/* Native-dependent code for Linux running on i386's, for GDB. 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., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA. */ #include "defs.h" #include "inferior.h" #include "gdbcore.h" /* For i386_linux_skip_solib_resolver. */ #include "symtab.h" #include "frame.h" #include "symfile.h" #include "objfiles.h" #include #include #include #ifdef HAVE_SYS_REG_H #include #endif /* On Linux, threads are implemented as pseudo-processes, in which case we may be tracing more than one process at a time. In that case, inferior_pid will contain the main process ID and the individual thread (process) ID mashed together. These macros are used to separate them out. These definitions should be overridden if thread support is included. */ #if !defined (PIDGET) /* Default definition for PIDGET/TIDGET. */ #define PIDGET(PID) PID #define TIDGET(PID) 0 #endif /* The register sets used in Linux ELF core-dumps are identical to the register sets in `struct user' that is used for a.out core-dumps, and is also used by `ptrace'. The corresponding types are `elf_gregset_t' for the general-purpose registers (with `elf_greg_t' the type of a single GP register) and `elf_fpregset_t' for the floating-point registers. Those types used to be available under the names `gregset_t' and `fpregset_t' too, and this file used those names in the past. But those names are now used for the register sets used in the `mcontext_t' type, and have a different size and layout. */ /* Mapping between the general-purpose registers in `struct user' format and GDB's register array layout. */ static int regmap[] = { EAX, ECX, EDX, EBX, UESP, EBP, ESI, EDI, EIP, EFL, CS, SS, DS, ES, FS, GS }; /* Which ptrace request retrieves which registers? These apply to the corresponding SET requests as well. */ #define GETREGS_SUPPLIES(regno) \ (0 <= (regno) && (regno) <= 15) #define GETFPREGS_SUPPLIES(regno) \ (FP0_REGNUM <= (regno) && (regno) <= LAST_FPU_CTRL_REGNUM) #define GETXFPREGS_SUPPLIES(regno) \ (FP0_REGNUM <= (regno) && (regno) <= MXCSR_REGNUM) /* Does the current host support the GETREGS request? */ int have_ptrace_getregs = #ifdef HAVE_PTRACE_GETREGS 1 #else 0 #endif ; /* Does the current host support the GETXFPREGS request? The header file may or may not define it, and even if it is defined, the kernel will return EIO if it's running on a pre-SSE processor. PTRACE_GETXFPREGS is a Cygnus invention, since we wrote our own Linux kernel patch for SSE support. That patch may or may not actually make it into the official distribution. If you find that years have gone by since this stuff was added, and Linux isn't using PTRACE_GETXFPREGS, that means that our patch didn't make it, and you can delete this, and the related code. My instinct is to attach this to some architecture- or target-specific data structure, but really, a particular GDB process can only run on top of one kernel at a time. So it's okay for this to be a simple variable. */ int have_ptrace_getxfpregs = #ifdef HAVE_PTRACE_GETXFPREGS 1 #else 0 #endif ; /* Fetching registers directly from the U area, one at a time. */ /* FIXME: kettenis/2000-03-05: This duplicates code from `inptrace.c'. The problem is that we define FETCH_INFERIOR_REGISTERS since we want to use our own versions of {fetch,store}_inferior_registers that use the GETREGS request. This means that the code in `infptrace.c' is #ifdef'd out. But we need to fall back on that code when GDB is running on top of a kernel that doesn't support the GETREGS request. I want to avoid changing `infptrace.c' right now. */ /* Default the type of the ptrace transfer to int. */ #ifndef PTRACE_XFER_TYPE #define PTRACE_XFER_TYPE int #endif /* Registers we shouldn't try to fetch. */ #if !defined (CANNOT_FETCH_REGISTER) #define CANNOT_FETCH_REGISTER(regno) 0 #endif /* Fetch one register. */ static void fetch_register (regno) int regno; { /* This isn't really an address. But ptrace thinks of it as one. */ CORE_ADDR regaddr; char mess[128]; /* For messages */ register int i; unsigned int offset; /* Offset of registers within the u area. */ char buf[MAX_REGISTER_RAW_SIZE]; int tid; if (CANNOT_FETCH_REGISTER (regno)) { memset (buf, '\0', REGISTER_RAW_SIZE (regno)); /* Supply zeroes */ supply_register (regno, buf); return; } /* Overload thread id onto process id */ if ((tid = TIDGET (inferior_pid)) == 0) tid = inferior_pid; /* no thread id, just use process id */ offset = U_REGS_OFFSET; regaddr = register_addr (regno, offset); for (i = 0; i < REGISTER_RAW_SIZE (regno); i += sizeof (PTRACE_XFER_TYPE)) { errno = 0; *(PTRACE_XFER_TYPE *) & buf[i] = ptrace (PT_READ_U, tid, (PTRACE_ARG3_TYPE) regaddr, 0); regaddr += sizeof (PTRACE_XFER_TYPE); if (errno != 0) { sprintf (mess, "reading register %s (#%d)", REGISTER_NAME (regno), regno); perror_with_name (mess); } } supply_register (regno, buf); } /* Fetch register values from the inferior. If REGNO is negative, do this for all registers. Otherwise, REGNO specifies which register (so we can save time). */ void old_fetch_inferior_registers (regno) int regno; { if (regno >= 0) { fetch_register (regno); } else { for (regno = 0; regno < ARCH_NUM_REGS; regno++) { fetch_register (regno); } } } /* Registers we shouldn't try to store. */ #if !defined (CANNOT_STORE_REGISTER) #define CANNOT_STORE_REGISTER(regno) 0 #endif /* Store one register. */ static void store_register (regno) int regno; { /* This isn't really an address. But ptrace thinks of it as one. */ CORE_ADDR regaddr; char mess[128]; /* For messages */ register int i; unsigned int offset; /* Offset of registers within the u area. */ int tid; if (CANNOT_STORE_REGISTER (regno)) { return; } /* Overload thread id onto process id */ if ((tid = TIDGET (inferior_pid)) == 0) tid = inferior_pid; /* no thread id, just use process id */ offset = U_REGS_OFFSET; regaddr = register_addr (regno, offset); for (i = 0; i < REGISTER_RAW_SIZE (regno); i += sizeof (PTRACE_XFER_TYPE)) { errno = 0; ptrace (PT_WRITE_U, tid, (PTRACE_ARG3_TYPE) regaddr, *(PTRACE_XFER_TYPE *) & registers[REGISTER_BYTE (regno) + i]); regaddr += sizeof (PTRACE_XFER_TYPE); if (errno != 0) { sprintf (mess, "writing register %s (#%d)", REGISTER_NAME (regno), regno); perror_with_name (mess); } } } /* Store our register values back into the inferior. If REGNO is negative, do this for all registers. Otherwise, REGNO specifies which register (so we can save time). */ void old_store_inferior_registers (regno) int regno; { if (regno >= 0) { store_register (regno); } else { for (regno = 0; regno < ARCH_NUM_REGS; regno++) { store_register (regno); } } } /* Transfering the general-purpose registers between GDB, inferiors and core files. */ /* Fill GDB's register array with the genereal-purpose register values in *GREGSETP. */ void supply_gregset (elf_gregset_t *gregsetp) { elf_greg_t *regp = (elf_greg_t *) gregsetp; int regi; for (regi = 0; regi < NUM_GREGS; regi++) supply_register (regi, (char *) (regp + regmap[regi])); } /* Convert the valid general-purpose register values in GDB's register array to `struct user' format and store them in *GREGSETP. The array VALID indicates which register values are valid. If VALID is NULL, all registers are assumed to be valid. */ static void convert_to_gregset (elf_gregset_t *gregsetp, signed char *valid) { elf_greg_t *regp = (elf_greg_t *) gregsetp; int regi; for (regi = 0; regi < NUM_GREGS; regi++) if (! valid || valid[regi]) *(regp + regmap[regi]) = * (int *) ®isters[REGISTER_BYTE (regi)]; } /* Fill register REGNO (if it is a general-purpose register) in *GREGSETPS with the value in GDB's register array. If REGNO is -1, do this for all registers. */ void fill_gregset (elf_gregset_t *gregsetp, int regno) { if (regno == -1) { convert_to_gregset (gregsetp, NULL); return; } if (GETREGS_SUPPLIES (regno)) { signed char valid[NUM_GREGS]; memset (valid, 0, sizeof (valid)); valid[regno] = 1; convert_to_gregset (gregsetp, valid); } } #ifdef HAVE_PTRACE_GETREGS /* Fetch all general-purpose registers from process/thread TID and store their values in GDB's register array. */ static void fetch_regs (int tid) { elf_gregset_t regs; int ret; ret = ptrace (PTRACE_GETREGS, tid, 0, (int) ®s); if (ret < 0) { if (errno == EIO) { /* The kernel we're running on doesn't support the GETREGS request. Reset `have_ptrace_getregs'. */ have_ptrace_getregs = 0; return; } warning ("Couldn't get registers."); return; } supply_gregset (®s); } /* Store all valid general-purpose registers in GDB's register array into the process/thread specified by TID. */ static void store_regs (int tid) { elf_gregset_t regs; int ret; ret = ptrace (PTRACE_GETREGS, tid, 0, (int) ®s); if (ret < 0) { warning ("Couldn't get registers."); return; } convert_to_gregset (®s, register_valid); ret = ptrace (PTRACE_SETREGS, tid, 0, (int) ®s); if (ret < 0) { warning ("Couldn't write registers."); return; } } #else static void fetch_regs (int tid) {} static void store_regs (int tid) {} #endif /* Transfering floating-point registers between GDB, inferiors and cores. */ /* What is the address of st(N) within the floating-point register set F? */ #define FPREG_ADDR(f, n) ((char *) &(f)->st_space + (n) * 10) /* Fill GDB's register array with the floating-point register values in *FPREGSETP. */ void supply_fpregset (elf_fpregset_t *fpregsetp) { int reg; long l; /* Supply the floating-point registers. */ for (reg = 0; reg < 8; reg++) supply_register (FP0_REGNUM + reg, FPREG_ADDR (fpregsetp, reg)); /* We have to mask off the reserved bits in *FPREGSETP before storing the values in GDB's register file. */ #define supply(REGNO, MEMBER) \ l = fpregsetp->MEMBER & 0xffff; \ supply_register (REGNO, (char *) &l) supply (FCTRL_REGNUM, cwd); supply (FSTAT_REGNUM, swd); supply (FTAG_REGNUM, twd); supply_register (FCOFF_REGNUM, (char *) &fpregsetp->fip); supply (FDS_REGNUM, fos); supply_register (FDOFF_REGNUM, (char *) &fpregsetp->foo); #undef supply /* Extract the code segment and opcode from the "fcs" member. */ l = fpregsetp->fcs & 0xffff; supply_register (FCS_REGNUM, (char *) &l); l = (fpregsetp->fcs >> 16) & ((1 << 11) - 1); supply_register (FOP_REGNUM, (char *) &l); } /* Convert the valid floating-point register values in GDB's register array to `struct user' format and store them in *FPREGSETP. The array VALID indicates which register values are valid. If VALID is NULL, all registers are assumed to be valid. */ static void convert_to_fpregset (elf_fpregset_t *fpregsetp, signed char *valid) { int reg; /* Fill in the floating-point registers. */ for (reg = 0; reg < 8; reg++) if (!valid || valid[reg]) memcpy (FPREG_ADDR (fpregsetp, reg), ®isters[REGISTER_BYTE (FP0_REGNUM + reg)], REGISTER_RAW_SIZE(FP0_REGNUM + reg)); /* We're not supposed to touch the reserved bits in *FPREGSETP. */ #define fill(MEMBER, REGNO) \ if (! valid || valid[(REGNO)]) \ fpregsetp->MEMBER \ = ((fpregsetp->MEMBER & ~0xffff) \ | (* (int *) ®isters[REGISTER_BYTE (REGNO)] & 0xffff)) #define fill_register(MEMBER, REGNO) \ if (! valid || valid[(REGNO)]) \ memcpy (&fpregsetp->MEMBER, ®isters[REGISTER_BYTE (REGNO)], \ sizeof (fpregsetp->MEMBER)) fill (cwd, FCTRL_REGNUM); fill (swd, FSTAT_REGNUM); fill (twd, FTAG_REGNUM); fill_register (fip, FCOFF_REGNUM); fill (foo, FDOFF_REGNUM); fill_register (fos, FDS_REGNUM); #undef fill #undef fill_register if (! valid || valid[FCS_REGNUM]) fpregsetp->fcs = ((fpregsetp->fcs & ~0xffff) | (* (int *) ®isters[REGISTER_BYTE (FCS_REGNUM)] & 0xffff)); if (! valid || valid[FOP_REGNUM]) fpregsetp->fcs = ((fpregsetp->fcs & 0xffff) | ((*(int *) ®isters[REGISTER_BYTE (FOP_REGNUM)] & ((1 << 11) - 1)) << 16)); } /* Fill register REGNO (if it is a floating-point register) in *FPREGSETP with the value in GDB's register array. If REGNO is -1, do this for all registers. */ void fill_fpregset (elf_fpregset_t *fpregsetp, int regno) { if (regno == -1) { convert_to_fpregset (fpregsetp, NULL); return; } if (GETFPREGS_SUPPLIES(regno)) { signed char valid[MAX_NUM_REGS]; memset (valid, 0, sizeof (valid)); valid[regno] = 1; convert_to_fpregset (fpregsetp, valid); } } #ifdef HAVE_PTRACE_GETREGS /* Fetch all floating-point registers from process/thread TID and store thier values in GDB's register array. */ static void fetch_fpregs (int tid) { elf_fpregset_t fpregs; int ret; ret = ptrace (PTRACE_GETFPREGS, tid, 0, (int) &fpregs); if (ret < 0) { warning ("Couldn't get floating point status."); return; } supply_fpregset (&fpregs); } /* Store all valid floating-point registers in GDB's register array into the process/thread specified by TID. */ static void store_fpregs (int tid) { elf_fpregset_t fpregs; int ret; ret = ptrace (PTRACE_GETFPREGS, tid, 0, (int) &fpregs); if (ret < 0) { warning ("Couldn't get floating point status."); return; } convert_to_fpregset (&fpregs, register_valid); ret = ptrace (PTRACE_SETFPREGS, tid, 0, (int) &fpregs); if (ret < 0) { warning ("Couldn't write floating point status."); return; } } #else static void fetch_fpregs (int tid) {} static void store_fpregs (int tid) {} #endif /* Transfering floating-point and SSE registers to and from GDB. */ /* PTRACE_GETXFPREGS is a Cygnus invention, since we wrote our own Linux kernel patch for SSE support. That patch may or may not actually make it into the official distribution. If you find that years have gone by since this code was added, and Linux isn't using PTRACE_GETXFPREGS, that means that our patch didn't make it, and you can delete this code. */ #ifdef HAVE_PTRACE_GETXFPREGS /* Fill GDB's register array with the floating-point and SSE register values in *XFPREGS. */ static void supply_xfpregset (struct user_xfpregs_struct *xfpregs) { int reg; /* Supply the floating-point registers. */ for (reg = 0; reg < 8; reg++) supply_register (FP0_REGNUM + reg, (char *) &xfpregs->st_space[reg]); { supply_register (FCTRL_REGNUM, (char *) &xfpregs->cwd); supply_register (FSTAT_REGNUM, (char *) &xfpregs->swd); supply_register (FTAG_REGNUM, (char *) &xfpregs->twd); supply_register (FCOFF_REGNUM, (char *) &xfpregs->fip); supply_register (FDS_REGNUM, (char *) &xfpregs->fos); supply_register (FDOFF_REGNUM, (char *) &xfpregs->foo); /* Extract the code segment and opcode from the "fcs" member. */ { long l; l = xfpregs->fcs & 0xffff; supply_register (FCS_REGNUM, (char *) &l); l = (xfpregs->fcs >> 16) & ((1 << 11) - 1); supply_register (FOP_REGNUM, (char *) &l); } } /* Supply the SSE registers. */ for (reg = 0; reg < 8; reg++) supply_register (XMM0_REGNUM + reg, (char *) &xfpregs->xmm_space[reg]); supply_register (MXCSR_REGNUM, (char *) &xfpregs->mxcsr); } /* Convert the valid floating-point and SSE registers in GDB's register array to `struct user' format and store them in *XFPREGS. The array VALID indicates which registers are valid. If VALID is NULL, all registers are assumed to be valid. */ static void convert_to_xfpregset (struct user_xfpregs_struct *xfpregs, signed char *valid) { int reg; /* Fill in the floating-point registers. */ for (reg = 0; reg < 8; reg++) if (!valid || valid[reg]) memcpy (&xfpregs->st_space[reg], ®isters[REGISTER_BYTE (FP0_REGNUM + reg)], REGISTER_RAW_SIZE(FP0_REGNUM + reg)); #define fill(MEMBER, REGNO) \ if (! valid || valid[(REGNO)]) \ memcpy (&xfpregs->MEMBER, ®isters[REGISTER_BYTE (REGNO)], \ sizeof (xfpregs->MEMBER)) fill (cwd, FCTRL_REGNUM); fill (swd, FSTAT_REGNUM); fill (twd, FTAG_REGNUM); fill (fip, FCOFF_REGNUM); fill (foo, FDOFF_REGNUM); fill (fos, FDS_REGNUM); #undef fill if (! valid || valid[FCS_REGNUM]) xfpregs->fcs = ((xfpregs->fcs & ~0xffff) | (* (int *) ®isters[REGISTER_BYTE (FCS_REGNUM)] & 0xffff)); if (! valid || valid[FOP_REGNUM]) xfpregs->fcs = ((xfpregs->fcs & 0xffff) | ((*(int *) ®isters[REGISTER_BYTE (FOP_REGNUM)] & ((1 << 11) - 1)) << 16)); /* Fill in the XMM registers. */ for (reg = 0; reg < 8; reg++) if (! valid || valid[reg]) memcpy (&xfpregs->xmm_space[reg], ®isters[REGISTER_BYTE (XMM0_REGNUM + reg)], REGISTER_RAW_SIZE (XMM0_REGNUM + reg)); } /* Fetch all registers covered by the PTRACE_SETXFPREGS request from process/thread TID and store their values in GDB's register array. Return non-zero if successful, zero otherwise. */ static int fetch_xfpregs (int tid) { struct user_xfpregs_struct xfpregs; int ret; if (! have_ptrace_getxfpregs) return 0; ret = ptrace (PTRACE_GETXFPREGS, tid, 0, &xfpregs); if (ret == -1) { if (errno == EIO) { have_ptrace_getxfpregs = 0; return 0; } warning ("Couldn't read floating-point and SSE registers."); return 0; } supply_xfpregset (&xfpregs); return 1; } /* Store all valid registers in GDB's register array covered by the PTRACE_SETXFPREGS request into the process/thread specified by TID. Return non-zero if successful, zero otherwise. */ static int store_xfpregs (int tid) { struct user_xfpregs_struct xfpregs; int ret; if (! have_ptrace_getxfpregs) return 0; ret = ptrace (PTRACE_GETXFPREGS, tid, 0, &xfpregs); if (ret == -1) { if (errno == EIO) { have_ptrace_getxfpregs = 0; return 0; } warning ("Couldn't read floating-point and SSE registers."); return 0; } convert_to_xfpregset (&xfpregs, register_valid); if (ptrace (PTRACE_SETXFPREGS, tid, 0, &xfpregs) < 0) { warning ("Couldn't write floating-point and SSE registers."); return 0; } return 1; } /* Fill the XMM registers in the register array with dummy values. For cases where we don't have access to the XMM registers. I think this is cleaner than printing a warning. For a cleaner solution, we should gdbarchify the i386 family. */ static void dummy_sse_values (void) { /* C doesn't have a syntax for NaN's, so write it out as an array of longs. */ static long dummy[4] = { 0xffffffff, 0xffffffff, 0xffffffff, 0xffffffff }; static long mxcsr = 0x1f80; int reg; for (reg = 0; reg < 8; reg++) supply_register (XMM0_REGNUM + reg, (char *) dummy); supply_register (MXCSR_REGNUM, (char *) &mxcsr); } #else /* Stub versions of the above routines, for systems that don't have PTRACE_GETXFPREGS. */ static int store_xfpregs (int tid) { return 0; } static int fetch_xfpregs (int tid) { return 0; } static void dummy_sse_values (void) {} #endif /* Transferring arbitrary registers between GDB and inferior. */ /* Fetch register REGNO from the child process. If REGNO is -1, do this for all registers (including the floating point and SSE registers). */ void fetch_inferior_registers (int regno) { int tid; /* Use the old method of peeking around in `struct user' if the GETREGS request isn't available. */ if (! have_ptrace_getregs) { old_fetch_inferior_registers (regno); return; } /* Linux LWP ID's are process ID's. */ if ((tid = TIDGET (inferior_pid)) == 0) tid = inferior_pid; /* Not a threaded program. */ /* Use the PTRACE_GETXFPREGS request whenever possible, since it transfers more registers in one system call, and we'll cache the results. But remember that fetch_xfpregs can fail, and return zero. */ if (regno == -1) { fetch_regs (tid); /* The call above might reset `have_ptrace_getregs'. */ if (! have_ptrace_getregs) { old_fetch_inferior_registers (-1); return; } if (fetch_xfpregs (tid)) return; fetch_fpregs (tid); return; } if (GETREGS_SUPPLIES (regno)) { fetch_regs (tid); return; } if (GETXFPREGS_SUPPLIES (regno)) { if (fetch_xfpregs (tid)) return; /* Either our processor or our kernel doesn't support the SSE registers, so read the FP registers in the traditional way, and fill the SSE registers with dummy values. It would be more graceful to handle differences in the register set using gdbarch. Until then, this will at least make things work plausibly. */ fetch_fpregs (tid); dummy_sse_values (); return; } internal_error ("i386-linux-nat.c (fetch_inferior_registers): " "got request for bad register number %d", regno); } /* Store register REGNO back into the child process. If REGNO is -1, do this for all registers (including the floating point and SSE registers). */ void store_inferior_registers (int regno) { int tid; /* Use the old method of poking around in `struct user' if the SETREGS request isn't available. */ if (! have_ptrace_getregs) { old_store_inferior_registers (regno); return; } /* Linux LWP ID's are process ID's. */ if ((tid = TIDGET (inferior_pid)) == 0) tid = inferior_pid; /* Not a threaded program. */ /* Use the PTRACE_SETXFPREGS requests whenever possibl, since it transfers more registers in one system call. But remember that store_xfpregs can fail, and return zero. */ if (regno == -1) { store_regs (tid); if (store_xfpregs (tid)) return; store_fpregs (tid); return; } if (GETREGS_SUPPLIES (regno)) { store_regs (tid); return; } if (GETXFPREGS_SUPPLIES (regno)) { if (store_xfpregs (tid)) return; /* Either our processor or our kernel doesn't support the SSE registers, so just write the FP registers in the traditional way. */ store_fpregs (tid); return; } internal_error ("Got request to store bad register number %d.", regno); } /* Interpreting register set info found in core files. */ /* Provide registers to GDB from a core file. (We can't use the generic version of this function in core-regset.c, because Linux has *three* different kinds of register set notes. core-regset.c would have to call supply_xfpregset, which most platforms don't have.) CORE_REG_SECT points to an array of bytes, which are the contents of a `note' from a core file which BFD thinks might contain register contents. CORE_REG_SIZE is its size. WHICH says which register set corelow suspects this is: 0 --- the general-purpose register set, in elf_gregset_t format 2 --- the floating-point register set, in elf_fpregset_t format 3 --- the extended floating-point register set, in struct user_xfpregs_struct format REG_ADDR isn't used on Linux. */ static void fetch_core_registers (char *core_reg_sect, unsigned core_reg_size, int which, CORE_ADDR reg_addr) { elf_gregset_t gregset; elf_fpregset_t fpregset; switch (which) { case 0: if (core_reg_size != sizeof (gregset)) warning ("Wrong size gregset in core file."); else { memcpy (&gregset, core_reg_sect, sizeof (gregset)); supply_gregset (&gregset); } break; case 2: if (core_reg_size != sizeof (fpregset)) warning ("Wrong size fpregset in core file."); else { memcpy (&fpregset, core_reg_sect, sizeof (fpregset)); supply_fpregset (&fpregset); } break; #ifdef HAVE_PTRACE_GETXFPREGS { struct user_xfpregs_struct xfpregset; case 3: if (core_reg_size != sizeof (xfpregset)) warning ("Wrong size user_xfpregs_struct in core file."); else { memcpy (&xfpregset, core_reg_sect, sizeof (xfpregset)); supply_xfpregset (&xfpregset); } break; } #endif default: /* We've covered all the kinds of registers we know about here, so this must be something we wouldn't know what to do with anyway. Just ignore it. */ break; } } /* Calling functions in shared libraries. */ /* FIXME: kettenis/2000-03-05: Doesn't this belong in a target-dependent file? The function `i386_linux_skip_solib_resolver' is mentioned in `config/i386/tm-linux.h'. */ /* Find the minimal symbol named NAME, and return both the minsym struct and its objfile. This probably ought to be in minsym.c, but everything there is trying to deal with things like C++ and SOFUN_ADDRESS_MAYBE_TURQUOISE, ... Since this is so simple, it may be considered too special-purpose for general consumption. */ static struct minimal_symbol * find_minsym_and_objfile (char *name, struct objfile **objfile_p) { struct objfile *objfile; ALL_OBJFILES (objfile) { struct minimal_symbol *msym; ALL_OBJFILE_MSYMBOLS (objfile, msym) { if (SYMBOL_NAME (msym) && STREQ (SYMBOL_NAME (msym), name)) { *objfile_p = objfile; return msym; } } } return 0; } static CORE_ADDR skip_hurd_resolver (CORE_ADDR pc) { /* The HURD dynamic linker is part of the GNU C library, so many GNU/Linux distributions use it. (All ELF versions, as far as I know.) An unresolved PLT entry points to "_dl_runtime_resolve", which calls "fixup" to patch the PLT, and then passes control to the function. We look for the symbol `_dl_runtime_resolve', and find `fixup' in the same objfile. If we are at the entry point of `fixup', then we set a breakpoint at the return address (at the top of the stack), and continue. It's kind of gross to do all these checks every time we're called, since they don't change once the executable has gotten started. But this is only a temporary hack --- upcoming versions of Linux will provide a portable, efficient interface for debugging programs that use shared libraries. */ struct objfile *objfile; struct minimal_symbol *resolver = find_minsym_and_objfile ("_dl_runtime_resolve", &objfile); if (resolver) { struct minimal_symbol *fixup = lookup_minimal_symbol ("fixup", 0, objfile); if (fixup && SYMBOL_VALUE_ADDRESS (fixup) == pc) return (SAVED_PC_AFTER_CALL (get_current_frame ())); } return 0; } /* See the comments for SKIP_SOLIB_RESOLVER at the top of infrun.c. This function: 1) decides whether a PLT has sent us into the linker to resolve a function reference, and 2) if so, tells us where to set a temporary breakpoint that will trigger when the dynamic linker is done. */ CORE_ADDR i386_linux_skip_solib_resolver (CORE_ADDR pc) { CORE_ADDR result; /* Plug in functions for other kinds of resolvers here. */ result = skip_hurd_resolver (pc); if (result) return result; return 0; } /* Recognizing signal handler frames. */ /* Linux has two flavors of signals. Normal signal handlers, and "realtime" (RT) signals. The RT signals can provide additional information to the signal handler if the SA_SIGINFO flag is set when establishing a signal handler using `sigaction'. It is not unlikely that future versions of Linux will support SA_SIGINFO for normal signals too. */ /* When the i386 Linux kernel calls a signal handler and the SA_RESTORER flag isn't set, the return address points to a bit of code on the stack. This function returns whether the PC appears to be within this bit of code. The instruction sequence for normal signals is pop %eax mov $0x77,%eax int $0x80 or 0x58 0xb8 0x77 0x00 0x00 0x00 0xcd 0x80. Checking for the code sequence should be somewhat reliable, because the effect is to call the system call sigreturn. This is unlikely to occur anywhere other than a signal trampoline. It kind of sucks that we have to read memory from the process in order to identify a signal trampoline, but there doesn't seem to be any other way. The IN_SIGTRAMP macro in tm-linux.h arranges to only call us if no function name could be identified, which should be the case since the code is on the stack. Detection of signal trampolines for handlers that set the SA_RESTORER flag is in general not possible. Unfortunately this is what the GNU C Library has been doing for quite some time now. However, as of version 2.1.2, the GNU C Library uses signal trampolines (named __restore and __restore_rt) that are identical to the ones used by the kernel. Therefore, these trampolines are supported too. */ #define LINUX_SIGTRAMP_INSN0 (0x58) /* pop %eax */ #define LINUX_SIGTRAMP_OFFSET0 (0) #define LINUX_SIGTRAMP_INSN1 (0xb8) /* mov $NNNN,%eax */ #define LINUX_SIGTRAMP_OFFSET1 (1) #define LINUX_SIGTRAMP_INSN2 (0xcd) /* int */ #define LINUX_SIGTRAMP_OFFSET2 (6) static const unsigned char linux_sigtramp_code[] = { LINUX_SIGTRAMP_INSN0, /* pop %eax */ LINUX_SIGTRAMP_INSN1, 0x77, 0x00, 0x00, 0x00, /* mov $0x77,%eax */ LINUX_SIGTRAMP_INSN2, 0x80 /* int $0x80 */ }; #define LINUX_SIGTRAMP_LEN (sizeof linux_sigtramp_code) /* If PC is in a sigtramp routine, return the address of the start of the routine. Otherwise, return 0. */ static CORE_ADDR i386_linux_sigtramp_start (CORE_ADDR pc) { unsigned char buf[LINUX_SIGTRAMP_LEN]; /* We only recognize a signal trampoline if PC is at the start of one of the three instructions. We optimize for finding the PC at the start, as will be the case when the trampoline is not the first frame on the stack. We assume that in the case where the PC is not at the start of the instruction sequence, there will be a few trailing readable bytes on the stack. */ if (read_memory_nobpt (pc, (char *) buf, LINUX_SIGTRAMP_LEN) != 0) return 0; if (buf[0] != LINUX_SIGTRAMP_INSN0) { int adjust; switch (buf[0]) { case LINUX_SIGTRAMP_INSN1: adjust = LINUX_SIGTRAMP_OFFSET1; break; case LINUX_SIGTRAMP_INSN2: adjust = LINUX_SIGTRAMP_OFFSET2; break; default: return 0; } pc -= adjust; if (read_memory_nobpt (pc, (char *) buf, LINUX_SIGTRAMP_LEN) != 0) return 0; } if (memcmp (buf, linux_sigtramp_code, LINUX_SIGTRAMP_LEN) != 0) return 0; return pc; } /* This function does the same for RT signals. Here the instruction sequence is mov $0xad,%eax int $0x80 or 0xb8 0xad 0x00 0x00 0x00 0xcd 0x80. The effect is to call the system call rt_sigreturn. */ #define LINUX_RT_SIGTRAMP_INSN0 (0xb8) /* mov $NNNN,%eax */ #define LINUX_RT_SIGTRAMP_OFFSET0 (0) #define LINUX_RT_SIGTRAMP_INSN1 (0xcd) /* int */ #define LINUX_RT_SIGTRAMP_OFFSET1 (5) static const unsigned char linux_rt_sigtramp_code[] = { LINUX_RT_SIGTRAMP_INSN0, 0xad, 0x00, 0x00, 0x00, /* mov $0xad,%eax */ LINUX_RT_SIGTRAMP_INSN1, 0x80 /* int $0x80 */ }; #define LINUX_RT_SIGTRAMP_LEN (sizeof linux_rt_sigtramp_code) /* If PC is in a RT sigtramp routine, return the address of the start of the routine. Otherwise, return 0. */ static CORE_ADDR i386_linux_rt_sigtramp_start (CORE_ADDR pc) { unsigned char buf[LINUX_RT_SIGTRAMP_LEN]; /* We only recognize a signal trampoline if PC is at the start of one of the two instructions. We optimize for finding the PC at the start, as will be the case when the trampoline is not the first frame on the stack. We assume that in the case where the PC is not at the start of the instruction sequence, there will be a few trailing readable bytes on the stack. */ if (read_memory_nobpt (pc, (char *) buf, LINUX_RT_SIGTRAMP_LEN) != 0) return 0; if (buf[0] != LINUX_RT_SIGTRAMP_INSN0) { if (buf[0] != LINUX_RT_SIGTRAMP_INSN1) return 0; pc -= LINUX_RT_SIGTRAMP_OFFSET1; if (read_memory_nobpt (pc, (char *) buf, LINUX_RT_SIGTRAMP_LEN) != 0) return 0; } if (memcmp (buf, linux_rt_sigtramp_code, LINUX_RT_SIGTRAMP_LEN) != 0) return 0; return pc; } /* Return whether PC is in a Linux sigtramp routine. */ int i386_linux_in_sigtramp (CORE_ADDR pc, char *name) { if (name) return STREQ ("__restore", name) || STREQ ("__restore_rt", name); return (i386_linux_sigtramp_start (pc) != 0 || i386_linux_rt_sigtramp_start (pc) != 0); } /* Assuming FRAME is for a Linux sigtramp routine, return the address of the associated sigcontext structure. */ CORE_ADDR i386_linux_sigcontext_addr (struct frame_info *frame) { CORE_ADDR pc; pc = i386_linux_sigtramp_start (frame->pc); if (pc) { CORE_ADDR sp; if (frame->next) /* If this isn't the top frame, the next frame must be for the signal handler itself. The sigcontext structure lives on the stack, right after the signum argument. */ return frame->next->frame + 12; /* This is the top frame. We'll have to find the address of the sigcontext structure by looking at the stack pointer. Keep in mind that the first instruction of the sigtramp code is "pop %eax". If the PC is at this instruction, adjust the returned value accordingly. */ sp = read_register (SP_REGNUM); if (pc == frame->pc) return sp + 4; return sp; } pc = i386_linux_rt_sigtramp_start (frame->pc); if (pc) { if (frame->next) /* If this isn't the top frame, the next frame must be for the signal handler itself. The sigcontext structure is part of the user context. A pointer to the user context is passed as the third argument to the signal handler. */ return read_memory_integer (frame->next->frame + 16, 4) + 20; /* This is the top frame. Again, use the stack pointer to find the address of the sigcontext structure. */ return read_memory_integer (read_register (SP_REGNUM) + 8, 4) + 20; } error ("Couldn't recognize signal trampoline."); return 0; } /* Offset to saved PC in sigcontext, from . */ #define LINUX_SIGCONTEXT_PC_OFFSET (56) /* Assuming FRAME is for a Linux sigtramp routine, return the saved program counter. */ CORE_ADDR i386_linux_sigtramp_saved_pc (struct frame_info *frame) { CORE_ADDR addr; addr = i386_linux_sigcontext_addr (frame); return read_memory_integer (addr + LINUX_SIGCONTEXT_PC_OFFSET, 4); } /* Offset to saved SP in sigcontext, from . */ #define LINUX_SIGCONTEXT_SP_OFFSET (28) /* Assuming FRAME is for a Linux sigtramp routine, return the saved stack pointer. */ CORE_ADDR i386_linux_sigtramp_saved_sp (struct frame_info *frame) { CORE_ADDR addr; addr = i386_linux_sigcontext_addr (frame); return read_memory_integer (addr + LINUX_SIGCONTEXT_SP_OFFSET, 4); } /* Immediately after a function call, return the saved pc. */ CORE_ADDR i386_linux_saved_pc_after_call (struct frame_info *frame) { if (frame->signal_handler_caller) return i386_linux_sigtramp_saved_pc (frame); return read_memory_integer (read_register (SP_REGNUM), 4); } /* Register that we are able to handle Linux ELF core file formats. */ static struct core_fns linux_elf_core_fns = { bfd_target_elf_flavour, /* core_flavour */ default_check_format, /* check_format */ default_core_sniffer, /* core_sniffer */ fetch_core_registers, /* core_read_registers */ NULL /* next */ }; void _initialize_i386_linux_nat () { add_core_fns (&linux_elf_core_fns); }