My name is in this file 

 In time decreasing order:

1. Decimal Floating Point Support in GDB

Finally, this huge patch got accepted into mainstream gdb. A big thanks to Thiago Jung Bauermann for achieving this. Here is changelog.

link


2007-10-25  Wu Zhou  <woodzltc@cn.ibm.com
            Thiago Jung Bauermann  <bauerman@br.ibm.com>

        * c-exp.y (YYSTYPE): Add typed_val_decfloat for decimal
        floating point in YYSTYPE union.
        (DECFLOAT) Add token and expression element handling code.
        (parse_number): Parse DFP constants, which end with suffix 'df',
        'dd' or 'dl'.  Return DECFLOAT.
        * eval.c (evaluate_subexp_standard): Call value_from_decfloat to
        handle OP_DECFLOAT.
        * expression.h (enum exp_opcode): Add an opcode (OP_DECFLOAT)
        for DFP constants.
        (union exp_element): Add decfloatconst to represent DFP
        elements, which is 16 bytes by default.
        * parse.c (write_exp_elt_decfloatcst): New function to write a
        decimal float const into the expression.
        (operator_length_standard): Set operator length for OP_DECFLOAT
        to 4.
        * parser-defs.h (write_exp_elt_decfloatcst): Prototype.
        * valarith.c (value_neg): Add code to handle the negation
        operation of DFP values.
        * value.c (value_from_decfloat): New function to get the value
        from a decimal floating point.
        * value.h (value_from_decfloat): Prototype.

2007-10-25  Wu Zhou  <woodzltc@cn.ibm.com
            Thiago Jung Bauermann  <bauerman@br.ibm.com>

        * c-lang.c (c_create_fundamental_type): Create fundamental
        types for DFP.
        * c-typeprint.c (c_type_print_varspec_prefix): Add
        TYPE_CODE_DECFLOAT to no prefix needed case.
        (c_type_print_varspec_suffix): Add TYPE_CODE_DECFLOAT to no
        suffix needed case.
        * c-valprint.c (c_val_print): Call print_decimal_floating to
        print DFP values.
        * dwarf2read.c (read_base_type): Read DW_ATE_decimal_float
        attribute code and return TYPE_CODE_DECFLOAT.
        (dwarf_base_type): Set dwarf2_fundamental_type for DFP values.
        * gdbtypes.c (gdbtypes_post_init): Initialize builtin_decfloat,
        builtin_decdouble and builtin_declong. 
        * gdbtypes.h (enum type_code): Add TYPE_CODE_DECFLOAT as a
        type code for DFP.
        (FT_DECFLOAT, FT_DBL_PREC_DECFLOAT, FT_EXT_PREC_DECFLOAT): New
        types, for decimal floating point.
        (FT_NUM_MEMBERS):  Increment, new types added.
        (struct builtin_type): Add builtin_decfloat, builtin_decdouble
        and builtin_declong.
        * valprint.c (print_decimal_floating): New function to print DFP
        values.
        * value.h (print_decimal_floating): Prototype.

2007-10-25  Wu Zhou  <woodzltc@cn.ibm.com
            Thiago Jung Bauermann  <bauerman@br.ibm.com>

        * Makefile.in (LIBDECNUMBER_DIR, LIBDECNUMBER, LIBDECNUMBER_SRC
        LIBDECNUMBER_CFLAGS): New macros for libdecnumber.
        (INTERNAL_CFLAGS_BASE): Add LIBDECNUMBER_CFLAGS in.
        (INSTALLED_LIBS): Add -ldecnumber in.
        (CLIBS): Add LIBDECNUMBER in.
        (decimal128_h, decimal64_h, decimal32_h): New macros for decimal
        headers.
        (dfp_h): New macros for decimal floating point.
        (dfp.o): New target.
        (COMMON_OBS): Add dfp.o in.
        (c-exp.o): Add dfp_h as dependency.
        (valprint.o): Add dfp_h as dependency.
        (value.o): Add dfp_h as dependency.
        * dfp.h: New header file for decimal floating point support in
        GDB.
        * dfp.c: New source file for decimal floating point support in
        GDB.  Implement decimal_from_string and decimal_to_string based
        on libdecnumber API.
        * configure.ac: Add AC_C_BIGENDIAN test.
        * config.in, configure: Regenerate.
 

 2. check the value of DW_AT_calling_convention, to find the main entry of Fortran program

The changelog 
2007-10-25  Wu Zhou  <woodzltc@cn.ibm.com>
            Carlos Eduardo Seo  <cseo@linux.vnet.ibm.com>
            Jim Blandy  <jimb@codesourcery.com>

        * dwarf2read.c (read_partial_die): check the value
        of DW_AT_calling_convention in Fortran programs.
        case DW_AT_calling_convention:
          /* DWARF doesn't provide a way to identify a program's source-level
             entry point.  DW_AT_calling_convention attributes are only meant
             to describe functions' calling conventions.

             However, because it's a necessary piece of information in
             Fortran, and because DW_CC_program is the only piece of debugging
             information whose definition refers to a 'main program' at all,
             several compilers have begun marking Fortran main programs with
             DW_CC_program --- even when those functions use the standard
             calling conventions.

             So until DWARF specifies a way to provide this information and
             compilers pick up the new representation, we'll support this
             practice.  */
          if (DW_UNSND (&attr) == DW_CC_program
              && cu->language == language_fortran)
            set_main_name (part_die->name);
          break;
 
3. Old code contributions in 2006

2006-03-01  Wu Zhou  <woodzltc@cn.ibm.com>

         * f-typeprint.c (f_type_print_base): Delete the redundant space.
 
2006-02-24  Wu Zhou  <woodzltc@cn.ibm.com>

        * f-exp.y: Symbol '%' is not used as the modulus operator in
        Fortran.  Delete this from Fortran expression.
        It is now used by Fortran 90 and later to access the member
        of derived type.  Add this into Fortran expression.
        * f-valprint.c (f_val_print): Add code to handle TYPE_CODE_STRUCT.
        Print each elements in the derived type.
        * f-typeprint.c (print_equivalent_f77_float_type): Add a parameter
        level into the function definition to do indented printing.  And
        call fprintfi_filtered instead to do indented printing.
        (f_type_print_base): Replace fprintf_filtered with the indented
        version (fprintfi_filtered).
        (f_type_print_base): Call indented print_equivalent_f77_float_type.
        (f_type_print_base): Add code to handle TYPE_CODE_STRUCT.  Print
        the definition of the derived type. 
 
2006-02-08  Wu Zhou  <woodzltc@cn.ibm.com>

        * config/i386/nm-i386sol2.h: Update copyright year.
        * config/mips/nm-irix5.h: Ditto.
        * config/sparc/nm-sol2.h: Ditto.
        * s390-nat.c: Ditto.
        * ppc-linux-nat.c: Ditto.

2006-02-08  Wu Zhou  <woodzltc@cn.ibm.com>

        * breakpoint.c (TARGET_REGION_OK_FOR_HW_WATCHPOINT): Delete.
        * config/i386/nm-i386sol2.h (TARGET_REGION_OK_FOR_HW_WATCHPOINT): New.
        (TARGET_REGION_SIZE_OK_FOR_HW_WATCHPOINT): Delete.
        * config/mips/nm-irix5.h (TARGET_REGION_OK_FOR_HW_WATCHPOINT): New.
        (TARGET_REGION_SIZE_OK_FOR_HW_WATCHPOINT): Delete.
        * config/sparc/nm-sol2.h (TARGET_REGION_OK_FOR_HW_WATCHPOINT): New.
        (TARGET_REGION_SIZE_OK_FOR_HW_WATCHPOINT): Delete.
        * inf-ttrace.c (inf_ttrace_region_ok_for_hw_watchpoint): New.
        (inf_ttrace_region_size_ok_for_hw_watchpoint): Delete.
        (inf_ttrace_target): Delete to_region_size_ok_for_hw_watchpoint and
        add to_region_ok_for_hw_watchpoint.
        * s390-nat.c (s390_region_size_ok_for_hw_watchpoint): Delete.
        (s390_region_ok_for_hw_watchpoint): New.
        (_initialize_s390_nat): Delete to_region_size_ok_for_hw_watchpoint
        and add to_region_ok_for_hw_watchpoint.
        * target.c (default_region_size_ok_for_hw_watchpoint, 
        debug_to_region_size_ok_for_hw_watchpoint): Delete prototype.
        (update_current_target): Delete to_region_size_ok_for_hw_watchpoint
        inheritance and default_region_size_ok_for_hw_watchpoint.
        (default_region_ok_for_hw_watchpoint): If len is less than or equal
        the length of void pointer, return ok.
        (default_region_size_ok_for_hw_watchpoint): Delete.
        (debug_to_region_size_ok_for_hw_watchpoint): Delete.
        (setup_target_debug): Delete to_region_size_ok_for_hw_watchpoint.
        * target.h (struct target_ops): Delete 
        to_region_size_ok_for_hw_watchpoint.
        (TARGET_REGION_SIZE_OK_FOR_HW_WATCHPOINT): Delete.
        
2006-02-08  Ben Elliston  <bje@au1.ibm.com>
            Wu Zhou  <woodzltc@cn.ibm.com>

        * ppc-linux-nat.c (PTRACE_GET_DEBUGREG, PTRACE_SET_DEBUGREG,
        PTRACE_GETSIGINFO): Define.
        (last_stopped_data_address): New.
        (ppc_linux_check_watch_resources): New function.
        (ppc_linux_region_ok_for_hw_watchpoint): New function.
        (ppc_linux_insert_watchpoint): New function.
        (ppc_linux_remove_watchpoint): New function.
        (ppc_linux_stopped_data_address): New function.
        (ppc_linux_stopped_by_watchpoint): New function.
        (_initialize_ppc_linux_nat): Set the above hardware watchpoint
        related target vectors.
        * rs6000-tdep.c (rs6000_gdbarch_init): Set PPC architectures
        to have nonsteppable watchpoint.
        * target.c (default_region_ok_for_hw_watchpoint,
        debug_to_region_ok_for_hw_watchpoint): New prototypes.
        (update_current_target): Inherit to_region_ok_for_hw_watchpoint
        and set default to_region_ok_for_hw_watchpoint.
        (default_region_ok_for_hw_watchpoint): New function.
        (debug_to_region_ok_for_hw_watchpoint): New function.
        (setup_target_debug): Set to_region_ok_for_hw_watchpoint of 
        debug_target.
        * target.h (struct target_ops): Add a new target vector 
        to_region_ok_for_hw_watchpoint.
        (TARGET_REGION_OK_FOR_HW_WATCHPOINT): Define this if it is not
        defined anyplace else. 
 
4. Old code contributions in 2005
 
The changelog 
 
2005-11-03  Wu Zhou  <woodzltc@cn.ibm.com>

        Checked in by Elena Zannoni  <ezannoni@redhat.com>

        * dwarf2read.c (read_structure_type): Add IBM XL C++
        specific code to set TYPE_VPTR_FIELDNO and TYPE_VPTR_BASETYPE
        of a virtual class if a field named "__vfp" is found.  

2005-09-20  Wu Zhou  <woodzltc@cn.ibm.com>

        * c-exp.y (parse-number): Modify the float parsing logic to let it 
        recognize a suffix.

2005-09-20  Wu Zhou  <woodzltc@cn.ibm.com>

        * expression.h (enum exp_opcode): Fix a format error of a comment.
 
2005-09-20  Wu Zhou  <woodzltc@cn.ibm.com>

        * expression.h (enum exp_opcode): Add a new operator for F90
        subrange.
        * f-lang.h (enum f90_range_type): New enumeration type to identify
        F90 subrange type.
        * f-exp.y (yyparse): Add support for parsing F90 subrange and
        change substring parsing to subrange parsing.
        * parse.c (operator_length_standard): Set the operator length
        and args number for OP_F90_RANGE.
        * eval.c (evaluate_subexp_standard): Add code to evaluate F90
        array section and substring.
        (value_f90_subarray): New function to evaluate F90 array section.
        (evaluate_subexp_standard): Delete label op_f77_substr and its code
        because the logic is implemented by function value_f90_subarray now.
 

2005-07-20  Wu Zhou  <woodzltc@cn.ibm.com>

        * remote.c (remote_pid_to_str): Use xsnprintf instead of snprintf.
 
2005-07-06  Wu Zhou  <woodzltc@cn.ibm.com>

        * tui/tui-win.c (parse_scrolling_args): Fix SEGV error with invalid
        scroll command.

2005-07-06  Wu Zhou  <woodzltc@cn.ibm.com>

        * f-exp.y (yyparse): Add code to support exponentiation expression.
        (yylex): Add code to scan exponentiation operator.
        * eval.c (evaluate_subexp_standard): Add support for BINOP_EXP.
        * valarith.c (value_binop): Reset errno to 0 before calling pow
        to do exponentiation operation. 

2005-06-27  Wu Zhou  <woodzltc@cn.ibm.com>

        * gdbtypes.c (get_discrete_bounds): Fix a comment typo.

2005-06-21  Wu Zhou  <woodzltc@cn.ibm.com>

        * MAINTAINERS (Write After Approval): Add myself.
 
Many thanks to my manager Geoff Smith, who helped me handle the paperwork 
and brought me into this wonderful world of GDB hacking. :-)
 
2005-06-17  Wu Zhou  <woodzltc@cn.ibm.com>

        * remote.c (set_thread, remote_thread_alive): Replace sprintf
        call with xsnprintf.
        (remote_threads_extra_info, extended_remote_restart)
        (remote_check_symbols, store_register_using_P)
        (compare_sections_command): Ditto. 
 
 2005-06-13  Wu Zhou  <woodzltc@cn.ibm.com>

        * eval.c (evaluate_subexp_standard): Add code to check the target
        type of a TYPE_CODE_PTR value when we encounter a f77 undetermined
        arglist.  If it is array, string or function, work on the target
        value instead.
 
 2005-06-08  Wu Zhou  <woodzltc@cn.ibm.com>

        * expression.h (enum exp_opcode): Fix a comment typo.

2005-05-28  Daniel Jacobowitz  <dan@codesourcery.com>

        From Wu Zhou <woodzltc@cn.ibm.com>:
        * remote.c (struct gdb_ext_thread_info): Fix a comment typo.
        (remote_unpack_thread_info_response): Likewise.
 
... Started with comment typo... hehe...