What has changed in GDB? (Organized release by release) *** Changes since GDB 14 * Building GDB and GDBserver now requires a C++17 compiler. For example, GCC 9 or later. * GDB index now contains information about the main function. This speeds up startup when it is being used for some large binaries. * Changed commands disassemble Attempting to use both the 'r' and 'b' flags with the disassemble command will now give an error. Previously the 'b' flag would always override the 'r' flag. * New commands info missing-debug-handler List all the registered missing debug handlers. enable missing-debug-handler LOCUS HANDLER disable missing-debug-handler LOCUS HANDLER Enable or disable a missing debug handler with a name matching the regular expression HANDLER, in LOCUS. LOCUS can be 'global' to operate on global missing debug handler, 'progspace' to operate on handlers within the current program space, or can be a regular expression which is matched against the filename of the primary executable in each program space. maintenance info linux-lwps List all LWPs under control of the linux-nat target. set remote thread-options-packet show remote thread-options-packet Set/show the use of the thread options packet. * New features in the GDB remote stub, GDBserver ** The --remote-debug and --event-loop-debug command line options have been removed. ** The --debug command line option now takes an optional comma separated list of components to emit debug for. The currently supported components are: all, threads, event-loop, and remote. If no components are given then threads is assumed. ** The 'monitor set remote-debug' and 'monitor set event-loop-debug' command have been removed. ** The 'monitor set debug 0|1' command has been extended to take a component name, e.g.: 'monitor set debug COMPONENT off|on'. Possible component names are: all, threads, event-loop, and remote. * Python API ** New function gdb.notify_mi(NAME, DATA), that emits custom GDB/MI async notification. ** New read/write attribute gdb.Value.bytes that contains a bytes object holding the contents of this value. ** New module gdb.missing_debug that facilitates dealing with objfiles that are missing any debug information. ** New function gdb.missing_debug.register_handler that can register an instance of a sub-class of gdb.missing_debug.MissingDebugInfo as a handler for objfiles that are missing debug information. ** New class gdb.missing_debug.MissingDebugInfo which can be sub-classed to create handlers for objfiles with missing debug information. ** Stop events now have a "details" attribute that holds a dictionary that carries the same information as an MI "*stopped" event. ** New function gdb.interrupt(), that interrupts GDB as if the user typed control-c. * Debugger Adapter Protocol changes ** GDB now emits the "process" event. * New remote packets New stop reason: clone Indicates that a clone system call was executed. QThreadOptions Enable/disable optional event reporting, on a per-thread basis. Currently supported options are GDB_THREAD_OPTION_CLONE, to enable clone event reporting, and GDB_THREAD_OPTION_EXIT to enable thread exit event reporting. QThreadOptions in qSupported The qSupported packet allows GDB to inform the stub it supports the QThreadOptions packet, and the qSupported response can contain the set of thread options the remote stub supports. *** Changes in GDB 14 * GDB now supports the AArch64 Scalable Matrix Extension 2 (SME2), which includes a new 512 bit lookup table register named ZT0. * GDB now supports the AArch64 Scalable Matrix Extension (SME), which includes a new matrix register named ZA, a new thread register TPIDR2 and a new vector length register SVG (streaming vector granule). GDB also supports tracking ZA state across signal frames. Some features are still under development or are dependent on ABI specs that are still in alpha stage. For example, manual function calls with ZA state don't have any special handling, and tracking of SVG changes based on DWARF information is still not implemented, but there are plans to do so in the future. * GDB now recognizes the NO_COLOR environment variable and disables styling according to the spec. See https://no-color.org/. Styling can be re-enabled with "set style enabled on". * The AArch64 'org.gnu.gdb.aarch64.pauth' Pointer Authentication feature string has been deprecated in favor of the 'org.gnu.gdb.aarch64.pauth_v2' feature string. * GDB now has some support for integer types larger than 64 bits. * Removed targets and native configurations GDB no longer supports AIX 4.x, AIX 5.x and AIX 6.x. The minimum supported AIX version is now AIX 7.1. * Multi-target feature configuration GDB now supports the individual configuration of remote targets' feature sets. Based on the current selection of a target, the commands 'set remote -packet (on|off|auto)' and 'show remote -packet' can be used to configure a target's feature packet and to display its configuration, respectively. The individual packet sizes can be configured and shown using the commands ** 'set remote memory-read-packet-size (number of bytes|fixed|limit)' ** 'set remote memory-write-packet-size (number of bytes|fixed|limit)' ** 'show remote memory-read-packet-size' ** 'show remote memory-write-packet-size'. The configuration of the packet itself, as well as the size of a memory-read or memory-write packet applies to the currently selected target (if available). If no target is selected, it applies to future remote connections. Similarly, the show commands print the configuration of the currently selected target. If no remote target is selected, the default configuration for future connections is shown. * GDB has initial built-in support for the Debugger Adapter Protocol. This support requires that GDB be built with Python scripting enabled. * For the break command, multiple uses of the 'thread' or 'task' keywords will now give an error instead of just using the thread or task id from the last instance of the keyword. E.g.: break foo thread 1 thread 2 will now give an error rather than using 'thread 2'. * For the watch command, multiple uses of the 'task' keyword will now give an error instead of just using the task id from the last instance of the keyword. E.g.: watch my_var task 1 task 2 will now give an error rather than using 'task 2'. The 'thread' keyword already gave an error when used multiple times with the watch command, this remains unchanged. * The 'set print elements' setting now helps when printing large arrays. If an array would otherwise exceed max-value-size, but 'print elements' is set such that the size of elements to print is less than or equal to 'max-value-size', GDB will now still print the array, however only 'max-value-size' worth of data will be added into the value history. * For both the break and watch commands, it is now invalid to use both the 'thread' and 'task' keywords within the same command. For example the following commnds will now give an error: break foo thread 1 task 1 watch var thread 2 task 3 * The printf command now accepts a '%V' output format which will format an expression just as the 'print' command would. Print options can be placed withing '[...]' after the '%V' to modify how the value is printed. E.g: printf "%V", some_array printf "%V[-array-indexes on]", some_array will print the array without, or with array indexes included, just as the array would be printed by the 'print' command. This functionality is also available for dprintf when dprintf-style is 'gdb'. * When the printf command requires a string to be fetched from the inferior, GDB now uses the existing 'max-value-size' setting to the limit the memory allocated within GDB. The default 'max-value-size' is 64k. To print longer strings you should increase 'max-value-size'. * The Ada 2022 Enum_Rep and Enum_Val attributes are now supported. * The Ada 2022 target name symbol ('@') is now supported by the Ada expression parser. * The 'list' command now accepts '.' as an argument, which tells GDB to print the location around the point of execution within the current frame. If the inferior hasn't started yet, the command will print around the beginning of the 'main' function. * Using the 'list' command with no arguments in a situation where the command would attempt to list past the end of the file now warns the user that the end of file has been reached, refers the user to the newly added '.' argument * Breakpoints can now be inferior-specific. This is similar to the existing thread-specific breakpoint support. Breakpoint conditions can include the 'inferior' keyword followed by an inferior id (as displayed in the 'info inferiors' output). It is invalid to use the 'inferior' keyword with either the 'thread' or 'task' keywords when creating a breakpoint. * New convenience function "$_shell", to execute a shell command and return the result. This lets you run shell commands in expressions. Some examples: (gdb) p $_shell("true") $1 = 0 (gdb) p $_shell("false") $2 = 1 (gdb) break func if $_shell("some command") == 0 * Configure changes --additional-debug-dirs=PATHs Provide a colon-separated list of additional directories to search for separate debug info. These directories are added to the default value of the 'debug-file-directory' GDB parameter. * New commands set debug breakpoint on|off show debug breakpoint Print additional debug messages about breakpoint insertion and removal. maintenance print record-instruction [ N ] Print the recorded information for a given instruction. If N is not given prints how GDB would undo the last instruction executed. If N is negative, prints how GDB would undo the N-th previous instruction, and if N is positive, it prints how GDB will redo the N-th following instruction. maintenance info frame-unwinders List the frame unwinders currently in effect, starting with the highest priority. maintenance wait-for-index-cache Wait until all pending writes to the index cache have completed. set always-read-ctf on|off show always-read-ctf When off, CTF is only read if DWARF is not present. When on, CTF is read regardless of whether DWARF is present. Off by default. info main Get main symbol to identify entry point into program. set tui mouse-events [on|off] show tui mouse-events When on (default), mouse clicks control the TUI and can be accessed by Python extensions. When off, mouse clicks are handled by the terminal, enabling terminal-native text selection. * MI changes ** MI version 1 has been removed. ** mi now reports 'no-history' as a stop reason when hitting the end of the reverse execution history. ** When creating a thread-specific breakpoint using the '-p' option, the -break-insert command would report the 'thread' field twice in the reply. The content of both fields was always identical. This has now been fixed; the 'thread' field will be reported just once for thread-specific breakpoints, or not at all for breakpoints without a thread restriction. The same is also true for the 'task' field of an Ada task-specific breakpoint. ** It is no longer possible to create a thread-specific breakpoint for a thread that doesn't exist using '-break-insert -p ID'. Creating breakpoints for non-existent threads is not allowed when using the CLI, that the MI allowed it was a long standing bug, which has now been fixed. ** The '--simple-values' argument to the '-stack-list-arguments', '-stack-list-locals', '-stack-list-variables', and '-var-list-children' commands now takes reference types into account: that is, a value is now considered simple if it is neither an array, structure, or union, nor a reference to an array, structure, or union. (Previously all references were considered simple.) Support for this feature can be verified by using the '-list-features' command, which should contain "simple-values-ref-types". ** The -break-insert command now accepts a '-g thread-group-id' option to allow for the creation of inferior-specific breakpoints. ** The bkpt tuple, which appears in breakpoint-created notifications, and in the result of the -break-insert command can now include an optional 'inferior' field for both the main breakpoint, and each location, when the breakpoint is inferior-specific. * Python API ** gdb.ThreadExitedEvent added. Emits a ThreadEvent. ** The gdb.unwinder.Unwinder.name attribute is now read-only. ** The name argument passed to gdb.unwinder.Unwinder.__init__ must now be of type 'str' otherwise a TypeError will be raised. ** The gdb.unwinder.Unwinder.enabled attribute can now only accept values of type 'bool'. Changing this attribute will now invalidate GDB's frame-cache, which means GDB will need to rebuild its frame-cache when next required - either with, or without the particular unwinder, depending on how 'enabled' was changed. ** New methods added to the gdb.PendingFrame class. These methods have the same behaviour as the corresponding methods on gdb.Frame. The new methods are: - gdb.PendingFrame.name: Return the name for the frame's function, or None. - gdb.PendingFrame.is_valid: Return True if the pending frame object is valid. - gdb.PendingFrame.pc: Return the $pc register value for this frame. - gdb.PendingFrame.language: Return a string containing the language for this frame, or None. - gdb.PendingFrame.find_sal: Return a gdb.Symtab_and_line object for the current location within the pending frame, or None. - gdb.PendingFrame.block: Return a gdb.Block for the current pending frame, or None. - gdb.PendingFrame.function: Return a gdb.Symbol for the current pending frame, or None. ** The frame-id passed to gdb.PendingFrame.create_unwind_info can now use either an integer or a gdb.Value object for each of its 'sp', 'pc', and 'special' attributes. ** A new class gdb.unwinder.FrameId has been added. Instances of this class are constructed with 'sp' (stack-pointer) and 'pc' (program-counter) values, and can be used as the frame-id when calling gdb.PendingFrame.create_unwind_info. ** It is now no longer possible to sub-class the gdb.disassembler.DisassemblerResult type. ** The Disassembler API from the gdb.disassembler module has been extended to include styling support: - The DisassemblerResult class can now be initialized with a list of parts. Each part represents part of the disassembled instruction along with the associated style information. This list of parts can be accessed with the new DisassemblerResult.parts property. - New constants gdb.disassembler.STYLE_* representing all the different styles part of an instruction might have. - New methods DisassembleInfo.text_part and DisassembleInfo.address_part which are used to create the new styled parts of a disassembled instruction. - Changes are backwards compatible, the older API can still be used to disassemble instructions without styling. ** New function gdb.execute_mi(COMMAND, [ARG]...), that invokes a GDB/MI command and returns the output as a Python dictionary. ** New function gdb.block_signals(). This returns a context manager that blocks any signals that GDB needs to handle itself. ** New class gdb.Thread. This is a subclass of threading.Thread that calls gdb.block_signals in its "start" method. ** gdb.parse_and_eval now has a new "global_context" parameter. This can be used to request that the parse only examine global symbols. ** gdb.Inferior now has a new "arguments" attribute. This holds the command-line arguments to the inferior, if known. ** gdb.Inferior now has a new "main_name" attribute. This holds the name of the inferior's "main", if known. ** gdb.Inferior now has new methods "clear_env", "set_env", and "unset_env". These can be used to modify the inferior's environment before it is started. ** gdb.Value now has the 'assign' method. ** gdb.Value now has the 'to_array' method. This converts an array-like Value to an array. ** gdb.Progspace now has the new method "objfile_for_address". This returns the gdb.Objfile, if any, that covers a given address. ** gdb.Breakpoint now has an "inferior" attribute. If the Breakpoint object is inferior specific then this attribute holds the inferior-id (an integer). If the Breakpoint object is not inferior specific, then this field contains None. This field can be written too. ** gdb.Type now has the "is_array_like" and "is_string_like" methods. These reflect GDB's internal idea of whether a type might be array- or string-like, even if they do not have the corresponding type code. ** gdb.ValuePrinter is a new class that can be used as the base class for the result of applying a pretty-printer. As a base class, it signals to gdb that the printer may implement new pretty-printer methods. ** New attribute Progspace.symbol_file. This attribute holds the gdb.Objfile that corresponds to Progspace.filename (when Progspace.filename is not None), otherwise, this attribute is itself None. ** New attribute Progspace.executable_filename. This attribute holds a string containing a file name set by the "exec-file" or "file" commands, or None if no executable file is set. This isn't the exact string passed by the user to these commands; the file name will have been partially resolved to an absolute file name. ** A new executable_changed event registry is available. This event emits ExecutableChangedEvent objects, which have 'progspace' (a gdb.Progspace) and 'reload' (a Boolean) attributes. This event is emitted when gdb.Progspace.executable_filename changes. ** New event registries gdb.events.new_progspace and gdb.events.free_progspace, these emit NewProgspaceEvent and FreeProgspaceEvent event types respectively. Both of these event types have a single 'progspace' attribute, which is the gdb.Progspace that is either being added to GDB, or removed from GDB. ** gdb.LazyString now implements the __str__ method. ** New method gdb.Frame.static_link that returns the outer frame of a nested function frame. *** Changes in GDB 13 * MI version 1 is deprecated, and will be removed in GDB 14. * GDB now supports dumping memory tag data for AArch64 MTE. It also supports reading memory tag data for AArch64 MTE from core files generated by the gcore command or the Linux kernel. When a process uses memory-mapped pages protected by memory tags (for example, AArch64 MTE), this additional information will be recorded in the core file in the event of a crash or if GDB generates a core file from the current process state. GDB will show this additional information automatically, or through one of the memory-tag subcommands. * Scheduler-locking and new threads When scheduler-locking is in effect, only the current thread may run when the inferior is resumed. However, previously, new threads created by the resumed thread would still be able to run free. Now, they are held stopped. * "info breakpoints" now displays enabled breakpoint locations of disabled breakpoints as in the "y-" state. For example: (gdb) info breakpoints Num Type Disp Enb Address What 1 breakpoint keep n 1.1 y- 0x00000000000011b6 in ... 1.2 y- 0x00000000000011c2 in ... 1.3 n 0x00000000000011ce in ... * Support for Thread Local Storage (TLS) variables on FreeBSD arm and aarch64 architectures. * GDB now supports hardware watchpoints on FreeBSD/Aarch64. * Remove support for building against Python 2, it is now only possible to build GDB against Python 3. * DBX mode has been removed. * GDB now honours the DWARF prologue_end line-table entry flag the compiler can emit to indicate where a breakpoint should be placed to break in a function past its prologue. * Completion now also offers "NUMBER" for "set" commands that accept a numeric argument and the "unlimited" keyword. For example: (gdb) set width NUMBER unlimited and consequently: (gdb) complete set width set width NUMBER set width unlimited * Disassembler styling using libopcodes. GDB now supports disassembler styling using libopcodes. This is only available for some targets (currently x86 and RISC-V). For unsupported targets Python Pygments is still used. For supported targets, libopcodes styling is used by default. * The Windows native target now supports target async. * gdb now supports zstd compressed debug sections (ELFCOMPRESS_ZSTD) for ELF. * The format of 'disassemble /r' and 'record instruction-history /r' has changed. The instruction bytes could now be grouped together, and displayed in the endianness of the instruction. This is the same layout as used by GNU objdump when disassembling. There is now 'disassemble /b' and 'record instruction-history /b' which will always display the instructions bytes one at a time in memory order, that is, the byte at the lowest address first. For both /r and /b GDB is now better at using whitespace in order to align the disassembled instruction text. * The TUI no longer styles the source and assembly code highlighted by the current position indicator by default. You can however re-enable styling using the new "set style tui-current-position" command. * New convenience variable $_inferior_thread_count contains the number of live threads in the current inferior. * When a breakpoint with multiple code locations is hit, GDB now prints the code location using the syntax . such as in: Thread 1 "zeoes" hit Breakpoint 2.3, some_func () at zeoes.c:8 * When a breakpoint is hit, GDB now sets the convenience variables $_hit_bpnum and $_hit_locno to the hit breakpoint number and code location number. This allows to disable the last hit breakpoint using (gdb) disable $_hit_bpnum or disable only the specific breakpoint code location using (gdb) disable $_hit_bpnum.$_hit_locno These commands can be used inside the command list of a breakpoint to automatically disable the just encountered breakpoint (or the just encountered specific breakpoint code location). When a breakpoint has only one location, $_hit_locno is set to 1 so that (gdb) disable $_hit_bpnum.$_hit_locno and (gdb) disable $_hit_bpnum are both disabling the breakpoint. * New commands maintenance set ignore-prologue-end-flag on|off maintenance show ignore-prologue-end-flag This setting, which is off by default, controls whether GDB ignores the PROLOGUE-END flag from the line-table when skipping prologue. This can be used to force GDB to use prologue analyzers if the line-table is constructed from erroneous debug information. set print nibbles [on|off] show print nibbles This controls whether the 'print/t' command will display binary values in groups of four bits, known as "nibbles". The default is 'off'. maintenance set libopcodes-styling on|off maintenance show libopcodes-styling These can be used to force off libopcodes based styling, the Python Pygments styling will then be used instead. set style disassembler comment show style disassembler comment set style disassembler immediate show style disassembler immediate set style disassembler mnemonic show style disassembler mnemonic set style disassembler register show style disassembler register set style disassembler address show style disassembler address set style disassembler symbol show style disassembler symbol For targets that support libopcodes based styling, these settings control how various aspects of the disassembler output are styled. The 'disassembler address' and 'disassembler symbol' styles are aliases for the 'address' and 'function' styles respectively. maintenance print frame-id [ LEVEL ] Print GDB's internal frame-id for the frame at LEVEL. If LEVEL is not given, then print the frame-id for the currently selected frame. set debug infcall on|off show debug infcall Print additional debug messages about inferior function calls. set debug solib on|off show debug solib Print additional debug messages about shared library handling. set style tui-current-position [on|off] Whether to style the source and assembly code highlighted by the TUI's current position indicator. The default is off. set print characters LIMIT show print characters This new setting is like 'set print elements', but controls how many characters of a string are printed. This functionality used to be covered by 'set print elements', but it can be controlled separately now. LIMIT can be set to a numerical value to request that particular character count, to 'unlimited' to print all characters of a string, or to 'elements', which is also the default, to follow the setting of 'set print elements' as it used to be. print -characters LIMIT This new option to the 'print' command has the same effect as a temporary use of 'set print characters'. * Changed commands document user-defined It is now possible to document user-defined aliases. When a user-defined alias is documented, the help and apropos commands use the provided documentation instead of the documentation of the aliased command. Documenting a user-defined alias is particularly useful when the alias is a set of nested 'with' commands to avoid showing the help of the with command for an alias that will in fact launch the last command given in the nested commands. maintenance info line-table Add a PROLOGUE-END column to the output which indicates that an entry corresponds to an address where a breakpoint should be placed to be at the first instruction past a function's prologue. * Removed commands set debug aix-solib on|off show debug aix-solib set debug solib-frv on|off show debug solib-frv Removed in favor of "set/show debug solib". maintenance info program-spaces This command now includes a 'Core File' column which indicates the name of the core file associated with each program space. * New targets GNU/Linux/LoongArch (gdbserver) loongarch*-*-linux* GNU/Linux/CSKY (gdbserver) csky*-*linux* AMDGPU amdgcn-*-* * MI changes ** The async record stating the stopped reason 'breakpoint-hit' now contains an optional field locno giving the code location number when the breakpoint has multiple code locations. * Python API ** GDB will now reformat the doc string for gdb.Command and gdb.Parameter sub-classes to remove unnecessary leading whitespace from each line before using the string as the help output. ** New function gdb.format_address(ADDRESS, PROGSPACE, ARCHITECTURE), that formats ADDRESS as 'address ', where symbol is looked up in PROGSPACE, and ARCHITECTURE is used to format address. This is the same format that GDB uses when printing address, symbol, and offset information from the disassembler. ** New function gdb.current_language that returns the name of the current language. Unlike gdb.parameter('language'), this will never return 'auto'. ** New method gdb.Frame.language that returns the name of the frame's language. ** New Python API for wrapping GDB's disassembler: - gdb.disassembler.register_disassembler(DISASSEMBLER, ARCH). DISASSEMBLER is a sub-class of gdb.disassembler.Disassembler. ARCH is either None or a string containing a bfd architecture name. DISASSEMBLER is registered as a disassembler for architecture ARCH, or for all architectures if ARCH is None. The previous disassembler registered for ARCH is returned, this can be None if no previous disassembler was registered. - gdb.disassembler.Disassembler is the class from which all disassemblers should inherit. Its constructor takes a string, a name for the disassembler, which is currently only used in some debug output. Sub-classes should override the __call__ method to perform disassembly, invoking __call__ on this base class will raise an exception. - gdb.disassembler.DisassembleInfo is the class used to describe a single disassembly request from GDB. An instance of this class is passed to the __call__ method of gdb.disassembler.Disassembler and has the following read-only attributes: 'address', and 'architecture', as well as the following method: 'read_memory'. - gdb.disassembler.builtin_disassemble(INFO, MEMORY_SOURCE), calls GDB's builtin disassembler on INFO, which is a gdb.disassembler.DisassembleInfo object. MEMORY_SOURCE is optional, its default value is None. If MEMORY_SOURCE is not None then it must be an object that has a 'read_memory' method. - gdb.disassembler.DisassemblerResult is a class that can be used to wrap the result of a call to a Disassembler. It has read-only attributes 'length' and 'string'. ** gdb.Objfile now has an attribute named "is_file". This is True if the objfile comes from a file, and False otherwise. ** New function gdb.print_options that returns a dictionary of the prevailing print options, in the form accepted by gdb.Value.format_string. ** gdb.Value.format_string now uses the format provided by 'print', if it is called during a 'print' or other similar operation. ** gdb.Value.format_string now accepts the 'summary' keyword. This can be used to request a shorter representation of a value, the way that 'set print frame-arguments scalars' does. ** New Python type gdb.BreakpointLocation. The new attribute 'locations' of gdb.Breakpoint returns a list of gdb.BreakpointLocation objects specifying the locations where the breakpoint is inserted into the debuggee. ** The gdb.register_window_type method now restricts the set of acceptable window names. The first character of a window's name must start with a character in the set [a-zA-Z], every subsequent character of a window's name must be in the set [-_.a-zA-Z0-9]. * New features in the GDB remote stub, GDBserver ** GDBserver is now supported on LoongArch GNU/Linux. ** GDBserver is now supported on CSKY GNU/Linux. * LoongArch floating-point support GDB now supports floating-point on LoongArch GNU/Linux. * AMD GPU ROCm debugging support GDB now supports debugging programs offloaded to AMD GPUs using the ROCm platform. *** Changes in GDB 12 * DBX mode is deprecated, and will be removed in GDB 13 * GDB 12 is the last release of GDB that will support building against Python 2. From GDB 13, it will only be possible to build GDB itself with Python 3 support. * The disable-randomization setting now works on Windows. * Improved C++ template support GDB now treats functions/types involving C++ templates like it does function overloads. Users may omit parameter lists to set breakpoints on families of template functions, including types/functions composed of multiple template types: (gdb) break template_func(template_1, int) The above will set breakpoints at every function `template_func' where the first function parameter is any template type named `template_1' and the second function parameter is `int'. TAB completion also gains similar improvements. * The FreeBSD native target now supports async mode. * Configure changes --enable-threading Enable or disable multithreaded symbol loading. This is enabled by default, but passing --disable-threading or --enable-threading=no to configure will disable it. Disabling this can cause a performance penalty when there are a lot of symbols to load, but is useful for debugging purposes. * New commands maint set backtrace-on-fatal-signal on|off maint show backtrace-on-fatal-signal This setting is 'on' by default. When 'on' GDB will print a limited backtrace to stderr in the situation where GDB terminates with a fatal signal. This only supported on some platforms where the backtrace and backtrace_symbols_fd functions are available. set source open on|off show source open This setting, which is on by default, controls whether GDB will try to open source code files. Switching this off will stop GDB trying to open and read source code files, which can be useful if the files are located over a slow network connection. set varsize-limit show varsize-limit These are now deprecated aliases for "set max-value-size" and "show max-value-size". task apply [all | TASK-IDS...] [FLAG]... COMMAND Like "thread apply", but applies COMMAND to Ada tasks. watch [...] task ID Watchpoints can now be restricted to a specific Ada task. maint set internal-error backtrace on|off maint show internal-error backtrace maint set internal-warning backtrace on|off maint show internal-warning backtrace GDB can now print a backtrace of itself when it encounters either an internal-error, or an internal-warning. This is on by default for internal-error and off by default for internal-warning. set logging on|off Deprecated and replaced by "set logging enabled on|off". set logging enabled on|off show logging enabled These commands set or show whether logging is enabled or disabled. exit You can now exit GDB by using the new command "exit", in addition to the existing "quit" command. set debug threads on|off show debug threads Print additional debug messages about thread creation and deletion. set debug linux-nat on|off show debug linux-nat These new commands replaced the old 'set debug lin-lwp' and 'show debug lin-lwp' respectively. Turning this setting on prints debug messages relating to GDB's handling of native Linux inferiors. maint flush source-cache Flush the contents of the source code cache. maint set gnu-source-highlight enabled on|off maint show gnu-source-highlight enabled Whether GDB should use the GNU Source Highlight library for adding styling to source code. When off, the library will not be used, even when available. When GNU Source Highlight isn't used, or can't add styling to a particular source file, then the Python Pygments library will be used instead. set suppress-cli-notifications (on|off) show suppress-cli-notifications This controls whether printing the notifications is suppressed for CLI. CLI notifications occur when you change the selected context (i.e., the current inferior, thread and/or the frame), or when the program being debugged stops (e.g., because of hitting a breakpoint, completing source-stepping, an interrupt, etc.). set style disassembler enabled on|off show style disassembler enabled If GDB is compiled with Python support, and the Python Pygments package is available, then, when this setting is on, disassembler output will have styling applied. set ada source-charset show ada source-charset Set the character set encoding that is assumed for Ada symbols. Valid values for this follow the values that can be passed to the GNAT compiler via the '-gnati' option. The default is ISO-8859-1. tui layout tui focus tui refresh tui window height These are the new names for the old 'layout', 'focus', 'refresh', and 'winheight' tui commands respectively. The old names still exist as aliases to these new commands. tui window width winwidth The new command 'tui window width', and the alias 'winwidth' allow the width of a tui window to be adjusted when windows are laid out in horizontal mode. set debug tui on|off show debug tui Control the display of debug output about GDB's tui. * Changed commands print Printing of floating-point values with base-modifying formats like /x has been changed to display the underlying bytes of the value in the desired base. This was GDB's documented behavior, but was never implemented correctly. maint packet This command can now print a reply, if the reply includes non-printable characters. Any non-printable characters are printed as escaped hex, e.g. \x?? where '??' is replaces with the value of the non-printable character. clone-inferior The clone-inferior command now ensures that the TTY, CMD and ARGS settings are copied from the original inferior to the new one. All modifications to the environment variables done using the 'set environment' or 'unset environment' commands are also copied to the new inferior. set debug lin-lwp on|off show debug lin-lwp These commands have been removed from GDB. The new command 'set debug linux-nat' and 'show debug linux-nat' should be used instead. info win This command now includes information about the width of the tui windows in its output. layout focus refresh winheight These commands are now aliases for the 'tui layout', 'tui focus', 'tui refresh', and 'tui window height' commands respectively. * GDB's Ada parser now supports an extension for specifying the exact byte contents of a floating-point literal. This can be useful for setting floating-point registers to a precise value without loss of precision. The syntax is an extension of the based literal syntax. Use, e.g., "16lf#0123abcd#" -- the number of "l"s controls the width of the floating-point type, and the "f" is the marker for floating point. * MI changes ** The '-add-inferior' with no option flags now inherits the connection of the current inferior, this restores the behaviour of GDB as it was prior to GDB 10. ** The '-add-inferior' command now accepts a '--no-connection' option, which causes the new inferior to start without a connection. ** The default version of the MI interpreter is now 4 (-i=mi4). ** The "script" field in breakpoint output (which is syntactically incorrect in MI 3 and below) has changed in MI 4 to become a list. This affects the following commands and events: - -break-insert - -break-info - =breakpoint-created - =breakpoint-modified The -fix-breakpoint-script-output command can be used to enable this behavior with previous MI versions. * New targets GNU/Linux/LoongArch loongarch*-*-linux* * Removed targets S+core score-*-* * Python API ** New function gdb.add_history(), which takes a gdb.Value object and adds the value it represents to GDB's history list. An integer, the index of the new item in the history list, is returned. ** New function gdb.history_count(), which returns the number of values in GDB's value history. ** New gdb.events.gdb_exiting event. This event is called with a gdb.GdbExitingEvent object which has the read-only attribute 'exit_code', which contains the value of the GDB exit code. This event is triggered once GDB decides it is going to exit, but before GDB starts to clean up its internal state. ** New function gdb.architecture_names(), which returns a list containing all of the possible Architecture.name() values. Each entry is a string. ** New function gdb.Architecture.integer_type(), which returns an integer type given a size and a signed-ness. ** New gdb.TargetConnection object type that represents a connection (as displayed by the 'info connections' command). A sub-class, gdb.RemoteTargetConnection, is used to represent 'remote' and 'extended-remote' connections. ** The gdb.Inferior type now has a 'connection' property which is an instance of gdb.TargetConnection, the connection used by this inferior. This can be None if the inferior has no connection. ** New 'gdb.events.connection_removed' event registry, which emits a 'gdb.ConnectionEvent' when a connection is removed from GDB. This event has a 'connection' property, a gdb.TargetConnection object for the connection being removed. ** New gdb.connections() function that returns a list of all currently active connections. ** New gdb.RemoteTargetConnection.send_packet(PACKET) method. This is equivalent to the existing 'maint packet' CLI command; it allows a user specified packet to be sent to the remote target. ** New function gdb.host_charset(), returns a string, which is the name of the current host charset. ** New gdb.set_parameter(NAME, VALUE). This sets the gdb parameter NAME to VALUE. ** New gdb.with_parameter(NAME, VALUE). This returns a context manager that temporarily sets the gdb parameter NAME to VALUE, then resets it when the context is exited. ** The gdb.Value.format_string method now takes a 'styling' argument, which is a boolean. When true, the returned string can include escape sequences to apply styling. The styling will only be present if styling is otherwise turned on in GDB (see 'help set styling'). When false, which is the default if the argument is not given, then no styling is applied to the returned string. ** New read-only attribute gdb.InferiorThread.details, which is either a string, containing additional, target specific thread state information, or None, if there is no such additional information. ** New read-only attribute gdb.Type.is_scalar, which is True for scalar types, and False for all other types. ** New read-only attribute gdb.Type.is_signed. This attribute should only be read when Type.is_scalar is True, and will be True for signed types, and False for all other types. Attempting to read this attribute for non-scalar types will raise a ValueError. ** It is now possible to add GDB/MI commands implemented in Python. * New features in the GDB remote stub, GDBserver ** GDBserver is now supported on OpenRISC GNU/Linux. * New native configurations GNU/Linux/OpenRISC or1k*-*-linux* *** Changes in GDB 11 * The 'set disassembler-options' command now supports specifying options for the ARC target. * GDB now supports general memory tagging functionality if the underlying architecture supports the proper primitives and hooks. Currently this is enabled only for AArch64 MTE. This includes: - Additional information when the inferior crashes with a SIGSEGV caused by a memory tag violation. - A new modifier 'm' for the "x" command, which displays allocation tags for a particular memory range. - Display of memory tag mismatches by "print", for addresses and pointers, if memory tagging is supported by the architecture. * Building GDB now requires GMP (The GNU Multiple Precision Arithmetic Library). * MI changes ** '-break-insert --qualified' and '-dprintf-insert --qualified' The MI -break-insert and -dprintf-insert commands now support a new "--qualified" option that makes GDB interpret a specified function name as a complete fully-qualified name. This is the equivalent of the CLI's "break -qualified" and "dprintf -qualified". ** '-break-insert --force-condition' and '-dprintf-insert --force-condition' The MI -break-insert and -dprintf-insert commands now support a '--force-condition' flag to forcibly define a condition even when the condition is invalid at all locations of the breakpoint. This is equivalent to the '-force-condition' flag of the CLI's "break" command. ** '-break-condition --force' The MI -break-condition command now supports a '--force' flag to forcibly define a condition even when the condition is invalid at all locations of the selected breakpoint. This is equivalent to the '-force' flag of the CLI's "cond" command. ** '-file-list-exec-source-files [--group-by-objfile] [--basename | --dirname] [--] [REGEXP]' The existing -file-list-exec-source-files command now takes an optional REGEXP which is used to filter the source files that are included in the results. By default REGEXP is matched against the full filename of the source file. When one of --basename or --dirname is given then REGEXP is only matched against the specified part of the full source filename. When the optional --group-by-objfile flag is used the output format is changed, the results are now a list of object files (executable and libraries) with the source files that are associated with each object file. The results from -file-list-exec-source-files now include a 'debug-fully-read' field which takes the value 'true' or 'false'. A 'true' value indicates the source file is from a compilation unit that has had its debug information fully read in by GDB, a value of 'false' indicates GDB has only performed a partial scan of the debug information so far. * GDB now supports core file debugging for x86_64 Cygwin programs. * GDB will now look for the .gdbinit file in a config directory before looking for ~/.gdbinit. The file is searched for in the following locations: $XDG_CONFIG_HOME/gdb/gdbinit, $HOME/.config/gdb/gdbinit, $HOME/.gdbinit. On Apple hosts the search order is instead: $HOME/Library/Preferences/gdb/gdbinit, $HOME/.gdbinit. * GDB now supports fixed point types which are described in DWARF as base types with a fixed-point encoding. Additionally, support for the DW_AT_GNU_numerator and DW_AT_GNU_denominator has also been added. For Ada, this allows support for fixed point types without requiring the use of the GNAT encoding (based on information added to the type's name following a GNAT-specific format). * GDB will now load and process commands from ~/.config/gdb/gdbearlyinit or ~/.gdbearlyinit if these files are present. These files are processed earlier than any of the other initialization files and can affect parts of GDB's startup that previously had already been completed before the initialization files were read, for example styling of the initial GDB greeting. * GDB now has two new options "--early-init-command" and "--early-init-eval-command" with corresponding short options "-eix" and "-eiex" that allow options (that would normally appear in a gdbearlyinit file) to be passed on the command line. * For RISC-V targets, the target feature "org.gnu.gdb.riscv.vector" is now understood by GDB, and can be used to describe the vector registers of a target. The precise requirements of this register feature are documented in the GDB manual. * For ARM targets, the "org.gnu.gdb.arm.m-profile-mve" feature is now supported by GDB and describes a new VPR register from the ARM MVE (Helium) extension. See the GDB manual for more information. * TUI improvements ** TUI windows now support mouse actions. The mouse wheel scrolls the appropriate window. ** Key combinations that do not have a specific action on the focused window are passed to GDB. For example, you now can use Ctrl-Left/Ctrl-Right to move between words in the command window regardless of which window is in focus. Previously you would need to focus on the command window for such key combinations to work. * New commands set debug event-loop show debug event-loop Control the display of debug output about GDB's event loop. set print memory-tag-violations show print memory-tag-violations Control whether to display additional information about memory tag violations when printing pointers and addresses. Architecture support for memory tagging is required for this option to have an effect. maintenance flush symbol-cache maintenance flush register-cache These new commands are equivalent to the already existing commands 'maintenance flush-symbol-cache' and 'flushregs' respectively. maintenance flush dcache A new command to flush the dcache. maintenance info target-sections Print GDB's internal target sections table. maintenance info jit Print the JIT code objects in the inferior known to GDB. memory-tag show-logical-tag POINTER Print the logical tag for POINTER. memory-tag with-logical-tag POINTER TAG Print POINTER with logical tag TAG. memory-tag show-allocation-tag ADDRESS Print the allocation tag for ADDRESS. memory-tag set-allocation-tag ADDRESS LENGTH TAGS Set the allocation tag for [ADDRESS, ADDRESS + LENGTH) to TAGS. memory-tag check POINTER Validate that POINTER's logical tag matches the allocation tag. set startup-quietly on|off show startup-quietly When 'on', this causes GDB to act as if "-silent" were passed on the command line. This command needs to be added to an early initialization file (e.g. ~/.config/gdb/gdbearlyinit) in order to affect GDB. set print type hex on|off show print type hex When 'on', the 'ptype' command uses hexadecimal notation to print sizes and offsets of struct members. When 'off', decimal notation is used. set python ignore-environment on|off show python ignore-environment When 'on', this causes GDB's builtin Python to ignore any environment variables that would otherwise affect how Python behaves. This command needs to be added to an early initialization file (e.g. ~/.config/gdb/gdbearlyinit) in order to affect GDB. set python dont-write-bytecode auto|on|off show python dont-write-bytecode When 'on', this causes GDB's builtin Python to not write any byte-code (.pyc files) to disk. This command needs to be added to an early initialization file (e.g. ~/.config/gdb/gdbearlyinit) in order to affect GDB. When 'off' byte-code will always be written. When set to 'auto' (the default) Python will check the PYTHONDONTWRITEBYTECODE environment variable. * Changed commands break [PROBE_MODIFIER] [LOCATION] [thread THREADNUM] [-force-condition] [if CONDITION] This command would previously refuse setting a breakpoint if the CONDITION expression is invalid at a location. It now accepts and defines the breakpoint if there is at least one location at which the CONDITION is valid. The locations for which the CONDITION is invalid, are automatically disabled. If CONDITION is invalid at all of the locations, setting the breakpoint is still rejected. However, the '-force-condition' flag can be used in this case for forcing GDB to define the breakpoint, making all the current locations automatically disabled. This may be useful if the user knows the condition will become meaningful at a future location, e.g. due to a shared library load. condition [-force] N COND The behavior of this command is changed the same way for the 'break' command as explained above. The '-force' flag can be used to force GDB into defining the condition even when COND is invalid for all the current locations of breakpoint N. flushregs maintenance flush-symbol-cache These commands are deprecated in favor of the new commands 'maintenance flush register-cache' and 'maintenance flush symbol-cache' respectively. set style version foreground COLOR set style version background COLOR set style version intensity VALUE Control the styling of GDB's version number text. inferior [ID] When the ID parameter is omitted, then this command prints information about the current inferior. When the ID parameter is present, the behavior of the command is unchanged and have the inferior ID become the current inferior. maintenance info sections The ALLOBJ keyword has been replaced with an -all-objects command line flag. It is now possible to filter which sections are printed even when -all-objects is passed. ptype[/FLAGS] TYPE | EXPRESSION The 'ptype' command has two new flags. When '/x' is set, hexadecimal notation is used when printing sizes and offsets of struct members. When '/d' is set, decimal notation is used when printing sizes and offsets of struct members. Default behavior is given by 'show print type hex'. info sources The info sources command output has been restructured. The results are now based around a list of objfiles (executable and libraries), and for each objfile the source files that are part of that objfile are listed. * Removed targets and native configurations ARM Symbian arm*-*-symbianelf* * New remote packets qMemTags Request the remote to send allocation tags for a particular memory range. QMemTags Request the remote to store the specified allocation tags to the requested memory range. * Guile API ** Improved support for rvalue reference values: TYPE_CODE_RVALUE_REF is now exported as part of the API and the value-referenced-value procedure now handles rvalue reference values. ** New procedures for obtaining value variants: value-reference-value, value-rvalue-reference-value and value-const-value. ** Temporary breakpoints can now be created with make-breakpoint and tested for using breakpoint-temporary?. * Python API ** Inferior objects now contain a read-only 'connection_num' attribute that gives the connection number as seen in 'info connections' and 'info inferiors'. ** New method gdb.Frame.level() which returns the stack level of the frame object. ** New method gdb.PendingFrame.level() which returns the stack level of the frame object. ** When hitting a catchpoint, the Python API will now emit a gdb.BreakpointEvent rather than a gdb.StopEvent. The gdb.Breakpoint attached to the event will have type BP_CATCHPOINT. ** Python TUI windows can now receive mouse click events. If the Window object implements the click method, it is called for each mouse click event in this window. *** Changes in GDB 10 * There are new feature names for ARC targets: "org.gnu.gdb.arc.core" and "org.gnu.gdb.arc.aux". The old names are still supported but must be considered obsolete. They will be deprecated after some grace period. * Help and apropos commands will now show the documentation of a command only once, even if that command has one or more aliases. These commands now show the command name, then all of its aliases, and finally the description of the command. * 'help aliases' now shows only the user defined aliases. GDB predefined aliases are shown together with their aliased command. * GDB now supports debuginfod, an HTTP server for distributing ELF/DWARF debugging information as well as source code. When built with debuginfod, GDB can automatically query debuginfod servers for the separate debug files and source code of the executable being debugged. To build GDB with debuginfod, pass --with-debuginfod to configure (this requires libdebuginfod, the debuginfod client library). debuginfod is distributed with elfutils, starting with version 0.178. You can get the latest version from https://sourceware.org/elfutils. * Multi-target debugging support GDB now supports debugging multiple target connections simultaneously. For example, you can now have each inferior connected to different remote servers running in different machines, or have one inferior debugging a local native process, an inferior debugging a core dump, etc. This support is experimental and comes with some limitations -- you can only resume multiple targets simultaneously if all targets support non-stop mode, and all remote stubs or servers must support the same set of remote protocol features exactly. See also "info connections" and "add-inferior -no-connection" below, and "maint set target-non-stop" in the user manual. * New features in the GDB remote stub, GDBserver ** GDBserver is now supported on ARC GNU/Linux. ** GDBserver is now supported on RISC-V GNU/Linux. ** GDBserver no longer supports these host triplets: i[34567]86-*-lynxos* powerpc-*-lynxos* i[34567]86-*-nto* bfin-*-*linux* crisv32-*-linux* cris-*-linux* m32r*-*-linux* tilegx-*-linux* arm*-*-mingw32ce* i[34567]86-*-mingw32ce* * Debugging MS-Windows processes now sets $_exitsignal when the inferior is terminated by a signal, instead of setting $_exitcode. * Multithreaded symbol loading has now been enabled by default on systems that support it (see entry for GDB 9, below), providing faster performance for programs with many symbols. * The $_siginfo convenience variable now also works on Windows targets, and will display the EXCEPTION_RECORD of the last handled exception. * TUI windows can now be arranged horizontally. * The command history filename can now be set to the empty string either using 'set history filename' or by setting 'GDBHISTFILE=' in the environment. The effect of setting this filename to the empty string is that GDB will not try to load any previous command history. * On Windows targets, it is now possible to debug 32-bit programs with a 64-bit GDB. * New commands set exec-file-mismatch -- Set exec-file-mismatch handling (ask|warn|off). show exec-file-mismatch -- Show exec-file-mismatch handling (ask|warn|off). Set or show the option 'exec-file-mismatch'. When GDB attaches to a running process, this new option indicates whether to detect a mismatch between the current executable file loaded by GDB and the executable file used to start the process. If 'ask', the default, display a warning and ask the user whether to load the process executable file; if 'warn', just display a warning; if 'off', don't attempt to detect a mismatch. tui new-layout NAME WINDOW WEIGHT [WINDOW WEIGHT]... Define a new TUI layout, specifying its name and the windows that will be displayed. maintenance print xml-tdesc [FILE] Prints the current target description as an XML document. If the optional FILE is provided (which is an XML target description) then the target description is read from FILE into GDB, and then reprinted. maintenance print core-file-backed-mappings Prints file-backed mappings loaded from a core file's note section. Output is expected to be similar to that of "info proc mappings". set debug fortran-array-slicing on|off show debug fortran-array-slicing Print debugging when taking slices of Fortran arrays. set fortran repack-array-slices on|off show fortran repack-array-slices When taking slices from Fortran arrays and strings, if the slice is non-contiguous within the original value then, when this option is on, the new value will be repacked into a single contiguous value. When this option is off, then the value returned will consist of a descriptor that describes the slice within the memory of the original parent value. * Changed commands alias [-a] [--] ALIAS = COMMAND [DEFAULT-ARGS...] The alias command can now specify default args for an alias. GDB automatically prepends the alias default args to the argument list provided explicitly by the user. For example, to have a backtrace with full details, you can define an alias 'bt_ALL' as 'alias bt_ALL = backtrace -entry-values both -frame-arg all -past-main -past-entry -full'. Alias default arguments can also use a set of nested 'with' commands, e.g. 'alias pp10 = with print pretty -- with print elem 10 -- print' defines the alias pp10 that will pretty print a maximum of 10 elements of the given expression (if the expression is an array). * New targets GNU/Linux/RISC-V (gdbserver) riscv*-*-linux* BPF bpf-unknown-none Z80 z80-unknown-* * Python API ** gdb.register_window_type can be used to implement new TUI windows in Python. ** Dynamic types can now be queried. gdb.Type has a new attribute, "dynamic", and gdb.Type.sizeof can be None for a dynamic type. A field of a dynamic type may have None for its "bitpos" attribute as well. ** Commands written in Python can be in the "TUI" help class by registering with the new constant gdb.COMMAND_TUI. ** New method gdb.PendingFrame.architecture () to retrieve the architecture of the pending frame. ** New gdb.Architecture.registers method that returns a gdb.RegisterDescriptorIterator object, an iterator that returns gdb.RegisterDescriptor objects. The new RegisterDescriptor is a way to query the registers available for an architecture. ** New gdb.Architecture.register_groups method that returns a gdb.RegisterGroupIterator object, an iterator that returns gdb.RegisterGroup objects. The new RegisterGroup is a way to discover the available register groups. * Guile API ** GDB can now be built with GNU Guile 3.0 and 2.2 in addition to 2.0. ** Procedures 'memory-port-read-buffer-size', 'set-memory-port-read-buffer-size!', 'memory-port-write-buffer-size', and 'set-memory-port-write-buffer-size!' are deprecated. When using Guile 2.2 and later, users who need to control the size of a memory port's internal buffer can use the 'setvbuf' procedure. *** Changes in GDB 9 * 'thread-exited' event is now available in the annotations interface. * New built-in convenience variables $_gdb_major and $_gdb_minor provide the GDB version. They are handy for conditionally using features available only in or since specific GDB versions, in scripts that should work error-free with many different versions, such as in system-wide init files. * New built-in convenience functions $_gdb_setting, $_gdb_setting_str, $_gdb_maint_setting and $_gdb_maint_setting_str provide access to values of the GDB settings and the GDB maintenance settings. They are handy for changing the logic of user defined commands depending on the current GDB settings. * GDB now supports Thread Local Storage (TLS) variables on several FreeBSD architectures (amd64, i386, powerpc, riscv). Other architectures require kernel changes. TLS is not yet supported for amd64 and i386 process core dumps. * Support for Pointer Authentication (PAC) on AArch64 Linux. Return addresses that required unmasking are shown in the backtrace with the postfix [PAC]. * Two new convenience functions $_cimag and $_creal that extract the imaginary and real parts respectively from complex numbers. * New built-in convenience variables $_shell_exitcode and $_shell_exitsignal provide the exitcode or exit status of the shell commands launched by GDB commands such as "shell", "pipe" and "make". * The command define-prefix can now define user defined prefix commands. User defined commands can now be defined using these user defined prefix commands. * Command names can now use the . character. * The RX port now supports XML target descriptions. * GDB now shows the Ada task names at more places, e.g. in task switching messages. * GDB can now be compiled with Python 3 on Windows. * New convenience variable $_ada_exception holds the address of the Ada exception being thrown. This is set by Ada-related catchpoints. * GDB can now place breakpoints on nested functions and subroutines in Fortran code. The '::' operator can be used between parent and child scopes when placing breakpoints, for example: (gdb) break outer_function::inner_function The 'outer_function::' prefix is only needed if 'inner_function' is not visible in the current scope. * In addition to the system-wide gdbinit file, if configured with --with-system-gdbinit-dir, GDB will now also load files in that directory as system gdbinit files, unless the -nx or -n flag is provided. Files with extensions .gdb, .py and .scm are supported as long as GDB was compiled with support for that language. * GDB now supports multithreaded symbol loading for higher performance. This feature is still in testing, so it is disabled by default. You can turn it on using 'maint set worker-threads unlimited'. * Python API ** The gdb.Value type has a new method 'format_string' which returns a string representing the value. The formatting is controlled by the optional keyword arguments: 'raw', 'pretty_arrays', 'pretty_structs', 'array_indexes', 'symbols', 'unions', 'deref_refs', 'actual_objects', 'static_members', 'max_elements', 'repeat_threshold', and 'format'. ** gdb.Type has a new property 'objfile' which returns the objfile the type was defined in. ** The frame information printed by the python frame filtering code is now consistent with what the 'backtrace' command prints when there are no filters, or when the 'backtrace' '-no-filters' option is given. ** The new function gdb.lookup_static_symbol can be used to look up symbols with static linkage. ** The new function gdb.lookup_static_symbols can be used to look up all static symbols with static linkage. ** gdb.Objfile has new methods 'lookup_global_symbol' and 'lookup_static_symbol' to lookup a symbol from this objfile only. ** gdb.Block now supports the dictionary syntax for accessing symbols in this block (e.g. block['local_variable']). * New commands | [COMMAND] | SHELL_COMMAND | -d DELIM COMMAND DELIM SHELL_COMMAND pipe [COMMAND] | SHELL_COMMAND pipe -d DELIM COMMAND DELIM SHELL_COMMAND Executes COMMAND and sends its output to SHELL_COMMAND. With no COMMAND, repeat the last executed command and send its output to SHELL_COMMAND. define-prefix COMMAND Define or mark a command as a user-defined prefix command. with SETTING [VALUE] [-- COMMAND] w SETTING [VALUE] [-- COMMAND] Temporarily set SETTING, run COMMAND, and restore SETTING. Usage: with SETTING -- COMMAND With no COMMAND, repeats the last executed command. SETTING is any GDB setting you can change with the "set" subcommands. For example, 'with language c -- print someobj' temporarily switches to the C language in order to print someobj. Settings can be combined: 'w lang c -- w print elements unlimited -- usercmd' switches to the C language and runs usercmd with no limit of array elements to print. maint with SETTING [VALUE] [-- COMMAND] Like "with", but works with "maintenance set" settings. set may-call-functions [on|off] show may-call-functions This controls whether GDB will attempt to call functions in the program, such as with expressions in the print command. It defaults to on. Calling functions in the program being debugged can have undesired side effects. It is now possible to forbid such function calls. If function calls are forbidden, GDB will throw an error when a command (such as print expression) calls a function in the program. set print finish [on|off] show print finish This controls whether the `finish' command will display the value that is returned by the current function. When `off', the value is still entered into the value history, but it is not printed. The default is `on'. set print max-depth show print max-depth Allows deeply nested structures to be simplified when printing by replacing deeply nested parts (beyond the max-depth) with ellipses. The default max-depth is 20, but this can be set to unlimited to get the old behavior back. set print raw-values [on|off] show print raw-values By default, GDB applies the enabled pretty printers when printing a value. This allows to ignore the enabled pretty printers for a series of commands. The default is 'off'. set logging debugredirect [on|off] By default, GDB debug output will go to both the terminal and the logfile. Set if you want debug output to go only to the log file. set style title foreground COLOR set style title background COLOR set style title intensity VALUE Control the styling of titles. set style highlight foreground COLOR set style highlight background COLOR set style highlight intensity VALUE Control the styling of highlightings. maint set worker-threads maint show worker-threads Control the number of worker threads that can be used by GDB. The default is 0. "unlimited" lets GDB choose a number that is reasonable. Currently worker threads are only used when demangling the names of linker symbols. set style tui-border foreground COLOR set style tui-border background COLOR Control the styling of TUI borders. set style tui-active-border foreground COLOR set style tui-active-border background COLOR Control the styling of the active TUI border. maint set test-settings KIND maint show test-settings KIND A set of commands used by the testsuite for exercising the settings infrastructure. maint set tui-resize-message [on|off] maint show tui-resize-message Control whether GDB prints a message each time the terminal is resized when in TUI mode. This is primarily useful for testing the TUI. set print frame-info [short-location|location|location-and-address |source-and-location|source-line|auto] show print frame-info This controls what frame information is printed by the commands printing a frame. This setting will e.g. influence the behaviour of 'backtrace', 'frame', 'stepi'. The python frame filtering also respect this setting. The 'backtrace' '-frame-info' option can override this global setting. set tui compact-source show tui compact-source Enable the "compact" display mode for the TUI source window. The compact display uses only as much space as is needed for the line numbers in the current file, and only a single space to separate the line numbers from the source. info modules [-q] [REGEXP] Return a list of Fortran modules matching REGEXP, or all modules if no REGEXP is given. info module functions [-q] [-m MODULE_REGEXP] [-t TYPE_REGEXP] [REGEXP] Return a list of functions within all modules, grouped by module. The list of functions can be restricted with the optional regular expressions. MODULE_REGEXP matches against the module name, TYPE_REGEXP matches against the function type signature, and REGEXP matches against the function name. info module variables [-q] [-m MODULE_REGEXP] [-t TYPE_REGEXP] [REGEXP] Return a list of variables within all modules, grouped by module. The list of variables can be restricted with the optional regular expressions. MODULE_REGEXP matches against the module name, TYPE_REGEXP matches against the variable type, and REGEXP matches against the variable name. set debug remote-packet-max-chars show debug remote-packet-max-chars Controls the number of characters to output in a remote packet when using "set debug remote". The default is 512 bytes. info connections Lists the target connections currently in use. * Changed commands help The "help" command uses the title style to enhance the readibility of its output by styling the classes and command names. apropos [-v] REGEXP Similarly to "help", the "apropos" command also uses the title style for the command names. "apropos" accepts now a flag "-v" (verbose) to show the full documentation of matching commands and to use the highlight style to mark the documentation parts matching REGEXP. printf eval The GDB printf and eval commands can now print C-style and Ada-style string convenience variables without calling functions in the program. This allows to do formatted printing of strings without having a running inferior, or when debugging a core dump. info sources [-dirname | -basename] [--] [REGEXP] This command has now optional arguments to only print the files whose names match REGEXP. The arguments -dirname and -basename allow to restrict matching respectively to the dirname and basename parts of the files. show style The "show style" and its subcommands are now styling a style name in their output using its own style, to help the user visualize the different styles. set print frame-arguments The new value 'presence' indicates to only indicate the presence of arguments using ..., instead of printing argument names and values. set print raw-frame-arguments show print raw-frame-arguments These commands replace the similarly-named "set/show print raw frame-arguments" commands (now with a dash instead of a space). The old commands are now deprecated and may be removed in a future release. add-inferior [-no-connection] The add-inferior command now supports a "-no-connection" flag that makes the new inferior start with no target connection associated. By default, the new inferior inherits the target connection of the current inferior. See also "info connections". info inferior This command's output now includes a new "Connection" column indicating which target connection an inferior is bound to. See "info connections" above. maint test-options require-delimiter maint test-options unknown-is-error maint test-options unknown-is-operand maint show test-options-completion-result Commands used by the testsuite to validate the command options framework. focus, winheight, +, -, >, < These commands are now case-sensitive. * New command options, command completion GDB now has a standard infrastructure to support dash-style command options ('-OPT'). One benefit is that commands that use it can easily support completion of command line arguments. Try "CMD -[TAB]" or "help CMD" to find options supported by a command. Over time, we intend to migrate most commands to this infrastructure. A number of commands got support for new command options in this release: ** The "print" and "compile print" commands now support a number of options that allow overriding relevant global print settings as set by "set print" subcommands: -address [on|off] -array [on|off] -array-indexes [on|off] -elements NUMBER|unlimited -null-stop [on|off] -object [on|off] -pretty [on|off] -raw-values [on|off] -repeats NUMBER|unlimited -static-members [on|off] -symbol [on|off] -union [on|off] -vtbl [on|off] Note that because the "print"/"compile print" commands accept arbitrary expressions which may look like options (including abbreviations), if you specify any command option, then you must use a double dash ("--") to mark the end of argument processing. ** The "backtrace" command now supports a number of options that allow overriding relevant global print settings as set by "set backtrace" and "set print" subcommands: -entry-values no|only|preferred|if-needed|both|compact|default -frame-arguments all|scalars|none -raw-frame-arguments [on|off] -frame-info auto|source-line|location|source-and-location |location-and-address|short-location -past-main [on|off] -past-entry [on|off] In addition, the full/no-filters/hide qualifiers are now also exposed as command options too: -full -no-filters -hide ** The "frame apply", "tfaas" and "faas" commands similarly now support the following options: -past-main [on|off] -past-entry [on|off] ** The new "info sources" options -dirname and -basename options are using the standard '-OPT' infrastructure. All options above can also be abbreviated. The argument of boolean (on/off) options can be 0/1 too, and also the argument is assumed "on" if omitted. This allows writing compact command invocations, like for example: (gdb) p -ra -p -o 0 -- *myptr The above is equivalent to: (gdb) print -raw-values -pretty -object off -- *myptr ** The "info types" command now supports the '-q' flag to disable printing of some header information in a similar fashion to "info variables" and "info functions". ** The "info variables", "info functions", and "whereis" commands now take a '-n' flag that excludes non-debug symbols (symbols from the symbol table, not from the debug info such as DWARF) from the results. * Completion improvements ** GDB can now complete the options of the "thread apply all" and "taas" commands, and their "-ascending" option can now be abbreviated. ** GDB can now complete the options of the "info threads", "info functions", "info variables", "info locals", and "info args" commands. ** GDB can now complete the options of the "compile file" and "compile code" commands. The "compile file" command now completes on filenames. ** GDB can now complete the backtrace command's "full/no-filters/hide" qualifiers. * In settings, you can now abbreviate "unlimited". E.g., "set print elements u" is now equivalent to "set print elements unlimited". * New MI commands -complete This lists all the possible completions for the rest of the line, if it were to be given as a command itself. This is intended for use by MI frontends in cases when separate CLI and MI channels cannot be used. -catch-throw, -catch-rethrow, and -catch-catch These can be used to catch C++ exceptions in a similar fashion to the CLI commands 'catch throw', 'catch rethrow', and 'catch catch'. -symbol-info-functions, -symbol-info-types, and -symbol-info-variables These commands are the MI equivalent of the CLI commands 'info functions', 'info types', and 'info variables' respectively. -symbol-info-modules, this is the MI equivalent of the CLI 'info modules' command. -symbol-info-module-functions and -symbol-info-module-variables. These commands are the MI equivalent of the CLI commands 'info module functions' and 'info module variables'. * Other MI changes ** The default version of the MI interpreter is now 3 (-i=mi3). ** The output of information about multi-location breakpoints (which is syntactically incorrect in MI 2) has changed in MI 3. This affects the following commands and events: - -break-insert - -break-info - =breakpoint-created - =breakpoint-modified The -fix-multi-location-breakpoint-output command can be used to enable this behavior with previous MI versions. ** Backtraces and frames include a new optional field addr_flags which is given after the addr field. On AArch64 this contains PAC if the address has been masked in the frame. On all other targets the field is not present. * Testsuite The testsuite now creates the files gdb.cmd (containing the arguments used to launch GDB) and gdb.in (containing all the commands sent to GDB) in the output directory for each test script. Multiple invocations are appended with .1, .2, .3 etc. * Building GDB and GDBserver now requires GNU make >= 3.82. Using another implementation of the make program or an earlier version of GNU make to build GDB or GDBserver is not supported. * Building GDB now requires GNU readline >= 7.0. GDB now bundles GNU readline 8.0, but if you choose to use --with-system-readline, only readline >= 7.0 can be used. * The TUI SingleKey keymap is now named "SingleKey". This can be used from .inputrc to bind keys in this keymap. This feature is only available when gdb is built against GNU readline 8.0 or later. * Removed targets and native configurations GDB no longer supports debugging the Cell Broadband Engine. This includes both debugging standalone Cell/B.E. SPU applications and integrated debugging of Cell/B.E. applications that use both the PPU and SPU architectures. * New Simulators TI PRU pru-*-elf * Removed targets and native configurations Solaris 10 i?86-*-solaris2.10, x86_64-*-solaris2.10, sparc*-*-solaris2.10 *** Changes in GDB 8.3 * GDB and GDBserver now support access to additional registers on PowerPC GNU/Linux targets: PPR, DSCR, TAR, EBB/PMU registers, and HTM registers. * GDB now has experimental support for the compilation and injection of C++ source code into the inferior. This beta release does not include support for several language features, such as templates, constructors, and operators. This feature requires GCC 7.1 or higher built with libcp1.so (the C++ plug-in). * GDB and GDBserver now support IPv6 connections. IPv6 addresses can be passed using the '[ADDRESS]:PORT' notation, or the regular 'ADDRESS:PORT' method. * DWARF index cache: GDB can now automatically save indices of DWARF symbols on disk to speed up further loading of the same binaries. * Ada task switching is now supported on aarch64-elf targets when debugging a program using the Ravenscar Profile. For more information, see the "Tasking Support when using the Ravenscar Profile" section in the GDB user manual. * GDB in batch mode now exits with status 1 if the last command to be executed failed. * The RISC-V target now supports target descriptions. * System call catchpoints now support system call aliases on FreeBSD. When the ABI of a system call changes in FreeBSD, this is implemented by leaving a compatibility system call using the old ABI at the existing number and allocating a new system call number for the new ABI. For example, FreeBSD 12 altered the layout of 'struct kevent' used by the 'kevent' system call. As a result, FreeBSD 12 kernels ship with both 'kevent' and 'freebsd11_kevent' system calls. The 'freebsd11_kevent' system call is assigned an alias of 'kevent' so that a system call catchpoint for the 'kevent' system call will catch invocations of both the 'kevent' and 'freebsd11_kevent' binaries. This ensures that 'kevent' system calls are caught for binaries using either the old or new ABIs. * Terminal styling is now available for the CLI and the TUI. GNU Source Highlight can additionally be used to provide styling of source code snippets. See the "set style" commands, below, for more information. * Removed support for old demangling styles arm, edg, gnu, hp and lucid. * New commands set debug compile-cplus-types show debug compile-cplus-types Control the display of debug output about type conversion in the C++ compile feature. Commands have no effect while compiling for other languages. set debug skip show debug skip Control whether debug output about files/functions skipping is displayed. frame apply [all | COUNT | -COUNT | level LEVEL...] [FLAG]... COMMAND Apply a command to some frames. FLAG arguments allow to control what output to produce and how to handle errors raised when applying COMMAND to a frame. taas COMMAND Apply a command to all threads (ignoring errors and empty output). Shortcut for 'thread apply all -s COMMAND'. faas COMMAND Apply a command to all frames (ignoring errors and empty output). Shortcut for 'frame apply all -s COMMAND'. tfaas COMMAND Apply a command to all frames of all threads (ignoring errors and empty output). Shortcut for 'thread apply all -s frame apply all -s COMMAND'. maint set dwarf unwinders (on|off) maint show dwarf unwinders Control whether DWARF unwinders can be used. info proc files Display a list of open files for a process. * Changed commands Changes to the "frame", "select-frame", and "info frame" CLI commands. These commands all now take a frame specification which is either a frame level, or one of the keywords 'level', 'address', 'function', or 'view' followed by a parameter. Selecting a frame by address, or viewing a frame outside the current backtrace now requires the use of a keyword. Selecting a frame by level is unchanged. The MI comment "-stack-select-frame" is unchanged. target remote FILENAME target extended-remote FILENAME If FILENAME is a Unix domain socket, GDB will attempt to connect to this socket instead of opening FILENAME as a character device. info args [-q] [-t TYPEREGEXP] [NAMEREGEXP] info functions [-q] [-t TYPEREGEXP] [NAMEREGEXP] info locals [-q] [-t TYPEREGEXP] [NAMEREGEXP] info variables [-q] [-t TYPEREGEXP] [NAMEREGEXP] These commands can now print only the searched entities matching the provided regexp(s), giving a condition on the entity names or entity types. The flag -q disables printing headers or informations messages. info functions info types info variables rbreak These commands now determine the syntax for the shown entities according to the language chosen by `set language'. In particular, `set language auto' means to automatically choose the language of the shown entities. thread apply [all | COUNT | -COUNT] [FLAG]... COMMAND The 'thread apply' command accepts new FLAG arguments. FLAG arguments allow to control what output to produce and how to handle errors raised when applying COMMAND to a thread. set tui tab-width NCHARS show tui tab-width NCHARS "set tui tab-width" replaces the "tabset" command, which has been deprecated. set style enabled [on|off] show style enabled Enable or disable terminal styling. Styling is enabled by default on most hosts, but disabled by default when in batch mode. set style sources [on|off] show style sources Enable or disable source code styling. Source code styling is enabled by default, but only takes effect if styling in general is enabled, and if GDB was linked with GNU Source Highlight. set style filename foreground COLOR set style filename background COLOR set style filename intensity VALUE Control the styling of file names. set style function foreground COLOR set style function background COLOR set style function intensity VALUE Control the styling of function names. set style variable foreground COLOR set style variable background COLOR set style variable intensity VALUE Control the styling of variable names. set style address foreground COLOR set style address background COLOR set style address intensity VALUE Control the styling of addresses. * MI changes ** The '-data-disassemble' MI command now accepts an '-a' option to disassemble the whole function surrounding the given program counter value or function name. Support for this feature can be verified by using the "-list-features" command, which should contain "data-disassemble-a-option". ** Command responses and notifications that include a frame now include the frame's architecture in a new "arch" attribute. * New native configurations GNU/Linux/RISC-V riscv*-*-linux* FreeBSD/riscv riscv*-*-freebsd* * New targets GNU/Linux/RISC-V riscv*-*-linux* CSKY ELF csky*-*-elf CSKY GNU/LINUX csky*-*-linux FreeBSD/riscv riscv*-*-freebsd* NXP S12Z s12z-*-elf GNU/Linux/OpenRISC or1k*-*-linux* * Removed targets GDB no longer supports native debugging on versions of MS-Windows before Windows XP. * Python API ** GDB no longer supports Python versions less than 2.6. ** The gdb.Inferior type has a new 'progspace' property, which is the program space associated to that inferior. ** The gdb.Progspace type has a new 'objfiles' method, which returns the list of objfiles associated to that program space. ** gdb.SYMBOL_LOC_COMMON_BLOCK, gdb.SYMBOL_MODULE_DOMAIN, and gdb.SYMBOL_COMMON_BLOCK_DOMAIN were added to reflect changes to the gdb core. ** gdb.SYMBOL_VARIABLES_DOMAIN, gdb.SYMBOL_FUNCTIONS_DOMAIN, and gdb.SYMBOL_TYPES_DOMAIN are now deprecated. These were never correct and did not work properly. ** The gdb.Value type has a new constructor, which is used to construct a gdb.Value from a Python buffer object and a gdb.Type. * Configure changes --enable-ubsan Enable or disable the undefined behavior sanitizer. This is disabled by default, but passing --enable-ubsan=yes or --enable-ubsan=auto to configure will enable it. Enabling this can cause a performance penalty. The undefined behavior sanitizer was first introduced in GCC 4.9. *** Changes in GDB 8.2 * The 'set disassembler-options' command now supports specifying options for the MIPS target. * The 'symbol-file' command now accepts an '-o' option to add a relative offset to all sections. * Similarly, the 'add-symbol-file' command also accepts an '-o' option to add a relative offset to all sections, but it allows to override the load address of individual sections using '-s'. * The 'add-symbol-file' command no longer requires the second argument (address of the text section). * The endianness used with the 'set endian auto' mode in the absence of an executable selected for debugging is now the last endianness chosen either by one of the 'set endian big' and 'set endian little' commands or by inferring from the last executable used, rather than the startup default. * The pager now allows a "c" response, meaning to disable the pager for the rest of the current command. * The commands 'info variables/functions/types' now show the source line numbers of symbol definitions when available. * 'info proc' now works on running processes on FreeBSD systems and core files created on FreeBSD systems. * C expressions can now use _Alignof, and C++ expressions can now use alignof. * Support for SVE on AArch64 Linux. Note that GDB does not detect changes to the vector length while the process is running. * New commands set debug fbsd-nat show debug fbsd-nat Control display of debugging info regarding the FreeBSD native target. set|show varsize-limit This new setting allows the user to control the maximum size of Ada objects being printed when those objects have a variable type, instead of that maximum size being hardcoded to 65536 bytes. set|show record btrace cpu Controls the processor to be used for enabling errata workarounds for branch trace decode. maint check libthread-db Run integrity checks on the current inferior's thread debugging library maint set check-libthread-db (on|off) maint show check-libthread-db Control whether to run integrity checks on inferior specific thread debugging libraries as they are loaded. The default is not to perform such checks. * Python API ** Type alignment is now exposed via the "align" attribute of a gdb.Type. ** The commands attached to a breakpoint can be set by assigning to the breakpoint's "commands" field. ** gdb.execute can now execute multi-line gdb commands. ** The new functions gdb.convenience_variable and gdb.set_convenience_variable can be used to get and set the value of convenience variables. ** A gdb.Parameter will no longer print the "set" help text on an ordinary "set"; instead by default a "set" will be silent unless the get_set_string method returns a non-empty string. * New targets RiscV ELF riscv*-*-elf * Removed targets and native configurations m88k running OpenBSD m88*-*-openbsd* SH-5/SH64 ELF sh64-*-elf*, SH-5/SH64 support in sh* SH-5/SH64 running GNU/Linux SH-5/SH64 support in sh*-*-linux* SH-5/SH64 running OpenBSD SH-5/SH64 support in sh*-*-openbsd* * Aarch64/Linux hardware watchpoints improvements Hardware watchpoints on unaligned addresses are now properly supported when running Linux kernel 4.10 or higher: read and access watchpoints are no longer spuriously missed, and all watchpoints lengths between 1 and 8 bytes are supported. On older kernels, watchpoints set on unaligned addresses are no longer missed, with the tradeoff that there is a possibility of false hits being reported. * Configure changes --enable-codesign=CERT This can be used to invoke "codesign -s CERT" after building gdb. This option is useful on macOS, where code signing is required for gdb to work properly. --disable-gdbcli has been removed This is now silently accepted, but does nothing. *** Changes in GDB 8.1 * GDB now supports dynamically creating arbitrary register groups specified in XML target descriptions. This allows for finer grain grouping of registers on systems with a large amount of registers. * The 'ptype' command now accepts a '/o' flag, which prints the offsets and sizes of fields in a struct, like the pahole(1) tool. * New "--readnever" command line option instructs GDB to not read each symbol file's symbolic debug information. This makes startup faster but at the expense of not being able to perform symbolic debugging. This option is intended for use cases where symbolic debugging will not be used, e.g., when you only need to dump the debuggee's core. * GDB now uses the GNU MPFR library, if available, to emulate target floating-point arithmetic during expression evaluation when the target uses different floating-point formats than the host. At least version 3.1 of GNU MPFR is required. * GDB now supports access to the guarded-storage-control registers and the software-based guarded-storage broadcast control registers on IBM z14. * On Unix systems, GDB now supports transmitting environment variables that are to be set or unset to GDBserver. These variables will affect the environment to be passed to the remote inferior. To inform GDB of environment variables that are to be transmitted to GDBserver, use the "set environment" command. Only user set environment variables are sent to GDBserver. To inform GDB of environment variables that are to be unset before the remote inferior is started by the GDBserver, use the "unset environment" command. * Completion improvements ** GDB can now complete function parameters in linespecs and explicit locations without quoting. When setting breakpoints, quoting around functions names to help with TAB-completion is generally no longer necessary. For example, this now completes correctly: (gdb) b function(in[TAB] (gdb) b function(int) Related, GDB is no longer confused with completing functions in C++ anonymous namespaces: (gdb) b (anon[TAB] (gdb) b (anonymous namespace)::[TAB][TAB] (anonymous namespace)::a_function() (anonymous namespace)::b_function() ** GDB now has much improved linespec and explicit locations TAB completion support, that better understands what you're completing and offers better suggestions. For example, GDB no longer offers data symbols as possible completions when you're setting a breakpoint. ** GDB now TAB-completes label symbol names. ** The "complete" command now mimics TAB completion accurately. * New command line options (gcore) -a Dump all memory mappings. * Breakpoints on C++ functions are now set on all scopes by default By default, breakpoints on functions/methods are now interpreted as specifying all functions with the given name ignoring missing leading scopes (namespaces and classes). For example, assuming a C++ program with symbols named: A::B::func() B::func() both commands "break func()" and "break B::func()" set a breakpoint on both symbols. You can use the new flag "-qualified" to override this. This makes GDB interpret the specified function name as a complete fully-qualified name instead. For example, using the same C++ program, the "break -q B::func" command sets a breakpoint on "B::func", only. A parameter has been added to the Python gdb.Breakpoint constructor to achieve the same result when creating a breakpoint from Python. * Breakpoints on functions marked with C++ ABI tags GDB can now set breakpoints on functions marked with C++ ABI tags (e.g., [abi:cxx11]). See here for a description of ABI tags: https://developers.redhat.com/blog/2015/02/05/gcc5-and-the-c11-abi/ Functions with a C++11 abi tag are demangled/displayed like this: function[abi:cxx11](int) ^^^^^^^^^^^ You can now set a breakpoint on such functions simply as if they had no tag, like: (gdb) b function(int) Or if you need to disambiguate between tags, like: (gdb) b function[abi:other_tag](int) Tab completion was adjusted accordingly as well. * Python Scripting ** New events gdb.new_inferior, gdb.inferior_deleted, and gdb.new_thread are emitted. See the manual for further description of these. ** A new function, "gdb.rbreak" has been added to the Python API. This function allows the setting of a large number of breakpoints via a regex pattern in Python. See the manual for further details. ** Python breakpoints can now accept explicit locations. See the manual for a further description of this feature. * New features in the GDB remote stub, GDBserver ** GDBserver is now able to start inferior processes with a specified initial working directory. The user can set the desired working directory to be used from GDB using the new "set cwd" command. ** New "--selftest" command line option runs some GDBserver self tests. These self tests are disabled in releases. ** On Unix systems, GDBserver now does globbing expansion and variable substitution in inferior command line arguments. This is done by starting inferiors using a shell, like GDB does. See "set startup-with-shell" in the user manual for how to disable this from GDB when using "target extended-remote". When using "target remote", you can disable the startup with shell by using the new "--no-startup-with-shell" GDBserver command line option. ** On Unix systems, GDBserver now supports receiving environment variables that are to be set or unset from GDB. These variables will affect the environment to be passed to the inferior. * When catching an Ada exception raised with a message, GDB now prints the message in the catchpoint hit notification. In GDB/MI mode, that information is provided as an extra field named "exception-message" in the *stopped notification. * Trait objects can now be inspected When debugging Rust code. This requires compiler support which will appear in Rust 1.24. * New remote packets QEnvironmentHexEncoded Inform GDBserver of an environment variable that is to be passed to the inferior when starting it. QEnvironmentUnset Inform GDBserver of an environment variable that is to be unset before starting the remote inferior. QEnvironmentReset Inform GDBserver that the environment should be reset (i.e., user-set environment variables should be unset). QStartupWithShell Indicates whether the inferior must be started with a shell or not. QSetWorkingDir Tell GDBserver that the inferior to be started should use a specific working directory. * The "maintenance print c-tdesc" command now takes an optional argument which is the file name of XML target description. * The "maintenance selftest" command now takes an optional argument to filter the tests to be run. * The "enable", and "disable" commands now accept a range of breakpoint locations, e.g. "enable 1.3-5". * New commands set|show cwd Set and show the current working directory for the inferior. set|show compile-gcc Set and show compilation command used for compiling and injecting code with the 'compile' commands. set debug separate-debug-file show debug separate-debug-file Control the display of debug output about separate debug file search. set dump-excluded-mappings show dump-excluded-mappings Control whether mappings marked with the VM_DONTDUMP flag should be dumped when generating a core file. maint info selftests List the registered selftests. starti Start the debugged program stopping at the first instruction. set|show debug or1k Control display of debugging messages related to OpenRISC targets. set|show print type nested-type-limit Set and show the limit of nesting level for nested types that the type printer will show. * TUI Single-Key mode now supports two new shortcut keys: `i' for stepi and `o' for nexti. * Safer/improved support for debugging with no debug info GDB no longer assumes functions with no debug information return 'int'. This means that GDB now refuses to call such functions unless you tell it the function's type, by either casting the call to the declared return type, or by casting the function to a function pointer of the right type, and calling that: (gdb) p getenv ("PATH") 'getenv' has unknown return type; cast the call to its declared return type (gdb) p (char *) getenv ("PATH") $1 = 0x7fffffffe "/usr/local/bin:/"... (gdb) p ((char * (*) (const char *)) getenv) ("PATH") $2 = 0x7fffffffe "/usr/local/bin:/"... Similarly, GDB no longer assumes that global variables with no debug info have type 'int', and refuses to print the variable's value unless you tell it the variable's type: (gdb) p var 'var' has unknown type; cast it to its declared type (gdb) p (float) var $3 = 3.14 * New native configurations FreeBSD/aarch64 aarch64*-*-freebsd* FreeBSD/arm arm*-*-freebsd* * New targets FreeBSD/aarch64 aarch64*-*-freebsd* FreeBSD/arm arm*-*-freebsd* OpenRISC ELF or1k*-*-elf * Removed targets and native configurations Solaris 2.0-9 i?86-*-solaris2.[0-9], sparc*-*-solaris2.[0-9] *** Changes in GDB 8.0 * GDB now supports access to the PKU register on GNU/Linux. The register is added by the Memory Protection Keys for Userspace feature which will be available in future Intel CPUs. * GDB now supports C++11 rvalue references. * Python Scripting ** New functions to start, stop and access a running btrace recording. ** Rvalue references are now supported in gdb.Type. * GDB now supports recording and replaying rdrand and rdseed Intel 64 instructions. * Building GDB and GDBserver now requires a C++11 compiler. For example, GCC 4.8 or later. It is no longer possible to build GDB or GDBserver with a C compiler. The --disable-build-with-cxx configure option has been removed. * Building GDB and GDBserver now requires GNU make >= 3.81. It is no longer supported to build GDB or GDBserver with another implementation of the make program or an earlier version of GNU make. * Native debugging on MS-Windows supports command-line redirection Command-line arguments used for starting programs on MS-Windows can now include redirection symbols supported by native Windows shells, such as '<', '>', '>>', '2>&1', etc. This affects GDB commands such as "run", "start", and "set args", as well as the corresponding MI features. * Support for thread names on MS-Windows. GDB now catches and handles the special exception that programs running on MS-Windows use to assign names to threads in the debugger. * Support for Java programs compiled with gcj has been removed. * User commands now accept an unlimited number of arguments. Previously, only up to 10 was accepted. * The "eval" command now expands user-defined command arguments. This makes it easier to process a variable number of arguments: define mycommand set $i = 0 while $i < $argc eval "print $arg%d", $i set $i = $i + 1 end end * Target descriptions can now describe registers for sparc32 and sparc64. * GDB now supports DWARF version 5 (debug information format). Its .debug_names index is not yet supported. * New native configurations FreeBSD/mips mips*-*-freebsd * New targets Synopsys ARC arc*-*-elf32 FreeBSD/mips mips*-*-freebsd * Removed targets and native configurations Alpha running FreeBSD alpha*-*-freebsd* Alpha running GNU/kFreeBSD alpha*-*-kfreebsd*-gnu * New commands flash-erase Erases all the flash memory regions reported by the target. maint print arc arc-instruction address Print internal disassembler information about instruction at a given address. * New options set disassembler-options show disassembler-options Controls the passing of target specific information to the disassembler. If it is necessary to specify more than one disassembler option then multiple options can be placed together into a comma separated list. The default value is the empty string. Currently, the only supported targets are ARM, PowerPC and S/390. * New MI commands -target-flash-erase Erases all the flash memory regions reported by the target. This is equivalent to the CLI command flash-erase. -file-list-shared-libraries List the shared libraries in the program. This is equivalent to the CLI command "info shared". -catch-handlers Catchpoints stopping the program when Ada exceptions are handled. This is equivalent to the CLI command "catch handlers". *** Changes in GDB 7.12 * GDB and GDBserver now build with a C++ compiler by default. The --enable-build-with-cxx configure option is now enabled by default. One must now explicitly configure with --disable-build-with-cxx in order to build with a C compiler. This option will be removed in a future release. * GDBserver now supports recording btrace without maintaining an active GDB connection. * GDB now supports a negative repeat count in the 'x' command to examine memory backward from the given address. For example: (gdb) bt #0 Func1 (n=42, p=0x40061c "hogehoge") at main.cpp:4 #1 0x400580 in main (argc=1, argv=0x7fffffffe5c8) at main.cpp:8 (gdb) x/-5i 0x0000000000400580 0x40056a : mov %edi,-0x4(%rbp) 0x40056d : mov %rsi,-0x10(%rbp) 0x400571 : mov $0x40061c,%esi 0x400576 : mov $0x2a,%edi 0x40057b : callq 0x400536 * Fortran: Support structures with fields of dynamic types and arrays of dynamic types. * The symbol dumping maintenance commands have new syntax. maint print symbols [-pc address] [--] [filename] maint print symbols [-objfile objfile] [-source source] [--] [filename] maint print psymbols [-objfile objfile] [-pc address] [--] [filename] maint print psymbols [-objfile objfile] [-source source] [--] [filename] maint print msymbols [-objfile objfile] [--] [filename] * GDB now supports multibit bitfields and enums in target register descriptions. * New Python-based convenience function $_as_string(val), which returns the textual representation of a value. This function is especially useful to obtain the text label of an enum value. * Intel MPX bound violation handling. Segmentation faults caused by a Intel MPX boundary violation now display the kind of violation (upper or lower), the memory address accessed and the memory bounds, along with the usual signal received and code location. For example: Program received signal SIGSEGV, Segmentation fault Upper bound violation while accessing address 0x7fffffffc3b3 Bounds: [lower = 0x7fffffffc390, upper = 0x7fffffffc3a3] 0x0000000000400d7c in upper () at i386-mpx-sigsegv.c:68 * Rust language support. GDB now supports debugging programs written in the Rust programming language. See https://www.rust-lang.org/ for more information about Rust. * Support for running interpreters on specified input/output devices GDB now supports a new mechanism that allows frontends to provide fully featured GDB console views, as a better alternative to building such views on top of the "-interpreter-exec console" command. See the new "new-ui" command below. With that command, frontends can now start GDB in the traditional command-line mode running in an embedded terminal emulator widget, and create a separate MI interpreter running on a specified i/o device. In this way, GDB handles line editing, history, tab completion, etc. in the console all by itself, and the GUI uses the separate MI interpreter for its own control and synchronization, invisible to the command line. * The "catch syscall" command catches groups of related syscalls. The "catch syscall" command now supports catching a group of related syscalls using the 'group:' or 'g:' prefix. * New commands skip -file file skip -gfile file-glob-pattern skip -function function skip -rfunction regular-expression A generalized form of the skip command, with new support for glob-style file names and regular expressions for function names. Additionally, a file spec and a function spec may now be combined. maint info line-table REGEXP Display the contents of GDB's internal line table data structure. maint selftest Run any GDB unit tests that were compiled in. new-ui INTERP TTY Start a new user interface instance running INTERP as interpreter, using the TTY file for input/output. * Python Scripting ** gdb.Breakpoint objects have a new attribute "pending", which indicates whether the breakpoint is pending. ** Three new breakpoint-related events have been added: gdb.breakpoint_created, gdb.breakpoint_modified, and gdb.breakpoint_deleted. signal-event EVENTID Signal ("set") the given MS-Windows event object. This is used in conjunction with the Windows JIT debugging (AeDebug) support, where the OS suspends a crashing process until a debugger can attach to it. Resuming the crashing process, in order to debug it, is done by signalling an event. * Support for tracepoints and fast tracepoints on s390-linux and s390x-linux was added in GDBserver, including JIT compiling fast tracepoint's conditional expression bytecode into native code. * Support for various remote target protocols and ROM monitors has been removed: target m32rsdi Remote M32R debugging over SDI target mips MIPS remote debugging protocol target pmon PMON ROM monitor target ddb NEC's DDB variant of PMON for Vr4300 target rockhopper NEC RockHopper variant of PMON target lsi LSI variant of PMO * Support for tracepoints and fast tracepoints on powerpc-linux, powerpc64-linux, and powerpc64le-linux was added in GDBserver, including JIT compiling fast tracepoint's conditional expression bytecode into native code. * MI async record =record-started now includes the method and format used for recording. For example: =record-started,thread-group="i1",method="btrace",format="bts" * MI async record =thread-selected now includes the frame field. For example: =thread-selected,id="3",frame={level="0",addr="0x00000000004007c0"} * New targets Andes NDS32 nds32*-*-elf *** Changes in GDB 7.11 * GDB now supports debugging kernel-based threads on FreeBSD. * Per-inferior thread numbers Thread numbers are now per inferior instead of global. If you're debugging multiple inferiors, GDB displays thread IDs using a qualified INF_NUM.THR_NUM form. For example: (gdb) info threads Id Target Id Frame 1.1 Thread 0x7ffff7fc2740 (LWP 8155) (running) 1.2 Thread 0x7ffff7fc1700 (LWP 8168) (running) * 2.1 Thread 0x7ffff7fc2740 (LWP 8157) (running) 2.2 Thread 0x7ffff7fc1700 (LWP 8190) (running) As consequence, thread numbers as visible in the $_thread convenience variable and in Python's InferiorThread.num attribute are no longer unique between inferiors. GDB now maintains a second thread ID per thread, referred to as the global thread ID, which is the new equivalent of thread numbers in previous releases. See also $_gthread below. For backwards compatibility, MI's thread IDs always refer to global IDs. * Commands that accept thread IDs now accept the qualified INF_NUM.THR_NUM form as well. For example: (gdb) thread 2.1 [Switching to thread 2.1 (Thread 0x7ffff7fc2740 (LWP 8157))] (running) (gdb) * In commands that accept a list of thread IDs, you can now refer to all threads of an inferior using a star wildcard. GDB accepts "INF_NUM.*", to refer to all threads of inferior INF_NUM, and "*" to refer to all threads of the current inferior. For example, "info threads 2.*". * You can use "info threads -gid" to display the global thread ID of all threads. * The new convenience variable $_gthread holds the global number of the current thread. * The new convenience variable $_inferior holds the number of the current inferior. * GDB now displays the ID and name of the thread that hit a breakpoint or received a signal, if your program is multi-threaded. For example: Thread 3 "bar" hit Breakpoint 1 at 0x40087a: file program.c, line 20. Thread 1 "main" received signal SIGINT, Interrupt. * Record btrace now supports non-stop mode. * Support for tracepoints on aarch64-linux was added in GDBserver. * The 'record instruction-history' command now indicates speculative execution when using the Intel Processor Trace recording format. * GDB now allows users to specify explicit locations, bypassing the linespec parser. This feature is also available to GDB/MI clients. * Multi-architecture debugging is supported on AArch64 GNU/Linux. GDB now is able to debug both AArch64 applications and ARM applications at the same time. * Support for fast tracepoints on aarch64-linux was added in GDBserver, including JIT compiling fast tracepoint's conditional expression bytecode into native code. * GDB now supports displaced stepping on AArch64 GNU/Linux. * "info threads", "info inferiors", "info display", "info checkpoints" and "maint info program-spaces" now list the corresponding items in ascending ID order, for consistency with all other "info" commands. * In Ada, the overloads selection menu has been enhanced to display the parameter types and the return types for the matching overloaded subprograms. * New commands maint set target-non-stop (on|off|auto) maint show target-non-stop Control whether GDB targets always operate in non-stop mode even if "set non-stop" is "off". The default is "auto", meaning non-stop mode is enabled if supported by the target. maint set bfd-sharing maint show bfd-sharing Control the reuse of bfd objects. set debug bfd-cache show debug bfd-cache Control display of debugging info regarding bfd caching. set debug fbsd-lwp show debug fbsd-lwp Control display of debugging info regarding FreeBSD threads. set remote multiprocess-extensions-packet show remote multiprocess-extensions-packet Set/show the use of the remote protocol multiprocess extensions. set remote thread-events show remote thread-events Set/show the use of thread create/exit events. set ada print-signatures on|off show ada print-signatures" Control whether parameter types and return types are displayed in overloads selection menus. It is activated (@code{on}) by default. set max-value-size show max-value-size Controls the maximum size of memory, in bytes, that GDB will allocate for value contents. Prevents incorrect programs from causing GDB to allocate overly large buffers. Default is 64k. * The "disassemble" command accepts a new modifier: /s. It prints mixed source+disassembly like /m with two differences: - disassembled instructions are now printed in program order, and - and source for all relevant files is now printed. The "/m" option is now considered deprecated: its "source-centric" output hasn't proved useful in practice. * The "record instruction-history" command accepts a new modifier: /s. It behaves exactly like /m and prints mixed source+disassembly. * The "set scheduler-locking" command supports a new mode "replay". It behaves like "off" in record mode and like "on" in replay mode. * Support for various ROM monitors has been removed: target dbug dBUG ROM monitor for Motorola ColdFire target picobug Motorola picobug monitor target dink32 DINK32 ROM monitor for PowerPC target m32r Renesas M32R/D ROM monitor target mon2000 mon2000 ROM monitor target ppcbug PPCBUG ROM monitor for PowerPC * Support for reading/writing memory and extracting values on architectures whose memory is addressable in units of any integral multiple of 8 bits. catch handlers Allows to break when an Ada exception is handled. * New remote packets exec stop reason Indicates that an exec system call was executed. exec-events feature in qSupported The qSupported packet allows GDB to request support for exec events using the new 'gdbfeature' exec-event, and the qSupported response can contain the corresponding 'stubfeature'. Set and show commands can be used to display whether these features are enabled. vCtrlC Equivalent to interrupting with the ^C character, but works in non-stop mode. thread created stop reason (T05 create:...) Indicates that the thread was just created and is stopped at entry. thread exit stop reply (w exitcode;tid) Indicates that the thread has terminated. QThreadEvents Enables/disables thread create and exit event reporting. For example, this is used in non-stop mode when GDB stops a set of threads and synchronously waits for the their corresponding stop replies. Without exit events, if one of the threads exits, GDB would hang forever not knowing that it should no longer expect a stop for that same thread. N stop reply Indicates that there are no resumed threads left in the target (all threads are stopped). The remote stub reports support for this stop reply to GDB's qSupported query. QCatchSyscalls Enables/disables catching syscalls from the inferior process. The remote stub reports support for this packet to GDB's qSupported query. syscall_entry stop reason Indicates that a syscall was just called. syscall_return stop reason Indicates that a syscall just returned. * Extended-remote exec events ** GDB now has support for exec events on extended-remote Linux targets. For such targets with Linux kernels 2.5.46 and later, this enables follow-exec-mode and exec catchpoints. set remote exec-event-feature-packet show remote exec-event-feature-packet Set/show the use of the remote exec event feature. * Thread names in remote protocol The reply to qXfer:threads:read may now include a name attribute for each thread. * Target remote mode fork and exec events ** GDB now has support for fork and exec events on target remote mode Linux targets. For such targets with Linux kernels 2.5.46 and later, this enables follow-fork-mode, detach-on-fork, follow-exec-mode, and fork and exec catchpoints. * Remote syscall events ** GDB now has support for catch syscall on remote Linux targets, currently enabled on x86/x86_64 architectures. set remote catch-syscall-packet show remote catch-syscall-packet Set/show the use of the remote catch syscall feature. * MI changes ** The -var-set-format command now accepts the zero-hexadecimal format. It outputs data in hexadecimal format with zero-padding on the left. * Python Scripting ** gdb.InferiorThread objects have a new attribute "global_num", which refers to the thread's global thread ID. The existing "num" attribute now refers to the thread's per-inferior number. See "Per-inferior thread numbers" above. ** gdb.InferiorThread objects have a new attribute "inferior", which is the Inferior object the thread belongs to. *** Changes in GDB 7.10 * Support for process record-replay and reverse debugging on aarch64*-linux* targets has been added. GDB now supports recording of A64 instruction set including advance SIMD instructions. * Support for Sun's version of the "stabs" debug file format has been removed. * GDB now honors the content of the file /proc/PID/coredump_filter (PID is the process ID) on GNU/Linux systems. This file can be used to specify the types of memory mappings that will be included in a corefile. For more information, please refer to the manual page of "core(5)". GDB also has a new command: "set use-coredump-filter on|off". It allows to set whether GDB will read the content of the /proc/PID/coredump_filter file when generating a corefile. * The "info os" command on GNU/Linux can now display information on cpu information : "info os cpus" Listing of all cpus/cores on the system * GDB has two new commands: "set serial parity odd|even|none" and "show serial parity". These allows to set or show parity for the remote serial I/O. * The "info source" command now displays the producer string if it was present in the debug info. This typically includes the compiler version and may include things like its command line arguments. * The "info dll", an alias of the "info sharedlibrary" command, is now available on all platforms. * Directory names supplied to the "set sysroot" commands may be prefixed with "target:" to tell GDB to access shared libraries from the target system, be it local or remote. This replaces the prefix "remote:". The default sysroot has been changed from "" to "target:". "remote:" is automatically converted to "target:" for backward compatibility. * The system root specified by "set sysroot" will be prepended to the filename of the main executable (if reported to GDB as absolute by the operating system) when starting processes remotely, and when attaching to already-running local or remote processes. * GDB now supports automatic location and retrieval of executable files from remote targets. Remote debugging can now be initiated using only a "target remote" or "target extended-remote" command (no "set sysroot" or "file" commands are required). See "New remote packets" below. * The "dump" command now supports verilog hex format. * GDB now supports the vector ABI on S/390 GNU/Linux targets. * On GNU/Linux, GDB and gdbserver are now able to access executable and shared library files without a "set sysroot" command when attaching to processes running in different mount namespaces from the debugger. This makes it possible to attach to processes in containers as simply as "gdb -p PID" or "gdbserver --attach PID". See "New remote packets" below. * The "tui reg" command now provides completion for all of the available register groups, including target specific groups. * The HISTSIZE environment variable is no longer read when determining the size of GDB's command history. GDB now instead reads the dedicated GDBHISTSIZE environment variable. Setting GDBHISTSIZE to "-1" or to "" now disables truncation of command history. Non-numeric values of GDBHISTSIZE are ignored. * Guile Scripting ** Memory ports can now be unbuffered. * Python Scripting ** gdb.Objfile objects have a new attribute "username", which is the name of the objfile as specified by the user, without, for example, resolving symlinks. ** You can now write frame unwinders in Python. ** gdb.Type objects have a new method "optimized_out", returning optimized out gdb.Value instance of this type. ** gdb.Value objects have new methods "reference_value" and "const_value" which return a reference to the value and a "const" version of the value respectively. * New commands maint print symbol-cache Print the contents of the symbol cache. maint print symbol-cache-statistics Print statistics of symbol cache usage. maint flush-symbol-cache Flush the contents of the symbol cache. record btrace bts record bts Start branch trace recording using Branch Trace Store (BTS) format. compile print Evaluate expression by using the compiler and print result. tui enable tui disable Explicit commands for enabling and disabling tui mode. show mpx bound set mpx bound on i386 and amd64 Support for bound table investigation on Intel MPX enabled applications. record btrace pt record pt Start branch trace recording using Intel Processor Trace format. maint info btrace Print information about branch tracing internals. maint btrace packet-history Print the raw branch tracing data. maint btrace clear-packet-history Discard the stored raw branch tracing data. maint btrace clear Discard all branch tracing data. It will be fetched and processed anew by the next "record" command. * New options set debug dwarf-die Renamed from "set debug dwarf2-die". show debug dwarf-die Renamed from "show debug dwarf2-die". set debug dwarf-read Renamed from "set debug dwarf2-read". show debug dwarf-read Renamed from "show debug dwarf2-read". maint set dwarf always-disassemble Renamed from "maint set dwarf2 always-disassemble". maint show dwarf always-disassemble Renamed from "maint show dwarf2 always-disassemble". maint set dwarf max-cache-age Renamed from "maint set dwarf2 max-cache-age". maint show dwarf max-cache-age Renamed from "maint show dwarf2 max-cache-age". set debug dwarf-line show debug dwarf-line Control display of debugging info regarding DWARF line processing. set max-completions show max-completions Set the maximum number of candidates to be considered during completion. The default value is 200. This limit allows GDB to avoid generating large completion lists, the computation of which can cause the debugger to become temporarily unresponsive. set history remove-duplicates show history remove-duplicates Control the removal of duplicate history entries. maint set symbol-cache-size maint show symbol-cache-size Control the size of the symbol cache. set|show record btrace bts buffer-size Set and show the size of the ring buffer used for branch tracing in BTS format. The obtained size may differ from the requested size. Use "info record" to see the obtained buffer size. set debug linux-namespaces show debug linux-namespaces Control display of debugging info regarding Linux namespaces. set|show record btrace pt buffer-size Set and show the size of the ring buffer used for branch tracing in Intel Processor Trace format. The obtained size may differ from the requested size. Use "info record" to see the obtained buffer size. maint set|show btrace pt skip-pad Set and show whether PAD packets are skipped when computing the packet history. * The command 'thread apply all' can now support new option '-ascending' to call its specified command for all threads in ascending order. * Python/Guile scripting ** GDB now supports auto-loading of Python/Guile scripts contained in the special section named `.debug_gdb_scripts'. * New remote packets qXfer:btrace-conf:read Return the branch trace configuration for the current thread. Qbtrace-conf:bts:size Set the requested ring buffer size for branch tracing in BTS format. Qbtrace:pt Enable Intel Processor Trace-based branch tracing for the current process. The remote stub reports support for this packet to GDB's qSupported query. Qbtrace-conf:pt:size Set the requested ring buffer size for branch tracing in Intel Processor Trace format. swbreak stop reason Indicates a memory breakpoint instruction was executed, irrespective of whether it was GDB that planted the breakpoint or the breakpoint is hardcoded in the program. This is required for correct non-stop mode operation. hwbreak stop reason Indicates the target stopped for a hardware breakpoint. This is required for correct non-stop mode operation. vFile:fstat: Return information about files on the remote system. qXfer:exec-file:read Return the full absolute name of the file that was executed to create a process running on the remote system. vFile:setfs: Select the filesystem on which vFile: operations with filename arguments will operate. This is required for GDB to be able to access files on remote targets where the remote stub does not share a common filesystem with the inferior(s). fork stop reason Indicates that a fork system call was executed. vfork stop reason Indicates that a vfork system call was executed. vforkdone stop reason Indicates that a vfork child of the specified process has executed an exec or exit, allowing the vfork parent to resume execution. fork-events and vfork-events features in qSupported The qSupported packet allows GDB to request support for fork and vfork events using new 'gdbfeatures' fork-events and vfork-events, and the qSupported response can contain the corresponding 'stubfeatures'. Set and show commands can be used to display whether these features are enabled. * Extended-remote fork events ** GDB now has support for fork events on extended-remote Linux targets. For targets with Linux kernels 2.5.60 and later, this enables follow-fork-mode and detach-on-fork for both fork and vfork, as well as fork and vfork catchpoints. * The info record command now shows the recording format and the branch tracing configuration for the current thread when using the btrace record target. For the BTS format, it shows the ring buffer size. * GDB now has support for DTrace USDT (Userland Static Defined Tracing) probes. The supported targets are x86_64-*-linux-gnu. * GDB now supports access to vector registers on S/390 GNU/Linux targets. * Removed command line options -xdb HP-UX XDB compatibility mode. * Removed targets and native configurations HP/PA running HP-UX hppa*-*-hpux* Itanium running HP-UX ia64-*-hpux* * New configure options --with-intel-pt This configure option allows the user to build GDB with support for Intel Processor Trace (default: auto). This requires libipt. --with-libipt-prefix=PATH Specify the path to the version of libipt that GDB should use. $PATH/include should contain the intel-pt.h header and $PATH/lib should contain the libipt.so library. *** Changes in GDB 7.9.1 * Python Scripting ** Xmethods can now specify a result type. *** Changes in GDB 7.9 * GDB now supports hardware watchpoints on x86 GNU Hurd. * Python Scripting ** You can now access frame registers from Python scripts. ** New attribute 'producer' for gdb.Symtab objects. ** gdb.Objfile objects have a new attribute "progspace", which is the gdb.Progspace object of the containing program space. ** gdb.Objfile objects have a new attribute "owner". ** gdb.Objfile objects have a new attribute "build_id", which is the build ID generated when the file was built. ** gdb.Objfile objects have a new method "add_separate_debug_file". ** A new event "gdb.clear_objfiles" has been added, triggered when selecting a new file to debug. ** You can now add attributes to gdb.Objfile and gdb.Progspace objects. ** New function gdb.lookup_objfile. New events which are triggered when GDB modifies the state of the inferior. ** gdb.events.inferior_call_pre: Function call is about to be made. ** gdb.events.inferior_call_post: Function call has just been made. ** gdb.events.memory_changed: A memory location has been altered. ** gdb.events.register_changed: A register has been altered. * New Python-based convenience functions: ** $_caller_is(name [, number_of_frames]) ** $_caller_matches(regexp [, number_of_frames]) ** $_any_caller_is(name [, number_of_frames]) ** $_any_caller_matches(regexp [, number_of_frames]) * GDB now supports the compilation and injection of source code into the inferior. GDB will use GCC 5.0 or higher built with libcc1.so to compile the source code to object code, and if successful, inject and execute that code within the current context of the inferior. Currently the C language is supported. The commands used to interface with this new feature are: compile code [-raw|-r] [--] [source code] compile file [-raw|-r] filename * New commands demangle [-l language] [--] name Demangle "name" in the specified language, or the current language if elided. This command is renamed from the "maint demangle" command. The latter is kept as a no-op to avoid "maint demangle" being interpreted as "maint demangler-warning". queue-signal signal-name-or-number Queue a signal to be delivered to the thread when it is resumed. add-auto-load-scripts-directory directory Add entries to the list of directories from which to load auto-loaded scripts. maint print user-registers List all currently available "user" registers. compile code [-r|-raw] [--] [source code] Compile, inject, and execute in the inferior the executable object code produced by compiling the provided source code. compile file [-r|-raw] filename Compile and inject into the inferior the executable object code produced by compiling the source code stored in the filename provided. * On resume, GDB now always passes the signal the program had stopped for to the thread the signal was sent to, even if the user changed threads before resuming. Previously GDB would often (but not always) deliver the signal to the thread that happens to be current at resume time. * Conversely, the "signal" command now consistently delivers the requested signal to the current thread. GDB now asks for confirmation if the program had stopped for a signal and the user switched threads meanwhile. * "breakpoint always-inserted" modes "off" and "auto" merged. Now, when 'breakpoint always-inserted mode' is set to "off", GDB won't remove breakpoints from the target until all threads stop, even in non-stop mode. The "auto" mode has been removed, and "off" is now the default mode. * New options set debug symbol-lookup show debug symbol-lookup Control display of debugging info regarding symbol lookup. * MI changes ** The -list-thread-groups command outputs an exit-code field for inferiors that have exited. * New targets MIPS SDE mips*-sde*-elf* * Removed targets Support for these obsolete configurations has been removed. Alpha running OSF/1 (or Tru64) alpha*-*-osf* SGI Irix-5.x mips-*-irix5* SGI Irix-6.x mips-*-irix6* VAX running (4.2 - 4.3 Reno) BSD vax-*-bsd* VAX running Ultrix vax-*-ultrix* * The "dll-symbols" command, and its two aliases ("add-shared-symbol-files" and "assf"), have been removed. Use the "sharedlibrary" command, or its alias "share", instead. *** Changes in GDB 7.8 * New command line options -D data-directory This is an alias for the --data-directory option. * GDB supports printing and modifying of variable length automatic arrays as specified in ISO C99. * The ARM simulator now supports instruction level tracing with or without disassembly. * Guile scripting GDB now has support for scripting using Guile. Whether this is available is determined at configure time. Guile version 2.0 or greater is required. Guile version 2.0.9 is well tested, earlier 2.0 versions are not. * New commands (for set/show, see "New options" below) guile [code] gu [code] Invoke CODE by passing it to the Guile interpreter. guile-repl gr Start a Guile interactive prompt (or "repl" for "read-eval-print loop"). info auto-load guile-scripts [regexp] Print the list of automatically loaded Guile scripts. * The source command is now capable of sourcing Guile scripts. This feature is dependent on the debugger being built with Guile support. * New options set print symbol-loading (off|brief|full) show print symbol-loading Control whether to print informational messages when loading symbol information for a file. The default is "full", but when debugging programs with large numbers of shared libraries the amount of output becomes less useful. set guile print-stack (none|message|full) show guile print-stack Show a stack trace when an error is encountered in a Guile script. set auto-load guile-scripts (on|off) show auto-load guile-scripts Control auto-loading of Guile script files. maint ada set ignore-descriptive-types (on|off) maint ada show ignore-descriptive-types Control whether the debugger should ignore descriptive types in Ada programs. The default is not to ignore the descriptive types. See the user manual for more details on descriptive types and the intended usage of this option. set auto-connect-native-target Control whether GDB is allowed to automatically connect to the native target for the run, attach, etc. commands when not connected to any target yet. See also "target native" below. set record btrace replay-memory-access (read-only|read-write) show record btrace replay-memory-access Control what memory accesses are allowed during replay. maint set target-async (on|off) maint show target-async This controls whether GDB targets operate in synchronous or asynchronous mode. Normally the default is asynchronous, if it is available; but this can be changed to more easily debug problems occurring only in synchronous mode. set mi-async (on|off) show mi-async Control whether MI asynchronous mode is preferred. This supersedes "set target-async" of previous GDB versions. * "set target-async" is deprecated as a CLI option and is now an alias for "set mi-async" (only puts MI into async mode). * Background execution commands (e.g., "c&", "s&", etc.) are now possible ``out of the box'' if the target supports them. Previously the user would need to explicitly enable the possibility with the "set target-async on" command. * New features in the GDB remote stub, GDBserver ** New option --debug-format=option1[,option2,...] allows one to add additional text to each output. At present only timestamps are supported: --debug-format=timestamps. Timestamps can also be turned on with the "monitor set debug-format timestamps" command from GDB. * The 'record instruction-history' command now starts counting instructions at one. This also affects the instruction ranges reported by the 'record function-call-history' command when given the /i modifier. * The command 'record function-call-history' supports a new modifier '/c' to indent the function names based on their call stack depth. The fields for the '/i' and '/l' modifier have been reordered. The source line range is now prefixed with 'at'. The instruction range is now prefixed with 'inst'. Both ranges are now printed as ', ' to allow copy&paste to the "record instruction-history" and "list" commands. * The ranges given as arguments to the 'record function-call-history' and 'record instruction-history' commands are now inclusive. * The btrace record target now supports the 'record goto' command. For locations inside the execution trace, the back trace is computed based on the information stored in the execution trace. * The btrace record target supports limited reverse execution and replay. The target does not record data and therefore does not allow reading memory or registers. * The "catch syscall" command now works on s390*-linux* targets. * The "compare-sections" command is no longer specific to target remote. It now works with all targets. * All native targets are now consistently called "native". Consequently, the "target child", "target GNU", "target djgpp", "target procfs" (Solaris/Irix/OSF/AIX) and "target darwin-child" commands have been replaced with "target native". The QNX/NTO port leaves the "procfs" target in place and adds a "native" target for consistency with other ports. The impact on users should be minimal as these commands previously either throwed an error, or were no-ops. The target's name is visible in the output of the following commands: "help target", "info target", "info files", "maint print target-stack". * The "target native" command now connects to the native target. This can be used to launch native programs even when "set auto-connect-native-target" is set to off. * GDB now supports access to Intel MPX registers on GNU/Linux. * Support for Intel AVX-512 registers on GNU/Linux. Support displaying and modifying Intel AVX-512 registers $zmm0 - $zmm31 and $k0 - $k7 on GNU/Linux. * New remote packets qXfer:btrace:read's annex The qXfer:btrace:read packet supports a new annex 'delta' to read branch trace incrementally. * Python Scripting ** Valid Python operations on gdb.Value objects representing structs/classes invoke the corresponding overloaded operators if available. ** New `Xmethods' feature in the Python API. Xmethods are additional methods or replacements for existing methods of a C++ class. This feature is useful for those cases where a method defined in C++ source code could be inlined or optimized out by the compiler, making it unavailable to GDB. * New targets PowerPC64 GNU/Linux little-endian powerpc64le-*-linux* * The "dll-symbols" command, and its two aliases ("add-shared-symbol-files" and "assf"), have been deprecated. Use the "sharedlibrary" command, or its alias "share", instead. * The commands "set remotebaud" and "show remotebaud" are no longer supported. Use "set serial baud" and "show serial baud" (respectively) instead. * MI changes ** A new option "-gdb-set mi-async" replaces "-gdb-set target-async". The latter is left as a deprecated alias of the former for backward compatibility. If the target supports it, CLI background execution commands are now always possible by default, independently of whether the frontend stated a preference for asynchronous execution with "-gdb-set mi-async". Previously "-gdb-set target-async off" affected both MI execution commands and CLI execution commands. *** Changes in GDB 7.7 * Improved support for process record-replay and reverse debugging on arm*-linux* targets. Support for thumb32 and syscall instruction recording has been added. * GDB now supports SystemTap SDT probes on AArch64 GNU/Linux. * GDB now supports Fission DWP file format version 2. http://gcc.gnu.org/wiki/DebugFission * New convenience function "$_isvoid", to check whether an expression is void. A void expression is an expression where the type of the result is "void". For example, some convenience variables may be "void" when evaluated (e.g., "$_exitcode" before the execution of the program being debugged; or an undefined convenience variable). Another example, when calling a function whose return type is "void". * The "maintenance print objfiles" command now takes an optional regexp. * The "catch syscall" command now works on arm*-linux* targets. * GDB now consistently shows "" when printing values of registers the debug info indicates have not been saved in the frame and there's nowhere to retrieve them from (callee-saved/call-clobbered registers): (gdb) p $rax $1 = (gdb) info registers rax rax Before, the former would print "", and the latter "*value not available*". * New script contrib/gdb-add-index.sh for adding .gdb_index sections to binaries. * Python scripting ** Frame filters and frame decorators have been added. ** Temporary breakpoints are now supported. ** Line tables representation has been added. ** New attribute 'parent_type' for gdb.Field objects. ** gdb.Field objects can be used as subscripts on gdb.Value objects. ** New attribute 'name' for gdb.Type objects. * New targets Nios II ELF nios2*-*-elf Nios II GNU/Linux nios2*-*-linux Texas Instruments MSP430 msp430*-*-elf * Removed native configurations Support for these a.out NetBSD and OpenBSD obsolete configurations has been removed. ELF variants of these configurations are kept supported. arm*-*-netbsd* but arm*-*-netbsdelf* is kept supported. i[34567]86-*-netbsd* but i[34567]86-*-netbsdelf* is kept supported. i[34567]86-*-openbsd[0-2].* but i[34567]86-*-openbsd* is kept supported. i[34567]86-*-openbsd3.[0-3] m68*-*-netbsd* but m68*-*-netbsdelf* is kept supported. sparc-*-netbsd* but sparc-*-netbsdelf* is kept supported. vax-*-netbsd* but vax-*-netbsdelf* is kept supported. * New commands: catch rethrow Like "catch throw", but catches a re-thrown exception. maint check-psymtabs Renamed from old "maint check-symtabs". maint check-symtabs Perform consistency checks on symtabs. maint expand-symtabs Expand symtabs matching an optional regexp. show configuration Display the details of GDB configure-time options. maint set|show per-command maint set|show per-command space maint set|show per-command time maint set|show per-command symtab Enable display of per-command gdb resource usage. remove-symbol-file FILENAME remove-symbol-file -a ADDRESS Remove a symbol file added via add-symbol-file. The file to remove can be identified by its filename or by an address that lies within the boundaries of this symbol file in memory. info exceptions info exceptions REGEXP Display the list of Ada exceptions defined in the program being debugged. If provided, only the exceptions whose names match REGEXP are listed. * New options set debug symfile off|on show debug symfile Control display of debugging info regarding reading symbol files and symbol tables within those files set print raw frame-arguments show print raw frame-arguments Set/show whether to print frame arguments in raw mode, disregarding any defined pretty-printers. set remote trace-status-packet show remote trace-status-packet Set/show the use of remote protocol qTStatus packet. set debug nios2 show debug nios2 Control display of debugging messages related to Nios II targets. set range-stepping show range-stepping Control whether target-assisted range stepping is enabled. set startup-with-shell show startup-with-shell Specifies whether Unix child processes are started via a shell or directly. set code-cache show code-cache Use the target memory cache for accesses to the code segment. This improves performance of remote debugging (particularly disassembly). * You can now use a literal value 'unlimited' for options that interpret 0 or -1 as meaning "unlimited". E.g., "set trace-buffer-size unlimited" is now an alias for "set trace-buffer-size -1" and "set height unlimited" is now an alias for "set height 0". * The "set debug symtab-create" debugging option of GDB has been changed to accept a verbosity level. 0 means "off", 1 provides basic debugging output, and values of 2 or greater provides more verbose output. * New command-line options --configuration Display the details of GDB configure-time options. * The command 'tsave' can now support new option '-ctf' to save trace buffer in Common Trace Format. * Newly installed $prefix/bin/gcore acts as a shell interface for the GDB command gcore. * GDB now implements the C++ 'typeid' operator. * The new convenience variable $_exception holds the exception being thrown or caught at an exception-related catchpoint. * The exception-related catchpoints, like "catch throw", now accept a regular expression which can be used to filter exceptions by type. * The new convenience variable $_exitsignal is automatically set to the terminating signal number when the program being debugged dies due to an uncaught signal. * MI changes ** All MI commands now accept an optional "--language" option. Support for this feature can be verified by using the "-list-features" command, which should contain "language-option". ** The new command -info-gdb-mi-command allows the user to determine whether a GDB/MI command is supported or not. ** The "^error" result record returned when trying to execute an undefined GDB/MI command now provides a variable named "code" whose content is the "undefined-command" error code. Support for this feature can be verified by using the "-list-features" command, which should contain "undefined-command-error-code". ** The -trace-save MI command can optionally save trace buffer in Common Trace Format now. ** The new command -dprintf-insert sets a dynamic printf breakpoint. ** The command -data-list-register-values now accepts an optional "--skip-unavailable" option. When used, only the available registers are displayed. ** The new command -trace-frame-collected dumps collected variables, computed expressions, tvars, memory and registers in a traceframe. ** The commands -stack-list-locals, -stack-list-arguments and -stack-list-variables now accept an option "--skip-unavailable". When used, only the available locals or arguments are displayed. ** The -exec-run command now accepts an optional "--start" option. When used, the command follows the same semantics as the "start" command, stopping the program's execution at the start of its main subprogram. Support for this feature can be verified using the "-list-features" command, which should contain "exec-run-start-option". ** The new commands -catch-assert and -catch-exceptions insert catchpoints stopping the program when Ada exceptions are raised. ** The new command -info-ada-exceptions provides the equivalent of the new "info exceptions" command. * New system-wide configuration scripts A GDB installation now provides scripts suitable for use as system-wide configuration scripts for the following systems: ** ElinOS ** Wind River Linux * GDB now supports target-assigned range stepping with remote targets. This improves the performance of stepping source lines by reducing the number of control packets from/to GDB. See "New remote packets" below. * GDB now understands the element 'tvar' in the XML traceframe info. It has the id of the collected trace state variables. * On S/390 targets that provide the transactional-execution feature, the program interruption transaction diagnostic block (TDB) is now represented as a number of additional "registers" in GDB. * New remote packets vCont;r The vCont packet supports a new 'r' action, that tells the remote stub to step through an address range itself, without GDB involvemement at each single-step. qXfer:libraries-svr4:read's annex The previously unused annex of the qXfer:libraries-svr4:read packet is now used to support passing an argument list. The remote stub reports support for this argument list to GDB's qSupported query. The defined arguments are "start" and "prev", used to reduce work necessary for library list updating, resulting in significant speedup. * New features in the GDB remote stub, GDBserver ** GDBserver now supports target-assisted range stepping. Currently enabled on x86/x86_64 GNU/Linux targets. ** GDBserver now adds element 'tvar' in the XML in the reply to 'qXfer:traceframe-info:read'. It has the id of the collected trace state variables. ** GDBserver now supports hardware watchpoints on the MIPS GNU/Linux target. * New 'z' formatter for printing and examining memory, this displays the value as hexadecimal zero padded on the left to the size of the type. * GDB can now use Windows x64 unwinding data. * The "set remotebaud" command has been replaced by "set serial baud". Similarly, "show remotebaud" has been replaced by "show serial baud". The "set remotebaud" and "show remotebaud" commands are still available to provide backward compatibility with older versions of GDB. *** Changes in GDB 7.6 * Target record has been renamed to record-full. Record/replay is now enabled with the "record full" command. This also affects settings that are associated with full record/replay that have been moved from "set/show record" to "set/show record full": set|show record full insn-number-max set|show record full stop-at-limit set|show record full memory-query * A new record target "record-btrace" has been added. The new target uses hardware support to record the control-flow of a process. It does not support replaying the execution, but it implements the below new commands for investigating the recorded execution log. This new recording method can be enabled using: record btrace The "record-btrace" target is only available on Intel Atom processors and requires a Linux kernel 2.6.32 or later. * Two new commands have been added for record/replay to give information about the recorded execution without having to replay the execution. The commands are only supported by "record btrace". record instruction-history prints the execution history at instruction granularity record function-call-history prints the execution history at function granularity * New native configurations ARM AArch64 GNU/Linux aarch64*-*-linux-gnu FreeBSD/powerpc powerpc*-*-freebsd x86_64/Cygwin x86_64-*-cygwin* Tilera TILE-Gx GNU/Linux tilegx*-*-linux-gnu * New targets ARM AArch64 aarch64*-*-elf ARM AArch64 GNU/Linux aarch64*-*-linux Lynx 178 PowerPC powerpc-*-lynx*178 x86_64/Cygwin x86_64-*-cygwin* Tilera TILE-Gx GNU/Linux tilegx*-*-linux * If the configured location of system.gdbinit file (as given by the --with-system-gdbinit option at configure time) is in the data-directory (as specified by --with-gdb-datadir at configure time) or in one of its subdirectories, then GDB will look for the system-wide init file in the directory specified by the --data-directory command-line option. * New command line options: -nh Disables auto-loading of ~/.gdbinit, but still executes all the other initialization files, unlike -nx which disables all of them. * Removed command line options -epoch This was used by the gdb mode in Epoch, an ancient fork of Emacs. * The 'ptype' and 'whatis' commands now accept an argument to control type formatting. * 'info proc' now works on some core files. * Python scripting ** Vectors can be created with gdb.Type.vector. ** Python's atexit.register now works in GDB. ** Types can be pretty-printed via a Python API. ** Python 3 is now supported (in addition to Python 2.4 or later) ** New class gdb.Architecture exposes GDB's internal representation of architecture in the Python API. ** New method Frame.architecture returns the gdb.Architecture object corresponding to the frame's architecture. * New Python-based convenience functions: ** $_memeq(buf1, buf2, length) ** $_streq(str1, str2) ** $_strlen(str) ** $_regex(str, regex) * The 'cd' command now defaults to using '~' (the home directory) if not given an argument. * The C++ ABI now defaults to the GNU v3 ABI. This has been the default for GCC since November 2000. * The command 'forward-search' can now be abbreviated as 'fo'. * The command 'info tracepoints' can now display 'installed on target' or 'not installed on target' for each non-pending location of tracepoint. * New configure options --enable-libmcheck/--disable-libmcheck By default, development versions are built with -lmcheck on hosts that support it, in order to help track memory corruption issues. Release versions, on the other hand, are built without -lmcheck by default. The --enable-libmcheck/--disable-libmcheck configure options allow the user to override that default. --with-babeltrace/--with-babeltrace-include/--with-babeltrace-lib This configure option allows the user to build GDB with libbabeltrace using which GDB can read Common Trace Format data. * New commands (for set/show, see "New options" below) catch signal Catch signals. This is similar to "handle", but allows commands and conditions to be attached. maint info bfds List the BFDs known to GDB. python-interactive [command] pi [command] Start a Python interactive prompt, or evaluate the optional command and print the result of expressions. py [command] "py" is a new alias for "python". enable type-printer [name]... disable type-printer [name]... Enable or disable type printers. * Removed commands ** For the Renesas Super-H architecture, the "regs" command has been removed (has been deprecated in GDB 7.5), and "info all-registers" should be used instead. * New options set print type methods (on|off) show print type methods Control whether method declarations are displayed by "ptype". The default is to show them. set print type typedefs (on|off) show print type typedefs Control whether typedef definitions are displayed by "ptype". The default is to show them. set filename-display basename|relative|absolute show filename-display Control the way in which filenames is displayed. The default is "relative", which preserves previous behavior. set trace-buffer-size show trace-buffer-size Request target to change the size of trace buffer. set remote trace-buffer-size-packet auto|on|off show remote trace-buffer-size-packet Control the use of the remote protocol `QTBuffer:size' packet. set debug aarch64 show debug aarch64 Control display of debugging messages related to ARM AArch64. The default is off. set debug coff-pe-read show debug coff-pe-read Control display of debugging messages related to reading of COFF/PE exported symbols. set debug mach-o show debug mach-o Control display of debugging messages related to Mach-O symbols processing. set debug notification show debug notification Control display of debugging info for async remote notification. * MI changes ** Command parameter changes are now notified using new async record "=cmd-param-changed". ** Trace frame changes caused by command "tfind" are now notified using new async record "=traceframe-changed". ** The creation, deletion and modification of trace state variables are now notified using new async records "=tsv-created", "=tsv-deleted" and "=tsv-modified". ** The start and stop of process record are now notified using new async record "=record-started" and "=record-stopped". ** Memory changes are now notified using new async record "=memory-changed". ** The data-disassemble command response will include a "fullname" field containing the absolute file name when source has been requested. ** New optional parameter COUNT added to the "-data-write-memory-bytes" command, to allow pattern filling of memory areas. ** New commands "-catch-load"/"-catch-unload" added for intercepting library load/unload events. ** The response to breakpoint commands and breakpoint async records includes an "installed" field containing a boolean state about each non-pending tracepoint location is whether installed on target or not. ** Output of the "-trace-status" command includes a "trace-file" field containing the name of the trace file being examined. This field is optional, and only present when examining a trace file. ** The "fullname" field is now always present along with the "file" field, even if the file cannot be found by GDB. * GDB now supports the "mini debuginfo" section, .gnu_debugdata. You must have the LZMA library available when configuring GDB for this feature to be enabled. For more information, see: http://fedoraproject.org/wiki/Features/MiniDebugInfo * New remote packets QTBuffer:size Set the size of trace buffer. The remote stub reports support for this packet to gdb's qSupported query. Qbtrace:bts Enable Branch Trace Store (BTS)-based branch tracing for the current thread. The remote stub reports support for this packet to gdb's qSupported query. Qbtrace:off Disable branch tracing for the current thread. The remote stub reports support for this packet to gdb's qSupported query. qXfer:btrace:read Read the traced branches for the current thread. The remote stub reports support for this packet to gdb's qSupported query. *** Changes in GDB 7.5 * GDB now supports x32 ABI. Visit for more x32 ABI info. * GDB now supports access to MIPS DSP registers on Linux targets. * GDB now supports debugging microMIPS binaries. * The "info os" command on GNU/Linux can now display information on several new classes of objects managed by the operating system: "info os procgroups" lists process groups "info os files" lists file descriptors "info os sockets" lists internet-domain sockets "info os shm" lists shared-memory regions "info os semaphores" lists semaphores "info os msg" lists message queues "info os modules" lists loaded kernel modules * GDB now has support for SDT (Static Defined Tracing) probes. Currently, the only implemented backend is for SystemTap probes (). You can set a breakpoint using the new "-probe, "-pstap" or "-probe-stap" options and inspect the probe arguments using the new $_probe_arg family of convenience variables. You can obtain more information about SystemTap in . * GDB now supports reversible debugging on ARM, it allows you to debug basic ARM and THUMB instructions, and provides record/replay support. * The option "symbol-reloading" has been deleted as it is no longer used. * Python scripting ** GDB commands implemented in Python can now be put in command class "gdb.COMMAND_USER". ** The "maint set python print-stack on|off" is now deleted. ** A new class, gdb.printing.FlagEnumerationPrinter, can be used to apply "flag enum"-style pretty-printing to any enum. ** gdb.lookup_symbol can now work when there is no current frame. ** gdb.Symbol now has a 'line' attribute, holding the line number in the source at which the symbol was defined. ** gdb.Symbol now has the new attribute 'needs_frame' and the new method 'value'. The former indicates whether the symbol needs a frame in order to compute its value, and the latter computes the symbol's value. ** A new method 'referenced_value' on gdb.Value objects which can dereference pointer as well as C++ reference values. ** New methods 'global_block' and 'static_block' on gdb.Symtab objects which return the global and static blocks (as gdb.Block objects), of the underlying symbol table, respectively. ** New function gdb.find_pc_line which returns the gdb.Symtab_and_line object associated with a PC value. ** gdb.Symtab_and_line has new attribute 'last' which holds the end of the address range occupied by code for the current source line. * Go language support. GDB now supports debugging programs written in the Go programming language. * GDBserver now supports stdio connections. E.g. (gdb) target remote | ssh myhost gdbserver - hello * The binary "gdbtui" can no longer be built or installed. Use "gdb -tui" instead. * GDB will now print "flag" enums specially. A flag enum is one where all the enumerator values have no bits in common when pairwise "and"ed. When printing a value whose type is a flag enum, GDB will show all the constants, e.g., for enum E { ONE = 1, TWO = 2}: (gdb) print (enum E) 3 $1 = (ONE | TWO) * The filename part of a linespec will now match trailing components of a source file name. For example, "break gcc/expr.c:1000" will now set a breakpoint in build/gcc/expr.c, but not build/libcpp/expr.c. * The "info proc" and "generate-core-file" commands will now also work on remote targets connected to GDBserver on Linux. * The command "info catch" has been removed. It has been disabled since December 2007. * The "catch exception" and "catch assert" commands now accept a condition at the end of the command, much like the "break" command does. For instance: (gdb) catch exception Constraint_Error if Barrier = True Previously, it was possible to add a condition to such catchpoints, but it had to be done as a second step, after the catchpoint had been created, using the "condition" command. * The "info static-tracepoint-marker" command will now also work on native Linux targets with in-process agent. * GDB can now set breakpoints on inlined functions. * The .gdb_index section has been updated to include symbols for inlined functions. GDB will ignore older .gdb_index sections by default, which could cause symbol files to be loaded more slowly until their .gdb_index sections can be recreated. The new command "set use-deprecated-index-sections on" will cause GDB to use any older .gdb_index sections it finds. This will restore performance, but the ability to set breakpoints on inlined functions will be lost in symbol files with older .gdb_index sections. The .gdb_index section has also been updated to record more information about each symbol. This speeds up the "info variables", "info functions" and "info types" commands when used with programs having the .gdb_index section, as well as speeding up debugging with shared libraries using the .gdb_index section. * Ada support for GDB/MI Variable Objects has been added. * GDB can now support 'breakpoint always-inserted mode' in 'record' target. * MI changes ** New command -info-os is the MI equivalent of "info os". ** Output logs ("set logging" and related) now include MI output. * New commands ** "set use-deprecated-index-sections on|off" "show use-deprecated-index-sections on|off" Controls the use of deprecated .gdb_index sections. ** "catch load" and "catch unload" can be used to stop when a shared library is loaded or unloaded, respectively. ** "enable count" can be used to auto-disable a breakpoint after several hits. ** "info vtbl" can be used to show the virtual method tables for C++ and Java objects. ** "explore" and its sub commands "explore value" and "explore type" can be used to recursively explore values and types of expressions. These commands are available only if GDB is configured with '--with-python'. ** "info auto-load" shows status of all kinds of auto-loaded files, "info auto-load gdb-scripts" shows status of auto-loading GDB canned sequences of commands files, "info auto-load python-scripts" shows status of auto-loading Python script files, "info auto-load local-gdbinit" shows status of loading init file (.gdbinit) from current directory and "info auto-load libthread-db" shows status of inferior specific thread debugging shared library loading. ** "info auto-load-scripts", "set auto-load-scripts on|off" and "show auto-load-scripts" commands have been deprecated, use their "info auto-load python-scripts", "set auto-load python-scripts on|off" and "show auto-load python-scripts" counterparts instead. ** "dprintf location,format,args..." creates a dynamic printf, which is basically a breakpoint that does a printf and immediately resumes your program's execution, so it is like a printf that you can insert dynamically at runtime instead of at compiletime. ** "set print symbol" "show print symbol" Controls whether GDB attempts to display the symbol, if any, corresponding to addresses it prints. This defaults to "on", but you can set it to "off" to restore GDB's previous behavior. * Deprecated commands ** For the Renesas Super-H architecture, the "regs" command has been deprecated, and "info all-registers" should be used instead. * New targets Renesas RL78 rl78-*-elf HP OpenVMS ia64 ia64-hp-openvms* * GDBserver supports evaluation of breakpoint conditions. When support is advertised by GDBserver, GDB may be told to send the breakpoint conditions in bytecode form to GDBserver. GDBserver will only report the breakpoint trigger to GDB when its condition evaluates to true. * New options set mips compression show mips compression Select the compressed ISA encoding used in functions that have no symbol information available. The encoding can be set to either of: mips16 micromips and is updated automatically from ELF file flags if available. set breakpoint condition-evaluation show breakpoint condition-evaluation Control whether breakpoint conditions are evaluated by GDB ("host") or by GDBserver ("target"). Default option "auto" chooses the most efficient available mode. This option can improve debugger efficiency depending on the speed of the target. set auto-load off Disable auto-loading globally. show auto-load Show auto-loading setting of all kinds of auto-loaded files. set auto-load gdb-scripts on|off show auto-load gdb-scripts Control auto-loading of GDB canned sequences of commands files. set auto-load python-scripts on|off show auto-load python-scripts Control auto-loading of Python script files. set auto-load local-gdbinit on|off show auto-load local-gdbinit Control loading of init file (.gdbinit) from current directory. set auto-load libthread-db on|off show auto-load libthread-db Control auto-loading of inferior specific thread debugging shared library. set auto-load scripts-directory [:...] show auto-load scripts-directory Set a list of directories from which to load auto-loaded scripts. Automatically loaded Python scripts and GDB scripts are located in one of the directories listed by this option. The delimiter (':' above) may differ according to the host platform. set auto-load safe-path [:...] show auto-load safe-path Set a list of directories from which it is safe to auto-load files. The delimiter (':' above) may differ according to the host platform. set debug auto-load on|off show debug auto-load Control display of debugging info for auto-loading the files above. set dprintf-style gdb|call|agent show dprintf-style Control the way in which a dynamic printf is performed; "gdb" requests a GDB printf command, while "call" causes dprintf to call a function in the inferior. "agent" requests that the target agent (such as GDBserver) do the printing. set dprintf-function show dprintf-function set dprintf-channel show dprintf-channel Set the function and optional first argument to the call when using the "call" style of dynamic printf. set disconnected-dprintf on|off show disconnected-dprintf Control whether agent-style dynamic printfs continue to be in effect after GDB disconnects. * New configure options --with-auto-load-dir Configure default value for the 'set auto-load scripts-directory' setting above. It defaults to '$debugdir:$datadir/auto-load', $debugdir representing global debugging info directories (available via 'show debug-file-directory') and $datadir representing GDB's data directory (available via 'show data-directory'). --with-auto-load-safe-path Configure default value for the 'set auto-load safe-path' setting above. It defaults to the --with-auto-load-dir setting. --without-auto-load-safe-path Set 'set auto-load safe-path' to '/', effectively disabling this security feature. * New remote packets z0/z1 conditional breakpoints extension The z0/z1 breakpoint insertion packets have been extended to carry a list of conditional expressions over to the remote stub depending on the condition evaluation mode. The use of this extension can be controlled via the "set remote conditional-breakpoints-packet" command. QProgramSignals: Specify the signals which the remote stub may pass to the debugged program without GDB involvement. * New command line options --init-command=FILE, -ix Like --command, -x but execute it before loading inferior. --init-eval-command=COMMAND, -iex Like --eval-command=COMMAND, -ex but execute it before loading inferior. *** Changes in GDB 7.4 * GDB now handles ambiguous linespecs more consistently; the existing FILE:LINE support has been expanded to other types of linespecs. A breakpoint will now be set on all matching locations in all inferiors, and locations will be added or removed according to inferior changes. * GDB now allows you to skip uninteresting functions and files when stepping with the "skip function" and "skip file" commands. * GDB has two new commands: "set remote hardware-watchpoint-length-limit" and "show remote hardware-watchpoint-length-limit". These allows to set or show the maximum length limit (in bytes) of a remote target hardware watchpoint. This allows e.g. to use "unlimited" hardware watchpoints with the gdbserver integrated in Valgrind version >= 3.7.0. Such Valgrind watchpoints are slower than real hardware watchpoints but are significantly faster than gdb software watchpoints. * Python scripting ** The register_pretty_printer function in module gdb.printing now takes an optional `replace' argument. If True, the new printer replaces any existing one. ** The "maint set python print-stack on|off" command has been deprecated and will be deleted in GDB 7.5. A new command: "set python print-stack none|full|message" has replaced it. Additionally, the default for "print-stack" is now "message", which just prints the error message without the stack trace. ** A prompt substitution hook (prompt_hook) is now available to the Python API. ** A new Python module, gdb.prompt has been added to the GDB Python modules library. This module provides functionality for escape sequences in prompts (used by set/show extended-prompt). These escape sequences are replaced by their corresponding value. ** Python commands and convenience-functions located in 'data-directory'/python/gdb/command and 'data-directory'/python/gdb/function are now automatically loaded on GDB start-up. ** Blocks now provide four new attributes. global_block and static_block will return the global and static blocks respectively. is_static and is_global are boolean attributes that indicate if the block is one of those two types. ** Symbols now provide the "type" attribute, the type of the symbol. ** The "gdb.breakpoint" function has been deprecated in favor of "gdb.breakpoints". ** A new class "gdb.FinishBreakpoint" is provided to catch the return of a function. This class is based on the "finish" command available in the CLI. ** Type objects for struct and union types now allow access to the fields using standard Python dictionary (mapping) methods. For example, "some_type['myfield']" now works, as does "some_type.items()". ** A new event "gdb.new_objfile" has been added, triggered by loading a new object file. ** A new function, "deep_items" has been added to the gdb.types module in the GDB Python modules library. This function returns an iterator over the fields of a struct or union type. Unlike the standard Python "iteritems" method, it will recursively traverse any anonymous fields. * MI changes ** "*stopped" events can report several new "reason"s, such as "solib-event". ** Breakpoint changes are now notified using new async records, like "=breakpoint-modified". ** New command -ada-task-info. * libthread-db-search-path now supports two special values: $sdir and $pdir. $sdir specifies the default system locations of shared libraries. $pdir specifies the directory where the libpthread used by the application lives. GDB no longer looks in $sdir and $pdir after it has searched the directories mentioned in libthread-db-search-path. If you want to search those directories, they must be specified in libthread-db-search-path. The default value of libthread-db-search-path on GNU/Linux and Solaris systems is now "$sdir:$pdir". $pdir is not supported by gdbserver, it is currently ignored. $sdir is supported by gdbserver. * New configure option --with-iconv-bin. When using the internationalization support like the one in the GNU C library, GDB will invoke the "iconv" program to get a list of supported character sets. If this program lives in a non-standard location, one can use this option to specify where to find it. * When natively debugging programs on PowerPC BookE processors running a Linux kernel version 2.6.34 or later, GDB supports masked hardware watchpoints, which specify a mask in addition to an address to watch. The mask specifies that some bits of an address (the bits which are reset in the mask) should be ignored when matching the address accessed by the inferior against the watchpoint address. See the "PowerPC Embedded" section in the user manual for more details. * The new option --once causes GDBserver to stop listening for connections once the first connection is made. The listening port used by GDBserver will become available after that. * New commands "info macros" and "alias" have been added. * New function parameters suffix @entry specifies value of function parameter at the time the function got called. Entry values are available only since gcc version 4.7. * New commands !SHELL COMMAND "!" is now an alias of the "shell" command. Note that no space is needed between "!" and SHELL COMMAND. * Changed commands watch EXPRESSION mask MASK_VALUE The watch command now supports the mask argument which allows creation of masked watchpoints, if the current architecture supports this feature. info auto-load-scripts [REGEXP] This command was formerly named "maintenance print section-scripts". It is now generally useful and is no longer a maintenance-only command. info macro [-all] [--] MACRO The info macro command has new options `-all' and `--'. The first for printing all definitions of a macro. The second for explicitly specifying the end of arguments and the beginning of the macro name in case the macro name starts with a hyphen. collect[/s] EXPRESSIONS The tracepoint collect command now takes an optional modifier "/s" that directs it to dereference pointer-to-character types and collect the bytes of memory up to a zero byte. The behavior is similar to what you see when you use the regular print command on a string. An optional integer following the "/s" sets a bound on the number of bytes that will be collected. tstart [NOTES] The trace start command now interprets any supplied arguments as a note to be recorded with the trace run, with an effect similar to setting the variable trace-notes. tstop [NOTES] The trace stop command now interprets any arguments as a note to be mentioned along with the tstatus report that the trace was stopped with a command. The effect is similar to setting the variable trace-stop-notes. * Tracepoints can now be enabled and disabled at any time after a trace experiment has been started using the standard "enable" and "disable" commands. It is now possible to start a trace experiment with no enabled tracepoints; GDB will display a warning, but will allow the experiment to begin, assuming that tracepoints will be enabled as needed while the trace is running. * Fast tracepoints on 32-bit x86-architectures can now be placed at locations with 4-byte instructions, when they were previously limited to locations with instructions of 5 bytes or longer. * New options set debug dwarf2-read show debug dwarf2-read Turns on or off display of debugging messages related to reading DWARF debug info. The default is off. set debug symtab-create show debug symtab-create Turns on or off display of debugging messages related to symbol table creation. The default is off. set extended-prompt show extended-prompt Set the GDB prompt, and allow escape sequences to be inserted to display miscellaneous information (see 'help set extended-prompt' for the list of sequences). This prompt (and any information accessed through the escape sequences) is updated every time the prompt is displayed. set print entry-values (both|compact|default|if-needed|no|only|preferred) show print entry-values Set printing of frame argument values at function entry. In some cases GDB can determine the value of function argument which was passed by the function caller, even if the value was modified inside the called function. set debug entry-values show debug entry-values Control display of debugging info for determining frame argument values at function entry and virtual tail call frames. set basenames-may-differ show basenames-may-differ Set whether a source file may have multiple base names. (A "base name" is the name of a file with the directory part removed. Example: The base name of "/home/user/hello.c" is "hello.c".) If set, GDB will canonicalize file names (e.g., expand symlinks) before comparing them. Canonicalization is an expensive operation, but it allows the same file be known by more than one base name. If not set (the default), all source files are assumed to have just one base name, and gdb will do file name comparisons more efficiently. set trace-user show trace-user set trace-notes show trace-notes Set a user name and notes for the current and any future trace runs. This is useful for long-running and/or disconnected traces, to inform others (or yourself) as to who is running the trace, supply contact information, or otherwise explain what is going on. set trace-stop-notes show trace-stop-notes Set a note attached to the trace run, that is displayed when the trace has been stopped by a tstop command. This is useful for instance as an explanation, if you are stopping a trace run that was started by someone else. * New remote packets QTEnable Dynamically enable a tracepoint in a started trace experiment. QTDisable Dynamically disable a tracepoint in a started trace experiment. QTNotes Set the user and notes of the trace run. qTP Query the current status of a tracepoint. qTMinFTPILen Query the minimum length of instruction at which a fast tracepoint may be placed. * Dcache size (number of lines) and line-size are now runtime-configurable via "set dcache line" and "set dcache line-size" commands. * New targets Texas Instruments TMS320C6x tic6x-*-* * New Simulators Renesas RL78 rl78-*-elf *** Changes in GDB 7.3.1 * The build failure for NetBSD and OpenBSD targets have now been fixed. *** Changes in GDB 7.3 * GDB has a new command: "thread find [REGEXP]". It finds the thread id whose name, target id, or thread extra info matches the given regular expression. * The "catch syscall" command now works on mips*-linux* targets. * The -data-disassemble MI command now supports modes 2 and 3 for dumping the instruction opcodes. * New command line options -data-directory DIR Specify DIR as the "data-directory". This is mostly for testing purposes. * The "maint set python auto-load on|off" command has been renamed to "set auto-load-scripts on|off". * GDB has a new command: "set directories". It is like the "dir" command except that it replaces the source path list instead of augmenting it. * GDB now understands thread names. On GNU/Linux, "info threads" will display the thread name as set by prctl or pthread_setname_np. There is also a new command, "thread name", which can be used to assign a name internally for GDB to display. * OpenCL C Initial support for the OpenCL C language (http://www.khronos.org/opencl) has been integrated into GDB. * Python scripting ** The function gdb.Write now accepts an optional keyword 'stream'. This keyword, when provided, will direct the output to either stdout, stderr, or GDB's logging output. ** Parameters can now be be sub-classed in Python, and in particular you may implement the get_set_doc and get_show_doc functions. This improves how Parameter set/show documentation is processed and allows for more dynamic content. ** Symbols, Symbol Table, Symbol Table and Line, Object Files, Inferior, Inferior Thread, Blocks, and Block Iterator APIs now have an is_valid method. ** Breakpoints can now be sub-classed in Python, and in particular you may implement a 'stop' function that is executed each time the inferior reaches that breakpoint. ** New function gdb.lookup_global_symbol looks up a global symbol. ** GDB values in Python are now callable if the value represents a function. For example, if 'some_value' represents a function that takes two integer parameters and returns a value, you can call that function like so: result = some_value (10,20) ** Module gdb.types has been added. It contains a collection of utilities for working with gdb.Types objects: get_basic_type, has_field, make_enum_dict. ** Module gdb.printing has been added. It contains utilities for writing and registering pretty-printers. New classes: PrettyPrinter, SubPrettyPrinter, RegexpCollectionPrettyPrinter. New function: register_pretty_printer. ** New commands "info pretty-printers", "enable pretty-printer" and "disable pretty-printer" have been added. ** gdb.parameter("directories") is now available. ** New function gdb.newest_frame returns the newest frame in the selected thread. ** The gdb.InferiorThread class has a new "name" attribute. This holds the thread's name. ** Python Support for Inferior events. Python scripts can add observers to be notified of events occurring in the process being debugged. The following events are currently supported: - gdb.events.cont Continue event. - gdb.events.exited Inferior exited event. - gdb.events.stop Signal received, and Breakpoint hit events. * C++ Improvements: ** GDB now puts template parameters in scope when debugging in an instantiation. For example, if you have: template int func (void) { return X; } then if you step into func<5>, "print X" will show "5". This feature requires proper debuginfo support from the compiler; it was added to GCC 4.5. ** The motion commands "next", "finish", "until", and "advance" now work better when exceptions are thrown. In particular, GDB will no longer lose control of the inferior; instead, the GDB will stop the inferior at the point at which the exception is caught. This functionality requires a change in the exception handling code that was introduced in GCC 4.5. * GDB now follows GCC's rules on accessing volatile objects when reading or writing target state during expression evaluation. One notable difference to prior behavior is that "print x = 0" no longer generates a read of x; the value of the assignment is now always taken directly from the value being assigned. * GDB now has some support for using labels in the program's source in linespecs. For instance, you can use "advance label" to continue execution to a label. * GDB now has support for reading and writing a new .gdb_index section. This section holds a fast index of DWARF debugging information and can be used to greatly speed up GDB startup and operation. See the documentation for `save gdb-index' for details. * The "watch" command now accepts an optional "-location" argument. When used, this causes GDB to watch the memory referred to by the expression. Such a watchpoint is never deleted due to it going out of scope. * GDB now supports thread debugging of core dumps on GNU/Linux. GDB now activates thread debugging using the libthread_db library when debugging GNU/Linux core dumps, similarly to when debugging live processes. As a result, when debugging a core dump file, GDB is now able to display pthread_t ids of threads. For example, "info threads" shows the same output as when debugging the process when it was live. In earlier releases, you'd see something like this: (gdb) info threads * 1 LWP 6780 main () at main.c:10 While now you see this: (gdb) info threads * 1 Thread 0x7f0f5712a700 (LWP 6780) main () at main.c:10 It is also now possible to inspect TLS variables when debugging core dumps. When debugging a core dump generated on a machine other than the one used to run GDB, you may need to point GDB at the correct libthread_db library with the "set libthread-db-search-path" command. See the user manual for more details on this command. * When natively debugging programs on PowerPC BookE processors running a Linux kernel version 2.6.34 or later, GDB supports ranged breakpoints, which stop execution of the inferior whenever it executes an instruction at any address within the specified range. See the "PowerPC Embedded" section in the user manual for more details. * New features in the GDB remote stub, GDBserver ** GDBserver is now supported on PowerPC LynxOS (versions 4.x and 5.x), and i686 LynxOS (version 5.x). ** GDBserver is now supported on Blackfin Linux. * New native configurations ia64 HP-UX ia64-*-hpux* * New targets: Analog Devices, Inc. Blackfin Processor bfin-* * Ada task switching is now supported on sparc-elf targets when debugging a program using the Ravenscar Profile. For more information, see the "Tasking Support when using the Ravenscar Profile" section in the GDB user manual. * Guile support was removed. * New features in the GNU simulator ** The --map-info flag lists all known core mappings. ** CFI flashes may be simulated via the "cfi" device. *** Changes in GDB 7.2 * Shared library support for remote targets by default When GDB is configured for a generic, non-OS specific target, like for example, --target=arm-eabi or one of the many *-*-elf targets, GDB now queries remote stubs for loaded shared libraries using the `qXfer:libraries:read' packet. Previously, shared library support was always disabled for such configurations. * C++ Improvements: ** Argument Dependent Lookup (ADL) In C++ ADL lookup directs function search to the namespaces of its arguments even if the namespace has not been imported. For example: namespace A { class B { }; void foo (B) { } } ... A::B b foo(b) Here the compiler will search for `foo' in the namespace of 'b' and find A::foo. GDB now supports this. This construct is commonly used in the Standard Template Library for operators. ** Improved User Defined Operator Support In addition to member operators, GDB now supports lookup of operators defined in a namespace and imported with a `using' directive, operators defined in the global scope, operators imported implicitly from an anonymous namespace, and the ADL operators mentioned in the previous entry. GDB now also supports proper overload resolution for all the previously mentioned flavors of operators. ** static const class members Printing of static const class members that are initialized in the class definition has been fixed. * Windows Thread Information Block access. On Windows targets, GDB now supports displaying the Windows Thread Information Block (TIB) structure. This structure is visible either by using the new command `info w32 thread-information-block' or, by dereferencing the new convenience variable named `$_tlb', a thread-specific pointer to the TIB. This feature is also supported when remote debugging using GDBserver. * Static tracepoints Static tracepoints are calls in the user program into a tracing library. One such library is a port of the LTTng kernel tracer to userspace --- UST (LTTng Userspace Tracer, http://lttng.org/ust). When debugging with GDBserver, GDB now supports combining the GDB tracepoint machinery with such libraries. For example: the user can use GDB to probe a static tracepoint marker (a call from the user program into the tracing library) with the new "strace" command (see "New commands" below). This creates a "static tracepoint" in the breakpoint list, that can be manipulated with the same feature set as fast and regular tracepoints. E.g., collect registers, local and global variables, collect trace state variables, and define tracepoint conditions. In addition, the user can collect extra static tracepoint marker specific data, by collecting the new $_sdata internal variable. When analyzing the trace buffer, you can inspect $_sdata like any other variable available to GDB. For more information, see the "Tracepoints" chapter in GDB user manual. New remote packets have been defined to support static tracepoints, see the "New remote packets" section below. * Better reconstruction of tracepoints after disconnected tracing GDB will attempt to download the original source form of tracepoint definitions when starting a trace run, and then will upload these upon reconnection to the target, resulting in a more accurate reconstruction of the tracepoints that are in use on the target. * Observer mode You can now exercise direct control over the ways that GDB can affect your program. For instance, you can disallow the setting of breakpoints, so that the program can run continuously (assuming non-stop mode). In addition, the "observer" variable is available to switch all of the different controls; in observer mode, GDB cannot affect the target's behavior at all, which is useful for tasks like diagnosing live systems in the field. * The new convenience variable $_thread holds the number of the current thread. * New remote packets qGetTIBAddr Return the address of the Windows Thread Information Block of a given thread. qRelocInsn In response to several of the tracepoint packets, the target may now also respond with a number of intermediate `qRelocInsn' request packets before the final result packet, to have GDB handle relocating an instruction to execute at a different address. This is particularly useful for stubs that support fast tracepoints. GDB reports support for this feature in the qSupported packet. qTfSTM, qTsSTM List static tracepoint markers in the target program. qTSTMat List static tracepoint markers at a given address in the target program. qXfer:statictrace:read Read the static trace data collected (by a `collect $_sdata' tracepoint action). The remote stub reports support for this packet to gdb's qSupported query. QAllow Send the current settings of GDB's permission flags. QTDPsrc Send part of the source (textual) form of a tracepoint definition, which includes location, conditional, and action list. * The source command now accepts a -s option to force searching for the script in the source search path even if the script name specifies a directory. * New features in the GDB remote stub, GDBserver - GDBserver now support tracepoints (including fast tracepoints, and static tracepoints). The feature is currently supported by the i386-linux and amd64-linux builds. See the "Tracepoints support in gdbserver" section in the manual for more information. GDBserver JIT compiles the tracepoint's conditional agent expression bytecode into native code whenever possible for low overhead dynamic tracepoints conditionals. For such tracepoints, an expression that examines program state is evaluated when the tracepoint is reached, in order to determine whether to capture trace data. If the condition is simple and false, processing the tracepoint finishes very quickly and no data is gathered. GDBserver interfaces with the UST (LTTng Userspace Tracer) library for static tracepoints support. - GDBserver now supports x86_64 Windows 64-bit debugging. * GDB now sends xmlRegisters= in qSupported packet to indicate that it understands register description. * The --batch flag now disables pagination and queries. * X86 general purpose registers GDB now supports reading/writing byte, word and double-word x86 general purpose registers directly. This means you can use, say, $ah or $ax to refer, respectively, to the byte register AH and 16-bit word register AX that are actually portions of the 32-bit register EAX or 64-bit register RAX. * The `commands' command now accepts a range of breakpoints to modify. A plain `commands' following a command that creates multiple breakpoints affects all the breakpoints set by that command. This applies to breakpoints set by `rbreak', and also applies when a single `break' command creates multiple breakpoints (e.g., breakpoints on overloaded c++ functions). * The `rbreak' command now accepts a filename specification as part of its argument, limiting the functions selected by the regex to those in the specified file. * Support for remote debugging Windows and SymbianOS shared libraries from Unix hosts has been improved. Non Windows GDB builds now can understand target reported file names that follow MS-DOS based file system semantics, such as file names that include drive letters and use the backslash character as directory separator. This makes it possible to transparently use the "set sysroot" and "set solib-search-path" on Unix hosts to point as host copies of the target's shared libraries. See the new command "set target-file-system-kind" described below, and the "Commands to specify files" section in the user manual for more information. * New commands eval template, expressions... Convert the values of one or more expressions under the control of the string template to a command line, and call it. set target-file-system-kind unix|dos-based|auto show target-file-system-kind Set or show the assumed file system kind for target reported file names. save breakpoints Save all current breakpoint definitions to a file suitable for use in a later debugging session. To read the saved breakpoint definitions, use the `source' command. `save tracepoints' is a new alias for `save-tracepoints'. The latter is now deprecated. info static-tracepoint-markers Display information about static tracepoint markers in the target. strace FN | FILE:LINE | *ADDR | -m MARKER_ID Define a static tracepoint by probing a marker at the given function, line, address, or marker ID. set observer on|off show observer Enable and disable observer mode. set may-write-registers on|off set may-write-memory on|off set may-insert-breakpoints on|off set may-insert-tracepoints on|off set may-insert-fast-tracepoints on|off set may-interrupt on|off Set individual permissions for GDB effects on the target. Note that some of these settings can have undesirable or surprising consequences, particularly when changed in the middle of a session. For instance, disabling the writing of memory can prevent breakpoints from being inserted, cause single-stepping to fail, or even crash your program, if you disable after breakpoints have been inserted. However, GDB should not crash. set record memory-query on|off show record memory-query Control whether to stop the inferior if memory changes caused by an instruction cannot be recorded. * Changed commands disassemble The disassemble command now supports "start,+length" form of two arguments. * Python scripting ** GDB now provides a new directory location, called the python directory, where Python scripts written for GDB can be installed. The location of that directory is /python, where is the GDB data directory. For more details, see section `Scripting GDB using Python' in the manual. ** The GDB Python API now has access to breakpoints, symbols, symbol tables, program spaces, inferiors, threads and frame's code blocks. Additionally, GDB Parameters can now be created from the API, and manipulated via set/show in the CLI. ** New functions gdb.target_charset, gdb.target_wide_charset, gdb.progspaces, gdb.current_progspace, and gdb.string_to_argv. ** New exception gdb.GdbError. ** Pretty-printers are now also looked up in the current program space. ** Pretty-printers can now be individually enabled and disabled. ** GDB now looks for names of Python scripts to auto-load in a special section named `.debug_gdb_scripts', in addition to looking for a OBJFILE-gdb.py script when OBJFILE is read by the debugger. * Tracepoint actions were unified with breakpoint commands. In particular, there are no longer differences in "info break" output for breakpoints and tracepoints and the "commands" command can be used for both tracepoints and regular breakpoints. * New targets ARM Symbian arm*-*-symbianelf* * D language support. GDB now supports debugging programs written in the D programming language. * GDB now supports the extended ptrace interface for PowerPC which is available since Linux kernel version 2.6.34. This automatically enables any hardware breakpoints and additional hardware watchpoints available in the processor. The old ptrace interface exposes just one hardware watchpoint and no hardware breakpoints. * GDB is now able to use the Data Value Compare (DVC) register available on embedded PowerPC processors to implement in hardware simple watchpoint conditions of the form: watch ADDRESS|VARIABLE if ADDRESS|VARIABLE == CONSTANT EXPRESSION This works in native GDB running on Linux kernels with the extended ptrace interface mentioned above. *** Changes in GDB 7.1 * C++ Improvements ** Namespace Support GDB now supports importing of namespaces in C++. This enables the user to inspect variables from imported namespaces. Support for namepace aliasing has also been added. So, if a namespace is aliased in the current scope (e.g. namepace C=A; ) the user can print variables using the alias (e.g. (gdb) print C::x). ** Bug Fixes All known bugs relating to the printing of virtual base class were fixed. It is now possible to call overloaded static methods using a qualified name. ** Cast Operators The C++ cast operators static_cast<>, dynamic_cast<>, const_cast<>, and reinterpret_cast<> are now handled by the C++ expression parser. * New targets Xilinx MicroBlaze microblaze-*-* Renesas RX rx-*-elf * New Simulators Xilinx MicroBlaze microblaze Renesas RX rx * Multi-program debugging. GDB now has support for multi-program (a.k.a. multi-executable or multi-exec) debugging. This allows for debugging multiple inferiors simultaneously each running a different program under the same GDB session. See "Debugging Multiple Inferiors and Programs" in the manual for more information. This implied some user visible changes in the multi-inferior support. For example, "info inferiors" now lists inferiors that are not running yet or that have exited already. See also "New commands" and "New options" below. * New tracing features GDB's tracepoint facility now includes several new features: ** Trace state variables GDB tracepoints now include support for trace state variables, which are variables managed by the target agent during a tracing experiment. They are useful for tracepoints that trigger each other, so for instance one tracepoint can count hits in a variable, and then a second tracepoint has a condition that is true when the count reaches a particular value. Trace state variables share the $-syntax of GDB convenience variables, and can appear in both tracepoint actions and condition expressions. Use the "tvariable" command to create, and "info tvariables" to view; see "Trace State Variables" in the manual for more detail. ** Fast tracepoints GDB now includes an option for defining fast tracepoints, which targets may implement more efficiently, such as by installing a jump into the target agent rather than a trap instruction. The resulting speedup can be by two orders of magnitude or more, although the tradeoff is that some program locations on some target architectures might not allow fast tracepoint installation, for instance if the instruction to be replaced is shorter than the jump. To request a fast tracepoint, use the "ftrace" command, with syntax identical to the regular trace command. ** Disconnected tracing It is now possible to detach GDB from the target while it is running a trace experiment, then reconnect later to see how the experiment is going. In addition, a new variable disconnected-tracing lets you tell the target agent whether to continue running a trace if the connection is lost unexpectedly. ** Trace files GDB now has the ability to save the trace buffer into a file, and then use that file as a target, similarly to you can do with corefiles. You can select trace frames, print data that was collected in them, and use tstatus to display the state of the tracing run at the moment that it was saved. To create a trace file, use "tsave ", and to use it, do "target tfile ". ** Circular trace buffer You can ask the target agent to handle the trace buffer as a circular buffer, discarding the oldest trace frames to make room for newer ones, by setting circular-trace-buffer to on. This feature may not be available for all target agents. * Changed commands disassemble The disassemble command, when invoked with two arguments, now requires the arguments to be comma-separated. info variables The info variables command now displays variable definitions. Files which only declare a variable are not shown. source The source command is now capable of sourcing Python scripts. This feature is dependent on the debugger being build with Python support. Related to this enhancement is also the introduction of a new command "set script-extension" (see below). * New commands (for set/show, see "New options" below) record save [] Save a file (in core file format) containing the process record execution log for replay debugging at a later time. record restore Restore the process record execution log that was saved at an earlier time, for replay debugging. add-inferior [-copies ] [-exec ] Add a new inferior. clone-inferior [-copies ] [ID] Make a new inferior ready to execute the same program another inferior has loaded. remove-inferior ID Remove an inferior. maint info program-spaces List the program spaces loaded into GDB. set remote interrupt-sequence [Ctrl-C | BREAK | BREAK-g] show remote interrupt-sequence Allow the user to select one of ^C, a BREAK signal or BREAK-g as the sequence to the remote target in order to interrupt the execution. Ctrl-C is a default. Some system prefers BREAK which is high level of serial line for some certain time. Linux kernel prefers BREAK-g, a.k.a Magic SysRq g. It is BREAK signal and character 'g'. set remote interrupt-on-connect [on | off] show remote interrupt-on-connect When interrupt-on-connect is ON, gdb sends interrupt-sequence to remote target when gdb connects to it. This is needed when you debug Linux kernel. set remotebreak [on | off] show remotebreak Deprecated. Use "set/show remote interrupt-sequence" instead. tvariable $NAME [ = EXP ] Create or modify a trace state variable. info tvariables List trace state variables and their values. delete tvariable $NAME ... Delete one or more trace state variables. teval EXPR, ... Evaluate the given expressions without collecting anything into the trace buffer. (Valid in tracepoint actions only.) ftrace FN / FILE:LINE / *ADDR Define a fast tracepoint at the given function, line, or address. * New expression syntax GDB now parses the 0b prefix of binary numbers the same way as GCC does. GDB now parses 0b101010 identically with 42. * New options set follow-exec-mode new|same show follow-exec-mode Control whether GDB reuses the same inferior across an exec call or creates a new one. This is useful to be able to restart the old executable after the inferior having done an exec call. set default-collect EXPR, ... show default-collect Define a list of expressions to be collected at each tracepoint. This is a useful way to ensure essential items are not overlooked, such as registers or a critical global variable. set disconnected-tracing show disconnected-tracing If set to 1, the target is instructed to continue tracing if it loses its connection to GDB. If 0, the target is to stop tracing upon disconnection. set circular-trace-buffer show circular-trace-buffer If set to on, the target is instructed to use a circular trace buffer and discard the oldest trace frames instead of stopping the trace due to a full trace buffer. If set to off, the trace stops when the buffer fills up. Some targets may not support this. set script-extension off|soft|strict show script-extension If set to "off", the debugger does not perform any script language recognition, and all sourced files are assumed to be GDB scripts. If set to "soft" (the default), files are sourced according to filename extension, falling back to GDB scripts if the first evaluation failed. If set to "strict", files are sourced according to filename extension. set ada trust-PAD-over-XVS on|off show ada trust-PAD-over-XVS If off, activate a workaround against a bug in the debugging information generated by the compiler for PAD types (see gcc/exp_dbug.ads in the GCC sources for more information about the GNAT encoding and PAD types in particular). It is always safe to set this option to off, but this introduces a slight performance penalty. The default is on. * Python API Improvements ** GDB provides the new class gdb.LazyString. This is useful in some pretty-printing cases. The new method gdb.Value.lazy_string provides a simple way to create objects of this type. ** The fields returned by gdb.Type.fields now have an `is_base_class' attribute. ** The new method gdb.Type.range returns the range of an array type. ** The new method gdb.parse_and_eval can be used to parse and evaluate an expression. * New remote packets QTDV Define a trace state variable. qTV Get the current value of a trace state variable. QTDisconnected Set desired tracing behavior upon disconnection. QTBuffer:circular Set the trace buffer to be linear or circular. qTfP, qTsP Get data about the tracepoints currently in use. * Bug fixes Process record now works correctly with hardware watchpoints. Multiple bug fixes have been made to the mips-irix port, making it much more reliable. In particular: - Debugging threaded applications is now possible again. Previously, GDB would hang while starting the program, or while waiting for the program to stop at a breakpoint. - Attaching to a running process no longer hangs. - An error occurring while loading a core file has been fixed. - Changing the value of the PC register now works again. This fixes problems observed when using the "jump" command, or when calling a function from GDB, or even when assigning a new value to $pc. - With the "finish" and "return" commands, the return value for functions returning a small array is now correctly printed. - It is now possible to break on shared library code which gets executed during a shared library init phase (code executed while executing their .init section). Previously, the breakpoint would have no effect. - GDB is now able to backtrace through the signal handler for non-threaded programs. PIE (Position Independent Executable) programs debugging is now supported. This includes debugging execution of PIC (Position Independent Code) shared libraries although for that, it should be possible to run such libraries as an executable program. *** Changes in GDB 7.0 * GDB now has an interface for JIT compilation. Applications that dynamically generate code can create symbol files in memory and register them with GDB. For users, the feature should work transparently, and for JIT developers, the interface is documented in the GDB manual in the "JIT Compilation Interface" chapter. * Tracepoints may now be conditional. The syntax is as for breakpoints; either an "if" clause appended to the "trace" command, or the "condition" command is available. GDB sends the condition to the target for evaluation using the same bytecode format as is used for tracepoint actions. * The disassemble command now supports: an optional /r modifier, print the raw instructions in hex as well as in symbolic form, and an optional /m modifier to print mixed source+assembly. * Process record and replay In a architecture environment that supports ``process record and replay'', ``process record and replay'' target can record a log of the process execution, and replay it with both forward and reverse execute commands. * Reverse debugging: GDB now has new commands reverse-continue, reverse- step, reverse-next, reverse-finish, reverse-stepi, reverse-nexti, and set execution-direction {forward|reverse}, for targets that support reverse execution. * GDB now supports hardware watchpoints on MIPS/Linux systems. This feature is available with a native GDB running on kernel version 2.6.28 or later. * GDB now has support for multi-byte and wide character sets on the target. Strings whose character type is wchar_t, char16_t, or char32_t are now correctly printed. GDB supports wide- and unicode- literals in C, that is, L'x', L"string", u'x', u"string", U'x', and U"string" syntax. And, GDB allows the "%ls" and "%lc" formats in `printf'. This feature requires iconv to work properly; if your system does not have a working iconv, GDB can use GNU libiconv. See the installation instructions for more information. * GDB now supports automatic retrieval of shared library files from remote targets. To use this feature, specify a system root that begins with the `remote:' prefix, either via the `set sysroot' command or via the `--with-sysroot' configure-time option. * "info sharedlibrary" now takes an optional regex of libraries to show, and it now reports if a shared library has no debugging information. * Commands `set debug-file-directory', `set solib-search-path' and `set args' now complete on file names. * When completing in expressions, gdb will attempt to limit completions to allowable structure or union fields, where appropriate. For instance, consider: # struct example { int f1; double f2; }; # struct example variable; (gdb) p variable. If the user types TAB at the end of this command line, the available completions will be "f1" and "f2". * Inlined functions are now supported. They show up in backtraces, and the "step", "next", and "finish" commands handle them automatically. * GDB now supports the token-splicing (##) and stringification (#) operators when expanding macros. It also supports variable-arity macros. * GDB now supports inspecting extra signal information, exported by the new $_siginfo convenience variable. The feature is currently implemented on linux ARM, i386 and amd64. * GDB can now display the VFP floating point registers and NEON vector registers on ARM targets. Both ARM GNU/Linux native GDB and gdbserver can provide these registers (requires Linux 2.6.30 or later). Remote and simulator targets may also provide them. * New remote packets qSearch:memory: Search memory for a sequence of bytes. QStartNoAckMode Turn off `+'/`-' protocol acknowledgments to permit more efficient operation over reliable transport links. Use of this packet is controlled by the `set remote noack-packet' command. vKill Kill the process with the specified process ID. Use this in preference to `k' when multiprocess protocol extensions are supported. qXfer:osdata:read Obtains additional operating system information qXfer:siginfo:read qXfer:siginfo:write Read or write additional signal information. * Removed remote protocol undocumented extension An undocumented extension to the remote protocol's `S' stop reply packet that permitted the stub to pass a process id was removed. Remote servers should use the `T' stop reply packet instead. * GDB now supports multiple function calling conventions according to the DWARF-2 DW_AT_calling_convention function attribute. * The SH target utilizes the aforementioned change to distinguish between gcc and Renesas calling convention. It also adds the new CLI commands `set/show sh calling-convention'. * GDB can now read compressed debug sections, as produced by GNU gold with the --compress-debug-sections=zlib flag. * 64-bit core files are now supported on AIX. * Thread switching is now supported on Tru64. * Watchpoints can now be set on unreadable memory locations, e.g. addresses which will be allocated using malloc later in program execution. * The qXfer:libraries:read remote protocol packet now allows passing a list of section offsets. * On GNU/Linux, GDB can now attach to stopped processes. Several race conditions handling signals delivered during attach or thread creation have also been fixed. * GDB now supports the use of DWARF boolean types for Ada's type Boolean. From the user's standpoint, all unqualified instances of True and False are treated as the standard definitions, regardless of context. * GDB now parses C++ symbol and type names more flexibly. For example, given: template class C { }; C c; GDB will now correctly handle all of: ptype C ptype C ptype C ptype C * New features in the GDB remote stub, gdbserver - The "--wrapper" command-line argument tells gdbserver to use a wrapper program to launch programs for debugging. - On PowerPC and S/390 targets, it is now possible to use a single gdbserver executable to debug both 32-bit and 64-bit programs. (This requires gdbserver itself to be built as a 64-bit executable.) - gdbserver uses the new noack protocol mode for TCP connections to reduce communications latency, if also supported and enabled in GDB. - Support for the sparc64-linux-gnu target is now included in gdbserver. - The amd64-linux build of gdbserver now supports debugging both 32-bit and 64-bit programs. - The i386-linux, amd64-linux, and i386-win32 builds of gdbserver now support hardware watchpoints, and will use them automatically as appropriate. * Python scripting GDB now has support for scripting using Python. Whether this is available is determined at configure time. New GDB commands can now be written in Python. * Ada tasking support Ada tasks can now be inspected in GDB. The following commands have been introduced: info tasks Print the list of Ada tasks. info task N Print detailed information about task number N. task Print the task number of the current task. task N Switch the context of debugging to task number N. * Support for user-defined prefixed commands. The "define" command can add new commands to existing prefixes, e.g. "target". * Multi-inferior, multi-process debugging. GDB now has generalized support for multi-inferior debugging. See "Debugging Multiple Inferiors" in the manual for more information. Although availability still depends on target support, the command set is more uniform now. The GNU/Linux specific multi-forks support has been migrated to this new framework. This implied some user visible changes; see "New commands" and also "Removed commands" below. * Target descriptions can now describe the target OS ABI. See the "Target Description Format" section in the user manual for more information. * Target descriptions can now describe "compatible" architectures to indicate that the target can execute applications for a different architecture in addition to those for the main target architecture. See the "Target Description Format" section in the user manual for more information. * Multi-architecture debugging. GDB now includes general supports for debugging applications on hybrid systems that use more than one single processor architecture at the same time. Each such hybrid architecture still requires specific support to be added. The only hybrid architecture supported in this version of GDB is the Cell Broadband Engine. * GDB now supports integrated debugging of Cell/B.E. applications that use both the PPU and SPU architectures. To enable support for hybrid Cell/B.E. debugging, you need to configure GDB to support both the powerpc-linux or powerpc64-linux and the spu-elf targets, using the --enable-targets configure option. * Non-stop mode debugging. For some targets, GDB now supports an optional mode of operation in which you can examine stopped threads while other threads continue to execute freely. This is referred to as non-stop mode, with the old mode referred to as all-stop mode. See the "Non-Stop Mode" section in the user manual for more information. To be able to support remote non-stop debugging, a remote stub needs to implement the non-stop mode remote protocol extensions, as described in the "Remote Non-Stop" section of the user manual. The GDB remote stub, gdbserver, has been adjusted to support these extensions on linux targets. * New commands (for set/show, see "New options" below) catch syscall [NAME(S) | NUMBER(S)] Catch system calls. Arguments, which should be names of system calls or their numbers, mean catch only those syscalls. Without arguments, every syscall will be caught. When the inferior issues any of the specified syscalls, GDB will stop and announce the system call, both when it is called and when its call returns. This feature is currently available with a native GDB running on the Linux Kernel, under the following architectures: x86, x86_64, PowerPC and PowerPC64. find [/size-char] [/max-count] start-address, end-address|+search-space-size, val1 [, val2, ...] Search memory for a sequence of bytes. maint set python print-stack maint show python print-stack Show a stack trace when an error is encountered in a Python script. python [CODE] Invoke CODE by passing it to the Python interpreter. macro define macro list macro undef These allow macros to be defined, undefined, and listed interactively. info os processes Show operating system information about processes. info inferiors List the inferiors currently under GDB's control. inferior NUM Switch focus to inferior number NUM. detach inferior NUM Detach from inferior number NUM. kill inferior NUM Kill inferior number NUM. * New options set spu stop-on-load show spu stop-on-load Control whether to stop for new SPE threads during Cell/B.E. debugging. set spu auto-flush-cache show spu auto-flush-cache Control whether to automatically flush the software-managed cache during Cell/B.E. debugging. set sh calling-convention show sh calling-convention Control the calling convention used when calling SH target functions. set debug timestamp show debug timestamp Control display of timestamps with GDB debugging output. set disassemble-next-line show disassemble-next-line Control display of disassembled source lines or instructions when the debuggee stops. set remote noack-packet show remote noack-packet Set/show the use of remote protocol QStartNoAckMode packet. See above under "New remote packets." set remote query-attached-packet show remote query-attached-packet Control use of remote protocol `qAttached' (query-attached) packet. set remote read-siginfo-object show remote read-siginfo-object Control use of remote protocol `qXfer:siginfo:read' (read-siginfo-object) packet. set remote write-siginfo-object show remote write-siginfo-object Control use of remote protocol `qXfer:siginfo:write' (write-siginfo-object) packet. set remote reverse-continue show remote reverse-continue Control use of remote protocol 'bc' (reverse-continue) packet. set remote reverse-step show remote reverse-step Control use of remote protocol 'bs' (reverse-step) packet. set displaced-stepping show displaced-stepping Control displaced stepping mode. Displaced stepping is a way to single-step over breakpoints without removing them from the debuggee. Also known as "out-of-line single-stepping". set debug displaced show debug displaced Control display of debugging info for displaced stepping. maint set internal-error maint show internal-error Control what GDB does when an internal error is detected. maint set internal-warning maint show internal-warning Control what GDB does when an internal warning is detected. set exec-wrapper show exec-wrapper unset exec-wrapper Use a wrapper program to launch programs for debugging. set multiple-symbols (all|ask|cancel) show multiple-symbols The value of this variable can be changed to adjust the debugger behavior when an expression or a breakpoint location contains an ambiguous symbol name (an overloaded function name, for instance). set breakpoint always-inserted show breakpoint always-inserted Keep breakpoints always inserted in the target, as opposed to inserting them when resuming the target, and removing them when the target stops. This option can improve debugger performance on slow remote targets. set arm fallback-mode (arm|thumb|auto) show arm fallback-mode set arm force-mode (arm|thumb|auto) show arm force-mode These commands control how ARM GDB determines whether instructions are ARM or Thumb. The default for both settings is auto, which uses the current CPSR value for instructions without symbols; previous versions of GDB behaved as if "set arm fallback-mode arm". set arm unwind-secure-frames Enable unwinding from Non-secure to Secure mode on Cortex-M with Security extension. This can trigger security exceptions when unwinding exception stacks. set disable-randomization show disable-randomization Standalone programs run with the virtual address space randomization enabled by default on some platforms. This option keeps the addresses stable across multiple debugging sessions. set non-stop show non-stop Control whether other threads are stopped or not when some thread hits a breakpoint. set target-async show target-async Requests that asynchronous execution is enabled in the target, if available. In this case, it's possible to resume target in the background, and interact with GDB while the target is running. "show target-async" displays the current state of asynchronous execution of the target. set target-wide-charset show target-wide-charset The target-wide-charset is the name of the character set that GDB uses when printing characters whose type is wchar_t. set tcp auto-retry (on|off) show tcp auto-retry set tcp connect-timeout show tcp connect-timeout These commands allow GDB to retry failed TCP connections to a remote stub with a specified timeout period; this is useful if the stub is launched in parallel with GDB but may not be ready to accept connections immediately. set libthread-db-search-path show libthread-db-search-path Control list of directories which GDB will search for appropriate libthread_db. set schedule-multiple (on|off) show schedule-multiple Allow GDB to resume all threads of all processes or only threads of the current process. set stack-cache show stack-cache Use more aggressive caching for accesses to the stack. This improves performance of remote debugging (particularly backtraces) without affecting correctness. set interactive-mode (on|off|auto) show interactive-mode Control whether GDB runs in interactive mode (on) or not (off). When in interactive mode, GDB waits for the user to answer all queries. Otherwise, GDB does not wait and assumes the default answer. When set to auto (the default), GDB determines which mode to use based on the stdin settings. * Removed commands info forks For program forks, this is replaced by the new more generic `info inferiors' command. To list checkpoints, you can still use the `info checkpoints' command, which was an alias for the `info forks' command. fork NUM Replaced by the new `inferior' command. To switch between checkpoints, you can still use the `restart' command, which was an alias for the `fork' command. process PID This is removed, since some targets don't have a notion of processes. To switch between processes, you can still use the `inferior' command using GDB's own inferior number. delete fork NUM For program forks, this is replaced by the new more generic `kill inferior' command. To delete a checkpoint, you can still use the `delete checkpoint' command, which was an alias for the `delete fork' command. detach fork NUM For program forks, this is replaced by the new more generic `detach inferior' command. To detach a checkpoint, you can still use the `detach checkpoint' command, which was an alias for the `detach fork' command. * New native configurations x86/x86_64 Darwin i[34567]86-*-darwin* x86_64 MinGW x86_64-*-mingw* * New targets Lattice Mico32 lm32-* x86 DICOS i[34567]86-*-dicos* x86_64 DICOS x86_64-*-dicos* S+core 3 score-*-* * The GDB remote stub, gdbserver, now supports x86 Windows CE (mingw32ce) debugging. * Removed commands catch load catch unload These commands were actually not implemented on any target. *** Changes in GDB 6.8 * New native configurations NetBSD/hppa hppa*-*netbsd* Xtensa GNU/Linux xtensa*-*-linux* * New targets NetBSD/hppa hppa*-*-netbsd* Xtensa GNU/Linux xtensa*-*-linux* * Change in command line behavior -- corefiles vs. process ids. When the '-p NUMBER' or '--pid NUMBER' options are used, and attaching to process NUMBER fails, GDB no longer attempts to open a core file named NUMBER. Attaching to a program using the -c option is no longer supported. Instead, use the '-p' or '--pid' options. * GDB can now be built as a native debugger for debugging Windows x86 (mingw32) Portable Executable (PE) programs. * Pending breakpoints no longer change their number when their address is resolved. * GDB now supports breakpoints with multiple locations, including breakpoints on C++ constructors, inside C++ templates, and in inlined functions. * GDB's ability to debug optimized code has been improved. GDB more accurately identifies function bodies and lexical blocks that occupy more than one contiguous range of addresses. * Target descriptions can now describe registers for PowerPC. * The GDB remote stub, gdbserver, now supports the AltiVec and SPE registers on PowerPC targets. * The GDB remote stub, gdbserver, now supports thread debugging on GNU/Linux targets even when the libthread_db library is not available. * The GDB remote stub, gdbserver, now supports the new file transfer commands (remote put, remote get, and remote delete). * The GDB remote stub, gdbserver, now supports run and attach in extended-remote mode. * hppa*64*-*-hpux11* target broken The debugger is unable to start a program and fails with the following error: "Error trying to get information about dynamic linker". The gdb-6.7 release is also affected. * GDB now supports the --enable-targets= configure option to allow building a single GDB executable that supports multiple remote target architectures. * GDB now supports debugging C and C++ programs which use the Decimal Floating Point extension. In addition, the PowerPC target now has a set of pseudo-registers to inspect decimal float values stored in two consecutive float registers. * The -break-insert MI command can optionally create pending breakpoints now. * Improved support for debugging Ada Many improvements to the Ada language support have been made. These include: - Better support for Ada2005 interface types - Improved handling of arrays and slices in general - Better support for Taft-amendment types - The '{type} ADDRESS' expression is now allowed on the left hand-side of an assignment - Improved command completion in Ada - Several bug fixes * GDB on GNU/Linux and HP/UX can now debug through "exec" of a new process. * New commands set print frame-arguments (all|scalars|none) show print frame-arguments The value of this variable can be changed to control which argument values should be printed by the debugger when displaying a frame. remote put remote get remote delete Transfer files to and from a remote target, and delete remote files. * New MI commands -target-file-put -target-file-get -target-file-delete Transfer files to and from a remote target, and delete remote files. * New remote packets vFile:open: vFile:close: vFile:pread: vFile:pwrite: vFile:unlink: Open, close, read, write, and delete files on the remote system. vAttach Attach to an existing process on the remote system, in extended-remote mode. vRun Run a new process on the remote system, in extended-remote mode. *** Changes in GDB 6.7 * Resolved 101 resource leaks, null pointer dereferences, etc. in gdb, bfd, libiberty and opcodes, as revealed by static analysis donated by Coverity, Inc. (http://scan.coverity.com). * When looking up multiply-defined global symbols, GDB will now prefer the symbol definition in the current shared library if it was built using the -Bsymbolic linker option. * When the Text User Interface (TUI) is not configured, GDB will now recognize the -tui command-line option and print a message that the TUI is not supported. * The GDB remote stub, gdbserver, now has lower overhead for high frequency signals (e.g. SIGALRM) via the QPassSignals packet. * GDB for MIPS targets now autodetects whether a remote target provides 32-bit or 64-bit register values. * Support for C++ member pointers has been improved. * GDB now understands XML target descriptions, which specify the target's overall architecture. GDB can read a description from a local file or over the remote serial protocol. * Vectors of single-byte data use a new integer type which is not automatically displayed as character or string data. * The /s format now works with the print command. It displays arrays of single-byte integers and pointers to single-byte integers as strings. * Target descriptions can now describe target-specific registers, for architectures which have implemented the support (currently only ARM, M68K, and MIPS). * GDB and the GDB remote stub, gdbserver, now support the XScale iWMMXt coprocessor. * The GDB remote stub, gdbserver, has been updated to support ARM Windows CE (mingw32ce) debugging, and GDB Windows CE support has been rewritten to use the standard GDB remote protocol. * GDB can now step into C++ functions which are called through thunks. * GDB for the Cell/B.E. SPU now supports overlay debugging. * The GDB remote protocol "qOffsets" packet can now honor ELF segment layout. It also supports a TextSeg= and DataSeg= response when only segment base addresses (rather than offsets) are available. * The /i format now outputs any trailing branch delay slot instructions immediately following the last instruction within the count specified. * The GDB remote protocol "T" stop reply packet now supports a "library" response. Combined with the new "qXfer:libraries:read" packet, this response allows GDB to debug shared libraries on targets where the operating system manages the list of loaded libraries (e.g. Windows and SymbianOS). * The GDB remote stub, gdbserver, now supports dynamic link libraries (DLLs) on Windows and Windows CE targets. * GDB now supports a faster verification that a .debug file matches its binary according to its build-id signature, if the signature is present. * New commands set remoteflow show remoteflow Enable or disable hardware flow control (RTS/CTS) on the serial port when debugging using remote targets. set mem inaccessible-by-default show mem inaccessible-by-default If the target supplies a memory map, for instance via the remote protocol's "qXfer:memory-map:read" packet, setting this variable prevents GDB from accessing memory outside the memory map. This is useful for targets with memory mapped registers or which react badly to accesses of unmapped address space. set breakpoint auto-hw show breakpoint auto-hw If the target supplies a memory map, for instance via the remote protocol's "qXfer:memory-map:read" packet, setting this variable lets GDB use hardware breakpoints automatically for memory regions where it can not use software breakpoints. This covers both the "break" command and internal breakpoints used for other commands including "next" and "finish". catch exception catch exception unhandled Stop the program execution when Ada exceptions are raised. catch assert Stop the program execution when an Ada assertion failed. set sysroot show sysroot Set an alternate system root for target files. This is a more general version of "set solib-absolute-prefix", which is now an alias to "set sysroot". info spu Provide extended SPU facility status information. This set of commands is available only when debugging the Cell/B.E. SPU architecture. * New native configurations OpenBSD/sh sh*-*openbsd* set tdesc filename unset tdesc filename show tdesc filename Use the specified local file as an XML target description, and do not query the target for its built-in description. * New targets OpenBSD/sh sh*-*-openbsd* MIPS64 GNU/Linux (gdbserver) mips64-linux-gnu Toshiba Media Processor mep-elf * New remote packets QPassSignals: Ignore the specified signals; pass them directly to the debugged program without stopping other threads or reporting them to GDB. qXfer:features:read: Read an XML target description from the target, which describes its features. qXfer:spu:read: qXfer:spu:write: Read or write contents of an spufs file on the target system. These packets are available only on the Cell/B.E. SPU architecture. qXfer:libraries:read: Report the loaded shared libraries. Combined with new "T" packet response, this packet allows GDB to debug shared libraries on targets where the operating system manages the list of loaded libraries (e.g. Windows and SymbianOS). * Removed targets Support for these obsolete configurations has been removed. alpha*-*-osf1* alpha*-*-osf2* d10v-*-* hppa*-*-hiux* i[34567]86-ncr-* i[34567]86-*-dgux* i[34567]86-*-lynxos* i[34567]86-*-netware* i[34567]86-*-sco3.2v5* i[34567]86-*-sco3.2v4* i[34567]86-*-sco* i[34567]86-*-sysv4.2* i[34567]86-*-sysv4* i[34567]86-*-sysv5* i[34567]86-*-unixware2* i[34567]86-*-unixware* i[34567]86-*-sysv* i[34567]86-*-isc* m68*-cisco*-* m68*-tandem-* mips*-*-pe rs6000-*-lynxos* sh*-*-pe * Other removed features target abug target cpu32bug target est target rom68k Various m68k-only ROM monitors. target hms target e7000 target sh3 target sh3e Various Renesas ROM monitors and debugging interfaces for SH and H8/300. target ocd Support for a Macraigor serial interface to on-chip debugging. GDB does not directly support the newer parallel or USB interfaces. DWARF 1 support A debug information format. The predecessor to DWARF 2 and DWARF 3, which are still supported. Support for the HP aCC compiler on HP-UX/PA-RISC SOM-encapsulated symbolic debugging information, automatic invocation of pxdb, and the aCC custom C++ ABI. This does not affect HP-UX for Itanium or GCC for HP-UX/PA-RISC. Code compiled with aCC can still be debugged on an assembly level. MIPS ".pdr" sections A MIPS-specific format used to describe stack frame layout in debugging information. Scheme support GDB could work with an older version of Guile to debug the interpreter and Scheme programs running in it. set mips stack-arg-size set mips saved-gpreg-size Use "set mips abi" to control parameter passing for MIPS. *** Changes in GDB 6.6 * New targets Xtensa xtensa-elf Cell Broadband Engine SPU spu-elf * GDB can now be configured as a cross-debugger targeting native Windows (mingw32) or Cygwin. It can communicate with a remote debugging stub running on a Windows system over TCP/IP to debug Windows programs. * The GDB remote stub, gdbserver, has been updated to support Windows and Cygwin debugging. Both single-threaded and multi-threaded programs are supported. * The "set trust-readonly-sections" command works again. This command was broken in GDB 6.3, 6.4, and 6.5. * The "load" command now supports writing to flash memory, if the remote stub provides the required support. * Support for GNU/Linux Thread Local Storage (TLS, per-thread variables) no longer requires symbolic debug information (e.g. DWARF-2). * New commands set substitute-path unset substitute-path show substitute-path Manage a list of substitution rules that GDB uses to rewrite the name of the directories where the sources are located. This can be useful for instance when the sources were moved to a different location between compilation and debugging. set trace-commands show trace-commands Print each CLI command as it is executed. Each command is prefixed with a number of `+' symbols representing the nesting depth. The source command now has a `-v' option to enable the same feature. * REMOVED features The ARM Demon monitor support (RDP protocol, "target rdp"). Kernel Object Display, an embedded debugging feature which only worked with an obsolete version of Cisco IOS. The 'set download-write-size' and 'show download-write-size' commands. * New remote packets qSupported: Tell a stub about GDB client features, and request remote target features. The first feature implemented is PacketSize, which allows the target to specify the size of packets it can handle - to minimize the number of packets required and improve performance when connected to a remote target. qXfer:auxv:read: Fetch an OS auxilliary vector from the remote stub. This packet is a more efficient replacement for qPart:auxv:read. qXfer:memory-map:read: Fetch a memory map from the remote stub, including information about RAM, ROM, and flash memory devices. vFlashErase: vFlashWrite: vFlashDone: Erase and program a flash memory device. * Removed remote packets qPart:auxv:read: This packet has been replaced by qXfer:auxv:read. Only GDB 6.4 and 6.5 used it, and only gdbserver implemented it. *** Changes in GDB 6.5 * New targets Renesas M32C/M16C m32c-elf Morpho Technologies ms1 ms1-elf * New commands init-if-undefined Initialize a convenience variable, but only if it doesn't already have a value. The following commands are presently only implemented for native GNU/Linux: checkpoint Save a snapshot of the program state. restart Return the program state to a previously saved state. info checkpoints List currently saved checkpoints. delete-checkpoint Delete a previously saved checkpoint. set|show detach-on-fork Tell gdb whether to detach from a newly forked process, or to keep debugging it. info forks List forks of the user program that are available to be debugged. fork Switch to debugging one of several forks of the user program that are available to be debugged. delete-fork Delete a fork from the list of forks that are available to be debugged (and kill the forked process). detach-fork Delete a fork from the list of forks that are available to be debugged (and allow the process to continue). * New architecture Morpho Technologies ms2 ms1-elf * Improved Windows host support GDB now builds as a cross debugger hosted on i686-mingw32, including native console support, and remote communications using either network sockets or serial ports. * Improved Modula-2 language support GDB can now print most types in the Modula-2 syntax. This includes: basic types, set types, record types, enumerated types, range types, pointer types and ARRAY types. Procedure var parameters are correctly printed and hexadecimal addresses and character constants are also written in the Modula-2 syntax. Best results can be obtained by using GNU Modula-2 together with the -gdwarf-2 command line option. * REMOVED features The ARM rdi-share module. The Netware NLM debug server. *** Changes in GDB 6.4 * New native configurations OpenBSD/arm arm*-*-openbsd* OpenBSD/mips64 mips64-*-openbsd* * New targets Morpho Technologies ms1 ms1-elf * New command line options --batch-silent As for --batch, but totally silent. --return-child-result The debugger will exist with the same value the child (debugged) program exited with. --eval-command COMMAND, -ex COMMAND Execute a single GDB CLI command. This may be specified multiple times and in conjunction with the --command (-x) option. * Deprecated commands removed The following commands, that were deprecated in 2000, have been removed: Command Replacement set|show arm disassembly-flavor set|show arm disassembler othernames set arm disassembler set|show remotedebug set|show debug remote set|show archdebug set|show debug arch set|show eventdebug set|show debug event regs info registers * New BSD user-level threads support It is now possible to debug programs using the user-level threads library on OpenBSD and FreeBSD. Currently supported (target) configurations are: FreeBSD/amd64 x86_64-*-freebsd* FreeBSD/i386 i386-*-freebsd* OpenBSD/i386 i386-*-openbsd* Note that the new kernel threads libraries introduced in FreeBSD 5.x are not yet supported. * New support for Matsushita MN10300 w/sim added (Work in progress). mn10300-elf. * REMOVED configurations and files VxWorks and the XDR protocol *-*-vxworks Motorola MCORE mcore-*-* National Semiconductor NS32000 ns32k-*-* * New "set print array-indexes" command After turning this setting "on", GDB prints the index of each element when displaying arrays. The default is "off" to preserve the previous behavior. * VAX floating point support GDB now supports the not-quite-ieee VAX F and D floating point formats. * User-defined command support In addition to using $arg0..$arg9 for argument passing, it is now possible to use $argc to determine now many arguments have been passed. See the section on user-defined commands in the user manual for more information. *** Changes in GDB 6.3: * New command line option GDB now accepts -l followed by a number to set the timeout for remote debugging. * GDB works with GCC -feliminate-dwarf2-dups GDB now supports a more compact representation of DWARF-2 debug information using DW_FORM_ref_addr references. These are produced by GCC with the option -feliminate-dwarf2-dups and also by some proprietary compilers. With GCC, you must use GCC 3.3.4 or later to use -feliminate-dwarf2-dups. * Internationalization When supported by the host system, GDB will be built with internationalization (libintl). The task of marking up the sources is continued, we're looking forward to our first translation. * Ada Initial support for debugging programs compiled with the GNAT implementation of the Ada programming language has been integrated into GDB. In this release, support is limited to expression evaluation. * New native configurations GNU/Linux/m32r m32r-*-linux-gnu * Remote 'p' packet GDB's remote protocol now includes support for the 'p' packet. This packet is used to fetch individual registers from a remote inferior. * END-OF-LIFE registers[] compatibility module GDB's internal register infrastructure has been completely rewritten. The new infrastructure making possible the implementation of key new features including 32x64 (e.g., 64-bit amd64 GDB debugging a 32-bit i386 application). GDB 6.3 will be the last release to include the registers[] compatibility module that allowed out-of-date configurations to continue to work. This change directly impacts the following configurations: hppa-*-hpux ia64-*-aix mips-*-irix* *-*-lynx mips-*-linux-gnu sds protocol xdr protocol powerpc bdm protocol Unless there is activity to revive these configurations, they will be made OBSOLETE in GDB 6.4, and REMOVED from GDB 6.5. * OBSOLETE configurations and files Configurations that have been declared obsolete in this release have been commented out. Unless there is activity to revive these configurations, the next release of GDB will have their sources permanently REMOVED. h8300-*-* mcore-*-* mn10300-*-* ns32k-*-* sh64-*-* v850-*-* *** Changes in GDB 6.2.1: * MIPS `break main; run' gave an heuristic-fence-post warning When attempting to run even a simple program, a warning about heuristic-fence-post being hit would be reported. This problem has been fixed. * MIPS IRIX 'long double' crashed GDB When examining a long double variable, GDB would get a segmentation fault. The crash has been fixed (but GDB 6.2 cannot correctly examine IRIX long double values). * VAX and "next" A bug in the VAX stack code was causing problems with the "next" command. This problem has been fixed. *** Changes in GDB 6.2: * Fix for ``many threads'' On GNU/Linux systems that use the NPTL threads library, a program rapidly creating and deleting threads would confuse GDB leading to the error message: ptrace: No such process. thread_db_get_info: cannot get thread info: generic error This problem has been fixed. * "-async" and "-noasync" options removed. Support for the broken "-noasync" option has been removed (it caused GDB to dump core). * New ``start'' command. This command runs the program until the beginning of the main procedure. * New BSD Kernel Data Access Library (libkvm) interface Using ``target kvm'' it is now possible to debug kernel core dumps and live kernel memory images on various FreeBSD, NetBSD and OpenBSD platforms. Currently supported (native-only) configurations are: FreeBSD/amd64 x86_64-*-freebsd* FreeBSD/i386 i?86-*-freebsd* NetBSD/i386 i?86-*-netbsd* NetBSD/m68k m68*-*-netbsd* NetBSD/sparc sparc-*-netbsd* OpenBSD/amd64 x86_64-*-openbsd* OpenBSD/i386 i?86-*-openbsd* OpenBSD/m68k m68*-openbsd* OpenBSD/sparc sparc-*-openbsd* * Signal trampoline code overhauled Many generic problems with GDB's signal handling code have been fixed. These include: backtraces through non-contiguous stacks; recognition of sa_sigaction signal trampolines; backtrace from a NULL pointer call; backtrace through a signal trampoline; step into and out of signal handlers; and single-stepping in the signal trampoline. Please note that kernel bugs are a limiting factor here. These features have been shown to work on an s390 GNU/Linux system that include a 2.6.8-rc1 kernel. Ref PR breakpoints/1702. * Cygwin support for DWARF 2 added. * New native configurations GNU/Linux/hppa hppa*-*-linux* OpenBSD/hppa hppa*-*-openbsd* OpenBSD/m68k m68*-*-openbsd* OpenBSD/m88k m88*-*-openbsd* OpenBSD/powerpc powerpc-*-openbsd* NetBSD/vax vax-*-netbsd* OpenBSD/vax vax-*-openbsd* * END-OF-LIFE frame compatibility module GDB's internal frame infrastructure has been completely rewritten. The new infrastructure making it possible to support key new features including DWARF 2 Call Frame Information. To aid in the task of migrating old configurations to this new infrastructure, a compatibility module, that allowed old configurations to continue to work, was also included. GDB 6.2 will be the last release to include this frame compatibility module. This change directly impacts the following configurations: h8300-*-* mcore-*-* mn10300-*-* ns32k-*-* sh64-*-* v850-*-* xstormy16-*-* Unless there is activity to revive these configurations, they will be made OBSOLETE in GDB 6.3, and REMOVED from GDB 6.4. * REMOVED configurations and files Sun 3, running SunOS 3 m68*-*-sunos3* Sun 3, running SunOS 4 m68*-*-sunos4* Sun 2, running SunOS 3 m68000-*-sunos3* Sun 2, running SunOS 4 m68000-*-sunos4* Motorola 680x0 running LynxOS m68*-*-lynxos* AT&T 3b1/Unix pc m68*-att-* Bull DPX2 (68k, System V release 3) m68*-bull-sysv* decstation mips-dec-* mips-little-* riscos mips-*-riscos* mips-*-sysv* sonymips mips-sony-* sysv mips*-*-sysv4* (IRIX 5/6 not included) *** Changes in GDB 6.1.1: * TUI (Text-mode User Interface) built-in (also included in GDB 6.1) The TUI (Text-mode User Interface) is now built as part of a default GDB configuration. It is enabled by either selecting the TUI with the command line option "-i=tui" or by running the separate "gdbtui" program. For more information on the TUI, see the manual "Debugging with GDB". * Pending breakpoint support (also included in GDB 6.1) Support has been added to allow you to specify breakpoints in shared libraries that have not yet been loaded. If a breakpoint location cannot be found, and the "breakpoint pending" option is set to auto, GDB queries you if you wish to make the breakpoint pending on a future shared-library load. If and when GDB resolves the breakpoint symbol, the pending breakpoint is removed as one or more regular breakpoints are created. Pending breakpoints are very useful for GCJ Java debugging. * Fixed ISO-C build problems The files bfd/elf-bfd.h, gdb/dictionary.c and gdb/types.c contained non ISO-C code that stopped them being built using a more strict ISO-C compiler (e.g., IBM's C compiler). * Fixed build problem on IRIX 5 Due to header problems with , the file gdb/proc-api.c wasn't able to compile compile on an IRIX 5 system. * Added execute permission to gdb/gdbserver/configure The shell script gdb/testsuite/gdb.stabs/configure lacked execute permission. This bug would cause configure to fail on a number of systems (Solaris, IRIX). Ref: server/519. * Fixed build problem on hpux2.0w-hp-hpux11.00 using the HP ANSI C compiler Older HPUX ANSI C compilers did not accept variable array sizes. somsolib.c has been updated to use constant array sizes. * Fixed a panic in the DWARF Call Frame Info code on Solaris 2.7 GCC 3.3.2, on Solaris 2.7, includes the DW_EH_PE_funcrel encoding in its generated DWARF Call Frame Info. This encoding was causing GDB to panic, that panic has been fixed. Ref: gdb/1628. * Fixed a problem when examining parameters in shared library code. When examining parameters in optimized shared library code generated by a mainline GCC, GDB would incorrectly report ``Variable "..." is not available''. GDB now correctly displays the variable's value. *** Changes in GDB 6.1: * Removed --with-mmalloc Support for the mmalloc memory manager has been removed, as it conflicted with the internal gdb byte cache. * Changes in AMD64 configurations The AMD64 target now includes the %cs and %ss registers. As a result the AMD64 remote protocol has changed; this affects the floating-point and SSE registers. If you rely on those registers for your debugging, you should upgrade gdbserver on the remote side. * Revised SPARC target The SPARC target has been completely revised, incorporating the FreeBSD/sparc64 support that was added for GDB 6.0. As a result support for LynxOS and SunOS 4 has been dropped. Calling functions from within GDB on operating systems with a non-executable stack (Solaris, OpenBSD) now works. * New C++ demangler GDB has a new C++ demangler which does a better job on the mangled names generated by current versions of g++. It also runs faster, so with this and other changes gdb should now start faster on large C++ programs. * DWARF 2 Location Expressions GDB support for location expressions has been extended to support function arguments and frame bases. Older versions of GDB could crash when they encountered these. * C++ nested types and namespaces GDB's support for nested types and namespaces in C++ has been improved, especially if you use the DWARF 2 debugging format. (This is the default for recent versions of GCC on most platforms.) Specifically, if you have a class "Inner" defined within a class or namespace "Outer", then GDB realizes that the class's name is "Outer::Inner", not simply "Inner". This should greatly reduce the frequency of complaints about not finding RTTI symbols. In addition, if you are stopped at inside of a function defined within a namespace, GDB modifies its name lookup accordingly. * New native configurations NetBSD/amd64 x86_64-*-netbsd* OpenBSD/amd64 x86_64-*-openbsd* OpenBSD/alpha alpha*-*-openbsd* OpenBSD/sparc sparc-*-openbsd* OpenBSD/sparc64 sparc64-*-openbsd* * New debugging protocols M32R with SDI protocol m32r-*-elf* * "set prompt-escape-char" command deleted. The command "set prompt-escape-char" has been deleted. This command, and its very obscure effect on GDB's prompt, was never documented, tested, nor mentioned in the NEWS file. * OBSOLETE configurations and files Configurations that have been declared obsolete in this release have been commented out. Unless there is activity to revive these configurations, the next release of GDB will have their sources permanently REMOVED. Sun 3, running SunOS 3 m68*-*-sunos3* Sun 3, running SunOS 4 m68*-*-sunos4* Sun 2, running SunOS 3 m68000-*-sunos3* Sun 2, running SunOS 4 m68000-*-sunos4* Motorola 680x0 running LynxOS m68*-*-lynxos* AT&T 3b1/Unix pc m68*-att-* Bull DPX2 (68k, System V release 3) m68*-bull-sysv* decstation mips-dec-* mips-little-* riscos mips-*-riscos* mips-*-sysv* sonymips mips-sony-* sysv mips*-*-sysv4* (IRIX 5/6 not included) * REMOVED configurations and files SGI Irix-4.x mips-sgi-irix4 or iris4 SGI Iris (MIPS) running Irix V3: mips-sgi-irix or iris Z8000 simulator z8k-zilog-none or z8ksim Matsushita MN10200 w/simulator mn10200-*-* H8/500 simulator h8500-hitachi-hms or h8500hms HP/PA running BSD hppa*-*-bsd* HP/PA running OSF/1 hppa*-*-osf* HP/PA Pro target hppa*-*-pro* PMAX (MIPS) running Mach 3.0 mips*-*-mach3* 386BSD i[3456]86-*-bsd* Sequent family i[3456]86-sequent-sysv4* i[3456]86-sequent-sysv* i[3456]86-sequent-bsd* SPARC running LynxOS sparc-*-lynxos* SPARC running SunOS 4 sparc-*-sunos4* Tsqware Sparclet sparclet-*-* Fujitsu SPARClite sparclite-fujitsu-none or sparclite *** Changes in GDB 6.0: * Objective-C Support for debugging the Objective-C programming language has been integrated into GDB. * New backtrace mechanism (includes DWARF 2 Call Frame Information). DWARF 2's Call Frame Information makes available compiler generated information that more exactly describes the program's run-time stack. By using this information, GDB is able to provide more robust stack backtraces. The i386, amd64 (nee, x86-64), Alpha, m68hc11, ia64, and m32r targets have been updated to use a new backtrace mechanism which includes DWARF 2 CFI support. * Hosted file I/O. GDB's remote protocol has been extended to include support for hosted file I/O (where the remote target uses GDB's file system). See GDB's remote protocol documentation for details. * All targets using the new architecture framework. All of GDB's targets have been updated to use the new internal architecture framework. The way is now open for future GDB releases to include cross-architecture native debugging support (i386 on amd64, ppc32 on ppc64). * GNU/Linux's Thread Local Storage (TLS) GDB now includes support for for the GNU/Linux implementation of per-thread variables. * GNU/Linux's Native POSIX Thread Library (NPTL) GDB's thread code has been updated to work with either the new GNU/Linux NPTL thread library or the older "LinuxThreads" library. * Separate debug info. GDB, in conjunction with BINUTILS, now supports a mechanism for automatically loading debug information from a separate file. Instead of shipping full debug and non-debug versions of system libraries, system integrators can now instead ship just the stripped libraries and optional debug files. * DWARF 2 Location Expressions DWARF 2 Location Expressions allow the compiler to more completely describe the location of variables (even in optimized code) to the debugger. GDB now includes preliminary support for location expressions (support for DW_OP_piece is still missing). * Java A number of long standing bugs that caused GDB to die while starting a Java application have been fixed. GDB's Java support is now considered "useable". * GNU/Linux support for fork, vfork, and exec. The "catch fork", "catch exec", "catch vfork", and "set follow-fork-mode" commands are now implemented for GNU/Linux. They require a 2.5.x or later kernel. * GDB supports logging output to a file There are two new commands, "set logging" and "show logging", which can be used to capture GDB's output to a file. * The meaning of "detach" has changed for gdbserver The "detach" command will now resume the application, as documented. To disconnect from gdbserver and leave it stopped, use the new "disconnect" command. * d10v, m68hc11 `regs' command deprecated The `info registers' command has been updated so that it displays the registers using a format identical to the old `regs' command. * Profiling support A new command, "maint set profile on/off", has been added. This command can be used to enable or disable profiling while running GDB, to profile a session or a set of commands. In addition there is a new configure switch, "--enable-profiling", which will cause GDB to be compiled with profiling data, for more informative profiling results. * Default MI syntax changed to "mi2". The default MI (machine interface) syntax, enabled by the command line option "-i=mi", has been changed to "mi2". The previous MI syntax, "mi1", can be enabled by specifying the option "-i=mi1". Support for the original "mi0" syntax (included in GDB 5.0) has been removed. Fix for gdb/192: removed extraneous space when displaying frame level. Fix for gdb/672: update changelist is now output in mi list format. Fix for gdb/702: a -var-assign that updates the value now shows up in a subsequent -var-update. * New native configurations. FreeBSD/amd64 x86_64-*-freebsd* * Multi-arched targets. HP/PA HPUX11 hppa*-*-hpux* Renesas M32R/D w/simulator m32r-*-elf* * OBSOLETE configurations and files Configurations that have been declared obsolete in this release have been commented out. Unless there is activity to revive these configurations, the next release of GDB will have their sources permanently REMOVED. Z8000 simulator z8k-zilog-none or z8ksim Matsushita MN10200 w/simulator mn10200-*-* H8/500 simulator h8500-hitachi-hms or h8500hms HP/PA running BSD hppa*-*-bsd* HP/PA running OSF/1 hppa*-*-osf* HP/PA Pro target hppa*-*-pro* PMAX (MIPS) running Mach 3.0 mips*-*-mach3* Sequent family i[3456]86-sequent-sysv4* i[3456]86-sequent-sysv* i[3456]86-sequent-bsd* Tsqware Sparclet sparclet-*-* Fujitsu SPARClite sparclite-fujitsu-none or sparclite * REMOVED configurations and files V850EA ISA Motorola Delta 88000 running Sys V m88k-motorola-sysv or delta88 IBM AIX PS/2 i[3456]86-*-aix i386 running Mach 3.0 i[3456]86-*-mach3* i386 running Mach i[3456]86-*-mach* i386 running OSF/1 i[3456]86-*osf1mk* HP/Apollo 68k Family m68*-apollo*-sysv*, m68*-apollo*-bsd*, m68*-hp-bsd*, m68*-hp-hpux* Argonaut Risc Chip (ARC) arc-*-* Mitsubishi D30V d30v-*-* Fujitsu FR30 fr30-*-elf* OS/9000 i[34]86-*-os9k I960 with MON960 i960-*-coff * MIPS $fp behavior changed The convenience variable $fp, for the MIPS, now consistently returns the address of the current frame's base. Previously, depending on the context, $fp could refer to either $sp or the current frame's base address. See ``8.10 Registers'' in the manual ``Debugging with GDB: The GNU Source-Level Debugger''. *** Changes in GDB 5.3: * GNU/Linux shared library multi-threaded performance improved. When debugging a multi-threaded application on GNU/Linux, GDB now uses `/proc', in preference to `ptrace' for memory reads. This may result in an improvement in the start-up time of multi-threaded, shared library applications when run under GDB. One GDB user writes: ``loads shared libs like mad''. * ``gdbserver'' now supports multi-threaded applications on some targets Support for debugging multi-threaded applications which use the GNU/Linux LinuxThreads package has been added for arm*-*-linux*-gnu*, i[3456]86-*-linux*-gnu*, mips*-*-linux*-gnu*, powerpc*-*-linux*-gnu*, and sh*-*-linux*-gnu*. * GDB now supports C/C++ preprocessor macros. GDB now expands preprocessor macro invocations in C/C++ expressions, and provides various commands for showing macro definitions and how they expand. The new command `macro expand EXPRESSION' expands any macro invocations in expression, and shows the result. The new command `show macro MACRO-NAME' shows the definition of the macro named MACRO-NAME, and where it was defined. Most compilers don't include information about macros in the debugging information by default. In GCC 3.1, for example, you need to compile your program with the options `-gdwarf-2 -g3'. If the macro information is present in the executable, GDB will read it. * Multi-arched targets. DEC Alpha (partial) alpha*-*-* DEC VAX (partial) vax-*-* NEC V850 v850-*-* National Semiconductor NS32000 (partial) ns32k-*-* Motorola 68000 (partial) m68k-*-* Motorola MCORE mcore-*-* * New targets. Fujitsu FRV architecture added by Red Hat frv*-*-* * New native configurations Alpha NetBSD alpha*-*-netbsd* SH NetBSD sh*-*-netbsdelf* MIPS NetBSD mips*-*-netbsd* UltraSPARC NetBSD sparc64-*-netbsd* * OBSOLETE configurations and files Configurations that have been declared obsolete in this release have been commented out. Unless there is activity to revive these configurations, the next release of GDB will have their sources permanently REMOVED. Mitsubishi D30V d30v-*-* OS/9000 i[34]86-*-os9k IBM AIX PS/2 i[3456]86-*-aix Fujitsu FR30 fr30-*-elf* Motorola Delta 88000 running Sys V m88k-motorola-sysv or delta88 Argonaut Risc Chip (ARC) arc-*-* i386 running Mach 3.0 i[3456]86-*-mach3* i386 running Mach i[3456]86-*-mach* i386 running OSF/1 i[3456]86-*osf1mk* HP/Apollo 68k Family m68*-apollo*-sysv*, m68*-apollo*-bsd*, m68*-hp-bsd*, m68*-hp-hpux* I960 with MON960 i960-*-coff * OBSOLETE languages CHILL, a Pascal like language used by telecommunications companies. * REMOVED configurations and files AMD 29k family via UDI a29k-amd-udi, udi29k A29K VxWorks a29k-*-vxworks AMD 29000 embedded, using EBMON a29k-none-none AMD 29000 embedded with COFF a29k-none-coff AMD 29000 embedded with a.out a29k-none-aout testsuite/gdb.hp/gdb.threads-hp/ directory * New command "set max-user-call-depth " This command allows the user to limit the call depth of user-defined commands. The default is 1024. * Changes in FreeBSD/i386 native debugging. Support for the "generate-core-file" has been added. * New commands "dump", "append", and "restore". These commands allow data to be copied from target memory to a bfd-format or binary file (dump and append), and back from a file into memory (restore). * Improved "next/step" support on multi-processor Alpha Tru64. The previous single-step mechanism could cause unpredictable problems, including the random appearance of SIGSEGV or SIGTRAP signals. The use of a software single-step mechanism prevents this. *** Changes in GDB 5.2.1: * New targets. Atmel AVR avr*-*-* * Bug fixes gdb/182: gdb/323: gdb/237: On alpha, gdb was reporting: mdebugread.c:2443: gdb-internal-error: sect_index_data not initialized Fix, by Joel Brobecker imported from mainline. gdb/439: gdb/291: On some ELF object files, gdb was reporting: dwarf2read.c:1072: gdb-internal-error: sect_index_text not initialize Fix, by Fred Fish, imported from mainline. Dwarf2 .debug_frame & .eh_frame handler improved in many ways. Surprisingly enough, it works now. By Michal Ludvig, imported from mainline. i386 hardware watchpoint support: avoid misses on second run for some targets. By Pierre Muller, imported from mainline. *** Changes in GDB 5.2: * New command "set trust-readonly-sections on[off]". This command is a hint that tells gdb that read-only sections really are read-only (ie. that their contents will not change). In this mode, gdb will go to the object file rather than the target to read memory from read-only sections (such as ".text"). This can be a significant performance improvement on some (notably embedded) targets. * New command "generate-core-file" (or "gcore"). This new gdb command allows the user to drop a core file of the child process state at any time. So far it's been implemented only for GNU/Linux and Solaris, but should be relatively easily ported to other hosts. Argument is core file name (defaults to core.). * New command line option GDB now accepts --pid or -p followed by a process id. * Change in command line behavior -- corefiles vs. process ids. There is a subtle behavior in the way in which GDB handles command line arguments. The first non-flag argument is always a program to debug, but the second non-flag argument may either be a corefile or a process id. Previously, GDB would attempt to open the second argument as a corefile, and if that failed, would issue a superfluous error message and then attempt to attach it as a process. Now, if the second argument begins with a non-digit, it will be treated as a corefile. If it begins with a digit, GDB will attempt to attach it as a process, and if no such process is found, will then attempt to open it as a corefile. * Changes in ARM configurations. Multi-arch support is enabled for all ARM configurations. The ARM/NetBSD configuration is fully multi-arch. * New native configurations ARM NetBSD arm*-*-netbsd* x86 OpenBSD i[3456]86-*-openbsd* AMD x86-64 running GNU/Linux x86_64-*-linux-* Sparc64 running FreeBSD sparc64-*-freebsd* * New targets Sanyo XStormy16 xstormy16-elf * OBSOLETE configurations and files Configurations that have been declared obsolete in this release have been commented out. Unless there is activity to revive these configurations, the next release of GDB will have their sources permanently REMOVED. AMD 29k family via UDI a29k-amd-udi, udi29k A29K VxWorks a29k-*-vxworks AMD 29000 embedded, using EBMON a29k-none-none AMD 29000 embedded with COFF a29k-none-coff AMD 29000 embedded with a.out a29k-none-aout testsuite/gdb.hp/gdb.threads-hp/ directory * REMOVED configurations and files TI TMS320C80 tic80-*-* WDC 65816 w65-*-* PowerPC Solaris powerpcle-*-solaris* PowerPC Windows NT powerpcle-*-cygwin32 PowerPC Netware powerpc-*-netware* Harris/CXUX m88k m88*-harris-cxux* Most ns32k hosts and targets ns32k-*-mach3* ns32k-umax-* ns32k-utek-sysv* ns32k-utek-* SunOS 4.0.Xi on i386 i[3456]86-*-sunos* Ultracomputer (29K) running Sym1 a29k-nyu-sym1 a29k-*-kern* Sony NEWS (68K) running NEWSOS 3.x m68*-sony-sysv news ISI Optimum V (3.05) under 4.3bsd. m68*-isi-* Apple Macintosh (MPW) host and target N/A host, powerpc-*-macos* * Changes to command line processing The new `--args' feature can be used to specify command-line arguments for the inferior from gdb's command line. * Changes to key bindings There is a new `operate-and-get-next' function bound to `C-o'. *** Changes in GDB 5.1.1 Fix compile problem on DJGPP. Fix a problem with floating-point registers on the i386 being corrupted. Fix to stop GDB crashing on .debug_str debug info. Numerous documentation fixes. Numerous testsuite fixes. *** Changes in GDB 5.1: * New native configurations Alpha FreeBSD alpha*-*-freebsd* x86 FreeBSD 3.x and 4.x i[3456]86*-freebsd[34]* MIPS GNU/Linux mips*-*-linux* MIPS SGI Irix 6.x mips*-sgi-irix6* ia64 AIX ia64-*-aix* s390 and s390x GNU/Linux {s390,s390x}-*-linux* * New targets Motorola 68HC11 and 68HC12 m68hc11-elf CRIS cris-axis UltraSparc running GNU/Linux sparc64-*-linux* * OBSOLETE configurations and files x86 FreeBSD before 2.2 i[3456]86*-freebsd{1,2.[01]}*, Harris/CXUX m88k m88*-harris-cxux* Most ns32k hosts and targets ns32k-*-mach3* ns32k-umax-* ns32k-utek-sysv* ns32k-utek-* TI TMS320C80 tic80-*-* WDC 65816 w65-*-* Ultracomputer (29K) running Sym1 a29k-nyu-sym1 a29k-*-kern* PowerPC Solaris powerpcle-*-solaris* PowerPC Windows NT powerpcle-*-cygwin32 PowerPC Netware powerpc-*-netware* SunOS 4.0.Xi on i386 i[3456]86-*-sunos* Sony NEWS (68K) running NEWSOS 3.x m68*-sony-sysv news ISI Optimum V (3.05) under 4.3bsd. m68*-isi-* Apple Macintosh (MPW) host N/A stuff.c (Program to stuff files into a specially prepared space in kdb) kdb-start.c (Main loop for the standalone kernel debugger) Configurations that have been declared obsolete in this release have been commented out. Unless there is activity to revive these configurations, the next release of GDB will have their sources permanently REMOVED. * REMOVED configurations and files Altos 3068 m68*-altos-* Convex c1-*-*, c2-*-* Pyramid pyramid-*-* ARM RISCix arm-*-* (as host) Tahoe tahoe-*-* ser-ocd.c *-*-* * GDB has been converted to ISO C. GDB's source code has been converted to ISO C. In particular, the sources are fully protoized, and rely on standard headers being present. * Other news: * "info symbol" works on platforms which use COFF, ECOFF, XCOFF, and NLM. * The MI enabled by default. The new machine oriented interface (MI) introduced in GDB 5.0 has been revised and enabled by default. Packages which use GDB as a debugging engine behind a UI or another front end are encouraged to switch to using the GDB/MI interface, instead of the old annotations interface which is now deprecated. * Support for debugging Pascal programs. GDB now includes support for debugging Pascal programs. The following main features are supported: - Pascal-specific data types such as sets; - automatic recognition of Pascal sources based on file-name extension; - Pascal-style display of data types, variables, and functions; - a Pascal expression parser. However, some important features are not yet supported. - Pascal string operations are not supported at all; - there are some problems with boolean types; - Pascal type hexadecimal constants are not supported because they conflict with the internal variables format; - support for Pascal objects and classes is not full yet; - unlike Pascal, GDB is case-sensitive for symbol names. * Changes in completion. Commands such as `shell', `run' and `set args', which pass arguments to inferior programs, now complete on file names, similar to what users expect at the shell prompt. Commands which accept locations, such as `disassemble', `print', `breakpoint', `until', etc. now complete on filenames as well as program symbols. Thus, if you type "break foob TAB", and the source files linked into the programs include `foobar.c', that file name will be one of the candidates for completion. However, file names are not considered for completion after you typed a colon that delimits a file name from a name of a function in that file, as in "break foo.c:bar". `set demangle-style' completes on available demangling styles. * New platform-independent commands: It is now possible to define a post-hook for a command as well as a hook that runs before the command. For more details, see the documentation of `hookpost' in the GDB manual. * Changes in GNU/Linux native debugging. Support for debugging multi-threaded programs has been completely revised for all platforms except m68k and sparc. You can now debug as many threads as your system allows you to have. Attach/detach is supported for multi-threaded programs. Support for SSE registers was added for x86. This doesn't work for multi-threaded programs though. * Changes in MIPS configurations. Multi-arch support is enabled for all MIPS configurations. GDB can now be built as native debugger on SGI Irix 6.x systems for debugging n32 executables. (Debugging 64-bit executables is not yet supported.) * Unified support for hardware watchpoints in all x86 configurations. Most (if not all) native x86 configurations support hardware-assisted breakpoints and watchpoints in a unified manner. This support implements debug register sharing between watchpoints, which allows to put a virtually infinite number of watchpoints on the same address, and also supports watching regions up to 16 bytes with several debug registers. The new maintenance command `maintenance show-debug-regs' toggles debugging print-outs in functions that insert, remove, and test watchpoints and hardware breakpoints. * Changes in the DJGPP native configuration. New command ``info dos sysinfo'' displays assorted information about the CPU, OS, memory, and DPMI server. New commands ``info dos gdt'', ``info dos ldt'', and ``info dos idt'' display information about segment descriptors stored in GDT, LDT, and IDT. New commands ``info dos pde'' and ``info dos pte'' display entries from Page Directory and Page Tables (for now works with CWSDPMI only). New command ``info dos address-pte'' displays the Page Table entry for a given linear address. GDB can now pass command lines longer than 126 characters to the program being debugged (requires an update to the libdbg.a library which is part of the DJGPP development kit). DWARF2 debug info is now supported. It is now possible to `step' and `next' through calls to `longjmp'. * Changes in documentation. All GDB documentation was converted to GFDL, the GNU Free Documentation License. Tracepoints-related commands are now fully documented in the GDB manual. TUI, the Text-mode User Interface, is now documented in the manual. Tracepoints-related commands are now fully documented in the GDB manual. The "GDB Internals" manual now has an index. It also includes documentation of `ui_out' functions, GDB coding standards, x86 hardware watchpoints, and memory region attributes. * GDB's version number moved to ``version.in'' The Makefile variable VERSION has been replaced by the file ``version.in''. People creating GDB distributions should update the contents of this file. * gdba.el deleted GUD support is now a standard part of the EMACS distribution. *** Changes in GDB 5.0: * Improved support for debugging FP programs on x86 targets Unified and much-improved support for debugging floating-point programs on all x86 targets. In particular, ``info float'' now displays the FP registers in the same format on all x86 targets, with greater level of detail. * Improvements and bugfixes in hardware-assisted watchpoints It is now possible to watch array elements, struct members, and bitfields with hardware-assisted watchpoints. Data-read watchpoints on x86 targets no longer erroneously trigger when the address is written. * Improvements in the native DJGPP version of GDB The distribution now includes all the scripts and auxiliary files necessary to build the native DJGPP version on MS-DOS/MS-Windows machines ``out of the box''. The DJGPP version can now debug programs that use signals. It is possible to catch signals that happened in the debuggee, deliver signals to it, interrupt it with Ctrl-C, etc. (Previously, a signal would kill the program being debugged.) Programs that hook hardware interrupts (keyboard, timer, etc.) can also be debugged. It is now possible to debug DJGPP programs that redirect their standard handles or switch them to raw (as opposed to cooked) mode, or even close them. The command ``run < foo > bar'' works as expected, and ``info terminal'' reports useful information about the debuggee's terminal, including raw/cooked mode, redirection, etc. The DJGPP version now uses termios functions for console I/O, which enables debugging graphics programs. Interrupting GDB with Ctrl-C also works. DOS-style file names with drive letters are now fully supported by GDB. It is now possible to debug DJGPP programs that switch their working directory. It is also possible to rerun the debuggee any number of times without restarting GDB; thus, you can use the same setup, breakpoints, etc. for many debugging sessions. * New native configurations ARM GNU/Linux arm*-*-linux* PowerPC GNU/Linux powerpc-*-linux* * New targets Motorola MCore mcore-*-* x86 VxWorks i[3456]86-*-vxworks* PowerPC VxWorks powerpc-*-vxworks* TI TMS320C80 tic80-*-* * OBSOLETE configurations Altos 3068 m68*-altos-* Convex c1-*-*, c2-*-* Pyramid pyramid-*-* ARM RISCix arm-*-* (as host) Tahoe tahoe-*-* Configurations that have been declared obsolete will be commented out, but the code will be left in place. If there is no activity to revive these configurations before the next release of GDB, the sources will be permanently REMOVED. * Gould support removed Support for the Gould PowerNode and NP1 has been removed. * New features for SVR4 On SVR4 native platforms (such as Solaris), if you attach to a process without first loading a symbol file, GDB will now attempt to locate and load symbols from the running process's executable file. * Many C++ enhancements C++ support has been greatly improved. Overload resolution now works properly in almost all cases. RTTI support is on the way. * Remote targets can connect to a sub-program A popen(3) style serial-device has been added. This device starts a sub-process (such as a stand-alone simulator) and then communicates with that. The sub-program to run is specified using the syntax ``| '' vis: (gdb) set remotedebug 1 (gdb) target extended-remote |mn10300-elf-sim program-args * MIPS 64 remote protocol A long standing bug in the mips64 remote protocol where by GDB expected certain 32 bit registers (ex SR) to be transfered as 32 instead of 64 bits has been fixed. The command ``set remote-mips64-transfers-32bit-regs on'' has been added to provide backward compatibility with older versions of GDB. * ``set remotebinarydownload'' replaced by ``set remote X-packet'' The command ``set remotebinarydownload'' command has been replaced by ``set remote X-packet''. Other commands in ``set remote'' family include ``set remote P-packet''. * Breakpoint commands accept ranges. The breakpoint commands ``enable'', ``disable'', and ``delete'' now accept a range of breakpoints, e.g. ``5-7''. The tracepoint command ``tracepoint passcount'' also accepts a range of tracepoints. * ``apropos'' command added. The ``apropos'' command searches through command names and documentation strings, printing out matches, making it much easier to try to find a command that does what you are looking for. * New MI interface A new machine oriented interface (MI) has been added to GDB. This interface is designed for debug environments running GDB as a separate process. This is part of the long term libGDB project. See the "GDB/MI" chapter of the GDB manual for further information. It can be enabled by configuring with: .../configure --enable-gdbmi *** Changes in GDB-4.18: * New native configurations HP-UX 10.20 hppa*-*-hpux10.20 HP-UX 11.x hppa*-*-hpux11.0* M68K GNU/Linux m68*-*-linux* * New targets Fujitsu FR30 fr30-*-elf* Intel StrongARM strongarm-*-* Mitsubishi D30V d30v-*-* * OBSOLETE configurations Gould PowerNode, NP1 np1-*-*, pn-*-* Configurations that have been declared obsolete will be commented out, but the code will be left in place. If there is no activity to revive these configurations before the next release of GDB, the sources will be permanently REMOVED. * ANSI/ISO C As a compatibility experiment, GDB's source files buildsym.h and buildsym.c have been converted to pure standard C, no longer containing any K&R compatibility code. We believe that all systems in use today either come with a standard C compiler, or have a GCC port available. If this is not true, please report the affected configuration to bug-gdb@gnu.org immediately. See the README file for information about getting a standard C compiler if you don't have one already. * Readline 2.2 GDB now uses readline 2.2. * set extension-language You can now control the mapping between filename extensions and source languages by using the `set extension-language' command. For instance, you can ask GDB to treat .c files as C++ by saying set extension-language .c c++ The command `info extensions' lists all of the recognized extensions and their associated languages. * Setting processor type for PowerPC and RS/6000 When GDB is configured for a powerpc*-*-* or an rs6000*-*-* target, you can use the `set processor' command to specify what variant of the PowerPC family you are debugging. The command set processor NAME sets the PowerPC/RS6000 variant to NAME. GDB knows about the following PowerPC and RS6000 variants: ppc-uisa PowerPC UISA - a PPC processor as viewed by user-level code rs6000 IBM RS6000 ("POWER") architecture, user-level view 403 IBM PowerPC 403 403GC IBM PowerPC 403GC 505 Motorola PowerPC 505 860 Motorola PowerPC 860 or 850 601 Motorola PowerPC 601 602 Motorola PowerPC 602 603 Motorola/IBM PowerPC 603 or 603e 604 Motorola PowerPC 604 or 604e 750 Motorola/IBM PowerPC 750 or 750 At the moment, this command just tells GDB what to name the special-purpose processor registers. Since almost all the affected registers are inaccessible to user-level programs, this command is only useful for remote debugging in its present form. * HP-UX support Thanks to a major code donation from Hewlett-Packard, GDB now has much more extensive support for HP-UX. Added features include shared library support, kernel threads and hardware watchpoints for 11.00, support for HP's ANSI C and C++ compilers, and a compatibility mode for xdb and dbx commands. * Catchpoints HP's donation includes the new concept of catchpoints, which is a generalization of the old catch command. On HP-UX, it is now possible to catch exec, fork, and vfork, as well as library loading. This means that the existing catch command has changed; its first argument now specifies the type of catch to be set up. See the output of "help catch" for a list of catchpoint types. * Debugging across forks On HP-UX, you can choose which process to debug when a fork() happens in the inferior. * TUI HP has donated a curses-based terminal user interface (TUI). To get it, build with --enable-tui. Although this can be enabled for any configuration, at present it only works for native HP debugging. * GDB remote protocol additions A new protocol packet 'X' that writes binary data is now available. Default behavior is to try 'X', then drop back to 'M' if the stub fails to respond. The settable variable `remotebinarydownload' allows explicit control over the use of 'X'. For 64-bit targets, the memory packets ('M' and 'm') can now contain a full 64-bit address. The command set remoteaddresssize 32 can be used to revert to the old behaviour. For existing remote stubs the change should not be noticed, as the additional address information will be discarded. In order to assist in debugging stubs, you may use the maintenance command `packet' to send any text string to the stub. For instance, maint packet heythere sends the packet "$heythere#". Note that it is very easy to disrupt a debugging session by sending the wrong packet at the wrong time. The compare-sections command allows you to compare section data on the target to what is in the executable file without uploading or downloading, by comparing CRC checksums. * Tracing can collect general expressions You may now collect general expressions at tracepoints. This requires further additions to the target-side stub; see tracepoint.c and doc/agentexpr.texi for further details. * mask-address variable for Mips For Mips targets, you may control the zeroing of the upper 32 bits of a 64-bit address by entering `set mask-address on'. This is mainly of interest to users of embedded R4xxx and R5xxx processors. * Higher serial baud rates GDB's serial code now allows you to specify baud rates 57600, 115200, 230400, and 460800 baud. (Note that your host system may not be able to achieve all of these rates.) * i960 simulator The i960 configuration now includes an initial implementation of a builtin simulator, contributed by Jim Wilson. *** Changes in GDB-4.17: * New native configurations Alpha GNU/Linux alpha*-*-linux* Unixware 2.x i[3456]86-unixware2* Irix 6.x mips*-sgi-irix6* PowerPC GNU/Linux powerpc-*-linux* PowerPC Solaris powerpcle-*-solaris* Sparc GNU/Linux sparc-*-linux* Motorola sysV68 R3V7.1 m68k-motorola-sysv * New targets Argonaut Risc Chip (ARC) arc-*-* Hitachi H8/300S h8300*-*-* Matsushita MN10200 w/simulator mn10200-*-* Matsushita MN10300 w/simulator mn10300-*-* MIPS NEC VR4100 mips64*vr4100*{,el}-*-elf* MIPS NEC VR5000 mips64*vr5000*{,el}-*-elf* MIPS Toshiba TX39 mips64*tx39*{,el}-*-elf* Mitsubishi D10V w/simulator d10v-*-* Mitsubishi M32R/D w/simulator m32r-*-elf* Tsqware Sparclet sparclet-*-* NEC V850 w/simulator v850-*-* * New debugging protocols ARM with RDI protocol arm*-*-* M68K with dBUG monitor m68*-*-{aout,coff,elf} DDB and LSI variants of PMON protocol mips*-*-* PowerPC with DINK32 monitor powerpc{,le}-*-eabi PowerPC with SDS protocol powerpc{,le}-*-eabi Macraigor OCD (Wiggler) devices powerpc{,le}-*-eabi * DWARF 2 All configurations can now understand and use the DWARF 2 debugging format. The choice is automatic, if the symbol file contains DWARF 2 information. * Java frontend GDB now includes basic Java language support. This support is only useful with Java compilers that produce native machine code. * solib-absolute-prefix and solib-search-path For SunOS and SVR4 shared libraries, you may now set the prefix for loading absolute shared library symbol files, and the search path for locating non-absolute shared library symbol files. * Live range splitting GDB can now effectively debug code for which GCC has performed live range splitting as part of its optimization. See gdb/doc/LRS for more details on the expected format of the stabs information. * Hurd support GDB's support for the GNU Hurd, including thread debugging, has been updated to work with current versions of the Hurd. * ARM Thumb support GDB's ARM target configuration now handles the ARM7T (Thumb) 16-bit instruction set. ARM GDB automatically detects when Thumb instructions are in use, and adjusts disassembly and backtracing accordingly. * MIPS16 support GDB's MIPS target configurations now handle the MIP16 16-bit instruction set. * Overlay support GDB now includes support for overlays; if an executable has been linked such that multiple sections are based at the same address, GDB will decide which section to use for symbolic info. You can choose to control the decision manually, using overlay commands, or implement additional target-side support and use "overlay load-target" to bring in the overlay mapping. Do "help overlay" for more detail. * info symbol The command "info symbol
" displays information about the symbol at the specified address. * Trace support The standard remote protocol now includes an extension that allows asynchronous collection and display of trace data. This requires extensive support in the target-side debugging stub. Tracing mode includes a new interaction mode in GDB and new commands: see the file tracepoint.c for more details. * MIPS simulator Configurations for embedded MIPS now include a simulator contributed by Cygnus Solutions. The simulator supports the instruction sets of most MIPS variants. * Sparc simulator Sparc configurations may now include the ERC32 simulator contributed by the European Space Agency. The simulator is not built into Sparc targets by default; configure with --enable-sim to include it. * set architecture For target configurations that may include multiple variants of a basic architecture (such as MIPS and SH), you may now set the architecture explicitly. "set arch" sets, "info arch" lists the possible architectures. *** Changes in GDB-4.16: * New native configurations Windows 95, x86 Windows NT i[345]86-*-cygwin32 M68K NetBSD m68k-*-netbsd* PowerPC AIX 4.x powerpc-*-aix* PowerPC MacOS powerpc-*-macos* PowerPC Windows NT powerpcle-*-cygwin32 RS/6000 AIX 4.x rs6000-*-aix4* * New targets ARM with RDP protocol arm-*-* I960 with MON960 i960-*-coff MIPS VxWorks mips*-*-vxworks* MIPS VR4300 with PMON mips64*vr4300{,el}-*-elf* PowerPC with PPCBUG monitor powerpc{,le}-*-eabi* Hitachi SH3 sh-*-* Matra Sparclet sparclet-*-* * PowerPC simulator The powerpc-eabi configuration now includes the PSIM simulator, contributed by Andrew Cagney, with assistance from Mike Meissner. PSIM is a very elaborate model of the PowerPC, including not only basic instruction set execution, but also details of execution unit performance and I/O hardware. See sim/ppc/README for more details. * Solaris 2.5 GDB now works with Solaris 2.5. * Windows 95/NT native GDB will now work as a native debugger on Windows 95 and Windows NT. To build it from source, you must use the "gnu-win32" environment, which uses a DLL to emulate enough of Unix to run the GNU tools. Further information, binaries, and sources are available at ftp.cygnus.com, under pub/gnu-win32. * dont-repeat command If a user-defined command includes the command `dont-repeat', then the command will not be repeated if the user just types return. This is useful if the command is time-consuming to run, so that accidental extra keystrokes don't run the same command many times. * Send break instead of ^C The standard remote protocol now includes an option to send a break rather than a ^C to the target in order to interrupt it. By default, GDB will send ^C; to send a break, set the variable `remotebreak' to 1. * Remote protocol timeout The standard remote protocol includes a new variable `remotetimeout' that allows you to set the number of seconds before GDB gives up trying to read from the target. The default value is 2. * Automatic tracking of dynamic object loading (HPUX and Solaris only) By default GDB will automatically keep track of objects as they are loaded and unloaded by the dynamic linker. By using the command `set stop-on-solib-events 1' you can arrange for GDB to stop the inferior when shared library events occur, thus allowing you to set breakpoints in shared libraries which are explicitly loaded by the inferior. Note this feature does not work on hpux8. On hpux9 you must link /usr/lib/end.o into your program. This feature should work automatically on hpux10. * Irix 5.x hardware watchpoint support Irix 5 configurations now support the use of hardware watchpoints. * Mips protocol "SYN garbage limit" When debugging a Mips target using the `target mips' protocol, you may set the number of characters that GDB will ignore by setting the `syn-garbage-limit'. A value of -1 means that GDB will ignore every character. The default value is 1050. * Recording and replaying remote debug sessions If you set `remotelogfile' to the name of a file, gdb will write to it a recording of a remote debug session. This recording may then be replayed back to gdb using "gdbreplay". See gdbserver/README for details. This is useful when you have a problem with GDB while doing remote debugging; you can make a recording of the session and send it to someone else, who can then recreate the problem. * Speedups for remote debugging GDB includes speedups for downloading and stepping MIPS systems using the IDT monitor, fast downloads to the Hitachi SH E7000 emulator, and more efficient S-record downloading. * Memory use reductions and statistics collection GDB now uses less memory and reports statistics about memory usage. Try the `maint print statistics' command, for example. *** Changes in GDB-4.15: * Psymtabs for XCOFF The symbol reader for AIX GDB now uses partial symbol tables. This can greatly improve startup time, especially for large executables. * Remote targets use caching Remote targets now use a data cache to speed up communication with the remote side. The data cache could lead to incorrect results because it doesn't know about volatile variables, thus making it impossible to debug targets which use memory mapped I/O devices. `set remotecache off' turns the data cache off. * Remote targets may have threads The standard remote protocol now includes support for multiple threads in the target system, using new protocol commands 'H' and 'T'. See gdb/remote.c for details. * NetROM support If GDB is configured with `--enable-netrom', then it will include support for the NetROM ROM emulator from XLNT Designs. The NetROM acts as though it is a bank of ROM on the target board, but you can write into it over the network. GDB's support consists only of support for fast loading into the emulated ROM; to debug, you must use another protocol, such as standard remote protocol. The usual sequence is something like target nrom load target remote :1235 * Macintosh host GDB now includes support for the Apple Macintosh, as a host only. It may be run as either an MPW tool or as a standalone application, and it can debug through the serial port. All the usual GDB commands are available, but to the target command, you must supply "serial" as the device type instead of "/dev/ttyXX". See mpw-README in the main directory for more information on how to build. The MPW configuration scripts */mpw-config.in support only a few targets, and only the mips-idt-ecoff target has been tested. * Autoconf GDB configuration now uses autoconf. This is not user-visible, but does simplify configuration and building. * hpux10 GDB now supports hpux10. *** Changes in GDB-4.14: * New native configurations x86 FreeBSD i[345]86-*-freebsd x86 NetBSD i[345]86-*-netbsd NS32k NetBSD ns32k-*-netbsd Sparc NetBSD sparc-*-netbsd * New targets A29K VxWorks a29k-*-vxworks HP PA PRO embedded (WinBond W89K & Oki OP50N) hppa*-*-pro* CPU32 EST-300 emulator m68*-*-est* PowerPC ELF powerpc-*-elf WDC 65816 w65-*-* * Alpha OSF/1 support for procfs GDB now supports procfs under OSF/1-2.x and higher, which makes it possible to attach to running processes. As the mounting of the /proc filesystem is optional on the Alpha, GDB automatically determines the availability of /proc during startup. This can lead to problems if /proc is unmounted after GDB has been started. * Arguments to user-defined commands User commands may accept up to 10 arguments separated by whitespace. Arguments are accessed within the user command via $arg0..$arg9. A trivial example: define adder print $arg0 + $arg1 + $arg2 To execute the command use: adder 1 2 3 Defines the command "adder" which prints the sum of its three arguments. Note the arguments are text substitutions, so they may reference variables, use complex expressions, or even perform inferior function calls. * New `if' and `while' commands This makes it possible to write more sophisticated user-defined commands. Both commands take a single argument, which is the expression to evaluate, and must be followed by the commands to execute, one per line, if the expression is nonzero, the list being terminated by the word `end'. The `if' command list may include an `else' word, which causes the following commands to be executed only if the expression is zero. * Fortran source language mode GDB now includes partial support for Fortran 77. It will recognize Fortran programs and can evaluate a subset of Fortran expressions, but variables and functions may not be handled correctly. GDB will work with G77, but does not yet know much about symbols emitted by other Fortran compilers. * Better HPUX support Most debugging facilities now work on dynamic executables for HPPAs running hpux9 or later. You can attach to running dynamically linked processes, but by default the dynamic libraries will be read-only, so for instance you won't be able to put breakpoints in them. To change that behavior do the following before running the program: adb -w a.out __dld_flags?W 0x5 control-d This will cause the libraries to be mapped private and read-write. To revert to the normal behavior, do this: adb -w a.out __dld_flags?W 0x4 control-d You cannot set breakpoints or examine data in the library until after the library is loaded if the function/data symbols do not have external linkage. GDB can now also read debug symbols produced by the HP C compiler on HPPAs (sorry, no C++, Fortran or 68k support). * Target byte order now dynamically selectable You can choose which byte order to use with a target system, via the commands "set endian big" and "set endian little", and you can see the current setting by using "show endian". You can also give the command "set endian auto", in which case GDB will use the byte order associated with the executable. Currently, only embedded MIPS configurations support dynamic selection of target byte order. * New DOS host serial code This version uses DPMI interrupts to handle buffered I/O, so you no longer need to run asynctsr when debugging boards connected to a PC's serial port. *** Changes in GDB-4.13: * New "complete" command This lists all the possible completions for the rest of the line, if it were to be given as a command itself. This is intended for use by emacs. * Trailing space optional in prompt "set prompt" no longer adds a space for you after the prompt you set. This allows you to set a prompt which ends in a space or one that does not. * Breakpoint hit counts "info break" now displays a count of the number of times the breakpoint has been hit. This is especially useful in conjunction with "ignore"; you can ignore a large number of breakpoint hits, look at the breakpoint info to see how many times the breakpoint was hit, then run again, ignoring one less than that number, and this will get you quickly to the last hit of that breakpoint. * Ability to stop printing at NULL character "set print null-stop" will cause GDB to stop printing the characters of an array when the first NULL is encountered. This is useful when large arrays actually contain only short strings. * Shared library breakpoints In SunOS 4.x, SVR4, and Alpha OSF/1 configurations, you can now set breakpoints in shared libraries before the executable is run. * Hardware watchpoints There is a new hardware breakpoint for the watch command for sparclite targets. See gdb/sparclite/hw_breakpoint.note. Hardware watchpoints are also now supported under GNU/Linux. * Annotations Annotations have been added. These are for use with graphical interfaces, and are still experimental. Currently only gdba.el uses these. * Improved Irix 5 support GDB now works properly with Irix 5.2. * Improved HPPA support GDB now works properly with the latest GCC and GAS. * New native configurations Sequent PTX4 i[34]86-sequent-ptx4 HPPA running OSF/1 hppa*-*-osf* Atari TT running SVR4 m68*-*-sysv4* RS/6000 LynxOS rs6000-*-lynxos* * New targets OS/9000 i[34]86-*-os9k MIPS R4000 mips64*{,el}-*-{ecoff,elf} Sparc64 sparc64-*-* * Hitachi SH7000 and E7000-PC ICE support There is now support for communicating with the Hitachi E7000-PC ICE. This is available automatically when GDB is configured for the SH. * Fixes As usual, a variety of small fixes and improvements, both generic and configuration-specific. See the ChangeLog for more detail. *** Changes in GDB-4.12: * Irix 5 is now supported * HPPA support GDB-4.12 on the HPPA has a number of changes which make it unable to debug the output from the currently released versions of GCC and GAS (GCC 2.5.8 and GAS-2.2 or PAGAS-1.36). Until the next major release of GCC and GAS, versions of these tools designed to work with GDB-4.12 can be retrieved via anonymous ftp from jaguar.cs.utah.edu:/dist. *** Changes in GDB-4.11: * User visible changes: * Remote Debugging The "set remotedebug" option is now consistent between the mips remote target, remote targets using the gdb-specific protocol, UDI (AMD's debug protocol for the 29k) and the 88k bug monitor. It is now an integer specifying a debug level (normally 0 or 1, but 2 means more debugging info for the mips target). * DEC Alpha native support GDB now works on the DEC Alpha. GCC 2.4.5 does not produce usable debug info, but GDB works fairly well with the DEC compiler and should work with a future GCC release. See the README file for a few Alpha-specific notes. * Preliminary thread implementation GDB now has preliminary thread support for both SGI/Irix and LynxOS. * LynxOS native and target support for 386 This release has been hosted on LynxOS 2.2, and also can be configured to remotely debug programs running under LynxOS (see gdb/gdbserver/README for details). * Improvements in C++ mangling/demangling. This release has much better g++ debugging, specifically in name mangling/demangling, virtual function calls, print virtual table, call methods, ...etc. *** Changes in GDB-4.10: * User visible changes: Remote debugging using the GDB-specific (`target remote') protocol now supports the `load' command. This is only useful if you have some other way of getting the stub to the target system, and you can put it somewhere in memory where it won't get clobbered by the download. Filename completion now works. When run under emacs mode, the "info line" command now causes the arrow to point to the line specified. Also, "info line" prints addresses in symbolic form (as well as hex). All vxworks based targets now support a user settable option, called vxworks-timeout. This option represents the number of seconds gdb should wait for responses to rpc's. You might want to use this if your vxworks target is, perhaps, a slow software simulator or happens to be on the far side of a thin network line. * DEC alpha support This release contains support for using a DEC alpha as a GDB host for cross debugging. Native alpha debugging is not supported yet. *** Changes in GDB-4.9: * Testsuite This is the first GDB release which is accompanied by a matching testsuite. The testsuite requires installation of dejagnu, which should be available via ftp from most sites that carry GNU software. * C++ demangling 'Cfront' style demangling has had its name changed to 'ARM' style, to emphasize that it was written from the specifications in the C++ Annotated Reference Manual, not necessarily to be compatible with AT&T cfront. Despite disclaimers, it still generated too much confusion with users attempting to use gdb with AT&T cfront. * Simulators GDB now uses a standard remote interface to a simulator library. So far, the library contains simulators for the Zilog Z8001/2, the Hitachi H8/300, H8/500 and Super-H. * New targets supported H8/300 simulator h8300-hitachi-hms or h8300hms H8/500 simulator h8500-hitachi-hms or h8500hms SH simulator sh-hitachi-hms or sh Z8000 simulator z8k-zilog-none or z8ksim IDT MIPS board over serial line mips-idt-ecoff Cross-debugging to GO32 targets is supported. It requires a custom version of the i386-stub.c module which is integrated with the GO32 memory extender. * New remote protocols MIPS remote debugging protocol. * New source languages supported This version includes preliminary support for Chill, a Pascal like language used by telecommunications companies. Chill support is also being integrated into the GNU compiler, but we don't know when it will be publically available. *** Changes in GDB-4.8: * HP Precision Architecture supported GDB now supports HP PA-RISC machines running HPUX. A preliminary version of this support was available as a set of patches from the University of Utah. GDB does not support debugging of programs compiled with the HP compiler, because HP will not document their file format. Instead, you must use GCC (version 2.3.2 or later) and PA-GAS (as available from jaguar.cs.utah.edu:/dist/pa-gas.u4.tar.Z). Many problems in the preliminary version have been fixed. * Faster and better demangling We have improved template demangling and fixed numerous bugs in the GNU style demangler. It can now handle type modifiers such as `static' or `const'. Wide character types (wchar_t) are now supported. Demangling of each symbol is now only done once, and is cached when the symbol table for a file is read in. This results in a small increase in memory usage for C programs, a moderate increase in memory usage for C++ programs, and a fantastic speedup in symbol lookups. `Cfront' style demangling still doesn't work with AT&T cfront. It was written from the specifications in the Annotated Reference Manual, which AT&T's compiler does not actually implement. * G++ multiple inheritance compiler problem In the 2.3.2 release of gcc/g++, how the compiler resolves multiple inheritance lattices was reworked to properly discover ambiguities. We recently found an example which causes this new algorithm to fail in a very subtle way, producing bad debug information for those classes. The file 'gcc.patch' (in this directory) can be applied to gcc to circumvent the problem. A future GCC release will contain a complete fix. The previous G++ debug info problem (mentioned below for the gdb-4.7 release) is fixed in gcc version 2.3.2. * Improved configure script The `configure' script will now attempt to guess your system type if you don't supply a host system type. The old scheme of supplying a host system triplet is preferable over using this. All the magic is done in the new `config.guess' script. Examine it for details. We have also brought our configure script much more in line with the FSF's version. It now supports the --with-xxx options. In particular, `--with-minimal-bfd' can be used to make the GDB binary image smaller. The resulting GDB will not be able to read arbitrary object file formats -- only the format ``expected'' to be used on the configured target system. We hope to make this the default in a future release. * Documentation improvements There's new internal documentation on how to modify GDB, and how to produce clean changes to the code. We implore people to read it before submitting changes. The GDB manual uses new, sexy Texinfo conditionals, rather than arcane M4 macros. The new texinfo.tex is provided in this release. Pre-built `info' files are also provided. To build `info' files from scratch, you will need the latest `makeinfo' release, which will be available in a future texinfo-X.Y release. *NOTE* The new texinfo.tex can cause old versions of TeX to hang. We're not sure exactly which versions have this problem, but it has been seen in 3.0. We highly recommend upgrading to TeX version 3.141 or better. If that isn't possible, there is a patch in `texinfo/tex3patch' that will modify `texinfo/texinfo.tex' to work around this problem. * New features GDB now supports array constants that can be used in expressions typed in by the user. The syntax is `{element, element, ...}'. Ie: you can now type `print {1, 2, 3}', and it will build up an array in memory malloc'd in the target program. The new directory `gdb/sparclite' contains a program that demonstrates how the sparc-stub.c remote stub runs on a Fujitsu SPARClite processor. * New native hosts supported HP/PA-RISC under HPUX using GNU tools hppa1.1-hp-hpux 386 CPUs running SCO Unix 3.2v4 i386-unknown-sco3.2v4 * New targets supported AMD 29k family via UDI a29k-amd-udi or udi29k * New file formats supported BFD now supports reading HP/PA-RISC executables (SOM file format?), HPUX core files, and SCO 3.2v2 core files. * Major bug fixes Attaching to processes now works again; thanks for the many bug reports. We have also stomped on a bunch of core dumps caused by printf_filtered("%s") problems. We eliminated a copyright problem on the rpc and ptrace header files for VxWorks, which was discovered at the last minute during the 4.7 release. You should now be able to build a VxWorks GDB. You can now interrupt gdb while an attached process is running. This will cause the attached process to stop, and give control back to GDB. We fixed problems caused by using too many file descriptors for reading symbols from object files and libraries. This was especially a problem for programs that used many (~100) shared libraries. The `step' command now only enters a subroutine if there is line number information for the subroutine. Otherwise it acts like the `next' command. Previously, `step' would enter subroutines if there was any debugging information about the routine. This avoids problems when using `cc -g1' on MIPS machines. * Internal improvements GDB's internal interfaces have been improved to make it easier to support debugging of multiple languages in the future. GDB now uses a common structure for symbol information internally. Minimal symbols (derived from linkage symbols in object files), partial symbols (from a quick scan of debug information), and full symbols contain a common subset of information, making it easier to write shared code that handles any of them. * New command line options We now accept --silent as an alias for --quiet. * Mmalloc licensing The memory-mapped-malloc library is now licensed under the GNU Library General Public License. *** Changes in GDB-4.7: * Host/native/target split GDB has had some major internal surgery to untangle the support for hosts and remote targets. Now, when you configure GDB for a remote target, it will no longer load in all of the support for debugging local programs on the host. When fully completed and tested, this will ensure that arbitrary host/target combinations are possible. The primary conceptual shift is to separate the non-portable code in GDB into three categories. Host specific code is required any time GDB is compiled on that host, regardless of the target. Target specific code relates to the peculiarities of the target, but can be compiled on any host. Native specific code is everything else: it can only be built when the host and target are the same system. Child process handling and core file support are two common `native' examples. GDB's use of /proc for controlling Unix child processes is now cleaner. It has been split out into a single module under the `target_ops' vector, plus two native-dependent functions for each system that uses /proc. * New hosts supported HP/Apollo 68k (under the BSD domain) m68k-apollo-bsd or apollo68bsd 386 CPUs running various BSD ports i386-unknown-bsd or 386bsd 386 CPUs running SCO Unix i386-unknown-scosysv322 or i386sco * New targets supported Fujitsu SPARClite sparclite-fujitsu-none or sparclite 68030 and CPU32 m68030-*-*, m68332-*-* * New native hosts supported 386 CPUs running various BSD ports i386-unknown-bsd or 386bsd (386bsd is not well tested yet) 386 CPUs running SCO Unix i386-unknown-scosysv322 or sco * New file formats supported BFD now supports COFF files for the Zilog Z8000 microprocessor. It supports reading of `a.out.adobe' object files, which are an a.out format extended with minimal information about multiple sections. * New commands `show copying' is the same as the old `info copying'. `show warranty' is the same as `info warrantee'. These were renamed for consistency. The old commands continue to work. `info handle' is a new alias for `info signals'. You can now define pre-command hooks, which attach arbitrary command scripts to any command. The commands in the hook will be executed prior to the user's command. You can also create a hook which will be executed whenever the program stops. See gdb.texinfo. * C++ improvements We now deal with Cfront style name mangling, and can even extract type info from mangled symbols. GDB can automatically figure out which symbol mangling style your C++ compiler uses. Calling of methods and virtual functions has been improved as well. * Major bug fixes The crash that occurred when debugging Sun Ansi-C compiled binaries is fixed. This was due to mishandling of the extra N_SO stabs output by the compiler. We also finally got Ultrix 4.2 running in house, and fixed core file support, with help from a dozen people on the net. John M. Farrell discovered that the reason that single-stepping was so slow on all of the Mips based platforms (primarily SGI and DEC) was that we were trying to demangle and lookup a symbol used for internal purposes on every instruction that was being stepped through. Changing the name of that symbol so that it couldn't be mistaken for a C++ mangled symbol sped things up a great deal. Rich Pixley sped up symbol lookups in general by getting much smarter about when C++ symbol mangling is necessary. This should make symbol completion (TAB on the command line) much faster. It's not as fast as we'd like, but it's significantly faster than gdb-4.6. * AMD 29k support A new user controllable variable 'call_scratch_address' can specify the location of a scratch area to be used when GDB calls a function in the target. This is necessary because the usual method of putting the scratch area on the stack does not work in systems that have separate instruction and data spaces. We integrated changes to support the 29k UDI (Universal Debugger Interface), but discovered at the last minute that we didn't have all of the appropriate copyright paperwork. We are working with AMD to resolve this, and hope to have it available soon. * Remote interfaces We have sped up the remote serial line protocol, especially for targets with lots of registers. It now supports a new `expedited status' ('T') message which can be used in place of the existing 'S' status message. This allows the remote stub to send only the registers that GDB needs to make a quick decision about single-stepping or conditional breakpoints, eliminating the need to fetch the entire register set for each instruction being stepped through. The GDB remote serial protocol now implements a write-through cache for registers, only re-reading the registers if the target has run. There is also a new remote serial stub for SPARC processors. You can find it in gdb-4.7/gdb/sparc-stub.c. This was written to support the Fujitsu SPARClite processor, but will run on any stand-alone SPARC processor with a serial port. * Configuration Configure.in files have become much easier to read and modify. A new `table driven' format makes it more obvious what configurations are supported, and what files each one uses. * Library changes There is a new opcodes library which will eventually contain all of the disassembly routines and opcode tables. At present, it only contains Sparc and Z8000 routines. This will allow the assembler, debugger, and disassembler (binutils/objdump) to share these routines. The libiberty library is now copylefted under the GNU Library General Public License. This allows more liberal use, and was done so libg++ can use it. This makes no difference to GDB, since the Library License grants all the rights from the General Public License. * Documentation The file gdb-4.7/gdb/doc/stabs.texinfo is a (relatively) complete reference to the stabs symbol info used by the debugger. It is (as far as we know) the only published document on this fascinating topic. We encourage you to read it, compare it to the stabs information on your system, and send improvements on the document in general (to bug-gdb@prep.ai.mit.edu). And, of course, many bugs have been fixed. *** Changes in GDB-4.6: * Better support for C++ function names GDB now accepts as input the "demangled form" of C++ overloaded function names and member function names, and can do command completion on such names (using TAB, TAB-TAB, and ESC-?). The names have to be quoted with a pair of single quotes. Examples are 'func (int, long)' and 'obj::operator==(obj&)'. Make use of command completion, it is your friend. GDB also now accepts a variety of C++ mangled symbol formats. They are the GNU g++ style, the Cfront (ARM) style, and the Lucid (lcc) style. You can tell GDB which format to use by doing a 'set demangle-style {gnu, lucid, cfront, auto}'. 'gnu' is the default. Do a 'set demangle-style foo' for the list of formats. * G++ symbol mangling problem Recent versions of gcc have a bug in how they emit debugging information for C++ methods (when using dbx-style stabs). The file 'gcc.patch' (in this directory) can be applied to gcc to fix the problem. Alternatively, if you can't fix gcc, you can #define GCC_MANGLE_BUG when compiling gdb/symtab.c. The usual symptom is difficulty with setting breakpoints on methods. GDB complains about the method being non-existent. (We believe that version 2.2.2 of GCC has this problem.) * New 'maintenance' command All of the commands related to hacking GDB internals have been moved out of the main command set, and now live behind the 'maintenance' command. This can also be abbreviated as 'mt'. The following changes were made: dump-me -> maintenance dump-me info all-breakpoints -> maintenance info breakpoints printmsyms -> maintenance print msyms printobjfiles -> maintenance print objfiles printpsyms -> maintenance print psymbols printsyms -> maintenance print symbols The following commands are new: maintenance demangle Call internal GDB demangler routine to demangle a C++ link name and prints the result. maintenance print type Print a type chain for a given symbol * Change to .gdbinit file processing We now read the $HOME/.gdbinit file before processing the argv arguments (e.g. reading symbol files or core files). This allows global parameters to be set, which will apply during the symbol reading. The ./.gdbinit is still read after argv processing. * New hosts supported Solaris-2.0 !!! sparc-sun-solaris2 or sun4sol2 GNU/Linux support i386-unknown-linux or linux We are also including code to support the HP/PA running BSD and HPUX. This is almost guaranteed not to work, as we didn't have time to test or build it for this release. We are including it so that the more adventurous (or masochistic) of you can play with it. We also had major problems with the fact that the compiler that we got from HP doesn't support the -g option. It costs extra. * New targets supported Hitachi H8/300 h8300-hitachi-hms or h8300hms * More smarts about finding #include files GDB now remembers the compilation directory for all include files, and for all files from which C is generated (like yacc and lex sources). This greatly improves GDB's ability to find yacc/lex sources, and include files, especially if you are debugging your program from a directory different from the one that contains your sources. We also fixed a bug which caused difficulty with listing and setting breakpoints in include files which contain C code. (In the past, you had to try twice in order to list an include file that you hadn't looked at before.) * Interesting infernals change GDB now deals with arbitrary numbers of sections, where the symbols for each section must be relocated relative to that section's landing place in the target's address space. This work was needed to support ELF with embedded stabs used by Solaris-2.0. * Bug fixes (of course!) There have been loads of fixes for the following things: mips, rs6000, 29k/udi, m68k, g++, type handling, elf/dwarf, m88k, i960, stabs, DOS(GO32), procfs, etc... See the ChangeLog for details. *** Changes in GDB-4.5: * New machines supported (host and target) IBM RS6000 running AIX rs6000-ibm-aix or rs6000 SGI Irix-4.x mips-sgi-irix4 or iris4 * New malloc package GDB now uses a new memory manager called mmalloc, based on gmalloc. Mmalloc is capable of handling multiple heaps of memory. It is also capable of saving a heap to a file, and then mapping it back in later. This can be used to greatly speedup the startup of GDB by using a pre-parsed symbol table which lives in a mmalloc managed heap. For more details, please read mmalloc/mmalloc.texi. * info proc The 'info proc' command (SVR4 only) has been enhanced quite a bit. See 'help info proc' for details. * MIPS ecoff symbol table format The code that reads MIPS symbol table format is now supported on all hosts. Thanks to MIPS for releasing the sym.h and symconst.h files to make this possible. * File name changes for MS-DOS Many files in the config directories have been renamed to make it easier to support GDB on MS-DOSe systems (which have very restrictive file name conventions :-( ). MS-DOSe host support (under DJ Delorie's GO32 environment) is close to working but has some remaining problems. Note that debugging of DOS programs is not supported, due to limitations in the ``operating system'', but it can be used to host cross-debugging. * Cross byte order fixes Many fixes have been made to support cross debugging of Sparc and MIPS targets from hosts whose byte order differs. * New -mapped and -readnow options If memory-mapped files are available on your system through the 'mmap' system call, you can use the -mapped option on the `file' or `symbol-file' commands to cause GDB to write the symbols from your program into a reusable file. If the program you are debugging is called `/path/fred', the mapped symbol file will be `./fred.syms'. Future GDB debugging sessions will notice the presence of this file, and will quickly map in symbol information from it, rather than reading the symbol table from the executable program. Using the '-mapped' option in a GDB `file' or `symbol-file' command has the same effect as starting GDB with the '-mapped' command-line option. You can cause GDB to read the entire symbol table immediately by using the '-readnow' option with any of the commands that load symbol table information (or on the GDB command line). This makes the command slower, but makes future operations faster. The -mapped and -readnow options are typically combined in order to build a `fred.syms' file that contains complete symbol information. A simple GDB invocation to do nothing but build a `.syms' file for future use is: gdb -batch -nx -mapped -readnow programname The `.syms' file is specific to the host machine on which GDB is run. It holds an exact image of GDB's internal symbol table. It cannot be shared across multiple host platforms. * longjmp() handling GDB is now capable of stepping and nexting over longjmp(), _longjmp(), and siglongjmp() without losing control. This feature has not yet been ported to all systems. It currently works on many 386 platforms, all MIPS-based platforms (SGI, DECstation, etc), and Sun3/4. * Solaris 2.0 Preliminary work has been put in to support the new Solaris OS from Sun. At this time, it can control and debug processes, but it is not capable of reading symbols. * Bug fixes As always, many many bug fixes. The major areas were with g++, and mipsread. People using the MIPS-based platforms should experience fewer mysterious crashes and trashed symbol tables. *** Changes in GDB-4.4: * New machines supported (host and target) SCO Unix on i386 IBM PC clones i386-sco-sysv or i386sco (except core files) BSD Reno on Vax vax-dec-bsd Ultrix on Vax vax-dec-ultrix * New machines supported (target) AMD 29000 embedded, using EBMON a29k-none-none * C++ support GDB continues to improve its handling of C++. `References' work better. The demangler has also been improved, and now deals with symbols mangled as per the Annotated C++ Reference Guide. GDB also now handles `stabs' symbol information embedded in MIPS `ecoff' symbol tables. Since the ecoff format was not easily extensible to handle new languages such as C++, this appeared to be a good way to put C++ debugging info into MIPS binaries. This option will be supported in the GNU C compiler, version 2, when it is released. * New features for SVR4 GDB now handles SVR4 shared libraries, in the same fashion as SunOS shared libraries. Debugging dynamically linked programs should present only minor differences from debugging statically linked programs. The `info proc' command will print out information about any process on an SVR4 system (including the one you are debugging). At the moment, it prints the address mappings of the process. If you bring up GDB on another SVR4 system, please send mail to bug-gdb@prep.ai.mit.edu to let us know what changes were reqired (if any). * Better dynamic linking support in SunOS Reading symbols from shared libraries which contain debugging symbols now works properly. However, there remain issues such as automatic skipping of `transfer vector' code during function calls, which make it harder to debug code in a shared library, than to debug the same code linked statically. * New Getopt GDB is now using the latest `getopt' routines from the FSF. This version accepts the -- prefix for options with long names. GDB will continue to accept the old forms (-option and +option) as well. Various single letter abbreviations for options have been explicity added to the option table so that they won't get overshadowed in the future by other options that begin with the same letter. * Bugs fixed The `cleanup_undefined_types' bug that many of you noticed has been squashed. Many assorted bugs have been handled. Many more remain to be handled. See the various ChangeLog files (primarily in gdb and bfd) for details. *** Changes in GDB-4.3: * New machines supported (host and target) Amiga 3000 running Amix m68k-cbm-svr4 or amix NCR 3000 386 running SVR4 i386-ncr-svr4 or ncr3000 Motorola Delta 88000 running Sys V m88k-motorola-sysv or delta88 * Almost SCO Unix support We had hoped to support: SCO Unix on i386 IBM PC clones i386-sco-sysv or i386sco (except for core file support), but we discovered very late in the release that it has problems with process groups that render gdb unusable. Sorry about that. I encourage people to fix it and post the fixes. * Preliminary ELF and DWARF support GDB can read ELF object files on System V Release 4, and can handle debugging records for C, in DWARF format, in ELF files. This support is preliminary. If you bring up GDB on another SVR4 system, please send mail to bug-gdb@prep.ai.mit.edu to let us know what changes were reqired (if any). * New Readline GDB now uses the latest `readline' library. One user-visible change is that two tabs will list possible command completions, which previously required typing M-? (meta-question mark, or ESC ?). * Bugs fixed The `stepi' bug that many of you noticed has been squashed. Many bugs in C++ have been handled. Many more remain to be handled. See the various ChangeLog files (primarily in gdb and bfd) for details. * State of the MIPS world (in case you wondered): GDB can understand the symbol tables emitted by the compilers supplied by most vendors of MIPS-based machines, including DEC. These symbol tables are in a format that essentially nobody else uses. Some versions of gcc come with an assembler post-processor called mips-tfile. This program is required if you want to do source-level debugging of gcc-compiled programs. I believe FSF does not ship mips-tfile with gcc version 1, but it will eventually come with gcc version 2. Debugging of g++ output remains a problem. g++ version 1.xx does not really support it at all. (If you're lucky, you should be able to get line numbers and stack traces to work, but no parameters or local variables.) With some work it should be possible to improve the situation somewhat. When gcc version 2 is released, you will have somewhat better luck. However, even then you will get confusing results for inheritance and methods. We will eventually provide full debugging of g++ output on DECstations. This will probably involve some kind of stabs-in-ecoff encapulation, but the details have not been worked out yet. *** Changes in GDB-4.2: * Improved configuration Only one copy of `configure' exists now, and it is not self-modifying. Porting BFD is simpler. * Stepping improved The `step' and `next' commands now only stop at the first instruction of a source line. This prevents the multiple stops that used to occur in switch statements, for-loops, etc. `Step' continues to stop if a function that has debugging information is called within the line. * Bug fixing Lots of small bugs fixed. More remain. * New host supported (not target) Intel 386 PC clone running Mach i386-none-mach *** Changes in GDB-4.1: * Multiple source language support GDB now has internal scaffolding to handle several source languages. It determines the type of each source file from its filename extension, and will switch expression parsing and number formatting to match the language of the function in the currently selected stack frame. You can also specifically set the language to be used, with `set language c' or `set language modula-2'. * GDB and Modula-2 GDB now has preliminary support for the GNU Modula-2 compiler, currently under development at the State University of New York at Buffalo. Development of both GDB and the GNU Modula-2 compiler will continue through the fall of 1991 and into 1992. Other Modula-2 compilers are currently not supported, and attempting to debug programs compiled with them will likely result in an error as the symbol table is read. Feel free to work on it, though! There are hooks in GDB for strict type checking and range checking, in the `Modula-2 philosophy', but they do not currently work. * set write on/off GDB can now write to executable and core files (e.g. patch a variable's value). You must turn this switch on, specify the file ("exec foo" or "core foo"), *then* modify it, e.g. by assigning a new value to a variable. Modifications take effect immediately. * Automatic SunOS shared library reading When you run your program, GDB automatically determines where its shared libraries (if any) have been loaded, and reads their symbols. The `share' command is no longer needed. This also works when examining core files. * set listsize You can specify the number of lines that the `list' command shows. The default is 10. * New machines supported (host and target) SGI Iris (MIPS) running Irix V3: mips-sgi-irix or iris Sony NEWS (68K) running NEWSOS 3.x: m68k-sony-sysv or news Ultracomputer (29K) running Sym1: a29k-nyu-sym1 or ultra3 * New hosts supported (not targets) IBM RT/PC: romp-ibm-aix or rtpc * New targets supported (not hosts) AMD 29000 embedded with COFF a29k-none-coff AMD 29000 embedded with a.out a29k-none-aout Ultracomputer remote kernel debug a29k-nyu-kern * New remote interfaces AMD 29000 Adapt AMD 29000 Minimon *** Changes in GDB-4.0: * New Facilities Wide output is wrapped at good places to make the output more readable. Gdb now supports cross-debugging from a host machine of one type to a target machine of another type. Communication with the target system is over serial lines. The ``target'' command handles connecting to the remote system; the ``load'' command will download a program into the remote system. Serial stubs for the m68k and i386 are provided. Gdb also supports debugging of realtime processes running under VxWorks, using SunRPC Remote Procedure Calls over TCP/IP to talk to a debugger stub on the target system. New CPUs supported include the AMD 29000 and Intel 960. GDB now reads object files and symbol tables via a ``binary file'' library, which allows a single copy of GDB to debug programs of multiple object file types such as a.out and coff. There is now a GDB reference card in "doc/refcard.tex". (Make targets refcard.dvi and refcard.ps are available to format it). * Control-Variable user interface simplified All variables that control the operation of the debugger can be set by the ``set'' command, and displayed by the ``show'' command. For example, ``set prompt new-gdb=>'' will change your prompt to new-gdb=>. ``Show prompt'' produces the response: Gdb's prompt is new-gdb=>. What follows are the NEW set commands. The command ``help set'' will print a complete list of old and new set commands. ``help set FOO'' will give a longer description of the variable FOO. ``show'' will show all of the variable descriptions and their current settings. confirm on/off: Enables warning questions for operations that are hard to recover from, e.g. rerunning the program while it is already running. Default is ON. editing on/off: Enables EMACS style command line editing of input. Previous lines can be recalled with control-P, the current line can be edited with control-B, you can search for commands with control-R, etc. Default is ON. history filename NAME: NAME is where the gdb command history will be stored. The default is .gdb_history, or the value of the environment variable GDBHISTFILE. history size N: The size, in commands, of the command history. The default is 256, or the value of the environment variable HISTSIZE. history save on/off: If this value is set to ON, the history file will be saved after exiting gdb. If set to OFF, the file will not be saved. The default is OFF. history expansion on/off: If this value is set to ON, then csh-like history expansion will be performed on command line input. The default is OFF. radix N: Sets the default radix for input and output. It can be set to 8, 10, or 16. Note that the argument to "radix" is interpreted in the current radix, so "set radix 10" is always a no-op. height N: This integer value is the number of lines on a page. Default is 24, the current `stty rows'' setting, or the ``li#'' setting from the termcap entry matching the environment variable TERM. width N: This integer value is the number of characters on a line. Default is 80, the current `stty cols'' setting, or the ``co#'' setting from the termcap entry matching the environment variable TERM. Note: ``set screensize'' is obsolete. Use ``set height'' and ``set width'' instead. print address on/off: Print memory addresses in various command displays, such as stack traces and structure values. Gdb looks more ``symbolic'' if you turn this off; it looks more ``machine level'' with it on. Default is ON. print array on/off: Prettyprint arrays. New convenient format! Default is OFF. print demangle on/off: Print C++ symbols in "source" form if on, "raw" form if off. print asm-demangle on/off: Same, for assembler level printouts like instructions. print vtbl on/off: Prettyprint C++ virtual function tables. Default is OFF. * Support for Epoch Environment. The epoch environment is a version of Emacs v18 with windowing. One new command, ``inspect'', is identical to ``print'', except that if you are running in the epoch environment, the value is printed in its own window. * Support for Shared Libraries GDB can now debug programs and core files that use SunOS shared libraries. Symbols from a shared library cannot be referenced before the shared library has been linked with the program (this happens after you type ``run'' and before the function main() is entered). At any time after this linking (including when examining core files from dynamically linked programs), gdb reads the symbols from each shared library when you type the ``sharedlibrary'' command. It can be abbreviated ``share''. sharedlibrary REGEXP: Load shared object library symbols for files matching a unix regular expression. No argument indicates to load symbols for all shared libraries. info sharedlibrary: Status of loaded shared libraries. * Watchpoints A watchpoint stops execution of a program whenever the value of an expression changes. Checking for this slows down execution tremendously whenever you are in the scope of the expression, but is quite useful for catching tough ``bit-spreader'' or pointer misuse problems. Some machines such as the 386 have hardware for doing this more quickly, and future versions of gdb will use this hardware. watch EXP: Set a watchpoint (breakpoint) for an expression. info watchpoints: Information about your watchpoints. delete N: Deletes watchpoint number N (same as breakpoints). disable N: Temporarily turns off watchpoint number N (same as breakpoints). enable N: Re-enables watchpoint number N (same as breakpoints). * C++ multiple inheritance When used with a GCC version 2 compiler, GDB supports multiple inheritance for C++ programs. * C++ exception handling Gdb now supports limited C++ exception handling. Besides the existing ability to breakpoint on an exception handler, gdb can breakpoint on the raising of an exception (before the stack is peeled back to the handler's context). catch FOO: If there is a FOO exception handler in the dynamic scope, set a breakpoint to catch exceptions which may be raised there. Multiple exceptions (``catch foo bar baz'') may be caught. info catch: Lists all exceptions which may be caught in the current stack frame. * Minor command changes The command ``call func (arg, arg, ...)'' now acts like the print command, except it does not print or save a value if the function's result is void. This is similar to dbx usage. The ``up'' and ``down'' commands now always print the frame they end up at; ``up-silently'' and `down-silently'' can be used in scripts to change frames without printing. * New directory command 'dir' now adds directories to the FRONT of the source search path. The path starts off empty. Source files that contain debug information about the directory in which they were compiled can be found even with an empty path; Sun CC and GCC include this information. If GDB can't find your source file in the current directory, type "dir .". * Configuring GDB for compilation For normal use, type ``./configure host''. See README or gdb.texinfo for more details. GDB now handles cross debugging. If you are remotely debugging between two different machines, type ``./configure host -target=targ''. Host is the machine where GDB will run; targ is the machine where the program that you are debugging will run.