You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Always meant to ask -- why is the context module implemented so strangely? It implements a ModuleType object and replaces itself with a singleton. Why was this design used rather than a normal object?
The text was updated successfully, but these errors were encountered:
* Added support for using different malloc algorithms and implemented
ptmalloc commands
- Added support for the mp_ structure from glibc
- Fixed the bug when the base of [heap] is not sbrk_base (occurs
when bss is very large).
- Added an interface for interacting with ptmalloc heaps
* Added fastbins, unsortedbins, smallbins, largebins
- Implemented the bin functions both in the ptmalloc interface in
the heap/ directory and the commands function
- Completed the bins command which prints all the commands
- Added option to bin commands `verbose` specifies the users desire
to print empty bins - linked lists with out any nodes from malloc
- Changed bin representation to OrderedDict
* remove old bins command
* Ptmalloc bug fixes
- switched constants.ptmalloc from using pwndbg.typeinfo to a gdb
api function to prevent bugs when libc-dbg isn't installed
- Other minor changes
* Added some functionality for when libc-dbg is not present
- Added checks for finding symbols
* Ran isort on pwndbg
* Made few changes based on feedback from pull requests
- Cleaned up spaces_table
- Syntax changes
* Moved formatting code away from logic
- Moved format_bin from pwndbg.heap.ptmalloc to pwndbg.commands.heap
* Fixed bug in spaces_table for 32 bit archs
* Changed malloc_chunk command to display FASTBIN instead of PREV_INUSE
- Fixed local isort version problems
* Small bug fixes, tweaks, and added features list.
- Changed pwndbg.chain.format to use the offset for the very last
index in the chain.
- Fixed bug with ptmalloc's main_arena property.
- Added images to cap and links to FEATURES.md.
- Changed bin commands to print hex instead of base 10.
* Modified docstring to match other docstrings in pwndbg
* Memoize properties, fix merge conflicst, fix constants
- Moved the macros from the constants directory into the Heap
class, left the constants there.
- Switched from lazy loading to the memoize events.
* Fixed double spaces after images from a previous pull request.
* Added more examples to FEATURES.md for heaps
* Fix typos in FEATURES.md
Always meant to ask -- why is the
context
module implemented so strangely? It implements aModuleType
object and replaces itself with a singleton. Why was this design used rather than a normal object?The text was updated successfully, but these errors were encountered: