mirror/binutils-gdb
2
0
Fork 0
mirror of https://sourceware.org/git/binutils-gdb.git synced 2025-01-06 12:09:26 +08:00
Code Issues Packages Projects Releases Wiki Activity
3b3e209011
binutils-gdb/gdb/gdbarch-gen.h

1754 lines
103 KiB
C
Raw Normal View History

Remove old GNU indent directives Now that gdb_indent.sh has been removed, I think it makes sense to also remove the directives intended for GNU indent.
2023-02-09 02:26:24 +08:00
/* THIS FILE IS GENERATED -*- buffer-read-only: t -*- */
Split gdbarch.h into two files This patch splits gdbarch.h into two files -- gdbarch.h now is editable and hand-maintained, and the new gdbarch-gen.h file is the only thing generated by gdbarch.sh. This lets us avoid maintaining boilerplate in the gdbarch.sh file. Note that gdbarch.sh still generates gdbarch.h after this patch. This makes it easier to re-run when rebasing. This code is removed in a subsequent patch.
2021-12-15 07:48:45 +08:00
/* vi:set ro: */
/* Dynamic architecture support for GDB, the GNU debugger.
Update copyright year range in header of all files managed by GDB This commit is the result of the following actions: - Running gdb/copyright.py to update all of the copyright headers to include 2024, - Manually updating a few files the copyright.py script told me to update, these files had copyright headers embedded within the file, - Regenerating gdbsupport/Makefile.in to refresh it's copyright date, - Using grep to find other files that still mentioned 2023. If these files were updated last year from 2022 to 2023 then I've updated them this year to 2024. I'm sure I've probably missed some dates. Feel free to fix them up as you spot them.
2024-01-12 23:30:44 +08:00
Copyright (C) 1998-2024 Free Software Foundation, Inc.
Split gdbarch.h into two files This patch splits gdbarch.h into two files -- gdbarch.h now is editable and hand-maintained, and the new gdbarch-gen.h file is the only thing generated by gdbarch.sh. This lets us avoid maintaining boilerplate in the gdbarch.sh file. Note that gdbarch.sh still generates gdbarch.h after this patch. This makes it easier to re-run when rebasing. This code is removed in a subsequent patch.
2021-12-15 07:48:45 +08:00
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 3 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, see <http://www.gnu.org/licenses/>. */
Move copyright code from gdbarch.py to new file This moves the copyright code from gdbarch.py to a new Python source file, gdbcopyright.py. The function in this file will find the copyright dates by scanning the calling script. This will be reused in a future patch. This involved minor changes to the output of gdbarch.py. Also, I've updated copyright.py to remove the reference to gdbarch.sh. We don't need to mention gdbarch.py there, either.
2022-02-14 23:26:32 +08:00
/* To regenerate this file, run:
./gdbarch.py
*/
Split gdbarch.h into two files This patch splits gdbarch.h into two files -- gdbarch.h now is editable and hand-maintained, and the new gdbarch-gen.h file is the only thing generated by gdbarch.sh. This lets us avoid maintaining boilerplate in the gdbarch.sh file. Note that gdbarch.sh still generates gdbarch.h after this patch. This makes it easier to re-run when rebasing. This code is removed in a subsequent patch.
2021-12-15 07:48:45 +08:00
/* The following are pre-initialized by GDBARCH. */
extern const struct bfd_arch_info * gdbarch_bfd_arch_info (struct gdbarch *gdbarch);
/* set_gdbarch_bfd_arch_info() - not applicable - pre-initialized. */
extern enum bfd_endian gdbarch_byte_order (struct gdbarch *gdbarch);
/* set_gdbarch_byte_order() - not applicable - pre-initialized. */
extern enum bfd_endian gdbarch_byte_order_for_code (struct gdbarch *gdbarch);
/* set_gdbarch_byte_order_for_code() - not applicable - pre-initialized. */
extern enum gdb_osabi gdbarch_osabi (struct gdbarch *gdbarch);
/* set_gdbarch_osabi() - not applicable - pre-initialized. */
extern const struct target_desc * gdbarch_target_desc (struct gdbarch *gdbarch);
/* set_gdbarch_target_desc() - not applicable - pre-initialized. */
/* The following are initialized by the target dependent code. */
/* Number of bits in a short or unsigned short for the target machine. */
extern int gdbarch_short_bit (struct gdbarch *gdbarch);
extern void set_gdbarch_short_bit (struct gdbarch *gdbarch, int short_bit);
/* Number of bits in an int or unsigned int for the target machine. */
extern int gdbarch_int_bit (struct gdbarch *gdbarch);
extern void set_gdbarch_int_bit (struct gdbarch *gdbarch, int int_bit);
/* Number of bits in a long or unsigned long for the target machine. */
extern int gdbarch_long_bit (struct gdbarch *gdbarch);
extern void set_gdbarch_long_bit (struct gdbarch *gdbarch, int long_bit);
/* Number of bits in a long long or unsigned long long for the target
machine. */
extern int gdbarch_long_long_bit (struct gdbarch *gdbarch);
extern void set_gdbarch_long_long_bit (struct gdbarch *gdbarch, int long_long_bit);
/* The ABI default bit-size and format for "bfloat16", "half", "float", "double", and
"long double". These bit/format pairs should eventually be combined
into a single object. For the moment, just initialize them as a pair.
Each format describes both the big and little endian layouts (if
useful). */
extern int gdbarch_bfloat16_bit (struct gdbarch *gdbarch);
extern void set_gdbarch_bfloat16_bit (struct gdbarch *gdbarch, int bfloat16_bit);
extern const struct floatformat ** gdbarch_bfloat16_format (struct gdbarch *gdbarch);
extern void set_gdbarch_bfloat16_format (struct gdbarch *gdbarch, const struct floatformat ** bfloat16_format);
extern int gdbarch_half_bit (struct gdbarch *gdbarch);
extern void set_gdbarch_half_bit (struct gdbarch *gdbarch, int half_bit);
extern const struct floatformat ** gdbarch_half_format (struct gdbarch *gdbarch);
extern void set_gdbarch_half_format (struct gdbarch *gdbarch, const struct floatformat ** half_format);
extern int gdbarch_float_bit (struct gdbarch *gdbarch);
extern void set_gdbarch_float_bit (struct gdbarch *gdbarch, int float_bit);
extern const struct floatformat ** gdbarch_float_format (struct gdbarch *gdbarch);
extern void set_gdbarch_float_format (struct gdbarch *gdbarch, const struct floatformat ** float_format);
extern int gdbarch_double_bit (struct gdbarch *gdbarch);
extern void set_gdbarch_double_bit (struct gdbarch *gdbarch, int double_bit);
extern const struct floatformat ** gdbarch_double_format (struct gdbarch *gdbarch);
extern void set_gdbarch_double_format (struct gdbarch *gdbarch, const struct floatformat ** double_format);
extern int gdbarch_long_double_bit (struct gdbarch *gdbarch);
extern void set_gdbarch_long_double_bit (struct gdbarch *gdbarch, int long_double_bit);
extern const struct floatformat ** gdbarch_long_double_format (struct gdbarch *gdbarch);
extern void set_gdbarch_long_double_format (struct gdbarch *gdbarch, const struct floatformat ** long_double_format);
/* The ABI default bit-size for "wchar_t". wchar_t is a built-in type
starting with C++11. */
extern int gdbarch_wchar_bit (struct gdbarch *gdbarch);
extern void set_gdbarch_wchar_bit (struct gdbarch *gdbarch, int wchar_bit);
/* One if `wchar_t' is signed, zero if unsigned. */
extern int gdbarch_wchar_signed (struct gdbarch *gdbarch);
extern void set_gdbarch_wchar_signed (struct gdbarch *gdbarch, int wchar_signed);
/* Returns the floating-point format to be used for values of length LENGTH.
NAME, if non-NULL, is the type name, which may be used to distinguish
different target formats of the same length. */
typedef const struct floatformat ** (gdbarch_floatformat_for_type_ftype) (struct gdbarch *gdbarch, const char *name, int length);
extern const struct floatformat ** gdbarch_floatformat_for_type (struct gdbarch *gdbarch, const char *name, int length);
extern void set_gdbarch_floatformat_for_type (struct gdbarch *gdbarch, gdbarch_floatformat_for_type_ftype *floatformat_for_type);
/* For most targets, a pointer on the target and its representation as an
address in GDB have the same size and "look the same". For such a
target, you need only set gdbarch_ptr_bit and gdbarch_addr_bit
/ addr_bit will be set from it.
Remove gdbarch.sh This patch runs gdbarch.py and removes gdbarch.sh.
2021-12-16 05:38:30 +08:00
Split gdbarch.h into two files This patch splits gdbarch.h into two files -- gdbarch.h now is editable and hand-maintained, and the new gdbarch-gen.h file is the only thing generated by gdbarch.sh. This lets us avoid maintaining boilerplate in the gdbarch.sh file. Note that gdbarch.sh still generates gdbarch.h after this patch. This makes it easier to re-run when rebasing. This code is removed in a subsequent patch.
2021-12-15 07:48:45 +08:00
If gdbarch_ptr_bit and gdbarch_addr_bit are different, you'll probably
also need to set gdbarch_dwarf2_addr_size, gdbarch_pointer_to_address and
gdbarch_address_to_pointer as well.
Remove gdbarch.sh This patch runs gdbarch.py and removes gdbarch.sh.
2021-12-16 05:38:30 +08:00
Split gdbarch.h into two files This patch splits gdbarch.h into two files -- gdbarch.h now is editable and hand-maintained, and the new gdbarch-gen.h file is the only thing generated by gdbarch.sh. This lets us avoid maintaining boilerplate in the gdbarch.sh file. Note that gdbarch.sh still generates gdbarch.h after this patch. This makes it easier to re-run when rebasing. This code is removed in a subsequent patch.
2021-12-15 07:48:45 +08:00
ptr_bit is the size of a pointer on the target */
extern int gdbarch_ptr_bit (struct gdbarch *gdbarch);
extern void set_gdbarch_ptr_bit (struct gdbarch *gdbarch, int ptr_bit);
/* addr_bit is the size of a target address as represented in gdb */
extern int gdbarch_addr_bit (struct gdbarch *gdbarch);
extern void set_gdbarch_addr_bit (struct gdbarch *gdbarch, int addr_bit);
/* dwarf2_addr_size is the target address size as used in the Dwarf debug
info. For .debug_frame FDEs, this is supposed to be the target address
size from the associated CU header, and which is equivalent to the
DWARF2_ADDR_SIZE as defined by the target specific GCC back-end.
Unfortunately there is no good way to determine this value. Therefore
dwarf2_addr_size simply defaults to the target pointer size.
Remove gdbarch.sh This patch runs gdbarch.py and removes gdbarch.sh.
2021-12-16 05:38:30 +08:00
Split gdbarch.h into two files This patch splits gdbarch.h into two files -- gdbarch.h now is editable and hand-maintained, and the new gdbarch-gen.h file is the only thing generated by gdbarch.sh. This lets us avoid maintaining boilerplate in the gdbarch.sh file. Note that gdbarch.sh still generates gdbarch.h after this patch. This makes it easier to re-run when rebasing. This code is removed in a subsequent patch.
2021-12-15 07:48:45 +08:00
dwarf2_addr_size is not used for .eh_frame FDEs, which are generally
defined using the target's pointer size so far.
Remove gdbarch.sh This patch runs gdbarch.py and removes gdbarch.sh.
2021-12-16 05:38:30 +08:00
Split gdbarch.h into two files This patch splits gdbarch.h into two files -- gdbarch.h now is editable and hand-maintained, and the new gdbarch-gen.h file is the only thing generated by gdbarch.sh. This lets us avoid maintaining boilerplate in the gdbarch.sh file. Note that gdbarch.sh still generates gdbarch.h after this patch. This makes it easier to re-run when rebasing. This code is removed in a subsequent patch.
2021-12-15 07:48:45 +08:00
Note that dwarf2_addr_size only needs to be redefined by a target if the
GCC back-end defines a DWARF2_ADDR_SIZE other than the target pointer size,
and if Dwarf versions < 4 need to be supported. */
extern int gdbarch_dwarf2_addr_size (struct gdbarch *gdbarch);
extern void set_gdbarch_dwarf2_addr_size (struct gdbarch *gdbarch, int dwarf2_addr_size);
/* One if `char' acts like `signed char', zero if `unsigned char'. */
extern int gdbarch_char_signed (struct gdbarch *gdbarch);
extern void set_gdbarch_char_signed (struct gdbarch *gdbarch, int char_signed);
extern bool gdbarch_read_pc_p (struct gdbarch *gdbarch);
typedef CORE_ADDR (gdbarch_read_pc_ftype) (readable_regcache *regcache);
extern CORE_ADDR gdbarch_read_pc (struct gdbarch *gdbarch, readable_regcache *regcache);
extern void set_gdbarch_read_pc (struct gdbarch *gdbarch, gdbarch_read_pc_ftype *read_pc);
extern bool gdbarch_write_pc_p (struct gdbarch *gdbarch);
typedef void (gdbarch_write_pc_ftype) (struct regcache *regcache, CORE_ADDR val);
extern void gdbarch_write_pc (struct gdbarch *gdbarch, struct regcache *regcache, CORE_ADDR val);
extern void set_gdbarch_write_pc (struct gdbarch *gdbarch, gdbarch_write_pc_ftype *write_pc);
/* Function for getting target's idea of a frame pointer. FIXME: GDB's
whole scheme for dealing with "frames" and "frame pointers" needs a
serious shakedown. */
typedef void (gdbarch_virtual_frame_pointer_ftype) (struct gdbarch *gdbarch, CORE_ADDR pc, int *frame_regnum, LONGEST *frame_offset);
extern void gdbarch_virtual_frame_pointer (struct gdbarch *gdbarch, CORE_ADDR pc, int *frame_regnum, LONGEST *frame_offset);
extern void set_gdbarch_virtual_frame_pointer (struct gdbarch *gdbarch, gdbarch_virtual_frame_pointer_ftype *virtual_frame_pointer);
extern bool gdbarch_pseudo_register_read_p (struct gdbarch *gdbarch);
typedef enum register_status (gdbarch_pseudo_register_read_ftype) (struct gdbarch *gdbarch, readable_regcache *regcache, int cookednum, gdb_byte *buf);
extern enum register_status gdbarch_pseudo_register_read (struct gdbarch *gdbarch, readable_regcache *regcache, int cookednum, gdb_byte *buf);
extern void set_gdbarch_pseudo_register_read (struct gdbarch *gdbarch, gdbarch_pseudo_register_read_ftype *pseudo_register_read);
/* Read a register into a new struct value. If the register is wholly
or partly unavailable, this should call mark_value_bytes_unavailable
as appropriate. If this is defined, then pseudo_register_read will
never be called. */
extern bool gdbarch_pseudo_register_read_value_p (struct gdbarch *gdbarch);
gdb: pass frames as `const frame_info_ptr &` We currently pass frames to function by value, as `frame_info_ptr`. This is somewhat expensive: - the size of `frame_info_ptr` is 64 bytes, which is a bit big to pass by value - the constructors and destructor link/unlink the object in the global `frame_info_ptr::frame_list` list. This is an `intrusive_list`, so it's not so bad: it's just assigning a few points, there's no memory allocation as if it was `std::list`, but still it's useless to do that over and over. As suggested by Tom Tromey, change many function signatures to accept `const frame_info_ptr &` instead of `frame_info_ptr`. Some functions reassign their `frame_info_ptr` parameter, like: void the_func (frame_info_ptr frame) { for (; frame != nullptr; frame = get_prev_frame (frame)) { ... } } I wondered what to do about them, do I leave them as-is or change them (and need to introduce a separate local variable that can be re-assigned). I opted for the later for consistency. It might not be clear why some functions take `const frame_info_ptr &` while others take `frame_info_ptr`. Also, if a function took a `frame_info_ptr` because it did re-assign its parameter, I doubt that we would think to change it to `const frame_info_ptr &` should the implementation change such that it doesn't need to take `frame_info_ptr` anymore. It seems better to have a simple rule and apply it everywhere. Change-Id: I59d10addef687d157f82ccf4d54f5dde9a963fd0 Approved-By: Andrew Burgess <aburgess@redhat.com>
2024-02-20 02:07:47 +08:00
typedef struct value * (gdbarch_pseudo_register_read_value_ftype) (struct gdbarch *gdbarch, const frame_info_ptr &next_frame, int cookednum);
extern struct value * gdbarch_pseudo_register_read_value (struct gdbarch *gdbarch, const frame_info_ptr &next_frame, int cookednum);
Split gdbarch.h into two files This patch splits gdbarch.h into two files -- gdbarch.h now is editable and hand-maintained, and the new gdbarch-gen.h file is the only thing generated by gdbarch.sh. This lets us avoid maintaining boilerplate in the gdbarch.sh file. Note that gdbarch.sh still generates gdbarch.h after this patch. This makes it easier to re-run when rebasing. This code is removed in a subsequent patch.
2021-12-15 07:48:45 +08:00
extern void set_gdbarch_pseudo_register_read_value (struct gdbarch *gdbarch, gdbarch_pseudo_register_read_value_ftype *pseudo_register_read_value);
gdb: add gdbarch_pseudo_register_write that takes a frame Add a new variant of gdbarch_pseudo_register_write that takes a frame_info in order to write raw registers. Use this new method when available: - in put_frame_register, when trying to write a pseudo register to a given frame - in regcache::cooked_write No implementation is migrated to use this new method (that will come in subsequent patches), so no behavior change is expected here. The objective is to fix writing pseudo registers to non-current frames. See previous commit "gdb: read pseudo register through frame" for a more detailed explanation. Change-Id: Ie7fe364a15a4d86c2ecb09de2b4baa08c45555ac Reviewed-By: John Baldwin <jhb@FreeBSD.org>
2023-12-02 00:27:30 +08:00
/* Write bytes in BUF to pseudo register with number PSEUDO_REG_NUM.
Raw registers backing the pseudo register should be written to using
NEXT_FRAME. */
extern bool gdbarch_pseudo_register_write_p (struct gdbarch *gdbarch);
gdb: pass frames as `const frame_info_ptr &` We currently pass frames to function by value, as `frame_info_ptr`. This is somewhat expensive: - the size of `frame_info_ptr` is 64 bytes, which is a bit big to pass by value - the constructors and destructor link/unlink the object in the global `frame_info_ptr::frame_list` list. This is an `intrusive_list`, so it's not so bad: it's just assigning a few points, there's no memory allocation as if it was `std::list`, but still it's useless to do that over and over. As suggested by Tom Tromey, change many function signatures to accept `const frame_info_ptr &` instead of `frame_info_ptr`. Some functions reassign their `frame_info_ptr` parameter, like: void the_func (frame_info_ptr frame) { for (; frame != nullptr; frame = get_prev_frame (frame)) { ... } } I wondered what to do about them, do I leave them as-is or change them (and need to introduce a separate local variable that can be re-assigned). I opted for the later for consistency. It might not be clear why some functions take `const frame_info_ptr &` while others take `frame_info_ptr`. Also, if a function took a `frame_info_ptr` because it did re-assign its parameter, I doubt that we would think to change it to `const frame_info_ptr &` should the implementation change such that it doesn't need to take `frame_info_ptr` anymore. It seems better to have a simple rule and apply it everywhere. Change-Id: I59d10addef687d157f82ccf4d54f5dde9a963fd0 Approved-By: Andrew Burgess <aburgess@redhat.com>
2024-02-20 02:07:47 +08:00
typedef void (gdbarch_pseudo_register_write_ftype) (struct gdbarch *gdbarch, const frame_info_ptr &next_frame, int pseudo_reg_num, gdb::array_view<const gdb_byte> buf);
extern void gdbarch_pseudo_register_write (struct gdbarch *gdbarch, const frame_info_ptr &next_frame, int pseudo_reg_num, gdb::array_view<const gdb_byte> buf);
gdb: add gdbarch_pseudo_register_write that takes a frame Add a new variant of gdbarch_pseudo_register_write that takes a frame_info in order to write raw registers. Use this new method when available: - in put_frame_register, when trying to write a pseudo register to a given frame - in regcache::cooked_write No implementation is migrated to use this new method (that will come in subsequent patches), so no behavior change is expected here. The objective is to fix writing pseudo registers to non-current frames. See previous commit "gdb: read pseudo register through frame" for a more detailed explanation. Change-Id: Ie7fe364a15a4d86c2ecb09de2b4baa08c45555ac Reviewed-By: John Baldwin <jhb@FreeBSD.org>
2023-12-02 00:27:30 +08:00
extern void set_gdbarch_pseudo_register_write (struct gdbarch *gdbarch, gdbarch_pseudo_register_write_ftype *pseudo_register_write);
gdb: rename gdbarch_pseudo_register_write to gdbarch_deprecated_pseudo_register_write The next patch introduces a new variant of gdbarch_pseudo_register_write that takes a frame instead of a regcache for implementations to write raw registers. Rename to old one to make it clear it's deprecated. Change-Id: If8872c89c6f8a1edfcab983eb064248fd5ff3115 Reviewed-By: John Baldwin <jhb@FreeBSD.org>
2023-12-02 00:27:29 +08:00
/* Write bytes to a pseudo register.
Split gdbarch.h into two files This patch splits gdbarch.h into two files -- gdbarch.h now is editable and hand-maintained, and the new gdbarch-gen.h file is the only thing generated by gdbarch.sh. This lets us avoid maintaining boilerplate in the gdbarch.sh file. Note that gdbarch.sh still generates gdbarch.h after this patch. This makes it easier to re-run when rebasing. This code is removed in a subsequent patch.
2021-12-15 07:48:45 +08:00
gdb: rename gdbarch_pseudo_register_write to gdbarch_deprecated_pseudo_register_write The next patch introduces a new variant of gdbarch_pseudo_register_write that takes a frame instead of a regcache for implementations to write raw registers. Rename to old one to make it clear it's deprecated. Change-Id: If8872c89c6f8a1edfcab983eb064248fd5ff3115 Reviewed-By: John Baldwin <jhb@FreeBSD.org>
2023-12-02 00:27:29 +08:00
This is marked as deprecated because it gets passed a regcache for
implementations to write raw registers in. This doesn't work for unwound
frames, where the raw registers backing the pseudo registers may have been
gdb: add gdbarch_pseudo_register_write that takes a frame Add a new variant of gdbarch_pseudo_register_write that takes a frame_info in order to write raw registers. Use this new method when available: - in put_frame_register, when trying to write a pseudo register to a given frame - in regcache::cooked_write No implementation is migrated to use this new method (that will come in subsequent patches), so no behavior change is expected here. The objective is to fix writing pseudo registers to non-current frames. See previous commit "gdb: read pseudo register through frame" for a more detailed explanation. Change-Id: Ie7fe364a15a4d86c2ecb09de2b4baa08c45555ac Reviewed-By: John Baldwin <jhb@FreeBSD.org>
2023-12-02 00:27:30 +08:00
saved elsewhere.
Implementations should be migrated to implement pseudo_register_write instead. */
gdb: rename gdbarch_pseudo_register_write to gdbarch_deprecated_pseudo_register_write The next patch introduces a new variant of gdbarch_pseudo_register_write that takes a frame instead of a regcache for implementations to write raw registers. Rename to old one to make it clear it's deprecated. Change-Id: If8872c89c6f8a1edfcab983eb064248fd5ff3115 Reviewed-By: John Baldwin <jhb@FreeBSD.org>
2023-12-02 00:27:29 +08:00
extern bool gdbarch_deprecated_pseudo_register_write_p (struct gdbarch *gdbarch);
typedef void (gdbarch_deprecated_pseudo_register_write_ftype) (struct gdbarch *gdbarch, struct regcache *regcache, int cookednum, const gdb_byte *buf);
extern void gdbarch_deprecated_pseudo_register_write (struct gdbarch *gdbarch, struct regcache *regcache, int cookednum, const gdb_byte *buf);
extern void set_gdbarch_deprecated_pseudo_register_write (struct gdbarch *gdbarch, gdbarch_deprecated_pseudo_register_write_ftype *deprecated_pseudo_register_write);
Split gdbarch.h into two files This patch splits gdbarch.h into two files -- gdbarch.h now is editable and hand-maintained, and the new gdbarch-gen.h file is the only thing generated by gdbarch.sh. This lets us avoid maintaining boilerplate in the gdbarch.sh file. Note that gdbarch.sh still generates gdbarch.h after this patch. This makes it easier to re-run when rebasing. This code is removed in a subsequent patch.
2021-12-15 07:48:45 +08:00
extern int gdbarch_num_regs (struct gdbarch *gdbarch);
extern void set_gdbarch_num_regs (struct gdbarch *gdbarch, int num_regs);
/* This macro gives the number of pseudo-registers that live in the
register namespace but do not get fetched or stored on the target.
These pseudo-registers may be aliases for other registers,
combinations of other registers, or they may be computed by GDB. */
extern int gdbarch_num_pseudo_regs (struct gdbarch *gdbarch);
extern void set_gdbarch_num_pseudo_regs (struct gdbarch *gdbarch, int num_pseudo_regs);
/* Assemble agent expression bytecode to collect pseudo-register REG.
Return -1 if something goes wrong, 0 otherwise. */
extern bool gdbarch_ax_pseudo_register_collect_p (struct gdbarch *gdbarch);
typedef int (gdbarch_ax_pseudo_register_collect_ftype) (struct gdbarch *gdbarch, struct agent_expr *ax, int reg);
extern int gdbarch_ax_pseudo_register_collect (struct gdbarch *gdbarch, struct agent_expr *ax, int reg);
extern void set_gdbarch_ax_pseudo_register_collect (struct gdbarch *gdbarch, gdbarch_ax_pseudo_register_collect_ftype *ax_pseudo_register_collect);
/* Assemble agent expression bytecode to push the value of pseudo-register
REG on the interpreter stack.
Return -1 if something goes wrong, 0 otherwise. */
extern bool gdbarch_ax_pseudo_register_push_stack_p (struct gdbarch *gdbarch);
typedef int (gdbarch_ax_pseudo_register_push_stack_ftype) (struct gdbarch *gdbarch, struct agent_expr *ax, int reg);
extern int gdbarch_ax_pseudo_register_push_stack (struct gdbarch *gdbarch, struct agent_expr *ax, int reg);
extern void set_gdbarch_ax_pseudo_register_push_stack (struct gdbarch *gdbarch, gdbarch_ax_pseudo_register_push_stack_ftype *ax_pseudo_register_push_stack);
/* Some architectures can display additional information for specific
signals.
UIOUT is the output stream where the handler will place information. */
extern bool gdbarch_report_signal_info_p (struct gdbarch *gdbarch);
typedef void (gdbarch_report_signal_info_ftype) (struct gdbarch *gdbarch, struct ui_out *uiout, enum gdb_signal siggnal);
extern void gdbarch_report_signal_info (struct gdbarch *gdbarch, struct ui_out *uiout, enum gdb_signal siggnal);
extern void set_gdbarch_report_signal_info (struct gdbarch *gdbarch, gdbarch_report_signal_info_ftype *report_signal_info);
/* GDB's standard (or well known) register numbers. These can map onto
a real register or a pseudo (computed) register or not be defined at
all (-1).
gdbarch_sp_regnum will hopefully be replaced by UNWIND_SP. */
extern int gdbarch_sp_regnum (struct gdbarch *gdbarch);
extern void set_gdbarch_sp_regnum (struct gdbarch *gdbarch, int sp_regnum);
extern int gdbarch_pc_regnum (struct gdbarch *gdbarch);
extern void set_gdbarch_pc_regnum (struct gdbarch *gdbarch, int pc_regnum);
extern int gdbarch_ps_regnum (struct gdbarch *gdbarch);
extern void set_gdbarch_ps_regnum (struct gdbarch *gdbarch, int ps_regnum);
extern int gdbarch_fp0_regnum (struct gdbarch *gdbarch);
extern void set_gdbarch_fp0_regnum (struct gdbarch *gdbarch, int fp0_regnum);
/* Convert stab register number (from `r' declaration) to a gdb REGNUM. */
typedef int (gdbarch_stab_reg_to_regnum_ftype) (struct gdbarch *gdbarch, int stab_regnr);
extern int gdbarch_stab_reg_to_regnum (struct gdbarch *gdbarch, int stab_regnr);
extern void set_gdbarch_stab_reg_to_regnum (struct gdbarch *gdbarch, gdbarch_stab_reg_to_regnum_ftype *stab_reg_to_regnum);
/* Provide a default mapping from a ecoff register number to a gdb REGNUM. */
typedef int (gdbarch_ecoff_reg_to_regnum_ftype) (struct gdbarch *gdbarch, int ecoff_regnr);
extern int gdbarch_ecoff_reg_to_regnum (struct gdbarch *gdbarch, int ecoff_regnr);
extern void set_gdbarch_ecoff_reg_to_regnum (struct gdbarch *gdbarch, gdbarch_ecoff_reg_to_regnum_ftype *ecoff_reg_to_regnum);
/* Convert from an sdb register number to an internal gdb register number. */
typedef int (gdbarch_sdb_reg_to_regnum_ftype) (struct gdbarch *gdbarch, int sdb_regnr);
extern int gdbarch_sdb_reg_to_regnum (struct gdbarch *gdbarch, int sdb_regnr);
extern void set_gdbarch_sdb_reg_to_regnum (struct gdbarch *gdbarch, gdbarch_sdb_reg_to_regnum_ftype *sdb_reg_to_regnum);
/* Provide a default mapping from a DWARF2 register number to a gdb REGNUM.
Return -1 for bad REGNUM. Note: Several targets get this wrong. */
typedef int (gdbarch_dwarf2_reg_to_regnum_ftype) (struct gdbarch *gdbarch, int dwarf2_regnr);
extern int gdbarch_dwarf2_reg_to_regnum (struct gdbarch *gdbarch, int dwarf2_regnr);
extern void set_gdbarch_dwarf2_reg_to_regnum (struct gdbarch *gdbarch, gdbarch_dwarf2_reg_to_regnum_ftype *dwarf2_reg_to_regnum);
gdb/gdbarch: add a comment to gdbarch_register_name After the previous commit, this commit sets out to formalise the API for gdbarch_register_name. Not every architecture is actually in compliance with the API I set out here, but I believe that most are. I think architectures that don't comply with the API laid out here will fail the gdb.base/completion.exp test. The claims in the comment are I feel, best demonstrated with the asserts in this code: const char * gdbarch_register_name (struct gdbarch *gdbarch, int regnr) { gdb_assert (regnr >= 0); gdb_assert (regnr < gdbarch_num_cooked_regs (gdbarch)); const char *name = gdbarch->register_name (gdbarch, regnr); gdb_assert (name != nullptr); return name; } Like I said, I don't believe every architecture follows these rules right now, which is why I'm not actually adding any asserts. Instead, this commit adds a comment to gdbarch_register_name, this comment is where I'd like to get to, rather than where we are right now. Subsequent commits will fix all targets to be in compliance with this comment, and will even add the asserts shown above to gdbarch_register_name.
2022-08-29 04:19:50 +08:00
/* Return the name of register REGNR for the specified architecture.
REGNR can be any value greater than, or equal to zero, and less than
'gdbarch_num_cooked_regs (GDBARCH)'. If REGNR is not supported for
GDBARCH, then this function will return an empty string, this function
should never return nullptr. */
Split gdbarch.h into two files This patch splits gdbarch.h into two files -- gdbarch.h now is editable and hand-maintained, and the new gdbarch-gen.h file is the only thing generated by gdbarch.sh. This lets us avoid maintaining boilerplate in the gdbarch.sh file. Note that gdbarch.sh still generates gdbarch.h after this patch. This makes it easier to re-run when rebasing. This code is removed in a subsequent patch.
2021-12-15 07:48:45 +08:00
typedef const char * (gdbarch_register_name_ftype) (struct gdbarch *gdbarch, int regnr);
extern const char * gdbarch_register_name (struct gdbarch *gdbarch, int regnr);
extern void set_gdbarch_register_name (struct gdbarch *gdbarch, gdbarch_register_name_ftype *register_name);
/* Return the type of a register specified by the architecture. Only
the register cache should call this function directly; others should
use "register_type". */
typedef struct type * (gdbarch_register_type_ftype) (struct gdbarch *gdbarch, int reg_nr);
extern struct type * gdbarch_register_type (struct gdbarch *gdbarch, int reg_nr);
extern void set_gdbarch_register_type (struct gdbarch *gdbarch, gdbarch_register_type_ftype *register_type);
/* Generate a dummy frame_id for THIS_FRAME assuming that the frame is
a dummy frame. A dummy frame is created before an inferior call,
the frame_id returned here must match the frame_id that was built
for the inferior call. Usually this means the returned frame_id's
stack address should match the address returned by
gdbarch_push_dummy_call, and the returned frame_id's code address
should match the address at which the breakpoint was set in the dummy
frame. */
gdb: pass frames as `const frame_info_ptr &` We currently pass frames to function by value, as `frame_info_ptr`. This is somewhat expensive: - the size of `frame_info_ptr` is 64 bytes, which is a bit big to pass by value - the constructors and destructor link/unlink the object in the global `frame_info_ptr::frame_list` list. This is an `intrusive_list`, so it's not so bad: it's just assigning a few points, there's no memory allocation as if it was `std::list`, but still it's useless to do that over and over. As suggested by Tom Tromey, change many function signatures to accept `const frame_info_ptr &` instead of `frame_info_ptr`. Some functions reassign their `frame_info_ptr` parameter, like: void the_func (frame_info_ptr frame) { for (; frame != nullptr; frame = get_prev_frame (frame)) { ... } } I wondered what to do about them, do I leave them as-is or change them (and need to introduce a separate local variable that can be re-assigned). I opted for the later for consistency. It might not be clear why some functions take `const frame_info_ptr &` while others take `frame_info_ptr`. Also, if a function took a `frame_info_ptr` because it did re-assign its parameter, I doubt that we would think to change it to `const frame_info_ptr &` should the implementation change such that it doesn't need to take `frame_info_ptr` anymore. It seems better to have a simple rule and apply it everywhere. Change-Id: I59d10addef687d157f82ccf4d54f5dde9a963fd0 Approved-By: Andrew Burgess <aburgess@redhat.com>
2024-02-20 02:07:47 +08:00
typedef struct frame_id (gdbarch_dummy_id_ftype) (struct gdbarch *gdbarch, const frame_info_ptr &this_frame);
extern struct frame_id gdbarch_dummy_id (struct gdbarch *gdbarch, const frame_info_ptr &this_frame);
Split gdbarch.h into two files This patch splits gdbarch.h into two files -- gdbarch.h now is editable and hand-maintained, and the new gdbarch-gen.h file is the only thing generated by gdbarch.sh. This lets us avoid maintaining boilerplate in the gdbarch.sh file. Note that gdbarch.sh still generates gdbarch.h after this patch. This makes it easier to re-run when rebasing. This code is removed in a subsequent patch.
2021-12-15 07:48:45 +08:00
extern void set_gdbarch_dummy_id (struct gdbarch *gdbarch, gdbarch_dummy_id_ftype *dummy_id);
/* Implement DUMMY_ID and PUSH_DUMMY_CALL, then delete
deprecated_fp_regnum. */
extern int gdbarch_deprecated_fp_regnum (struct gdbarch *gdbarch);
extern void set_gdbarch_deprecated_fp_regnum (struct gdbarch *gdbarch, int deprecated_fp_regnum);
extern bool gdbarch_push_dummy_call_p (struct gdbarch *gdbarch);
typedef CORE_ADDR (gdbarch_push_dummy_call_ftype) (struct gdbarch *gdbarch, struct value *function, struct regcache *regcache, CORE_ADDR bp_addr, int nargs, struct value **args, CORE_ADDR sp, function_call_return_method return_method, CORE_ADDR struct_addr);
extern CORE_ADDR gdbarch_push_dummy_call (struct gdbarch *gdbarch, struct value *function, struct regcache *regcache, CORE_ADDR bp_addr, int nargs, struct value **args, CORE_ADDR sp, function_call_return_method return_method, CORE_ADDR struct_addr);
extern void set_gdbarch_push_dummy_call (struct gdbarch *gdbarch, gdbarch_push_dummy_call_ftype *push_dummy_call);
Use enum for gdbarch's call_dummy_location This changes gdbarch to use an enum for call_dummy_location, providing a little more type safety.
2022-09-12 21:04:21 +08:00
extern enum call_dummy_location_type gdbarch_call_dummy_location (struct gdbarch *gdbarch);
extern void set_gdbarch_call_dummy_location (struct gdbarch *gdbarch, enum call_dummy_location_type call_dummy_location);
Split gdbarch.h into two files This patch splits gdbarch.h into two files -- gdbarch.h now is editable and hand-maintained, and the new gdbarch-gen.h file is the only thing generated by gdbarch.sh. This lets us avoid maintaining boilerplate in the gdbarch.sh file. Note that gdbarch.sh still generates gdbarch.h after this patch. This makes it easier to re-run when rebasing. This code is removed in a subsequent patch.
2021-12-15 07:48:45 +08:00
extern bool gdbarch_push_dummy_code_p (struct gdbarch *gdbarch);
typedef CORE_ADDR (gdbarch_push_dummy_code_ftype) (struct gdbarch *gdbarch, CORE_ADDR sp, CORE_ADDR funaddr, struct value **args, int nargs, struct type *value_type, CORE_ADDR *real_pc, CORE_ADDR *bp_addr, struct regcache *regcache);
extern CORE_ADDR gdbarch_push_dummy_code (struct gdbarch *gdbarch, CORE_ADDR sp, CORE_ADDR funaddr, struct value **args, int nargs, struct type *value_type, CORE_ADDR *real_pc, CORE_ADDR *bp_addr, struct regcache *regcache);
extern void set_gdbarch_push_dummy_code (struct gdbarch *gdbarch, gdbarch_push_dummy_code_ftype *push_dummy_code);
/* Return true if the code of FRAME is writable. */
gdb: pass frames as `const frame_info_ptr &` We currently pass frames to function by value, as `frame_info_ptr`. This is somewhat expensive: - the size of `frame_info_ptr` is 64 bytes, which is a bit big to pass by value - the constructors and destructor link/unlink the object in the global `frame_info_ptr::frame_list` list. This is an `intrusive_list`, so it's not so bad: it's just assigning a few points, there's no memory allocation as if it was `std::list`, but still it's useless to do that over and over. As suggested by Tom Tromey, change many function signatures to accept `const frame_info_ptr &` instead of `frame_info_ptr`. Some functions reassign their `frame_info_ptr` parameter, like: void the_func (frame_info_ptr frame) { for (; frame != nullptr; frame = get_prev_frame (frame)) { ... } } I wondered what to do about them, do I leave them as-is or change them (and need to introduce a separate local variable that can be re-assigned). I opted for the later for consistency. It might not be clear why some functions take `const frame_info_ptr &` while others take `frame_info_ptr`. Also, if a function took a `frame_info_ptr` because it did re-assign its parameter, I doubt that we would think to change it to `const frame_info_ptr &` should the implementation change such that it doesn't need to take `frame_info_ptr` anymore. It seems better to have a simple rule and apply it everywhere. Change-Id: I59d10addef687d157f82ccf4d54f5dde9a963fd0 Approved-By: Andrew Burgess <aburgess@redhat.com>
2024-02-20 02:07:47 +08:00
typedef int (gdbarch_code_of_frame_writable_ftype) (struct gdbarch *gdbarch, const frame_info_ptr &frame);
extern int gdbarch_code_of_frame_writable (struct gdbarch *gdbarch, const frame_info_ptr &frame);
Split gdbarch.h into two files This patch splits gdbarch.h into two files -- gdbarch.h now is editable and hand-maintained, and the new gdbarch-gen.h file is the only thing generated by gdbarch.sh. This lets us avoid maintaining boilerplate in the gdbarch.sh file. Note that gdbarch.sh still generates gdbarch.h after this patch. This makes it easier to re-run when rebasing. This code is removed in a subsequent patch.
2021-12-15 07:48:45 +08:00
extern void set_gdbarch_code_of_frame_writable (struct gdbarch *gdbarch, gdbarch_code_of_frame_writable_ftype *code_of_frame_writable);
gdb: pass frames as `const frame_info_ptr &` We currently pass frames to function by value, as `frame_info_ptr`. This is somewhat expensive: - the size of `frame_info_ptr` is 64 bytes, which is a bit big to pass by value - the constructors and destructor link/unlink the object in the global `frame_info_ptr::frame_list` list. This is an `intrusive_list`, so it's not so bad: it's just assigning a few points, there's no memory allocation as if it was `std::list`, but still it's useless to do that over and over. As suggested by Tom Tromey, change many function signatures to accept `const frame_info_ptr &` instead of `frame_info_ptr`. Some functions reassign their `frame_info_ptr` parameter, like: void the_func (frame_info_ptr frame) { for (; frame != nullptr; frame = get_prev_frame (frame)) { ... } } I wondered what to do about them, do I leave them as-is or change them (and need to introduce a separate local variable that can be re-assigned). I opted for the later for consistency. It might not be clear why some functions take `const frame_info_ptr &` while others take `frame_info_ptr`. Also, if a function took a `frame_info_ptr` because it did re-assign its parameter, I doubt that we would think to change it to `const frame_info_ptr &` should the implementation change such that it doesn't need to take `frame_info_ptr` anymore. It seems better to have a simple rule and apply it everywhere. Change-Id: I59d10addef687d157f82ccf4d54f5dde9a963fd0 Approved-By: Andrew Burgess <aburgess@redhat.com>
2024-02-20 02:07:47 +08:00
typedef void (gdbarch_print_registers_info_ftype) (struct gdbarch *gdbarch, struct ui_file *file, const frame_info_ptr &frame, int regnum, int all);
extern void gdbarch_print_registers_info (struct gdbarch *gdbarch, struct ui_file *file, const frame_info_ptr &frame, int regnum, int all);
Split gdbarch.h into two files This patch splits gdbarch.h into two files -- gdbarch.h now is editable and hand-maintained, and the new gdbarch-gen.h file is the only thing generated by gdbarch.sh. This lets us avoid maintaining boilerplate in the gdbarch.sh file. Note that gdbarch.sh still generates gdbarch.h after this patch. This makes it easier to re-run when rebasing. This code is removed in a subsequent patch.
2021-12-15 07:48:45 +08:00
extern void set_gdbarch_print_registers_info (struct gdbarch *gdbarch, gdbarch_print_registers_info_ftype *print_registers_info);
gdb: pass frames as `const frame_info_ptr &` We currently pass frames to function by value, as `frame_info_ptr`. This is somewhat expensive: - the size of `frame_info_ptr` is 64 bytes, which is a bit big to pass by value - the constructors and destructor link/unlink the object in the global `frame_info_ptr::frame_list` list. This is an `intrusive_list`, so it's not so bad: it's just assigning a few points, there's no memory allocation as if it was `std::list`, but still it's useless to do that over and over. As suggested by Tom Tromey, change many function signatures to accept `const frame_info_ptr &` instead of `frame_info_ptr`. Some functions reassign their `frame_info_ptr` parameter, like: void the_func (frame_info_ptr frame) { for (; frame != nullptr; frame = get_prev_frame (frame)) { ... } } I wondered what to do about them, do I leave them as-is or change them (and need to introduce a separate local variable that can be re-assigned). I opted for the later for consistency. It might not be clear why some functions take `const frame_info_ptr &` while others take `frame_info_ptr`. Also, if a function took a `frame_info_ptr` because it did re-assign its parameter, I doubt that we would think to change it to `const frame_info_ptr &` should the implementation change such that it doesn't need to take `frame_info_ptr` anymore. It seems better to have a simple rule and apply it everywhere. Change-Id: I59d10addef687d157f82ccf4d54f5dde9a963fd0 Approved-By: Andrew Burgess <aburgess@redhat.com>
2024-02-20 02:07:47 +08:00
typedef void (gdbarch_print_float_info_ftype) (struct gdbarch *gdbarch, struct ui_file *file, const frame_info_ptr &frame, const char *args);
extern void gdbarch_print_float_info (struct gdbarch *gdbarch, struct ui_file *file, const frame_info_ptr &frame, const char *args);
Split gdbarch.h into two files This patch splits gdbarch.h into two files -- gdbarch.h now is editable and hand-maintained, and the new gdbarch-gen.h file is the only thing generated by gdbarch.sh. This lets us avoid maintaining boilerplate in the gdbarch.sh file. Note that gdbarch.sh still generates gdbarch.h after this patch. This makes it easier to re-run when rebasing. This code is removed in a subsequent patch.
2021-12-15 07:48:45 +08:00
extern void set_gdbarch_print_float_info (struct gdbarch *gdbarch, gdbarch_print_float_info_ftype *print_float_info);
extern bool gdbarch_print_vector_info_p (struct gdbarch *gdbarch);
gdb: pass frames as `const frame_info_ptr &` We currently pass frames to function by value, as `frame_info_ptr`. This is somewhat expensive: - the size of `frame_info_ptr` is 64 bytes, which is a bit big to pass by value - the constructors and destructor link/unlink the object in the global `frame_info_ptr::frame_list` list. This is an `intrusive_list`, so it's not so bad: it's just assigning a few points, there's no memory allocation as if it was `std::list`, but still it's useless to do that over and over. As suggested by Tom Tromey, change many function signatures to accept `const frame_info_ptr &` instead of `frame_info_ptr`. Some functions reassign their `frame_info_ptr` parameter, like: void the_func (frame_info_ptr frame) { for (; frame != nullptr; frame = get_prev_frame (frame)) { ... } } I wondered what to do about them, do I leave them as-is or change them (and need to introduce a separate local variable that can be re-assigned). I opted for the later for consistency. It might not be clear why some functions take `const frame_info_ptr &` while others take `frame_info_ptr`. Also, if a function took a `frame_info_ptr` because it did re-assign its parameter, I doubt that we would think to change it to `const frame_info_ptr &` should the implementation change such that it doesn't need to take `frame_info_ptr` anymore. It seems better to have a simple rule and apply it everywhere. Change-Id: I59d10addef687d157f82ccf4d54f5dde9a963fd0 Approved-By: Andrew Burgess <aburgess@redhat.com>
2024-02-20 02:07:47 +08:00
typedef void (gdbarch_print_vector_info_ftype) (struct gdbarch *gdbarch, struct ui_file *file, const frame_info_ptr &frame, const char *args);
extern void gdbarch_print_vector_info (struct gdbarch *gdbarch, struct ui_file *file, const frame_info_ptr &frame, const char *args);
Split gdbarch.h into two files This patch splits gdbarch.h into two files -- gdbarch.h now is editable and hand-maintained, and the new gdbarch-gen.h file is the only thing generated by gdbarch.sh. This lets us avoid maintaining boilerplate in the gdbarch.sh file. Note that gdbarch.sh still generates gdbarch.h after this patch. This makes it easier to re-run when rebasing. This code is removed in a subsequent patch.
2021-12-15 07:48:45 +08:00
extern void set_gdbarch_print_vector_info (struct gdbarch *gdbarch, gdbarch_print_vector_info_ftype *print_vector_info);
/* MAP a GDB RAW register number onto a simulator register number. See
also include/...-sim.h. */
typedef int (gdbarch_register_sim_regno_ftype) (struct gdbarch *gdbarch, int reg_nr);
extern int gdbarch_register_sim_regno (struct gdbarch *gdbarch, int reg_nr);
extern void set_gdbarch_register_sim_regno (struct gdbarch *gdbarch, gdbarch_register_sim_regno_ftype *register_sim_regno);
typedef int (gdbarch_cannot_fetch_register_ftype) (struct gdbarch *gdbarch, int regnum);
extern int gdbarch_cannot_fetch_register (struct gdbarch *gdbarch, int regnum);
extern void set_gdbarch_cannot_fetch_register (struct gdbarch *gdbarch, gdbarch_cannot_fetch_register_ftype *cannot_fetch_register);
typedef int (gdbarch_cannot_store_register_ftype) (struct gdbarch *gdbarch, int regnum);
extern int gdbarch_cannot_store_register (struct gdbarch *gdbarch, int regnum);
extern void set_gdbarch_cannot_store_register (struct gdbarch *gdbarch, gdbarch_cannot_store_register_ftype *cannot_store_register);
/* Determine the address where a longjmp will land and save this address
in PC. Return nonzero on success.
Remove gdbarch.sh This patch runs gdbarch.py and removes gdbarch.sh.
2021-12-16 05:38:30 +08:00
Split gdbarch.h into two files This patch splits gdbarch.h into two files -- gdbarch.h now is editable and hand-maintained, and the new gdbarch-gen.h file is the only thing generated by gdbarch.sh. This lets us avoid maintaining boilerplate in the gdbarch.sh file. Note that gdbarch.sh still generates gdbarch.h after this patch. This makes it easier to re-run when rebasing. This code is removed in a subsequent patch.
2021-12-15 07:48:45 +08:00
FRAME corresponds to the longjmp frame. */
extern bool gdbarch_get_longjmp_target_p (struct gdbarch *gdbarch);
gdb: pass frames as `const frame_info_ptr &` We currently pass frames to function by value, as `frame_info_ptr`. This is somewhat expensive: - the size of `frame_info_ptr` is 64 bytes, which is a bit big to pass by value - the constructors and destructor link/unlink the object in the global `frame_info_ptr::frame_list` list. This is an `intrusive_list`, so it's not so bad: it's just assigning a few points, there's no memory allocation as if it was `std::list`, but still it's useless to do that over and over. As suggested by Tom Tromey, change many function signatures to accept `const frame_info_ptr &` instead of `frame_info_ptr`. Some functions reassign their `frame_info_ptr` parameter, like: void the_func (frame_info_ptr frame) { for (; frame != nullptr; frame = get_prev_frame (frame)) { ... } } I wondered what to do about them, do I leave them as-is or change them (and need to introduce a separate local variable that can be re-assigned). I opted for the later for consistency. It might not be clear why some functions take `const frame_info_ptr &` while others take `frame_info_ptr`. Also, if a function took a `frame_info_ptr` because it did re-assign its parameter, I doubt that we would think to change it to `const frame_info_ptr &` should the implementation change such that it doesn't need to take `frame_info_ptr` anymore. It seems better to have a simple rule and apply it everywhere. Change-Id: I59d10addef687d157f82ccf4d54f5dde9a963fd0 Approved-By: Andrew Burgess <aburgess@redhat.com>
2024-02-20 02:07:47 +08:00
typedef int (gdbarch_get_longjmp_target_ftype) (const frame_info_ptr &frame, CORE_ADDR *pc);
extern int gdbarch_get_longjmp_target (struct gdbarch *gdbarch, const frame_info_ptr &frame, CORE_ADDR *pc);
Split gdbarch.h into two files This patch splits gdbarch.h into two files -- gdbarch.h now is editable and hand-maintained, and the new gdbarch-gen.h file is the only thing generated by gdbarch.sh. This lets us avoid maintaining boilerplate in the gdbarch.sh file. Note that gdbarch.sh still generates gdbarch.h after this patch. This makes it easier to re-run when rebasing. This code is removed in a subsequent patch.
2021-12-15 07:48:45 +08:00
extern void set_gdbarch_get_longjmp_target (struct gdbarch *gdbarch, gdbarch_get_longjmp_target_ftype *get_longjmp_target);
extern int gdbarch_believe_pcc_promotion (struct gdbarch *gdbarch);
extern void set_gdbarch_believe_pcc_promotion (struct gdbarch *gdbarch, int believe_pcc_promotion);
typedef int (gdbarch_convert_register_p_ftype) (struct gdbarch *gdbarch, int regnum, struct type *type);
extern int gdbarch_convert_register_p (struct gdbarch *gdbarch, int regnum, struct type *type);
extern void set_gdbarch_convert_register_p (struct gdbarch *gdbarch, gdbarch_convert_register_p_ftype *convert_register_p);
gdb: pass frames as `const frame_info_ptr &` We currently pass frames to function by value, as `frame_info_ptr`. This is somewhat expensive: - the size of `frame_info_ptr` is 64 bytes, which is a bit big to pass by value - the constructors and destructor link/unlink the object in the global `frame_info_ptr::frame_list` list. This is an `intrusive_list`, so it's not so bad: it's just assigning a few points, there's no memory allocation as if it was `std::list`, but still it's useless to do that over and over. As suggested by Tom Tromey, change many function signatures to accept `const frame_info_ptr &` instead of `frame_info_ptr`. Some functions reassign their `frame_info_ptr` parameter, like: void the_func (frame_info_ptr frame) { for (; frame != nullptr; frame = get_prev_frame (frame)) { ... } } I wondered what to do about them, do I leave them as-is or change them (and need to introduce a separate local variable that can be re-assigned). I opted for the later for consistency. It might not be clear why some functions take `const frame_info_ptr &` while others take `frame_info_ptr`. Also, if a function took a `frame_info_ptr` because it did re-assign its parameter, I doubt that we would think to change it to `const frame_info_ptr &` should the implementation change such that it doesn't need to take `frame_info_ptr` anymore. It seems better to have a simple rule and apply it everywhere. Change-Id: I59d10addef687d157f82ccf4d54f5dde9a963fd0 Approved-By: Andrew Burgess <aburgess@redhat.com>
2024-02-20 02:07:47 +08:00
typedef int (gdbarch_register_to_value_ftype) (const frame_info_ptr &frame, int regnum, struct type *type, gdb_byte *buf, int *optimizedp, int *unavailablep);
extern int gdbarch_register_to_value (struct gdbarch *gdbarch, const frame_info_ptr &frame, int regnum, struct type *type, gdb_byte *buf, int *optimizedp, int *unavailablep);
Split gdbarch.h into two files This patch splits gdbarch.h into two files -- gdbarch.h now is editable and hand-maintained, and the new gdbarch-gen.h file is the only thing generated by gdbarch.sh. This lets us avoid maintaining boilerplate in the gdbarch.sh file. Note that gdbarch.sh still generates gdbarch.h after this patch. This makes it easier to re-run when rebasing. This code is removed in a subsequent patch.
2021-12-15 07:48:45 +08:00
extern void set_gdbarch_register_to_value (struct gdbarch *gdbarch, gdbarch_register_to_value_ftype *register_to_value);
gdb: pass frames as `const frame_info_ptr &` We currently pass frames to function by value, as `frame_info_ptr`. This is somewhat expensive: - the size of `frame_info_ptr` is 64 bytes, which is a bit big to pass by value - the constructors and destructor link/unlink the object in the global `frame_info_ptr::frame_list` list. This is an `intrusive_list`, so it's not so bad: it's just assigning a few points, there's no memory allocation as if it was `std::list`, but still it's useless to do that over and over. As suggested by Tom Tromey, change many function signatures to accept `const frame_info_ptr &` instead of `frame_info_ptr`. Some functions reassign their `frame_info_ptr` parameter, like: void the_func (frame_info_ptr frame) { for (; frame != nullptr; frame = get_prev_frame (frame)) { ... } } I wondered what to do about them, do I leave them as-is or change them (and need to introduce a separate local variable that can be re-assigned). I opted for the later for consistency. It might not be clear why some functions take `const frame_info_ptr &` while others take `frame_info_ptr`. Also, if a function took a `frame_info_ptr` because it did re-assign its parameter, I doubt that we would think to change it to `const frame_info_ptr &` should the implementation change such that it doesn't need to take `frame_info_ptr` anymore. It seems better to have a simple rule and apply it everywhere. Change-Id: I59d10addef687d157f82ccf4d54f5dde9a963fd0 Approved-By: Andrew Burgess <aburgess@redhat.com>
2024-02-20 02:07:47 +08:00
typedef void (gdbarch_value_to_register_ftype) (const frame_info_ptr &frame, int regnum, struct type *type, const gdb_byte *buf);
extern void gdbarch_value_to_register (struct gdbarch *gdbarch, const frame_info_ptr &frame, int regnum, struct type *type, const gdb_byte *buf);
Split gdbarch.h into two files This patch splits gdbarch.h into two files -- gdbarch.h now is editable and hand-maintained, and the new gdbarch-gen.h file is the only thing generated by gdbarch.sh. This lets us avoid maintaining boilerplate in the gdbarch.sh file. Note that gdbarch.sh still generates gdbarch.h after this patch. This makes it easier to re-run when rebasing. This code is removed in a subsequent patch.
2021-12-15 07:48:45 +08:00
extern void set_gdbarch_value_to_register (struct gdbarch *gdbarch, gdbarch_value_to_register_ftype *value_to_register);
/* Construct a value representing the contents of register REGNUM in
gdb: pass frame_info_ptr to gdbarch_value_from_register Pass a frame_info_ptr rather than a frame_id. This avoids having to do a frame lookup on the callee side, when we can just pass the frame down directly. I think this fixes a bug in rs6000-tdep.c where the id of the wrong frame was set to `VALUE_NEXT_FRAME_ID (v)`. Change-Id: I77039bc87ea8fc5262f16d0e1446515efa21c565
2023-12-22 00:03:02 +08:00
frame THIS_FRAME, interpreted as type TYPE. The routine needs to
Split gdbarch.h into two files This patch splits gdbarch.h into two files -- gdbarch.h now is editable and hand-maintained, and the new gdbarch-gen.h file is the only thing generated by gdbarch.sh. This lets us avoid maintaining boilerplate in the gdbarch.sh file. Note that gdbarch.sh still generates gdbarch.h after this patch. This makes it easier to re-run when rebasing. This code is removed in a subsequent patch.
2021-12-15 07:48:45 +08:00
allocate and return a struct value with all value attributes
(but not the value contents) filled in. */
gdb: pass frame_info_ptr to gdbarch_value_from_register Pass a frame_info_ptr rather than a frame_id. This avoids having to do a frame lookup on the callee side, when we can just pass the frame down directly. I think this fixes a bug in rs6000-tdep.c where the id of the wrong frame was set to `VALUE_NEXT_FRAME_ID (v)`. Change-Id: I77039bc87ea8fc5262f16d0e1446515efa21c565
2023-12-22 00:03:02 +08:00
typedef struct value * (gdbarch_value_from_register_ftype) (struct gdbarch *gdbarch, struct type *type, int regnum, const frame_info_ptr &this_frame);
extern struct value * gdbarch_value_from_register (struct gdbarch *gdbarch, struct type *type, int regnum, const frame_info_ptr &this_frame);
Split gdbarch.h into two files This patch splits gdbarch.h into two files -- gdbarch.h now is editable and hand-maintained, and the new gdbarch-gen.h file is the only thing generated by gdbarch.sh. This lets us avoid maintaining boilerplate in the gdbarch.sh file. Note that gdbarch.sh still generates gdbarch.h after this patch. This makes it easier to re-run when rebasing. This code is removed in a subsequent patch.
2021-12-15 07:48:45 +08:00
extern void set_gdbarch_value_from_register (struct gdbarch *gdbarch, gdbarch_value_from_register_ftype *value_from_register);
typedef CORE_ADDR (gdbarch_pointer_to_address_ftype) (struct gdbarch *gdbarch, struct type *type, const gdb_byte *buf);
extern CORE_ADDR gdbarch_pointer_to_address (struct gdbarch *gdbarch, struct type *type, const gdb_byte *buf);
extern void set_gdbarch_pointer_to_address (struct gdbarch *gdbarch, gdbarch_pointer_to_address_ftype *pointer_to_address);
typedef void (gdbarch_address_to_pointer_ftype) (struct gdbarch *gdbarch, struct type *type, gdb_byte *buf, CORE_ADDR addr);
extern void gdbarch_address_to_pointer (struct gdbarch *gdbarch, struct type *type, gdb_byte *buf, CORE_ADDR addr);
extern void set_gdbarch_address_to_pointer (struct gdbarch *gdbarch, gdbarch_address_to_pointer_ftype *address_to_pointer);
extern bool gdbarch_integer_to_address_p (struct gdbarch *gdbarch);
typedef CORE_ADDR (gdbarch_integer_to_address_ftype) (struct gdbarch *gdbarch, struct type *type, const gdb_byte *buf);
extern CORE_ADDR gdbarch_integer_to_address (struct gdbarch *gdbarch, struct type *type, const gdb_byte *buf);
extern void set_gdbarch_integer_to_address (struct gdbarch *gdbarch, gdbarch_integer_to_address_ftype *integer_to_address);
/* Return the return-value convention that will be used by FUNCTION
to return a value of type VALTYPE. FUNCTION may be NULL in which
case the return convention is computed based only on VALTYPE.
Remove gdbarch.sh This patch runs gdbarch.py and removes gdbarch.sh.
2021-12-16 05:38:30 +08:00
Split gdbarch.h into two files This patch splits gdbarch.h into two files -- gdbarch.h now is editable and hand-maintained, and the new gdbarch-gen.h file is the only thing generated by gdbarch.sh. This lets us avoid maintaining boilerplate in the gdbarch.sh file. Note that gdbarch.sh still generates gdbarch.h after this patch. This makes it easier to re-run when rebasing. This code is removed in a subsequent patch.
2021-12-15 07:48:45 +08:00
If READBUF is not NULL, extract the return value and save it in this buffer.
Remove gdbarch.sh This patch runs gdbarch.py and removes gdbarch.sh.
2021-12-16 05:38:30 +08:00
Split gdbarch.h into two files This patch splits gdbarch.h into two files -- gdbarch.h now is editable and hand-maintained, and the new gdbarch-gen.h file is the only thing generated by gdbarch.sh. This lets us avoid maintaining boilerplate in the gdbarch.sh file. Note that gdbarch.sh still generates gdbarch.h after this patch. This makes it easier to re-run when rebasing. This code is removed in a subsequent patch.
2021-12-15 07:48:45 +08:00
If WRITEBUF is not NULL, it contains a return value which will be
stored into the appropriate register. This can be used when we want
to force the value returned by a function (see the "return" command
Add new overload of gdbarch_return_value The gdbarch "return_value" can't correctly handle variably-sized types. The problem here is that the TYPE_LENGTH of such a type is 0, until the type is resolved, which requires reading memory. However, gdbarch_return_value only accepts a buffer as an out parameter. Fixing this requires letting the implementation of the gdbarch method resolve the type and return a value -- that is, both the contents and the new type. After an attempt at this, I realized I wouldn't be able to correctly update all implementations (there are ~80) of this method. So, instead, this patch adds a new method that falls back to the current method, and it updates gdb to only call the new method. This way it's possible to incrementally convert the architectures that I am able to test.
2022-09-07 22:39:52 +08:00
for instance).
Split gdbarch.h into two files This patch splits gdbarch.h into two files -- gdbarch.h now is editable and hand-maintained, and the new gdbarch-gen.h file is the only thing generated by gdbarch.sh. This lets us avoid maintaining boilerplate in the gdbarch.sh file. Note that gdbarch.sh still generates gdbarch.h after this patch. This makes it easier to re-run when rebasing. This code is removed in a subsequent patch.
2021-12-15 07:48:45 +08:00
Add new overload of gdbarch_return_value The gdbarch "return_value" can't correctly handle variably-sized types. The problem here is that the TYPE_LENGTH of such a type is 0, until the type is resolved, which requires reading memory. However, gdbarch_return_value only accepts a buffer as an out parameter. Fixing this requires letting the implementation of the gdbarch method resolve the type and return a value -- that is, both the contents and the new type. After an attempt at this, I realized I wouldn't be able to correctly update all implementations (there are ~80) of this method. So, instead, this patch adds a new method that falls back to the current method, and it updates gdb to only call the new method. This way it's possible to incrementally convert the architectures that I am able to test.
2022-09-07 22:39:52 +08:00
NOTE: it is better to implement return_value_as_value instead, as that
method can properly handle variably-sized types. */
Split gdbarch.h into two files This patch splits gdbarch.h into two files -- gdbarch.h now is editable and hand-maintained, and the new gdbarch-gen.h file is the only thing generated by gdbarch.sh. This lets us avoid maintaining boilerplate in the gdbarch.sh file. Note that gdbarch.sh still generates gdbarch.h after this patch. This makes it easier to re-run when rebasing. This code is removed in a subsequent patch.
2021-12-15 07:48:45 +08:00
typedef enum return_value_convention (gdbarch_return_value_ftype) (struct gdbarch *gdbarch, struct value *function, struct type *valtype, struct regcache *regcache, gdb_byte *readbuf, const gdb_byte *writebuf);
extern void set_gdbarch_return_value (struct gdbarch *gdbarch, gdbarch_return_value_ftype *return_value);
Add new overload of gdbarch_return_value The gdbarch "return_value" can't correctly handle variably-sized types. The problem here is that the TYPE_LENGTH of such a type is 0, until the type is resolved, which requires reading memory. However, gdbarch_return_value only accepts a buffer as an out parameter. Fixing this requires letting the implementation of the gdbarch method resolve the type and return a value -- that is, both the contents and the new type. After an attempt at this, I realized I wouldn't be able to correctly update all implementations (there are ~80) of this method. So, instead, this patch adds a new method that falls back to the current method, and it updates gdb to only call the new method. This way it's possible to incrementally convert the architectures that I am able to test.
2022-09-07 22:39:52 +08:00
/* Return the return-value convention that will be used by FUNCTION
to return a value of type VALTYPE. FUNCTION may be NULL in which
case the return convention is computed based only on VALTYPE.
If READ_VALUE is not NULL, extract the return value and save it in
this pointer.
If WRITEBUF is not NULL, it contains a return value which will be
stored into the appropriate register. This can be used when we want
to force the value returned by a function (see the "return" command
for instance). */
typedef enum return_value_convention (gdbarch_return_value_as_value_ftype) (struct gdbarch *gdbarch, struct value *function, struct type *valtype, struct regcache *regcache, struct value **read_value, const gdb_byte *writebuf);
extern enum return_value_convention gdbarch_return_value_as_value (struct gdbarch *gdbarch, struct value *function, struct type *valtype, struct regcache *regcache, struct value **read_value, const gdb_byte *writebuf);
extern void set_gdbarch_return_value_as_value (struct gdbarch *gdbarch, gdbarch_return_value_as_value_ftype *return_value_as_value);
PowerPC, fix support for printing the function return value for non-trivial values. Currently, a non-trivial return value from a function cannot currently be reliably determined on PowerPC. This is due to the fact that the PowerPC ABI uses register r3 to store the address of the buffer containing the non-trivial return value when the function is called. The PowerPC ABI does not guarantee the value in register r3 is not modified in the function. Thus the value in r3 cannot be reliably used to obtain the return addreses on exit from the function. This patch adds a new gdbarch method to allow PowerPC to access the value of r3 on entry to a function. On PowerPC, the new gdbarch method attempts to use the DW_OP_entry_value for the DWARF entries, when exiting the function, to determine the value of r3 on entry to the function. This requires the use of the -fvar-tracking compiler option to compile the user application thus generating the DW_OP_entry_value in the binary. The DW_OP_entry_value entries in the binary file allows GDB to resolve the DW_TAG_call_site entries. This new gdbarch method is used to get the return buffer address, in the case of a function returning a nontrivial data type, on exit from the function. The GDB function should_stop checks to see if RETURN_BUF is non-zero. By default, RETURN_BUF will be set to zero by the new gdbarch method call for all architectures except PowerPC. The get_return_value function will be used to obtain the return value on all other architectures as is currently being done if RETURN_BUF is zero. On PowerPC, the new gdbarch method will return a nonzero address in RETURN_BUF if the value can be determined. The value_at function uses the return buffer address to get the return value. This patch fixes five testcase failures in gdb.cp/non-trivial-retval.exp. The correct function return values are now reported. Note this patch is dependent on patch: "PowerPC, function ppc64_sysv_abi_return_value add missing return value convention". This patch has been tested on Power 10 and x86-64 with no regressions.
2022-11-15 05:22:37 +08:00
/* Return the address at which the value being returned from
the current function will be stored. This routine is only
called if the current function uses the the "struct return
convention".
May return 0 when unable to determine that address. */
gdb: pass frames as `const frame_info_ptr &` We currently pass frames to function by value, as `frame_info_ptr`. This is somewhat expensive: - the size of `frame_info_ptr` is 64 bytes, which is a bit big to pass by value - the constructors and destructor link/unlink the object in the global `frame_info_ptr::frame_list` list. This is an `intrusive_list`, so it's not so bad: it's just assigning a few points, there's no memory allocation as if it was `std::list`, but still it's useless to do that over and over. As suggested by Tom Tromey, change many function signatures to accept `const frame_info_ptr &` instead of `frame_info_ptr`. Some functions reassign their `frame_info_ptr` parameter, like: void the_func (frame_info_ptr frame) { for (; frame != nullptr; frame = get_prev_frame (frame)) { ... } } I wondered what to do about them, do I leave them as-is or change them (and need to introduce a separate local variable that can be re-assigned). I opted for the later for consistency. It might not be clear why some functions take `const frame_info_ptr &` while others take `frame_info_ptr`. Also, if a function took a `frame_info_ptr` because it did re-assign its parameter, I doubt that we would think to change it to `const frame_info_ptr &` should the implementation change such that it doesn't need to take `frame_info_ptr` anymore. It seems better to have a simple rule and apply it everywhere. Change-Id: I59d10addef687d157f82ccf4d54f5dde9a963fd0 Approved-By: Andrew Burgess <aburgess@redhat.com>
2024-02-20 02:07:47 +08:00
typedef CORE_ADDR (gdbarch_get_return_buf_addr_ftype) (struct type *val_type, const frame_info_ptr &cur_frame);
extern CORE_ADDR gdbarch_get_return_buf_addr (struct gdbarch *gdbarch, struct type *val_type, const frame_info_ptr &cur_frame);
PowerPC, fix support for printing the function return value for non-trivial values. Currently, a non-trivial return value from a function cannot currently be reliably determined on PowerPC. This is due to the fact that the PowerPC ABI uses register r3 to store the address of the buffer containing the non-trivial return value when the function is called. The PowerPC ABI does not guarantee the value in register r3 is not modified in the function. Thus the value in r3 cannot be reliably used to obtain the return addreses on exit from the function. This patch adds a new gdbarch method to allow PowerPC to access the value of r3 on entry to a function. On PowerPC, the new gdbarch method attempts to use the DW_OP_entry_value for the DWARF entries, when exiting the function, to determine the value of r3 on entry to the function. This requires the use of the -fvar-tracking compiler option to compile the user application thus generating the DW_OP_entry_value in the binary. The DW_OP_entry_value entries in the binary file allows GDB to resolve the DW_TAG_call_site entries. This new gdbarch method is used to get the return buffer address, in the case of a function returning a nontrivial data type, on exit from the function. The GDB function should_stop checks to see if RETURN_BUF is non-zero. By default, RETURN_BUF will be set to zero by the new gdbarch method call for all architectures except PowerPC. The get_return_value function will be used to obtain the return value on all other architectures as is currently being done if RETURN_BUF is zero. On PowerPC, the new gdbarch method will return a nonzero address in RETURN_BUF if the value can be determined. The value_at function uses the return buffer address to get the return value. This patch fixes five testcase failures in gdb.cp/non-trivial-retval.exp. The correct function return values are now reported. Note this patch is dependent on patch: "PowerPC, function ppc64_sysv_abi_return_value add missing return value convention". This patch has been tested on Power 10 and x86-64 with no regressions.
2022-11-15 05:22:37 +08:00
extern void set_gdbarch_get_return_buf_addr (struct gdbarch *gdbarch, gdbarch_get_return_buf_addr_ftype *get_return_buf_addr);
PowerPC: fix _Float128 type output string PowerPC supports two 128-bit floating point formats, the IBM long double and IEEE 128-bit float. The issue is the DWARF information does not distinguish between the two. There have been proposals of how to extend the DWARF information as discussed in https://gcc.gnu.org/bugzilla/show_bug.cgi?id=104194 but has not been fully implemented. GCC introduced the _Float128 internal type as a work around for the issue. The workaround is not transparent to GDB. The internal _Float128 type name is printed rather then the user specified long double type. This patch adds a new gdbarch method to allow PowerPC to detect the GCC workaround. The workaround checks for "_Float128" name when reading the base typedef from the die_info. If the workaround is detected, the type and format fields from the _Float128 typedef are copied to the long double typedef. The same is done for the complex long double typedef. This patch fixes 74 regression test failures in gdb.base/whatis-ptype-typedefs.exp on PowerPC with IEEE float 128 as the default on GCC. It fixes one regression test failure in gdb.base/complex-parts.exp. The patch has been tested on Power 10 where GCC defaults to IEEE Float 128-bit and on Power 10 where GCC defaults to the IBM 128-bit float. The patch as also been tested on X86-64 with no new regression failures.
2023-03-24 06:23:05 +08:00
/* Return true if the typedef record needs to be replaced.".
Return 0 by default */
typedef bool (gdbarch_dwarf2_omit_typedef_p_ftype) (struct type *target_type, const char *producer, const char *name);
extern bool gdbarch_dwarf2_omit_typedef_p (struct gdbarch *gdbarch, struct type *target_type, const char *producer, const char *name);
extern void set_gdbarch_dwarf2_omit_typedef_p (struct gdbarch *gdbarch, gdbarch_dwarf2_omit_typedef_p_ftype *dwarf2_omit_typedef_p);
Handle erroneous DW_AT_call_return_pc On PPC64, with the test case included in an earlier patch, we found that "finish" would still not correctly find the return value via entry values. The issue is simple. The compiler emits: 0x00000000100032b8 <+28>: bl 0x1000320c <pck__create_large> 0x00000000100032bc <+32>: nop 0x00000000100032c0 <+36>: li r9,42 ... but the DWARF says: <162a> DW_AT_call_return_pc: 0x100032c0 That is, the declared return PC is one instruction past the actual return PC. This patch adds a new arch hook to handle this scenario, and implements it for PPC64. Some care is taken so that GDB will continue to work if this compiler bug is fixed. A GCC patch is here: https://gcc.gnu.org/pipermail/gcc-patches/2023-March/613336.html No check for 'nop' is done, as subsequent discussion revealed that the linker might replace this with another instruction.
2023-03-09 01:58:35 +08:00
/* Update PC when trying to find a call site. This is useful on
architectures where the call site PC, as reported in the DWARF, can be
incorrect for some reason.
The passed-in PC will be an address in the inferior. GDB will have
already failed to find a call site at this PC. This function may
simply return its parameter if it thinks that should be the correct
address. */
typedef CORE_ADDR (gdbarch_update_call_site_pc_ftype) (struct gdbarch *gdbarch, CORE_ADDR pc);
extern CORE_ADDR gdbarch_update_call_site_pc (struct gdbarch *gdbarch, CORE_ADDR pc);
extern void set_gdbarch_update_call_site_pc (struct gdbarch *gdbarch, gdbarch_update_call_site_pc_ftype *update_call_site_pc);
Split gdbarch.h into two files This patch splits gdbarch.h into two files -- gdbarch.h now is editable and hand-maintained, and the new gdbarch-gen.h file is the only thing generated by gdbarch.sh. This lets us avoid maintaining boilerplate in the gdbarch.sh file. Note that gdbarch.sh still generates gdbarch.h after this patch. This makes it easier to re-run when rebasing. This code is removed in a subsequent patch.
2021-12-15 07:48:45 +08:00
/* Return true if the return value of function is stored in the first hidden
parameter. In theory, this feature should be language-dependent, specified
by language and its ABI, such as C++. Unfortunately, compiler may
implement it to a target-dependent feature. So that we need such hook here
to be aware of this in GDB. */
typedef int (gdbarch_return_in_first_hidden_param_p_ftype) (struct gdbarch *gdbarch, struct type *type);
extern int gdbarch_return_in_first_hidden_param_p (struct gdbarch *gdbarch, struct type *type);
extern void set_gdbarch_return_in_first_hidden_param_p (struct gdbarch *gdbarch, gdbarch_return_in_first_hidden_param_p_ftype *return_in_first_hidden_param_p);
typedef CORE_ADDR (gdbarch_skip_prologue_ftype) (struct gdbarch *gdbarch, CORE_ADDR ip);
extern CORE_ADDR gdbarch_skip_prologue (struct gdbarch *gdbarch, CORE_ADDR ip);
extern void set_gdbarch_skip_prologue (struct gdbarch *gdbarch, gdbarch_skip_prologue_ftype *skip_prologue);
extern bool gdbarch_skip_main_prologue_p (struct gdbarch *gdbarch);
typedef CORE_ADDR (gdbarch_skip_main_prologue_ftype) (struct gdbarch *gdbarch, CORE_ADDR ip);
extern CORE_ADDR gdbarch_skip_main_prologue (struct gdbarch *gdbarch, CORE_ADDR ip);
extern void set_gdbarch_skip_main_prologue (struct gdbarch *gdbarch, gdbarch_skip_main_prologue_ftype *skip_main_prologue);
/* On some platforms, a single function may provide multiple entry points,
e.g. one that is used for function-pointer calls and a different one
that is used for direct function calls.
In order to ensure that breakpoints set on the function will trigger
no matter via which entry point the function is entered, a platform
may provide the skip_entrypoint callback. It is called with IP set
to the main entry point of a function (as determined by the symbol table),
and should return the address of the innermost entry point, where the
actual breakpoint needs to be set. Note that skip_entrypoint is used
by GDB common code even when debugging optimized code, where skip_prologue
is not used. */
extern bool gdbarch_skip_entrypoint_p (struct gdbarch *gdbarch);
typedef CORE_ADDR (gdbarch_skip_entrypoint_ftype) (struct gdbarch *gdbarch, CORE_ADDR ip);
extern CORE_ADDR gdbarch_skip_entrypoint (struct gdbarch *gdbarch, CORE_ADDR ip);
extern void set_gdbarch_skip_entrypoint (struct gdbarch *gdbarch, gdbarch_skip_entrypoint_ftype *skip_entrypoint);
typedef int (gdbarch_inner_than_ftype) (CORE_ADDR lhs, CORE_ADDR rhs);
extern int gdbarch_inner_than (struct gdbarch *gdbarch, CORE_ADDR lhs, CORE_ADDR rhs);
extern void set_gdbarch_inner_than (struct gdbarch *gdbarch, gdbarch_inner_than_ftype *inner_than);
typedef const gdb_byte * (gdbarch_breakpoint_from_pc_ftype) (struct gdbarch *gdbarch, CORE_ADDR *pcptr, int *lenptr);
extern const gdb_byte * gdbarch_breakpoint_from_pc (struct gdbarch *gdbarch, CORE_ADDR *pcptr, int *lenptr);
extern void set_gdbarch_breakpoint_from_pc (struct gdbarch *gdbarch, gdbarch_breakpoint_from_pc_ftype *breakpoint_from_pc);
/* Return the breakpoint kind for this target based on *PCPTR. */
typedef int (gdbarch_breakpoint_kind_from_pc_ftype) (struct gdbarch *gdbarch, CORE_ADDR *pcptr);
extern int gdbarch_breakpoint_kind_from_pc (struct gdbarch *gdbarch, CORE_ADDR *pcptr);
extern void set_gdbarch_breakpoint_kind_from_pc (struct gdbarch *gdbarch, gdbarch_breakpoint_kind_from_pc_ftype *breakpoint_kind_from_pc);
/* Return the software breakpoint from KIND. KIND can have target
specific meaning like the Z0 kind parameter.
SIZE is set to the software breakpoint's length in memory. */
typedef const gdb_byte * (gdbarch_sw_breakpoint_from_kind_ftype) (struct gdbarch *gdbarch, int kind, int *size);
extern const gdb_byte * gdbarch_sw_breakpoint_from_kind (struct gdbarch *gdbarch, int kind, int *size);
extern void set_gdbarch_sw_breakpoint_from_kind (struct gdbarch *gdbarch, gdbarch_sw_breakpoint_from_kind_ftype *sw_breakpoint_from_kind);
/* Return the breakpoint kind for this target based on the current
processor state (e.g. the current instruction mode on ARM) and the
*PCPTR. In default, it is gdbarch->breakpoint_kind_from_pc. */
typedef int (gdbarch_breakpoint_kind_from_current_state_ftype) (struct gdbarch *gdbarch, struct regcache *regcache, CORE_ADDR *pcptr);
extern int gdbarch_breakpoint_kind_from_current_state (struct gdbarch *gdbarch, struct regcache *regcache, CORE_ADDR *pcptr);
extern void set_gdbarch_breakpoint_kind_from_current_state (struct gdbarch *gdbarch, gdbarch_breakpoint_kind_from_current_state_ftype *breakpoint_kind_from_current_state);
extern bool gdbarch_adjust_breakpoint_address_p (struct gdbarch *gdbarch);
typedef CORE_ADDR (gdbarch_adjust_breakpoint_address_ftype) (struct gdbarch *gdbarch, CORE_ADDR bpaddr);
extern CORE_ADDR gdbarch_adjust_breakpoint_address (struct gdbarch *gdbarch, CORE_ADDR bpaddr);
extern void set_gdbarch_adjust_breakpoint_address (struct gdbarch *gdbarch, gdbarch_adjust_breakpoint_address_ftype *adjust_breakpoint_address);
typedef int (gdbarch_memory_insert_breakpoint_ftype) (struct gdbarch *gdbarch, struct bp_target_info *bp_tgt);
extern int gdbarch_memory_insert_breakpoint (struct gdbarch *gdbarch, struct bp_target_info *bp_tgt);
extern void set_gdbarch_memory_insert_breakpoint (struct gdbarch *gdbarch, gdbarch_memory_insert_breakpoint_ftype *memory_insert_breakpoint);
typedef int (gdbarch_memory_remove_breakpoint_ftype) (struct gdbarch *gdbarch, struct bp_target_info *bp_tgt);
extern int gdbarch_memory_remove_breakpoint (struct gdbarch *gdbarch, struct bp_target_info *bp_tgt);
extern void set_gdbarch_memory_remove_breakpoint (struct gdbarch *gdbarch, gdbarch_memory_remove_breakpoint_ftype *memory_remove_breakpoint);
extern CORE_ADDR gdbarch_decr_pc_after_break (struct gdbarch *gdbarch);
extern void set_gdbarch_decr_pc_after_break (struct gdbarch *gdbarch, CORE_ADDR decr_pc_after_break);
[gdb] Fix grammar in comments and docs Fix grammar in some comments and docs: - machines that doesn't -> machines that don't - its a -> it's a - its the -> it's the - if does its not -> if it does it's not - one more instructions if doesn't match -> one more instruction if it doesn't match - it's own -> its own - it's first -> its first - it's pointer -> its pointer I also came across "it's performance" in gdb/stubs/*-stub.c in the HP public domain notice, I've left that alone. Tested on x86_64-linux.
2023-06-05 18:53:15 +08:00
/* A function can be addressed by either its "pointer" (possibly a
Split gdbarch.h into two files This patch splits gdbarch.h into two files -- gdbarch.h now is editable and hand-maintained, and the new gdbarch-gen.h file is the only thing generated by gdbarch.sh. This lets us avoid maintaining boilerplate in the gdbarch.sh file. Note that gdbarch.sh still generates gdbarch.h after this patch. This makes it easier to re-run when rebasing. This code is removed in a subsequent patch.
2021-12-15 07:48:45 +08:00
descriptor address) or "entry point" (first executable instruction).
The method "convert_from_func_ptr_addr" converting the former to the
latter. gdbarch_deprecated_function_start_offset is being used to implement
a simplified subset of that functionality - the function's address
corresponds to the "function pointer" and the function's start
corresponds to the "function entry point" - and hence is redundant. */
extern CORE_ADDR gdbarch_deprecated_function_start_offset (struct gdbarch *gdbarch);
extern void set_gdbarch_deprecated_function_start_offset (struct gdbarch *gdbarch, CORE_ADDR deprecated_function_start_offset);
/* Return the remote protocol register number associated with this
register. Normally the identity mapping. */
typedef int (gdbarch_remote_register_number_ftype) (struct gdbarch *gdbarch, int regno);
extern int gdbarch_remote_register_number (struct gdbarch *gdbarch, int regno);
extern void set_gdbarch_remote_register_number (struct gdbarch *gdbarch, gdbarch_remote_register_number_ftype *remote_register_number);
/* Fetch the target specific address used to represent a load module. */
extern bool gdbarch_fetch_tls_load_module_address_p (struct gdbarch *gdbarch);
typedef CORE_ADDR (gdbarch_fetch_tls_load_module_address_ftype) (struct objfile *objfile);
extern CORE_ADDR gdbarch_fetch_tls_load_module_address (struct gdbarch *gdbarch, struct objfile *objfile);
extern void set_gdbarch_fetch_tls_load_module_address (struct gdbarch *gdbarch, gdbarch_fetch_tls_load_module_address_ftype *fetch_tls_load_module_address);
/* Return the thread-local address at OFFSET in the thread-local
storage for the thread PTID and the shared library or executable
file given by LM_ADDR. If that block of thread-local storage hasn't
been allocated yet, this function may throw an error. LM_ADDR may
be zero for statically linked multithreaded inferiors. */
extern bool gdbarch_get_thread_local_address_p (struct gdbarch *gdbarch);
typedef CORE_ADDR (gdbarch_get_thread_local_address_ftype) (struct gdbarch *gdbarch, ptid_t ptid, CORE_ADDR lm_addr, CORE_ADDR offset);
extern CORE_ADDR gdbarch_get_thread_local_address (struct gdbarch *gdbarch, ptid_t ptid, CORE_ADDR lm_addr, CORE_ADDR offset);
extern void set_gdbarch_get_thread_local_address (struct gdbarch *gdbarch, gdbarch_get_thread_local_address_ftype *get_thread_local_address);
extern CORE_ADDR gdbarch_frame_args_skip (struct gdbarch *gdbarch);
extern void set_gdbarch_frame_args_skip (struct gdbarch *gdbarch, CORE_ADDR frame_args_skip);
gdb: pass frames as `const frame_info_ptr &` We currently pass frames to function by value, as `frame_info_ptr`. This is somewhat expensive: - the size of `frame_info_ptr` is 64 bytes, which is a bit big to pass by value - the constructors and destructor link/unlink the object in the global `frame_info_ptr::frame_list` list. This is an `intrusive_list`, so it's not so bad: it's just assigning a few points, there's no memory allocation as if it was `std::list`, but still it's useless to do that over and over. As suggested by Tom Tromey, change many function signatures to accept `const frame_info_ptr &` instead of `frame_info_ptr`. Some functions reassign their `frame_info_ptr` parameter, like: void the_func (frame_info_ptr frame) { for (; frame != nullptr; frame = get_prev_frame (frame)) { ... } } I wondered what to do about them, do I leave them as-is or change them (and need to introduce a separate local variable that can be re-assigned). I opted for the later for consistency. It might not be clear why some functions take `const frame_info_ptr &` while others take `frame_info_ptr`. Also, if a function took a `frame_info_ptr` because it did re-assign its parameter, I doubt that we would think to change it to `const frame_info_ptr &` should the implementation change such that it doesn't need to take `frame_info_ptr` anymore. It seems better to have a simple rule and apply it everywhere. Change-Id: I59d10addef687d157f82ccf4d54f5dde9a963fd0 Approved-By: Andrew Burgess <aburgess@redhat.com>
2024-02-20 02:07:47 +08:00
typedef CORE_ADDR (gdbarch_unwind_pc_ftype) (struct gdbarch *gdbarch, const frame_info_ptr &next_frame);
extern CORE_ADDR gdbarch_unwind_pc (struct gdbarch *gdbarch, const frame_info_ptr &next_frame);
Split gdbarch.h into two files This patch splits gdbarch.h into two files -- gdbarch.h now is editable and hand-maintained, and the new gdbarch-gen.h file is the only thing generated by gdbarch.sh. This lets us avoid maintaining boilerplate in the gdbarch.sh file. Note that gdbarch.sh still generates gdbarch.h after this patch. This makes it easier to re-run when rebasing. This code is removed in a subsequent patch.
2021-12-15 07:48:45 +08:00
extern void set_gdbarch_unwind_pc (struct gdbarch *gdbarch, gdbarch_unwind_pc_ftype *unwind_pc);
gdb: pass frames as `const frame_info_ptr &` We currently pass frames to function by value, as `frame_info_ptr`. This is somewhat expensive: - the size of `frame_info_ptr` is 64 bytes, which is a bit big to pass by value - the constructors and destructor link/unlink the object in the global `frame_info_ptr::frame_list` list. This is an `intrusive_list`, so it's not so bad: it's just assigning a few points, there's no memory allocation as if it was `std::list`, but still it's useless to do that over and over. As suggested by Tom Tromey, change many function signatures to accept `const frame_info_ptr &` instead of `frame_info_ptr`. Some functions reassign their `frame_info_ptr` parameter, like: void the_func (frame_info_ptr frame) { for (; frame != nullptr; frame = get_prev_frame (frame)) { ... } } I wondered what to do about them, do I leave them as-is or change them (and need to introduce a separate local variable that can be re-assigned). I opted for the later for consistency. It might not be clear why some functions take `const frame_info_ptr &` while others take `frame_info_ptr`. Also, if a function took a `frame_info_ptr` because it did re-assign its parameter, I doubt that we would think to change it to `const frame_info_ptr &` should the implementation change such that it doesn't need to take `frame_info_ptr` anymore. It seems better to have a simple rule and apply it everywhere. Change-Id: I59d10addef687d157f82ccf4d54f5dde9a963fd0 Approved-By: Andrew Burgess <aburgess@redhat.com>
2024-02-20 02:07:47 +08:00
typedef CORE_ADDR (gdbarch_unwind_sp_ftype) (struct gdbarch *gdbarch, const frame_info_ptr &next_frame);
extern CORE_ADDR gdbarch_unwind_sp (struct gdbarch *gdbarch, const frame_info_ptr &next_frame);
Split gdbarch.h into two files This patch splits gdbarch.h into two files -- gdbarch.h now is editable and hand-maintained, and the new gdbarch-gen.h file is the only thing generated by gdbarch.sh. This lets us avoid maintaining boilerplate in the gdbarch.sh file. Note that gdbarch.sh still generates gdbarch.h after this patch. This makes it easier to re-run when rebasing. This code is removed in a subsequent patch.
2021-12-15 07:48:45 +08:00
extern void set_gdbarch_unwind_sp (struct gdbarch *gdbarch, gdbarch_unwind_sp_ftype *unwind_sp);
/* DEPRECATED_FRAME_LOCALS_ADDRESS as been replaced by the per-frame
frame-base. Enable frame-base before frame-unwind. */
extern bool gdbarch_frame_num_args_p (struct gdbarch *gdbarch);
gdb: pass frames as `const frame_info_ptr &` We currently pass frames to function by value, as `frame_info_ptr`. This is somewhat expensive: - the size of `frame_info_ptr` is 64 bytes, which is a bit big to pass by value - the constructors and destructor link/unlink the object in the global `frame_info_ptr::frame_list` list. This is an `intrusive_list`, so it's not so bad: it's just assigning a few points, there's no memory allocation as if it was `std::list`, but still it's useless to do that over and over. As suggested by Tom Tromey, change many function signatures to accept `const frame_info_ptr &` instead of `frame_info_ptr`. Some functions reassign their `frame_info_ptr` parameter, like: void the_func (frame_info_ptr frame) { for (; frame != nullptr; frame = get_prev_frame (frame)) { ... } } I wondered what to do about them, do I leave them as-is or change them (and need to introduce a separate local variable that can be re-assigned). I opted for the later for consistency. It might not be clear why some functions take `const frame_info_ptr &` while others take `frame_info_ptr`. Also, if a function took a `frame_info_ptr` because it did re-assign its parameter, I doubt that we would think to change it to `const frame_info_ptr &` should the implementation change such that it doesn't need to take `frame_info_ptr` anymore. It seems better to have a simple rule and apply it everywhere. Change-Id: I59d10addef687d157f82ccf4d54f5dde9a963fd0 Approved-By: Andrew Burgess <aburgess@redhat.com>
2024-02-20 02:07:47 +08:00
typedef int (gdbarch_frame_num_args_ftype) (const frame_info_ptr &frame);
extern int gdbarch_frame_num_args (struct gdbarch *gdbarch, const frame_info_ptr &frame);
Split gdbarch.h into two files This patch splits gdbarch.h into two files -- gdbarch.h now is editable and hand-maintained, and the new gdbarch-gen.h file is the only thing generated by gdbarch.sh. This lets us avoid maintaining boilerplate in the gdbarch.sh file. Note that gdbarch.sh still generates gdbarch.h after this patch. This makes it easier to re-run when rebasing. This code is removed in a subsequent patch.
2021-12-15 07:48:45 +08:00
extern void set_gdbarch_frame_num_args (struct gdbarch *gdbarch, gdbarch_frame_num_args_ftype *frame_num_args);
extern bool gdbarch_frame_align_p (struct gdbarch *gdbarch);
typedef CORE_ADDR (gdbarch_frame_align_ftype) (struct gdbarch *gdbarch, CORE_ADDR address);
extern CORE_ADDR gdbarch_frame_align (struct gdbarch *gdbarch, CORE_ADDR address);
extern void set_gdbarch_frame_align (struct gdbarch *gdbarch, gdbarch_frame_align_ftype *frame_align);
typedef int (gdbarch_stabs_argument_has_addr_ftype) (struct gdbarch *gdbarch, struct type *type);
extern int gdbarch_stabs_argument_has_addr (struct gdbarch *gdbarch, struct type *type);
extern void set_gdbarch_stabs_argument_has_addr (struct gdbarch *gdbarch, gdbarch_stabs_argument_has_addr_ftype *stabs_argument_has_addr);
extern int gdbarch_frame_red_zone_size (struct gdbarch *gdbarch);
extern void set_gdbarch_frame_red_zone_size (struct gdbarch *gdbarch, int frame_red_zone_size);
typedef CORE_ADDR (gdbarch_convert_from_func_ptr_addr_ftype) (struct gdbarch *gdbarch, CORE_ADDR addr, struct target_ops *targ);
extern CORE_ADDR gdbarch_convert_from_func_ptr_addr (struct gdbarch *gdbarch, CORE_ADDR addr, struct target_ops *targ);
extern void set_gdbarch_convert_from_func_ptr_addr (struct gdbarch *gdbarch, gdbarch_convert_from_func_ptr_addr_ftype *convert_from_func_ptr_addr);
/* On some machines there are bits in addresses which are not really
part of the address, but are used by the kernel, the hardware, etc.
for special purposes. gdbarch_addr_bits_remove takes out any such bits so
we get a "real" address such as one would find in a symbol table.
This is used only for addresses of instructions, and even then I'm
not sure it's used in all contexts. It exists to deal with there
being a few stray bits in the PC which would mislead us, not as some
sort of generic thing to handle alignment or segmentation (it's
possible it should be in TARGET_READ_PC instead). */
typedef CORE_ADDR (gdbarch_addr_bits_remove_ftype) (struct gdbarch *gdbarch, CORE_ADDR addr);
extern CORE_ADDR gdbarch_addr_bits_remove (struct gdbarch *gdbarch, CORE_ADDR addr);
extern void set_gdbarch_addr_bits_remove (struct gdbarch *gdbarch, gdbarch_addr_bits_remove_ftype *addr_bits_remove);
[aarch64] Fix removal of non-address bits for PAuth PR gdb/28947 The address_significant gdbarch setting was introduced as a way to remove non-address bits from pointers, and it is specified by a constant. This constant represents the number of address bits in a pointer. Right now AArch64 is the only architecture that uses it, and 56 was a correct option so far. But if we are using Pointer Authentication (PAuth), we might use up to 2 bytes from the address space to store the required information. We could also have cases where we're using both PAuth and MTE. We could adjust the constant to 48 to cover those cases, but this doesn't cover the case where GDB needs to sign-extend kernel addresses after removal of the non-address bits. This has worked so far because bit 55 is used to select between kernel-space and user-space addresses. But trying to clear a range of bits crossing the bit 55 boundary requires the hook to be smarter. The following patch renames the gdbarch hook from significant_addr_bit to remove_non_address_bits and passes a pointer as opposed to the number of bits. The hook is now responsible for removing the required non-address bits and sign-extending the address if needed. While at it, make GDB and GDBServer share some more code for aarch64 and add a new arch-specific testcase gdb.arch/aarch64-non-address-bits.exp. Bug-url: https://sourceware.org/bugzilla/show_bug.cgi?id=28947 Approved-By: Simon Marchi <simon.marchi@efficios.com>
2022-05-25 06:31:09 +08:00
/* On some architectures, not all bits of a pointer are significant.
On AArch64, for example, the top bits of a pointer may carry a "tag", which
can be ignored by the kernel and the hardware. The "tag" can be regarded as
additional data associated with the pointer, but it is not part of the address.
Given a pointer for the architecture, this hook removes all the
non-significant bits and sign-extends things as needed. It gets used to remove
non-address bits from data pointers (for example, removing the AArch64 MTE tag
bits from a pointer) and from code pointers (removing the AArch64 PAC signature
from a pointer containing the return address). */
typedef CORE_ADDR (gdbarch_remove_non_address_bits_ftype) (struct gdbarch *gdbarch, CORE_ADDR pointer);
extern CORE_ADDR gdbarch_remove_non_address_bits (struct gdbarch *gdbarch, CORE_ADDR pointer);
extern void set_gdbarch_remove_non_address_bits (struct gdbarch *gdbarch, gdbarch_remove_non_address_bits_ftype *remove_non_address_bits);
Split gdbarch.h into two files This patch splits gdbarch.h into two files -- gdbarch.h now is editable and hand-maintained, and the new gdbarch-gen.h file is the only thing generated by gdbarch.sh. This lets us avoid maintaining boilerplate in the gdbarch.sh file. Note that gdbarch.sh still generates gdbarch.h after this patch. This makes it easier to re-run when rebasing. This code is removed in a subsequent patch.
2021-12-15 07:48:45 +08:00
/* Return a string representation of the memory tag TAG. */
typedef std::string (gdbarch_memtag_to_string_ftype) (struct gdbarch *gdbarch, struct value *tag);
extern std::string gdbarch_memtag_to_string (struct gdbarch *gdbarch, struct value *tag);
extern void set_gdbarch_memtag_to_string (struct gdbarch *gdbarch, gdbarch_memtag_to_string_ftype *memtag_to_string);
/* Return true if ADDRESS contains a tag and false otherwise. ADDRESS
must be either a pointer or a reference type. */
gdb: Introduce is_address_tagged target hook This commit introduces a new target hook, target_is_address_tagged, which is used instead of the gdbarch_tagged_address_p gdbarch hook in the upper layer (printcmd.c). This change enables easy specialization of memory tagging address check per target in the future. As target_is_address_tagged continues to utilize the gdbarch_tagged_address_p hook, there is no change in behavior for all the targets that use the new target hook (i.e., the remote.c, aarch64-linux-nat.c, and corelow.c targets). Just the gdbarch_tagged_address_p signature is changed for convenience, since target_is_address_tagged takes the address to be checked as a CORE_ADDR type. Signed-off-by: Gustavo Romero <gustavo.romero@linaro.org> Approved-By: Luis Machado <luis.machado@arm.com> Tested-By: Luis Machado <luis.machado@arm.com>
2024-04-19 04:10:36 +08:00
typedef bool (gdbarch_tagged_address_p_ftype) (struct gdbarch *gdbarch, CORE_ADDR address);
extern bool gdbarch_tagged_address_p (struct gdbarch *gdbarch, CORE_ADDR address);
Split gdbarch.h into two files This patch splits gdbarch.h into two files -- gdbarch.h now is editable and hand-maintained, and the new gdbarch-gen.h file is the only thing generated by gdbarch.sh. This lets us avoid maintaining boilerplate in the gdbarch.sh file. Note that gdbarch.sh still generates gdbarch.h after this patch. This makes it easier to re-run when rebasing. This code is removed in a subsequent patch.
2021-12-15 07:48:45 +08:00
extern void set_gdbarch_tagged_address_p (struct gdbarch *gdbarch, gdbarch_tagged_address_p_ftype *tagged_address_p);
/* Return true if the tag from ADDRESS matches the memory tag for that
particular address. Return false otherwise. */
typedef bool (gdbarch_memtag_matches_p_ftype) (struct gdbarch *gdbarch, struct value *address);
extern bool gdbarch_memtag_matches_p (struct gdbarch *gdbarch, struct value *address);
extern void set_gdbarch_memtag_matches_p (struct gdbarch *gdbarch, gdbarch_memtag_matches_p_ftype *memtag_matches_p);
/* Set the tags of type TAG_TYPE, for the memory address range
[ADDRESS, ADDRESS + LENGTH) to TAGS.
Return true if successful and false otherwise. */
typedef bool (gdbarch_set_memtags_ftype) (struct gdbarch *gdbarch, struct value *address, size_t length, const gdb::byte_vector &tags, memtag_type tag_type);
extern bool gdbarch_set_memtags (struct gdbarch *gdbarch, struct value *address, size_t length, const gdb::byte_vector &tags, memtag_type tag_type);
extern void set_gdbarch_set_memtags (struct gdbarch *gdbarch, gdbarch_set_memtags_ftype *set_memtags);
/* Return the tag of type TAG_TYPE associated with the memory address ADDRESS,
assuming ADDRESS is tagged. */
typedef struct value * (gdbarch_get_memtag_ftype) (struct gdbarch *gdbarch, struct value *address, memtag_type tag_type);
extern struct value * gdbarch_get_memtag (struct gdbarch *gdbarch, struct value *address, memtag_type tag_type);
extern void set_gdbarch_get_memtag (struct gdbarch *gdbarch, gdbarch_get_memtag_ftype *get_memtag);
/* memtag_granule_size is the size of the allocation tag granule, for
architectures that support memory tagging.
This is 0 for architectures that do not support memory tagging.
For a non-zero value, this represents the number of bytes of memory per tag. */
extern CORE_ADDR gdbarch_memtag_granule_size (struct gdbarch *gdbarch);
extern void set_gdbarch_memtag_granule_size (struct gdbarch *gdbarch, CORE_ADDR memtag_granule_size);
/* FIXME/cagney/2001-01-18: This should be split in two. A target method that
indicates if the target needs software single step. An ISA method to
implement it.
Remove gdbarch.sh This patch runs gdbarch.py and removes gdbarch.sh.
2021-12-16 05:38:30 +08:00
Split gdbarch.h into two files This patch splits gdbarch.h into two files -- gdbarch.h now is editable and hand-maintained, and the new gdbarch-gen.h file is the only thing generated by gdbarch.sh. This lets us avoid maintaining boilerplate in the gdbarch.sh file. Note that gdbarch.sh still generates gdbarch.h after this patch. This makes it easier to re-run when rebasing. This code is removed in a subsequent patch.
2021-12-15 07:48:45 +08:00
FIXME/cagney/2001-01-18: The logic is backwards. It should be asking if the
target can single step. If not, then implement single step using breakpoints.
Remove gdbarch.sh This patch runs gdbarch.py and removes gdbarch.sh.
2021-12-16 05:38:30 +08:00
Split gdbarch.h into two files This patch splits gdbarch.h into two files -- gdbarch.h now is editable and hand-maintained, and the new gdbarch-gen.h file is the only thing generated by gdbarch.sh. This lets us avoid maintaining boilerplate in the gdbarch.sh file. Note that gdbarch.sh still generates gdbarch.h after this patch. This makes it easier to re-run when rebasing. This code is removed in a subsequent patch.
2021-12-15 07:48:45 +08:00
Return a vector of addresses on which the software single step
breakpoints should be inserted. NULL means software single step is
not used.
Multiple breakpoints may be inserted for some instructions such as
conditional branch. However, each implementation must always evaluate
the condition and only put the breakpoint at the branch destination if
the condition is true, so that we ensure forward progress when stepping
past a conditional branch to self. */
extern bool gdbarch_software_single_step_p (struct gdbarch *gdbarch);
typedef std::vector<CORE_ADDR> (gdbarch_software_single_step_ftype) (struct regcache *regcache);
extern std::vector<CORE_ADDR> gdbarch_software_single_step (struct gdbarch *gdbarch, struct regcache *regcache);
extern void set_gdbarch_software_single_step (struct gdbarch *gdbarch, gdbarch_software_single_step_ftype *software_single_step);
/* Return non-zero if the processor is executing a delay slot and a
further single-step is needed before the instruction finishes. */
extern bool gdbarch_single_step_through_delay_p (struct gdbarch *gdbarch);
gdb: pass frames as `const frame_info_ptr &` We currently pass frames to function by value, as `frame_info_ptr`. This is somewhat expensive: - the size of `frame_info_ptr` is 64 bytes, which is a bit big to pass by value - the constructors and destructor link/unlink the object in the global `frame_info_ptr::frame_list` list. This is an `intrusive_list`, so it's not so bad: it's just assigning a few points, there's no memory allocation as if it was `std::list`, but still it's useless to do that over and over. As suggested by Tom Tromey, change many function signatures to accept `const frame_info_ptr &` instead of `frame_info_ptr`. Some functions reassign their `frame_info_ptr` parameter, like: void the_func (frame_info_ptr frame) { for (; frame != nullptr; frame = get_prev_frame (frame)) { ... } } I wondered what to do about them, do I leave them as-is or change them (and need to introduce a separate local variable that can be re-assigned). I opted for the later for consistency. It might not be clear why some functions take `const frame_info_ptr &` while others take `frame_info_ptr`. Also, if a function took a `frame_info_ptr` because it did re-assign its parameter, I doubt that we would think to change it to `const frame_info_ptr &` should the implementation change such that it doesn't need to take `frame_info_ptr` anymore. It seems better to have a simple rule and apply it everywhere. Change-Id: I59d10addef687d157f82ccf4d54f5dde9a963fd0 Approved-By: Andrew Burgess <aburgess@redhat.com>
2024-02-20 02:07:47 +08:00
typedef int (gdbarch_single_step_through_delay_ftype) (struct gdbarch *gdbarch, const frame_info_ptr &frame);
extern int gdbarch_single_step_through_delay (struct gdbarch *gdbarch, const frame_info_ptr &frame);
Split gdbarch.h into two files This patch splits gdbarch.h into two files -- gdbarch.h now is editable and hand-maintained, and the new gdbarch-gen.h file is the only thing generated by gdbarch.sh. This lets us avoid maintaining boilerplate in the gdbarch.sh file. Note that gdbarch.sh still generates gdbarch.h after this patch. This makes it easier to re-run when rebasing. This code is removed in a subsequent patch.
2021-12-15 07:48:45 +08:00
extern void set_gdbarch_single_step_through_delay (struct gdbarch *gdbarch, gdbarch_single_step_through_delay_ftype *single_step_through_delay);
/* FIXME: cagney/2003-08-28: Need to find a better way of selecting the
disassembler. Perhaps objdump can handle it? */
typedef int (gdbarch_print_insn_ftype) (bfd_vma vma, struct disassemble_info *info);
extern int gdbarch_print_insn (struct gdbarch *gdbarch, bfd_vma vma, struct disassemble_info *info);
extern void set_gdbarch_print_insn (struct gdbarch *gdbarch, gdbarch_print_insn_ftype *print_insn);
gdb: pass frames as `const frame_info_ptr &` We currently pass frames to function by value, as `frame_info_ptr`. This is somewhat expensive: - the size of `frame_info_ptr` is 64 bytes, which is a bit big to pass by value - the constructors and destructor link/unlink the object in the global `frame_info_ptr::frame_list` list. This is an `intrusive_list`, so it's not so bad: it's just assigning a few points, there's no memory allocation as if it was `std::list`, but still it's useless to do that over and over. As suggested by Tom Tromey, change many function signatures to accept `const frame_info_ptr &` instead of `frame_info_ptr`. Some functions reassign their `frame_info_ptr` parameter, like: void the_func (frame_info_ptr frame) { for (; frame != nullptr; frame = get_prev_frame (frame)) { ... } } I wondered what to do about them, do I leave them as-is or change them (and need to introduce a separate local variable that can be re-assigned). I opted for the later for consistency. It might not be clear why some functions take `const frame_info_ptr &` while others take `frame_info_ptr`. Also, if a function took a `frame_info_ptr` because it did re-assign its parameter, I doubt that we would think to change it to `const frame_info_ptr &` should the implementation change such that it doesn't need to take `frame_info_ptr` anymore. It seems better to have a simple rule and apply it everywhere. Change-Id: I59d10addef687d157f82ccf4d54f5dde9a963fd0 Approved-By: Andrew Burgess <aburgess@redhat.com>
2024-02-20 02:07:47 +08:00
typedef CORE_ADDR (gdbarch_skip_trampoline_code_ftype) (const frame_info_ptr &frame, CORE_ADDR pc);
extern CORE_ADDR gdbarch_skip_trampoline_code (struct gdbarch *gdbarch, const frame_info_ptr &frame, CORE_ADDR pc);
Split gdbarch.h into two files This patch splits gdbarch.h into two files -- gdbarch.h now is editable and hand-maintained, and the new gdbarch-gen.h file is the only thing generated by gdbarch.sh. This lets us avoid maintaining boilerplate in the gdbarch.sh file. Note that gdbarch.sh still generates gdbarch.h after this patch. This makes it easier to re-run when rebasing. This code is removed in a subsequent patch.
2021-12-15 07:48:45 +08:00
extern void set_gdbarch_skip_trampoline_code (struct gdbarch *gdbarch, gdbarch_skip_trampoline_code_ftype *skip_trampoline_code);
Move solib_ops into gdbarch This changs solib_ops to be an ordinary gdbarch value and updates all the uses. This removes a longstanding FIXME and makes the code somewhat cleaner as well.
2022-08-09 00:41:43 +08:00
/* Vtable of solib operations functions. */
gdb: rename target_so_ops to solib_ops I don't like the name `target_so_ops`, because: - The name `target` is so overloaded, and in this case it's not even related to target_ops or anything else called "target". - We do have an implementation that actually fetches solibs from the target (solib_target_so_op in solib-target.c), so it's confusing for the "base class" to be called target_something as well. Rename to solib_ops. Change-Id: I46a983d44e81400470e22deb09aaf26ad8a3587f Approved-By: Tom Tromey <tom@tromey.com>
2024-02-06 04:18:34 +08:00
extern const solib_ops * gdbarch_so_ops (struct gdbarch *gdbarch);
extern void set_gdbarch_so_ops (struct gdbarch *gdbarch, const solib_ops * so_ops);
Move solib_ops into gdbarch This changs solib_ops to be an ordinary gdbarch value and updates all the uses. This removes a longstanding FIXME and makes the code somewhat cleaner as well.
2022-08-09 00:41:43 +08:00
Split gdbarch.h into two files This patch splits gdbarch.h into two files -- gdbarch.h now is editable and hand-maintained, and the new gdbarch-gen.h file is the only thing generated by gdbarch.sh. This lets us avoid maintaining boilerplate in the gdbarch.sh file. Note that gdbarch.sh still generates gdbarch.h after this patch. This makes it easier to re-run when rebasing. This code is removed in a subsequent patch.
2021-12-15 07:48:45 +08:00
/* If in_solib_dynsym_resolve_code() returns true, and SKIP_SOLIB_RESOLVER
evaluates non-zero, this is the address where the debugger will place
a step-resume breakpoint to get us past the dynamic linker. */
typedef CORE_ADDR (gdbarch_skip_solib_resolver_ftype) (struct gdbarch *gdbarch, CORE_ADDR pc);
extern CORE_ADDR gdbarch_skip_solib_resolver (struct gdbarch *gdbarch, CORE_ADDR pc);
extern void set_gdbarch_skip_solib_resolver (struct gdbarch *gdbarch, gdbarch_skip_solib_resolver_ftype *skip_solib_resolver);
/* Some systems also have trampoline code for returning from shared libs. */
typedef int (gdbarch_in_solib_return_trampoline_ftype) (struct gdbarch *gdbarch, CORE_ADDR pc, const char *name);
extern int gdbarch_in_solib_return_trampoline (struct gdbarch *gdbarch, CORE_ADDR pc, const char *name);
extern void set_gdbarch_in_solib_return_trampoline (struct gdbarch *gdbarch, gdbarch_in_solib_return_trampoline_ftype *in_solib_return_trampoline);
/* Return true if PC lies inside an indirect branch thunk. */
typedef bool (gdbarch_in_indirect_branch_thunk_ftype) (struct gdbarch *gdbarch, CORE_ADDR pc);
extern bool gdbarch_in_indirect_branch_thunk (struct gdbarch *gdbarch, CORE_ADDR pc);
extern void set_gdbarch_in_indirect_branch_thunk (struct gdbarch *gdbarch, gdbarch_in_indirect_branch_thunk_ftype *in_indirect_branch_thunk);
/* A target might have problems with watchpoints as soon as the stack
frame of the current function has been destroyed. This mostly happens
as the first action in a function's epilogue. stack_frame_destroyed_p()
is defined to return a non-zero value if either the given addr is one
instruction after the stack destroying instruction up to the trailing
return instruction or if we can figure out that the stack frame has
already been invalidated regardless of the value of addr. Targets
which don't suffer from that problem could just let this functionality
untouched. */
typedef int (gdbarch_stack_frame_destroyed_p_ftype) (struct gdbarch *gdbarch, CORE_ADDR addr);
extern int gdbarch_stack_frame_destroyed_p (struct gdbarch *gdbarch, CORE_ADDR addr);
extern void set_gdbarch_stack_frame_destroyed_p (struct gdbarch *gdbarch, gdbarch_stack_frame_destroyed_p_ftype *stack_frame_destroyed_p);
/* Process an ELF symbol in the minimal symbol table in a backend-specific
way. Normally this hook is supposed to do nothing, however if required,
then this hook can be used to apply tranformations to symbols that are
considered special in some way. For example the MIPS backend uses it
to interpret `st_other' information to mark compressed code symbols so
that they can be treated in the appropriate manner in the processing of
the main symbol table and DWARF-2 records. */
extern bool gdbarch_elf_make_msymbol_special_p (struct gdbarch *gdbarch);
typedef void (gdbarch_elf_make_msymbol_special_ftype) (asymbol *sym, struct minimal_symbol *msym);
extern void gdbarch_elf_make_msymbol_special (struct gdbarch *gdbarch, asymbol *sym, struct minimal_symbol *msym);
extern void set_gdbarch_elf_make_msymbol_special (struct gdbarch *gdbarch, gdbarch_elf_make_msymbol_special_ftype *elf_make_msymbol_special);
typedef void (gdbarch_coff_make_msymbol_special_ftype) (int val, struct minimal_symbol *msym);
extern void gdbarch_coff_make_msymbol_special (struct gdbarch *gdbarch, int val, struct minimal_symbol *msym);
extern void set_gdbarch_coff_make_msymbol_special (struct gdbarch *gdbarch, gdbarch_coff_make_msymbol_special_ftype *coff_make_msymbol_special);
/* Process a symbol in the main symbol table in a backend-specific way.
Normally this hook is supposed to do nothing, however if required,
then this hook can be used to apply tranformations to symbols that
are considered special in some way. This is currently used by the
MIPS backend to make sure compressed code symbols have the ISA bit
set. This in turn is needed for symbol values seen in GDB to match
the values used at the runtime by the program itself, for function
and label references. */
typedef void (gdbarch_make_symbol_special_ftype) (struct symbol *sym, struct objfile *objfile);
extern void gdbarch_make_symbol_special (struct gdbarch *gdbarch, struct symbol *sym, struct objfile *objfile);
extern void set_gdbarch_make_symbol_special (struct gdbarch *gdbarch, gdbarch_make_symbol_special_ftype *make_symbol_special);
/* Adjust the address retrieved from a DWARF-2 record other than a line
entry in a backend-specific way. Normally this hook is supposed to
return the address passed unchanged, however if that is incorrect for
any reason, then this hook can be used to fix the address up in the
required manner. This is currently used by the MIPS backend to make
sure addresses in FDE, range records, etc. referring to compressed
code have the ISA bit set, matching line information and the symbol
table. */
typedef CORE_ADDR (gdbarch_adjust_dwarf2_addr_ftype) (CORE_ADDR pc);
extern CORE_ADDR gdbarch_adjust_dwarf2_addr (struct gdbarch *gdbarch, CORE_ADDR pc);
extern void set_gdbarch_adjust_dwarf2_addr (struct gdbarch *gdbarch, gdbarch_adjust_dwarf2_addr_ftype *adjust_dwarf2_addr);
/* Adjust the address updated by a line entry in a backend-specific way.
Normally this hook is supposed to return the address passed unchanged,
however in the case of inconsistencies in these records, this hook can
be used to fix them up in the required manner. This is currently used
by the MIPS backend to make sure all line addresses in compressed code
are presented with the ISA bit set, which is not always the case. This
in turn ensures breakpoint addresses are correctly matched against the
stop PC. */
typedef CORE_ADDR (gdbarch_adjust_dwarf2_line_ftype) (CORE_ADDR addr, int rel);
extern CORE_ADDR gdbarch_adjust_dwarf2_line (struct gdbarch *gdbarch, CORE_ADDR addr, int rel);
extern void set_gdbarch_adjust_dwarf2_line (struct gdbarch *gdbarch, gdbarch_adjust_dwarf2_line_ftype *adjust_dwarf2_line);
extern int gdbarch_cannot_step_breakpoint (struct gdbarch *gdbarch);
extern void set_gdbarch_cannot_step_breakpoint (struct gdbarch *gdbarch, int cannot_step_breakpoint);
/* See comment in target.h about continuable, steppable and
non-steppable watchpoints. */
extern int gdbarch_have_nonsteppable_watchpoint (struct gdbarch *gdbarch);
extern void set_gdbarch_have_nonsteppable_watchpoint (struct gdbarch *gdbarch, int have_nonsteppable_watchpoint);
extern bool gdbarch_address_class_type_flags_p (struct gdbarch *gdbarch);
typedef type_instance_flags (gdbarch_address_class_type_flags_ftype) (int byte_size, int dwarf2_addr_class);
extern type_instance_flags gdbarch_address_class_type_flags (struct gdbarch *gdbarch, int byte_size, int dwarf2_addr_class);
extern void set_gdbarch_address_class_type_flags (struct gdbarch *gdbarch, gdbarch_address_class_type_flags_ftype *address_class_type_flags);
extern bool gdbarch_address_class_type_flags_to_name_p (struct gdbarch *gdbarch);
typedef const char * (gdbarch_address_class_type_flags_to_name_ftype) (struct gdbarch *gdbarch, type_instance_flags type_flags);
extern const char * gdbarch_address_class_type_flags_to_name (struct gdbarch *gdbarch, type_instance_flags type_flags);
extern void set_gdbarch_address_class_type_flags_to_name (struct gdbarch *gdbarch, gdbarch_address_class_type_flags_to_name_ftype *address_class_type_flags_to_name);
/* Execute vendor-specific DWARF Call Frame Instruction. OP is the instruction.
FS are passed from the generic execute_cfa_program function. */
typedef bool (gdbarch_execute_dwarf_cfa_vendor_op_ftype) (struct gdbarch *gdbarch, gdb_byte op, struct dwarf2_frame_state *fs);
extern bool gdbarch_execute_dwarf_cfa_vendor_op (struct gdbarch *gdbarch, gdb_byte op, struct dwarf2_frame_state *fs);
extern void set_gdbarch_execute_dwarf_cfa_vendor_op (struct gdbarch *gdbarch, gdbarch_execute_dwarf_cfa_vendor_op_ftype *execute_dwarf_cfa_vendor_op);
/* Return the appropriate type_flags for the supplied address class.
This function should return true if the address class was recognized and
type_flags was set, false otherwise. */
extern bool gdbarch_address_class_name_to_type_flags_p (struct gdbarch *gdbarch);
typedef bool (gdbarch_address_class_name_to_type_flags_ftype) (struct gdbarch *gdbarch, const char *name, type_instance_flags *type_flags_ptr);
extern bool gdbarch_address_class_name_to_type_flags (struct gdbarch *gdbarch, const char *name, type_instance_flags *type_flags_ptr);
extern void set_gdbarch_address_class_name_to_type_flags (struct gdbarch *gdbarch, gdbarch_address_class_name_to_type_flags_ftype *address_class_name_to_type_flags);
/* Is a register in a group */
gdb: make gdbarch_register_reggroup_p take a const reggroup * Change gdbarch_register_reggroup_p to take a 'const struct reggroup *' argument. This requires a change to the gdb/gdbarch-components.py script, regeneration of gdbarch.{c,h}, and then updates to all the architectures that implement this method. There should be no user visible changes after this commit.
2022-03-31 19:36:06 +08:00
typedef int (gdbarch_register_reggroup_p_ftype) (struct gdbarch *gdbarch, int regnum, const struct reggroup *reggroup);
extern int gdbarch_register_reggroup_p (struct gdbarch *gdbarch, int regnum, const struct reggroup *reggroup);
Split gdbarch.h into two files This patch splits gdbarch.h into two files -- gdbarch.h now is editable and hand-maintained, and the new gdbarch-gen.h file is the only thing generated by gdbarch.sh. This lets us avoid maintaining boilerplate in the gdbarch.sh file. Note that gdbarch.sh still generates gdbarch.h after this patch. This makes it easier to re-run when rebasing. This code is removed in a subsequent patch.
2021-12-15 07:48:45 +08:00
extern void set_gdbarch_register_reggroup_p (struct gdbarch *gdbarch, gdbarch_register_reggroup_p_ftype *register_reggroup_p);
/* Fetch the pointer to the ith function argument. */
extern bool gdbarch_fetch_pointer_argument_p (struct gdbarch *gdbarch);
gdb: pass frames as `const frame_info_ptr &` We currently pass frames to function by value, as `frame_info_ptr`. This is somewhat expensive: - the size of `frame_info_ptr` is 64 bytes, which is a bit big to pass by value - the constructors and destructor link/unlink the object in the global `frame_info_ptr::frame_list` list. This is an `intrusive_list`, so it's not so bad: it's just assigning a few points, there's no memory allocation as if it was `std::list`, but still it's useless to do that over and over. As suggested by Tom Tromey, change many function signatures to accept `const frame_info_ptr &` instead of `frame_info_ptr`. Some functions reassign their `frame_info_ptr` parameter, like: void the_func (frame_info_ptr frame) { for (; frame != nullptr; frame = get_prev_frame (frame)) { ... } } I wondered what to do about them, do I leave them as-is or change them (and need to introduce a separate local variable that can be re-assigned). I opted for the later for consistency. It might not be clear why some functions take `const frame_info_ptr &` while others take `frame_info_ptr`. Also, if a function took a `frame_info_ptr` because it did re-assign its parameter, I doubt that we would think to change it to `const frame_info_ptr &` should the implementation change such that it doesn't need to take `frame_info_ptr` anymore. It seems better to have a simple rule and apply it everywhere. Change-Id: I59d10addef687d157f82ccf4d54f5dde9a963fd0 Approved-By: Andrew Burgess <aburgess@redhat.com>
2024-02-20 02:07:47 +08:00
typedef CORE_ADDR (gdbarch_fetch_pointer_argument_ftype) (const frame_info_ptr &frame, int argi, struct type *type);
extern CORE_ADDR gdbarch_fetch_pointer_argument (struct gdbarch *gdbarch, const frame_info_ptr &frame, int argi, struct type *type);
Split gdbarch.h into two files This patch splits gdbarch.h into two files -- gdbarch.h now is editable and hand-maintained, and the new gdbarch-gen.h file is the only thing generated by gdbarch.sh. This lets us avoid maintaining boilerplate in the gdbarch.sh file. Note that gdbarch.sh still generates gdbarch.h after this patch. This makes it easier to re-run when rebasing. This code is removed in a subsequent patch.
2021-12-15 07:48:45 +08:00
extern void set_gdbarch_fetch_pointer_argument (struct gdbarch *gdbarch, gdbarch_fetch_pointer_argument_ftype *fetch_pointer_argument);
/* Iterate over all supported register notes in a core file. For each
supported register note section, the iterator must call CB and pass
CB_DATA unchanged. If REGCACHE is not NULL, the iterator can limit
the supported register note sections based on the current register
values. Otherwise it should enumerate all supported register note
sections. */
extern bool gdbarch_iterate_over_regset_sections_p (struct gdbarch *gdbarch);
typedef void (gdbarch_iterate_over_regset_sections_ftype) (struct gdbarch *gdbarch, iterate_over_regset_sections_cb *cb, void *cb_data, const struct regcache *regcache);
extern void gdbarch_iterate_over_regset_sections (struct gdbarch *gdbarch, iterate_over_regset_sections_cb *cb, void *cb_data, const struct regcache *regcache);
extern void set_gdbarch_iterate_over_regset_sections (struct gdbarch *gdbarch, gdbarch_iterate_over_regset_sections_ftype *iterate_over_regset_sections);
/* Create core file notes */
extern bool gdbarch_make_corefile_notes_p (struct gdbarch *gdbarch);
typedef gdb::unique_xmalloc_ptr<char> (gdbarch_make_corefile_notes_ftype) (struct gdbarch *gdbarch, bfd *obfd, int *note_size);
extern gdb::unique_xmalloc_ptr<char> gdbarch_make_corefile_notes (struct gdbarch *gdbarch, bfd *obfd, int *note_size);
extern void set_gdbarch_make_corefile_notes (struct gdbarch *gdbarch, gdbarch_make_corefile_notes_ftype *make_corefile_notes);
/* Find core file memory regions */
extern bool gdbarch_find_memory_regions_p (struct gdbarch *gdbarch);
typedef int (gdbarch_find_memory_regions_ftype) (struct gdbarch *gdbarch, find_memory_region_ftype func, void *data);
extern int gdbarch_find_memory_regions (struct gdbarch *gdbarch, find_memory_region_ftype func, void *data);
extern void set_gdbarch_find_memory_regions (struct gdbarch *gdbarch, gdbarch_find_memory_regions_ftype *find_memory_regions);
[AArch64] MTE corefile support Teach GDB how to dump memory tags for AArch64 when using the gcore command and how to read memory tag data back from a core file generated by GDB (via gcore) or by the Linux kernel. The format is documented in the Linux Kernel documentation [1]. Each tagged memory range (listed in /proc/<pid>/smaps) gets dumped to its own PT_AARCH64_MEMTAG_MTE segment. A section named ".memtag" is created for each of those segments when reading the core file back. To save a little bit of space, given MTE tags only take 4 bits, the memory tags are stored packed as 2 tags per byte. When reading the data back, the tags are unpacked. I've added a new testcase to exercise the feature. Build-tested with --enable-targets=all and regression tested on aarch64-linux Ubuntu 20.04. [1] Documentation/arm64/memory-tagging-extension.rst (Core Dump Support)
2022-03-31 18:42:35 +08:00
/* Given a bfd OBFD, segment ADDRESS and SIZE, create a memory tag section to be dumped to a core file */
extern bool gdbarch_create_memtag_section_p (struct gdbarch *gdbarch);
typedef asection * (gdbarch_create_memtag_section_ftype) (struct gdbarch *gdbarch, bfd *obfd, CORE_ADDR address, size_t size);
extern asection * gdbarch_create_memtag_section (struct gdbarch *gdbarch, bfd *obfd, CORE_ADDR address, size_t size);
extern void set_gdbarch_create_memtag_section (struct gdbarch *gdbarch, gdbarch_create_memtag_section_ftype *create_memtag_section);
/* Given a memory tag section OSEC, fill OSEC's contents with the appropriate tag data */
extern bool gdbarch_fill_memtag_section_p (struct gdbarch *gdbarch);
typedef bool (gdbarch_fill_memtag_section_ftype) (struct gdbarch *gdbarch, asection *osec);
extern bool gdbarch_fill_memtag_section (struct gdbarch *gdbarch, asection *osec);
extern void set_gdbarch_fill_memtag_section (struct gdbarch *gdbarch, gdbarch_fill_memtag_section_ftype *fill_memtag_section);
/* Decode a memory tag SECTION and return the tags of type TYPE contained in
the memory range [ADDRESS, ADDRESS + LENGTH).
If no tags were found, return an empty vector. */
extern bool gdbarch_decode_memtag_section_p (struct gdbarch *gdbarch);
typedef gdb::byte_vector (gdbarch_decode_memtag_section_ftype) (struct gdbarch *gdbarch, bfd_section *section, int type, CORE_ADDR address, size_t length);
extern gdb::byte_vector gdbarch_decode_memtag_section (struct gdbarch *gdbarch, bfd_section *section, int type, CORE_ADDR address, size_t length);
extern void set_gdbarch_decode_memtag_section (struct gdbarch *gdbarch, gdbarch_decode_memtag_section_ftype *decode_memtag_section);
Split gdbarch.h into two files This patch splits gdbarch.h into two files -- gdbarch.h now is editable and hand-maintained, and the new gdbarch-gen.h file is the only thing generated by gdbarch.sh. This lets us avoid maintaining boilerplate in the gdbarch.sh file. Note that gdbarch.sh still generates gdbarch.h after this patch. This makes it easier to re-run when rebasing. This code is removed in a subsequent patch.
2021-12-15 07:48:45 +08:00
/* Read offset OFFSET of TARGET_OBJECT_LIBRARIES formatted shared libraries list from
core file into buffer READBUF with length LEN. Return the number of bytes read
(zero indicates failure).
failed, otherwise, return the red length of READBUF. */
extern bool gdbarch_core_xfer_shared_libraries_p (struct gdbarch *gdbarch);
typedef ULONGEST (gdbarch_core_xfer_shared_libraries_ftype) (struct gdbarch *gdbarch, gdb_byte *readbuf, ULONGEST offset, ULONGEST len);
extern ULONGEST gdbarch_core_xfer_shared_libraries (struct gdbarch *gdbarch, gdb_byte *readbuf, ULONGEST offset, ULONGEST len);
extern void set_gdbarch_core_xfer_shared_libraries (struct gdbarch *gdbarch, gdbarch_core_xfer_shared_libraries_ftype *core_xfer_shared_libraries);
/* Read offset OFFSET of TARGET_OBJECT_LIBRARIES_AIX formatted shared
libraries list from core file into buffer READBUF with length LEN.
Return the number of bytes read (zero indicates failure). */
extern bool gdbarch_core_xfer_shared_libraries_aix_p (struct gdbarch *gdbarch);
typedef ULONGEST (gdbarch_core_xfer_shared_libraries_aix_ftype) (struct gdbarch *gdbarch, gdb_byte *readbuf, ULONGEST offset, ULONGEST len);
extern ULONGEST gdbarch_core_xfer_shared_libraries_aix (struct gdbarch *gdbarch, gdb_byte *readbuf, ULONGEST offset, ULONGEST len);
extern void set_gdbarch_core_xfer_shared_libraries_aix (struct gdbarch *gdbarch, gdbarch_core_xfer_shared_libraries_aix_ftype *core_xfer_shared_libraries_aix);
/* How the core target converts a PTID from a core file to a string. */
extern bool gdbarch_core_pid_to_str_p (struct gdbarch *gdbarch);
typedef std::string (gdbarch_core_pid_to_str_ftype) (struct gdbarch *gdbarch, ptid_t ptid);
extern std::string gdbarch_core_pid_to_str (struct gdbarch *gdbarch, ptid_t ptid);
extern void set_gdbarch_core_pid_to_str (struct gdbarch *gdbarch, gdbarch_core_pid_to_str_ftype *core_pid_to_str);
/* How the core target extracts the name of a thread from a core file. */
extern bool gdbarch_core_thread_name_p (struct gdbarch *gdbarch);
typedef const char * (gdbarch_core_thread_name_ftype) (struct gdbarch *gdbarch, struct thread_info *thr);
extern const char * gdbarch_core_thread_name (struct gdbarch *gdbarch, struct thread_info *thr);
extern void set_gdbarch_core_thread_name (struct gdbarch *gdbarch, gdbarch_core_thread_name_ftype *core_thread_name);
/* Read offset OFFSET of TARGET_OBJECT_SIGNAL_INFO signal information
from core file into buffer READBUF with length LEN. Return the number
of bytes read (zero indicates EOF, a negative value indicates failure). */
extern bool gdbarch_core_xfer_siginfo_p (struct gdbarch *gdbarch);
typedef LONGEST (gdbarch_core_xfer_siginfo_ftype) (struct gdbarch *gdbarch, gdb_byte *readbuf, ULONGEST offset, ULONGEST len);
extern LONGEST gdbarch_core_xfer_siginfo (struct gdbarch *gdbarch, gdb_byte *readbuf, ULONGEST offset, ULONGEST len);
extern void set_gdbarch_core_xfer_siginfo (struct gdbarch *gdbarch, gdbarch_core_xfer_siginfo_ftype *core_xfer_siginfo);
core: Support fetching x86 XSAVE layout from architectures. Add gdbarch_core_read_x86_xsave_layout to fetch the x86 XSAVE layout structure from a core file. Current OS's do not export the offsets of XSAVE state components in core dumps, so provide an i387_guess_xsave_layout helper function to set offsets based on known combinations of XCR0 masks and total state sizes. Eventually when core dumps do contain this information this function should only be used as a fall back for older core dumps. Approved-By: Simon Marchi <simon.marchi@efficios.com>
2023-08-29 05:18:19 +08:00
/* Read x86 XSAVE layout information from core file into XSAVE_LAYOUT.
Returns true if the layout was read successfully. */
extern bool gdbarch_core_read_x86_xsave_layout_p (struct gdbarch *gdbarch);
typedef bool (gdbarch_core_read_x86_xsave_layout_ftype) (struct gdbarch *gdbarch, x86_xsave_layout &xsave_layout);
extern bool gdbarch_core_read_x86_xsave_layout (struct gdbarch *gdbarch, x86_xsave_layout &xsave_layout);
extern void set_gdbarch_core_read_x86_xsave_layout (struct gdbarch *gdbarch, gdbarch_core_read_x86_xsave_layout_ftype *core_read_x86_xsave_layout);
Split gdbarch.h into two files This patch splits gdbarch.h into two files -- gdbarch.h now is editable and hand-maintained, and the new gdbarch-gen.h file is the only thing generated by gdbarch.sh. This lets us avoid maintaining boilerplate in the gdbarch.sh file. Note that gdbarch.sh still generates gdbarch.h after this patch. This makes it easier to re-run when rebasing. This code is removed in a subsequent patch.
2021-12-15 07:48:45 +08:00
/* BFD target to use when generating a core file. */
extern bool gdbarch_gcore_bfd_target_p (struct gdbarch *gdbarch);
extern const char * gdbarch_gcore_bfd_target (struct gdbarch *gdbarch);
extern void set_gdbarch_gcore_bfd_target (struct gdbarch *gdbarch, const char * gcore_bfd_target);
/* If the elements of C++ vtables are in-place function descriptors rather
than normal function pointers (which may point to code or a descriptor),
set this to one. */
extern int gdbarch_vtable_function_descriptors (struct gdbarch *gdbarch);
extern void set_gdbarch_vtable_function_descriptors (struct gdbarch *gdbarch, int vtable_function_descriptors);
/* Set if the least significant bit of the delta is used instead of the least
significant bit of the pfn for pointers to virtual member functions. */
extern int gdbarch_vbit_in_delta (struct gdbarch *gdbarch);
extern void set_gdbarch_vbit_in_delta (struct gdbarch *gdbarch, int vbit_in_delta);
/* Advance PC to next instruction in order to skip a permanent breakpoint. */
typedef void (gdbarch_skip_permanent_breakpoint_ftype) (struct regcache *regcache);
extern void gdbarch_skip_permanent_breakpoint (struct gdbarch *gdbarch, struct regcache *regcache);
extern void set_gdbarch_skip_permanent_breakpoint (struct gdbarch *gdbarch, gdbarch_skip_permanent_breakpoint_ftype *skip_permanent_breakpoint);
/* The maximum length of an instruction on this architecture in bytes. */
extern bool gdbarch_max_insn_length_p (struct gdbarch *gdbarch);
extern ULONGEST gdbarch_max_insn_length (struct gdbarch *gdbarch);
extern void set_gdbarch_max_insn_length (struct gdbarch *gdbarch, ULONGEST max_insn_length);
/* Copy the instruction at FROM to TO, and make any adjustments
necessary to single-step it at that address.
Remove gdbarch.sh This patch runs gdbarch.py and removes gdbarch.sh.
2021-12-16 05:38:30 +08:00
Split gdbarch.h into two files This patch splits gdbarch.h into two files -- gdbarch.h now is editable and hand-maintained, and the new gdbarch-gen.h file is the only thing generated by gdbarch.sh. This lets us avoid maintaining boilerplate in the gdbarch.sh file. Note that gdbarch.sh still generates gdbarch.h after this patch. This makes it easier to re-run when rebasing. This code is removed in a subsequent patch.
2021-12-15 07:48:45 +08:00
REGS holds the state the thread's registers will have before
executing the copied instruction; the PC in REGS will refer to FROM,
not the copy at TO. The caller should update it to point at TO later.
Remove gdbarch.sh This patch runs gdbarch.py and removes gdbarch.sh.
2021-12-16 05:38:30 +08:00
Split gdbarch.h into two files This patch splits gdbarch.h into two files -- gdbarch.h now is editable and hand-maintained, and the new gdbarch-gen.h file is the only thing generated by gdbarch.sh. This lets us avoid maintaining boilerplate in the gdbarch.sh file. Note that gdbarch.sh still generates gdbarch.h after this patch. This makes it easier to re-run when rebasing. This code is removed in a subsequent patch.
2021-12-15 07:48:45 +08:00
Return a pointer to data of the architecture's choice to be passed
to gdbarch_displaced_step_fixup.
Remove gdbarch.sh This patch runs gdbarch.py and removes gdbarch.sh.
2021-12-16 05:38:30 +08:00
Split gdbarch.h into two files This patch splits gdbarch.h into two files -- gdbarch.h now is editable and hand-maintained, and the new gdbarch-gen.h file is the only thing generated by gdbarch.sh. This lets us avoid maintaining boilerplate in the gdbarch.sh file. Note that gdbarch.sh still generates gdbarch.h after this patch. This makes it easier to re-run when rebasing. This code is removed in a subsequent patch.
2021-12-15 07:48:45 +08:00
For a general explanation of displaced stepping and how GDB uses it,
see the comments in infrun.c.
Remove gdbarch.sh This patch runs gdbarch.py and removes gdbarch.sh.
2021-12-16 05:38:30 +08:00
Split gdbarch.h into two files This patch splits gdbarch.h into two files -- gdbarch.h now is editable and hand-maintained, and the new gdbarch-gen.h file is the only thing generated by gdbarch.sh. This lets us avoid maintaining boilerplate in the gdbarch.sh file. Note that gdbarch.sh still generates gdbarch.h after this patch. This makes it easier to re-run when rebasing. This code is removed in a subsequent patch.
2021-12-15 07:48:45 +08:00
The TO area is only guaranteed to have space for
gdb: add gdbarch::displaced_step_buffer_length The gdbarch::max_insn_length field is used mostly to support displaced stepping; it controls the size of the buffers allocated for the displaced-step instruction, and is also used when first copying the instruction, and later, when fixing up the instruction, in order to read in and parse the instruction being stepped. However, it has started to be used in other places in GDB, for example, it's used in the Python disassembler API, and it is used on amd64 as part of branch-tracing instruction classification. The problem is that the value assigned to max_insn_length is not always the maximum instruction length, but sometimes is a multiple of that length, as required to support displaced stepping, see rs600, ARM, and AArch64 for examples of this. It seems to me that we are overloading the meaning of the max_insn_length field, and I think that could potentially lead to confusion. I propose that we add a new gdbarch field, gdbarch::displaced_step_buffer_length, this new field will do exactly what it says on the tin; represent the required displaced step buffer size. The max_insn_length field can then do exactly what it claims to do; represent the maximum length of a single instruction. As some architectures (e.g. i386, and amd64) only require their displaced step buffers to be a single instruction in size, I propose that the default for displaced_step_buffer_length will be the value of max_insn_length. Architectures than need more buffer space can then override this default as needed. I've updated all architectures to setup the new field if appropriate, and I've audited all calls to gdbarch_max_insn_length and switched to gdbarch_displaced_step_buffer_length where appropriate. There should be no user visible changes after this commit. Approved-By: Simon Marchi <simon.marchi@efficios.com>
2023-02-23 19:45:11 +08:00
gdbarch_displaced_step_buffer_length (arch) octets, so this
function must not write more octets than that to this area.
Remove gdbarch.sh This patch runs gdbarch.py and removes gdbarch.sh.
2021-12-16 05:38:30 +08:00
Split gdbarch.h into two files This patch splits gdbarch.h into two files -- gdbarch.h now is editable and hand-maintained, and the new gdbarch-gen.h file is the only thing generated by gdbarch.sh. This lets us avoid maintaining boilerplate in the gdbarch.sh file. Note that gdbarch.sh still generates gdbarch.h after this patch. This makes it easier to re-run when rebasing. This code is removed in a subsequent patch.
2021-12-15 07:48:45 +08:00
If you do not provide this function, GDB assumes that the
architecture does not support displaced stepping.
Remove gdbarch.sh This patch runs gdbarch.py and removes gdbarch.sh.
2021-12-16 05:38:30 +08:00
Split gdbarch.h into two files This patch splits gdbarch.h into two files -- gdbarch.h now is editable and hand-maintained, and the new gdbarch-gen.h file is the only thing generated by gdbarch.sh. This lets us avoid maintaining boilerplate in the gdbarch.sh file. Note that gdbarch.sh still generates gdbarch.h after this patch. This makes it easier to re-run when rebasing. This code is removed in a subsequent patch.
2021-12-15 07:48:45 +08:00
If the instruction cannot execute out of line, return NULL. The
core falls back to stepping past the instruction in-line instead in
that case. */
extern bool gdbarch_displaced_step_copy_insn_p (struct gdbarch *gdbarch);
typedef displaced_step_copy_insn_closure_up (gdbarch_displaced_step_copy_insn_ftype) (struct gdbarch *gdbarch, CORE_ADDR from, CORE_ADDR to, struct regcache *regs);
extern displaced_step_copy_insn_closure_up gdbarch_displaced_step_copy_insn (struct gdbarch *gdbarch, CORE_ADDR from, CORE_ADDR to, struct regcache *regs);
extern void set_gdbarch_displaced_step_copy_insn (struct gdbarch *gdbarch, gdbarch_displaced_step_copy_insn_ftype *displaced_step_copy_insn);
/* Return true if GDB should use hardware single-stepping to execute a displaced
step instruction. If false, GDB will simply restart execution at the
displaced instruction location, and it is up to the target to ensure GDB will
receive control again (e.g. by placing a software breakpoint instruction into
the displaced instruction buffer).
Remove gdbarch.sh This patch runs gdbarch.py and removes gdbarch.sh.
2021-12-16 05:38:30 +08:00
Split gdbarch.h into two files This patch splits gdbarch.h into two files -- gdbarch.h now is editable and hand-maintained, and the new gdbarch-gen.h file is the only thing generated by gdbarch.sh. This lets us avoid maintaining boilerplate in the gdbarch.sh file. Note that gdbarch.sh still generates gdbarch.h after this patch. This makes it easier to re-run when rebasing. This code is removed in a subsequent patch.
2021-12-15 07:48:45 +08:00
The default implementation returns false on all targets that provide a
gdbarch_software_single_step routine, and true otherwise. */
typedef bool (gdbarch_displaced_step_hw_singlestep_ftype) (struct gdbarch *gdbarch);
extern bool gdbarch_displaced_step_hw_singlestep (struct gdbarch *gdbarch);
extern void set_gdbarch_displaced_step_hw_singlestep (struct gdbarch *gdbarch, gdbarch_displaced_step_hw_singlestep_ftype *displaced_step_hw_singlestep);
gdb: fix reg corruption from displaced stepping on amd64 This commit aims to address a problem that exists with the current approach to displaced stepping, and was identified in PR gdb/22921. Displaced stepping is currently supported on AArch64, ARM, amd64, i386, rs6000 (ppc), and s390. Of these, I believe there is a problem with the current approach which will impact amd64 and ARM, and can lead to random register corruption when the inferior makes use of asynchronous signals and GDB is using displaced stepping. The problem can be found in displaced_step_buffers::finish in displaced-stepping.c, and is this; after GDB tries to perform a displaced step, and the inferior stops, GDB classifies the stop into one of two states, either the displaced step succeeded, or the displaced step failed. If the displaced step succeeded then gdbarch_displaced_step_fixup is called, which has the job of fixing up the state of the current inferior as if the step had not been performed in a displaced manner. This all seems just fine. However, if the displaced step is considered to have not completed then GDB doesn't call gdbarch_displaced_step_fixup, instead GDB remains in displaced_step_buffers::finish and just performs a minimal fixup which involves adjusting the program counter back to its original value. The problem here is that for amd64 and ARM setting up for a displaced step can involve changing the values in some temporary registers. If the displaced step succeeds then this is fine; after the step the temporary registers are restored to their original values in the architecture specific code. But if the displaced step does not succeed then the temporary registers are never restored, and they retain their modified values. In this context a temporary register is simply any register that is not otherwise used by the instruction being stepped that the architecture specific code considers safe to borrow for the lifetime of the instruction being stepped. In the bug PR gdb/22921, the amd64 instruction being stepped is an rip-relative instruction like this: jmp *0x2fe2(%rip) When we displaced step this instruction we borrow a register, and modify the instruction to something like: jmp *0x2fe2(%rcx) with %rcx having its value adjusted to contain the original %rip value. Now if the displaced step does not succeed, then %rcx will be left with a corrupted value. Obviously corrupting any register is bad; in the bug report this problem was spotted because %rcx is used as a function argument register. And finally, why might a displaced step not succeed? Asynchronous signals provides one reason. GDB sets up for the displaced step and, at that precise moment, the OS delivers a signal (SIGALRM in the bug report), the signal stops the inferior at the address of the displaced instruction. GDB cancels the displaced instruction, handles the signal, and then tries again with the displaced step. But it is that first cancellation of the displaced step that causes the problem; in that case GDB (correctly) sees the displaced step as having not completed, and so does not perform the architecture specific fixup, leaving the register corrupted. The reason why I think AArch64, rs600, i386, and s390 are not effected by this problem is that I don't believe these architectures make use of any temporary registers, so when a displaced step is not completed successfully, the minimal fix up is sufficient. On amd64 we use at most one temporary register. On ARM, looking at arm_displaced_step_copy_insn_closure, we could modify up to 16 temporary registers, and the instruction being displaced stepped could be expanded to multiple replacement instructions, which increases the chances of this bug triggering. This commit only aims to address the issue on amd64 for now, though I believe that the approach I'm proposing here might be applicable for ARM too. What I propose is that we always call gdbarch_displaced_step_fixup. We will now pass an extra argument to gdbarch_displaced_step_fixup, this a boolean that indicates whether GDB thinks the displaced step completed successfully or not. When this flag is false this indicates that the displaced step halted for some "other" reason. On ARM GDB can potentially read the inferior's program counter in order figure out how far through the sequence of replacement instructions we got, and from that GDB can figure out what fixup needs to be performed. On targets like amd64 the problem is slightly easier as displaced stepping only uses a single replacement instruction. If the displaced step didn't complete the GDB knows that the single instruction didn't execute. The point is that by always calling gdbarch_displaced_step_fixup, each architecture can now ensure that the inferior state is fixed up correctly in all cases, not just the success case. On amd64 this ensures that we always restore the temporary register value, and so bug PR gdb/22921 is resolved. In order to move all architectures to this new API, I have moved the minimal roll-back version of the code inside the architecture specific fixup functions for AArch64, rs600, s390, and ARM. For all of these except ARM I think this is good enough, as no temporaries are used all that's needed is the program counter restore anyway. For ARM the minimal code is no worse than what we had before, though I do consider this architecture's displaced-stepping broken. I've updated the gdb.arch/amd64-disp-step.exp test to cover the 'jmpq*' instruction that was causing problems in the original bug, and also added support for testing the displaced step in the presence of asynchronous signal delivery. I've also added two new tests (for amd64 and i386) that check that GDB can correctly handle displaced stepping over a single instruction that branches to itself. I added these tests after a first version of this patch relied too much on checking the program-counter value in order to see if the displaced instruction had executed. This works fine in almost all cases, but when an instruction branches to itself a pure program counter check is not sufficient. The new tests expose this problem. Bug: https://sourceware.org/bugzilla/show_bug.cgi?id=22921 Approved-By: Pedro Alves <pedro@palves.net>
2023-02-22 20:15:34 +08:00
/* Fix up the state after attempting to single-step a displaced
instruction, to give the result we would have gotten from stepping the
instruction in its original location.
Remove gdbarch.sh This patch runs gdbarch.py and removes gdbarch.sh.
2021-12-16 05:38:30 +08:00
Split gdbarch.h into two files This patch splits gdbarch.h into two files -- gdbarch.h now is editable and hand-maintained, and the new gdbarch-gen.h file is the only thing generated by gdbarch.sh. This lets us avoid maintaining boilerplate in the gdbarch.sh file. Note that gdbarch.sh still generates gdbarch.h after this patch. This makes it easier to re-run when rebasing. This code is removed in a subsequent patch.
2021-12-15 07:48:45 +08:00
REGS is the register state resulting from single-stepping the
displaced instruction.
Remove gdbarch.sh This patch runs gdbarch.py and removes gdbarch.sh.
2021-12-16 05:38:30 +08:00
Split gdbarch.h into two files This patch splits gdbarch.h into two files -- gdbarch.h now is editable and hand-maintained, and the new gdbarch-gen.h file is the only thing generated by gdbarch.sh. This lets us avoid maintaining boilerplate in the gdbarch.sh file. Note that gdbarch.sh still generates gdbarch.h after this patch. This makes it easier to re-run when rebasing. This code is removed in a subsequent patch.
2021-12-15 07:48:45 +08:00
CLOSURE is the result from the matching call to
gdbarch_displaced_step_copy_insn.
Remove gdbarch.sh This patch runs gdbarch.py and removes gdbarch.sh.
2021-12-16 05:38:30 +08:00
gdb: fix reg corruption from displaced stepping on amd64 This commit aims to address a problem that exists with the current approach to displaced stepping, and was identified in PR gdb/22921. Displaced stepping is currently supported on AArch64, ARM, amd64, i386, rs6000 (ppc), and s390. Of these, I believe there is a problem with the current approach which will impact amd64 and ARM, and can lead to random register corruption when the inferior makes use of asynchronous signals and GDB is using displaced stepping. The problem can be found in displaced_step_buffers::finish in displaced-stepping.c, and is this; after GDB tries to perform a displaced step, and the inferior stops, GDB classifies the stop into one of two states, either the displaced step succeeded, or the displaced step failed. If the displaced step succeeded then gdbarch_displaced_step_fixup is called, which has the job of fixing up the state of the current inferior as if the step had not been performed in a displaced manner. This all seems just fine. However, if the displaced step is considered to have not completed then GDB doesn't call gdbarch_displaced_step_fixup, instead GDB remains in displaced_step_buffers::finish and just performs a minimal fixup which involves adjusting the program counter back to its original value. The problem here is that for amd64 and ARM setting up for a displaced step can involve changing the values in some temporary registers. If the displaced step succeeds then this is fine; after the step the temporary registers are restored to their original values in the architecture specific code. But if the displaced step does not succeed then the temporary registers are never restored, and they retain their modified values. In this context a temporary register is simply any register that is not otherwise used by the instruction being stepped that the architecture specific code considers safe to borrow for the lifetime of the instruction being stepped. In the bug PR gdb/22921, the amd64 instruction being stepped is an rip-relative instruction like this: jmp *0x2fe2(%rip) When we displaced step this instruction we borrow a register, and modify the instruction to something like: jmp *0x2fe2(%rcx) with %rcx having its value adjusted to contain the original %rip value. Now if the displaced step does not succeed, then %rcx will be left with a corrupted value. Obviously corrupting any register is bad; in the bug report this problem was spotted because %rcx is used as a function argument register. And finally, why might a displaced step not succeed? Asynchronous signals provides one reason. GDB sets up for the displaced step and, at that precise moment, the OS delivers a signal (SIGALRM in the bug report), the signal stops the inferior at the address of the displaced instruction. GDB cancels the displaced instruction, handles the signal, and then tries again with the displaced step. But it is that first cancellation of the displaced step that causes the problem; in that case GDB (correctly) sees the displaced step as having not completed, and so does not perform the architecture specific fixup, leaving the register corrupted. The reason why I think AArch64, rs600, i386, and s390 are not effected by this problem is that I don't believe these architectures make use of any temporary registers, so when a displaced step is not completed successfully, the minimal fix up is sufficient. On amd64 we use at most one temporary register. On ARM, looking at arm_displaced_step_copy_insn_closure, we could modify up to 16 temporary registers, and the instruction being displaced stepped could be expanded to multiple replacement instructions, which increases the chances of this bug triggering. This commit only aims to address the issue on amd64 for now, though I believe that the approach I'm proposing here might be applicable for ARM too. What I propose is that we always call gdbarch_displaced_step_fixup. We will now pass an extra argument to gdbarch_displaced_step_fixup, this a boolean that indicates whether GDB thinks the displaced step completed successfully or not. When this flag is false this indicates that the displaced step halted for some "other" reason. On ARM GDB can potentially read the inferior's program counter in order figure out how far through the sequence of replacement instructions we got, and from that GDB can figure out what fixup needs to be performed. On targets like amd64 the problem is slightly easier as displaced stepping only uses a single replacement instruction. If the displaced step didn't complete the GDB knows that the single instruction didn't execute. The point is that by always calling gdbarch_displaced_step_fixup, each architecture can now ensure that the inferior state is fixed up correctly in all cases, not just the success case. On amd64 this ensures that we always restore the temporary register value, and so bug PR gdb/22921 is resolved. In order to move all architectures to this new API, I have moved the minimal roll-back version of the code inside the architecture specific fixup functions for AArch64, rs600, s390, and ARM. For all of these except ARM I think this is good enough, as no temporaries are used all that's needed is the program counter restore anyway. For ARM the minimal code is no worse than what we had before, though I do consider this architecture's displaced-stepping broken. I've updated the gdb.arch/amd64-disp-step.exp test to cover the 'jmpq*' instruction that was causing problems in the original bug, and also added support for testing the displaced step in the presence of asynchronous signal delivery. I've also added two new tests (for amd64 and i386) that check that GDB can correctly handle displaced stepping over a single instruction that branches to itself. I added these tests after a first version of this patch relied too much on checking the program-counter value in order to see if the displaced instruction had executed. This works fine in almost all cases, but when an instruction branches to itself a pure program counter check is not sufficient. The new tests expose this problem. Bug: https://sourceware.org/bugzilla/show_bug.cgi?id=22921 Approved-By: Pedro Alves <pedro@palves.net>
2023-02-22 20:15:34 +08:00
FROM is the address where the instruction was original located, TO is
the address of the displaced buffer where the instruction was copied
to for stepping.
COMPLETED_P is true if GDB stopped as a result of the requested step
having completed (e.g. the inferior stopped with SIGTRAP), otherwise
COMPLETED_P is false and GDB stopped for some other reason. In the
case where a single instruction is expanded to multiple replacement
instructions for stepping then it may be necessary to read the current
program counter from REGS in order to decide how far through the
series of replacement instructions the inferior got before stopping,
this may impact what will need fixing up in this function.
Remove gdbarch.sh This patch runs gdbarch.py and removes gdbarch.sh.
2021-12-16 05:38:30 +08:00
Split gdbarch.h into two files This patch splits gdbarch.h into two files -- gdbarch.h now is editable and hand-maintained, and the new gdbarch-gen.h file is the only thing generated by gdbarch.sh. This lets us avoid maintaining boilerplate in the gdbarch.sh file. Note that gdbarch.sh still generates gdbarch.h after this patch. This makes it easier to re-run when rebasing. This code is removed in a subsequent patch.
2021-12-15 07:48:45 +08:00
For a general explanation of displaced stepping and how GDB uses it,
see the comments in infrun.c. */
gdb: fix reg corruption from displaced stepping on amd64 This commit aims to address a problem that exists with the current approach to displaced stepping, and was identified in PR gdb/22921. Displaced stepping is currently supported on AArch64, ARM, amd64, i386, rs6000 (ppc), and s390. Of these, I believe there is a problem with the current approach which will impact amd64 and ARM, and can lead to random register corruption when the inferior makes use of asynchronous signals and GDB is using displaced stepping. The problem can be found in displaced_step_buffers::finish in displaced-stepping.c, and is this; after GDB tries to perform a displaced step, and the inferior stops, GDB classifies the stop into one of two states, either the displaced step succeeded, or the displaced step failed. If the displaced step succeeded then gdbarch_displaced_step_fixup is called, which has the job of fixing up the state of the current inferior as if the step had not been performed in a displaced manner. This all seems just fine. However, if the displaced step is considered to have not completed then GDB doesn't call gdbarch_displaced_step_fixup, instead GDB remains in displaced_step_buffers::finish and just performs a minimal fixup which involves adjusting the program counter back to its original value. The problem here is that for amd64 and ARM setting up for a displaced step can involve changing the values in some temporary registers. If the displaced step succeeds then this is fine; after the step the temporary registers are restored to their original values in the architecture specific code. But if the displaced step does not succeed then the temporary registers are never restored, and they retain their modified values. In this context a temporary register is simply any register that is not otherwise used by the instruction being stepped that the architecture specific code considers safe to borrow for the lifetime of the instruction being stepped. In the bug PR gdb/22921, the amd64 instruction being stepped is an rip-relative instruction like this: jmp *0x2fe2(%rip) When we displaced step this instruction we borrow a register, and modify the instruction to something like: jmp *0x2fe2(%rcx) with %rcx having its value adjusted to contain the original %rip value. Now if the displaced step does not succeed, then %rcx will be left with a corrupted value. Obviously corrupting any register is bad; in the bug report this problem was spotted because %rcx is used as a function argument register. And finally, why might a displaced step not succeed? Asynchronous signals provides one reason. GDB sets up for the displaced step and, at that precise moment, the OS delivers a signal (SIGALRM in the bug report), the signal stops the inferior at the address of the displaced instruction. GDB cancels the displaced instruction, handles the signal, and then tries again with the displaced step. But it is that first cancellation of the displaced step that causes the problem; in that case GDB (correctly) sees the displaced step as having not completed, and so does not perform the architecture specific fixup, leaving the register corrupted. The reason why I think AArch64, rs600, i386, and s390 are not effected by this problem is that I don't believe these architectures make use of any temporary registers, so when a displaced step is not completed successfully, the minimal fix up is sufficient. On amd64 we use at most one temporary register. On ARM, looking at arm_displaced_step_copy_insn_closure, we could modify up to 16 temporary registers, and the instruction being displaced stepped could be expanded to multiple replacement instructions, which increases the chances of this bug triggering. This commit only aims to address the issue on amd64 for now, though I believe that the approach I'm proposing here might be applicable for ARM too. What I propose is that we always call gdbarch_displaced_step_fixup. We will now pass an extra argument to gdbarch_displaced_step_fixup, this a boolean that indicates whether GDB thinks the displaced step completed successfully or not. When this flag is false this indicates that the displaced step halted for some "other" reason. On ARM GDB can potentially read the inferior's program counter in order figure out how far through the sequence of replacement instructions we got, and from that GDB can figure out what fixup needs to be performed. On targets like amd64 the problem is slightly easier as displaced stepping only uses a single replacement instruction. If the displaced step didn't complete the GDB knows that the single instruction didn't execute. The point is that by always calling gdbarch_displaced_step_fixup, each architecture can now ensure that the inferior state is fixed up correctly in all cases, not just the success case. On amd64 this ensures that we always restore the temporary register value, and so bug PR gdb/22921 is resolved. In order to move all architectures to this new API, I have moved the minimal roll-back version of the code inside the architecture specific fixup functions for AArch64, rs600, s390, and ARM. For all of these except ARM I think this is good enough, as no temporaries are used all that's needed is the program counter restore anyway. For ARM the minimal code is no worse than what we had before, though I do consider this architecture's displaced-stepping broken. I've updated the gdb.arch/amd64-disp-step.exp test to cover the 'jmpq*' instruction that was causing problems in the original bug, and also added support for testing the displaced step in the presence of asynchronous signal delivery. I've also added two new tests (for amd64 and i386) that check that GDB can correctly handle displaced stepping over a single instruction that branches to itself. I added these tests after a first version of this patch relied too much on checking the program-counter value in order to see if the displaced instruction had executed. This works fine in almost all cases, but when an instruction branches to itself a pure program counter check is not sufficient. The new tests expose this problem. Bug: https://sourceware.org/bugzilla/show_bug.cgi?id=22921 Approved-By: Pedro Alves <pedro@palves.net>
2023-02-22 20:15:34 +08:00
typedef void (gdbarch_displaced_step_fixup_ftype) (struct gdbarch *gdbarch, struct displaced_step_copy_insn_closure *closure, CORE_ADDR from, CORE_ADDR to, struct regcache *regs, bool completed_p);
extern void gdbarch_displaced_step_fixup (struct gdbarch *gdbarch, struct displaced_step_copy_insn_closure *closure, CORE_ADDR from, CORE_ADDR to, struct regcache *regs, bool completed_p);
Split gdbarch.h into two files This patch splits gdbarch.h into two files -- gdbarch.h now is editable and hand-maintained, and the new gdbarch-gen.h file is the only thing generated by gdbarch.sh. This lets us avoid maintaining boilerplate in the gdbarch.sh file. Note that gdbarch.sh still generates gdbarch.h after this patch. This makes it easier to re-run when rebasing. This code is removed in a subsequent patch.
2021-12-15 07:48:45 +08:00
extern void set_gdbarch_displaced_step_fixup (struct gdbarch *gdbarch, gdbarch_displaced_step_fixup_ftype *displaced_step_fixup);
/* Prepare THREAD for it to displaced step the instruction at its current PC.
Remove gdbarch.sh This patch runs gdbarch.py and removes gdbarch.sh.
2021-12-16 05:38:30 +08:00
Split gdbarch.h into two files This patch splits gdbarch.h into two files -- gdbarch.h now is editable and hand-maintained, and the new gdbarch-gen.h file is the only thing generated by gdbarch.sh. This lets us avoid maintaining boilerplate in the gdbarch.sh file. Note that gdbarch.sh still generates gdbarch.h after this patch. This makes it easier to re-run when rebasing. This code is removed in a subsequent patch.
2021-12-15 07:48:45 +08:00
Throw an exception if any unexpected error happens. */
extern bool gdbarch_displaced_step_prepare_p (struct gdbarch *gdbarch);
typedef displaced_step_prepare_status (gdbarch_displaced_step_prepare_ftype) (struct gdbarch *gdbarch, thread_info *thread, CORE_ADDR &displaced_pc);
extern displaced_step_prepare_status gdbarch_displaced_step_prepare (struct gdbarch *gdbarch, thread_info *thread, CORE_ADDR &displaced_pc);
extern void set_gdbarch_displaced_step_prepare (struct gdbarch *gdbarch, gdbarch_displaced_step_prepare_ftype *displaced_step_prepare);
gdb: clear step over information on thread exit (PR gdb/27338) GDB doesn't handle correctly the case where a thread steps over a breakpoint (using either in-line or displaced stepping), and the executed instruction causes the thread to exit. Using the test program included later in the series, this is what it looks like with displaced-stepping, on x86-64 Linux, where we have two displaced-step buffers: $ ./gdb -q -nx --data-directory=data-directory build/binutils-gdb/gdb/testsuite/outputs/gdb.threads/step-over-thread-exit/step-over-thread-exit -ex "b my_exit_syscall" -ex r Reading symbols from build/binutils-gdb/gdb/testsuite/outputs/gdb.threads/step-over-thread-exit/step-over-thread-exit... Breakpoint 1 at 0x123c: file src/binutils-gdb/gdb/testsuite/lib/my-syscalls.S, line 68. Starting program: build/binutils-gdb/gdb/testsuite/outputs/gdb.threads/step-over-thread-exit/step-over-thread-exit [Thread debugging using libthread_db enabled] Using host libthread_db library "/usr/lib/../lib/libthread_db.so.1". [New Thread 0x7ffff7c5f640 (LWP 2915510)] [Switching to Thread 0x7ffff7c5f640 (LWP 2915510)] Thread 2 "step-over-threa" hit Breakpoint 1, my_exit_syscall () at src/binutils-gdb/gdb/testsuite/lib/my-syscalls.S:68 68 syscall (gdb) c Continuing. [New Thread 0x7ffff7c5f640 (LWP 2915524)] [Thread 0x7ffff7c5f640 (LWP 2915510) exited] [Switching to Thread 0x7ffff7c5f640 (LWP 2915524)] Thread 3 "step-over-threa" hit Breakpoint 1, my_exit_syscall () at src/binutils-gdb/gdb/testsuite/lib/my-syscalls.S:68 68 syscall (gdb) c Continuing. [New Thread 0x7ffff7c5f640 (LWP 2915616)] [Thread 0x7ffff7c5f640 (LWP 2915524) exited] [Switching to Thread 0x7ffff7c5f640 (LWP 2915616)] Thread 4 "step-over-threa" hit Breakpoint 1, my_exit_syscall () at src/binutils-gdb/gdb/testsuite/lib/my-syscalls.S:68 68 syscall (gdb) c Continuing. ... hangs ... The first two times we do "continue", we displaced-step the syscall instruction that causes the thread to exit. When the thread exits, the main thread, waiting on pthread_join, is unblocked. It spawns a new thread, which hits the breakpoint on the syscall again. However, infrun was never notified that the displaced-stepping threads are done using the displaced-step buffer, so now both buffers are marked as used. So when we do the third continue, there are no buffers available to displaced-step the syscall, so the thread waits forever for its turn. When trying the same but with in-line step over (displaced-stepping disabled): $ ./gdb -q -nx --data-directory=data-directory \ build/binutils-gdb/gdb/testsuite/outputs/gdb.threads/step-over-thread-exit/step-over-thread-exit \ -ex "b my_exit_syscall" -ex "set displaced-stepping off" -ex r Reading symbols from build/binutils-gdb/gdb/testsuite/outputs/gdb.threads/step-over-thread-exit/step-over-thread-exit... Breakpoint 1 at 0x123c: file src/binutils-gdb/gdb/testsuite/lib/my-syscalls.S, line 68. Starting program: build/binutils-gdb/gdb/testsuite/outputs/gdb.threads/step-over-thread-exit/step-over-thread-exit [Thread debugging using libthread_db enabled] Using host libthread_db library "/usr/lib/../lib/libthread_db.so.1". [New Thread 0x7ffff7c5f640 (LWP 2928290)] [Switching to Thread 0x7ffff7c5f640 (LWP 2928290)] Thread 2 "step-over-threa" hit Breakpoint 1, my_exit_syscall () at src/binutils-gdb/gdb/testsuite/lib/my-syscalls.S:68 68 syscall (gdb) c Continuing. [Thread 0x7ffff7c5f640 (LWP 2928290) exited] No unwaited-for children left. (gdb) i th Id Target Id Frame 1 Thread 0x7ffff7c60740 (LWP 2928285) "step-over-threa" 0x00007ffff7f7c9b7 in __pthread_clockjoin_ex () from /usr/lib/libpthread.so.0 The current thread <Thread ID 2> has terminated. See `help thread'. (gdb) thread 1 [Switching to thread 1 (Thread 0x7ffff7c60740 (LWP 2928285))] #0 0x00007ffff7f7c9b7 in __pthread_clockjoin_ex () from /usr/lib/libpthread.so.0 (gdb) c Continuing. ^C^C ... hangs ... The "continue" causes an in-line step to occur, meaning the main thread is stopped while we step the syscall. The stepped thread exits when executing the syscall, the linux-nat target notices there are no more resumed threads to be waited for, so returns TARGET_WAITKIND_NO_RESUMED, which causes the prompt to return. But infrun never clears the in-line step over info. So if we try continuing the main thread, GDB doesn't resume it, because it thinks there's an in-line step in progress that we need to wait for to finish, and we are stuck there. To fix this, infrun needs to be informed when a thread doing a displaced or in-line step over exits. We can do that with the new target_set_thread_options mechanism which is optimal for only enabling exit events of the thread that needs it; or, if that is not supported, by using target_thread_events, which enables thread exit events for all threads. This is done by this commit. This patch then modifies handle_inferior_event in infrun.c to clean up any step-over the exiting thread might have been doing at the time of the exit. The cases to consider are: - the exiting thread was doing an in-line step-over with an all-stop target - the exiting thread was doing an in-line step-over with a non-stop target - the exiting thread was doing a displaced step-over with a non-stop target The displaced-stepping buffer implementation in displaced-stepping.c is modified to account for the fact that it's possible that we "finish" a displaced step after a thread exit event. The buffer that the exiting thread was using is marked as available again and the original instructions under the scratch pad are restored. However, it skips applying the fixup, which wouldn't make sense since the thread does not exist anymore. Another case that needs handling is if a displaced-stepping thread exits, and the event is reported while we are in stop_all_threads. We should call displaced_step_finish in the handle_one function, in that case. It was already called in other code paths, just not the "thread exited" path. This commit doesn't make infrun ask the target to report the TARGET_WAITKIND_THREAD_EXITED events yet, that'll be done later in the series. Note that "stop_print_frame = false;" line is moved to normal_stop, because TARGET_WAITKIND_THREAD_EXITED can also end up with the event transmorphed into TARGET_WAITKIND_NO_RESUMED. Moving it to normal_stop keeps it centralized. Co-authored-by: Simon Marchi <simon.marchi@efficios.com> Bug: https://sourceware.org/bugzilla/show_bug.cgi?id=27338 Reviewed-By: Andrew Burgess <aburgess@redhat.com> Change-Id: I745c6955d7ef90beb83bcf0ff1d1ac8b9b6285a5
2021-02-06 05:42:32 +08:00
/* Clean up after a displaced step of THREAD.
It is possible for the displaced-stepped instruction to have caused
the thread to exit. The implementation can detect this case by
checking if WS.kind is TARGET_WAITKIND_THREAD_EXITED. */
Split gdbarch.h into two files This patch splits gdbarch.h into two files -- gdbarch.h now is editable and hand-maintained, and the new gdbarch-gen.h file is the only thing generated by gdbarch.sh. This lets us avoid maintaining boilerplate in the gdbarch.sh file. Note that gdbarch.sh still generates gdbarch.h after this patch. This makes it easier to re-run when rebasing. This code is removed in a subsequent patch.
2021-12-15 07:48:45 +08:00
displaced step: pass down target_waitstatus instead of gdb_signal This commit tweaks displaced_step_finish & friends to pass down a target_waitstatus instead of a gdb_signal. This is needed because a patch later in the step-over-{thread-exit,clone] series will want to make displaced_step_buffers::finish handle TARGET_WAITKIND_THREAD_EXITED. It also helps with the TARGET_WAITKIND_THREAD_CLONED patch later in that same series. It's also a bit more logical this way, as we don't have to pass down signals when the thread didn't actually stop for a signal. So we can also think of it as a clean up. Bug: https://sourceware.org/bugzilla/show_bug.cgi?id=27338 Change-Id: I4c5d338647b028071bc498c4e47063795a2db4c0 Approved-By: Andrew Burgess <aburgess@redhat.com>
2021-06-22 22:42:51 +08:00
typedef displaced_step_finish_status (gdbarch_displaced_step_finish_ftype) (struct gdbarch *gdbarch, thread_info *thread, const target_waitstatus &ws);
extern displaced_step_finish_status gdbarch_displaced_step_finish (struct gdbarch *gdbarch, thread_info *thread, const target_waitstatus &ws);
Split gdbarch.h into two files This patch splits gdbarch.h into two files -- gdbarch.h now is editable and hand-maintained, and the new gdbarch-gen.h file is the only thing generated by gdbarch.sh. This lets us avoid maintaining boilerplate in the gdbarch.sh file. Note that gdbarch.sh still generates gdbarch.h after this patch. This makes it easier to re-run when rebasing. This code is removed in a subsequent patch.
2021-12-15 07:48:45 +08:00
extern void set_gdbarch_displaced_step_finish (struct gdbarch *gdbarch, gdbarch_displaced_step_finish_ftype *displaced_step_finish);
/* Return the closure associated to the displaced step buffer that is at ADDR. */
extern bool gdbarch_displaced_step_copy_insn_closure_by_addr_p (struct gdbarch *gdbarch);
typedef const displaced_step_copy_insn_closure * (gdbarch_displaced_step_copy_insn_closure_by_addr_ftype) (inferior *inf, CORE_ADDR addr);
extern const displaced_step_copy_insn_closure * gdbarch_displaced_step_copy_insn_closure_by_addr (struct gdbarch *gdbarch, inferior *inf, CORE_ADDR addr);
extern void set_gdbarch_displaced_step_copy_insn_closure_by_addr (struct gdbarch *gdbarch, gdbarch_displaced_step_copy_insn_closure_by_addr_ftype *displaced_step_copy_insn_closure_by_addr);
/* PARENT_INF has forked and CHILD_PTID is the ptid of the child. Restore the
contents of all displaced step buffers in the child's address space. */
typedef void (gdbarch_displaced_step_restore_all_in_ptid_ftype) (inferior *parent_inf, ptid_t child_ptid);
extern void gdbarch_displaced_step_restore_all_in_ptid (struct gdbarch *gdbarch, inferior *parent_inf, ptid_t child_ptid);
extern void set_gdbarch_displaced_step_restore_all_in_ptid (struct gdbarch *gdbarch, gdbarch_displaced_step_restore_all_in_ptid_ftype *displaced_step_restore_all_in_ptid);
gdb: add gdbarch::displaced_step_buffer_length The gdbarch::max_insn_length field is used mostly to support displaced stepping; it controls the size of the buffers allocated for the displaced-step instruction, and is also used when first copying the instruction, and later, when fixing up the instruction, in order to read in and parse the instruction being stepped. However, it has started to be used in other places in GDB, for example, it's used in the Python disassembler API, and it is used on amd64 as part of branch-tracing instruction classification. The problem is that the value assigned to max_insn_length is not always the maximum instruction length, but sometimes is a multiple of that length, as required to support displaced stepping, see rs600, ARM, and AArch64 for examples of this. It seems to me that we are overloading the meaning of the max_insn_length field, and I think that could potentially lead to confusion. I propose that we add a new gdbarch field, gdbarch::displaced_step_buffer_length, this new field will do exactly what it says on the tin; represent the required displaced step buffer size. The max_insn_length field can then do exactly what it claims to do; represent the maximum length of a single instruction. As some architectures (e.g. i386, and amd64) only require their displaced step buffers to be a single instruction in size, I propose that the default for displaced_step_buffer_length will be the value of max_insn_length. Architectures than need more buffer space can then override this default as needed. I've updated all architectures to setup the new field if appropriate, and I've audited all calls to gdbarch_max_insn_length and switched to gdbarch_displaced_step_buffer_length where appropriate. There should be no user visible changes after this commit. Approved-By: Simon Marchi <simon.marchi@efficios.com>
2023-02-23 19:45:11 +08:00
/* The maximum length in octets required for a displaced-step instruction
buffer. By default this will be the same as gdbarch::max_insn_length,
but should be overridden for architectures that might expand a
displaced-step instruction to multiple replacement instructions. */
extern ULONGEST gdbarch_displaced_step_buffer_length (struct gdbarch *gdbarch);
extern void set_gdbarch_displaced_step_buffer_length (struct gdbarch *gdbarch, ULONGEST displaced_step_buffer_length);
Split gdbarch.h into two files This patch splits gdbarch.h into two files -- gdbarch.h now is editable and hand-maintained, and the new gdbarch-gen.h file is the only thing generated by gdbarch.sh. This lets us avoid maintaining boilerplate in the gdbarch.sh file. Note that gdbarch.sh still generates gdbarch.h after this patch. This makes it easier to re-run when rebasing. This code is removed in a subsequent patch.
2021-12-15 07:48:45 +08:00
/* Relocate an instruction to execute at a different address. OLDLOC
is the address in the inferior memory where the instruction to
relocate is currently at. On input, TO points to the destination
where we want the instruction to be copied (and possibly adjusted)
to. On output, it points to one past the end of the resulting
instruction(s). The effect of executing the instruction at TO shall
be the same as if executing it at FROM. For example, call
instructions that implicitly push the return address on the stack
should be adjusted to return to the instruction after OLDLOC;
relative branches, and other PC-relative instructions need the
offset adjusted; etc. */
extern bool gdbarch_relocate_instruction_p (struct gdbarch *gdbarch);
typedef void (gdbarch_relocate_instruction_ftype) (struct gdbarch *gdbarch, CORE_ADDR *to, CORE_ADDR from);
extern void gdbarch_relocate_instruction (struct gdbarch *gdbarch, CORE_ADDR *to, CORE_ADDR from);
extern void set_gdbarch_relocate_instruction (struct gdbarch *gdbarch, gdbarch_relocate_instruction_ftype *relocate_instruction);
/* Refresh overlay mapped state for section OSECT. */
extern bool gdbarch_overlay_update_p (struct gdbarch *gdbarch);
typedef void (gdbarch_overlay_update_ftype) (struct obj_section *osect);
extern void gdbarch_overlay_update (struct gdbarch *gdbarch, struct obj_section *osect);
extern void set_gdbarch_overlay_update (struct gdbarch *gdbarch, gdbarch_overlay_update_ftype *overlay_update);
extern bool gdbarch_core_read_description_p (struct gdbarch *gdbarch);
typedef const struct target_desc * (gdbarch_core_read_description_ftype) (struct gdbarch *gdbarch, struct target_ops *target, bfd *abfd);
extern const struct target_desc * gdbarch_core_read_description (struct gdbarch *gdbarch, struct target_ops *target, bfd *abfd);
extern void set_gdbarch_core_read_description (struct gdbarch *gdbarch, gdbarch_core_read_description_ftype *core_read_description);
/* Set if the address in N_SO or N_FUN stabs may be zero. */
extern int gdbarch_sofun_address_maybe_missing (struct gdbarch *gdbarch);
extern void set_gdbarch_sofun_address_maybe_missing (struct gdbarch *gdbarch, int sofun_address_maybe_missing);
/* Parse the instruction at ADDR storing in the record execution log
the registers REGCACHE and memory ranges that will be affected when
the instruction executes, along with their current values.
Return -1 if something goes wrong, 0 otherwise. */
extern bool gdbarch_process_record_p (struct gdbarch *gdbarch);
typedef int (gdbarch_process_record_ftype) (struct gdbarch *gdbarch, struct regcache *regcache, CORE_ADDR addr);
extern int gdbarch_process_record (struct gdbarch *gdbarch, struct regcache *regcache, CORE_ADDR addr);
extern void set_gdbarch_process_record (struct gdbarch *gdbarch, gdbarch_process_record_ftype *process_record);
/* Save process state after a signal.
Return -1 if something goes wrong, 0 otherwise. */
extern bool gdbarch_process_record_signal_p (struct gdbarch *gdbarch);
typedef int (gdbarch_process_record_signal_ftype) (struct gdbarch *gdbarch, struct regcache *regcache, enum gdb_signal signal);
extern int gdbarch_process_record_signal (struct gdbarch *gdbarch, struct regcache *regcache, enum gdb_signal signal);
extern void set_gdbarch_process_record_signal (struct gdbarch *gdbarch, gdbarch_process_record_signal_ftype *process_record_signal);
/* Signal translation: translate inferior's signal (target's) number
into GDB's representation. The implementation of this method must
be host independent. IOW, don't rely on symbols of the NAT_FILE
header (the nm-*.h files), the host <signal.h> header, or similar
headers. This is mainly used when cross-debugging core files ---
"Live" targets hide the translation behind the target interface
(target_wait, target_resume, etc.). */
extern bool gdbarch_gdb_signal_from_target_p (struct gdbarch *gdbarch);
typedef enum gdb_signal (gdbarch_gdb_signal_from_target_ftype) (struct gdbarch *gdbarch, int signo);
extern enum gdb_signal gdbarch_gdb_signal_from_target (struct gdbarch *gdbarch, int signo);
extern void set_gdbarch_gdb_signal_from_target (struct gdbarch *gdbarch, gdbarch_gdb_signal_from_target_ftype *gdb_signal_from_target);
/* Signal translation: translate the GDB's internal signal number into
the inferior's signal (target's) representation. The implementation
of this method must be host independent. IOW, don't rely on symbols
of the NAT_FILE header (the nm-*.h files), the host <signal.h>
header, or similar headers.
Return the target signal number if found, or -1 if the GDB internal
signal number is invalid. */
extern bool gdbarch_gdb_signal_to_target_p (struct gdbarch *gdbarch);
typedef int (gdbarch_gdb_signal_to_target_ftype) (struct gdbarch *gdbarch, enum gdb_signal signal);
extern int gdbarch_gdb_signal_to_target (struct gdbarch *gdbarch, enum gdb_signal signal);
extern void set_gdbarch_gdb_signal_to_target (struct gdbarch *gdbarch, gdbarch_gdb_signal_to_target_ftype *gdb_signal_to_target);
/* Extra signal info inspection.
Remove gdbarch.sh This patch runs gdbarch.py and removes gdbarch.sh.
2021-12-16 05:38:30 +08:00
Split gdbarch.h into two files This patch splits gdbarch.h into two files -- gdbarch.h now is editable and hand-maintained, and the new gdbarch-gen.h file is the only thing generated by gdbarch.sh. This lets us avoid maintaining boilerplate in the gdbarch.sh file. Note that gdbarch.sh still generates gdbarch.h after this patch. This makes it easier to re-run when rebasing. This code is removed in a subsequent patch.
2021-12-15 07:48:45 +08:00
Return a type suitable to inspect extra signal information. */
extern bool gdbarch_get_siginfo_type_p (struct gdbarch *gdbarch);
typedef struct type * (gdbarch_get_siginfo_type_ftype) (struct gdbarch *gdbarch);
extern struct type * gdbarch_get_siginfo_type (struct gdbarch *gdbarch);
extern void set_gdbarch_get_siginfo_type (struct gdbarch *gdbarch, gdbarch_get_siginfo_type_ftype *get_siginfo_type);
/* Record architecture-specific information from the symbol table. */
extern bool gdbarch_record_special_symbol_p (struct gdbarch *gdbarch);
typedef void (gdbarch_record_special_symbol_ftype) (struct gdbarch *gdbarch, struct objfile *objfile, asymbol *sym);
extern void gdbarch_record_special_symbol (struct gdbarch *gdbarch, struct objfile *objfile, asymbol *sym);
extern void set_gdbarch_record_special_symbol (struct gdbarch *gdbarch, gdbarch_record_special_symbol_ftype *record_special_symbol);
/* Function for the 'catch syscall' feature.
Get architecture-specific system calls information from registers. */
extern bool gdbarch_get_syscall_number_p (struct gdbarch *gdbarch);
typedef LONGEST (gdbarch_get_syscall_number_ftype) (struct gdbarch *gdbarch, thread_info *thread);
extern LONGEST gdbarch_get_syscall_number (struct gdbarch *gdbarch, thread_info *thread);
extern void set_gdbarch_get_syscall_number (struct gdbarch *gdbarch, gdbarch_get_syscall_number_ftype *get_syscall_number);
/* The filename of the XML syscall for this architecture. */
extern const char * gdbarch_xml_syscall_file (struct gdbarch *gdbarch);
extern void set_gdbarch_xml_syscall_file (struct gdbarch *gdbarch, const char * xml_syscall_file);
/* Information about system calls from this architecture */
extern struct syscalls_info * gdbarch_syscalls_info (struct gdbarch *gdbarch);
extern void set_gdbarch_syscalls_info (struct gdbarch *gdbarch, struct syscalls_info * syscalls_info);
/* SystemTap related fields and functions.
A NULL-terminated array of prefixes used to mark an integer constant
on the architecture's assembly.
For example, on x86 integer constants are written as:
Remove gdbarch.sh This patch runs gdbarch.py and removes gdbarch.sh.
2021-12-16 05:38:30 +08:00
$10 ;; integer constant 10
Split gdbarch.h into two files This patch splits gdbarch.h into two files -- gdbarch.h now is editable and hand-maintained, and the new gdbarch-gen.h file is the only thing generated by gdbarch.sh. This lets us avoid maintaining boilerplate in the gdbarch.sh file. Note that gdbarch.sh still generates gdbarch.h after this patch. This makes it easier to re-run when rebasing. This code is removed in a subsequent patch.
2021-12-15 07:48:45 +08:00
in this case, this prefix would be the character `$'. */
extern const char *const * gdbarch_stap_integer_prefixes (struct gdbarch *gdbarch);
extern void set_gdbarch_stap_integer_prefixes (struct gdbarch *gdbarch, const char *const * stap_integer_prefixes);
/* A NULL-terminated array of suffixes used to mark an integer constant
on the architecture's assembly. */
extern const char *const * gdbarch_stap_integer_suffixes (struct gdbarch *gdbarch);
extern void set_gdbarch_stap_integer_suffixes (struct gdbarch *gdbarch, const char *const * stap_integer_suffixes);
/* A NULL-terminated array of prefixes used to mark a register name on
the architecture's assembly.
For example, on x86 the register name is written as:
Remove gdbarch.sh This patch runs gdbarch.py and removes gdbarch.sh.
2021-12-16 05:38:30 +08:00
%eax ;; register eax
Split gdbarch.h into two files This patch splits gdbarch.h into two files -- gdbarch.h now is editable and hand-maintained, and the new gdbarch-gen.h file is the only thing generated by gdbarch.sh. This lets us avoid maintaining boilerplate in the gdbarch.sh file. Note that gdbarch.sh still generates gdbarch.h after this patch. This makes it easier to re-run when rebasing. This code is removed in a subsequent patch.
2021-12-15 07:48:45 +08:00
in this case, this prefix would be the character `%'. */
extern const char *const * gdbarch_stap_register_prefixes (struct gdbarch *gdbarch);
extern void set_gdbarch_stap_register_prefixes (struct gdbarch *gdbarch, const char *const * stap_register_prefixes);
/* A NULL-terminated array of suffixes used to mark a register name on
the architecture's assembly. */
extern const char *const * gdbarch_stap_register_suffixes (struct gdbarch *gdbarch);
extern void set_gdbarch_stap_register_suffixes (struct gdbarch *gdbarch, const char *const * stap_register_suffixes);
/* A NULL-terminated array of prefixes used to mark a register
indirection on the architecture's assembly.
For example, on x86 the register indirection is written as:
Remove gdbarch.sh This patch runs gdbarch.py and removes gdbarch.sh.
2021-12-16 05:38:30 +08:00
(%eax) ;; indirecting eax
Split gdbarch.h into two files This patch splits gdbarch.h into two files -- gdbarch.h now is editable and hand-maintained, and the new gdbarch-gen.h file is the only thing generated by gdbarch.sh. This lets us avoid maintaining boilerplate in the gdbarch.sh file. Note that gdbarch.sh still generates gdbarch.h after this patch. This makes it easier to re-run when rebasing. This code is removed in a subsequent patch.
2021-12-15 07:48:45 +08:00
in this case, this prefix would be the charater `('.
Remove gdbarch.sh This patch runs gdbarch.py and removes gdbarch.sh.
2021-12-16 05:38:30 +08:00
Split gdbarch.h into two files This patch splits gdbarch.h into two files -- gdbarch.h now is editable and hand-maintained, and the new gdbarch-gen.h file is the only thing generated by gdbarch.sh. This lets us avoid maintaining boilerplate in the gdbarch.sh file. Note that gdbarch.sh still generates gdbarch.h after this patch. This makes it easier to re-run when rebasing. This code is removed in a subsequent patch.
2021-12-15 07:48:45 +08:00
Please note that we use the indirection prefix also for register
displacement, e.g., `4(%eax)' on x86. */
extern const char *const * gdbarch_stap_register_indirection_prefixes (struct gdbarch *gdbarch);
extern void set_gdbarch_stap_register_indirection_prefixes (struct gdbarch *gdbarch, const char *const * stap_register_indirection_prefixes);
/* A NULL-terminated array of suffixes used to mark a register
indirection on the architecture's assembly.
For example, on x86 the register indirection is written as:
Remove gdbarch.sh This patch runs gdbarch.py and removes gdbarch.sh.
2021-12-16 05:38:30 +08:00
(%eax) ;; indirecting eax
Split gdbarch.h into two files This patch splits gdbarch.h into two files -- gdbarch.h now is editable and hand-maintained, and the new gdbarch-gen.h file is the only thing generated by gdbarch.sh. This lets us avoid maintaining boilerplate in the gdbarch.sh file. Note that gdbarch.sh still generates gdbarch.h after this patch. This makes it easier to re-run when rebasing. This code is removed in a subsequent patch.
2021-12-15 07:48:45 +08:00
in this case, this prefix would be the charater `)'.
Remove gdbarch.sh This patch runs gdbarch.py and removes gdbarch.sh.
2021-12-16 05:38:30 +08:00
Split gdbarch.h into two files This patch splits gdbarch.h into two files -- gdbarch.h now is editable and hand-maintained, and the new gdbarch-gen.h file is the only thing generated by gdbarch.sh. This lets us avoid maintaining boilerplate in the gdbarch.sh file. Note that gdbarch.sh still generates gdbarch.h after this patch. This makes it easier to re-run when rebasing. This code is removed in a subsequent patch.
2021-12-15 07:48:45 +08:00
Please note that we use the indirection suffix also for register
displacement, e.g., `4(%eax)' on x86. */
extern const char *const * gdbarch_stap_register_indirection_suffixes (struct gdbarch *gdbarch);
extern void set_gdbarch_stap_register_indirection_suffixes (struct gdbarch *gdbarch, const char *const * stap_register_indirection_suffixes);
/* Prefix(es) used to name a register using GDB's nomenclature.
Remove gdbarch.sh This patch runs gdbarch.py and removes gdbarch.sh.
2021-12-16 05:38:30 +08:00
Split gdbarch.h into two files This patch splits gdbarch.h into two files -- gdbarch.h now is editable and hand-maintained, and the new gdbarch-gen.h file is the only thing generated by gdbarch.sh. This lets us avoid maintaining boilerplate in the gdbarch.sh file. Note that gdbarch.sh still generates gdbarch.h after this patch. This makes it easier to re-run when rebasing. This code is removed in a subsequent patch.
2021-12-15 07:48:45 +08:00
For example, on PPC a register is represented by a number in the assembly
language (e.g., `10' is the 10th general-purpose register). However,
inside GDB this same register has an `r' appended to its name, so the 10th
register would be represented as `r10' internally. */
extern const char * gdbarch_stap_gdb_register_prefix (struct gdbarch *gdbarch);
extern void set_gdbarch_stap_gdb_register_prefix (struct gdbarch *gdbarch, const char * stap_gdb_register_prefix);
/* Suffix used to name a register using GDB's nomenclature. */
extern const char * gdbarch_stap_gdb_register_suffix (struct gdbarch *gdbarch);
extern void set_gdbarch_stap_gdb_register_suffix (struct gdbarch *gdbarch, const char * stap_gdb_register_suffix);
/* Check if S is a single operand.
Remove gdbarch.sh This patch runs gdbarch.py and removes gdbarch.sh.
2021-12-16 05:38:30 +08:00
Split gdbarch.h into two files This patch splits gdbarch.h into two files -- gdbarch.h now is editable and hand-maintained, and the new gdbarch-gen.h file is the only thing generated by gdbarch.sh. This lets us avoid maintaining boilerplate in the gdbarch.sh file. Note that gdbarch.sh still generates gdbarch.h after this patch. This makes it easier to re-run when rebasing. This code is removed in a subsequent patch.
2021-12-15 07:48:45 +08:00
Single operands can be:
Remove gdbarch.sh This patch runs gdbarch.py and removes gdbarch.sh.
2021-12-16 05:38:30 +08:00
- Literal integers, e.g. `$10' on x86
- Register access, e.g. `%eax' on x86
- Register indirection, e.g. `(%eax)' on x86
- Register displacement, e.g. `4(%eax)' on x86
Split gdbarch.h into two files This patch splits gdbarch.h into two files -- gdbarch.h now is editable and hand-maintained, and the new gdbarch-gen.h file is the only thing generated by gdbarch.sh. This lets us avoid maintaining boilerplate in the gdbarch.sh file. Note that gdbarch.sh still generates gdbarch.h after this patch. This makes it easier to re-run when rebasing. This code is removed in a subsequent patch.
2021-12-15 07:48:45 +08:00
This function should check for these patterns on the string
and return 1 if some were found, or zero otherwise. Please try to match
as much info as you can from the string, i.e., if you have to match
something like `(%', do not match just the `('. */
extern bool gdbarch_stap_is_single_operand_p (struct gdbarch *gdbarch);
typedef int (gdbarch_stap_is_single_operand_ftype) (struct gdbarch *gdbarch, const char *s);
extern int gdbarch_stap_is_single_operand (struct gdbarch *gdbarch, const char *s);
extern void set_gdbarch_stap_is_single_operand (struct gdbarch *gdbarch, gdbarch_stap_is_single_operand_ftype *stap_is_single_operand);
/* Function used to handle a "special case" in the parser.
Remove gdbarch.sh This patch runs gdbarch.py and removes gdbarch.sh.
2021-12-16 05:38:30 +08:00
Split gdbarch.h into two files This patch splits gdbarch.h into two files -- gdbarch.h now is editable and hand-maintained, and the new gdbarch-gen.h file is the only thing generated by gdbarch.sh. This lets us avoid maintaining boilerplate in the gdbarch.sh file. Note that gdbarch.sh still generates gdbarch.h after this patch. This makes it easier to re-run when rebasing. This code is removed in a subsequent patch.
2021-12-15 07:48:45 +08:00
A "special case" is considered to be an unknown token, i.e., a token
that the parser does not know how to parse. A good example of special
case would be ARM's register displacement syntax:
Remove gdbarch.sh This patch runs gdbarch.py and removes gdbarch.sh.
2021-12-16 05:38:30 +08:00
[R0, #4] ;; displacing R0 by 4
Split gdbarch.h into two files This patch splits gdbarch.h into two files -- gdbarch.h now is editable and hand-maintained, and the new gdbarch-gen.h file is the only thing generated by gdbarch.sh. This lets us avoid maintaining boilerplate in the gdbarch.sh file. Note that gdbarch.sh still generates gdbarch.h after this patch. This makes it easier to re-run when rebasing. This code is removed in a subsequent patch.
2021-12-15 07:48:45 +08:00
Since the parser assumes that a register displacement is of the form:
Remove gdbarch.sh This patch runs gdbarch.py and removes gdbarch.sh.
2021-12-16 05:38:30 +08:00
<number> <indirection_prefix> <register_name> <indirection_suffix>
Split gdbarch.h into two files This patch splits gdbarch.h into two files -- gdbarch.h now is editable and hand-maintained, and the new gdbarch-gen.h file is the only thing generated by gdbarch.sh. This lets us avoid maintaining boilerplate in the gdbarch.sh file. Note that gdbarch.sh still generates gdbarch.h after this patch. This makes it easier to re-run when rebasing. This code is removed in a subsequent patch.
2021-12-15 07:48:45 +08:00
it means that it will not be able to recognize and parse this odd syntax.
Therefore, we should add a special case function that will handle this token.
Remove gdbarch.sh This patch runs gdbarch.py and removes gdbarch.sh.
2021-12-16 05:38:30 +08:00
Split gdbarch.h into two files This patch splits gdbarch.h into two files -- gdbarch.h now is editable and hand-maintained, and the new gdbarch-gen.h file is the only thing generated by gdbarch.sh. This lets us avoid maintaining boilerplate in the gdbarch.sh file. Note that gdbarch.sh still generates gdbarch.h after this patch. This makes it easier to re-run when rebasing. This code is removed in a subsequent patch.
2021-12-15 07:48:45 +08:00
This function should generate the proper expression form of the expression
using GDB's internal expression mechanism (e.g., `write_exp_elt_opcode'
and so on). It should also return 1 if the parsing was successful, or zero
if the token was not recognized as a special token (in this case, returning
zero means that the special parser is deferring the parsing to the generic
parser), and should advance the buffer pointer (p->arg). */
extern bool gdbarch_stap_parse_special_token_p (struct gdbarch *gdbarch);
typedef expr::operation_up (gdbarch_stap_parse_special_token_ftype) (struct gdbarch *gdbarch, struct stap_parse_info *p);
extern expr::operation_up gdbarch_stap_parse_special_token (struct gdbarch *gdbarch, struct stap_parse_info *p);
extern void set_gdbarch_stap_parse_special_token (struct gdbarch *gdbarch, gdbarch_stap_parse_special_token_ftype *stap_parse_special_token);
/* Perform arch-dependent adjustments to a register name.
Remove gdbarch.sh This patch runs gdbarch.py and removes gdbarch.sh.
2021-12-16 05:38:30 +08:00
Split gdbarch.h into two files This patch splits gdbarch.h into two files -- gdbarch.h now is editable and hand-maintained, and the new gdbarch-gen.h file is the only thing generated by gdbarch.sh. This lets us avoid maintaining boilerplate in the gdbarch.sh file. Note that gdbarch.sh still generates gdbarch.h after this patch. This makes it easier to re-run when rebasing. This code is removed in a subsequent patch.
2021-12-15 07:48:45 +08:00
In very specific situations, it may be necessary for the register
name present in a SystemTap probe's argument to be handled in a
special way. For example, on i386, GCC may over-optimize the
register allocation and use smaller registers than necessary. In
such cases, the client that is reading and evaluating the SystemTap
probe (ourselves) will need to actually fetch values from the wider
version of the register in question.
Remove gdbarch.sh This patch runs gdbarch.py and removes gdbarch.sh.
2021-12-16 05:38:30 +08:00
Split gdbarch.h into two files This patch splits gdbarch.h into two files -- gdbarch.h now is editable and hand-maintained, and the new gdbarch-gen.h file is the only thing generated by gdbarch.sh. This lets us avoid maintaining boilerplate in the gdbarch.sh file. Note that gdbarch.sh still generates gdbarch.h after this patch. This makes it easier to re-run when rebasing. This code is removed in a subsequent patch.
2021-12-15 07:48:45 +08:00
To illustrate the example, consider the following probe argument
(i386):
Remove gdbarch.sh This patch runs gdbarch.py and removes gdbarch.sh.
2021-12-16 05:38:30 +08:00
4@%ax
Split gdbarch.h into two files This patch splits gdbarch.h into two files -- gdbarch.h now is editable and hand-maintained, and the new gdbarch-gen.h file is the only thing generated by gdbarch.sh. This lets us avoid maintaining boilerplate in the gdbarch.sh file. Note that gdbarch.sh still generates gdbarch.h after this patch. This makes it easier to re-run when rebasing. This code is removed in a subsequent patch.
2021-12-15 07:48:45 +08:00
This argument says that its value can be found at the %ax register,
which is a 16-bit register. However, the argument's prefix says
that its type is "uint32_t", which is 32-bit in size. Therefore, in
this case, GDB should actually fetch the probe's value from register
%eax, not %ax. In this scenario, this function would actually
replace the register name from %ax to %eax.
Remove gdbarch.sh This patch runs gdbarch.py and removes gdbarch.sh.
2021-12-16 05:38:30 +08:00
Split gdbarch.h into two files This patch splits gdbarch.h into two files -- gdbarch.h now is editable and hand-maintained, and the new gdbarch-gen.h file is the only thing generated by gdbarch.sh. This lets us avoid maintaining boilerplate in the gdbarch.sh file. Note that gdbarch.sh still generates gdbarch.h after this patch. This makes it easier to re-run when rebasing. This code is removed in a subsequent patch.
2021-12-15 07:48:45 +08:00
The rationale for this can be found at PR breakpoints/24541. */
extern bool gdbarch_stap_adjust_register_p (struct gdbarch *gdbarch);
typedef std::string (gdbarch_stap_adjust_register_ftype) (struct gdbarch *gdbarch, struct stap_parse_info *p, const std::string &regname, int regnum);
extern std::string gdbarch_stap_adjust_register (struct gdbarch *gdbarch, struct stap_parse_info *p, const std::string &regname, int regnum);
extern void set_gdbarch_stap_adjust_register (struct gdbarch *gdbarch, gdbarch_stap_adjust_register_ftype *stap_adjust_register);
/* DTrace related functions.
The expression to compute the NARTGth+1 argument to a DTrace USDT probe.
NARG must be >= 0. */
extern bool gdbarch_dtrace_parse_probe_argument_p (struct gdbarch *gdbarch);
typedef expr::operation_up (gdbarch_dtrace_parse_probe_argument_ftype) (struct gdbarch *gdbarch, int narg);
extern expr::operation_up gdbarch_dtrace_parse_probe_argument (struct gdbarch *gdbarch, int narg);
extern void set_gdbarch_dtrace_parse_probe_argument (struct gdbarch *gdbarch, gdbarch_dtrace_parse_probe_argument_ftype *dtrace_parse_probe_argument);
/* True if the given ADDR does not contain the instruction sequence
corresponding to a disabled DTrace is-enabled probe. */
extern bool gdbarch_dtrace_probe_is_enabled_p (struct gdbarch *gdbarch);
typedef int (gdbarch_dtrace_probe_is_enabled_ftype) (struct gdbarch *gdbarch, CORE_ADDR addr);
extern int gdbarch_dtrace_probe_is_enabled (struct gdbarch *gdbarch, CORE_ADDR addr);
extern void set_gdbarch_dtrace_probe_is_enabled (struct gdbarch *gdbarch, gdbarch_dtrace_probe_is_enabled_ftype *dtrace_probe_is_enabled);
/* Enable a DTrace is-enabled probe at ADDR. */
extern bool gdbarch_dtrace_enable_probe_p (struct gdbarch *gdbarch);
typedef void (gdbarch_dtrace_enable_probe_ftype) (struct gdbarch *gdbarch, CORE_ADDR addr);
extern void gdbarch_dtrace_enable_probe (struct gdbarch *gdbarch, CORE_ADDR addr);
extern void set_gdbarch_dtrace_enable_probe (struct gdbarch *gdbarch, gdbarch_dtrace_enable_probe_ftype *dtrace_enable_probe);
/* Disable a DTrace is-enabled probe at ADDR. */
extern bool gdbarch_dtrace_disable_probe_p (struct gdbarch *gdbarch);
typedef void (gdbarch_dtrace_disable_probe_ftype) (struct gdbarch *gdbarch, CORE_ADDR addr);
extern void gdbarch_dtrace_disable_probe (struct gdbarch *gdbarch, CORE_ADDR addr);
extern void set_gdbarch_dtrace_disable_probe (struct gdbarch *gdbarch, gdbarch_dtrace_disable_probe_ftype *dtrace_disable_probe);
/* True if the list of shared libraries is one and only for all
processes, as opposed to a list of shared libraries per inferior.
This usually means that all processes, although may or may not share
an address space, will see the same set of symbols at the same
addresses. */
extern int gdbarch_has_global_solist (struct gdbarch *gdbarch);
extern void set_gdbarch_has_global_solist (struct gdbarch *gdbarch, int has_global_solist);
/* On some targets, even though each inferior has its own private
address space, the debug interface takes care of making breakpoints
visible to all address spaces automatically. For such cases,
this property should be set to true. */
extern int gdbarch_has_global_breakpoints (struct gdbarch *gdbarch);
extern void set_gdbarch_has_global_breakpoints (struct gdbarch *gdbarch, int has_global_breakpoints);
/* True if inferiors share an address space (e.g., uClinux). */
typedef int (gdbarch_has_shared_address_space_ftype) (struct gdbarch *gdbarch);
extern int gdbarch_has_shared_address_space (struct gdbarch *gdbarch);
extern void set_gdbarch_has_shared_address_space (struct gdbarch *gdbarch, gdbarch_has_shared_address_space_ftype *has_shared_address_space);
/* True if a fast tracepoint can be set at an address. */
typedef int (gdbarch_fast_tracepoint_valid_at_ftype) (struct gdbarch *gdbarch, CORE_ADDR addr, std::string *msg);
extern int gdbarch_fast_tracepoint_valid_at (struct gdbarch *gdbarch, CORE_ADDR addr, std::string *msg);
extern void set_gdbarch_fast_tracepoint_valid_at (struct gdbarch *gdbarch, gdbarch_fast_tracepoint_valid_at_ftype *fast_tracepoint_valid_at);
/* Guess register state based on tracepoint location. Used for tracepoints
where no registers have been collected, but there's only one location,
allowing us to guess the PC value, and perhaps some other registers.
On entry, regcache has all registers marked as unavailable. */
typedef void (gdbarch_guess_tracepoint_registers_ftype) (struct gdbarch *gdbarch, struct regcache *regcache, CORE_ADDR addr);
extern void gdbarch_guess_tracepoint_registers (struct gdbarch *gdbarch, struct regcache *regcache, CORE_ADDR addr);
extern void set_gdbarch_guess_tracepoint_registers (struct gdbarch *gdbarch, gdbarch_guess_tracepoint_registers_ftype *guess_tracepoint_registers);
/* Return the "auto" target charset. */
Remove gdbarch.sh This patch runs gdbarch.py and removes gdbarch.sh.
2021-12-16 05:38:30 +08:00
typedef const char * (gdbarch_auto_charset_ftype) ();
Split gdbarch.h into two files This patch splits gdbarch.h into two files -- gdbarch.h now is editable and hand-maintained, and the new gdbarch-gen.h file is the only thing generated by gdbarch.sh. This lets us avoid maintaining boilerplate in the gdbarch.sh file. Note that gdbarch.sh still generates gdbarch.h after this patch. This makes it easier to re-run when rebasing. This code is removed in a subsequent patch.
2021-12-15 07:48:45 +08:00
extern const char * gdbarch_auto_charset (struct gdbarch *gdbarch);
extern void set_gdbarch_auto_charset (struct gdbarch *gdbarch, gdbarch_auto_charset_ftype *auto_charset);
/* Return the "auto" target wide charset. */
Remove gdbarch.sh This patch runs gdbarch.py and removes gdbarch.sh.
2021-12-16 05:38:30 +08:00
typedef const char * (gdbarch_auto_wide_charset_ftype) ();
Split gdbarch.h into two files This patch splits gdbarch.h into two files -- gdbarch.h now is editable and hand-maintained, and the new gdbarch-gen.h file is the only thing generated by gdbarch.sh. This lets us avoid maintaining boilerplate in the gdbarch.sh file. Note that gdbarch.sh still generates gdbarch.h after this patch. This makes it easier to re-run when rebasing. This code is removed in a subsequent patch.
2021-12-15 07:48:45 +08:00
extern const char * gdbarch_auto_wide_charset (struct gdbarch *gdbarch);
extern void set_gdbarch_auto_wide_charset (struct gdbarch *gdbarch, gdbarch_auto_wide_charset_ftype *auto_wide_charset);
/* If non-empty, this is a file extension that will be opened in place
of the file extension reported by the shared library list.
Remove gdbarch.sh This patch runs gdbarch.py and removes gdbarch.sh.
2021-12-16 05:38:30 +08:00
Split gdbarch.h into two files This patch splits gdbarch.h into two files -- gdbarch.h now is editable and hand-maintained, and the new gdbarch-gen.h file is the only thing generated by gdbarch.sh. This lets us avoid maintaining boilerplate in the gdbarch.sh file. Note that gdbarch.sh still generates gdbarch.h after this patch. This makes it easier to re-run when rebasing. This code is removed in a subsequent patch.
2021-12-15 07:48:45 +08:00
This is most useful for toolchains that use a post-linker tool,
where the names of the files run on the target differ in extension
compared to the names of the files GDB should load for debug info. */
extern const char * gdbarch_solib_symbols_extension (struct gdbarch *gdbarch);
extern void set_gdbarch_solib_symbols_extension (struct gdbarch *gdbarch, const char * solib_symbols_extension);
/* If true, the target OS has DOS-based file system semantics. That
is, absolute paths include a drive name, and the backslash is
considered a directory separator. */
extern int gdbarch_has_dos_based_file_system (struct gdbarch *gdbarch);
extern void set_gdbarch_has_dos_based_file_system (struct gdbarch *gdbarch, int has_dos_based_file_system);
/* Generate bytecodes to collect the return address in a frame.
Since the bytecodes run on the target, possibly with GDB not even
connected, the full unwinding machinery is not available, and
typically this function will issue bytecodes for one or more likely
places that the return address may be found. */
typedef void (gdbarch_gen_return_address_ftype) (struct gdbarch *gdbarch, struct agent_expr *ax, struct axs_value *value, CORE_ADDR scope);
extern void gdbarch_gen_return_address (struct gdbarch *gdbarch, struct agent_expr *ax, struct axs_value *value, CORE_ADDR scope);
extern void set_gdbarch_gen_return_address (struct gdbarch *gdbarch, gdbarch_gen_return_address_ftype *gen_return_address);
/* Implement the "info proc" command. */
extern bool gdbarch_info_proc_p (struct gdbarch *gdbarch);
typedef void (gdbarch_info_proc_ftype) (struct gdbarch *gdbarch, const char *args, enum info_proc_what what);
extern void gdbarch_info_proc (struct gdbarch *gdbarch, const char *args, enum info_proc_what what);
extern void set_gdbarch_info_proc (struct gdbarch *gdbarch, gdbarch_info_proc_ftype *info_proc);
/* Implement the "info proc" command for core files. Noe that there
are two "info_proc"-like methods on gdbarch -- one for core files,
one for live targets. */
extern bool gdbarch_core_info_proc_p (struct gdbarch *gdbarch);
typedef void (gdbarch_core_info_proc_ftype) (struct gdbarch *gdbarch, const char *args, enum info_proc_what what);
extern void gdbarch_core_info_proc (struct gdbarch *gdbarch, const char *args, enum info_proc_what what);
extern void set_gdbarch_core_info_proc (struct gdbarch *gdbarch, gdbarch_core_info_proc_ftype *core_info_proc);
/* Iterate over all objfiles in the order that makes the most sense
for the architecture to make global symbol searches.
Remove gdbarch.sh This patch runs gdbarch.py and removes gdbarch.sh.
2021-12-16 05:38:30 +08:00
gdb: use gdb::function_view for gdbarch_iterate_over_objfiles_in_search_order callback A rather straightforward patch to change an instance of callback + void pointer to gdb::function_view, allowing pasing lambdas that capture, and eliminating the need for the untyped pointer. Change-Id: I73ed644e7849945265a2c763f79f5456695b0037
2022-05-04 20:14:22 +08:00
CB is a callback function passed an objfile to be searched. The iteration stops
if this function returns nonzero.
Remove gdbarch.sh This patch runs gdbarch.py and removes gdbarch.sh.
2021-12-16 05:38:30 +08:00
Split gdbarch.h into two files This patch splits gdbarch.h into two files -- gdbarch.h now is editable and hand-maintained, and the new gdbarch-gen.h file is the only thing generated by gdbarch.sh. This lets us avoid maintaining boilerplate in the gdbarch.sh file. Note that gdbarch.sh still generates gdbarch.h after this patch. This makes it easier to re-run when rebasing. This code is removed in a subsequent patch.
2021-12-15 07:48:45 +08:00
If not NULL, CURRENT_OBJFILE corresponds to the objfile being
inspected when the symbol search was requested. */
gdb: use gdb::function_view for gdbarch_iterate_over_objfiles_in_search_order callback A rather straightforward patch to change an instance of callback + void pointer to gdb::function_view, allowing pasing lambdas that capture, and eliminating the need for the untyped pointer. Change-Id: I73ed644e7849945265a2c763f79f5456695b0037
2022-05-04 20:14:22 +08:00
typedef void (gdbarch_iterate_over_objfiles_in_search_order_ftype) (struct gdbarch *gdbarch, iterate_over_objfiles_in_search_order_cb_ftype cb, struct objfile *current_objfile);
extern void gdbarch_iterate_over_objfiles_in_search_order (struct gdbarch *gdbarch, iterate_over_objfiles_in_search_order_cb_ftype cb, struct objfile *current_objfile);
Split gdbarch.h into two files This patch splits gdbarch.h into two files -- gdbarch.h now is editable and hand-maintained, and the new gdbarch-gen.h file is the only thing generated by gdbarch.sh. This lets us avoid maintaining boilerplate in the gdbarch.sh file. Note that gdbarch.sh still generates gdbarch.h after this patch. This makes it easier to re-run when rebasing. This code is removed in a subsequent patch.
2021-12-15 07:48:45 +08:00
extern void set_gdbarch_iterate_over_objfiles_in_search_order (struct gdbarch *gdbarch, gdbarch_iterate_over_objfiles_in_search_order_ftype *iterate_over_objfiles_in_search_order);
/* Ravenscar arch-dependent ops. */
extern struct ravenscar_arch_ops * gdbarch_ravenscar_ops (struct gdbarch *gdbarch);
extern void set_gdbarch_ravenscar_ops (struct gdbarch *gdbarch, struct ravenscar_arch_ops * ravenscar_ops);
/* Return non-zero if the instruction at ADDR is a call; zero otherwise. */
typedef int (gdbarch_insn_is_call_ftype) (struct gdbarch *gdbarch, CORE_ADDR addr);
extern int gdbarch_insn_is_call (struct gdbarch *gdbarch, CORE_ADDR addr);
extern void set_gdbarch_insn_is_call (struct gdbarch *gdbarch, gdbarch_insn_is_call_ftype *insn_is_call);
/* Return non-zero if the instruction at ADDR is a return; zero otherwise. */
typedef int (gdbarch_insn_is_ret_ftype) (struct gdbarch *gdbarch, CORE_ADDR addr);
extern int gdbarch_insn_is_ret (struct gdbarch *gdbarch, CORE_ADDR addr);
extern void set_gdbarch_insn_is_ret (struct gdbarch *gdbarch, gdbarch_insn_is_ret_ftype *insn_is_ret);
/* Return non-zero if the instruction at ADDR is a jump; zero otherwise. */
typedef int (gdbarch_insn_is_jump_ftype) (struct gdbarch *gdbarch, CORE_ADDR addr);
extern int gdbarch_insn_is_jump (struct gdbarch *gdbarch, CORE_ADDR addr);
extern void set_gdbarch_insn_is_jump (struct gdbarch *gdbarch, gdbarch_insn_is_jump_ftype *insn_is_jump);
/* Return true if there's a program/permanent breakpoint planted in
memory at ADDRESS, return false otherwise. */
typedef bool (gdbarch_program_breakpoint_here_p_ftype) (struct gdbarch *gdbarch, CORE_ADDR address);
extern bool gdbarch_program_breakpoint_here_p (struct gdbarch *gdbarch, CORE_ADDR address);
extern void set_gdbarch_program_breakpoint_here_p (struct gdbarch *gdbarch, gdbarch_program_breakpoint_here_p_ftype *program_breakpoint_here_p);
/* Read one auxv entry from *READPTR, not reading locations >= ENDPTR.
Return 0 if *READPTR is already at the end of the buffer.
Return -1 if there is insufficient buffer for a whole entry.
Return 1 if an entry was read into *TYPEP and *VALP. */
extern bool gdbarch_auxv_parse_p (struct gdbarch *gdbarch);
gdb: constify auxv parse functions Constify the input parameters of the various auxv parse functions, they don't need to modify the raw auxv data. Change-Id: I13eacd5ab8e925ec2b5c1f7722cbab39c41516ec
2022-09-30 03:13:09 +08:00
typedef int (gdbarch_auxv_parse_ftype) (struct gdbarch *gdbarch, const gdb_byte **readptr, const gdb_byte *endptr, CORE_ADDR *typep, CORE_ADDR *valp);
extern int gdbarch_auxv_parse (struct gdbarch *gdbarch, const gdb_byte **readptr, const gdb_byte *endptr, CORE_ADDR *typep, CORE_ADDR *valp);
Split gdbarch.h into two files This patch splits gdbarch.h into two files -- gdbarch.h now is editable and hand-maintained, and the new gdbarch-gen.h file is the only thing generated by gdbarch.sh. This lets us avoid maintaining boilerplate in the gdbarch.sh file. Note that gdbarch.sh still generates gdbarch.h after this patch. This makes it easier to re-run when rebasing. This code is removed in a subsequent patch.
2021-12-15 07:48:45 +08:00
extern void set_gdbarch_auxv_parse (struct gdbarch *gdbarch, gdbarch_auxv_parse_ftype *auxv_parse);
/* Print the description of a single auxv entry described by TYPE and VAL
to FILE. */
typedef void (gdbarch_print_auxv_entry_ftype) (struct gdbarch *gdbarch, struct ui_file *file, CORE_ADDR type, CORE_ADDR val);
extern void gdbarch_print_auxv_entry (struct gdbarch *gdbarch, struct ui_file *file, CORE_ADDR type, CORE_ADDR val);
extern void set_gdbarch_print_auxv_entry (struct gdbarch *gdbarch, gdbarch_print_auxv_entry_ftype *print_auxv_entry);
/* Find the address range of the current inferior's vsyscall/vDSO, and
write it to *RANGE. If the vsyscall's length can't be determined, a
range with zero length is returned. Returns true if the vsyscall is
found, false otherwise. */
typedef int (gdbarch_vsyscall_range_ftype) (struct gdbarch *gdbarch, struct mem_range *range);
extern int gdbarch_vsyscall_range (struct gdbarch *gdbarch, struct mem_range *range);
extern void set_gdbarch_vsyscall_range (struct gdbarch *gdbarch, gdbarch_vsyscall_range_ftype *vsyscall_range);
/* Allocate SIZE bytes of PROT protected page aligned memory in inferior.
PROT has GDB_MMAP_PROT_* bitmask format.
Throw an error if it is not possible. Returned address is always valid. */
typedef CORE_ADDR (gdbarch_infcall_mmap_ftype) (CORE_ADDR size, unsigned prot);
extern CORE_ADDR gdbarch_infcall_mmap (struct gdbarch *gdbarch, CORE_ADDR size, unsigned prot);
extern void set_gdbarch_infcall_mmap (struct gdbarch *gdbarch, gdbarch_infcall_mmap_ftype *infcall_mmap);
/* Deallocate SIZE bytes of memory at ADDR in inferior from gdbarch_infcall_mmap.
Print a warning if it is not possible. */
typedef void (gdbarch_infcall_munmap_ftype) (CORE_ADDR addr, CORE_ADDR size);
extern void gdbarch_infcall_munmap (struct gdbarch *gdbarch, CORE_ADDR addr, CORE_ADDR size);
extern void set_gdbarch_infcall_munmap (struct gdbarch *gdbarch, gdbarch_infcall_munmap_ftype *infcall_munmap);
/* Return string (caller has to use xfree for it) with options for GCC
to produce code for this target, typically "-m64", "-m32" or "-m31".
These options are put before CU's DW_AT_producer compilation options so that
they can override it. */
typedef std::string (gdbarch_gcc_target_options_ftype) (struct gdbarch *gdbarch);
extern std::string gdbarch_gcc_target_options (struct gdbarch *gdbarch);
extern void set_gdbarch_gcc_target_options (struct gdbarch *gdbarch, gdbarch_gcc_target_options_ftype *gcc_target_options);
/* Return a regular expression that matches names used by this
architecture in GNU configury triplets. The result is statically
allocated and must not be freed. The default implementation simply
returns the BFD architecture name, which is correct in nearly every
case. */
typedef const char * (gdbarch_gnu_triplet_regexp_ftype) (struct gdbarch *gdbarch);
extern const char * gdbarch_gnu_triplet_regexp (struct gdbarch *gdbarch);
extern void set_gdbarch_gnu_triplet_regexp (struct gdbarch *gdbarch, gdbarch_gnu_triplet_regexp_ftype *gnu_triplet_regexp);
/* Return the size in 8-bit bytes of an addressable memory unit on this
architecture. This corresponds to the number of 8-bit bytes associated to
each address in memory. */
typedef int (gdbarch_addressable_memory_unit_size_ftype) (struct gdbarch *gdbarch);
extern int gdbarch_addressable_memory_unit_size (struct gdbarch *gdbarch);
extern void set_gdbarch_addressable_memory_unit_size (struct gdbarch *gdbarch, gdbarch_addressable_memory_unit_size_ftype *addressable_memory_unit_size);
/* Functions for allowing a target to modify its disassembler options. */
extern const char * gdbarch_disassembler_options_implicit (struct gdbarch *gdbarch);
extern void set_gdbarch_disassembler_options_implicit (struct gdbarch *gdbarch, const char * disassembler_options_implicit);
Use std::string for disassembler options I noticed that the disassembler_options code uses manual memory management. It seemed simpler to replace this with std::string. Approved-By: John Baldwin <jhb@FreeBSD.org>
2024-03-22 01:15:56 +08:00
extern std::string * gdbarch_disassembler_options (struct gdbarch *gdbarch);
extern void set_gdbarch_disassembler_options (struct gdbarch *gdbarch, std::string * disassembler_options);
Split gdbarch.h into two files This patch splits gdbarch.h into two files -- gdbarch.h now is editable and hand-maintained, and the new gdbarch-gen.h file is the only thing generated by gdbarch.sh. This lets us avoid maintaining boilerplate in the gdbarch.sh file. Note that gdbarch.sh still generates gdbarch.h after this patch. This makes it easier to re-run when rebasing. This code is removed in a subsequent patch.
2021-12-15 07:48:45 +08:00
extern const disasm_options_and_args_t * gdbarch_valid_disassembler_options (struct gdbarch *gdbarch);
extern void set_gdbarch_valid_disassembler_options (struct gdbarch *gdbarch, const disasm_options_and_args_t * valid_disassembler_options);
/* Type alignment override method. Return the architecture specific
alignment required for TYPE. If there is no special handling
required for TYPE then return the value 0, GDB will then apply the
default rules as laid out in gdbtypes.c:type_align. */
typedef ULONGEST (gdbarch_type_align_ftype) (struct gdbarch *gdbarch, struct type *type);
extern ULONGEST gdbarch_type_align (struct gdbarch *gdbarch, struct type *type);
extern void set_gdbarch_type_align (struct gdbarch *gdbarch, gdbarch_type_align_ftype *type_align);
/* Return a string containing any flags for the given PC in the given FRAME. */
gdb: pass frames as `const frame_info_ptr &` We currently pass frames to function by value, as `frame_info_ptr`. This is somewhat expensive: - the size of `frame_info_ptr` is 64 bytes, which is a bit big to pass by value - the constructors and destructor link/unlink the object in the global `frame_info_ptr::frame_list` list. This is an `intrusive_list`, so it's not so bad: it's just assigning a few points, there's no memory allocation as if it was `std::list`, but still it's useless to do that over and over. As suggested by Tom Tromey, change many function signatures to accept `const frame_info_ptr &` instead of `frame_info_ptr`. Some functions reassign their `frame_info_ptr` parameter, like: void the_func (frame_info_ptr frame) { for (; frame != nullptr; frame = get_prev_frame (frame)) { ... } } I wondered what to do about them, do I leave them as-is or change them (and need to introduce a separate local variable that can be re-assigned). I opted for the later for consistency. It might not be clear why some functions take `const frame_info_ptr &` while others take `frame_info_ptr`. Also, if a function took a `frame_info_ptr` because it did re-assign its parameter, I doubt that we would think to change it to `const frame_info_ptr &` should the implementation change such that it doesn't need to take `frame_info_ptr` anymore. It seems better to have a simple rule and apply it everywhere. Change-Id: I59d10addef687d157f82ccf4d54f5dde9a963fd0 Approved-By: Andrew Burgess <aburgess@redhat.com>
2024-02-20 02:07:47 +08:00
typedef std::string (gdbarch_get_pc_address_flags_ftype) (const frame_info_ptr &frame, CORE_ADDR pc);
extern std::string gdbarch_get_pc_address_flags (struct gdbarch *gdbarch, const frame_info_ptr &frame, CORE_ADDR pc);
Split gdbarch.h into two files This patch splits gdbarch.h into two files -- gdbarch.h now is editable and hand-maintained, and the new gdbarch-gen.h file is the only thing generated by gdbarch.sh. This lets us avoid maintaining boilerplate in the gdbarch.sh file. Note that gdbarch.sh still generates gdbarch.h after this patch. This makes it easier to re-run when rebasing. This code is removed in a subsequent patch.
2021-12-15 07:48:45 +08:00
extern void set_gdbarch_get_pc_address_flags (struct gdbarch *gdbarch, gdbarch_get_pc_address_flags_ftype *get_pc_address_flags);
/* Read core file mappings */
typedef void (gdbarch_read_core_file_mappings_ftype) (struct gdbarch *gdbarch, struct bfd *cbfd, read_core_file_mappings_pre_loop_ftype pre_loop_cb, read_core_file_mappings_loop_ftype loop_cb);
extern void gdbarch_read_core_file_mappings (struct gdbarch *gdbarch, struct bfd *cbfd, read_core_file_mappings_pre_loop_ftype pre_loop_cb, read_core_file_mappings_loop_ftype loop_cb);
extern void set_gdbarch_read_core_file_mappings (struct gdbarch *gdbarch, gdbarch_read_core_file_mappings_ftype *read_core_file_mappings);
corefile/bug: Add hook to control the use of target description notes from corefiles Due to the nature of the AArch64 SVE/SME extensions in GDB, each thread can potentially have distinct target descriptions/gdbarches. When loading a gcore-generated core file, at the moment GDB gives priority to the target description dumped to NT_GDB_TDESC. Though technically correct for most targets, it doesn't work correctly for AArch64 with SVE or SME support. The correct approach for AArch64/Linux is to either have per-thread target description notes in the corefiles or to rely on the gdbarch_core_read_description hook, so it can figure out the proper target description for a given thread based on the various available register notes. The former, although more correct, doesn't address the case of existing gdb's that only output a single target description note. This patch goes for the latter, and adds a new gdbarch hook to conditionalize the use of the corefile target description note. The hook is called use_target_description_from_corefile_notes. The hook defaults to returning true, meaning targets will use the corefile target description note. AArch64 Linux overrides the hook to return false when it detects any of the SVE or SME register notes in the corefile. Otherwise it should be fine for AArch64 Linux to use the corefile target description note. When we support per-thread target description notes, then we can augment the AArch64 Linux hook to rely on those notes. Regression-tested on aarch64-linux Ubuntu 22.04/20.04. Approved-By: Simon Marchi <simon.marchi@efficios.com> Reviewed-by: Thiago Jung Bauermann <thiago.bauermann@linaro.org>
2023-09-07 23:20:15 +08:00
/* Return true if the target description for all threads should be read from the
target description core file note(s). Return false if the target description
for all threads should be inferred from the core file contents/sections.
The corefile's bfd is passed through COREFILE_BFD. */
typedef bool (gdbarch_use_target_description_from_corefile_notes_ftype) (struct gdbarch *gdbarch, struct bfd *corefile_bfd);
extern bool gdbarch_use_target_description_from_corefile_notes (struct gdbarch *gdbarch, struct bfd *corefile_bfd);
extern void set_gdbarch_use_target_description_from_corefile_notes (struct gdbarch *gdbarch, gdbarch_use_target_description_from_corefile_notes_ftype *use_target_description_from_corefile_notes);
Reference in New Issue Copy Permalink