Ghidra 10.2 is fully backward compatible with project data from previous releases. However, programs and data type archives - which are created or modified in 10.2 will not be useable by an earlier Ghidra version.
+Ghidra 10.3 is fully backward compatible with project data from previous releases. + However, programs and data type archives which are created or modified in 10.3 will not be useable by an earlier Ghidra version.
This release includes many new features and capabilities, performance improvements, quite a few bug fixes, and many pull-request - contributions. Thanks to all those who have contributed their time, thoughts, and code. The Ghidra user community - thanks you too!
+ contributions. Thanks to all those who have contributed their time, thoughts, and code. The Ghidra user community thanks you too! -IMPORTANT: Ghidra requires Java 17 JDK to run. A newer version of Java may be acceptable, but has not been tested. Please see the +
IMPORTANT: Ghidra requires Java 17 JDK to run. A newer version of Java may be acceptable but has not been fully tested. Please see the Ghidra Installation Guide for additional information.
NOTE: Please note that any programs imported with a Ghidra beta versions or code built directly from source outside of a release tag may not be compatible - and may have flaws that have been corrected. Any programs analyzed from a beta or other local master source build should be considered experimental and - re-imported and analyzed with a release version. As an example, Ghidra 10.1 beta had an import flaw affecting symbol demangling that was not correctable. - Programs imported with previous release versions should upgrade correctly through various automatic upgrade mechanisms. Any program + and may have flaws that won't be corrected by using this new release. Any programs analyzed from a beta or other local master source build should be considered + experimental and re-imported and analyzed with a release version. As an example, Ghidra 10.1 beta had an import flaw affecting symbol demangling that was not + correctable. Programs imported with previous release versions should upgrade correctly through various automatic upgrade mechanisms. Any program you will continue to reverse engineer should be imported fresh with a release version or a build you trust with the latest code fixes.
-NOTE: Ghidra Server: The Ghidra 10.2 server is compatible with Ghidra 9.2 and later Ghidra clients. Ghidra 10.2 - clients are compatible with all 10.x and 9.x servers.
+NOTE: Ghidra Server: The Ghidra 10.3 server is compatible with Ghidra 9.2 and later Ghidra clients. Ghidra 10.3 + clients are compatible with all 10.x and 9.x servers. although due to potential Java version differences, it is strongly recommended + Although compatible in most cases, it is suggested that users using Ghidra versions before 10.2 upgrade their servers to 10.3. + Those using 10.2 and newer should not need a server upgrade.
NOTE: Platform-specific native executables can be built directly from a release distribution. The distribution currently provides Linux 64-bit, Windows 64-bit, and MacOS x86 binaries. If you have another platform, @@ -72,165 +73,174 @@ demangler, and legacy PDB executables for your plaform. Please see "Building Ghidra Native Components" section in the the Ghidra Installation Guide for additional information.
-A Software Bill of Materials (SBOM) is now included in the Ghidra release. The SBOM follows the CycloneDX standard, - and can be used with tools such as Dependency-Track to help identify risk in the software supply-chain.
+Ghidra now supports UI theming, which allows for full customization of colors, fonts, and icons used consistently throughout the application. + Ghidra themes are built on top of the various Java Look and Feel classes. Included are standard themes for all the supported + Look and Feels. The most notable is the Flat Dark theme, which is built using the FlatLaf, a modern open-source flat Look and Feel + library. Additionally, Ghidra includes various tools for editing and creating custom themes.
+ +Also, all the main display windows (Listing, Decompiler, and Bytes Viewer) support quickly changing the font size via < Ctrl > + or < Ctrl > -.
+ +See the Ghidra Help pages for full details on the theming feature.
The Debugger improvement highlights include:
+Perhaps the most exciting debugger change is the addition of new training course materials for the Debugger. The materials are written in + Markdown so they display right on GitHub, but they can also be rendered to nice HTML pages by Pandoc for offline viewing. They are suitable + both for self-paced learning and classroom environments. Even if you have used our Debugger before, we highly recommend reading these materials. + They are in the docs/GhidraClass directory with the other course materials.
+There are several changes to improve the user experience with the Emulator:
+-
+- FlatDebuggerAPI is introduced, providing a scripting API for Java-based GhidraScripts. An example DemoDebuggerScript.java - is included to get started.
-- P-code Emulation is improved, including numerous fixes, a new framework for system calls in emulation scripts, and a - prototype taint analyzer.
-- Compatibility is improved, including support for GDB versions 8.0.1 through 12.1, and LLDB version 14.0.
-- Support for memory/register editing is improved in Registers, Dynamic Listing, Memory, and Watches panels.
-- A new Frida connector is introduced, including support for debugging using Frida on USB/remote devices.
+- There is a dedicated Emulator tool. Previously, it was not apparent an Emulator GUI even existed in the Debugger tool. Most only + accessed it via scripting. The Emulator tool is the same as the Debugger tool, but without the back-end debugger management plugins. + This both showcases the Emulator and makes it safer to access, e.g., when examining malware. The launch buttons are removed, nearly + eliminating the risk of accidental detonation.
+- The control actions (step, suspend, resume, etc.) have been moved to the main toolbar. When toggled to control the emulator, it is now + possible to emulate to the next breakpoint. Before, it was only possible to step. If you were savvy, you could use the Go To Time + action to run many steps, but you had to predict precisely how many steps. These controls present the Emulator as a more traditional + trap-and-trace debugger and retain support for time travel.
+- Breakpoints are now applied to the Emulator. They also support injecting custom Sleigh semantics into the Emulator. This makes + it possible, e.g., to stub out external function calls. Breakpoints are now displayed in the Decompiler margin, too.
+- Regarding uninitialized/undefined memory, the Emulator will still treat undefined bytes as zeros. When decoding an + instruction; however, it will now interrupt if when encounters undefined bytes. Previously, it would just decode them as if + zeros, which was never useful.
+- Nascent support for stack unwinding has been added. Up to now, we have relied on the back-end debugger to unwind the stack, + which ruled out displaying accurate stack frames during emulation. There is still more work for full UI integration, but you can + unwind a stack (whether on target or emulated) using the Debugger -> Analysis menu and view the results by navigating the + Dynamic Listing to stack space. Please understand it may not work in most situations, yet.
+- We have added several miscellaneous actions: To invalidate the Emulator cache, use the Debugger -> Configure Emulator menu. + Use this whenever the Emulator seems to be ignoring configuration changes, especially when modifying custom Sleigh breakpoints. + To display all bytes (not just changed ones) in the Dynamic Listing, choose Load Bytes from Emulator in the Auto-Read drop-down. + To manually add or remove memory regions, e.g., to create and initialize a heap for emulation, use the new actions in the Regions window
+
There are several Debugger UI improvements:
++++
+- The control actions are duplicated in the main toolbar. Previously, these were only in the Objects window. (They remain there for + back-end connector/model development, troubleshooting, and diagnostics.) The actions in the main toolbar can be toggled to control a + live target or the Emulator. The Emulator stepping actions have been removed from the Threads panel. (They never really made sense there.) + Toggling these actions to the Emulator effectively forks an emulator from the target's live state, i.e., for extrapolation, just as the old + emulator stepping actions did.
+- We added text in the top right of the Dynamic Listing to display the current program counter (or whatever the listing is configured to + track). It will display in red if the address cannot be shown in the listing, e.g., because it is not mapped in memory. This provides better + feedback when the listings seem to be out of sync.
+- We added GDB's advance command to the Listing context menus as well as the equivalent actions for other debuggers. (More generally, + any command provided by a back-end connector that takes a single address parameter is presented in context menus where an address is + available.)
+- The Go To dialog in the Dynamic Listing can now take simple addresses in hexadecimal. Previously, it only took Sleigh expressions, + which are powerful, but made the common case too complicated. It still accepts Sleigh expressions, and those expressions can now refer + to labels (symbols) from any mapped program database (static image).
+- We added a new kind of hover on variables. If there is a debugger target (live or emulated) mapped to the current program, + the hover will display the variable’s current value. This applies to Listings and the Decompiler window.
+- You can now select a different thread, frame, or snapshot without activating it. Single-click to select. Double-click to activate.
+
There are a few small improvements to back-end debugger integration:
+++
- You can now set the working directory when launching a Windows target.
+- GADP agents now accept a single connection and automatically terminate when Ghidra disconnects.
+- We added launch scripts for starting a GADP agent from the command line.
+- There is now a script to build the Java bindings needed for the LLDB connector.
The Decompiler has a myriad of improvements in the latest-release. Many have been long-requested features or improvements. - Highlights of the changes include:
--+-
-- Support for union data-types. The Decompiler scores and displays the most likely field based on how code accesses the union. Alternately, a field access can be set manually.
-- Support for pointers with an offset relative to the start of a data type, usually a structure. Examples include windows LIST_ENTRY/CONTAINING_RECORD linked lists, - CString allocation data, and memory allocation records.
-- Support for pointers with a specified address space. Useful for targeting a specific address space such as SPI memory or - in Harvard architectures with multiple address spaces.
-- Improved reconciliation of overlapping views of data-types; for example, passing of sub members of a structure to a function.
-- Marker Margins, similar to the listing marker margins, have been added to display things like Debugger breakpoints.
-- A colored highlighting service has been added, allowing clients to create highlights in the form of background colors for syntax tokens in - the Decompiler UI through API calls.
-- Read-from and write-to access to a volatile variable now display as simple assignments, with a special token color, instead of as read- or write-volatile function calls.
-
Support has been added for expanding assignment statements on structures or arrays, where multiple fields or elements are moved as a + group by a single instruction. This is especially helpful for analyzing structure initialization code and stack strings.
+ +Support continues to improve for structures that are either stored across multiple registers or in a single register that is + accessed in pieces. Data types associated with the component fields are propagated more fully throughout the function, and assignments + to fields are displayed simply.
With this release of Ghidra, support for Pointer Typedefs has been expanded to facilitate the use of specialized - data type settings. Improvements have also been made to ensure that such settings are preserved within data type - archives and merge situations. These settings are not supported at the instance-level and are intended to be an - attribute of the associated pointer. The Typedef provides the ability to tailor a pointer for a specific use. It - is highly recommended that all required Pointer Typedef settings be applied prior to using the data type - (e.g., for defined data, data type components, and variables) since there is currently no change propagation for such modifications. +
Data Type Archives may now optionally target a specific architecture as specified by a processor and associated compiler specification + such as data organization. This has the advantage of better conveying datatype details for a desired architecture and preserving aspects + which may change when resolved into a program. In the future, this will also allow function definitions to retain architecture-specific + details.
-The following Pointer Typedef settings have been introduced with this release:
--- --
-- Address Space - allows the destination address space to be specified for a pointer. While this does not affect pointer dereferencing operations - dictated by instruction semantics, it can aid analysis and the generation of associated memory references.
-- Component Offset - provides the ability to specify an offset relative to the associated pointer's referenced data type such that: -
---<referenced-data-type-storage-address> = <pointer-offset> - <component-offset-setting>-- Offset Mask - bit-mask to be applied prior to any bit-shift (if specified) during the computation of an actual address offset
+Function definition data types have been improved to preserve calling convention names which may differ from the predefined generic + calling convention names to include those which may have originated from an extended compiler specification. In addition, function + definitions now support the noreturn attribute.
-- Offset Shift - bit-shift to be applied after any bit-mask (if specified) during the computation of an actual address - offset (positive: left-shift, negative: right-shift)
- -- Pointer Type - facilitates special interpretation of pointers
- ----
-- default - normal pointer
-- image-base-relative - pointers whose offset should be treated as relative to the program's image base (e.g., relative virtual address (RVA))
- -- relative - pointers whose offset is relative to the pointer's storage address
- -- file-offset - pointers whose offset corresponds to an offset within the loaded binary file (limited to single load file)
-
NOTE: The use and consumption of Pointer Typedef settings is in its early stages and may not be utilized by various analyzers. - In addition, some settings are not relevant to some analyzers where instruction semantics will dictate pointer dereferencing.
- -At the API level, the PointerTypedef and PointerTypedefBuilder classes have been added to simplify the creation of a Pointer Typedef. - While an explicit Typedef name may be used, Pointer Typedefs also support an auto-naming mechanism (constructed with a null/empty name) - which will simply use the pointer name followed by the settings as an attribute list; example:
-- --int * __((space(ram)))-
Within the GUI, using the New->Typedef on <pointer> action on a selected pointer within the Data Type Tree is the quickest way to create one. - Once this is done, use the Settings... action on the selected Pointer Typedef. The Settings dialog will be displayed allowing the various settings - to be applied to the Typedef. Settings should be made to Typedef prior to applying it since settings change propogation is very limited.
+Enum handling has been improved in the data type manager when creating new enums from an existing set of enum values, + for example “define_” enums parsed from header files. Enum values will be automatically sized to fit all the values contained + in the enum. Setting the size of an Enum will check if the values will fit within the new size. In addition, “define_” values + created as enums with a single value are sized to the minimum size to fit the value. Parsed enums from header files are sized based + on the declared size of an int from the data organization used to parse. A future version will have a setting to size all parsed enums + to the smallest size that will fit all the values.
+ +The C-Parser GUI has been refactored to remove include paths from the Options section done as –D define lines, to a new Include section. + This should make it easier to configure paths to the include files and has the added benefit of coloring the include file entries red if + they are not found within any include path. You may find creating and using a Ghidra Script instead of the GUI an easier repeatable process. + There are several included examples scripts, including ones to parse AVR8 header files, and Visual Studio version 22 files.
-C-Parser support has been added for missing C specification syntax from C11 and C23, such as tags, macros with varargs, and _NoReturn. - Numerous parsing errors have also been fixed, including for arrays of function pointers, array definitions, and placement of compiler directives. - In addition, parsing time of extremely large header files has been drastically reduced.
-Error handling and reporting from the Pre-Processor and C-Parser have been improved.
-Several scripts to parse header files outside of the GUI have been included, including one that specially parses AVR8 data types and memory-mapped register - definitions from header files for each AVR8 processor variant. The scripts are CreateAVR8GDTArchiveScript.java, CreateExampleGDTArchiveScript.java, CreateJNIArchivesScript.java, - and CreateDefaultGDTArchives.java.
-Finally, data types in open archives can be used during parsing for undefined data types in a header file. At the start of parsing, use of open - archives can be chosen or ignored without closing open archives. The header files must still parse without error, - however a missing data type or unfound header file may not cause the parsing to fail if an open archive contains a missing, but needed data type definition.
- +All supplied data type archive GDT files, except macOS, have been re-parsed to include the new processor architecture.
Mach-O binary analysis continues to improve. Support has been added for new file formats introduced in iOS 16 and macOS 13. - Improvements have also been made to function identification, symbol detection, and Objective-C support.
- -Import and analysis of the entire existing set of Android binaries up to version 13.x is now supported, including new support for the Multi-DEX format. - The type of binaries supported include: Android Run-Time (ART), Ahead-of-Time (OAT)/ELF, Dalvik Executables (DEX), Multi-DEX, Compact DEX (CDEX), - Verified DEX (VEX), Boot Image, and Boot Loader formats. Also included are Sleigh modules for DEX files covering each major release of Android; - the optimized instructions vary across versions.
- -A new Android APK loader will load all DEX files at one time and link the method_lookup
- sections using external references. The new APK loader uses the manifest file to determine the Android version.
The option Assume Contiguous Functions Only, for the Shared Return Analyzer, has been turned on by default. - The Shared Return Analyzer turns jump instructions into a call if the jump - target is, or should be, considered a function. When turned on, the option treats a jump - over a known function entry point to be a call, even if there is only one jump to that location. The option improves thunk function - recovery as well as decompilation results by using a call to the function instead of including the called functions code within the calling function.
+New ApplyDataArchives analyzer settings enable use of locally created GDT data type archive files or project archives in the + analysis pipeline. Used in conjunction with analysis options settings saved to a named analysis configuration you can easily switch to using a new + GDT file and associated analysis options for a given type of binary. For example, if you are working with AVR8 binaries and have + an associated AVR8.gdt file, create an AVR8 configuration and it will be used as the default analysis options configuration until + you change to a new configuration.
-The option has been turned on by default for all processor types except ARM. ARM Thumb binaries can sometimes use BL instructions, - normally used as calls, as an internal jump within a large function. If this option were on by default for such a binary it would cause - additional erroneous functions to be created. The option can be used on ARM binaries, however they should be all ARM code; otherwise any Thumb code - using BL for far jumps must be fixed using the Fix_ARM_Call_JumpsScript and Override_ARM_Call_JumpsScript.
+Constant Propagation now deals with constants passed as stack parameters. In addition, there are several new settings which can better + control when a constant is considered to be an address. For example, processors with small memory spaces, the setting “Require pointer param + data type,” will only create a reference if the parameter is declared with a data type that would be a pointer. This can be useful for Harvard + architectures with multiple address spaces used in conjunction with the PointerTypedef to specify the address space of the pointer. Currently, + once you change the parameter of a called function to be a pointer, you will need to re-run analysis to get the constants passed to the function + to be turned into a reference. This will be automated in the near future.
-An optional MachineLearning extension has been added containing the Random Forest Function Finder Plugin. - The plugin finds undiscovered functions within a binary using classifiers to identify potential function starts. - The plugin trains classifiers using data sets created from known functions within a binary. - These classifiers can then be used by the plugin on the original binary or other binaries to find additional functions - missed by initial analysis.
-The extension can be installed from the Ghidra Project Window via File->Install Extensions...
- +By default, pointer-to-pointer analysis is turned off for ARM binaries in the Operand and Data Reference analyzers. This can result in fewer + references created and can be turned back on if your ARM binaries use pointers data stored in memory instead of offset values from the current PC + to calculate all references.
+ +Added support for PE MinGW pseudo-relocation processing.
+ +Folder and file links to contents of another shared project repository may now be added to a Ghidra Project. This could allow a team to + include a program or subfolder that resides in another project rather than copying the program into your project for easy access. The linked + files are opened for read-only viewing.
+Updated ARM32 and AARCH64 to version v9.3 to include vfp4 instructions.
-Improvements and bug fixes have been made to many processors to include: AARCH64, ARM, AVR8, AVR32, Coldfire, JVM, MIPS, MSP430, PA-Risc, PowerPC, - RISC-V, SuperH, Tricore, V850, X86, 6502, and 68K.
- Sleigh now supports inst_next2 as well as inst_next to support branching around the next instruction when its length is unknown.
- Many processors have conditional skip instructions which can be used on any instruction, including another skip instruction.
- Some sleigh processor developers have tried to use the delayslot() directive to accomplish instruction skipping. Unfortunately, the use of the delayslot() directive
- can cause nested delay slots or the potential for branches into the delay slotted instruction, both of which are not supported.
Improvements and bug fixes have been made to many processors since 10.2 to include: + AARCH64, ARM, Coldfire, HCS12 MIPS X86, PowerPC, RISCV SPARC, SuperH, TriCore, V850, Z80, 6x09, 68K, and 8051.
+ +Two new user-submitted processors, eBPF and BPF, add support for two variants of Berkeley Packet Filter binaries.
+ +A user-submitted refactoring of X86 LOCK/UNLOCK decoding and semantics has been committed. There are currently some issues with the + Decompiler re-arranging code outside of the LOCK/UNLOCK which will be addressed an upcoming patch. If your analysis depends on + the LOCK/UNLOCK semantics, please be aware of the issue.
+ +A new “leading zeroes count” operator, called lzcount, has been added to p-code, and it can now be used by SLEIGH developers + to model processor instructions. The Decompiler can simplify common code idioms using these instructions, and emulation is supported.
The Go To... dialog now provides navigation to file offsets. In addition, a new File Offset field is available in the Listing. The
- field must be added to the Listing using Edit Listing Fields.
- These new features can greatly simplify correlating bytes in program memory with their original location within the file from which they were imported.
- Example: to go to the memory location which corresponds to the first byte in the original file, enter file(0) in the Go To... dialog.
Diff can now be performed between two open programs which may include remote files previously opened via a Ghidra-URL.
-Support has been added for loading WinDbg and APPORT dump files.
-Redesigned the Importer's library loading options to provide finer-grained control over where libraries are searched - for on disk and in the project, as well as where newly loaded libraries are saved to.
+An importer, Analyzer, and Internal changes have been made to support GoLang. Currently, only version 1.18 is supported; however slightly older or newer versions may work. + There are still some Decompiler issues with multiple return parameters to be worked out, however the implementation was thought complete enough + for initial real use. Please consider the feature an evolving initial implementation.
+ +Ghidra now remembers the last location of a program when it is closed. When that program is later re-opened, Ghidra will position the + program to that location. Also, there are options for where Ghidra should start for new programs and optionally when Ghidra completes + the initial analysis.
+ +Ghidra now has options for simplifying the display of symbol names, in both the Listing and Decompiler, with complex template information + embedded in them. The simplification should result in a much less busy display when dealing with templates.
Numerous other bug fixes and improvements are fully listed in the ChangeHistory file.