Adds /*<type>*/ comments and a linter check from rz-bindgen to enforce
their existence and consistency
Also includes the following fixes made when adding the annotations:
* removed unused intern_table arguments in pyc_dis.c, pyc_dis.h, asm_pyc.c
* removed unused classes argument from place_nodes in agraph.c
* removed unused recurse and recurse_bb functions in canalysis.c
* removed unused vars field from RzPrint struct
* removed unused RzAnalysisType* structs from rz_analysis.h
* removed unused list field from RzEgg struct
* fixed bug in bp_plugin.c where duplication-checking logic iterates over the wrong list
* removed unused q_regs field from RzDebug struct
* removed unused backtrace field from RzDebugPlugin struct
* removed unused classes_list field from RzBinNXOObj struct
* removed unused methods_list and classes_list fields from RzBinZimgObj struct
ARM_THREAD_STATE64 fetches arm_thread_state64_t but our struct was
arm_unified_thread_state_t, which adds a header. The pc in the register
profile was hacked to account for the shift, but all the x-regs were
not, so x7 had the value of x8 for example. Now we just use the specific
state for 32 or 64.
This fixes basic single-stepping on macOS/arm64.
dbg->bits is affected by all kinds of things going on in rizin, in
particular it currently randomly switches to 16 when calling pd during
an arm64 debug session, causing the stepping code to break.
We now fetch the cpu type of the process once during attach, similar to
how lldb does it. They also use 0 for the "invalid" value, so we do that
as well.
Getting the thread list with task_threads() allocates a mapping in the
caller's memory that should then be freed with vm_deallocate() later.
However rz_xnu_get_cur_thread() was passing the debuggee's task to the
deallocate rather than its own, meaning the thread list would probably
leak and the debuggee's maps would get broken if there was something at
that address.
ptrace with PT_ATTACHEXC will send a SIGSTOP to the process, which we then
sometimes received in the first mach_msg loop while waiting for hitting a
breakpoint, when having been already attached at this point.
See also bsd/kern/mach_process.c:287 in xnu-8020.121.3.
Correct attaching works like this, in exactly this order:
* posix_spawn the process with POSIX_SPAWN_START_SUSPENDED
* set up the Mach port for receiving exceptions
* ptrace(PT_ATTACHEXC) -> sends SIGSTOP
* wait on the Mach port until receiving the SIGSTOP msg
* call task_suspend
* reply to the msg
Then we have a beautiful process suspended right at the start in dyld.
Entire files are now avoided to be built by checking in meson rather
than using #if DEBUGGER. This now also includes bsd_debug.c, which
is currently slightly broken on OpenBSD/sparc64 and caused compile
errors even with -Ddebugger=false.
If the kernel/system where Rizin is currently run does not support
ptrace(PTRACE_GETREGSET, ..., NT_X86_XSTATE, ...), ENODEV is returned by
the kernel. In such cases, we print the error message once and never
execute the ptrace call again, assuming it is not supported.
Signed-off-by: Riccardo Schirone <sirmy15@gmail.com>
* Create `rz_windows.h`
* Move `RzThread` structure definitions to internal header
* Create and use `rz_th_get_user()` API to avoid exposing `RzThread` layout in external header
* Make `rz_th_self()` an IPI
* Use `_wmkdir()` in `rz_sys_mkdir()` on Windows
* Remove unused `winkd_break_read` function
* Remove `RzIORap` definition from `rz_io.h`
* Remove include of `Windows.h` from `rz_types.h`
* Use `rz_windows.h` when necessary
* Remove `RzCoreRtrHost` definition from `rz_core.h`
* Remove `rz_socket.h` include from `rz_core.h`
In practice, dbg.bpsize was always automatically set to
rz_bp_size(core->dbg->bp, core->analysis->bits) whenever the global bits
changed. But this breaks as soon as different bits are needed for where
the breakpoint is supposed to be set vs. the current value of asm.bits
(for example for thumb/arm32).
So this removes the global bpsize and always calculates it for the
specific breakpoint address.
As a sidenote, dbg.bpsize was not even used for manual db (for dcu it
was though).
* Support building for Windows on ARM
* Fix usage of reserved function name on ARM
* Fix architecture dependent `CONTEXT` fields access
* Fixes for debugging on Windows on ARM
* Fix breakpoint behavior on windows on ARM
* Fix register profile when debugging
* Remove `drx` error message if not supported
* Add support for hardware breakpoints
* Disable support for hardware single-stepping
* Fix Windows Message breakpoints on ARM
* Fix software single-stepping on return instruction on ARM
* Consider link register as destination for return instructions
* Fix setting debug bits to 16
* Rename `r12` to `ip` for windows arm register profile
* Remove debug bits hardcoding on windows
* Add 32bit registers to Windows ARM64 register profile
* Fix software step breakpoint alignment when entering thumb mode
* Define `RZ_TEST_ARCH` for `arm` and `arm64`
* Use `StackWalk64` API for backtracing on Windows
* Fix software single-stepping ARM in thumb mode
* Sync `asm.bits` with debugger
* Cleanup windows_debug.c
* Add SPDX for cross-arm64-windows.txt
* Add floating point control register flags
* Fix Windows Arm register profile alignment
* Keep `drx` warning
* Add `db` commands in `cmd_debug.yaml`
* Use bool instead of int as return value
* Port `db` commands to newshell
* Fix integration tests. Use `db- <addr>`, instead of `db-<addr>`
* Fix `dbW` handler and update `rz_w32_add_winmsg_breakpoint`
* Remove checks for `argc` in `cmd_debug.c` (from review)
* Fix failing Windows tests
* Change control flow in conditional expressions
* Use `RzCmdStateOutput` in handler for `dbt` to print backtrace in various formats
* Use forward slash (/) instead of pipe (|) for command descriptions
* Use output modes for printing breakpoints in `db` handler
* Remove extra newline while printing breakpoints in JSON format
* Change description for `dbc` command
* Update descriptions and handlers; Remove <addr> argument wherever necessary
* Use `RZ_CMD_ARG_TYPE_CHOICES` for perm in `dbw`
* Add details for `dbC` and `dbw`
* Add quiet printing mode for breakpoints
* Remove unnecessary <addr> argument for `db` commands
* Also add a quiet output mode for `dbl` and `dbt`
* Return if no breakpoint found in function `add_breakpoint`
* Update tests for new `db` commands
* Update `db` command descriptions
* Remove `dbi.` redundant command
* Add details for iterating through the list of breakpoints
* Extract `rz_debug_bp_add_noreturn_func` into `debug.c` as RZ_API
* Fix `dbil` handler and `dbg_bps` test
* Shift the noreturn function breakpoint to `cdebug.c`
* Format according to clang-format
* Fix failing integration tests
* TODO: Find why `db- @@c:dblq` is failing, and fix the bug
* Update `db` commands
* Use `db- @@c:dblq` instead of `db-*`
* This will also work as a test for command edits (see https://github.com/rizinorg/rizin/pull/1806)
from the 14.x (next) release, since the snapshot update, interface is available
and allows enough flexibility thus here we re trying to dump data as much
as possible.
* Add copyright and license to all librz/include/*.h files in dep5
* SPDX headers for most include files except bin/arch/debug
* Add SPDX headers in asm/arch/{8051,arm}
* Add SPDX header for shlr/gdb include files
* Add SPDX header for shlr/bochs include files
* Add SPDX header for librz/reg
* Add SPDX header for librz/debug/p/native/xnu
* Add SPDX header for librz/debug/p/native/windows
* Add SPDX header for librz/debug/p/native/reg include files
* Add SPDX header for librz/debug/p/native/maps include files
* Add SPDX header for librz/debug/p/native/linux include files
* Fix regression showing threads `rip` for `dpt` on Windows
* Improve `dpt` output
* Consider flags too when printing thread address
* Improve `RzCoreBind->getNameDelta` to use realname if `flags->realnames` is true
* Fix negative offsets
* Fix flagspace selection side-effects after calling `rz_core_bin_apply()` APIs
* Fix wrong register profile after attaching to a process with `dpa`
* Fix wrong base address when opening a process memory
* Fix wrong register profile after opening a process memory
* Implement `rz_debug_reg_profile_sync()`
* Fix diff tests
* Remove doubled `rz_reg_set_profile_string()` call in `rz_debug_use()`
* Remove `RzDebug->user` and use `RzDebug->plugin_data` instead
* Fix crashes when switching between `native` and `windbg` debug plugins
* Fix some handle leaks
* Fix some memory leaks
* Re-use handles if possible
* Avoid allocations inside loops in `w32_desc_list()`
* Use debug plugin functions instead of reimplementing logic in io_debug.c and io_w32dbg.c to maintain state consistency
* Remove unneeded code
* Rename some functions
