/* Target-dependent code for the MIPS architecture, for GDB, the GNU Debugger. Copyright (C) 1988, 1989, 1990, 1991, 1992, 1993, 1994, 1995, 1996, 1997, 1998, 1999, 2000, 2001, 2002, 2003, 2004, 2005, 2006, 2007 Free Software Foundation, Inc. Contributed by Alessandro Forin(af@cs.cmu.edu) at CMU and by Per Bothner(bothner@cs.wisc.edu) at U.Wisconsin. This file is part of GDB. This program is free software; you can redistribute it and/or modify it under the terms of the GNU General Public License as published by the Free Software Foundation; either version 2 of the License, or (at your option) any later version. This program is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for more details. You should have received a copy of the GNU General Public License along with this program; if not, write to the Free Software Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. */ #include "defs.h" #include "gdb_string.h" #include "gdb_assert.h" #include "frame.h" #include "inferior.h" #include "symtab.h" #include "value.h" #include "gdbcmd.h" #include "language.h" #include "gdbcore.h" #include "symfile.h" #include "objfiles.h" #include "gdbtypes.h" #include "target.h" #include "arch-utils.h" #include "regcache.h" #include "osabi.h" #include "mips-tdep.h" #include "block.h" #include "reggroups.h" #include "opcode/mips.h" #include "elf/mips.h" #include "elf-bfd.h" #include "symcat.h" #include "sim-regno.h" #include "dis-asm.h" #include "frame-unwind.h" #include "frame-base.h" #include "trad-frame.h" #include "infcall.h" #include "floatformat.h" #include "remote.h" #include "target-descriptions.h" #include "dwarf2-frame.h" #include "user-regs.h" static const struct objfile_data *mips_pdr_data; static struct type *mips_register_type (struct gdbarch *gdbarch, int regnum); /* A useful bit in the CP0 status register (MIPS_PS_REGNUM). */ /* This bit is set if we are emulating 32-bit FPRs on a 64-bit chip. */ #define ST0_FR (1 << 26) /* The sizes of floating point registers. */ enum { MIPS_FPU_SINGLE_REGSIZE = 4, MIPS_FPU_DOUBLE_REGSIZE = 8 }; enum { MIPS32_REGSIZE = 4, MIPS64_REGSIZE = 8 }; static const char *mips_abi_string; static const char *mips_abi_strings[] = { "auto", "n32", "o32", "n64", "o64", "eabi32", "eabi64", NULL }; /* The standard register names, and all the valid aliases for them. */ struct register_alias { const char *name; int regnum; }; /* Aliases for o32 and most other ABIs. */ const struct register_alias mips_o32_aliases[] = { { "ta0", 12 }, { "ta1", 13 }, { "ta2", 14 }, { "ta3", 15 } }; /* Aliases for n32 and n64. */ const struct register_alias mips_n32_n64_aliases[] = { { "ta0", 8 }, { "ta1", 9 }, { "ta2", 10 }, { "ta3", 11 } }; /* Aliases for ABI-independent registers. */ const struct register_alias mips_register_aliases[] = { /* The architecture manuals specify these ABI-independent names for the GPRs. */ #define R(n) { "r" #n, n } R(0), R(1), R(2), R(3), R(4), R(5), R(6), R(7), R(8), R(9), R(10), R(11), R(12), R(13), R(14), R(15), R(16), R(17), R(18), R(19), R(20), R(21), R(22), R(23), R(24), R(25), R(26), R(27), R(28), R(29), R(30), R(31), #undef R /* k0 and k1 are sometimes called these instead (for "kernel temp"). */ { "kt0", 26 }, { "kt1", 27 }, /* This is the traditional GDB name for the CP0 status register. */ { "sr", MIPS_PS_REGNUM }, /* This is the traditional GDB name for the CP0 BadVAddr register. */ { "bad", MIPS_EMBED_BADVADDR_REGNUM }, /* This is the traditional GDB name for the FCSR. */ { "fsr", MIPS_EMBED_FP0_REGNUM + 32 } }; /* Some MIPS boards don't support floating point while others only support single-precision floating-point operations. */ enum mips_fpu_type { MIPS_FPU_DOUBLE, /* Full double precision floating point. */ MIPS_FPU_SINGLE, /* Single precision floating point (R4650). */ MIPS_FPU_NONE /* No floating point. */ }; #ifndef MIPS_DEFAULT_FPU_TYPE #define MIPS_DEFAULT_FPU_TYPE MIPS_FPU_DOUBLE #endif static int mips_fpu_type_auto = 1; static enum mips_fpu_type mips_fpu_type = MIPS_DEFAULT_FPU_TYPE; static int mips_debug = 0; /* Properties (for struct target_desc) describing the g/G packet layout. */ #define PROPERTY_GP32 "internal: transfers-32bit-registers" #define PROPERTY_GP64 "internal: transfers-64bit-registers" /* MIPS specific per-architecture information */ struct gdbarch_tdep { /* from the elf header */ int elf_flags; /* mips options */ enum mips_abi mips_abi; enum mips_abi found_abi; enum mips_fpu_type mips_fpu_type; int mips_last_arg_regnum; int mips_last_fp_arg_regnum; int default_mask_address_p; /* Is the target using 64-bit raw integer registers but only storing a left-aligned 32-bit value in each? */ int mips64_transfers_32bit_regs_p; /* Indexes for various registers. IRIX and embedded have different values. This contains the "public" fields. Don't add any that do not need to be public. */ const struct mips_regnum *regnum; /* Register names table for the current register set. */ const char **mips_processor_reg_names; /* The size of register data available from the target, if known. This doesn't quite obsolete the manual mips64_transfers_32bit_regs_p, since that is documented to force left alignment even for big endian (very strange). */ int register_size_valid_p; int register_size; }; static int n32n64_floatformat_always_valid (const struct floatformat *fmt, const void *from) { return 1; } /* FIXME: brobecker/2004-08-08: Long Double values are 128 bit long. They are implemented as a pair of 64bit doubles where the high part holds the result of the operation rounded to double, and the low double holds the difference between the exact result and the rounded result. So "high" + "low" contains the result with added precision. Unfortunately, the floatformat structure used by GDB is not powerful enough to describe this format. As a temporary measure, we define a 128bit floatformat that only uses the high part. We lose a bit of precision but that's probably the best we can do for now with the current infrastructure. */ static const struct floatformat floatformat_n32n64_long_double_big = { floatformat_big, 128, 0, 1, 11, 1023, 2047, 12, 52, floatformat_intbit_no, "floatformat_n32n64_long_double_big", n32n64_floatformat_always_valid }; static const struct floatformat *floatformats_n32n64_long[BFD_ENDIAN_UNKNOWN] = { &floatformat_n32n64_long_double_big, &floatformat_n32n64_long_double_big }; const struct mips_regnum * mips_regnum (struct gdbarch *gdbarch) { return gdbarch_tdep (gdbarch)->regnum; } static int mips_fpa0_regnum (struct gdbarch *gdbarch) { return mips_regnum (gdbarch)->fp0 + 12; } #define MIPS_EABI (gdbarch_tdep (current_gdbarch)->mips_abi == MIPS_ABI_EABI32 \ || gdbarch_tdep (current_gdbarch)->mips_abi == MIPS_ABI_EABI64) #define MIPS_LAST_FP_ARG_REGNUM (gdbarch_tdep (current_gdbarch)->mips_last_fp_arg_regnum) #define MIPS_LAST_ARG_REGNUM (gdbarch_tdep (current_gdbarch)->mips_last_arg_regnum) #define MIPS_FPU_TYPE (gdbarch_tdep (current_gdbarch)->mips_fpu_type) /* MIPS16 function addresses are odd (bit 0 is set). Here are some functions to test, set, or clear bit 0 of addresses. */ static CORE_ADDR is_mips16_addr (CORE_ADDR addr) { return ((addr) & 1); } static CORE_ADDR unmake_mips16_addr (CORE_ADDR addr) { return ((addr) & ~(CORE_ADDR) 1); } /* Return the contents of register REGNUM as a signed integer. */ static LONGEST read_signed_register (int regnum) { LONGEST val; regcache_cooked_read_signed (current_regcache, regnum, &val); return val; } static LONGEST read_signed_register_pid (int regnum, ptid_t ptid) { ptid_t save_ptid; LONGEST retval; if (ptid_equal (ptid, inferior_ptid)) return read_signed_register (regnum); save_ptid = inferior_ptid; inferior_ptid = ptid; retval = read_signed_register (regnum); inferior_ptid = save_ptid; return retval; } /* Return the MIPS ABI associated with GDBARCH. */ enum mips_abi mips_abi (struct gdbarch *gdbarch) { return gdbarch_tdep (gdbarch)->mips_abi; } int mips_isa_regsize (struct gdbarch *gdbarch) { struct gdbarch_tdep *tdep = gdbarch_tdep (gdbarch); /* If we know how big the registers are, use that size. */ if (tdep->register_size_valid_p) return tdep->register_size; /* Fall back to the previous behavior. */ return (gdbarch_bfd_arch_info (gdbarch)->bits_per_word / gdbarch_bfd_arch_info (gdbarch)->bits_per_byte); } /* Return the currently configured (or set) saved register size. */ unsigned int mips_abi_regsize (struct gdbarch *gdbarch) { switch (mips_abi (gdbarch)) { case MIPS_ABI_EABI32: case MIPS_ABI_O32: return 4; case MIPS_ABI_N32: case MIPS_ABI_N64: case MIPS_ABI_O64: case MIPS_ABI_EABI64: return 8; case MIPS_ABI_UNKNOWN: case MIPS_ABI_LAST: default: internal_error (__FILE__, __LINE__, _("bad switch")); } } /* Functions for setting and testing a bit in a minimal symbol that marks it as 16-bit function. The MSB of the minimal symbol's "info" field is used for this purpose. gdbarch_elf_make_msymbol_special tests whether an ELF symbol is "special", i.e. refers to a 16-bit function, and sets a "special" bit in a minimal symbol to mark it as a 16-bit function MSYMBOL_IS_SPECIAL tests the "special" bit in a minimal symbol */ static void mips_elf_make_msymbol_special (asymbol * sym, struct minimal_symbol *msym) { if (((elf_symbol_type *) (sym))->internal_elf_sym.st_other == STO_MIPS16) { MSYMBOL_INFO (msym) = (char *) (((long) MSYMBOL_INFO (msym)) | 0x80000000); SYMBOL_VALUE_ADDRESS (msym) |= 1; } } static int msymbol_is_special (struct minimal_symbol *msym) { return (((long) MSYMBOL_INFO (msym) & 0x80000000) != 0); } /* XFER a value from the big/little/left end of the register. Depending on the size of the value it might occupy the entire register or just part of it. Make an allowance for this, aligning things accordingly. */ static void mips_xfer_register (struct regcache *regcache, int reg_num, int length, enum bfd_endian endian, gdb_byte *in, const gdb_byte *out, int buf_offset) { int reg_offset = 0; gdb_assert (reg_num >= gdbarch_num_regs (current_gdbarch)); /* Need to transfer the left or right part of the register, based on the targets byte order. */ switch (endian) { case BFD_ENDIAN_BIG: reg_offset = register_size (current_gdbarch, reg_num) - length; break; case BFD_ENDIAN_LITTLE: reg_offset = 0; break; case BFD_ENDIAN_UNKNOWN: /* Indicates no alignment. */ reg_offset = 0; break; default: internal_error (__FILE__, __LINE__, _("bad switch")); } if (mips_debug) fprintf_unfiltered (gdb_stderr, "xfer $%d, reg offset %d, buf offset %d, length %d, ", reg_num, reg_offset, buf_offset, length); if (mips_debug && out != NULL) { int i; fprintf_unfiltered (gdb_stdlog, "out "); for (i = 0; i < length; i++) fprintf_unfiltered (gdb_stdlog, "%02x", out[buf_offset + i]); } if (in != NULL) regcache_cooked_read_part (regcache, reg_num, reg_offset, length, in + buf_offset); if (out != NULL) regcache_cooked_write_part (regcache, reg_num, reg_offset, length, out + buf_offset); if (mips_debug && in != NULL) { int i; fprintf_unfiltered (gdb_stdlog, "in "); for (i = 0; i < length; i++) fprintf_unfiltered (gdb_stdlog, "%02x", in[buf_offset + i]); } if (mips_debug) fprintf_unfiltered (gdb_stdlog, "\n"); } /* Determine if a MIPS3 or later cpu is operating in MIPS{1,2} FPU compatiblity mode. A return value of 1 means that we have physical 64-bit registers, but should treat them as 32-bit registers. */ static int mips2_fp_compat (struct frame_info *frame) { /* MIPS1 and MIPS2 have only 32 bit FPRs, and the FR bit is not meaningful. */ if (register_size (current_gdbarch, mips_regnum (current_gdbarch)->fp0) == 4) return 0; #if 0 /* FIXME drow 2002-03-10: This is disabled until we can do it consistently, in all the places we deal with FP registers. PR gdb/413. */ /* Otherwise check the FR bit in the status register - it controls the FP compatiblity mode. If it is clear we are in compatibility mode. */ if ((get_frame_register_unsigned (frame, MIPS_PS_REGNUM) & ST0_FR) == 0) return 1; #endif return 0; } #define VM_MIN_ADDRESS (CORE_ADDR)0x400000 static CORE_ADDR heuristic_proc_start (CORE_ADDR); static void reinit_frame_cache_sfunc (char *, int, struct cmd_list_element *); static struct type *mips_float_register_type (void); static struct type *mips_double_register_type (void); /* The list of available "set mips " and "show mips " commands */ static struct cmd_list_element *setmipscmdlist = NULL; static struct cmd_list_element *showmipscmdlist = NULL; /* Integer registers 0 thru 31 are handled explicitly by mips_register_name(). Processor specific registers 32 and above are listed in the following tables. */ enum { NUM_MIPS_PROCESSOR_REGS = (90 - 32) }; /* Generic MIPS. */ static const char *mips_generic_reg_names[NUM_MIPS_PROCESSOR_REGS] = { "sr", "lo", "hi", "bad", "cause", "pc", "f0", "f1", "f2", "f3", "f4", "f5", "f6", "f7", "f8", "f9", "f10", "f11", "f12", "f13", "f14", "f15", "f16", "f17", "f18", "f19", "f20", "f21", "f22", "f23", "f24", "f25", "f26", "f27", "f28", "f29", "f30", "f31", "fsr", "fir", "" /*"fp" */ , "", "", "", "", "", "", "", "", "", "", "", "", "", "", "", "", "", }; /* Names of IDT R3041 registers. */ static const char *mips_r3041_reg_names[] = { "sr", "lo", "hi", "bad", "cause", "pc", "f0", "f1", "f2", "f3", "f4", "f5", "f6", "f7", "f8", "f9", "f10", "f11", "f12", "f13", "f14", "f15", "f16", "f17", "f18", "f19", "f20", "f21", "f22", "f23", "f24", "f25", "f26", "f27", "f28", "f29", "f30", "f31", "fsr", "fir", "", /*"fp" */ "", "", "", "bus", "ccfg", "", "", "", "", "", "", "port", "cmp", "", "", "epc", "prid", }; /* Names of tx39 registers. */ static const char *mips_tx39_reg_names[NUM_MIPS_PROCESSOR_REGS] = { "sr", "lo", "hi", "bad", "cause", "pc", "", "", "", "", "", "", "", "", "", "", "", "", "", "", "", "", "", "", "", "", "", "", "", "", "", "", "", "", "", "", "", "", "", "", "", "", "", "", "", "", "", "", "", "", "", "", "config", "cache", "debug", "depc", "epc", "" }; /* Names of IRIX registers. */ static const char *mips_irix_reg_names[NUM_MIPS_PROCESSOR_REGS] = { "f0", "f1", "f2", "f3", "f4", "f5", "f6", "f7", "f8", "f9", "f10", "f11", "f12", "f13", "f14", "f15", "f16", "f17", "f18", "f19", "f20", "f21", "f22", "f23", "f24", "f25", "f26", "f27", "f28", "f29", "f30", "f31", "pc", "cause", "bad", "hi", "lo", "fsr", "fir" }; /* Return the name of the register corresponding to REGNO. */ static const char * mips_register_name (int regno) { struct gdbarch_tdep *tdep = gdbarch_tdep (current_gdbarch); /* GPR names for all ABIs other than n32/n64. */ static char *mips_gpr_names[] = { "zero", "at", "v0", "v1", "a0", "a1", "a2", "a3", "t0", "t1", "t2", "t3", "t4", "t5", "t6", "t7", "s0", "s1", "s2", "s3", "s4", "s5", "s6", "s7", "t8", "t9", "k0", "k1", "gp", "sp", "s8", "ra", }; /* GPR names for n32 and n64 ABIs. */ static char *mips_n32_n64_gpr_names[] = { "zero", "at", "v0", "v1", "a0", "a1", "a2", "a3", "a4", "a5", "a6", "a7", "t0", "t1", "t2", "t3", "s0", "s1", "s2", "s3", "s4", "s5", "s6", "s7", "t8", "t9", "k0", "k1", "gp", "sp", "s8", "ra" }; enum mips_abi abi = mips_abi (current_gdbarch); /* Map [gdbarch_num_regs .. 2*gdbarch_num_regs) onto the raw registers, but then don't make the raw register names visible. */ int rawnum = regno % gdbarch_num_regs (current_gdbarch); if (regno < gdbarch_num_regs (current_gdbarch)) return ""; /* The MIPS integer registers are always mapped from 0 to 31. The names of the registers (which reflects the conventions regarding register use) vary depending on the ABI. */ if (0 <= rawnum && rawnum < 32) { if (abi == MIPS_ABI_N32 || abi == MIPS_ABI_N64) return mips_n32_n64_gpr_names[rawnum]; else return mips_gpr_names[rawnum]; } else if (tdesc_has_registers (gdbarch_target_desc (current_gdbarch))) return tdesc_register_name (rawnum); else if (32 <= rawnum && rawnum < gdbarch_num_regs (current_gdbarch)) { gdb_assert (rawnum - 32 < NUM_MIPS_PROCESSOR_REGS); return tdep->mips_processor_reg_names[rawnum - 32]; } else internal_error (__FILE__, __LINE__, _("mips_register_name: bad register number %d"), rawnum); } /* Return the groups that a MIPS register can be categorised into. */ static int mips_register_reggroup_p (struct gdbarch *gdbarch, int regnum, struct reggroup *reggroup) { int vector_p; int float_p; int raw_p; int rawnum = regnum % gdbarch_num_regs (current_gdbarch); int pseudo = regnum / gdbarch_num_regs (current_gdbarch); if (reggroup == all_reggroup) return pseudo; vector_p = TYPE_VECTOR (register_type (gdbarch, regnum)); float_p = TYPE_CODE (register_type (gdbarch, regnum)) == TYPE_CODE_FLT; /* FIXME: cagney/2003-04-13: Can't yet use gdbarch_num_regs (gdbarch), as not all architectures are multi-arch. */ raw_p = rawnum < gdbarch_num_regs (current_gdbarch); if (gdbarch_register_name (current_gdbarch, regnum) == NULL || gdbarch_register_name (current_gdbarch, regnum)[0] == '\0') return 0; if (reggroup == float_reggroup) return float_p && pseudo; if (reggroup == vector_reggroup) return vector_p && pseudo; if (reggroup == general_reggroup) return (!vector_p && !float_p) && pseudo; /* Save the pseudo registers. Need to make certain that any code extracting register values from a saved register cache also uses pseudo registers. */ if (reggroup == save_reggroup) return raw_p && pseudo; /* Restore the same pseudo register. */ if (reggroup == restore_reggroup) return raw_p && pseudo; return 0; } /* Return the groups that a MIPS register can be categorised into. This version is only used if we have a target description which describes real registers (and their groups). */ static int mips_tdesc_register_reggroup_p (struct gdbarch *gdbarch, int regnum, struct reggroup *reggroup) { int rawnum = regnum % gdbarch_num_regs (gdbarch); int pseudo = regnum / gdbarch_num_regs (gdbarch); int ret; /* Only save, restore, and display the pseudo registers. Need to make certain that any code extracting register values from a saved register cache also uses pseudo registers. Note: saving and restoring the pseudo registers is slightly strange; if we have 64 bits, we should save and restore all 64 bits. But this is hard and has little benefit. */ if (!pseudo) return 0; ret = tdesc_register_in_reggroup_p (gdbarch, rawnum, reggroup); if (ret != -1) return ret; return mips_register_reggroup_p (gdbarch, regnum, reggroup); } /* Map the symbol table registers which live in the range [1 * gdbarch_num_regs .. 2 * gdbarch_num_regs) back onto the corresponding raw registers. Take care of alignment and size problems. */ static void mips_pseudo_register_read (struct gdbarch *gdbarch, struct regcache *regcache, int cookednum, gdb_byte *buf) { int rawnum = cookednum % gdbarch_num_regs (current_gdbarch); gdb_assert (cookednum >= gdbarch_num_regs (current_gdbarch) && cookednum < 2 * gdbarch_num_regs (current_gdbarch)); if (register_size (gdbarch, rawnum) == register_size (gdbarch, cookednum)) regcache_raw_read (regcache, rawnum, buf); else if (register_size (gdbarch, rawnum) > register_size (gdbarch, cookednum)) { if (gdbarch_tdep (gdbarch)->mips64_transfers_32bit_regs_p || gdbarch_byte_order (current_gdbarch) == BFD_ENDIAN_LITTLE) regcache_raw_read_part (regcache, rawnum, 0, 4, buf); else regcache_raw_read_part (regcache, rawnum, 4, 4, buf); } else internal_error (__FILE__, __LINE__, _("bad register size")); } static void mips_pseudo_register_write (struct gdbarch *gdbarch, struct regcache *regcache, int cookednum, const gdb_byte *buf) { int rawnum = cookednum % gdbarch_num_regs (current_gdbarch); gdb_assert (cookednum >= gdbarch_num_regs (current_gdbarch) && cookednum < 2 * gdbarch_num_regs (current_gdbarch)); if (register_size (gdbarch, rawnum) == register_size (gdbarch, cookednum)) regcache_raw_write (regcache, rawnum, buf); else if (register_size (gdbarch, rawnum) > register_size (gdbarch, cookednum)) { if (gdbarch_tdep (gdbarch)->mips64_transfers_32bit_regs_p || gdbarch_byte_order (current_gdbarch) == BFD_ENDIAN_LITTLE) regcache_raw_write_part (regcache, rawnum, 0, 4, buf); else regcache_raw_write_part (regcache, rawnum, 4, 4, buf); } else internal_error (__FILE__, __LINE__, _("bad register size")); } /* Table to translate MIPS16 register field to actual register number. */ static int mips16_to_32_reg[8] = { 16, 17, 2, 3, 4, 5, 6, 7 }; /* Heuristic_proc_start may hunt through the text section for a long time across a 2400 baud serial line. Allows the user to limit this search. */ static unsigned int heuristic_fence_post = 0; /* Number of bytes of storage in the actual machine representation for register N. NOTE: This defines the pseudo register type so need to rebuild the architecture vector. */ static int mips64_transfers_32bit_regs_p = 0; static void set_mips64_transfers_32bit_regs (char *args, int from_tty, struct cmd_list_element *c) { struct gdbarch_info info; gdbarch_info_init (&info); /* FIXME: cagney/2003-11-15: Should be setting a field in "info" instead of relying on globals. Doing that would let generic code handle the search for this specific architecture. */ if (!gdbarch_update_p (info)) { mips64_transfers_32bit_regs_p = 0; error (_("32-bit compatibility mode not supported")); } } /* Convert to/from a register and the corresponding memory value. */ static int mips_convert_register_p (int regnum, struct type *type) { return (gdbarch_byte_order (current_gdbarch) == BFD_ENDIAN_BIG && register_size (current_gdbarch, regnum) == 4 && (regnum % gdbarch_num_regs (current_gdbarch)) >= mips_regnum (current_gdbarch)->fp0 && (regnum % gdbarch_num_regs (current_gdbarch)) < mips_regnum (current_gdbarch)->fp0 + 32 && TYPE_CODE (type) == TYPE_CODE_FLT && TYPE_LENGTH (type) == 8); } static void mips_register_to_value (struct frame_info *frame, int regnum, struct type *type, gdb_byte *to) { get_frame_register (frame, regnum + 0, to + 4); get_frame_register (frame, regnum + 1, to + 0); } static void mips_value_to_register (struct frame_info *frame, int regnum, struct type *type, const gdb_byte *from) { put_frame_register (frame, regnum + 0, from + 4); put_frame_register (frame, regnum + 1, from + 0); } /* Return the GDB type object for the "standard" data type of data in register REG. */ static struct type * mips_register_type (struct gdbarch *gdbarch, int regnum) { gdb_assert (regnum >= 0 && regnum < 2 * gdbarch_num_regs (current_gdbarch)); if ((regnum % gdbarch_num_regs (current_gdbarch)) >= mips_regnum (current_gdbarch)->fp0 && (regnum % gdbarch_num_regs (current_gdbarch)) < mips_regnum (current_gdbarch)->fp0 + 32) { /* The floating-point registers raw, or cooked, always match mips_isa_regsize(), and also map 1:1, byte for byte. */ if (mips_isa_regsize (gdbarch) == 4) return builtin_type_ieee_single; else return builtin_type_ieee_double; } else if (regnum < gdbarch_num_regs (current_gdbarch)) { /* The raw or ISA registers. These are all sized according to the ISA regsize. */ if (mips_isa_regsize (gdbarch) == 4) return builtin_type_int32; else return builtin_type_int64; } else { /* The cooked or ABI registers. These are sized according to the ABI (with a few complications). */ if (regnum >= (gdbarch_num_regs (current_gdbarch) + mips_regnum (current_gdbarch)->fp_control_status) && regnum <= gdbarch_num_regs (current_gdbarch) + MIPS_LAST_EMBED_REGNUM) /* The pseudo/cooked view of the embedded registers is always 32-bit. The raw view is handled below. */ return builtin_type_int32; else if (gdbarch_tdep (gdbarch)->mips64_transfers_32bit_regs_p) /* The target, while possibly using a 64-bit register buffer, is only transfering 32-bits of each integer register. Reflect this in the cooked/pseudo (ABI) register value. */ return builtin_type_int32; else if (mips_abi_regsize (gdbarch) == 4) /* The ABI is restricted to 32-bit registers (the ISA could be 32- or 64-bit). */ return builtin_type_int32; else /* 64-bit ABI. */ return builtin_type_int64; } } /* Return the GDB type for the pseudo register REGNUM, which is the ABI-level view. This function is only called if there is a target description which includes registers, so we know precisely the types of hardware registers. */ static struct type * mips_pseudo_register_type (struct gdbarch *gdbarch, int regnum) { const int num_regs = gdbarch_num_regs (gdbarch); struct gdbarch_tdep *tdep = gdbarch_tdep (gdbarch); int rawnum = regnum % num_regs; struct type *rawtype; gdb_assert (regnum >= num_regs && regnum < 2 * num_regs); /* Absent registers are still absent. */ rawtype = gdbarch_register_type (gdbarch, rawnum); if (TYPE_LENGTH (rawtype) == 0) return rawtype; if (rawnum >= MIPS_EMBED_FP0_REGNUM && rawnum < MIPS_EMBED_FP0_REGNUM + 32) /* Present the floating point registers however the hardware did; do not try to convert between FPU layouts. */ return rawtype; if (rawnum >= MIPS_EMBED_FP0_REGNUM + 32 && rawnum <= MIPS_LAST_EMBED_REGNUM) { /* The pseudo/cooked view of embedded registers is always 32-bit, even if the target transfers 64-bit values for them. New targets relying on XML descriptions should only transfer the necessary 32 bits, but older versions of GDB expected 64, so allow the target to provide 64 bits without interfering with the displayed type. */ return builtin_type_int32; } /* Use pointer types for registers if we can. For n32 we can not, since we do not have a 64-bit pointer type. */ if (mips_abi_regsize (gdbarch) == TYPE_LENGTH (builtin_type_void_data_ptr)) { if (rawnum == MIPS_SP_REGNUM || rawnum == MIPS_EMBED_BADVADDR_REGNUM) return builtin_type_void_data_ptr; else if (rawnum == MIPS_EMBED_PC_REGNUM) return builtin_type_void_func_ptr; } if (mips_abi_regsize (gdbarch) == 4 && TYPE_LENGTH (rawtype) == 8 && rawnum >= MIPS_ZERO_REGNUM && rawnum <= MIPS_EMBED_PC_REGNUM) return builtin_type_int32; /* For all other registers, pass through the hardware type. */ return rawtype; } /* Should the upper word of 64-bit addresses be zeroed? */ enum auto_boolean mask_address_var = AUTO_BOOLEAN_AUTO; static int mips_mask_address_p (struct gdbarch_tdep *tdep) { switch (mask_address_var) { case AUTO_BOOLEAN_TRUE: return 1; case AUTO_BOOLEAN_FALSE: return 0; break; case AUTO_BOOLEAN_AUTO: return tdep->default_mask_address_p; default: internal_error (__FILE__, __LINE__, _("mips_mask_address_p: bad switch")); return -1; } } static void show_mask_address (struct ui_file *file, int from_tty, struct cmd_list_element *c, const char *value) { struct gdbarch_tdep *tdep = gdbarch_tdep (current_gdbarch); deprecated_show_value_hack (file, from_tty, c, value); switch (mask_address_var) { case AUTO_BOOLEAN_TRUE: printf_filtered ("The 32 bit mips address mask is enabled\n"); break; case AUTO_BOOLEAN_FALSE: printf_filtered ("The 32 bit mips address mask is disabled\n"); break; case AUTO_BOOLEAN_AUTO: printf_filtered ("The 32 bit address mask is set automatically. Currently %s\n", mips_mask_address_p (tdep) ? "enabled" : "disabled"); break; default: internal_error (__FILE__, __LINE__, _("show_mask_address: bad switch")); break; } } /* Tell if the program counter value in MEMADDR is in a MIPS16 function. */ int mips_pc_is_mips16 (CORE_ADDR memaddr) { struct minimal_symbol *sym; /* If bit 0 of the address is set, assume this is a MIPS16 address. */ if (is_mips16_addr (memaddr)) return 1; /* A flag indicating that this is a MIPS16 function is stored by elfread.c in the high bit of the info field. Use this to decide if the function is MIPS16 or normal MIPS. */ sym = lookup_minimal_symbol_by_pc (memaddr); if (sym) return msymbol_is_special (sym); else return 0; } /* MIPS believes that the PC has a sign extended value. Perhaps the all registers should be sign extended for simplicity? */ static CORE_ADDR mips_read_pc (ptid_t ptid) { return read_signed_register_pid (mips_regnum (current_gdbarch)->pc, ptid); } static CORE_ADDR mips_unwind_pc (struct gdbarch *gdbarch, struct frame_info *next_frame) { return frame_unwind_register_signed (next_frame, gdbarch_num_regs (current_gdbarch) + mips_regnum (gdbarch)->pc); } static CORE_ADDR mips_unwind_sp (struct gdbarch *gdbarch, struct frame_info *next_frame) { return frame_unwind_register_signed (next_frame, gdbarch_num_regs (current_gdbarch) + MIPS_SP_REGNUM); } /* Assuming NEXT_FRAME->prev is a dummy, return the frame ID of that dummy frame. The frame ID's base needs to match the TOS value saved by save_dummy_frame_tos(), and the PC match the dummy frame's breakpoint. */ static struct frame_id mips_unwind_dummy_id (struct gdbarch *gdbarch, struct frame_info *next_frame) { return frame_id_build (frame_unwind_register_signed (next_frame, gdbarch_num_regs (current_gdbarch) + MIPS_SP_REGNUM), frame_pc_unwind (next_frame)); } static void mips_write_pc (CORE_ADDR pc, ptid_t ptid) { write_register_pid (mips_regnum (current_gdbarch)->pc, pc, ptid); } /* Fetch and return instruction from the specified location. If the PC is odd, assume it's a MIPS16 instruction; otherwise MIPS32. */ static ULONGEST mips_fetch_instruction (CORE_ADDR addr) { gdb_byte buf[MIPS_INSN32_SIZE]; int instlen; int status; if (mips_pc_is_mips16 (addr)) { instlen = MIPS_INSN16_SIZE; addr = unmake_mips16_addr (addr); } else instlen = MIPS_INSN32_SIZE; status = read_memory_nobpt (addr, buf, instlen); if (status) memory_error (status, addr); return extract_unsigned_integer (buf, instlen); } /* These the fields of 32 bit mips instructions */ #define mips32_op(x) (x >> 26) #define itype_op(x) (x >> 26) #define itype_rs(x) ((x >> 21) & 0x1f) #define itype_rt(x) ((x >> 16) & 0x1f) #define itype_immediate(x) (x & 0xffff) #define jtype_op(x) (x >> 26) #define jtype_target(x) (x & 0x03ffffff) #define rtype_op(x) (x >> 26) #define rtype_rs(x) ((x >> 21) & 0x1f) #define rtype_rt(x) ((x >> 16) & 0x1f) #define rtype_rd(x) ((x >> 11) & 0x1f) #define rtype_shamt(x) ((x >> 6) & 0x1f) #define rtype_funct(x) (x & 0x3f) static LONGEST mips32_relative_offset (ULONGEST inst) { return ((itype_immediate (inst) ^ 0x8000) - 0x8000) << 2; } /* Determine where to set a single step breakpoint while considering branch prediction. */ static CORE_ADDR mips32_next_pc (struct frame_info *frame, CORE_ADDR pc) { unsigned long inst; int op; inst = mips_fetch_instruction (pc); if ((inst & 0xe0000000) != 0) /* Not a special, jump or branch instruction */ { if (itype_op (inst) >> 2 == 5) /* BEQL, BNEL, BLEZL, BGTZL: bits 0101xx */ { op = (itype_op (inst) & 0x03); switch (op) { case 0: /* BEQL */ goto equal_branch; case 1: /* BNEL */ goto neq_branch; case 2: /* BLEZL */ goto less_branch; case 3: /* BGTZ */ goto greater_branch; default: pc += 4; } } else if (itype_op (inst) == 17 && itype_rs (inst) == 8) /* BC1F, BC1FL, BC1T, BC1TL: 010001 01000 */ { int tf = itype_rt (inst) & 0x01; int cnum = itype_rt (inst) >> 2; int fcrcs = get_frame_register_signed (frame, mips_regnum (current_gdbarch)-> fp_control_status); int cond = ((fcrcs >> 24) & 0x0e) | ((fcrcs >> 23) & 0x01); if (((cond >> cnum) & 0x01) == tf) pc += mips32_relative_offset (inst) + 4; else pc += 8; } else pc += 4; /* Not a branch, next instruction is easy */ } else { /* This gets way messy */ /* Further subdivide into SPECIAL, REGIMM and other */ switch (op = itype_op (inst) & 0x07) /* extract bits 28,27,26 */ { case 0: /* SPECIAL */ op = rtype_funct (inst); switch (op) { case 8: /* JR */ case 9: /* JALR */ /* Set PC to that address */ pc = get_frame_register_signed (frame, rtype_rs (inst)); break; default: pc += 4; } break; /* end SPECIAL */ case 1: /* REGIMM */ { op = itype_rt (inst); /* branch condition */ switch (op) { case 0: /* BLTZ */ case 2: /* BLTZL */ case 16: /* BLTZAL */ case 18: /* BLTZALL */ less_branch: if (get_frame_register_signed (frame, itype_rs (inst)) < 0) pc += mips32_relative_offset (inst) + 4; else pc += 8; /* after the delay slot */ break; case 1: /* BGEZ */ case 3: /* BGEZL */ case 17: /* BGEZAL */ case 19: /* BGEZALL */ if (get_frame_register_signed (frame, itype_rs (inst)) >= 0) pc += mips32_relative_offset (inst) + 4; else pc += 8; /* after the delay slot */ break; /* All of the other instructions in the REGIMM category */ default: pc += 4; } } break; /* end REGIMM */ case 2: /* J */ case 3: /* JAL */ { unsigned long reg; reg = jtype_target (inst) << 2; /* Upper four bits get never changed... */ pc = reg + ((pc + 4) & ~(CORE_ADDR) 0x0fffffff); } break; /* FIXME case JALX : */ { unsigned long reg; reg = jtype_target (inst) << 2; pc = reg + ((pc + 4) & ~(CORE_ADDR) 0x0fffffff) + 1; /* yes, +1 */ /* Add 1 to indicate 16 bit mode - Invert ISA mode */ } break; /* The new PC will be alternate mode */ case 4: /* BEQ, BEQL */ equal_branch: if (get_frame_register_signed (frame, itype_rs (inst)) == get_frame_register_signed (frame, itype_rt (inst))) pc += mips32_relative_offset (inst) + 4; else pc += 8; break; case 5: /* BNE, BNEL */ neq_branch: if (get_frame_register_signed (frame, itype_rs (inst)) != get_frame_register_signed (frame, itype_rt (inst))) pc += mips32_relative_offset (inst) + 4; else pc += 8; break; case 6: /* BLEZ, BLEZL */ if (get_frame_register_signed (frame, itype_rs (inst)) <= 0) pc += mips32_relative_offset (inst) + 4; else pc += 8; break; case 7: default: greater_branch: /* BGTZ, BGTZL */ if (get_frame_register_signed (frame, itype_rs (inst)) > 0) pc += mips32_relative_offset (inst) + 4; else pc += 8; break; } /* switch */ } /* else */ return pc; } /* mips32_next_pc */ /* Decoding the next place to set a breakpoint is irregular for the mips 16 variant, but fortunately, there fewer instructions. We have to cope ith extensions for 16 bit instructions and a pair of actual 32 bit instructions. We dont want to set a single step instruction on the extend instruction either. */ /* Lots of mips16 instruction formats */ /* Predicting jumps requires itype,ritype,i8type and their extensions extItype,extritype,extI8type */ enum mips16_inst_fmts { itype, /* 0 immediate 5,10 */ ritype, /* 1 5,3,8 */ rrtype, /* 2 5,3,3,5 */ rritype, /* 3 5,3,3,5 */ rrrtype, /* 4 5,3,3,3,2 */ rriatype, /* 5 5,3,3,1,4 */ shifttype, /* 6 5,3,3,3,2 */ i8type, /* 7 5,3,8 */ i8movtype, /* 8 5,3,3,5 */ i8mov32rtype, /* 9 5,3,5,3 */ i64type, /* 10 5,3,8 */ ri64type, /* 11 5,3,3,5 */ jalxtype, /* 12 5,1,5,5,16 - a 32 bit instruction */ exiItype, /* 13 5,6,5,5,1,1,1,1,1,1,5 */ extRitype, /* 14 5,6,5,5,3,1,1,1,5 */ extRRItype, /* 15 5,5,5,5,3,3,5 */ extRRIAtype, /* 16 5,7,4,5,3,3,1,4 */ EXTshifttype, /* 17 5,5,1,1,1,1,1,1,5,3,3,1,1,1,2 */ extI8type, /* 18 5,6,5,5,3,1,1,1,5 */ extI64type, /* 19 5,6,5,5,3,1,1,1,5 */ extRi64type, /* 20 5,6,5,5,3,3,5 */ extshift64type /* 21 5,5,1,1,1,1,1,1,5,1,1,1,3,5 */ }; /* I am heaping all the fields of the formats into one structure and then, only the fields which are involved in instruction extension */ struct upk_mips16 { CORE_ADDR offset; unsigned int regx; /* Function in i8 type */ unsigned int regy; }; /* The EXT-I, EXT-ri nad EXT-I8 instructions all have the same format for the bits which make up the immediatate extension. */ static CORE_ADDR extended_offset (unsigned int extension) { CORE_ADDR value; value = (extension >> 21) & 0x3f; /* * extract 15:11 */ value = value << 6; value |= (extension >> 16) & 0x1f; /* extrace 10:5 */ value = value << 5; value |= extension & 0x01f; /* extract 4:0 */ return value; } /* Only call this function if you know that this is an extendable instruction, It wont malfunction, but why make excess remote memory references? If the immediate operands get sign extended or somthing, do it after the extension is performed. */ /* FIXME: Every one of these cases needs to worry about sign extension when the offset is to be used in relative addressing */ static unsigned int fetch_mips_16 (CORE_ADDR pc) { gdb_byte buf[8]; pc &= 0xfffffffe; /* clear the low order bit */ target_read_memory (pc, buf, 2); return extract_unsigned_integer (buf, 2); } static void unpack_mips16 (CORE_ADDR pc, unsigned int extension, unsigned int inst, enum mips16_inst_fmts insn_format, struct upk_mips16 *upk) { CORE_ADDR offset; int regx; int regy; switch (insn_format) { case itype: { CORE_ADDR value; if (extension) { value = extended_offset (extension); value = value << 11; /* rom for the original value */ value |= inst & 0x7ff; /* eleven bits from instruction */ } else { value = inst & 0x7ff; /* FIXME : Consider sign extension */ } offset = value; regx = -1; regy = -1; } break; case ritype: case i8type: { /* A register identifier and an offset */ /* Most of the fields are the same as I type but the immediate value is of a different length */ CORE_ADDR value; if (extension) { value = extended_offset (extension); value = value << 8; /* from the original instruction */ value |= inst & 0xff; /* eleven bits from instruction */ regx = (extension >> 8) & 0x07; /* or i8 funct */ if (value & 0x4000) /* test the sign bit , bit 26 */ { value &= ~0x3fff; /* remove the sign bit */ value = -value; } } else { value = inst & 0xff; /* 8 bits */ regx = (inst >> 8) & 0x07; /* or i8 funct */ /* FIXME: Do sign extension , this format needs it */ if (value & 0x80) /* THIS CONFUSES ME */ { value &= 0xef; /* remove the sign bit */ value = -value; } } offset = value; regy = -1; break; } case jalxtype: { unsigned long value; unsigned int nexthalf; value = ((inst & 0x1f) << 5) | ((inst >> 5) & 0x1f); value = value << 16; nexthalf = mips_fetch_instruction (pc + 2); /* low bit still set */ value |= nexthalf; offset = value; regx = -1; regy = -1; break; } default: internal_error (__FILE__, __LINE__, _("bad switch")); } upk->offset = offset; upk->regx = regx; upk->regy = regy; } static CORE_ADDR add_offset_16 (CORE_ADDR pc, int offset) { return ((offset << 2) | ((pc + 2) & (~(CORE_ADDR) 0x0fffffff))); } static CORE_ADDR extended_mips16_next_pc (struct frame_info *frame, CORE_ADDR pc, unsigned int extension, unsigned int insn) { int op = (insn >> 11); switch (op) { case 2: /* Branch */ { CORE_ADDR offset; struct upk_mips16 upk; unpack_mips16 (pc, extension, insn, itype, &upk); offset = upk.offset; if (offset & 0x800) { offset &= 0xeff; offset = -offset; } pc += (offset << 1) + 2; break; } case 3: /* JAL , JALX - Watch out, these are 32 bit instruction */ { struct upk_mips16 upk; unpack_mips16 (pc, extension, insn, jalxtype, &upk); pc = add_offset_16 (pc, upk.offset); if ((insn >> 10) & 0x01) /* Exchange mode */ pc = pc & ~0x01; /* Clear low bit, indicate 32 bit mode */ else pc |= 0x01; break; } case 4: /* beqz */ { struct upk_mips16 upk; int reg; unpack_mips16 (pc, extension, insn, ritype, &upk); reg = get_frame_register_signed (frame, upk.regx); if (reg == 0) pc += (upk.offset << 1) + 2; else pc += 2; break; } case 5: /* bnez */ { struct upk_mips16 upk; int reg; unpack_mips16 (pc, extension, insn, ritype, &upk); reg = get_frame_register_signed (frame, upk.regx); if (reg != 0) pc += (upk.offset << 1) + 2; else pc += 2; break; } case 12: /* I8 Formats btez btnez */ { struct upk_mips16 upk; int reg; unpack_mips16 (pc, extension, insn, i8type, &upk); /* upk.regx contains the opcode */ reg = get_frame_register_signed (frame, 24); /* Test register is 24 */ if (((upk.regx == 0) && (reg == 0)) /* BTEZ */ || ((upk.regx == 1) && (reg != 0))) /* BTNEZ */ /* pc = add_offset_16(pc,upk.offset) ; */ pc += (upk.offset << 1) + 2; else pc += 2; break; } case 29: /* RR Formats JR, JALR, JALR-RA */ { struct upk_mips16 upk; /* upk.fmt = rrtype; */ op = insn & 0x1f; if (op == 0) { int reg; upk.regx = (insn >> 8) & 0x07; upk.regy = (insn >> 5) & 0x07; switch (upk.regy) { case 0: reg = upk.regx; break; case 1: reg = 31; break; /* Function return instruction */ case 2: reg = upk.regx; break; default: reg = 31; break; /* BOGUS Guess */ } pc = get_frame_register_signed (frame, reg); } else pc += 2; break; } case 30: /* This is an instruction extension. Fetch the real instruction (which follows the extension) and decode things based on that. */ { pc += 2; pc = extended_mips16_next_pc (frame, pc, insn, fetch_mips_16 (pc)); break; } default: { pc += 2; break; } } return pc; } static CORE_ADDR mips16_next_pc (struct frame_info *frame, CORE_ADDR pc) { unsigned int insn = fetch_mips_16 (pc); return extended_mips16_next_pc (frame, pc, 0, insn); } /* The mips_next_pc function supports single_step when the remote target monitor or stub is not developed enough to do a single_step. It works by decoding the current instruction and predicting where a branch will go. This isnt hard because all the data is available. The MIPS32 and MIPS16 variants are quite different */ static CORE_ADDR mips_next_pc (struct frame_info *frame, CORE_ADDR pc) { if (pc & 0x01) return mips16_next_pc (frame, pc); else return mips32_next_pc (frame, pc); } struct mips_frame_cache { CORE_ADDR base; struct trad_frame_saved_reg *saved_regs; }; /* Set a register's saved stack address in temp_saved_regs. If an address has already been set for this register, do nothing; this way we will only recognize the first save of a given register in a function prologue. For simplicity, save the address in both [0 .. gdbarch_num_regs) and [gdbarch_num_regs .. 2*gdbarch_num_regs). Strictly speaking, only the second range is used as it is only second range (the ABI instead of ISA registers) that comes into play when finding saved registers in a frame. */ static void set_reg_offset (struct mips_frame_cache *this_cache, int regnum, CORE_ADDR offset) { if (this_cache != NULL && this_cache->saved_regs[regnum].addr == -1) { this_cache->saved_regs[regnum + 0 * gdbarch_num_regs (current_gdbarch)].addr = offset; this_cache->saved_regs[regnum + 1 * gdbarch_num_regs (current_gdbarch)].addr = offset; } } /* Fetch the immediate value from a MIPS16 instruction. If the previous instruction was an EXTEND, use it to extend the upper bits of the immediate value. This is a helper function for mips16_scan_prologue. */ static int mips16_get_imm (unsigned short prev_inst, /* previous instruction */ unsigned short inst, /* current instruction */ int nbits, /* number of bits in imm field */ int scale, /* scale factor to be applied to imm */ int is_signed) /* is the imm field signed? */ { int offset; if ((prev_inst & 0xf800) == 0xf000) /* prev instruction was EXTEND? */ { offset = ((prev_inst & 0x1f) << 11) | (prev_inst & 0x7e0); if (offset & 0x8000) /* check for negative extend */ offset = 0 - (0x10000 - (offset & 0xffff)); return offset | (inst & 0x1f); } else { int max_imm = 1 << nbits; int mask = max_imm - 1; int sign_bit = max_imm >> 1; offset = inst & mask; if (is_signed && (offset & sign_bit)) offset = 0 - (max_imm - offset); return offset * scale; } } /* Analyze the function prologue from START_PC to LIMIT_PC. Builds the associated FRAME_CACHE if not null. Return the address of the first instruction past the prologue. */ static CORE_ADDR mips16_scan_prologue (CORE_ADDR start_pc, CORE_ADDR limit_pc, struct frame_info *next_frame, struct mips_frame_cache *this_cache) { CORE_ADDR cur_pc; CORE_ADDR frame_addr = 0; /* Value of $r17, used as frame pointer */ CORE_ADDR sp; long frame_offset = 0; /* Size of stack frame. */ long frame_adjust = 0; /* Offset of FP from SP. */ int frame_reg = MIPS_SP_REGNUM; unsigned short prev_inst = 0; /* saved copy of previous instruction */ unsigned inst = 0; /* current instruction */ unsigned entry_inst = 0; /* the entry instruction */ int reg, offset; int extend_bytes = 0; int prev_extend_bytes; CORE_ADDR end_prologue_addr = 0; /* Can be called when there's no process, and hence when there's no NEXT_FRAME. */ if (next_frame != NULL) sp = frame_unwind_register_signed (next_frame, gdbarch_num_regs (current_gdbarch) + MIPS_SP_REGNUM); else sp = 0; if (limit_pc > start_pc + 200) limit_pc = start_pc + 200; for (cur_pc = start_pc; cur_pc < limit_pc; cur_pc += MIPS_INSN16_SIZE) { /* Save the previous instruction. If it's an EXTEND, we'll extract the immediate offset extension from it in mips16_get_imm. */ prev_inst = inst; /* Fetch and decode the instruction. */ inst = (unsigned short) mips_fetch_instruction (cur_pc); /* Normally we ignore extend instructions. However, if it is not followed by a valid prologue instruction, then this instruction is not part of the prologue either. We must remember in this case to adjust the end_prologue_addr back over the extend. */ if ((inst & 0xf800) == 0xf000) /* extend */ { extend_bytes = MIPS_INSN16_SIZE; continue; } prev_extend_bytes = extend_bytes; extend_bytes = 0; if ((inst & 0xff00) == 0x6300 /* addiu sp */ || (inst & 0xff00) == 0xfb00) /* daddiu sp */ { offset = mips16_get_imm (prev_inst, inst, 8, 8, 1); if (offset < 0) /* negative stack adjustment? */ frame_offset -= offset; else /* Exit loop if a positive stack adjustment is found, which usually means that the stack cleanup code in the function epilogue is reached. */ break; } else if ((inst & 0xf800) == 0xd000) /* sw reg,n($sp) */ { offset = mips16_get_imm (prev_inst, inst, 8, 4, 0); reg = mips16_to_32_reg[(inst & 0x700) >> 8]; set_reg_offset (this_cache, reg, sp + offset); } else if ((inst & 0xff00) == 0xf900) /* sd reg,n($sp) */ { offset = mips16_get_imm (prev_inst, inst, 5, 8, 0); reg = mips16_to_32_reg[(inst & 0xe0) >> 5]; set_reg_offset (this_cache, reg, sp + offset); } else if ((inst & 0xff00) == 0x6200) /* sw $ra,n($sp) */ { offset = mips16_get_imm (prev_inst, inst, 8, 4, 0); set_reg_offset (this_cache, MIPS_RA_REGNUM, sp + offset); } else if ((inst & 0xff00) == 0xfa00) /* sd $ra,n($sp) */ { offset = mips16_get_imm (prev_inst, inst, 8, 8, 0); set_reg_offset (this_cache, MIPS_RA_REGNUM, sp + offset); } else if (inst == 0x673d) /* move $s1, $sp */ { frame_addr = sp; frame_reg = 17; } else if ((inst & 0xff00) == 0x0100) /* addiu $s1,sp,n */ { offset = mips16_get_imm (prev_inst, inst, 8, 4, 0); frame_addr = sp + offset; frame_reg = 17; frame_adjust = offset; } else if ((inst & 0xFF00) == 0xd900) /* sw reg,offset($s1) */ { offset = mips16_get_imm (prev_inst, inst, 5, 4, 0); reg = mips16_to_32_reg[(inst & 0xe0) >> 5]; set_reg_offset (this_cache, reg, frame_addr + offset); } else if ((inst & 0xFF00) == 0x7900) /* sd reg,offset($s1) */ { offset = mips16_get_imm (prev_inst, inst, 5, 8, 0); reg = mips16_to_32_reg[(inst & 0xe0) >> 5]; set_reg_offset (this_cache, reg, frame_addr + offset); } else if ((inst & 0xf81f) == 0xe809 && (inst & 0x700) != 0x700) /* entry */ entry_inst = inst; /* save for later processing */ else if ((inst & 0xf800) == 0x1800) /* jal(x) */ cur_pc += MIPS_INSN16_SIZE; /* 32-bit instruction */ else if ((inst & 0xff1c) == 0x6704) /* move reg,$a0-$a3 */ { /* This instruction is part of the prologue, but we don't need to do anything special to handle it. */ } else { /* This instruction is not an instruction typically found in a prologue, so we must have reached the end of the prologue. */ if (end_prologue_addr == 0) end_prologue_addr = cur_pc - prev_extend_bytes; } } /* The entry instruction is typically the first instruction in a function, and it stores registers at offsets relative to the value of the old SP (before the prologue). But the value of the sp parameter to this function is the new SP (after the prologue has been executed). So we can't calculate those offsets until we've seen the entire prologue, and can calculate what the old SP must have been. */ if (entry_inst != 0) { int areg_count = (entry_inst >> 8) & 7; int sreg_count = (entry_inst >> 6) & 3; /* The entry instruction always subtracts 32 from the SP. */ frame_offset += 32; /* Now we can calculate what the SP must have been at the start of the function prologue. */ sp += frame_offset; /* Check if a0-a3 were saved in the caller's argument save area. */ for (reg = 4, offset = 0; reg < areg_count + 4; reg++) { set_reg_offset (this_cache, reg, sp + offset); offset += mips_abi_regsize (current_gdbarch); } /* Check if the ra register was pushed on the stack. */ offset = -4; if (entry_inst & 0x20) { set_reg_offset (this_cache, MIPS_RA_REGNUM, sp + offset); offset -= mips_abi_regsize (current_gdbarch); } /* Check if the s0 and s1 registers were pushed on the stack. */ for (reg = 16; reg < sreg_count + 16; reg++) { set_reg_offset (this_cache, reg, sp + offset); offset -= mips_abi_regsize (current_gdbarch); } } if (this_cache != NULL) { this_cache->base = (frame_unwind_register_signed (next_frame, gdbarch_num_regs (current_gdbarch) + frame_reg) + frame_offset - frame_adjust); /* FIXME: brobecker/2004-10-10: Just as in the mips32 case, we should be able to get rid of the assignment below, evetually. But it's still needed for now. */ this_cache->saved_regs[gdbarch_num_regs (current_gdbarch) + mips_regnum (current_gdbarch)->pc] = this_cache->saved_regs[gdbarch_num_regs (current_gdbarch) + MIPS_RA_REGNUM]; } /* If we didn't reach the end of the prologue when scanning the function instructions, then set end_prologue_addr to the address of the instruction immediately after the last one we scanned. */ if (end_prologue_addr == 0) end_prologue_addr = cur_pc; return end_prologue_addr; } /* Heuristic unwinder for 16-bit MIPS instruction set (aka MIPS16). Procedures that use the 32-bit instruction set are handled by the mips_insn32 unwinder. */ static struct mips_frame_cache * mips_insn16_frame_cache (struct frame_info *next_frame, void **this_cache) { struct mips_frame_cache *cache; if ((*this_cache) != NULL) return (*this_cache); cache = FRAME_OBSTACK_ZALLOC (struct mips_frame_cache); (*this_cache) = cache; cache->saved_regs = trad_frame_alloc_saved_regs (next_frame); /* Analyze the function prologue. */ { const CORE_ADDR pc = frame_unwind_address_in_block (next_frame, NORMAL_FRAME); CORE_ADDR start_addr; find_pc_partial_function (pc, NULL, &start_addr, NULL); if (start_addr == 0) start_addr = heuristic_proc_start (pc); /* We can't analyze the prologue if we couldn't find the begining of the function. */ if (start_addr == 0) return cache; mips16_scan_prologue (start_addr, pc, next_frame, *this_cache); } /* SP_REGNUM, contains the value and not the address. */ trad_frame_set_value (cache->saved_regs, gdbarch_num_regs (current_gdbarch) + MIPS_SP_REGNUM, cache->base); return (*this_cache); } static void mips_insn16_frame_this_id (struct frame_info *next_frame, void **this_cache, struct frame_id *this_id) { struct mips_frame_cache *info = mips_insn16_frame_cache (next_frame, this_cache); (*this_id) = frame_id_build (info->base, frame_func_unwind (next_frame, NORMAL_FRAME)); } static void mips_insn16_frame_prev_register (struct frame_info *next_frame, void **this_cache, int regnum, int *optimizedp, enum lval_type *lvalp, CORE_ADDR *addrp, int *realnump, gdb_byte *valuep) { struct mips_frame_cache *info = mips_insn16_frame_cache (next_frame, this_cache); trad_frame_get_prev_register (next_frame, info->saved_regs, regnum, optimizedp, lvalp, addrp, realnump, valuep); } static const struct frame_unwind mips_insn16_frame_unwind = { NORMAL_FRAME, mips_insn16_frame_this_id, mips_insn16_frame_prev_register }; static const struct frame_unwind * mips_insn16_frame_sniffer (struct frame_info *next_frame) { CORE_ADDR pc = frame_pc_unwind (next_frame); if (mips_pc_is_mips16 (pc)) return &mips_insn16_frame_unwind; return NULL; } static CORE_ADDR mips_insn16_frame_base_address (struct frame_info *next_frame, void **this_cache) { struct mips_frame_cache *info = mips_insn16_frame_cache (next_frame, this_cache); return info->base; } static const struct frame_base mips_insn16_frame_base = { &mips_insn16_frame_unwind, mips_insn16_frame_base_address, mips_insn16_frame_base_address, mips_insn16_frame_base_address }; static const struct frame_base * mips_insn16_frame_base_sniffer (struct frame_info *next_frame) { if (mips_insn16_frame_sniffer (next_frame) != NULL) return &mips_insn16_frame_base; else return NULL; } /* Mark all the registers as unset in the saved_regs array of THIS_CACHE. Do nothing if THIS_CACHE is null. */ void reset_saved_regs (struct mips_frame_cache *this_cache) { if (this_cache == NULL || this_cache->saved_regs == NULL) return; { const int num_regs = gdbarch_num_regs (current_gdbarch); int i; for (i = 0; i < num_regs; i++) { this_cache->saved_regs[i].addr = -1; } } } /* Analyze the function prologue from START_PC to LIMIT_PC. Builds the associated FRAME_CACHE if not null. Return the address of the first instruction past the prologue. */ static CORE_ADDR mips32_scan_prologue (CORE_ADDR start_pc, CORE_ADDR limit_pc, struct frame_info *next_frame, struct mips_frame_cache *this_cache) { CORE_ADDR cur_pc; CORE_ADDR frame_addr = 0; /* Value of $r30. Used by gcc for frame-pointer */ CORE_ADDR sp; long frame_offset; int frame_reg = MIPS_SP_REGNUM; CORE_ADDR end_prologue_addr = 0; int seen_sp_adjust = 0; int load_immediate_bytes = 0; /* Can be called when there's no process, and hence when there's no NEXT_FRAME. */ if (next_frame != NULL) sp = frame_unwind_register_signed (next_frame, gdbarch_num_regs (current_gdbarch) + MIPS_SP_REGNUM); else sp = 0; if (limit_pc > start_pc + 200) limit_pc = start_pc + 200; restart: frame_offset = 0; for (cur_pc = start_pc; cur_pc < limit_pc; cur_pc += MIPS_INSN32_SIZE) { unsigned long inst, high_word, low_word; int reg; /* Fetch the instruction. */ inst = (unsigned long) mips_fetch_instruction (cur_pc); /* Save some code by pre-extracting some useful fields. */ high_word = (inst >> 16) & 0xffff; low_word = inst & 0xffff; reg = high_word & 0x1f; if (high_word == 0x27bd /* addiu $sp,$sp,-i */ || high_word == 0x23bd /* addi $sp,$sp,-i */ || high_word == 0x67bd) /* daddiu $sp,$sp,-i */ { if (low_word & 0x8000) /* negative stack adjustment? */ frame_offset += 0x10000 - low_word; else /* Exit loop if a positive stack adjustment is found, which usually means that the stack cleanup code in the function epilogue is reached. */ break; seen_sp_adjust = 1; } else if ((high_word & 0xFFE0) == 0xafa0) /* sw reg,offset($sp) */ { set_reg_offset (this_cache, reg, sp + low_word); } else if ((high_word & 0xFFE0) == 0xffa0) /* sd reg,offset($sp) */ { /* Irix 6.2 N32 ABI uses sd instructions for saving $gp and $ra. */ set_reg_offset (this_cache, reg, sp + low_word); } else if (high_word == 0x27be) /* addiu $30,$sp,size */ { /* Old gcc frame, r30 is virtual frame pointer. */ if ((long) low_word != frame_offset) frame_addr = sp + low_word; else if (next_frame && frame_reg == MIPS_SP_REGNUM) { unsigned alloca_adjust; frame_reg = 30; frame_addr = frame_unwind_register_signed (next_frame, gdbarch_num_regs (current_gdbarch) + 30); alloca_adjust = (unsigned) (frame_addr - (sp + low_word)); if (alloca_adjust > 0) { /* FP > SP + frame_size. This may be because of an alloca or somethings similar. Fix sp to "pre-alloca" value, and try again. */ sp += alloca_adjust; /* Need to reset the status of all registers. Otherwise, we will hit a guard that prevents the new address for each register to be recomputed during the second pass. */ reset_saved_regs (this_cache); goto restart; } } } /* move $30,$sp. With different versions of gas this will be either `addu $30,$sp,$zero' or `or $30,$sp,$zero' or `daddu 30,sp,$0'. Accept any one of these. */ else if (inst == 0x03A0F021 || inst == 0x03a0f025 || inst == 0x03a0f02d) { /* New gcc frame, virtual frame pointer is at r30 + frame_size. */ if (next_frame && frame_reg == MIPS_SP_REGNUM) { unsigned alloca_adjust; frame_reg = 30; frame_addr = frame_unwind_register_signed (next_frame, gdbarch_num_regs (current_gdbarch) + 30); alloca_adjust = (unsigned) (frame_addr - sp); if (alloca_adjust > 0) { /* FP > SP + frame_size. This may be because of an alloca or somethings similar. Fix sp to "pre-alloca" value, and try again. */ sp = frame_addr; /* Need to reset the status of all registers. Otherwise, we will hit a guard that prevents the new address for each register to be recomputed during the second pass. */ reset_saved_regs (this_cache); goto restart; } } } else if ((high_word & 0xFFE0) == 0xafc0) /* sw reg,offset($30) */ { set_reg_offset (this_cache, reg, frame_addr + low_word); } else if ((high_word & 0xFFE0) == 0xE7A0 /* swc1 freg,n($sp) */ || (high_word & 0xF3E0) == 0xA3C0 /* sx reg,n($s8) */ || (inst & 0xFF9F07FF) == 0x00800021 /* move reg,$a0-$a3 */ || high_word == 0x3c1c /* lui $gp,n */ || high_word == 0x279c /* addiu $gp,$gp,n */ || inst == 0x0399e021 /* addu $gp,$gp,$t9 */ || inst == 0x033ce021 /* addu $gp,$t9,$gp */ ) { /* These instructions are part of the prologue, but we don't need to do anything special to handle them. */ } /* The instructions below load $at or $t0 with an immediate value in preparation for a stack adjustment via subu $sp,$sp,[$at,$t0]. These instructions could also initialize a local variable, so we accept them only before a stack adjustment instruction was seen. */ else if (!seen_sp_adjust && (high_word == 0x3c01 /* lui $at,n */ || high_word == 0x3c08 /* lui $t0,n */ || high_word == 0x3421 /* ori $at,$at,n */ || high_word == 0x3508 /* ori $t0,$t0,n */ || high_word == 0x3401 /* ori $at,$zero,n */ || high_word == 0x3408 /* ori $t0,$zero,n */ )) { load_immediate_bytes += MIPS_INSN32_SIZE; /* FIXME! */ } else { /* This instruction is not an instruction typically found in a prologue, so we must have reached the end of the prologue. */ /* FIXME: brobecker/2004-10-10: Can't we just break out of this loop now? Why would we need to continue scanning the function instructions? */ if (end_prologue_addr == 0) end_prologue_addr = cur_pc; } } if (this_cache != NULL) { this_cache->base = (frame_unwind_register_signed (next_frame, gdbarch_num_regs (current_gdbarch) + frame_reg) + frame_offset); /* FIXME: brobecker/2004-09-15: We should be able to get rid of this assignment below, eventually. But it's still needed for now. */ this_cache->saved_regs[gdbarch_num_regs (current_gdbarch) + mips_regnum (current_gdbarch)->pc] = this_cache->saved_regs[gdbarch_num_regs (current_gdbarch) + MIPS_RA_REGNUM]; } /* If we didn't reach the end of the prologue when scanning the function instructions, then set end_prologue_addr to the address of the instruction immediately after the last one we scanned. */ /* brobecker/2004-10-10: I don't think this would ever happen, but we may as well be careful and do our best if we have a null end_prologue_addr. */ if (end_prologue_addr == 0) end_prologue_addr = cur_pc; /* In a frameless function, we might have incorrectly skipped some load immediate instructions. Undo the skipping if the load immediate was not followed by a stack adjustment. */ if (load_immediate_bytes && !seen_sp_adjust) end_prologue_addr -= load_immediate_bytes; return end_prologue_addr; } /* Heuristic unwinder for procedures using 32-bit instructions (covers both 32-bit and 64-bit MIPS ISAs). Procedures using 16-bit instructions (a.k.a. MIPS16) are handled by the mips_insn16 unwinder. */ static struct mips_frame_cache * mips_insn32_frame_cache (struct frame_info *next_frame, void **this_cache) { struct mips_frame_cache *cache; if ((*this_cache) != NULL) return (*this_cache); cache = FRAME_OBSTACK_ZALLOC (struct mips_frame_cache); (*this_cache) = cache; cache->saved_regs = trad_frame_alloc_saved_regs (next_frame); /* Analyze the function prologue. */ { const CORE_ADDR pc = frame_unwind_address_in_block (next_frame, NORMAL_FRAME); CORE_ADDR start_addr; find_pc_partial_function (pc, NULL, &start_addr, NULL); if (start_addr == 0) start_addr = heuristic_proc_start (pc); /* We can't analyze the prologue if we couldn't find the begining of the function. */ if (start_addr == 0) return cache; mips32_scan_prologue (start_addr, pc, next_frame, *this_cache); } /* SP_REGNUM, contains the value and not the address. */ trad_frame_set_value (cache->saved_regs, gdbarch_num_regs (current_gdbarch) + MIPS_SP_REGNUM, cache->base); return (*this_cache); } static void mips_insn32_frame_this_id (struct frame_info *next_frame, void **this_cache, struct frame_id *this_id) { struct mips_frame_cache *info = mips_insn32_frame_cache (next_frame, this_cache); (*this_id) = frame_id_build (info->base, frame_func_unwind (next_frame, NORMAL_FRAME)); } static void mips_insn32_frame_prev_register (struct frame_info *next_frame, void **this_cache, int regnum, int *optimizedp, enum lval_type *lvalp, CORE_ADDR *addrp, int *realnump, gdb_byte *valuep) { struct mips_frame_cache *info = mips_insn32_frame_cache (next_frame, this_cache); trad_frame_get_prev_register (next_frame, info->saved_regs, regnum, optimizedp, lvalp, addrp, realnump, valuep); } static const struct frame_unwind mips_insn32_frame_unwind = { NORMAL_FRAME, mips_insn32_frame_this_id, mips_insn32_frame_prev_register }; static const struct frame_unwind * mips_insn32_frame_sniffer (struct frame_info *next_frame) { CORE_ADDR pc = frame_pc_unwind (next_frame); if (! mips_pc_is_mips16 (pc)) return &mips_insn32_frame_unwind; return NULL; } static CORE_ADDR mips_insn32_frame_base_address (struct frame_info *next_frame, void **this_cache) { struct mips_frame_cache *info = mips_insn32_frame_cache (next_frame, this_cache); return info->base; } static const struct frame_base mips_insn32_frame_base = { &mips_insn32_frame_unwind, mips_insn32_frame_base_address, mips_insn32_frame_base_address, mips_insn32_frame_base_address }; static const struct frame_base * mips_insn32_frame_base_sniffer (struct frame_info *next_frame) { if (mips_insn32_frame_sniffer (next_frame) != NULL) return &mips_insn32_frame_base; else return NULL; } static struct trad_frame_cache * mips_stub_frame_cache (struct frame_info *next_frame, void **this_cache) { CORE_ADDR pc; CORE_ADDR start_addr; CORE_ADDR stack_addr; struct trad_frame_cache *this_trad_cache; if ((*this_cache) != NULL) return (*this_cache); this_trad_cache = trad_frame_cache_zalloc (next_frame); (*this_cache) = this_trad_cache; /* The return address is in the link register. */ trad_frame_set_reg_realreg (this_trad_cache, PC_REGNUM, MIPS_RA_REGNUM); /* Frame ID, since it's a frameless / stackless function, no stack space is allocated and SP on entry is the current SP. */ pc = frame_pc_unwind (next_frame); find_pc_partial_function (pc, NULL, &start_addr, NULL); stack_addr = frame_unwind_register_signed (next_frame, MIPS_SP_REGNUM); trad_frame_set_id (this_trad_cache, frame_id_build (start_addr, stack_addr)); /* Assume that the frame's base is the same as the stack-pointer. */ trad_frame_set_this_base (this_trad_cache, stack_addr); return this_trad_cache; } static void mips_stub_frame_this_id (struct frame_info *next_frame, void **this_cache, struct frame_id *this_id) { struct trad_frame_cache *this_trad_cache = mips_stub_frame_cache (next_frame, this_cache); trad_frame_get_id (this_trad_cache, this_id); } static void mips_stub_frame_prev_register (struct frame_info *next_frame, void **this_cache, int regnum, int *optimizedp, enum lval_type *lvalp, CORE_ADDR *addrp, int *realnump, gdb_byte *valuep) { struct trad_frame_cache *this_trad_cache = mips_stub_frame_cache (next_frame, this_cache); trad_frame_get_register (this_trad_cache, next_frame, regnum, optimizedp, lvalp, addrp, realnump, valuep); } static const struct frame_unwind mips_stub_frame_unwind = { NORMAL_FRAME, mips_stub_frame_this_id, mips_stub_frame_prev_register }; static const struct frame_unwind * mips_stub_frame_sniffer (struct frame_info *next_frame) { struct obj_section *s; CORE_ADDR pc = frame_unwind_address_in_block (next_frame, NORMAL_FRAME); if (in_plt_section (pc, NULL)) return &mips_stub_frame_unwind; /* Binutils for MIPS puts lazy resolution stubs into .MIPS.stubs. */ s = find_pc_section (pc); if (s != NULL && strcmp (bfd_get_section_name (s->objfile->obfd, s->the_bfd_section), ".MIPS.stubs") == 0) return &mips_stub_frame_unwind; return NULL; } static CORE_ADDR mips_stub_frame_base_address (struct frame_info *next_frame, void **this_cache) { struct trad_frame_cache *this_trad_cache = mips_stub_frame_cache (next_frame, this_cache); return trad_frame_get_this_base (this_trad_cache); } static const struct frame_base mips_stub_frame_base = { &mips_stub_frame_unwind, mips_stub_frame_base_address, mips_stub_frame_base_address, mips_stub_frame_base_address }; static const struct frame_base * mips_stub_frame_base_sniffer (struct frame_info *next_frame) { if (mips_stub_frame_sniffer (next_frame) != NULL) return &mips_stub_frame_base; else return NULL; } /* mips_addr_bits_remove - remove useless address bits */ static CORE_ADDR mips_addr_bits_remove (CORE_ADDR addr) { struct gdbarch_tdep *tdep = gdbarch_tdep (current_gdbarch); if (mips_mask_address_p (tdep) && (((ULONGEST) addr) >> 32 == 0xffffffffUL)) /* This hack is a work-around for existing boards using PMON, the simulator, and any other 64-bit targets that doesn't have true 64-bit addressing. On these targets, the upper 32 bits of addresses are ignored by the hardware. Thus, the PC or SP are likely to have been sign extended to all 1s by instruction sequences that load 32-bit addresses. For example, a typical piece of code that loads an address is this: lui $r2, ori $r2, But the lui sign-extends the value such that the upper 32 bits may be all 1s. The workaround is simply to mask off these bits. In the future, gcc may be changed to support true 64-bit addressing, and this masking will have to be disabled. */ return addr &= 0xffffffffUL; else return addr; } /* mips_software_single_step() is called just before we want to resume the inferior, if we want to single-step it but there is no hardware or kernel single-step support (MIPS on GNU/Linux for example). We find the target of the coming instruction and breakpoint it. */ int mips_software_single_step (struct frame_info *frame) { CORE_ADDR pc, next_pc; pc = get_frame_pc (frame); next_pc = mips_next_pc (frame, pc); insert_single_step_breakpoint (next_pc); return 1; } /* Test whether the PC points to the return instruction at the end of a function. */ static int mips_about_to_return (CORE_ADDR pc) { if (mips_pc_is_mips16 (pc)) /* This mips16 case isn't necessarily reliable. Sometimes the compiler generates a "jr $ra"; other times it generates code to load the return address from the stack to an accessible register (such as $a3), then a "jr" using that register. This second case is almost impossible to distinguish from an indirect jump used for switch statements, so we don't even try. */ return mips_fetch_instruction (pc) == 0xe820; /* jr $ra */ else return mips_fetch_instruction (pc) == 0x3e00008; /* jr $ra */ } /* This fencepost looks highly suspicious to me. Removing it also seems suspicious as it could affect remote debugging across serial lines. */ static CORE_ADDR heuristic_proc_start (CORE_ADDR pc) { CORE_ADDR start_pc; CORE_ADDR fence; int instlen; int seen_adjsp = 0; pc = gdbarch_addr_bits_remove (current_gdbarch, pc); start_pc = pc; fence = start_pc - heuristic_fence_post; if (start_pc == 0) return 0; if (heuristic_fence_post == UINT_MAX || fence < VM_MIN_ADDRESS) fence = VM_MIN_ADDRESS; instlen = mips_pc_is_mips16 (pc) ? MIPS_INSN16_SIZE : MIPS_INSN32_SIZE; /* search back for previous return */ for (start_pc -= instlen;; start_pc -= instlen) if (start_pc < fence) { /* It's not clear to me why we reach this point when stop_soon, but with this test, at least we don't print out warnings for every child forked (eg, on decstation). 22apr93 rich@cygnus.com. */ if (stop_soon == NO_STOP_QUIETLY) { static int blurb_printed = 0; warning (_("GDB can't find the start of the function at 0x%s."), paddr_nz (pc)); if (!blurb_printed) { /* This actually happens frequently in embedded development, when you first connect to a board and your stack pointer and pc are nowhere in particular. This message needs to give people in that situation enough information to determine that it's no big deal. */ printf_filtered ("\n\ GDB is unable to find the start of the function at 0x%s\n\ and thus can't determine the size of that function's stack frame.\n\ This means that GDB may be unable to access that stack frame, or\n\ the frames below it.\n\ This problem is most likely caused by an invalid program counter or\n\ stack pointer.\n\ However, if you think GDB should simply search farther back\n\ from 0x%s for code which looks like the beginning of a\n\ function, you can increase the range of the search using the `set\n\ heuristic-fence-post' command.\n", paddr_nz (pc), paddr_nz (pc)); blurb_printed = 1; } } return 0; } else if (mips_pc_is_mips16 (start_pc)) { unsigned short inst; /* On MIPS16, any one of the following is likely to be the start of a function: entry addiu sp,-n daddiu sp,-n extend -n followed by 'addiu sp,+n' or 'daddiu sp,+n' */ inst = mips_fetch_instruction (start_pc); if (((inst & 0xf81f) == 0xe809 && (inst & 0x700) != 0x700) /* entry */ || (inst & 0xff80) == 0x6380 /* addiu sp,-n */ || (inst & 0xff80) == 0xfb80 /* daddiu sp,-n */ || ((inst & 0xf810) == 0xf010 && seen_adjsp)) /* extend -n */ break; else if ((inst & 0xff00) == 0x6300 /* addiu sp */ || (inst & 0xff00) == 0xfb00) /* daddiu sp */ seen_adjsp = 1; else seen_adjsp = 0; } else if (mips_about_to_return (start_pc)) { /* Skip return and its delay slot. */ start_pc += 2 * MIPS_INSN32_SIZE; break; } return start_pc; } struct mips_objfile_private { bfd_size_type size; char *contents; }; /* According to the current ABI, should the type be passed in a floating-point register (assuming that there is space)? When there is no FPU, FP are not even considered as possible candidates for FP registers and, consequently this returns false - forces FP arguments into integer registers. */ static int fp_register_arg_p (enum type_code typecode, struct type *arg_type) { return ((typecode == TYPE_CODE_FLT || (MIPS_EABI && (typecode == TYPE_CODE_STRUCT || typecode == TYPE_CODE_UNION) && TYPE_NFIELDS (arg_type) == 1 && TYPE_CODE (check_typedef (TYPE_FIELD_TYPE (arg_type, 0))) == TYPE_CODE_FLT)) && MIPS_FPU_TYPE != MIPS_FPU_NONE); } /* On o32, argument passing in GPRs depends on the alignment of the type being passed. Return 1 if this type must be aligned to a doubleword boundary. */ static int mips_type_needs_double_align (struct type *type) { enum type_code typecode = TYPE_CODE (type); if (typecode == TYPE_CODE_FLT && TYPE_LENGTH (type) == 8) return 1; else if (typecode == TYPE_CODE_STRUCT) { if (TYPE_NFIELDS (type) < 1) return 0; return mips_type_needs_double_align (TYPE_FIELD_TYPE (type, 0)); } else if (typecode == TYPE_CODE_UNION) { int i, n; n = TYPE_NFIELDS (type); for (i = 0; i < n; i++) if (mips_type_needs_double_align (TYPE_FIELD_TYPE (type, i))) return 1; return 0; } return 0; } /* Adjust the address downward (direction of stack growth) so that it is correctly aligned for a new stack frame. */ static CORE_ADDR mips_frame_align (struct gdbarch *gdbarch, CORE_ADDR addr) { return align_down (addr, 16); } static CORE_ADDR mips_eabi_push_dummy_call (struct gdbarch *gdbarch, struct value *function, struct regcache *regcache, CORE_ADDR bp_addr, int nargs, struct value **args, CORE_ADDR sp, int struct_return, CORE_ADDR struct_addr) { int argreg; int float_argreg; int argnum; int len = 0; int stack_offset = 0; struct gdbarch_tdep *tdep = gdbarch_tdep (gdbarch); CORE_ADDR func_addr = find_function_addr (function, NULL); int regsize = mips_abi_regsize (gdbarch); /* For shared libraries, "t9" needs to point at the function address. */ regcache_cooked_write_signed (regcache, MIPS_T9_REGNUM, func_addr); /* Set the return address register to point to the entry point of the program, where a breakpoint lies in wait. */ regcache_cooked_write_signed (regcache, MIPS_RA_REGNUM, bp_addr); /* First ensure that the stack and structure return address (if any) are properly aligned. The stack has to be at least 64-bit aligned even on 32-bit machines, because doubles must be 64-bit aligned. For n32 and n64, stack frames need to be 128-bit aligned, so we round to this widest known alignment. */ sp = align_down (sp, 16); struct_addr = align_down (struct_addr, 16); /* Now make space on the stack for the args. We allocate more than necessary for EABI, because the first few arguments are passed in registers, but that's OK. */ for (argnum = 0; argnum < nargs; argnum++) len += align_up (TYPE_LENGTH (value_type (args[argnum])), regsize); sp -= align_up (len, 16); if (mips_debug) fprintf_unfiltered (gdb_stdlog, "mips_eabi_push_dummy_call: sp=0x%s allocated %ld\n", paddr_nz (sp), (long) align_up (len, 16)); /* Initialize the integer and float register pointers. */ argreg = MIPS_A0_REGNUM; float_argreg = mips_fpa0_regnum (current_gdbarch); /* The struct_return pointer occupies the first parameter-passing reg. */ if (struct_return) { if (mips_debug) fprintf_unfiltered (gdb_stdlog, "mips_eabi_push_dummy_call: struct_return reg=%d 0x%s\n", argreg, paddr_nz (struct_addr)); regcache_cooked_write_unsigned (regcache, argreg++, struct_addr); } /* Now load as many as possible of the first arguments into registers, and push the rest onto the stack. Loop thru args from first to last. */ for (argnum = 0; argnum < nargs; argnum++) { const gdb_byte *val; gdb_byte valbuf[MAX_REGISTER_SIZE]; struct value *arg = args[argnum]; struct type *arg_type = check_typedef (value_type (arg)); int len = TYPE_LENGTH (arg_type); enum type_code typecode = TYPE_CODE (arg_type); if (mips_debug) fprintf_unfiltered (gdb_stdlog, "mips_eabi_push_dummy_call: %d len=%d type=%d", argnum + 1, len, (int) typecode); /* The EABI passes structures that do not fit in a register by reference. */ if (len > regsize && (typecode == TYPE_CODE_STRUCT || typecode == TYPE_CODE_UNION)) { store_unsigned_integer (valbuf, regsize, VALUE_ADDRESS (arg)); typecode = TYPE_CODE_PTR; len = regsize; val = valbuf; if (mips_debug) fprintf_unfiltered (gdb_stdlog, " push"); } else val = value_contents (arg); /* 32-bit ABIs always start floating point arguments in an even-numbered floating point register. Round the FP register up before the check to see if there are any FP registers left. Non MIPS_EABI targets also pass the FP in the integer registers so also round up normal registers. */ if (regsize < 8 && fp_register_arg_p (typecode, arg_type)) { if ((float_argreg & 1)) float_argreg++; } /* Floating point arguments passed in registers have to be treated specially. On 32-bit architectures, doubles are passed in register pairs; the even register gets the low word, and the odd register gets the high word. On non-EABI processors, the first two floating point arguments are also copied to general registers, because MIPS16 functions don't use float registers for arguments. This duplication of arguments in general registers can't hurt non-MIPS16 functions because those registers are normally skipped. */ /* MIPS_EABI squeezes a struct that contains a single floating point value into an FP register instead of pushing it onto the stack. */ if (fp_register_arg_p (typecode, arg_type) && float_argreg <= MIPS_LAST_FP_ARG_REGNUM) { /* EABI32 will pass doubles in consecutive registers, even on 64-bit cores. At one time, we used to check the size of `float_argreg' to determine whether or not to pass doubles in consecutive registers, but this is not sufficient for making the ABI determination. */ if (len == 8 && mips_abi (gdbarch) == MIPS_ABI_EABI32) { int low_offset = gdbarch_byte_order (current_gdbarch) == BFD_ENDIAN_BIG ? 4 : 0; unsigned long regval; /* Write the low word of the double to the even register(s). */ regval = extract_unsigned_integer (val + low_offset, 4); if (mips_debug) fprintf_unfiltered (gdb_stdlog, " - fpreg=%d val=%s", float_argreg, phex (regval, 4)); regcache_cooked_write_unsigned (regcache, float_argreg++, regval); /* Write the high word of the double to the odd register(s). */ regval = extract_unsigned_integer (val + 4 - low_offset, 4); if (mips_debug) fprintf_unfiltered (gdb_stdlog, " - fpreg=%d val=%s", float_argreg, phex (regval, 4)); regcache_cooked_write_unsigned (regcache, float_argreg++, regval); } else { /* This is a floating point value that fits entirely in a single register. */ /* On 32 bit ABI's the float_argreg is further adjusted above to ensure that it is even register aligned. */ LONGEST regval = extract_unsigned_integer (val, len); if (mips_debug) fprintf_unfiltered (gdb_stdlog, " - fpreg=%d val=%s", float_argreg, phex (regval, len)); regcache_cooked_write_unsigned (regcache, float_argreg++, regval); } } else { /* Copy the argument to general registers or the stack in register-sized pieces. Large arguments are split between registers and stack. */ /* Note: structs whose size is not a multiple of regsize are treated specially: Irix cc passes them in registers where gcc sometimes puts them on the stack. For maximum compatibility, we will put them in both places. */ int odd_sized_struct = (len > regsize && len % regsize != 0); /* Note: Floating-point values that didn't fit into an FP register are only written to memory. */ while (len > 0) { /* Remember if the argument was written to the stack. */ int stack_used_p = 0; int partial_len = (len < regsize ? len : regsize); if (mips_debug) fprintf_unfiltered (gdb_stdlog, " -- partial=%d", partial_len); /* Write this portion of the argument to the stack. */ if (argreg > MIPS_LAST_ARG_REGNUM || odd_sized_struct || fp_register_arg_p (typecode, arg_type)) { /* Should shorter than int integer values be promoted to int before being stored? */ int longword_offset = 0; CORE_ADDR addr; stack_used_p = 1; if (gdbarch_byte_order (current_gdbarch) == BFD_ENDIAN_BIG) { if (regsize == 8 && (typecode == TYPE_CODE_INT || typecode == TYPE_CODE_PTR || typecode == TYPE_CODE_FLT) && len <= 4) longword_offset = regsize - len; else if ((typecode == TYPE_CODE_STRUCT || typecode == TYPE_CODE_UNION) && TYPE_LENGTH (arg_type) < regsize) longword_offset = regsize - len; } if (mips_debug) { fprintf_unfiltered (gdb_stdlog, " - stack_offset=0x%s", paddr_nz (stack_offset)); fprintf_unfiltered (gdb_stdlog, " longword_offset=0x%s", paddr_nz (longword_offset)); } addr = sp + stack_offset + longword_offset; if (mips_debug) { int i; fprintf_unfiltered (gdb_stdlog, " @0x%s ", paddr_nz (addr)); for (i = 0; i < partial_len; i++) { fprintf_unfiltered (gdb_stdlog, "%02x", val[i] & 0xff); } } write_memory (addr, val, partial_len); } /* Note!!! This is NOT an else clause. Odd sized structs may go thru BOTH paths. Floating point arguments will not. */ /* Write this portion of the argument to a general purpose register. */ if (argreg <= MIPS_LAST_ARG_REGNUM && !fp_register_arg_p (typecode, arg_type)) { LONGEST regval = extract_unsigned_integer (val, partial_len); if (mips_debug) fprintf_filtered (gdb_stdlog, " - reg=%d val=%s", argreg, phex (regval, regsize)); regcache_cooked_write_unsigned (regcache, argreg, regval); argreg++; } len -= partial_len; val += partial_len; /* Compute the the offset into the stack at which we will copy the next parameter. In the new EABI (and the NABI32), the stack_offset only needs to be adjusted when it has been used. */ if (stack_used_p) stack_offset += align_up (partial_len, regsize); } } if (mips_debug) fprintf_unfiltered (gdb_stdlog, "\n"); } regcache_cooked_write_signed (regcache, MIPS_SP_REGNUM, sp); /* Return adjusted stack pointer. */ return sp; } /* Determine the return value convention being used. */ static enum return_value_convention mips_eabi_return_value (struct gdbarch *gdbarch, struct type *type, struct regcache *regcache, gdb_byte *readbuf, const gdb_byte *writebuf) { if (TYPE_LENGTH (type) > 2 * mips_abi_regsize (gdbarch)) return RETURN_VALUE_STRUCT_CONVENTION; if (readbuf) memset (readbuf, 0, TYPE_LENGTH (type)); return RETURN_VALUE_REGISTER_CONVENTION; } /* N32/N64 ABI stuff. */ static CORE_ADDR mips_n32n64_push_dummy_call (struct gdbarch *gdbarch, struct value *function, struct regcache *regcache, CORE_ADDR bp_addr, int nargs, struct value **args, CORE_ADDR sp, int struct_return, CORE_ADDR struct_addr) { int argreg; int float_argreg; int argnum; int len = 0; int stack_offset = 0; struct gdbarch_tdep *tdep = gdbarch_tdep (gdbarch); CORE_ADDR func_addr = find_function_addr (function, NULL); /* For shared libraries, "t9" needs to point at the function address. */ regcache_cooked_write_signed (regcache, MIPS_T9_REGNUM, func_addr); /* Set the return address register to point to the entry point of the program, where a breakpoint lies in wait. */ regcache_cooked_write_signed (regcache, MIPS_RA_REGNUM, bp_addr); /* First ensure that the stack and structure return address (if any) are properly aligned. The stack has to be at least 64-bit aligned even on 32-bit machines, because doubles must be 64-bit aligned. For n32 and n64, stack frames need to be 128-bit aligned, so we round to this widest known alignment. */ sp = align_down (sp, 16); struct_addr = align_down (struct_addr, 16); /* Now make space on the stack for the args. */ for (argnum = 0; argnum < nargs; argnum++) len += align_up (TYPE_LENGTH (value_type (args[argnum])), MIPS64_REGSIZE); sp -= align_up (len, 16); if (mips_debug) fprintf_unfiltered (gdb_stdlog, "mips_n32n64_push_dummy_call: sp=0x%s allocated %ld\n", paddr_nz (sp), (long) align_up (len, 16)); /* Initialize the integer and float register pointers. */ argreg = MIPS_A0_REGNUM; float_argreg = mips_fpa0_regnum (current_gdbarch); /* The struct_return pointer occupies the first parameter-passing reg. */ if (struct_return) { if (mips_debug) fprintf_unfiltered (gdb_stdlog, "mips_n32n64_push_dummy_call: struct_return reg=%d 0x%s\n", argreg, paddr_nz (struct_addr)); regcache_cooked_write_unsigned (regcache, argreg++, struct_addr); } /* Now load as many as possible of the first arguments into registers, and push the rest onto the stack. Loop thru args from first to last. */ for (argnum = 0; argnum < nargs; argnum++) { const gdb_byte *val; struct value *arg = args[argnum]; struct type *arg_type = check_typedef (value_type (arg)); int len = TYPE_LENGTH (arg_type); enum type_code typecode = TYPE_CODE (arg_type); if (mips_debug) fprintf_unfiltered (gdb_stdlog, "mips_n32n64_push_dummy_call: %d len=%d type=%d", argnum + 1, len, (int) typecode); val = value_contents (arg); if (fp_register_arg_p (typecode, arg_type) && float_argreg <= MIPS_LAST_FP_ARG_REGNUM) { /* This is a floating point value that fits entirely in a single register. */ /* On 32 bit ABI's the float_argreg is further adjusted above to ensure that it is even register aligned. */ LONGEST regval = extract_unsigned_integer (val, len); if (mips_debug) fprintf_unfiltered (gdb_stdlog, " - fpreg=%d val=%s", float_argreg, phex (regval, len)); regcache_cooked_write_unsigned (regcache, float_argreg++, regval); if (mips_debug) fprintf_unfiltered (gdb_stdlog, " - reg=%d val=%s", argreg, phex (regval, len)); regcache_cooked_write_unsigned (regcache, argreg, regval); argreg += 1; } else { /* Copy the argument to general registers or the stack in register-sized pieces. Large arguments are split between registers and stack. */ /* Note: structs whose size is not a multiple of MIPS64_REGSIZE are treated specially: Irix cc passes them in registers where gcc sometimes puts them on the stack. For maximum compatibility, we will put them in both places. */ int odd_sized_struct = (len > MIPS64_REGSIZE && len % MIPS64_REGSIZE != 0); /* Note: Floating-point values that didn't fit into an FP register are only written to memory. */ while (len > 0) { /* Remember if the argument was written to the stack. */ int stack_used_p = 0; int partial_len = (len < MIPS64_REGSIZE ? len : MIPS64_REGSIZE); if (mips_debug) fprintf_unfiltered (gdb_stdlog, " -- partial=%d", partial_len); /* Write this portion of the argument to the stack. */ if (argreg > MIPS_LAST_ARG_REGNUM || odd_sized_struct || fp_register_arg_p (typecode, arg_type)) { /* Should shorter than int integer values be promoted to int before being stored? */ int longword_offset = 0; CORE_ADDR addr; stack_used_p = 1; if (gdbarch_byte_order (current_gdbarch) == BFD_ENDIAN_BIG) { if ((typecode == TYPE_CODE_INT || typecode == TYPE_CODE_PTR || typecode == TYPE_CODE_FLT) && len <= 4) longword_offset = MIPS64_REGSIZE - len; } if (mips_debug) { fprintf_unfiltered (gdb_stdlog, " - stack_offset=0x%s", paddr_nz (stack_offset)); fprintf_unfiltered (gdb_stdlog, " longword_offset=0x%s", paddr_nz (longword_offset)); } addr = sp + stack_offset + longword_offset; if (mips_debug) { int i; fprintf_unfiltered (gdb_stdlog, " @0x%s ", paddr_nz (addr)); for (i = 0; i < partial_len; i++) { fprintf_unfiltered (gdb_stdlog, "%02x", val[i] & 0xff); } } write_memory (addr, val, partial_len); } /* Note!!! This is NOT an else clause. Odd sized structs may go thru BOTH paths. Floating point arguments will not. */ /* Write this portion of the argument to a general purpose register. */ if (argreg <= MIPS_LAST_ARG_REGNUM && !fp_register_arg_p (typecode, arg_type)) { LONGEST regval = extract_unsigned_integer (val, partial_len); /* A non-floating-point argument being passed in a general register. If a struct or union, and if the remaining length is smaller than the register size, we have to adjust the register value on big endian targets. It does not seem to be necessary to do the same for integral types. */ if (gdbarch_byte_order (current_gdbarch) == BFD_ENDIAN_BIG && partial_len < MIPS64_REGSIZE && (typecode == TYPE_CODE_STRUCT || typecode == TYPE_CODE_UNION)) regval <<= ((MIPS64_REGSIZE - partial_len) * TARGET_CHAR_BIT); if (mips_debug) fprintf_filtered (gdb_stdlog, " - reg=%d val=%s", argreg, phex (regval, MIPS64_REGSIZE)); regcache_cooked_write_unsigned (regcache, argreg, regval); argreg++; } len -= partial_len; val += partial_len; /* Compute the the offset into the stack at which we will copy the next parameter. In N32 (N64?), the stack_offset only needs to be adjusted when it has been used. */ if (stack_used_p) stack_offset += align_up (partial_len, MIPS64_REGSIZE); } } if (mips_debug) fprintf_unfiltered (gdb_stdlog, "\n"); } regcache_cooked_write_signed (regcache, MIPS_SP_REGNUM, sp); /* Return adjusted stack pointer. */ return sp; } static enum return_value_convention mips_n32n64_return_value (struct gdbarch *gdbarch, struct type *type, struct regcache *regcache, gdb_byte *readbuf, const gdb_byte *writebuf) { struct gdbarch_tdep *tdep = gdbarch_tdep (current_gdbarch); if (TYPE_CODE (type) == TYPE_CODE_STRUCT || TYPE_CODE (type) == TYPE_CODE_UNION || TYPE_CODE (type) == TYPE_CODE_ARRAY || TYPE_LENGTH (type) > 2 * MIPS64_REGSIZE) return RETURN_VALUE_STRUCT_CONVENTION; else if (TYPE_CODE (type) == TYPE_CODE_FLT && TYPE_LENGTH (type) == 16 && tdep->mips_fpu_type != MIPS_FPU_NONE) { /* A 128-bit floating-point value fills both $f0 and $f2. The two registers are used in the same as memory order, so the eight bytes with the lower memory address are in $f0. */ if (mips_debug) fprintf_unfiltered (gdb_stderr, "Return float in $f0 and $f2\n"); mips_xfer_register (regcache, gdbarch_num_regs (current_gdbarch) + mips_regnum (current_gdbarch)->fp0, 8, gdbarch_byte_order (current_gdbarch), readbuf, writebuf, 0); mips_xfer_register (regcache, gdbarch_num_regs (current_gdbarch) + mips_regnum (current_gdbarch)->fp0 + 2, 8, gdbarch_byte_order (current_gdbarch), readbuf ? readbuf + 8 : readbuf, writebuf ? writebuf + 8 : writebuf, 0); return RETURN_VALUE_REGISTER_CONVENTION; } else if (TYPE_CODE (type) == TYPE_CODE_FLT && tdep->mips_fpu_type != MIPS_FPU_NONE) { /* A floating-point value belongs in the least significant part of FP0. */ if (mips_debug) fprintf_unfiltered (gdb_stderr, "Return float in $fp0\n"); mips_xfer_register (regcache, gdbarch_num_regs (current_gdbarch) + mips_regnum (current_gdbarch)->fp0, TYPE_LENGTH (type), gdbarch_byte_order (current_gdbarch), readbuf, writebuf, 0); return RETURN_VALUE_REGISTER_CONVENTION; } else if (TYPE_CODE (type) == TYPE_CODE_STRUCT && TYPE_NFIELDS (type) <= 2 && TYPE_NFIELDS (type) >= 1 && ((TYPE_NFIELDS (type) == 1 && (TYPE_CODE (TYPE_FIELD_TYPE (type, 0)) == TYPE_CODE_FLT)) || (TYPE_NFIELDS (type) == 2 && (TYPE_CODE (TYPE_FIELD_TYPE (type, 0)) == TYPE_CODE_FLT) && (TYPE_CODE (TYPE_FIELD_TYPE (type, 1)) == TYPE_CODE_FLT))) && tdep->mips_fpu_type != MIPS_FPU_NONE) { /* A struct that contains one or two floats. Each value is part in the least significant part of their floating point register.. */ int regnum; int field; for (field = 0, regnum = mips_regnum (current_gdbarch)->fp0; field < TYPE_NFIELDS (type); field++, regnum += 2) { int offset = (FIELD_BITPOS (TYPE_FIELDS (type)[field]) / TARGET_CHAR_BIT); if (mips_debug) fprintf_unfiltered (gdb_stderr, "Return float struct+%d\n", offset); mips_xfer_register (regcache, gdbarch_num_regs (current_gdbarch) + regnum, TYPE_LENGTH (TYPE_FIELD_TYPE (type, field)), gdbarch_byte_order (current_gdbarch), readbuf, writebuf, offset); } return RETURN_VALUE_REGISTER_CONVENTION; } else if (TYPE_CODE (type) == TYPE_CODE_STRUCT || TYPE_CODE (type) == TYPE_CODE_UNION) { /* A structure or union. Extract the left justified value, regardless of the byte order. I.e. DO NOT USE mips_xfer_lower. */ int offset; int regnum; for (offset = 0, regnum = MIPS_V0_REGNUM; offset < TYPE_LENGTH (type); offset += register_size (current_gdbarch, regnum), regnum++) { int xfer = register_size (current_gdbarch, regnum); if (offset + xfer > TYPE_LENGTH (type)) xfer = TYPE_LENGTH (type) - offset; if (mips_debug) fprintf_unfiltered (gdb_stderr, "Return struct+%d:%d in $%d\n", offset, xfer, regnum); mips_xfer_register (regcache, gdbarch_num_regs (current_gdbarch) + regnum, xfer, BFD_ENDIAN_UNKNOWN, readbuf, writebuf, offset); } return RETURN_VALUE_REGISTER_CONVENTION; } else { /* A scalar extract each part but least-significant-byte justified. */ int offset; int regnum; for (offset = 0, regnum = MIPS_V0_REGNUM; offset < TYPE_LENGTH (type); offset += register_size (current_gdbarch, regnum), regnum++) { int xfer = register_size (current_gdbarch, regnum); if (offset + xfer > TYPE_LENGTH (type)) xfer = TYPE_LENGTH (type) - offset; if (mips_debug) fprintf_unfiltered (gdb_stderr, "Return scalar+%d:%d in $%d\n", offset, xfer, regnum); mips_xfer_register (regcache, gdbarch_num_regs (current_gdbarch) + regnum, xfer, gdbarch_byte_order (current_gdbarch), readbuf, writebuf, offset); } return RETURN_VALUE_REGISTER_CONVENTION; } } /* O32 ABI stuff. */ static CORE_ADDR mips_o32_push_dummy_call (struct gdbarch *gdbarch, struct value *function, struct regcache *regcache, CORE_ADDR bp_addr, int nargs, struct value **args, CORE_ADDR sp, int struct_return, CORE_ADDR struct_addr) { int argreg; int float_argreg; int argnum; int len = 0; int stack_offset = 0; struct gdbarch_tdep *tdep = gdbarch_tdep (gdbarch); CORE_ADDR func_addr = find_function_addr (function, NULL); /* For shared libraries, "t9" needs to point at the function address. */ regcache_cooked_write_signed (regcache, MIPS_T9_REGNUM, func_addr); /* Set the return address register to point to the entry point of the program, where a breakpoint lies in wait. */ regcache_cooked_write_signed (regcache, MIPS_RA_REGNUM, bp_addr); /* First ensure that the stack and structure return address (if any) are properly aligned. The stack has to be at least 64-bit aligned even on 32-bit machines, because doubles must be 64-bit aligned. For n32 and n64, stack frames need to be 128-bit aligned, so we round to this widest known alignment. */ sp = align_down (sp, 16); struct_addr = align_down (struct_addr, 16); /* Now make space on the stack for the args. */ for (argnum = 0; argnum < nargs; argnum++) { struct type *arg_type = check_typedef (value_type (args[argnum])); int arglen = TYPE_LENGTH (arg_type); /* Align to double-word if necessary. */ if (mips_type_needs_double_align (arg_type)) len = align_up (len, MIPS32_REGSIZE * 2); /* Allocate space on the stack. */ len += align_up (arglen, MIPS32_REGSIZE); } sp -= align_up (len, 16); if (mips_debug) fprintf_unfiltered (gdb_stdlog, "mips_o32_push_dummy_call: sp=0x%s allocated %ld\n", paddr_nz (sp), (long) align_up (len, 16)); /* Initialize the integer and float register pointers. */ argreg = MIPS_A0_REGNUM; float_argreg = mips_fpa0_regnum (current_gdbarch); /* The struct_return pointer occupies the first parameter-passing reg. */ if (struct_return) { if (mips_debug) fprintf_unfiltered (gdb_stdlog, "mips_o32_push_dummy_call: struct_return reg=%d 0x%s\n", argreg, paddr_nz (struct_addr)); regcache_cooked_write_unsigned (regcache, argreg++, struct_addr); stack_offset += MIPS32_REGSIZE; } /* Now load as many as possible of the first arguments into registers, and push the rest onto the stack. Loop thru args from first to last. */ for (argnum = 0; argnum < nargs; argnum++) { const gdb_byte *val; struct value *arg = args[argnum]; struct type *arg_type = check_typedef (value_type (arg)); int len = TYPE_LENGTH (arg_type); enum type_code typecode = TYPE_CODE (arg_type); if (mips_debug) fprintf_unfiltered (gdb_stdlog, "mips_o32_push_dummy_call: %d len=%d type=%d", argnum + 1, len, (int) typecode); val = value_contents (arg); /* 32-bit ABIs always start floating point arguments in an even-numbered floating point register. Round the FP register up before the check to see if there are any FP registers left. O32/O64 targets also pass the FP in the integer registers so also round up normal registers. */ if (fp_register_arg_p (typecode, arg_type)) { if ((float_argreg & 1)) float_argreg++; } /* Floating point arguments passed in registers have to be treated specially. On 32-bit architectures, doubles are passed in register pairs; the even register gets the low word, and the odd register gets the high word. On O32/O64, the first two floating point arguments are also copied to general registers, because MIPS16 functions don't use float registers for arguments. This duplication of arguments in general registers can't hurt non-MIPS16 functions because those registers are normally skipped. */ if (fp_register_arg_p (typecode, arg_type) && float_argreg <= MIPS_LAST_FP_ARG_REGNUM) { if (register_size (gdbarch, float_argreg) < 8 && len == 8) { int low_offset = gdbarch_byte_order (current_gdbarch) == BFD_ENDIAN_BIG ? 4 : 0; unsigned long regval; /* Write the low word of the double to the even register(s). */ regval = extract_unsigned_integer (val + low_offset, 4); if (mips_debug) fprintf_unfiltered (gdb_stdlog, " - fpreg=%d val=%s", float_argreg, phex (regval, 4)); regcache_cooked_write_unsigned (regcache, float_argreg++, regval); if (mips_debug) fprintf_unfiltered (gdb_stdlog, " - reg=%d val=%s", argreg, phex (regval, 4)); regcache_cooked_write_unsigned (regcache, argreg++, regval); /* Write the high word of the double to the odd register(s). */ regval = extract_unsigned_integer (val + 4 - low_offset, 4); if (mips_debug) fprintf_unfiltered (gdb_stdlog, " - fpreg=%d val=%s", float_argreg, phex (regval, 4)); regcache_cooked_write_unsigned (regcache, float_argreg++, regval); if (mips_debug) fprintf_unfiltered (gdb_stdlog, " - reg=%d val=%s", argreg, phex (regval, 4)); regcache_cooked_write_unsigned (regcache, argreg++, regval); } else { /* This is a floating point value that fits entirely in a single register. */ /* On 32 bit ABI's the float_argreg is further adjusted above to ensure that it is even register aligned. */ LONGEST regval = extract_unsigned_integer (val, len); if (mips_debug) fprintf_unfiltered (gdb_stdlog, " - fpreg=%d val=%s", float_argreg, phex (regval, len)); regcache_cooked_write_unsigned (regcache, float_argreg++, regval); /* CAGNEY: 32 bit MIPS ABI's always reserve two FP registers for each argument. The below is (my guess) to ensure that the corresponding integer register has reserved the same space. */ if (mips_debug) fprintf_unfiltered (gdb_stdlog, " - reg=%d val=%s", argreg, phex (regval, len)); regcache_cooked_write_unsigned (regcache, argreg, regval); argreg += 2; } /* Reserve space for the FP register. */ stack_offset += align_up (len, MIPS32_REGSIZE); } else { /* Copy the argument to general registers or the stack in register-sized pieces. Large arguments are split between registers and stack. */ /* Note: structs whose size is not a multiple of MIPS32_REGSIZE are treated specially: Irix cc passes them in registers where gcc sometimes puts them on the stack. For maximum compatibility, we will put them in both places. */ int odd_sized_struct = (len > MIPS32_REGSIZE && len % MIPS32_REGSIZE != 0); /* Structures should be aligned to eight bytes (even arg registers) on MIPS_ABI_O32, if their first member has double precision. */ if (mips_type_needs_double_align (arg_type)) { if ((argreg & 1)) { argreg++; stack_offset += MIPS32_REGSIZE; } } while (len > 0) { /* Remember if the argument was written to the stack. */ int stack_used_p = 0; int partial_len = (len < MIPS32_REGSIZE ? len : MIPS32_REGSIZE); if (mips_debug) fprintf_unfiltered (gdb_stdlog, " -- partial=%d", partial_len); /* Write this portion of the argument to the stack. */ if (argreg > MIPS_LAST_ARG_REGNUM || odd_sized_struct) { /* Should shorter than int integer values be promoted to int before being stored? */ int longword_offset = 0; CORE_ADDR addr; stack_used_p = 1; if (mips_debug) { fprintf_unfiltered (gdb_stdlog, " - stack_offset=0x%s", paddr_nz (stack_offset)); fprintf_unfiltered (gdb_stdlog, " longword_offset=0x%s", paddr_nz (longword_offset)); } addr = sp + stack_offset + longword_offset; if (mips_debug) { int i; fprintf_unfiltered (gdb_stdlog, " @0x%s ", paddr_nz (addr)); for (i = 0; i < partial_len; i++) { fprintf_unfiltered (gdb_stdlog, "%02x", val[i] & 0xff); } } write_memory (addr, val, partial_len); } /* Note!!! This is NOT an else clause. Odd sized structs may go thru BOTH paths. */ /* Write this portion of the argument to a general purpose register. */ if (argreg <= MIPS_LAST_ARG_REGNUM) { LONGEST regval = extract_signed_integer (val, partial_len); /* Value may need to be sign extended, because mips_isa_regsize() != mips_abi_regsize(). */ /* A non-floating-point argument being passed in a general register. If a struct or union, and if the remaining length is smaller than the register size, we have to adjust the register value on big endian targets. It does not seem to be necessary to do the same for integral types. Also don't do this adjustment on O64 binaries. cagney/2001-07-23: gdb/179: Also, GCC, when outputting LE O32 with sizeof (struct) < mips_abi_regsize(), generates a left shift as part of storing the argument in a register (the left shift isn't generated when sizeof (struct) >= mips_abi_regsize()). Since it is quite possible that this is GCC contradicting the LE/O32 ABI, GDB has not been adjusted to accommodate this. Either someone needs to demonstrate that the LE/O32 ABI specifies such a left shift OR this new ABI gets identified as such and GDB gets tweaked accordingly. */ if (gdbarch_byte_order (current_gdbarch) == BFD_ENDIAN_BIG && partial_len < MIPS32_REGSIZE && (typecode == TYPE_CODE_STRUCT || typecode == TYPE_CODE_UNION)) regval <<= ((MIPS32_REGSIZE - partial_len) * TARGET_CHAR_BIT); if (mips_debug) fprintf_filtered (gdb_stdlog, " - reg=%d val=%s", argreg, phex (regval, MIPS32_REGSIZE)); regcache_cooked_write_unsigned (regcache, argreg, regval); argreg++; /* Prevent subsequent floating point arguments from being passed in floating point registers. */ float_argreg = MIPS_LAST_FP_ARG_REGNUM + 1; } len -= partial_len; val += partial_len; /* Compute the the offset into the stack at which we will copy the next parameter. In older ABIs, the caller reserved space for registers that contained arguments. This was loosely refered to as their "home". Consequently, space is always allocated. */ stack_offset += align_up (partial_len, MIPS32_REGSIZE); } } if (mips_debug) fprintf_unfiltered (gdb_stdlog, "\n"); } regcache_cooked_write_signed (regcache, MIPS_SP_REGNUM, sp); /* Return adjusted stack pointer. */ return sp; } static enum return_value_convention mips_o32_return_value (struct gdbarch *gdbarch, struct type *type, struct regcache *regcache, gdb_byte *readbuf, const gdb_byte *writebuf) { struct gdbarch_tdep *tdep = gdbarch_tdep (current_gdbarch); if (TYPE_CODE (type) == TYPE_CODE_STRUCT || TYPE_CODE (type) == TYPE_CODE_UNION || TYPE_CODE (type) == TYPE_CODE_ARRAY) return RETURN_VALUE_STRUCT_CONVENTION; else if (TYPE_CODE (type) == TYPE_CODE_FLT && TYPE_LENGTH (type) == 4 && tdep->mips_fpu_type != MIPS_FPU_NONE) { /* A single-precision floating-point value. It fits in the least significant part of FP0. */ if (mips_debug) fprintf_unfiltered (gdb_stderr, "Return float in $fp0\n"); mips_xfer_register (regcache, gdbarch_num_regs (current_gdbarch) + mips_regnum (current_gdbarch)->fp0, TYPE_LENGTH (type), gdbarch_byte_order (current_gdbarch), readbuf, writebuf, 0); return RETURN_VALUE_REGISTER_CONVENTION; } else if (TYPE_CODE (type) == TYPE_CODE_FLT && TYPE_LENGTH (type) == 8 && tdep->mips_fpu_type != MIPS_FPU_NONE) { /* A double-precision floating-point value. The most significant part goes in FP1, and the least significant in FP0. */ if (mips_debug) fprintf_unfiltered (gdb_stderr, "Return float in $fp1/$fp0\n"); switch (gdbarch_byte_order (current_gdbarch)) { case BFD_ENDIAN_LITTLE: mips_xfer_register (regcache, gdbarch_num_regs (current_gdbarch) + mips_regnum (current_gdbarch)->fp0 + 0, 4, gdbarch_byte_order (current_gdbarch), readbuf, writebuf, 0); mips_xfer_register (regcache, gdbarch_num_regs (current_gdbarch) + mips_regnum (current_gdbarch)->fp0 + 1, 4, gdbarch_byte_order (current_gdbarch), readbuf, writebuf, 4); break; case BFD_ENDIAN_BIG: mips_xfer_register (regcache, gdbarch_num_regs (current_gdbarch) + mips_regnum (current_gdbarch)->fp0 + 1, 4, gdbarch_byte_order (current_gdbarch), readbuf, writebuf, 0); mips_xfer_register (regcache, gdbarch_num_regs (current_gdbarch) + mips_regnum (current_gdbarch)->fp0 + 0, 4, gdbarch_byte_order (current_gdbarch), readbuf, writebuf, 4); break; default: internal_error (__FILE__, __LINE__, _("bad switch")); } return RETURN_VALUE_REGISTER_CONVENTION; } #if 0 else if (TYPE_CODE (type) == TYPE_CODE_STRUCT && TYPE_NFIELDS (type) <= 2 && TYPE_NFIELDS (type) >= 1 && ((TYPE_NFIELDS (type) == 1 && (TYPE_CODE (TYPE_FIELD_TYPE (type, 0)) == TYPE_CODE_FLT)) || (TYPE_NFIELDS (type) == 2 && (TYPE_CODE (TYPE_FIELD_TYPE (type, 0)) == TYPE_CODE_FLT) && (TYPE_CODE (TYPE_FIELD_TYPE (type, 1)) == TYPE_CODE_FLT))) && tdep->mips_fpu_type != MIPS_FPU_NONE) { /* A struct that contains one or two floats. Each value is part in the least significant part of their floating point register.. */ gdb_byte reg[MAX_REGISTER_SIZE]; int regnum; int field; for (field = 0, regnum = mips_regnum (current_gdbarch)->fp0; field < TYPE_NFIELDS (type); field++, regnum += 2) { int offset = (FIELD_BITPOS (TYPE_FIELDS (type)[field]) / TARGET_CHAR_BIT); if (mips_debug) fprintf_unfiltered (gdb_stderr, "Return float struct+%d\n", offset); mips_xfer_register (regcache, gdbarch_num_regs (current_gdbarch) + regnum, TYPE_LENGTH (TYPE_FIELD_TYPE (type, field)), gdbarch_byte_order (current_gdbarch), readbuf, writebuf, offset); } return RETURN_VALUE_REGISTER_CONVENTION; } #endif #if 0 else if (TYPE_CODE (type) == TYPE_CODE_STRUCT || TYPE_CODE (type) == TYPE_CODE_UNION) { /* A structure or union. Extract the left justified value, regardless of the byte order. I.e. DO NOT USE mips_xfer_lower. */ int offset; int regnum; for (offset = 0, regnum = MIPS_V0_REGNUM; offset < TYPE_LENGTH (type); offset += register_size (current_gdbarch, regnum), regnum++) { int xfer = register_size (current_gdbarch, regnum); if (offset + xfer > TYPE_LENGTH (type)) xfer = TYPE_LENGTH (type) - offset; if (mips_debug) fprintf_unfiltered (gdb_stderr, "Return struct+%d:%d in $%d\n", offset, xfer, regnum); mips_xfer_register (regcache, gdbarch_num_regs (current_gdbarch) + regnum, xfer, BFD_ENDIAN_UNKNOWN, readbuf, writebuf, offset); } return RETURN_VALUE_REGISTER_CONVENTION; } #endif else { /* A scalar extract each part but least-significant-byte justified. o32 thinks registers are 4 byte, regardless of the ISA. */ int offset; int regnum; for (offset = 0, regnum = MIPS_V0_REGNUM; offset < TYPE_LENGTH (type); offset += MIPS32_REGSIZE, regnum++) { int xfer = MIPS32_REGSIZE; if (offset + xfer > TYPE_LENGTH (type)) xfer = TYPE_LENGTH (type) - offset; if (mips_debug) fprintf_unfiltered (gdb_stderr, "Return scalar+%d:%d in $%d\n", offset, xfer, regnum); mips_xfer_register (regcache, gdbarch_num_regs (current_gdbarch) + regnum, xfer, gdbarch_byte_order (current_gdbarch), readbuf, writebuf, offset); } return RETURN_VALUE_REGISTER_CONVENTION; } } /* O64 ABI. This is a hacked up kind of 64-bit version of the o32 ABI. */ static CORE_ADDR mips_o64_push_dummy_call (struct gdbarch *gdbarch, struct value *function, struct regcache *regcache, CORE_ADDR bp_addr, int nargs, struct value **args, CORE_ADDR sp, int struct_return, CORE_ADDR struct_addr) { int argreg; int float_argreg; int argnum; int len = 0; int stack_offset = 0; struct gdbarch_tdep *tdep = gdbarch_tdep (gdbarch); CORE_ADDR func_addr = find_function_addr (function, NULL); /* For shared libraries, "t9" needs to point at the function address. */ regcache_cooked_write_signed (regcache, MIPS_T9_REGNUM, func_addr); /* Set the return address register to point to the entry point of the program, where a breakpoint lies in wait. */ regcache_cooked_write_signed (regcache, MIPS_RA_REGNUM, bp_addr); /* First ensure that the stack and structure return address (if any) are properly aligned. The stack has to be at least 64-bit aligned even on 32-bit machines, because doubles must be 64-bit aligned. For n32 and n64, stack frames need to be 128-bit aligned, so we round to this widest known alignment. */ sp = align_down (sp, 16); struct_addr = align_down (struct_addr, 16); /* Now make space on the stack for the args. */ for (argnum = 0; argnum < nargs; argnum++) { struct type *arg_type = check_typedef (value_type (args[argnum])); int arglen = TYPE_LENGTH (arg_type); /* Allocate space on the stack. */ len += align_up (arglen, MIPS64_REGSIZE); } sp -= align_up (len, 16); if (mips_debug) fprintf_unfiltered (gdb_stdlog, "mips_o64_push_dummy_call: sp=0x%s allocated %ld\n", paddr_nz (sp), (long) align_up (len, 16)); /* Initialize the integer and float register pointers. */ argreg = MIPS_A0_REGNUM; float_argreg = mips_fpa0_regnum (current_gdbarch); /* The struct_return pointer occupies the first parameter-passing reg. */ if (struct_return) { if (mips_debug) fprintf_unfiltered (gdb_stdlog, "mips_o64_push_dummy_call: struct_return reg=%d 0x%s\n", argreg, paddr_nz (struct_addr)); regcache_cooked_write_unsigned (regcache, argreg++, struct_addr); stack_offset += MIPS64_REGSIZE; } /* Now load as many as possible of the first arguments into registers, and push the rest onto the stack. Loop thru args from first to last. */ for (argnum = 0; argnum < nargs; argnum++) { const gdb_byte *val; struct value *arg = args[argnum]; struct type *arg_type = check_typedef (value_type (arg)); int len = TYPE_LENGTH (arg_type); enum type_code typecode = TYPE_CODE (arg_type); if (mips_debug) fprintf_unfiltered (gdb_stdlog, "mips_o64_push_dummy_call: %d len=%d type=%d", argnum + 1, len, (int) typecode); val = value_contents (arg); /* Floating point arguments passed in registers have to be treated specially. On 32-bit architectures, doubles are passed in register pairs; the even register gets the low word, and the odd register gets the high word. On O32/O64, the first two floating point arguments are also copied to general registers, because MIPS16 functions don't use float registers for arguments. This duplication of arguments in general registers can't hurt non-MIPS16 functions because those registers are normally skipped. */ if (fp_register_arg_p (typecode, arg_type) && float_argreg <= MIPS_LAST_FP_ARG_REGNUM) { LONGEST regval = extract_unsigned_integer (val, len); if (mips_debug) fprintf_unfiltered (gdb_stdlog, " - fpreg=%d val=%s", float_argreg, phex (regval, len)); regcache_cooked_write_unsigned (regcache, float_argreg++, regval); if (mips_debug) fprintf_unfiltered (gdb_stdlog, " - reg=%d val=%s", argreg, phex (regval, len)); regcache_cooked_write_unsigned (regcache, argreg, regval); argreg++; /* Reserve space for the FP register. */ stack_offset += align_up (len, MIPS64_REGSIZE); } else { /* Copy the argument to general registers or the stack in register-sized pieces. Large arguments are split between registers and stack. */ /* Note: structs whose size is not a multiple of MIPS64_REGSIZE are treated specially: Irix cc passes them in registers where gcc sometimes puts them on the stack. For maximum compatibility, we will put them in both places. */ int odd_sized_struct = (len > MIPS64_REGSIZE && len % MIPS64_REGSIZE != 0); while (len > 0) { /* Remember if the argument was written to the stack. */ int stack_used_p = 0; int partial_len = (len < MIPS64_REGSIZE ? len : MIPS64_REGSIZE); if (mips_debug) fprintf_unfiltered (gdb_stdlog, " -- partial=%d", partial_len); /* Write this portion of the argument to the stack. */ if (argreg > MIPS_LAST_ARG_REGNUM || odd_sized_struct) { /* Should shorter than int integer values be promoted to int before being stored? */ int longword_offset = 0; CORE_ADDR addr; stack_used_p = 1; if (gdbarch_byte_order (current_gdbarch) == BFD_ENDIAN_BIG) { if ((typecode == TYPE_CODE_INT || typecode == TYPE_CODE_PTR || typecode == TYPE_CODE_FLT) && len <= 4) longword_offset = MIPS64_REGSIZE - len; } if (mips_debug) { fprintf_unfiltered (gdb_stdlog, " - stack_offset=0x%s", paddr_nz (stack_offset)); fprintf_unfiltered (gdb_stdlog, " longword_offset=0x%s", paddr_nz (longword_offset)); } addr = sp + stack_offset + longword_offset; if (mips_debug) { int i; fprintf_unfiltered (gdb_stdlog, " @0x%s ", paddr_nz (addr)); for (i = 0; i < partial_len; i++) { fprintf_unfiltered (gdb_stdlog, "%02x", val[i] & 0xff); } } write_memory (addr, val, partial_len); } /* Note!!! This is NOT an else clause. Odd sized structs may go thru BOTH paths. */ /* Write this portion of the argument to a general purpose register. */ if (argreg <= MIPS_LAST_ARG_REGNUM) { LONGEST regval = extract_signed_integer (val, partial_len); /* Value may need to be sign extended, because mips_isa_regsize() != mips_abi_regsize(). */ /* A non-floating-point argument being passed in a general register. If a struct or union, and if the remaining length is smaller than the register size, we have to adjust the register value on big endian targets. It does not seem to be necessary to do the same for integral types. */ if (gdbarch_byte_order (current_gdbarch) == BFD_ENDIAN_BIG && partial_len < MIPS64_REGSIZE && (typecode == TYPE_CODE_STRUCT || typecode == TYPE_CODE_UNION)) regval <<= ((MIPS64_REGSIZE - partial_len) * TARGET_CHAR_BIT); if (mips_debug) fprintf_filtered (gdb_stdlog, " - reg=%d val=%s", argreg, phex (regval, MIPS64_REGSIZE)); regcache_cooked_write_unsigned (regcache, argreg, regval); argreg++; /* Prevent subsequent floating point arguments from being passed in floating point registers. */ float_argreg = MIPS_LAST_FP_ARG_REGNUM + 1; } len -= partial_len; val += partial_len; /* Compute the the offset into the stack at which we will copy the next parameter. In older ABIs, the caller reserved space for registers that contained arguments. This was loosely refered to as their "home". Consequently, space is always allocated. */ stack_offset += align_up (partial_len, MIPS64_REGSIZE); } } if (mips_debug) fprintf_unfiltered (gdb_stdlog, "\n"); } regcache_cooked_write_signed (regcache, MIPS_SP_REGNUM, sp); /* Return adjusted stack pointer. */ return sp; } static enum return_value_convention mips_o64_return_value (struct gdbarch *gdbarch, struct type *type, struct regcache *regcache, gdb_byte *readbuf, const gdb_byte *writebuf) { struct gdbarch_tdep *tdep = gdbarch_tdep (current_gdbarch); if (TYPE_CODE (type) == TYPE_CODE_STRUCT || TYPE_CODE (type) == TYPE_CODE_UNION || TYPE_CODE (type) == TYPE_CODE_ARRAY) return RETURN_VALUE_STRUCT_CONVENTION; else if (fp_register_arg_p (TYPE_CODE (type), type)) { /* A floating-point value. It fits in the least significant part of FP0. */ if (mips_debug) fprintf_unfiltered (gdb_stderr, "Return float in $fp0\n"); mips_xfer_register (regcache, gdbarch_num_regs (current_gdbarch) + mips_regnum (current_gdbarch)->fp0, TYPE_LENGTH (type), gdbarch_byte_order (current_gdbarch), readbuf, writebuf, 0); return RETURN_VALUE_REGISTER_CONVENTION; } else { /* A scalar extract each part but least-significant-byte justified. */ int offset; int regnum; for (offset = 0, regnum = MIPS_V0_REGNUM; offset < TYPE_LENGTH (type); offset += MIPS64_REGSIZE, regnum++) { int xfer = MIPS64_REGSIZE; if (offset + xfer > TYPE_LENGTH (type)) xfer = TYPE_LENGTH (type) - offset; if (mips_debug) fprintf_unfiltered (gdb_stderr, "Return scalar+%d:%d in $%d\n", offset, xfer, regnum); mips_xfer_register (regcache, gdbarch_num_regs (current_gdbarch) + regnum, xfer, gdbarch_byte_order (current_gdbarch), readbuf, writebuf, offset); } return RETURN_VALUE_REGISTER_CONVENTION; } } /* Floating point register management. Background: MIPS1 & 2 fp registers are 32 bits wide. To support 64bit operations, these early MIPS cpus treat fp register pairs (f0,f1) as a single register (d0). Later MIPS cpu's have 64 bit fp registers and offer a compatibility mode that emulates the MIPS2 fp model. When operating in MIPS2 fp compat mode, later cpu's split double precision floats into two 32-bit chunks and store them in consecutive fp regs. To display 64-bit floats stored in this fashion, we have to combine 32 bits from f0 and 32 bits from f1. Throw in user-configurable endianness and you have a real mess. The way this works is: - If we are in 32-bit mode or on a 32-bit processor, then a 64-bit double-precision value will be split across two logical registers. The lower-numbered logical register will hold the low-order bits, regardless of the processor's endianness. - If we are on a 64-bit processor, and we are looking for a single-precision value, it will be in the low ordered bits of a 64-bit GPR (after mfc1, for example) or a 64-bit register save slot in memory. - If we are in 64-bit mode, everything is straightforward. Note that this code only deals with "live" registers at the top of the stack. We will attempt to deal with saved registers later, when the raw/cooked register interface is in place. (We need a general interface that can deal with dynamic saved register sizes -- fp regs could be 32 bits wide in one frame and 64 on the frame above and below). */ static struct type * mips_float_register_type (void) { return builtin_type_ieee_single; } static struct type * mips_double_register_type (void) { return builtin_type_ieee_double; } /* Copy a 32-bit single-precision value from the current frame into rare_buffer. */ static void mips_read_fp_register_single (struct frame_info *frame, int regno, gdb_byte *rare_buffer) { int raw_size = register_size (current_gdbarch, regno); gdb_byte *raw_buffer = alloca (raw_size); if (!frame_register_read (frame, regno, raw_buffer)) error (_("can't read register %d (%s)"), regno, gdbarch_register_name (current_gdbarch, regno)); if (raw_size == 8) { /* We have a 64-bit value for this register. Find the low-order 32 bits. */ int offset; if (gdbarch_byte_order (current_gdbarch) == BFD_ENDIAN_BIG) offset = 4; else offset = 0; memcpy (rare_buffer, raw_buffer + offset, 4); } else { memcpy (rare_buffer, raw_buffer, 4); } } /* Copy a 64-bit double-precision value from the current frame into rare_buffer. This may include getting half of it from the next register. */ static void mips_read_fp_register_double (struct frame_info *frame, int regno, gdb_byte *rare_buffer) { int raw_size = register_size (current_gdbarch, regno); if (raw_size == 8 && !mips2_fp_compat (frame)) { /* We have a 64-bit value for this register, and we should use all 64 bits. */ if (!frame_register_read (frame, regno, rare_buffer)) error (_("can't read register %d (%s)"), regno, gdbarch_register_name (current_gdbarch, regno)); } else { if ((regno - mips_regnum (current_gdbarch)->fp0) & 1) internal_error (__FILE__, __LINE__, _("mips_read_fp_register_double: bad access to " "odd-numbered FP register")); /* mips_read_fp_register_single will find the correct 32 bits from each register. */ if (gdbarch_byte_order (current_gdbarch) == BFD_ENDIAN_BIG) { mips_read_fp_register_single (frame, regno, rare_buffer + 4); mips_read_fp_register_single (frame, regno + 1, rare_buffer); } else { mips_read_fp_register_single (frame, regno, rare_buffer); mips_read_fp_register_single (frame, regno + 1, rare_buffer + 4); } } } static void mips_print_fp_register (struct ui_file *file, struct frame_info *frame, int regnum) { /* do values for FP (float) regs */ gdb_byte *raw_buffer; double doub, flt1; /* doubles extracted from raw hex data */ int inv1, inv2; raw_buffer = alloca (2 * register_size (current_gdbarch, mips_regnum (current_gdbarch)->fp0)); fprintf_filtered (file, "%s:", gdbarch_register_name (current_gdbarch, regnum)); fprintf_filtered (file, "%*s", 4 - (int) strlen (gdbarch_register_name (current_gdbarch, regnum)), ""); if (register_size (current_gdbarch, regnum) == 4 || mips2_fp_compat (frame)) { /* 4-byte registers: Print hex and floating. Also print even numbered registers as doubles. */ mips_read_fp_register_single (frame, regnum, raw_buffer); flt1 = unpack_double (mips_float_register_type (), raw_buffer, &inv1); print_scalar_formatted (raw_buffer, builtin_type_uint32, 'x', 'w', file); fprintf_filtered (file, " flt: "); if (inv1) fprintf_filtered (file, " "); else fprintf_filtered (file, "%-17.9g", flt1); if (regnum % 2 == 0) { mips_read_fp_register_double (frame, regnum, raw_buffer); doub = unpack_double (mips_double_register_type (), raw_buffer, &inv2); fprintf_filtered (file, " dbl: "); if (inv2) fprintf_filtered (file, ""); else fprintf_filtered (file, "%-24.17g", doub); } } else { /* Eight byte registers: print each one as hex, float and double. */ mips_read_fp_register_single (frame, regnum, raw_buffer); flt1 = unpack_double (mips_float_register_type (), raw_buffer, &inv1); mips_read_fp_register_double (frame, regnum, raw_buffer); doub = unpack_double (mips_double_register_type (), raw_buffer, &inv2); print_scalar_formatted (raw_buffer, builtin_type_uint64, 'x', 'g', file); fprintf_filtered (file, " flt: "); if (inv1) fprintf_filtered (file, ""); else fprintf_filtered (file, "%-17.9g", flt1); fprintf_filtered (file, " dbl: "); if (inv2) fprintf_filtered (file, ""); else fprintf_filtered (file, "%-24.17g", doub); } } static void mips_print_register (struct ui_file *file, struct frame_info *frame, int regnum) { struct gdbarch *gdbarch = get_frame_arch (frame); gdb_byte raw_buffer[MAX_REGISTER_SIZE]; int offset; if (TYPE_CODE (register_type (gdbarch, regnum)) == TYPE_CODE_FLT) { mips_print_fp_register (file, frame, regnum); return; } /* Get the data in raw format. */ if (!frame_register_read (frame, regnum, raw_buffer)) { fprintf_filtered (file, "%s: [Invalid]", gdbarch_register_name (current_gdbarch, regnum)); return; } fputs_filtered (gdbarch_register_name (current_gdbarch, regnum), file); /* The problem with printing numeric register names (r26, etc.) is that the user can't use them on input. Probably the best solution is to fix it so that either the numeric or the funky (a2, etc.) names are accepted on input. */ if (regnum < MIPS_NUMREGS) fprintf_filtered (file, "(r%d): ", regnum); else fprintf_filtered (file, ": "); if (gdbarch_byte_order (current_gdbarch) == BFD_ENDIAN_BIG) offset = register_size (current_gdbarch, regnum) - register_size (current_gdbarch, regnum); else offset = 0; print_scalar_formatted (raw_buffer + offset, register_type (gdbarch, regnum), 'x', 0, file); } /* Replacement for generic do_registers_info. Print regs in pretty columns. */ static int print_fp_register_row (struct ui_file *file, struct frame_info *frame, int regnum) { fprintf_filtered (file, " "); mips_print_fp_register (file, frame, regnum); fprintf_filtered (file, "\n"); return regnum + 1; } /* Print a row's worth of GP (int) registers, with name labels above */ static int print_gp_register_row (struct ui_file *file, struct frame_info *frame, int start_regnum) { struct gdbarch *gdbarch = get_frame_arch (frame); /* do values for GP (int) regs */ gdb_byte raw_buffer[MAX_REGISTER_SIZE]; int ncols = (mips_abi_regsize (gdbarch) == 8 ? 4 : 8); /* display cols per row */ int col, byte; int regnum; /* For GP registers, we print a separate row of names above the vals */ for (col = 0, regnum = start_regnum; col < ncols && regnum < gdbarch_num_regs (current_gdbarch) + gdbarch_num_pseudo_regs (current_gdbarch); regnum++) { if (*gdbarch_register_name (current_gdbarch, regnum) == '\0') continue; /* unused register */ if (TYPE_CODE (register_type (gdbarch, regnum)) == TYPE_CODE_FLT) break; /* end the row: reached FP register */ /* Large registers are handled separately. */ if (register_size (current_gdbarch, regnum) > mips_abi_regsize (current_gdbarch)) { if (col > 0) break; /* End the row before this register. */ /* Print this register on a row by itself. */ mips_print_register (file, frame, regnum); fprintf_filtered (file, "\n"); return regnum + 1; } if (col == 0) fprintf_filtered (file, " "); fprintf_filtered (file, mips_abi_regsize (current_gdbarch) == 8 ? "%17s" : "%9s", gdbarch_register_name (current_gdbarch, regnum)); col++; } if (col == 0) return regnum; /* print the R0 to R31 names */ if ((start_regnum % gdbarch_num_regs (current_gdbarch)) < MIPS_NUMREGS) fprintf_filtered (file, "\n R%-4d", start_regnum % gdbarch_num_regs (current_gdbarch)); else fprintf_filtered (file, "\n "); /* now print the values in hex, 4 or 8 to the row */ for (col = 0, regnum = start_regnum; col < ncols && regnum < gdbarch_num_regs (current_gdbarch) + gdbarch_num_pseudo_regs (current_gdbarch); regnum++) { if (*gdbarch_register_name (current_gdbarch, regnum) == '\0') continue; /* unused register */ if (TYPE_CODE (register_type (gdbarch, regnum)) == TYPE_CODE_FLT) break; /* end row: reached FP register */ if (register_size (current_gdbarch, regnum) > mips_abi_regsize (current_gdbarch)) break; /* End row: large register. */ /* OK: get the data in raw format. */ if (!frame_register_read (frame, regnum, raw_buffer)) error (_("can't read register %d (%s)"), regnum, gdbarch_register_name (current_gdbarch, regnum)); /* pad small registers */ for (byte = 0; byte < (mips_abi_regsize (current_gdbarch) - register_size (current_gdbarch, regnum)); byte++) printf_filtered (" "); /* Now print the register value in hex, endian order. */ if (gdbarch_byte_order (current_gdbarch) == BFD_ENDIAN_BIG) for (byte = register_size (current_gdbarch, regnum) - register_size (current_gdbarch, regnum); byte < register_size (current_gdbarch, regnum); byte++) fprintf_filtered (file, "%02x", raw_buffer[byte]); else for (byte = register_size (current_gdbarch, regnum) - 1; byte >= 0; byte--) fprintf_filtered (file, "%02x", raw_buffer[byte]); fprintf_filtered (file, " "); col++; } if (col > 0) /* ie. if we actually printed anything... */ fprintf_filtered (file, "\n"); return regnum; } /* MIPS_DO_REGISTERS_INFO(): called by "info register" command */ static void mips_print_registers_info (struct gdbarch *gdbarch, struct ui_file *file, struct frame_info *frame, int regnum, int all) { if (regnum != -1) /* do one specified register */ { gdb_assert (regnum >= gdbarch_num_regs (current_gdbarch)); if (*(gdbarch_register_name (current_gdbarch, regnum)) == '\0') error (_("Not a valid register for the current processor type")); mips_print_register (file, frame, regnum); fprintf_filtered (file, "\n"); } else /* do all (or most) registers */ { regnum = gdbarch_num_regs (current_gdbarch); while (regnum < gdbarch_num_regs (current_gdbarch) + gdbarch_num_pseudo_regs (current_gdbarch)) { if (TYPE_CODE (register_type (gdbarch, regnum)) == TYPE_CODE_FLT) { if (all) /* true for "INFO ALL-REGISTERS" command */ regnum = print_fp_register_row (file, frame, regnum); else regnum += MIPS_NUMREGS; /* skip floating point regs */ } else regnum = print_gp_register_row (file, frame, regnum); } } } /* Is this a branch with a delay slot? */ static int is_delayed (unsigned long insn) { int i; for (i = 0; i < NUMOPCODES; ++i) if (mips_opcodes[i].pinfo != INSN_MACRO && (insn & mips_opcodes[i].mask) == mips_opcodes[i].match) break; return (i < NUMOPCODES && (mips_opcodes[i].pinfo & (INSN_UNCOND_BRANCH_DELAY | INSN_COND_BRANCH_DELAY | INSN_COND_BRANCH_LIKELY))); } int mips_single_step_through_delay (struct gdbarch *gdbarch, struct frame_info *frame) { CORE_ADDR pc = get_frame_pc (frame); gdb_byte buf[MIPS_INSN32_SIZE]; /* There is no branch delay slot on MIPS16. */ if (mips_pc_is_mips16 (pc)) return 0; if (!breakpoint_here_p (pc + 4)) return 0; if (!safe_frame_unwind_memory (frame, pc, buf, sizeof buf)) /* If error reading memory, guess that it is not a delayed branch. */ return 0; return is_delayed (extract_unsigned_integer (buf, sizeof buf)); } /* To skip prologues, I use this predicate. Returns either PC itself if the code at PC does not look like a function prologue; otherwise returns an address that (if we're lucky) follows the prologue. If LENIENT, then we must skip everything which is involved in setting up the frame (it's OK to skip more, just so long as we don't skip anything which might clobber the registers which are being saved. We must skip more in the case where part of the prologue is in the delay slot of a non-prologue instruction). */ static CORE_ADDR mips_skip_prologue (CORE_ADDR pc) { CORE_ADDR limit_pc; CORE_ADDR func_addr; /* See if we can determine the end of the prologue via the symbol table. If so, then return either PC, or the PC after the prologue, whichever is greater. */ if (find_pc_partial_function (pc, NULL, &func_addr, NULL)) { CORE_ADDR post_prologue_pc = skip_prologue_using_sal (func_addr); if (post_prologue_pc != 0) return max (pc, post_prologue_pc); } /* Can't determine prologue from the symbol table, need to examine instructions. */ /* Find an upper limit on the function prologue using the debug information. If the debug information could not be used to provide that bound, then use an arbitrary large number as the upper bound. */ limit_pc = skip_prologue_using_sal (pc); if (limit_pc == 0) limit_pc = pc + 100; /* Magic. */ if (mips_pc_is_mips16 (pc)) return mips16_scan_prologue (pc, limit_pc, NULL, NULL); else return mips32_scan_prologue (pc, limit_pc, NULL, NULL); } /* Root of all "set mips "/"show mips " commands. This will eventually be used for all MIPS-specific commands. */ static void show_mips_command (char *args, int from_tty) { help_list (showmipscmdlist, "show mips ", all_commands, gdb_stdout); } static void set_mips_command (char *args, int from_tty) { printf_unfiltered ("\"set mips\" must be followed by an appropriate subcommand.\n"); help_list (setmipscmdlist, "set mips ", all_commands, gdb_stdout); } /* Commands to show/set the MIPS FPU type. */ static void show_mipsfpu_command (char *args, int from_tty) { char *fpu; switch (MIPS_FPU_TYPE) { case MIPS_FPU_SINGLE: fpu = "single-precision"; break; case MIPS_FPU_DOUBLE: fpu = "double-precision"; break; case MIPS_FPU_NONE: fpu = "absent (none)"; break; default: internal_error (__FILE__, __LINE__, _("bad switch")); } if (mips_fpu_type_auto) printf_unfiltered ("The MIPS floating-point coprocessor is set automatically (currently %s)\n", fpu); else printf_unfiltered ("The MIPS floating-point coprocessor is assumed to be %s\n", fpu); } static void set_mipsfpu_command (char *args, int from_tty) { printf_unfiltered ("\"set mipsfpu\" must be followed by \"double\", \"single\",\"none\" or \"auto\".\n"); show_mipsfpu_command (args, from_tty); } static void set_mipsfpu_single_command (char *args, int from_tty) { struct gdbarch_info info; gdbarch_info_init (&info); mips_fpu_type = MIPS_FPU_SINGLE; mips_fpu_type_auto = 0; /* FIXME: cagney/2003-11-15: Should be setting a field in "info" instead of relying on globals. Doing that would let generic code handle the search for this specific architecture. */ if (!gdbarch_update_p (info)) internal_error (__FILE__, __LINE__, _("set mipsfpu failed")); } static void set_mipsfpu_double_command (char *args, int from_tty) { struct gdbarch_info info; gdbarch_info_init (&info); mips_fpu_type = MIPS_FPU_DOUBLE; mips_fpu_type_auto = 0; /* FIXME: cagney/2003-11-15: Should be setting a field in "info" instead of relying on globals. Doing that would let generic code handle the search for this specific architecture. */ if (!gdbarch_update_p (info)) internal_error (__FILE__, __LINE__, _("set mipsfpu failed")); } static void set_mipsfpu_none_command (char *args, int from_tty) { struct gdbarch_info info; gdbarch_info_init (&info); mips_fpu_type = MIPS_FPU_NONE; mips_fpu_type_auto = 0; /* FIXME: cagney/2003-11-15: Should be setting a field in "info" instead of relying on globals. Doing that would let generic code handle the search for this specific architecture. */ if (!gdbarch_update_p (info)) internal_error (__FILE__, __LINE__, _("set mipsfpu failed")); } static void set_mipsfpu_auto_command (char *args, int from_tty) { mips_fpu_type_auto = 1; } /* Attempt to identify the particular processor model by reading the processor id. NOTE: cagney/2003-11-15: Firstly it isn't clear that the relevant processor still exists (it dates back to '94) and secondly this is not the way to do this. The processor type should be set by forcing an architecture change. */ void deprecated_mips_set_processor_regs_hack (void) { struct gdbarch_tdep *tdep = gdbarch_tdep (current_gdbarch); ULONGEST prid; regcache_cooked_read_unsigned (current_regcache, MIPS_PRID_REGNUM, &prid); if ((prid & ~0xf) == 0x700) tdep->mips_processor_reg_names = mips_r3041_reg_names; } /* Just like reinit_frame_cache, but with the right arguments to be callable as an sfunc. */ static void reinit_frame_cache_sfunc (char *args, int from_tty, struct cmd_list_element *c) { reinit_frame_cache (); } static int gdb_print_insn_mips (bfd_vma memaddr, struct disassemble_info *info) { struct gdbarch_tdep *tdep = gdbarch_tdep (current_gdbarch); /* FIXME: cagney/2003-06-26: Is this even necessary? The disassembler needs to be able to locally determine the ISA, and not rely on GDB. Otherwize the stand-alone 'objdump -d' will not work. */ if (mips_pc_is_mips16 (memaddr)) info->mach = bfd_mach_mips16; /* Round down the instruction address to the appropriate boundary. */ memaddr &= (info->mach == bfd_mach_mips16 ? ~1 : ~3); /* Set the disassembler options. */ if (tdep->mips_abi == MIPS_ABI_N32 || tdep->mips_abi == MIPS_ABI_N64) { /* Set up the disassembler info, so that we get the right register names from libopcodes. */ if (tdep->mips_abi == MIPS_ABI_N32) info->disassembler_options = "gpr-names=n32"; else info->disassembler_options = "gpr-names=64"; info->flavour = bfd_target_elf_flavour; } else /* This string is not recognized explicitly by the disassembler, but it tells the disassembler to not try to guess the ABI from the bfd elf headers, such that, if the user overrides the ABI of a program linked as NewABI, the disassembly will follow the register naming conventions specified by the user. */ info->disassembler_options = "gpr-names=32"; /* Call the appropriate disassembler based on the target endian-ness. */ if (gdbarch_byte_order (current_gdbarch) == BFD_ENDIAN_BIG) return print_insn_big_mips (memaddr, info); else return print_insn_little_mips (memaddr, info); } /* This function implements gdbarch_breakpoint_from_pc. It uses the program counter value to determine whether a 16- or 32-bit breakpoint should be used. It returns a pointer to a string of bytes that encode a breakpoint instruction, stores the length of the string to *lenptr, and adjusts pc (if necessary) to point to the actual memory location where the breakpoint should be inserted. */ static const gdb_byte * mips_breakpoint_from_pc (CORE_ADDR *pcptr, int *lenptr) { if (gdbarch_byte_order (current_gdbarch) == BFD_ENDIAN_BIG) { if (mips_pc_is_mips16 (*pcptr)) { static gdb_byte mips16_big_breakpoint[] = { 0xe8, 0xa5 }; *pcptr = unmake_mips16_addr (*pcptr); *lenptr = sizeof (mips16_big_breakpoint); return mips16_big_breakpoint; } else { /* The IDT board uses an unusual breakpoint value, and sometimes gets confused when it sees the usual MIPS breakpoint instruction. */ static gdb_byte big_breakpoint[] = { 0, 0x5, 0, 0xd }; static gdb_byte pmon_big_breakpoint[] = { 0, 0, 0, 0xd }; static gdb_byte idt_big_breakpoint[] = { 0, 0, 0x0a, 0xd }; *lenptr = sizeof (big_breakpoint); if (strcmp (target_shortname, "mips") == 0) return idt_big_breakpoint; else if (strcmp (target_shortname, "ddb") == 0 || strcmp (target_shortname, "pmon") == 0 || strcmp (target_shortname, "lsi") == 0) return pmon_big_breakpoint; else return big_breakpoint; } } else { if (mips_pc_is_mips16 (*pcptr)) { static gdb_byte mips16_little_breakpoint[] = { 0xa5, 0xe8 }; *pcptr = unmake_mips16_addr (*pcptr); *lenptr = sizeof (mips16_little_breakpoint); return mips16_little_breakpoint; } else { static gdb_byte little_breakpoint[] = { 0xd, 0, 0x5, 0 }; static gdb_byte pmon_little_breakpoint[] = { 0xd, 0, 0, 0 }; static gdb_byte idt_little_breakpoint[] = { 0xd, 0x0a, 0, 0 }; *lenptr = sizeof (little_breakpoint); if (strcmp (target_shortname, "mips") == 0) return idt_little_breakpoint; else if (strcmp (target_shortname, "ddb") == 0 || strcmp (target_shortname, "pmon") == 0 || strcmp (target_shortname, "lsi") == 0) return pmon_little_breakpoint; else return little_breakpoint; } } } /* If PC is in a mips16 call or return stub, return the address of the target PC, which is either the callee or the caller. There are several cases which must be handled: * If the PC is in __mips16_ret_{d,s}f, this is a return stub and the target PC is in $31 ($ra). * If the PC is in __mips16_call_stub_{1..10}, this is a call stub and the target PC is in $2. * If the PC at the start of __mips16_call_stub_{s,d}f_{0..10}, i.e. before the jal instruction, this is effectively a call stub and the the target PC is in $2. Otherwise this is effectively a return stub and the target PC is in $18. See the source code for the stubs in gcc/config/mips/mips16.S for gory details. */ static CORE_ADDR mips_skip_trampoline_code (CORE_ADDR pc) { char *name; CORE_ADDR start_addr; /* Find the starting address and name of the function containing the PC. */ if (find_pc_partial_function (pc, &name, &start_addr, NULL) == 0) return 0; /* If the PC is in __mips16_ret_{d,s}f, this is a return stub and the target PC is in $31 ($ra). */ if (strcmp (name, "__mips16_ret_sf") == 0 || strcmp (name, "__mips16_ret_df") == 0) return read_signed_register (MIPS_RA_REGNUM); if (strncmp (name, "__mips16_call_stub_", 19) == 0) { /* If the PC is in __mips16_call_stub_{1..10}, this is a call stub and the target PC is in $2. */ if (name[19] >= '0' && name[19] <= '9') return read_signed_register (2); /* If the PC at the start of __mips16_call_stub_{s,d}f_{0..10}, i.e. before the jal instruction, this is effectively a call stub and the the target PC is in $2. Otherwise this is effectively a return stub and the target PC is in $18. */ else if (name[19] == 's' || name[19] == 'd') { if (pc == start_addr) { /* Check if the target of the stub is a compiler-generated stub. Such a stub for a function bar might have a name like __fn_stub_bar, and might look like this: mfc1 $4,$f13 mfc1 $5,$f12 mfc1 $6,$f15 mfc1 $7,$f14 la $1,bar (becomes a lui/addiu pair) jr $1 So scan down to the lui/addi and extract the target address from those two instructions. */ CORE_ADDR target_pc = read_signed_register (2); ULONGEST inst; int i; /* See if the name of the target function is __fn_stub_*. */ if (find_pc_partial_function (target_pc, &name, NULL, NULL) == 0) return target_pc; if (strncmp (name, "__fn_stub_", 10) != 0 && strcmp (name, "etext") != 0 && strcmp (name, "_etext") != 0) return target_pc; /* Scan through this _fn_stub_ code for the lui/addiu pair. The limit on the search is arbitrarily set to 20 instructions. FIXME. */ for (i = 0, pc = 0; i < 20; i++, target_pc += MIPS_INSN32_SIZE) { inst = mips_fetch_instruction (target_pc); if ((inst & 0xffff0000) == 0x3c010000) /* lui $at */ pc = (inst << 16) & 0xffff0000; /* high word */ else if ((inst & 0xffff0000) == 0x24210000) /* addiu $at */ return pc | (inst & 0xffff); /* low word */ } /* Couldn't find the lui/addui pair, so return stub address. */ return target_pc; } else /* This is the 'return' part of a call stub. The return address is in $r18. */ return read_signed_register (18); } } return 0; /* not a stub */ } /* Convert a dbx stab register number (from `r' declaration) to a GDB [1 * gdbarch_num_regs .. 2 * gdbarch_num_regs) REGNUM. */ static int mips_stab_reg_to_regnum (int num) { int regnum; if (num >= 0 && num < 32) regnum = num; else if (num >= 38 && num < 70) regnum = num + mips_regnum (current_gdbarch)->fp0 - 38; else if (num == 70) regnum = mips_regnum (current_gdbarch)->hi; else if (num == 71) regnum = mips_regnum (current_gdbarch)->lo; else /* This will hopefully (eventually) provoke a warning. Should we be calling complaint() here? */ return gdbarch_num_regs (current_gdbarch) + gdbarch_num_pseudo_regs (current_gdbarch); return gdbarch_num_regs (current_gdbarch) + regnum; } /* Convert a dwarf, dwarf2, or ecoff register number to a GDB [1 * gdbarch_num_regs .. 2 * gdbarch_num_regs) REGNUM. */ static int mips_dwarf_dwarf2_ecoff_reg_to_regnum (int num) { int regnum; if (num >= 0 && num < 32) regnum = num; else if (num >= 32 && num < 64) regnum = num + mips_regnum (current_gdbarch)->fp0 - 32; else if (num == 64) regnum = mips_regnum (current_gdbarch)->hi; else if (num == 65) regnum = mips_regnum (current_gdbarch)->lo; else /* This will hopefully (eventually) provoke a warning. Should we be calling complaint() here? */ return gdbarch_num_regs (current_gdbarch) + gdbarch_num_pseudo_regs (current_gdbarch); return gdbarch_num_regs (current_gdbarch) + regnum; } static int mips_register_sim_regno (int regnum) { /* Only makes sense to supply raw registers. */ gdb_assert (regnum >= 0 && regnum < gdbarch_num_regs (current_gdbarch)); /* FIXME: cagney/2002-05-13: Need to look at the pseudo register to decide if it is valid. Should instead define a standard sim/gdb register numbering scheme. */ if (gdbarch_register_name (current_gdbarch, gdbarch_num_regs (current_gdbarch) + regnum) != NULL && gdbarch_register_name (current_gdbarch, gdbarch_num_regs (current_gdbarch) + regnum)[0] != '\0') return regnum; else return LEGACY_SIM_REGNO_IGNORE; } /* Convert an integer into an address. Extracting the value signed guarantees a correctly sign extended address. */ static CORE_ADDR mips_integer_to_address (struct gdbarch *gdbarch, struct type *type, const gdb_byte *buf) { return (CORE_ADDR) extract_signed_integer (buf, TYPE_LENGTH (type)); } static void mips_find_abi_section (bfd *abfd, asection *sect, void *obj) { enum mips_abi *abip = (enum mips_abi *) obj; const char *name = bfd_get_section_name (abfd, sect); if (*abip != MIPS_ABI_UNKNOWN) return; if (strncmp (name, ".mdebug.", 8) != 0) return; if (strcmp (name, ".mdebug.abi32") == 0) *abip = MIPS_ABI_O32; else if (strcmp (name, ".mdebug.abiN32") == 0) *abip = MIPS_ABI_N32; else if (strcmp (name, ".mdebug.abi64") == 0) *abip = MIPS_ABI_N64; else if (strcmp (name, ".mdebug.abiO64") == 0) *abip = MIPS_ABI_O64; else if (strcmp (name, ".mdebug.eabi32") == 0) *abip = MIPS_ABI_EABI32; else if (strcmp (name, ".mdebug.eabi64") == 0) *abip = MIPS_ABI_EABI64; else warning (_("unsupported ABI %s."), name + 8); } static void mips_find_long_section (bfd *abfd, asection *sect, void *obj) { int *lbp = (int *) obj; const char *name = bfd_get_section_name (abfd, sect); if (strncmp (name, ".gcc_compiled_long32", 20) == 0) *lbp = 32; else if (strncmp (name, ".gcc_compiled_long64", 20) == 0) *lbp = 64; else if (strncmp (name, ".gcc_compiled_long", 18) == 0) warning (_("unrecognized .gcc_compiled_longXX")); } static enum mips_abi global_mips_abi (void) { int i; for (i = 0; mips_abi_strings[i] != NULL; i++) if (mips_abi_strings[i] == mips_abi_string) return (enum mips_abi) i; internal_error (__FILE__, __LINE__, _("unknown ABI string")); } static void mips_register_g_packet_guesses (struct gdbarch *gdbarch) { static struct target_desc *tdesc_gp32, *tdesc_gp64; if (tdesc_gp32 == NULL) { /* Create feature sets with the appropriate properties. The values are not important. */ tdesc_gp32 = allocate_target_description (); set_tdesc_property (tdesc_gp32, PROPERTY_GP32, ""); tdesc_gp64 = allocate_target_description (); set_tdesc_property (tdesc_gp64, PROPERTY_GP64, ""); } /* If the size matches the set of 32-bit or 64-bit integer registers, assume that's what we've got. */ register_remote_g_packet_guess (gdbarch, 38 * 4, tdesc_gp32); register_remote_g_packet_guess (gdbarch, 38 * 8, tdesc_gp64); /* If the size matches the full set of registers GDB traditionally knows about, including floating point, for either 32-bit or 64-bit, assume that's what we've got. */ register_remote_g_packet_guess (gdbarch, 90 * 4, tdesc_gp32); register_remote_g_packet_guess (gdbarch, 90 * 8, tdesc_gp64); /* Otherwise we don't have a useful guess. */ } static struct value * value_of_mips_user_reg (struct frame_info *frame, const void *baton) { const int *reg_p = baton; return value_of_register (*reg_p, frame); } static struct gdbarch * mips_gdbarch_init (struct gdbarch_info info, struct gdbarch_list *arches) { struct gdbarch *gdbarch; struct gdbarch_tdep *tdep; int elf_flags; enum mips_abi mips_abi, found_abi, wanted_abi; int i, num_regs; enum mips_fpu_type fpu_type; struct tdesc_arch_data *tdesc_data = NULL; /* Check any target description for validity. */ if (tdesc_has_registers (info.target_desc)) { static const char *const mips_gprs[] = { "r0", "r1", "r2", "r3", "r4", "r5", "r6", "r7", "r8", "r9", "r10", "r11", "r12", "r13", "r14", "r15", "r16", "r17", "r18", "r19", "r20", "r21", "r22", "r23", "r24", "r25", "r26", "r27", "r28", "r29", "r30", "r31" }; static const char *const mips_fprs[] = { "f0", "f1", "f2", "f3", "f4", "f5", "f6", "f7", "f8", "f9", "f10", "f11", "f12", "f13", "f14", "f15", "f16", "f17", "f18", "f19", "f20", "f21", "f22", "f23", "f24", "f25", "f26", "f27", "f28", "f29", "f30", "f31", }; const struct tdesc_feature *feature; int valid_p; feature = tdesc_find_feature (info.target_desc, "org.gnu.gdb.mips.cpu"); if (feature == NULL) return NULL; tdesc_data = tdesc_data_alloc (); valid_p = 1; for (i = MIPS_ZERO_REGNUM; i <= MIPS_RA_REGNUM; i++) valid_p &= tdesc_numbered_register (feature, tdesc_data, i, mips_gprs[i]); valid_p &= tdesc_numbered_register (feature, tdesc_data, MIPS_EMBED_LO_REGNUM, "lo"); valid_p &= tdesc_numbered_register (feature, tdesc_data, MIPS_EMBED_HI_REGNUM, "hi"); valid_p &= tdesc_numbered_register (feature, tdesc_data, MIPS_EMBED_PC_REGNUM, "pc"); if (!valid_p) { tdesc_data_cleanup (tdesc_data); return NULL; } feature = tdesc_find_feature (info.target_desc, "org.gnu.gdb.mips.cp0"); if (feature == NULL) { tdesc_data_cleanup (tdesc_data); return NULL; } valid_p = 1; valid_p &= tdesc_numbered_register (feature, tdesc_data, MIPS_EMBED_BADVADDR_REGNUM, "badvaddr"); valid_p &= tdesc_numbered_register (feature, tdesc_data, MIPS_PS_REGNUM, "status"); valid_p &= tdesc_numbered_register (feature, tdesc_data, MIPS_EMBED_CAUSE_REGNUM, "cause"); if (!valid_p) { tdesc_data_cleanup (tdesc_data); return NULL; } /* FIXME drow/2007-05-17: The FPU should be optional. The MIPS backend is not prepared for that, though. */ feature = tdesc_find_feature (info.target_desc, "org.gnu.gdb.mips.fpu"); if (feature == NULL) { tdesc_data_cleanup (tdesc_data); return NULL; } valid_p = 1; for (i = 0; i < 32; i++) valid_p &= tdesc_numbered_register (feature, tdesc_data, i + MIPS_EMBED_FP0_REGNUM, mips_fprs[i]); valid_p &= tdesc_numbered_register (feature, tdesc_data, MIPS_EMBED_FP0_REGNUM + 32, "fcsr"); valid_p &= tdesc_numbered_register (feature, tdesc_data, MIPS_EMBED_FP0_REGNUM + 33, "fir"); if (!valid_p) { tdesc_data_cleanup (tdesc_data); return NULL; } /* It would be nice to detect an attempt to use a 64-bit ABI when only 32-bit registers are provided. */ } /* First of all, extract the elf_flags, if available. */ if (info.abfd && bfd_get_flavour (info.abfd) == bfd_target_elf_flavour) elf_flags = elf_elfheader (info.abfd)->e_flags; else if (arches != NULL) elf_flags = gdbarch_tdep (arches->gdbarch)->elf_flags; else elf_flags = 0; if (gdbarch_debug) fprintf_unfiltered (gdb_stdlog, "mips_gdbarch_init: elf_flags = 0x%08x\n", elf_flags); /* Check ELF_FLAGS to see if it specifies the ABI being used. */ switch ((elf_flags & EF_MIPS_ABI)) { case E_MIPS_ABI_O32: found_abi = MIPS_ABI_O32; break; case E_MIPS_ABI_O64: found_abi = MIPS_ABI_O64; break; case E_MIPS_ABI_EABI32: found_abi = MIPS_ABI_EABI32; break; case E_MIPS_ABI_EABI64: found_abi = MIPS_ABI_EABI64; break; default: if ((elf_flags & EF_MIPS_ABI2)) found_abi = MIPS_ABI_N32; else found_abi = MIPS_ABI_UNKNOWN; break; } /* GCC creates a pseudo-section whose name describes the ABI. */ if (found_abi == MIPS_ABI_UNKNOWN && info.abfd != NULL) bfd_map_over_sections (info.abfd, mips_find_abi_section, &found_abi); /* If we have no useful BFD information, use the ABI from the last MIPS architecture (if there is one). */ if (found_abi == MIPS_ABI_UNKNOWN && info.abfd == NULL && arches != NULL) found_abi = gdbarch_tdep (arches->gdbarch)->found_abi; /* Try the architecture for any hint of the correct ABI. */ if (found_abi == MIPS_ABI_UNKNOWN && info.bfd_arch_info != NULL && info.bfd_arch_info->arch == bfd_arch_mips) { switch (info.bfd_arch_info->mach) { case bfd_mach_mips3900: found_abi = MIPS_ABI_EABI32; break; case bfd_mach_mips4100: case bfd_mach_mips5000: found_abi = MIPS_ABI_EABI64; break; case bfd_mach_mips8000: case bfd_mach_mips10000: /* On Irix, ELF64 executables use the N64 ABI. The pseudo-sections which describe the ABI aren't present on IRIX. (Even for executables created by gcc.) */ if (bfd_get_flavour (info.abfd) == bfd_target_elf_flavour && elf_elfheader (info.abfd)->e_ident[EI_CLASS] == ELFCLASS64) found_abi = MIPS_ABI_N64; else found_abi = MIPS_ABI_N32; break; } } /* Default 64-bit objects to N64 instead of O32. */ if (found_abi == MIPS_ABI_UNKNOWN && info.abfd != NULL && bfd_get_flavour (info.abfd) == bfd_target_elf_flavour && elf_elfheader (info.abfd)->e_ident[EI_CLASS] == ELFCLASS64) found_abi = MIPS_ABI_N64; if (gdbarch_debug) fprintf_unfiltered (gdb_stdlog, "mips_gdbarch_init: found_abi = %d\n", found_abi); /* What has the user specified from the command line? */ wanted_abi = global_mips_abi (); if (gdbarch_debug) fprintf_unfiltered (gdb_stdlog, "mips_gdbarch_init: wanted_abi = %d\n", wanted_abi); /* Now that we have found what the ABI for this binary would be, check whether the user is overriding it. */ if (wanted_abi != MIPS_ABI_UNKNOWN) mips_abi = wanted_abi; else if (found_abi != MIPS_ABI_UNKNOWN) mips_abi = found_abi; else mips_abi = MIPS_ABI_O32; if (gdbarch_debug) fprintf_unfiltered (gdb_stdlog, "mips_gdbarch_init: mips_abi = %d\n", mips_abi); /* Also used when doing an architecture lookup. */ if (gdbarch_debug) fprintf_unfiltered (gdb_stdlog, "mips_gdbarch_init: mips64_transfers_32bit_regs_p = %d\n", mips64_transfers_32bit_regs_p); /* Determine the MIPS FPU type. */ if (!mips_fpu_type_auto) fpu_type = mips_fpu_type; else if (info.bfd_arch_info != NULL && info.bfd_arch_info->arch == bfd_arch_mips) switch (info.bfd_arch_info->mach) { case bfd_mach_mips3900: case bfd_mach_mips4100: case bfd_mach_mips4111: case bfd_mach_mips4120: fpu_type = MIPS_FPU_NONE; break; case bfd_mach_mips4650: fpu_type = MIPS_FPU_SINGLE; break; default: fpu_type = MIPS_FPU_DOUBLE; break; } else if (arches != NULL) fpu_type = gdbarch_tdep (arches->gdbarch)->mips_fpu_type; else fpu_type = MIPS_FPU_DOUBLE; if (gdbarch_debug) fprintf_unfiltered (gdb_stdlog, "mips_gdbarch_init: fpu_type = %d\n", fpu_type); /* Check for blatant incompatibilities. */ /* If we have only 32-bit registers, then we can't debug a 64-bit ABI. */ if (info.target_desc && tdesc_property (info.target_desc, PROPERTY_GP32) != NULL && mips_abi != MIPS_ABI_EABI32 && mips_abi != MIPS_ABI_O32) { if (tdesc_data != NULL) tdesc_data_cleanup (tdesc_data); return NULL; } /* try to find a pre-existing architecture */ for (arches = gdbarch_list_lookup_by_info (arches, &info); arches != NULL; arches = gdbarch_list_lookup_by_info (arches->next, &info)) { /* MIPS needs to be pedantic about which ABI the object is using. */ if (gdbarch_tdep (arches->gdbarch)->elf_flags != elf_flags) continue; if (gdbarch_tdep (arches->gdbarch)->mips_abi != mips_abi) continue; /* Need to be pedantic about which register virtual size is used. */ if (gdbarch_tdep (arches->gdbarch)->mips64_transfers_32bit_regs_p != mips64_transfers_32bit_regs_p) continue; /* Be pedantic about which FPU is selected. */ if (gdbarch_tdep (arches->gdbarch)->mips_fpu_type != fpu_type) continue; if (tdesc_data != NULL) tdesc_data_cleanup (tdesc_data); return arches->gdbarch; } /* Need a new architecture. Fill in a target specific vector. */ tdep = (struct gdbarch_tdep *) xmalloc (sizeof (struct gdbarch_tdep)); gdbarch = gdbarch_alloc (&info, tdep); tdep->elf_flags = elf_flags; tdep->mips64_transfers_32bit_regs_p = mips64_transfers_32bit_regs_p; tdep->found_abi = found_abi; tdep->mips_abi = mips_abi; tdep->mips_fpu_type = fpu_type; tdep->register_size_valid_p = 0; tdep->register_size = 0; if (info.target_desc) { /* Some useful properties can be inferred from the target. */ if (tdesc_property (info.target_desc, PROPERTY_GP32) != NULL) { tdep->register_size_valid_p = 1; tdep->register_size = 4; } else if (tdesc_property (info.target_desc, PROPERTY_GP64) != NULL) { tdep->register_size_valid_p = 1; tdep->register_size = 8; } } /* Initially set everything according to the default ABI/ISA. */ set_gdbarch_short_bit (gdbarch, 16); set_gdbarch_int_bit (gdbarch, 32); set_gdbarch_float_bit (gdbarch, 32); set_gdbarch_double_bit (gdbarch, 64); set_gdbarch_long_double_bit (gdbarch, 64); set_gdbarch_register_reggroup_p (gdbarch, mips_register_reggroup_p); set_gdbarch_pseudo_register_read (gdbarch, mips_pseudo_register_read); set_gdbarch_pseudo_register_write (gdbarch, mips_pseudo_register_write); set_gdbarch_elf_make_msymbol_special (gdbarch, mips_elf_make_msymbol_special); /* Fill in the OS dependant register numbers and names. */ { const char **reg_names; struct mips_regnum *regnum = GDBARCH_OBSTACK_ZALLOC (gdbarch, struct mips_regnum); if (tdesc_has_registers (info.target_desc)) { regnum->lo = MIPS_EMBED_LO_REGNUM; regnum->hi = MIPS_EMBED_HI_REGNUM; regnum->badvaddr = MIPS_EMBED_BADVADDR_REGNUM; regnum->cause = MIPS_EMBED_CAUSE_REGNUM; regnum->pc = MIPS_EMBED_PC_REGNUM; regnum->fp0 = MIPS_EMBED_FP0_REGNUM; regnum->fp_control_status = 70; regnum->fp_implementation_revision = 71; num_regs = MIPS_LAST_EMBED_REGNUM + 1; reg_names = NULL; } else if (info.osabi == GDB_OSABI_IRIX) { regnum->fp0 = 32; regnum->pc = 64; regnum->cause = 65; regnum->badvaddr = 66; regnum->hi = 67; regnum->lo = 68; regnum->fp_control_status = 69; regnum->fp_implementation_revision = 70; num_regs = 71; reg_names = mips_irix_reg_names; } else { regnum->lo = MIPS_EMBED_LO_REGNUM; regnum->hi = MIPS_EMBED_HI_REGNUM; regnum->badvaddr = MIPS_EMBED_BADVADDR_REGNUM; regnum->cause = MIPS_EMBED_CAUSE_REGNUM; regnum->pc = MIPS_EMBED_PC_REGNUM; regnum->fp0 = MIPS_EMBED_FP0_REGNUM; regnum->fp_control_status = 70; regnum->fp_implementation_revision = 71; num_regs = 90; if (info.bfd_arch_info != NULL && info.bfd_arch_info->mach == bfd_mach_mips3900) reg_names = mips_tx39_reg_names; else reg_names = mips_generic_reg_names; } /* FIXME: cagney/2003-11-15: For MIPS, hasn't PC_REGNUM been replaced by read_pc? */ set_gdbarch_pc_regnum (gdbarch, regnum->pc + num_regs); set_gdbarch_sp_regnum (gdbarch, MIPS_SP_REGNUM + num_regs); set_gdbarch_fp0_regnum (gdbarch, regnum->fp0); set_gdbarch_num_regs (gdbarch, num_regs); set_gdbarch_num_pseudo_regs (gdbarch, num_regs); set_gdbarch_register_name (gdbarch, mips_register_name); tdep->mips_processor_reg_names = reg_names; tdep->regnum = regnum; } switch (mips_abi) { case MIPS_ABI_O32: set_gdbarch_push_dummy_call (gdbarch, mips_o32_push_dummy_call); set_gdbarch_return_value (gdbarch, mips_o32_return_value); tdep->mips_last_arg_regnum = MIPS_A0_REGNUM + 4 - 1; tdep->mips_last_fp_arg_regnum = tdep->regnum->fp0 + 12 + 4 - 1; tdep->default_mask_address_p = 0; set_gdbarch_long_bit (gdbarch, 32); set_gdbarch_ptr_bit (gdbarch, 32); set_gdbarch_long_long_bit (gdbarch, 64); break; case MIPS_ABI_O64: set_gdbarch_push_dummy_call (gdbarch, mips_o64_push_dummy_call); set_gdbarch_return_value (gdbarch, mips_o64_return_value); tdep->mips_last_arg_regnum = MIPS_A0_REGNUM + 4 - 1; tdep->mips_last_fp_arg_regnum = tdep->regnum->fp0 + 12 + 4 - 1; tdep->default_mask_address_p = 0; set_gdbarch_long_bit (gdbarch, 32); set_gdbarch_ptr_bit (gdbarch, 32); set_gdbarch_long_long_bit (gdbarch, 64); break; case MIPS_ABI_EABI32: set_gdbarch_push_dummy_call (gdbarch, mips_eabi_push_dummy_call); set_gdbarch_return_value (gdbarch, mips_eabi_return_value); tdep->mips_last_arg_regnum = MIPS_A0_REGNUM + 8 - 1; tdep->mips_last_fp_arg_regnum = tdep->regnum->fp0 + 12 + 8 - 1; tdep->default_mask_address_p = 0; set_gdbarch_long_bit (gdbarch, 32); set_gdbarch_ptr_bit (gdbarch, 32); set_gdbarch_long_long_bit (gdbarch, 64); break; case MIPS_ABI_EABI64: set_gdbarch_push_dummy_call (gdbarch, mips_eabi_push_dummy_call); set_gdbarch_return_value (gdbarch, mips_eabi_return_value); tdep->mips_last_arg_regnum = MIPS_A0_REGNUM + 8 - 1; tdep->mips_last_fp_arg_regnum = tdep->regnum->fp0 + 12 + 8 - 1; tdep->default_mask_address_p = 0; set_gdbarch_long_bit (gdbarch, 64); set_gdbarch_ptr_bit (gdbarch, 64); set_gdbarch_long_long_bit (gdbarch, 64); break; case MIPS_ABI_N32: set_gdbarch_push_dummy_call (gdbarch, mips_n32n64_push_dummy_call); set_gdbarch_return_value (gdbarch, mips_n32n64_return_value); tdep->mips_last_arg_regnum = MIPS_A0_REGNUM + 8 - 1; tdep->mips_last_fp_arg_regnum = tdep->regnum->fp0 + 12 + 8 - 1; tdep->default_mask_address_p = 0; set_gdbarch_long_bit (gdbarch, 32); set_gdbarch_ptr_bit (gdbarch, 32); set_gdbarch_long_long_bit (gdbarch, 64); set_gdbarch_long_double_bit (gdbarch, 128); set_gdbarch_long_double_format (gdbarch, floatformats_n32n64_long); break; case MIPS_ABI_N64: set_gdbarch_push_dummy_call (gdbarch, mips_n32n64_push_dummy_call); set_gdbarch_return_value (gdbarch, mips_n32n64_return_value); tdep->mips_last_arg_regnum = MIPS_A0_REGNUM + 8 - 1; tdep->mips_last_fp_arg_regnum = tdep->regnum->fp0 + 12 + 8 - 1; tdep->default_mask_address_p = 0; set_gdbarch_long_bit (gdbarch, 64); set_gdbarch_ptr_bit (gdbarch, 64); set_gdbarch_long_long_bit (gdbarch, 64); set_gdbarch_long_double_bit (gdbarch, 128); set_gdbarch_long_double_format (gdbarch, floatformats_n32n64_long); break; default: internal_error (__FILE__, __LINE__, _("unknown ABI in switch")); } /* GCC creates a pseudo-section whose name specifies the size of longs, since -mlong32 or -mlong64 may be used independent of other options. How those options affect pointer sizes is ABI and architecture dependent, so use them to override the default sizes set by the ABI. This table shows the relationship between ABI, -mlongXX, and size of pointers: ABI -mlongXX ptr bits --- -------- -------- o32 32 32 o32 64 32 n32 32 32 n32 64 64 o64 32 32 o64 64 64 n64 32 32 n64 64 64 eabi32 32 32 eabi32 64 32 eabi64 32 32 eabi64 64 64 Note that for o32 and eabi32, pointers are always 32 bits regardless of any -mlongXX option. For all others, pointers and longs are the same, as set by -mlongXX or set by defaults. */ if (info.abfd != NULL) { int long_bit = 0; bfd_map_over_sections (info.abfd, mips_find_long_section, &long_bit); if (long_bit) { set_gdbarch_long_bit (gdbarch, long_bit); switch (mips_abi) { case MIPS_ABI_O32: case MIPS_ABI_EABI32: break; case MIPS_ABI_N32: case MIPS_ABI_O64: case MIPS_ABI_N64: case MIPS_ABI_EABI64: set_gdbarch_ptr_bit (gdbarch, long_bit); break; default: internal_error (__FILE__, __LINE__, _("unknown ABI in switch")); } } } /* FIXME: jlarmour/2000-04-07: There *is* a flag EF_MIPS_32BIT_MODE that could indicate -gp32 BUT gas/config/tc-mips.c contains the comment: ``We deliberately don't allow "-gp32" to set the MIPS_32BITMODE flag in object files because to do so would make it impossible to link with libraries compiled without "-gp32". This is unnecessarily restrictive. We could solve this problem by adding "-gp32" multilibs to gcc, but to set this flag before gcc is built with such multilibs will break too many systems.'' But even more unhelpfully, the default linker output target for mips64-elf is elf32-bigmips, and has EF_MIPS_32BIT_MODE set, even for 64-bit programs - you need to change the ABI to change this, and not all gcc targets support that currently. Therefore using this flag to detect 32-bit mode would do the wrong thing given the current gcc - it would make GDB treat these 64-bit programs as 32-bit programs by default. */ set_gdbarch_read_pc (gdbarch, mips_read_pc); set_gdbarch_write_pc (gdbarch, mips_write_pc); /* Add/remove bits from an address. The MIPS needs be careful to ensure that all 32 bit addresses are sign extended to 64 bits. */ set_gdbarch_addr_bits_remove (gdbarch, mips_addr_bits_remove); /* Unwind the frame. */ set_gdbarch_unwind_pc (gdbarch, mips_unwind_pc); set_gdbarch_unwind_sp (gdbarch, mips_unwind_sp); set_gdbarch_unwind_dummy_id (gdbarch, mips_unwind_dummy_id); /* Map debug register numbers onto internal register numbers. */ set_gdbarch_stab_reg_to_regnum (gdbarch, mips_stab_reg_to_regnum); set_gdbarch_ecoff_reg_to_regnum (gdbarch, mips_dwarf_dwarf2_ecoff_reg_to_regnum); set_gdbarch_dwarf_reg_to_regnum (gdbarch, mips_dwarf_dwarf2_ecoff_reg_to_regnum); set_gdbarch_dwarf2_reg_to_regnum (gdbarch, mips_dwarf_dwarf2_ecoff_reg_to_regnum); set_gdbarch_register_sim_regno (gdbarch, mips_register_sim_regno); /* MIPS version of CALL_DUMMY */ /* NOTE: cagney/2003-08-05: Eventually call dummy location will be replaced by a command, and all targets will default to on stack (regardless of the stack's execute status). */ set_gdbarch_call_dummy_location (gdbarch, AT_SYMBOL); set_gdbarch_frame_align (gdbarch, mips_frame_align); set_gdbarch_convert_register_p (gdbarch, mips_convert_register_p); set_gdbarch_register_to_value (gdbarch, mips_register_to_value); set_gdbarch_value_to_register (gdbarch, mips_value_to_register); set_gdbarch_inner_than (gdbarch, core_addr_lessthan); set_gdbarch_breakpoint_from_pc (gdbarch, mips_breakpoint_from_pc); set_gdbarch_skip_prologue (gdbarch, mips_skip_prologue); set_gdbarch_pointer_to_address (gdbarch, signed_pointer_to_address); set_gdbarch_address_to_pointer (gdbarch, address_to_signed_pointer); set_gdbarch_integer_to_address (gdbarch, mips_integer_to_address); set_gdbarch_register_type (gdbarch, mips_register_type); set_gdbarch_print_registers_info (gdbarch, mips_print_registers_info); set_gdbarch_print_insn (gdbarch, gdb_print_insn_mips); /* FIXME: cagney/2003-08-29: The macros HAVE_STEPPABLE_WATCHPOINT, HAVE_NONSTEPPABLE_WATCHPOINT, and HAVE_CONTINUABLE_WATCHPOINT need to all be folded into the target vector. Since they are being used as guards for STOPPED_BY_WATCHPOINT, why not have STOPPED_BY_WATCHPOINT return the type of watchpoint that the code is sitting on? */ set_gdbarch_have_nonsteppable_watchpoint (gdbarch, 1); set_gdbarch_skip_trampoline_code (gdbarch, mips_skip_trampoline_code); set_gdbarch_single_step_through_delay (gdbarch, mips_single_step_through_delay); /* Virtual tables. */ set_gdbarch_vbit_in_delta (gdbarch, 1); mips_register_g_packet_guesses (gdbarch); /* Hook in OS ABI-specific overrides, if they have been registered. */ info.tdep_info = (void *) tdesc_data; gdbarch_init_osabi (info, gdbarch); /* Unwind the frame. */ frame_unwind_append_sniffer (gdbarch, dwarf2_frame_sniffer); frame_unwind_append_sniffer (gdbarch, mips_stub_frame_sniffer); frame_unwind_append_sniffer (gdbarch, mips_insn16_frame_sniffer); frame_unwind_append_sniffer (gdbarch, mips_insn32_frame_sniffer); frame_base_append_sniffer (gdbarch, dwarf2_frame_base_sniffer); frame_base_append_sniffer (gdbarch, mips_stub_frame_base_sniffer); frame_base_append_sniffer (gdbarch, mips_insn16_frame_base_sniffer); frame_base_append_sniffer (gdbarch, mips_insn32_frame_base_sniffer); if (tdesc_data) { set_tdesc_pseudo_register_type (gdbarch, mips_pseudo_register_type); tdesc_use_registers (gdbarch, tdesc_data); /* Override the normal target description methods to handle our dual real and pseudo registers. */ set_gdbarch_register_name (gdbarch, mips_register_name); set_gdbarch_register_reggroup_p (gdbarch, mips_tdesc_register_reggroup_p); num_regs = gdbarch_num_regs (gdbarch); set_gdbarch_num_pseudo_regs (gdbarch, num_regs); set_gdbarch_pc_regnum (gdbarch, tdep->regnum->pc + num_regs); set_gdbarch_sp_regnum (gdbarch, MIPS_SP_REGNUM + num_regs); } /* Add ABI-specific aliases for the registers. */ if (mips_abi == MIPS_ABI_N32 || mips_abi == MIPS_ABI_N64) for (i = 0; i < ARRAY_SIZE (mips_n32_n64_aliases); i++) user_reg_add (gdbarch, mips_n32_n64_aliases[i].name, value_of_mips_user_reg, &mips_n32_n64_aliases[i].regnum); else for (i = 0; i < ARRAY_SIZE (mips_o32_aliases); i++) user_reg_add (gdbarch, mips_o32_aliases[i].name, value_of_mips_user_reg, &mips_o32_aliases[i].regnum); /* Add some other standard aliases. */ for (i = 0; i < ARRAY_SIZE (mips_register_aliases); i++) user_reg_add (gdbarch, mips_register_aliases[i].name, value_of_mips_user_reg, &mips_register_aliases[i].regnum); return gdbarch; } static void mips_abi_update (char *ignore_args, int from_tty, struct cmd_list_element *c) { struct gdbarch_info info; /* Force the architecture to update, and (if it's a MIPS architecture) mips_gdbarch_init will take care of the rest. */ gdbarch_info_init (&info); gdbarch_update_p (info); } /* Print out which MIPS ABI is in use. */ static void show_mips_abi (struct ui_file *file, int from_tty, struct cmd_list_element *ignored_cmd, const char *ignored_value) { if (gdbarch_bfd_arch_info (current_gdbarch)->arch != bfd_arch_mips) fprintf_filtered (file, "The MIPS ABI is unknown because the current architecture " "is not MIPS.\n"); else { enum mips_abi global_abi = global_mips_abi (); enum mips_abi actual_abi = mips_abi (current_gdbarch); const char *actual_abi_str = mips_abi_strings[actual_abi]; if (global_abi == MIPS_ABI_UNKNOWN) fprintf_filtered (file, "The MIPS ABI is set automatically (currently \"%s\").\n", actual_abi_str); else if (global_abi == actual_abi) fprintf_filtered (file, "The MIPS ABI is assumed to be \"%s\" (due to user setting).\n", actual_abi_str); else { /* Probably shouldn't happen... */ fprintf_filtered (file, "The (auto detected) MIPS ABI \"%s\" is in use even though the user setting was \"%s\".\n", actual_abi_str, mips_abi_strings[global_abi]); } } } static void mips_dump_tdep (struct gdbarch *current_gdbarch, struct ui_file *file) { struct gdbarch_tdep *tdep = gdbarch_tdep (current_gdbarch); if (tdep != NULL) { int ef_mips_arch; int ef_mips_32bitmode; /* Determine the ISA. */ switch (tdep->elf_flags & EF_MIPS_ARCH) { case E_MIPS_ARCH_1: ef_mips_arch = 1; break; case E_MIPS_ARCH_2: ef_mips_arch = 2; break; case E_MIPS_ARCH_3: ef_mips_arch = 3; break; case E_MIPS_ARCH_4: ef_mips_arch = 4; break; default: ef_mips_arch = 0; break; } /* Determine the size of a pointer. */ ef_mips_32bitmode = (tdep->elf_flags & EF_MIPS_32BITMODE); fprintf_unfiltered (file, "mips_dump_tdep: tdep->elf_flags = 0x%x\n", tdep->elf_flags); fprintf_unfiltered (file, "mips_dump_tdep: ef_mips_32bitmode = %d\n", ef_mips_32bitmode); fprintf_unfiltered (file, "mips_dump_tdep: ef_mips_arch = %d\n", ef_mips_arch); fprintf_unfiltered (file, "mips_dump_tdep: tdep->mips_abi = %d (%s)\n", tdep->mips_abi, mips_abi_strings[tdep->mips_abi]); fprintf_unfiltered (file, "mips_dump_tdep: mips_mask_address_p() %d (default %d)\n", mips_mask_address_p (tdep), tdep->default_mask_address_p); } fprintf_unfiltered (file, "mips_dump_tdep: MIPS_DEFAULT_FPU_TYPE = %d (%s)\n", MIPS_DEFAULT_FPU_TYPE, (MIPS_DEFAULT_FPU_TYPE == MIPS_FPU_NONE ? "none" : MIPS_DEFAULT_FPU_TYPE == MIPS_FPU_SINGLE ? "single" : MIPS_DEFAULT_FPU_TYPE == MIPS_FPU_DOUBLE ? "double" : "???")); fprintf_unfiltered (file, "mips_dump_tdep: MIPS_EABI = %d\n", MIPS_EABI); fprintf_unfiltered (file, "mips_dump_tdep: MIPS_FPU_TYPE = %d (%s)\n", MIPS_FPU_TYPE, (MIPS_FPU_TYPE == MIPS_FPU_NONE ? "none" : MIPS_FPU_TYPE == MIPS_FPU_SINGLE ? "single" : MIPS_FPU_TYPE == MIPS_FPU_DOUBLE ? "double" : "???")); } extern initialize_file_ftype _initialize_mips_tdep; /* -Wmissing-prototypes */ void _initialize_mips_tdep (void) { static struct cmd_list_element *mipsfpulist = NULL; struct cmd_list_element *c; mips_abi_string = mips_abi_strings[MIPS_ABI_UNKNOWN]; if (MIPS_ABI_LAST + 1 != sizeof (mips_abi_strings) / sizeof (mips_abi_strings[0])) internal_error (__FILE__, __LINE__, _("mips_abi_strings out of sync")); gdbarch_register (bfd_arch_mips, mips_gdbarch_init, mips_dump_tdep); mips_pdr_data = register_objfile_data (); /* Add root prefix command for all "set mips"/"show mips" commands */ add_prefix_cmd ("mips", no_class, set_mips_command, _("Various MIPS specific commands."), &setmipscmdlist, "set mips ", 0, &setlist); add_prefix_cmd ("mips", no_class, show_mips_command, _("Various MIPS specific commands."), &showmipscmdlist, "show mips ", 0, &showlist); /* Allow the user to override the ABI. */ add_setshow_enum_cmd ("abi", class_obscure, mips_abi_strings, &mips_abi_string, _("\ Set the MIPS ABI used by this program."), _("\ Show the MIPS ABI used by this program."), _("\ This option can be set to one of:\n\ auto - the default ABI associated with the current binary\n\ o32\n\ o64\n\ n32\n\ n64\n\ eabi32\n\ eabi64"), mips_abi_update, show_mips_abi, &setmipscmdlist, &showmipscmdlist); /* Let the user turn off floating point and set the fence post for heuristic_proc_start. */ add_prefix_cmd ("mipsfpu", class_support, set_mipsfpu_command, _("Set use of MIPS floating-point coprocessor."), &mipsfpulist, "set mipsfpu ", 0, &setlist); add_cmd ("single", class_support, set_mipsfpu_single_command, _("Select single-precision MIPS floating-point coprocessor."), &mipsfpulist); add_cmd ("double", class_support, set_mipsfpu_double_command, _("Select double-precision MIPS floating-point coprocessor."), &mipsfpulist); add_alias_cmd ("on", "double", class_support, 1, &mipsfpulist); add_alias_cmd ("yes", "double", class_support, 1, &mipsfpulist); add_alias_cmd ("1", "double", class_support, 1, &mipsfpulist); add_cmd ("none", class_support, set_mipsfpu_none_command, _("Select no MIPS floating-point coprocessor."), &mipsfpulist); add_alias_cmd ("off", "none", class_support, 1, &mipsfpulist); add_alias_cmd ("no", "none", class_support, 1, &mipsfpulist); add_alias_cmd ("0", "none", class_support, 1, &mipsfpulist); add_cmd ("auto", class_support, set_mipsfpu_auto_command, _("Select MIPS floating-point coprocessor automatically."), &mipsfpulist); add_cmd ("mipsfpu", class_support, show_mipsfpu_command, _("Show current use of MIPS floating-point coprocessor target."), &showlist); /* We really would like to have both "0" and "unlimited" work, but command.c doesn't deal with that. So make it a var_zinteger because the user can always use "999999" or some such for unlimited. */ add_setshow_zinteger_cmd ("heuristic-fence-post", class_support, &heuristic_fence_post, _("\ Set the distance searched for the start of a function."), _("\ Show the distance searched for the start of a function."), _("\ If you are debugging a stripped executable, GDB needs to search through the\n\ program for the start of a function. This command sets the distance of the\n\ search. The only need to set it is when debugging a stripped executable."), reinit_frame_cache_sfunc, NULL, /* FIXME: i18n: The distance searched for the start of a function is %s. */ &setlist, &showlist); /* Allow the user to control whether the upper bits of 64-bit addresses should be zeroed. */ add_setshow_auto_boolean_cmd ("mask-address", no_class, &mask_address_var, _("\ Set zeroing of upper 32 bits of 64-bit addresses."), _("\ Show zeroing of upper 32 bits of 64-bit addresses."), _("\ Use \"on\" to enable the masking, \"off\" to disable it and \"auto\" to \n\ allow GDB to determine the correct value."), NULL, show_mask_address, &setmipscmdlist, &showmipscmdlist); /* Allow the user to control the size of 32 bit registers within the raw remote packet. */ add_setshow_boolean_cmd ("remote-mips64-transfers-32bit-regs", class_obscure, &mips64_transfers_32bit_regs_p, _("\ Set compatibility with 64-bit MIPS target that transfers 32-bit quantities."), _("\ Show compatibility with 64-bit MIPS target that transfers 32-bit quantities."), _("\ Use \"on\" to enable backward compatibility with older MIPS 64 GDB+target\n\ that would transfer 32 bits for some registers (e.g. SR, FSR) and\n\ 64 bits for others. Use \"off\" to disable compatibility mode"), set_mips64_transfers_32bit_regs, NULL, /* FIXME: i18n: Compatibility with 64-bit MIPS target that transfers 32-bit quantities is %s. */ &setlist, &showlist); /* Debug this files internals. */ add_setshow_zinteger_cmd ("mips", class_maintenance, &mips_debug, _("\ Set mips debugging."), _("\ Show mips debugging."), _("\ When non-zero, mips specific debugging is enabled."), NULL, NULL, /* FIXME: i18n: Mips debugging is currently %s. */ &setdebuglist, &showdebuglist); }