exp: Add tests for combinations
with MIPS64r2.

[ opcodes/ChangeLog ]
2003-09-30  Chris Demetriou  <cgd@broadcom.com>

* mips-dis.c (mips_arch_choices): Add entry for "mips64r2"
(print_insn_args): Add handing for +E, +F, +G, and +H.
* mips-opc.c (I65): New define for MIPS64r2.
(mips_builtin_opcodes): Add "dext", "dextm", "dextu", "dins",
"dinsm", "dinsu", "drotl", "drotr", "drotr32", "drotrv", "dsbh",
and "dshd" for MIPS64r2.  Adjust "dror", "dror32", and "drorv" to
be supported on MIPS64r2.

2003-09-29  Andrew Cagney  <cagney@redhat.com>

* gdbarch.sh (DEPRECATED_REGISTER_VIRTUAL_TYPE): Rename
DEPRECATED_REGISTER_VIRTUAL_TYPE.
* gdbarch.h, gdbarch.c: Regenerate.
* arch-utils.c, hppa-tdep.c, regcache.c, regcache.h: Update.
* sh64-tdep.c, sparc-tdep.c: Update.

2003-09-30  Andrew Cagney  <cagney@redhat.com>

* gdbint.texinfo (Target Architecture Definition): Rename
REGISTER_VIRTUAL_TYPE to DEPRECATED_REGISTER_VIRTUAL_TYPE.

2003-09-30  Andrew Cagney  <cagney@redhat.com>

* remote-vxsparc.c (vx_read_register): Replace bzero with memset.
* remote-vxmips.c (vx_read_register): Ditto.
* remote-vx68.c (vx_read_register): Ditto.
* gnu-nat.c (inf_validate_procs): Ditto.

2003-09-29  H.J. Lu  <hongjiu.lu@intel.com>

* lib/ld-lib.exp (proc is_elf64): New.

* ld-scripts/phdrs.exp: Use is_elf_format and is_elf64.

2003-09-29  H.J. Lu  <hongjiu.lu@intel.com>

* readelf.c (dump_relocations): Support SHN_IA_64_ANSI_COMMON.
(get_symbol_index_type): Likewise.

testsuites/

2003-09-29  H.J. Lu  <hongjiu.lu@intel.com>

* binutils-all/readelf.exp: Use is_elf_format.

* binutils-all/readelf.h: Support IA64/ILP32.
* binutils-all/readelf.s: Likewise.
* binutils-all/readelf.ss: Likewise.

* lib/utils-lib.exp (proc is_elf_format): Copy from ld
testsuite. Add ia64-*-hpux*.

*** empty log message ***

daily update

2003-09-29  H.J. Lu  <hongjiu.lu@intel.com>

* elfxx-ia64.c (elfNN_hpux_backend_symbol_processing): New.
Handle SHN_IA_64_ANSI_COMMON.
(elf_backend_section_from_bfd_section): Defined.

    * infcall.c (call_function_by_hand): Fix build failure
        introduced in the previous change to this file.

2003-09-29  Andrew Cagney  <cagney@redhat.com>

* gdbint.texinfo (Target Architecture Definition): Delete
documentation for NEED_TEXT_START_END.

2003-09-29  Andrew Cagney  <cagney@redhat.com>

* gdbint.texinfo (Target Architecture Definition): Rename
NPC_REGNUM to DEPRECATED_NPC_REGNUM.  Add cross reference to
TARGET_WRITE_PC.

2003-09-29  Andrew Cagney  <cagney@redhat.com>

* NEWS: Mention Objective-C.

2003-09-29  Dave Brolley  <brolley@redhat.com>

        * gas/frv/allinsn.s (nldqi): Removed. Replaced with nop.
        * gas/frv/allinsn.d (nldqi): Ditto.

* ChangeLog: Correct an entry command.exp -> commands.exp.
* gdb.base/commands.exp (bp_deleted_in_command_test)
(temporary_breakpoint_commands): Check noargs.

2003-09-29  Jerome Guitton  <guitton@act-europe.fr>

* arm-tdep.c (arm_make_prologue_cache): Use trad_frame_addr_p to
test if the register has been saved on the stack.
(arm_scan_prologue_cache): When analysing the instruction
"str lr, [sp, #-4]", save the address where lr has been stored.

daily update

*** empty log message ***

2003-09-28  Andrew Cagney  <cagney@redhat.com>

* frame.c (frame_read_unsigned_register): Delete function.
* frame.h (frame_read_unsigned_register): Delete declaration.
* sparc-tdep.c (sparc_init_extra_frame_info): Use
get_frame_register_unsigned.
(sparc_frame_saved_pc, sparc_pop_frame): Ditto.
* m68hc11-tdep.c (m68hc11_print_register): Ditto.
* d10v-tdep.c (d10v_print_registers_info): Ditto.

* frame.h (frame_read_register): Delete declaration.
* frame.c (frame_read_register): Delete function.
* arch-utils.c (legacy_register_to_value): Use get_frame_register.
* sparc-tdep.c (sparc_fetch_pointer_argument): Ditto.
* rs6000-tdep.c (rs6000_fetch_pointer_argument): Ditto.
* mips-tdep.c (mips_register_to_value): Ditto.
* hppa-tdep.c (hppa_fetch_pointer_argument): Ditto.
* d10v-tdep.c (d10v_print_registers_info): Ditto.

* frame.c (frame_read_signed_register): Delete function.
(frame_read_unsigned_register): Update comments.
* frame.h (frame_read_signed_register): Delete declaration.
* h8300-tdep.c (h8300_print_register): Use
get_frame_register_signed.
* m68hc11-tdep.c (m68hc11_print_register): Ditto.

2003-09-28  Andrew Cagney  <cagney@redhat.com>

* config/pa/tm-hppa.h (DEPRECATED_VALUE_RETURNED_FROM_STACK):
Rename VALUE_RETURNED_FROM_STACK.
* infcmd.c (print_return_value): Update.
* infcall.c (call_function_by_hand): Update.

* i387-tdep.c (i387_supply_fsave, i387_supply_fxsave): Add
regcache argument and reverse the order of the other two
arguments.  Remove local regcache variable.  Determine
architecture from REGCACHE.  Update comments.
* x86-64-tdep.c (x86_64_supply_fxsave): Add regcache argument and
reverse the order of the other two arguments.  Remove local
regcache variable.  Determine architecture from REGCACHE.  Update
comments.
* i387-tdep.h (i387_supply_fsave, i387_supply_fxsave): Adjust
prototypes.  Update comments.
* x86-64-tdep.c (x86_64_supply_fxsave): Adjust prototype.  Adjust
comment.
* amd64fbsd-nat.c (supply_fpregset, fetch_inferior_registers):
Update.
* go32-nat.c (fetch_register, go32_fetch_registers): Update.
* i386-interix-nat.c (supply_fpregset): Update.
* i386-linux-nat.c (supply_fpregset, supply_fpxregset): Update.
* i386-nto-tdep.c (i386nto_supply_fpregset): Update.
* i386gnu-nat.c (fetch_fpregs, supply_fpregset): Update.
* i386bsd-nat.c (supply_fpregset, fetch_inferior_registers): Update.
* i386nbsd-tdep.c (fetch_core_registers, fetch_elfcore_registers):
Update.
* i386obsd-tdep.c (fetch_core_registers): Update.
* i386v4-nat.c (supply_fpregset): Update.
* x86-64-linux-nat.c (supply_fpregset): Update.
* x86-64-linux-tdep.c (fetch_core_registers): Update.

*** empty log message ***

daily update

* i386-tdep.h: Put opaque declarations in alphabetical
order.  Remove spurious whitespace.
(struct gdbarch_tdep): add st0_regnum and mm0_regnum members.
(i386_sse_regnum_p, i386_mxcsr_regnum_p): Remove prototypes.
* i386-tdep.c (MM0_REGNUM): Remove define.
(i386_mmx_regnum_p): Add gdbarch argument.
(i386_sse_regnum_p, i386_mxcsr_regnum_p): Add gdbarch argument.
Rewrite using new macro definitions for FPU/SSE registers.
(i386_fp_regnum_p, i386_fpc_regnum_p): Rewrite using new macro
definitions from i387-tdep.h.
(i386_register_name): Update.
(i386_stab_reg_to_regnum, i386_dwarf_reg_to_regnum): Update to use
new macro definitions for FPU/SSE registers.
(i386_extract_return_value): Determine whether floating-point
registers are present by examining REGCACHE's architecture.
(i386_store_return_value): Likewise.  Use I386_MAX_REGISTER_SIZE
instead of FPU_REG_RAW_SIZE.  Use new macro definitions for
FPU/SSE registers.
(i386_register_type): Update.
(i386_mmx_regnum_to_fp_regnum): Rewrite using new macro
definitions for FPU registers.  Use REGCACHE's architecture to
determine the appropriate register numbers.
(i386_pseudo_register_read, i386_pseudo_register_write,
i386_register_reggroup_p): Update.
(i386_gdbarch_init): Initialize TDEP->st0_regnum and
TDEP->mm0_regnum.
* i387-tdep.h (I387_FCTRL_REGNUM, I387_FSTAT_REGNUM,
I387_FTAG_REGNUM, I387_FISEG_REGNUM, I387_FIOFF_REGNUM,
I387_FOSEG_REGNUM, I387_FOOFF_REGNUM, I387_FOP_REGNUM,
I387_XMM0_REGNUM, I387_MXCSR_REGNUM): New defines.
(i387_supply_fsave, i387_fill_fsave, i387_supply_fxsave,
i387_fill_fxsave): Change type of fsave/fxsave argument from `char
*' to `void *'.
* i387-tdep.c (i387_print_float_info, fsave_offset, FSAVE_ADDR,
i387_supply_fsave, i387_fill_fsave, fxsave_offset, FXSAVE_ADDR,
i387_supply_fxsave, i387_fill_fxsave): Update to use new macro
definitions for FPU/SSE registers.
(FXSAVE_MXCSR_ADDR): New define.
* x86-64-tdep.c (x86_64_init_abi): Override TDEP->st0_regnum and
TDEP->mm0_regnum.
(I387_FISEG_REGNUM, I387_FOSEG_REGNUM): Remove defines.
(I387_ST0_REGNUM): Define.

* regcache.h (get_regcache_arch): New prototype.
* regcache.c (get_regcache_arch): New function.

* config/tc-hppa.c (pa_ip): Check for invalid 64-bit conditions.

* x86-64-tdep.c (x86_64_store_return_value): Remove spurious
whitespace.

* i386-tdep.c (i386_num_register_names, i386_num_mmx_regs):
Initialize using ARRAY_SIZE.

2003-09-27  Andrew Cagney  <cagney@redhat.com>

* arch-utils.c (deprecated_init_frame_pc_default): Rename
"init_frame_pc_default".
* arch-utils.h (deprecated_init_frame_pc_default): Update.
* xstormy16-tdep.c (xstormy16_gdbarch_init): Update.
* vax-tdep.c (vax_gdbarch_init): Update.
* v850-tdep.c (v850_gdbarch_init): Update.
* sh64-tdep.c (sh64_gdbarch_init): Update.
* s390-tdep.c (s390_gdbarch_init): Update.
* ns32k-tdep.c (ns32k_gdbarch_init): Update.
* mcore-tdep.c (mcore_gdbarch_init): Update.
* h8300-tdep.c (h8300_gdbarch_init): Update.
* cris-tdep.c (cris_gdbarch_init): Update.
* config/pa/tm-hppa.h (DEPRECATED_INIT_FRAME_PC): Update.

daily update

*** empty log message ***

2003-09-26  Matt Thomas  <matt@3am-software.com>

* config/tc-vax.c (md_shortopts): Fix a typo.  Remove 'K'.
(OPTION_PIC): Define.
(md_longopts): Add "pic" option.
(md_parse_option): Change 'K' to OPTION_PIC.

* regset.h: New file.
* Makefile.in (regset_h): Add.

2003-09-25  H.J. Lu  <hongjiu.lu@intel.com>

* gas/ia64/alias-ilp32.d: New file.
* gas/ia64/secname-ilp32.d: Likewise.
* gas/ia64/unwind-ilp32.d: Likewise.

* gas/ia64/dependency-1.d: Support ILP32.
* gas/ia64/dv-branch.d: Likewise.
* gas/ia64/dv-imply.d: Likewise.
* gas/ia64/dv-mutex.d: Likewise.
* gas/ia64/dv-safe.d: Likewise.
* gas/ia64/dv-srlz.d: Likewise.
* gas/ia64/global.d: Likewise.
* gas/ia64/ldxmov-1.d: Likewise.
* gas/ia64/ltoff22x-1.d:Likewise.
* gas/ia64/opc-a.d: Likewise.
* gas/ia64/opc-b.d: Likewise.
* gas/ia64/opc-f.d: Likewise.
* gas/ia64/opc-i.d: Likewise.
* gas/ia64/opc-m.d: Likewise.
* gas/ia64/opc-x.d: Likewise.
* gas/ia64/order.d: Likewise.
* gas/ia64/regs.d: Likewise.
* gas/ia64/tls.d: Likewise.

* gas/ia64/ia64.exp: Run "secname-ilp32", "unwind-ilp32" and
"alias-ilp32" for ia64-*-hpux*.

*** empty log message ***

2003-09-25  Dave Brolley  <brolley@redhat.com>

        * reset.c (frv_initialize): Call frv_register_control_init first.

2003-09-25  Andrew Cagney  <cagney@redhat.com>

* frame.h (deprecated_frame_saved_regs): Rename
get_frame_saved_regs.
* cris-tdep.c, frame.c, h8300-tdep.c, hppa-tdep.c: Update.
* mcore-tdep.c, mips-tdep.c, mn10300-tdep.c: Update.
* ns32k-tdep.c, ppc-linux-tdep.c, rs6000-tdep.c: Update.
* s390-tdep.c, sh64-tdep.c, stack.c: Update.
* v850-tdep.c, vax-tdep.c, xstormy16-tdep.c: Update.

2003-09-25  H.J. Lu  <hongjiu.lu@intel.com>

* readelf.c (process_version_sections): Indent version names.

2003-09-25  Andrew Cagney  <cagney@redhat.com>

* NEWS: Mention the new backtrace mechanism, DWARF 2 CFI, hosted
file I/O, multi-arch, TLS and NPTL, DWARF 2 Location Expressions,
and Java.
* PROBLEMS: Mention that mips*-*-*, powerpc*-*-*, sparc*-*-* and
arm*-*-* do not use the new frame code.

2003-09-25  David Carlton  <carlton@kealia.com>

* c-exp.y: Remove 'register' declarations.
* f-exp.y, jv-exp.y, m2-exp.y, objc-exp.y, p-exp.y: Ditto.

2003-09-25  David Carlton  <carlton@kealia.com>

* c-exp.y: Include cp-support.h.  Add qualified_type.
(yylex): Delete nested type hack; add comments.
* cp-namespace.c (cp_lookup_nested_type): New function.
* cp-support.h: Declare cp_lookup_nested_type.
* eval.c (evaluate_subexp_standard): Call value_aggregate_elt
instead of value_struct_elt_for_reference.
* valops.c: Include cp-support.h.
(value_aggregate_elt): New function.
(value_namespace_elt): Ditto.
(value_struct_elt_for_reference): Make static.
* value.h: Delete declaration of value_struct_elt_for_reference;
add declaration for value_aggregate_elt.
* Makefile.in (c-exp.tab.o): Depend on $(cp_support_h).
(valops.o): Ditto.

2003-09-25  David Carlton  <carlton@kealia.com>

* gdb.cp/namespace.exp: Tweak comments.  Add non-quoted versions
of some print tests, where appropriate.  Add tests for C::D::cd,
E::ce, F::cXfX, G::XgX.
* gdb.cp/namespace.cc: Add XgX, cXfX, ce.

* lib/java.exp (java_init): Import target_alias before using it.

* stack.c: Include "reggroups.h".
(frame_info): Only display registers in all_reggroup.
* Makefile.in (stack.o): Update dependencies.

2003-09-25  David Carlton  <carlton@kealia.com>

* gdb.base/corefile.exp: Delete obsolete setup_xfail.

* acx.m4 (AC_PROG_CPP_WERROR): New.

* configure.in: Use AC_PROG_CPP_WERROR.
* configure: Regenerated.

* arm-tdep.c (arm_skip_prologue): Handle "sub ip, sp #n" and
"add ip, sp #n" in the prologue.
(arm_scan_prologue): Ditto.

* MAINTAINERS (write after approval): Add myself.

* gas/cris/rd-pcrel2.d: Update for changes in section flag
defaults.

* m68k-tdep.c: Include "dwarf2-frame.h".
(m68k_gdbarch_init): Add the DWARF CFI frame unwinder.
* Makefile.in (m68k-tdep.o): Update dependencies.

* sh-tdep.c (struct frame_extra_info): Remove.
(struct sh_frame_cache): New structure.
(GET_SOURCE_REG): New macro extracting source register of an opcode.
(GET_TARGET_REG): Ditto but target register.
(GET_PUSHED_REG): Remove.
(IS_MOV_ARG_TO_REG): New macro.
(IS_MOV_ARG_TO_IND_R14): New macro.
(IS_MOV_ARG_TO_IND_R14_WITH_DISP): New macro.
(IS_MOVW_PCREL_TO_REG): New macro.
(IS_MOVL_PCREL_TO_REG): New macro.
(IS_SUB_REG_FROM_SP): New macro.
(IS_ARG_MOV): Remove.
(IS_MOV_TO_R14): Remove.
(IS_RESTORE_FP): New macro.
(IS_RTS): New macro.
(IS_LDS): New macro.
(IS_MOV_FP_SP): New macro.
(IS_ADD_REG_TO_FP): New macro.
(IS_ADD_IMM_FP): New macro.
(sh_skip_prologue_hard_way): Remove.
(sh_saved_pc_after_call): Remove.
(sh_frame_chain): Remove.
(sh_find_callers_reg): Remove.
(sh_nofp_frame_init_saved_regs): Remove.
(sh_fp_frame_init_saved_regs): Remove.
(sh_init_extra_frame_info): Remove.
(sh_analyze_prologue): New function.
(sh_skip_prologue): Remove deprecated code.  Rely on new function
sh_analyze_prologue when after_prologue fails.
(sh_frame_saved_pc): Remove.
(sh_alloc_frame_cache): New function.
(sh_frame_cache): Ditto.
(sh_frame_prev_register): Ditto.
(sh_frame_this_id): Ditto.
(sh_frame_unwind): New structure defining the heuristic frame
sniffer interface.
(sh_frame_sniffer): New function.
(sh_unwind_sp): Ditto.
(sh_unwind_pc): Ditto.
(sh_unwind_dummy_id): Ditto.
(sh_frame_base_address): Ditto.
(sh_frame_base): New structure defining new frame base code.
(sh_in_function_epilogue_p): New function.
(sh_gdbarch_init): Restructure and simplify to eliminate deprecated
code and to call all new code instead.  Initialize dwarf2 and
heuristic frame sniffer.

* parser-defs.h (struct exp_descriptor): New definition, containing
language-specific info for printing, prefixifying, dumping, and
evaluating expressions.
(exp_descriptor_standard): Declare new variable.
(print_subexp): Make global and declare here (from expprint.c).
(dump_subexp): Ditto.
(dump_subexp_body_standard): Declare.
(operator_length_standard): Declare.
(op_name_standard): Declare.
(print_subexp): Declare.
(print_subexp_standard): Declare.

* language.h (struct language_defn): Add la_exp_desc field to hold
pointer to table for language-specific operators.
Remove evaluate_exp field, which is now in struct exp_descriptor.

* parse.c (operator_length): Move most code to new
operator_length_standard function.  Use language-specific information.
(operator_length_standard): New function taking most code from
operator_length.
(exp_descriptor_standard): New constant.

* expression.h (enum exp_opcode): Add definitions of OP_EXTENDED0
and OP_EXTENDED_LAST.

* expprint.c (print_subexp): Use language-specific print_subexp.
Make global; remove static declaration.
Move most code to print_subexp_standard.
(print_subexp_standard): New function, containing code formerly in
print_subexp.
(op_name): Add expression to argument signature.
Use langauge-specific op_name.
Move most code to op_name_standard.
(op_name_standard): New function, containing code formerly in op_name.
(dump_subexp):  Use new version of op_name function.
Use language-specific dump_subexp_body, and move most existing code to
dump_subexp_body_standard.
(dump_raw_expression): Use new op_name interface.
(dump_subexp_body): Move most code to dump_subexp_body_standard.
(dump_subexp_body_standard): New function, containing code formerly
in dump_subexp_body.

* language.c (unknown_language): Add default la_exp_desc field and
remove evaluate_exp field.
(auto_language): Ditto.
(local_language): Ditto.
* f-lang.c (f_language_defn): Ditto.
* c-lang.c (c_language_defn): Ditto.
(cplus_language_defn): Ditto.
(asm_language_defn): Ditto.
(minimal_language_defn): Ditto.
* p-lang.c (pascal_language_defn): Ditto.
* m2-lang.c (m2_language_defn): Ditto.
* objc-lang.c (objc_language_defn): Ditto.

* jv-lang.c (exp_descriptor_java): New variable, containing
Java-specific expression evaluator.
(java_language_defn): Add la_exp_desc field and remove evaluate_exp
field.
* scm-lang.c (exp_descriptor_scm): New variable, containing
Scheme-specific expression evaluator.
(scm_language_defn): Add la_exp_desc field and remove evaluate_exp
field.
* objc-lang.c (print_object_command): Take evaluate_exp from the
la_exp_desc field.

* Makefile.in (eval.o): Add dependency on parser-defs.h.

* eval.c: Include parser-defs.h for the full declaration of
la_exp_desc's type.
(evaluate_subexp): Get evaluate_exp out of la_exp_desc field.

2003-09-26  H.J. Lu  <hongjiu.lu@intel.com>

* elflink.c (elf_link_read_relocs_from_section): Add an argument
of a pointer to section. Check bad symbol index.
(_bfd_elf_link_read_relocs): Modify calls to
elf_link_read_relocs_from_section.

*** empty log message ***

2003-09-24  Dave Brolley  <brolley@redhat.com>

        * frv-desc.c, frv-opc.c, frv-opc.h: Regenerated.

2003-09-24  Dave Brolley  <brolley@redhat.com>

        * profile.h (update_FR_ptime): New prototype.
        (update_FRdouble_ptime): Ditto.
        (update_SPR_ptime): Ditto.
        (increase_ACC_busy): Ditto.
        (enforce_full_acc_latency): Ditto.
        (post_wait_for_SPR): Ditto.
        * profile.c (update_FR_ptime): Moved here from profile-fr500.c.
        (update_FRdouble_ptime): Ditto.
        (update_SPR_ptime): New function.
        (increase_ACC_busy): Ditto.
        (enforce_full_acc_latency): Ditto.
        (vliw_wait_for_fdiv_resource): Correct resource name.
        (vliw_wait_for_fsqrt_resource): Ditto.
        (post_wait_for_SPR): New function.
        * profile-fr500.c (frvbf_model_fr500_u_commit): New function.
        (frvbf_model_fr500_u_gr2fr): Pass out_FRk as output register to
        adjust_float_register_busy.
        (frvbf_model_fr500_u_gr_load): Record latency of SPR registers.
        (frvbf_model_fr500_u_fr_load): Wait for and record latency of SPR
        registers.
        (frvbf_model_fr500_u_float_arith): Ditto.
        (frvbf_model_fr500_u_float_dual_arith): Ditto.
        (frvbf_model_fr500_u_float_div): Ditto.
        (frvbf_model_fr500_u_float_sqrt): Ditto.
        (frvbf_model_fr500_u_float_convert): Ditto.
        (update_FR_ptime): Moved to profile.c
        (update_FRdouble_ptime): Moved to profile.c
        * profile-fr400.c (update_FR_ptime): Removed. Identical to functions
        for other machines.
        (update_FRdouble_ptime): Ditto.
        * arch.h,cpu.h,sem.c,decode.[ch],model.c,sem.c: Regenerated.

2003-09-24  Dave Brolley  <brolley@redhat.com>

        * frv.cpu (u-commit): New modelling unit for fr500.
        (mwtaccg): Use frv_ref_SI to reference ACC40Sk as an input operand.
        (commit-r): Use u-commit model for fr500.
        (commit): Ditto.
        (conditional-float-binary-op): Take profiling data as an argument.
        Update callers.
        (ne-float-binary-op): Ditto.

* parser-defs.h (operator_length): Declare.

* parse.c (length_of_subexp): Use operator_length to get operator
lengths and arities for operators.
Move most code to new operator_length function.
(operator_length): New function absorbing most code from
length_of_subexp.
(prefixify_subexp): Remove large case and use operator_length instead.
(parse_exp_1): Use renamings:
dump_prefix_expression => dump_raw_expression and
dump_postfix_expression => dump_prefix_expression.

* expression.h (dump_prefix_expression): Rename to ...
(dump_raw_expression): New name.
(dump_postfix_expression): Rename to ...
(dump_prefix_expression): New name.

* expprint.c (dump_subexp): Make global.  Add comment.
Move most existing code to dump_subexp_body.
(dump_subexp_body): New function.
(dump_prefix_expression): Rename to dump_raw_expression.
Remove attempt to print the expression via print_expression: it can't
work before the expression is prefixified.
(dump_raw_expression): Renamed from dump_prefix_expression.
(dump_postfix_expression): Rename to dump_prefix_expression, since
that's what it does.
Remove 'note' parameter, since this routine must be used on
prefixified expression.
(dump_prefix_expression): Renamed from dump_postfix_expression.

* sh.h (R_SH_SWITCH8, R_SH_GNU_VTINHERIT, R_SH_GNU_VTENTRY,
R_SH_LOOP_START,R_SH_LOOP_END): Move to "reserved" spaces.
(R_SH_DIR16, R_SH_DIR8, R_SH_DIR8UL, R_SH_DIR8UW, R_SH_DIR8U,
R_SH_DIR8SW, R_SH_DIR8S, R_SH_DIR4UL, R_SH_DIR4UW, R_SH_DIR4U,
R_SH_PSHA, R_SH_PSHL): New.

* elf32-sh.c (sh_elf_howto_table): R_SH_SWITCH8,
R_SH_GNU_VTINHERIT, R_SH_GNU_VTENTRY,
R_SH_LOOP_START,R_SH_LOOP_END moved to "reserved" spaces,
R_SH_DIR16, R_SH_DIR8, R_SH_DIR8UL, R_SH_DIR8UW, R_SH_DIR8U,
R_SH_DIR8SW, R_SH_DIR8S, R_SH_DIR4UL, R_SH_DIR4UW, R_SH_DIR4U,
R_SH_PSHA, R_SH_PSHL added.
(sh_reloc_map): Add R_SH_DIR16 and R_SH_DIR8.
(sh_elf_relocate_section): Support new relocs.

*** empty log message ***

* elf32-xtensa.c (elf_xtensa_relocate_section): Fix typo that clobbered
dynamic relocation offsets.

2003-09-23  Elena Zannoni  <ezannoni@redhat.com>

* gdb.base/selftest.exp: Accomodate more instruction reordering
        weirdness.

* dwarf2read.c (read_array_type): When building the type for an
array of unspecified length, make sure to choose the upper bound
so that the array's total length comes out to be zero --- that's
how we represent such arrays.

* simple.c (bfd_simple_get_relocated_section_contents): Free the
hash table using _bfd_generic_link_hash_table_free.

o * ld-discard/exit.s: Correct .text.exit attributes.
* ld-discard/extern.s: Likewise.
* ld-discard/static.s: Likewise.

* config/obj-elf.c (obj_elf_change_section): Adjust for
_bfd_elf_get_sec_type_attr changes.  Allow SHF_MERGE and SHF_STRINGS
to be set when defaults are not.  Don't set attr from defaults if
old_sec.

* elf-bfd.h (struct bfd_elf_special_section): Remove "suffix".  Change
type of prefix_length and suffix_length to int.  Rename "attributes"
to "attr".  Comment.
(_bfd_elf_get_sec_type_attr): Update prototype.
* elf.c (get_special_section): Rewrite.
(_bfd_elf_get_sec_type_attr): Return struct rather than passing in
attr and type pointers.
(_bfd_elf_new_section_hook): Adjust for above.
(special_sections): Merge suffix with prefix.  Set
prefix_length for all entries.  Set suffix_length appropriately.
* elf32-m32r.c (m32r_elf_special_sections): Likewise.
* elf32-m68hc11.c (elf32_m68hc11_special_sections): Likewise.
* elf32-m68hc12.c (elf32_m68hc12_special_sections): Likewise.
* elf32-mcore.c (mcore_elf_special_sections): Likewise.
* elf32-sh64.c (sh64_elf_special_sections): Likewise.
* elf32-v850.c (v850_elf_special_sections): Likewise.
* elf32-xtensa.c (elf_xtensa_special_sections): Likewise.
* elf64-alpha.c (elf64_alpha_special_sections): Likewise.
* elf64-hppa.c (elf64_hppa_special_sections): Likewise.
* elf64-ppc.c (ppc64_elf_special_sections): Likewise.
* elf64-sh64.c (sh64_elf64_special_sections): Likewise.
* elfxx-ia64.c (elfNN_ia64_special_sections): Likewise.
* elfxx-mips.c (_bfd_mips_elf_special_sections): Likewise.
* elf32-ppc.c (ppc_elf_special_sections): Likewise.  Fix .plt flags.

*** empty log message ***

2003-09-22  Michael Chastain  <mec@shout.net>

* gdbint.texinfo (Testsuite Organization): Change gdb.c++ to gdb.cp.

2003-09-22  Michael Chastain  <mec@shout.net>

* MAINTAINERS: Rename gdb.c++ to gdb.cp.

2003-09-22  Jeff Johnston  <jjohnstn@redhat.com>

        * top.c (quit_force): Fix indirect call to quit_target so
        a struct qt_args pointer is passed.

2003-09-22  Andrew Cagney  <cagney@redhat.com>

* arch-utils.h (init_frame_pc_noop): Delete declaration.
* arch-utils.c (init_frame_pc_noop): Delete function.
* mn10300-tdep.c (mn10300_gdbarch_init): Do not set
"init_frame_pc".
* mips-tdep.c (mips_gdbarch_init): Ditto.
* i386-interix-tdep.c (i386_interix_init_abi): Ditto.
* config/sparc/tm-sparc.h (init_frame_pc_noop): Delete
declaration.
(DEPRECATED_INIT_FRAME_PC): Delete macro.
* config/rs6000/tm-rs6000.h (init_frame_pc_noop): Delete
declaration.
(DEPRECATED_INIT_FRAME_PC): Delete macro.

2003-09-22  Andrew Cagney  <cagney@redhat.com>

* floatformat.h (struct floatformat): Add field "is_valid".

2003-09-22  Andrew Cagney  <cagney@redhat.com>

* floatformat.c (floatformat_i387_ext_is_valid): New function.
(floatformat_always_valid): New function.
(floatformat_i387_ext): Initialize new "is_valid" field to
"floatformat_i387_ext_is_valid".
(floatformat_ieee_single_little): Initialize "is_valid" field to
floatformat_always_valid.
(floatformat_ieee_double_big): Ditto.
(floatformat_ieee_double_little): Ditto.
(floatformat_ieee_double_little): Ditto.
(floatformat_ieee_double_littlebyte_bigword): Ditto.
(floatformat_i960_ext): Ditto.
(floatformat_m88110_ext): Ditto.
(floatformat_m88110_harris_ext): Ditto.
(floatformat_arm_ext_big): Ditto.
(floatformat_arm_ext_littlebyte_bigword): Ditto.
(floatformat_ia64_spill_big): Ditto.
(floatformat_ia64_spill_little): Ditto.
(floatformat_ia64_quad_big): Ditto.
(floatformat_ia64_quad_little): Ditto.
(floatformat_ia64_quad_little): Ditto.
(floatformat_is_valid): Call "is_valid".