* fix: format with clang-format-20
* fix: fix clang-format linter with rewrite '^#define.*/\*.*\\$'
* fix: update clang-format to version 20 in workflows and documentation
* fix: add SPDX license information to .git-blame-ignore-revs
- separate plugin archos wise.
- move all register profiles under reg directory.
- move architecture specific from linux_debug.c and bsd_debug.c to respective files.
- replace eprintf with rz_sys_perror and RZ_LOG_*
- remove debug_native.c and reg.c
- refactor librz/debug/meson.build to compile the native plugins archos wise.
This improves code readability and maintainibility.
* Fix sys.c unused variable.
* Add native/unsupported.c
* Clean rz_debug.h from comments and functions which does not exists
* Add fall-thru comment for haiku os case
* Fix warning: "MIN" redefined /boot/system/develop/headers/posix/sys/param.h:18: note: this is the location of the previous definition
* Port `dmhb` to RzShell
* Port `dmhf` to RzShell
* Remove 'to' argument from rz_core_debug_continue_until(). It was never used in the whole function. Hence, whatever it was supposed to do, it didn't
* Port `dcu` to RzShell
* Fix tests of `dmh[bf]` commands
* Port `dmi` commands to RzShell
* Port `dmx` to RzShell
* Remove `dta` command.
Originally it limited the addresses to trace.
But it was not used in any test, and the implementation was not optimal.
It allowed only specific addresses (no range).
Also worked on strings (not ut64 values).
Also, the usefulness is limited, considering the alternative
to just not tracing a certain address.
* Add quiet and quietest mode to `dmi` commands.
This also removes the test for `dmias`.
The command was not documented and it doesn't seem to
do anything special `dmia` can't do.
* Refactor: Use HtSP for storing plugins.
This saves iterations when searching for a plugin of a given name.
It also allows to save plugin configurations and data in hash table,
all addressed by their name. Hence, usage is more streamlined.
* Fix: iterators iterate over pointers of pointers.
* Soften the NULL check.
If arch is passed on from another struct, NULL is an allowed invalid value.
* Fix some leaks
* Fix NULL checks and incorrect inits
* Fix more leaks
* Add warning if plugin with same name was added before and is overwritten.
* Document rz_str_cmp
* Add function to create new list from iterator.
* Sort plugins alphabetically when listed.
* Fix parameter attributes, NULL checks, leaks and single rename.
Use the p_stat of the main thread to decide in which state the process is.
For multi-threaded applications this is still not quite correct but at
least this compiles on OpenBSD-current.
See also 5d4f90ee10
Fix for #4576
* Refactor sdb_foreach and related APIs
* Strengthen bounds check in sdb_dump_next()
* Switch to RzPVector
* Use length info in filter_noreturn()
* Fix cmp function
* Remove unused sdb_like() API
* Pass SdbKv to SdbForeachCallback
* Add sdbkv_dup_value() helper
* Fix memory leaks in debug path
* Fix memory leaks in mach and mdmp format
* Update the fix for mach
* Add support for aarch64 tcache parsing
* Test case fix
* Comment update
---------
Co-authored-by: Giridhar Prasath R <giridh1337@gmail.com>
- All `ht_*_new0()` are removed
- `freefn` field of HT options was renamed to `finiKV`
- `finiKV_user` fields was added to HT options
- Added `ht_*_new_opt_size` API
- `HtSP` and `HtUP` are created with `ValueFree` callback that allows to reduce extra LOC and prevent bugs
- `SetP` replaced with `SetS` (based on `HtSP`)
- `rz_th_ht_*_new0()` and `rz_th_ht_*_new_opt()` are replaced with `rz_th_ht_*_new(HtXX *)`
Some more APIs were incorrectly checked for with __POWERPC__, even
though they have nothing to do with the architecture, but were only
introduced in Mac OS X 10.6.
- Add tcache heap heuristics
- Fetch glibc version from libc
- Add unit testcases for glibc version
Co-authored-by: Giridhar Prasath R <giridh1337@gmail.com>
If a function calls 'rz_cons_break_push()' but never calls 'rz_cons_break_pop()' before return,
the stack count of 'RzConsContext->break_stack' contains too many elements (each time one too much).
This in turn will lead to not resetting 'RzConsContext->breaked' flag.
Because the flag is only set to false, if 'rz_stack_is_empty(context->break_stack) == true'
(in 'rz_cons_context_break_push()').
This wasn't a problem so far, because 'RzConsContext->breaked' is simply never set to true
(exceptions are some timeout cases as far as I can see).
Also these cases when 'rz_cons_break_pop()' was forgetten to be called, were edge error cases.
So not often hit.
But if Rizin is usd by Cutter 'RzConsContext->breaked' is set to 'true',
if an `AnalysisTask` interrupt is handled (in 'AnalysisTask::interrupt()').
This interrupt is triggered for example, when the introduction dialog is closed
and the main Cutter window opens (after the optional 'aaa').
Now, if the binary file was analysed with 'aaa', and a lot of error cases were hit,
those error cases sometimes never called 'rz_cons_break_pop()' before returning from their function.
Although, of course, they should have to the `RzConsContext->break_stack` is in a proper state.
This means, when the main Cutter window opens binary files which trigger many error edge cases,
the `RzConsContext->break_stack` is not empty
(because of the not executed 'rz_cons_break_pop()').
This also means, that the last thing done, was setting 'RzConsContext->breaked = true'
(by 'AnalysisTask::interrupt()').
If Cutter wants to show some disassembly, it calls 'rz_core_print_disasm()' which checks
'RzConsContext == false' via 'rz_cons_is_breaked()'. This condition is never true, because
the flag was not reset to `false` because the stack was never empty.
So it returns before anything was disassembled.
Hence Cutter gets no disassembly text.
Fixes https://github.com/rizinorg/cutter/issues/2552
Fixes https://github.com/rizinorg/cutter/issues/3275
