hanchenye-llvm-project/lldb/source/Commands/CommandObjectProcess.cpp

1837 lines
66 KiB
C++
Raw Normal View History

//===-- CommandObjectProcess.cpp --------------------------------*- C++ -*-===//
//
// The LLVM Compiler Infrastructure
//
// This file is distributed under the University of Illinois Open Source
// License. See LICENSE.TXT for details.
//
//===----------------------------------------------------------------------===//
#include "CommandObjectProcess.h"
// C Includes
// C++ Includes
// Other libraries and framework includes
// Project includes
#include "lldb/Interpreter/Args.h"
#include "lldb/Interpreter/Options.h"
#include "lldb/Core/State.h"
Centralized a lot of the status information for processes, threads, and stack frame down in the lldb_private::Process, lldb_private::Thread, lldb_private::StackFrameList and the lldb_private::StackFrame classes. We had some command line commands that had duplicate versions of the process status output ("thread list" and "process status" for example). Removed the "file" command and placed it where it should have been: "target create". Made an alias for "file" to "target create" so we stay compatible with GDB commands. We can now have multple usable targets in lldb at the same time. This is nice for comparing two runs of a program or debugging more than one binary at the same time. The new command is "target select <target-idx>" and also to see a list of the current targets you can use the new "target list" command. The flow in a debug session can be: (lldb) target create /path/to/exe/a.out (lldb) breakpoint set --name main (lldb) run ... hit breakpoint (lldb) target create /bin/ls (lldb) run /tmp Process 36001 exited with status = 0 (0x00000000) (lldb) target list Current targets: target #0: /tmp/args/a.out ( arch=x86_64-apple-darwin, platform=localhost, pid=35999, state=stopped ) * target #1: /bin/ls ( arch=x86_64-apple-darwin, platform=localhost, pid=36001, state=exited ) (lldb) target select 0 Current targets: * target #0: /tmp/args/a.out ( arch=x86_64-apple-darwin, platform=localhost, pid=35999, state=stopped ) target #1: /bin/ls ( arch=x86_64-apple-darwin, platform=localhost, pid=36001, state=exited ) (lldb) bt * thread #1: tid = 0x2d03, 0x0000000100000b9a a.out`main + 42 at main.c:16, stop reason = breakpoint 1.1 frame #0: 0x0000000100000b9a a.out`main + 42 at main.c:16 frame #1: 0x0000000100000b64 a.out`start + 52 Above we created a target for "a.out" and ran and hit a breakpoint at "main". Then we created a new target for /bin/ls and ran it. Then we listed the targest and selected our original "a.out" program, so we showed two concurent debug sessions going on at the same time. llvm-svn: 129695
2011-04-18 16:33:37 +08:00
#include "lldb/Host/Host.h"
#include "lldb/Interpreter/CommandInterpreter.h"
#include "lldb/Interpreter/CommandReturnObject.h"
LLDB now has "Platform" plug-ins. Platform plug-ins are plug-ins that provide an interface to a local or remote debugging platform. By default each host OS that supports LLDB should be registering a "default" platform that will be used unless a new platform is selected. Platforms are responsible for things such as: - getting process information by name or by processs ID - finding platform files. This is useful for remote debugging where there is an SDK with files that might already or need to be cached for debug access. - getting a list of platform supported architectures in the exact order they should be selected. This helps the native x86 platform on MacOSX select the correct x86_64/i386 slice from universal binaries. - Connect to remote platforms for remote debugging - Resolving an executable including finding an executable inside platform specific bundles (macosx uses .app bundles that contain files) and also selecting the appropriate slice of universal files for a given platform. So by default there is always a local platform, but remote platforms can be connected to. I will soon be adding a new "platform" command that will support the following commands: (lldb) platform connect --name machine1 macosx connect://host:port Connected to "machine1" platform. (lldb) platform disconnect macosx This allows LLDB to be well setup to do remote debugging and also once connected process listing and finding for things like: (lldb) process attach --name x<TAB> The currently selected platform plug-in can now auto complete any available processes that start with "x". The responsibilities for the platform plug-in will soon grow and expand. llvm-svn: 127286
2011-03-09 06:40:15 +08:00
#include "lldb/Target/Platform.h"
#include "lldb/Target/Process.h"
#include "lldb/Target/Target.h"
#include "lldb/Target/Thread.h"
using namespace lldb;
using namespace lldb_private;
//-------------------------------------------------------------------------
// CommandObjectProcessLaunch
//-------------------------------------------------------------------------
#pragma mark CommandObjectProjectLaunch
class CommandObjectProcessLaunch : public CommandObject
{
public:
class CommandOptions : public Options
{
public:
CommandOptions (CommandInterpreter &interpreter) :
Options(interpreter)
{
Added two new classes for command options: lldb_private::OptionGroup lldb_private::OptionGroupOptions OptionGroup lets you define a class that encapsulates settings that you want to reuse in multiple commands. It contains only the option definitions and the ability to set the option values, but it doesn't directly interface with the lldb_private::Options class that is the front end to all of the CommandObject option parsing. For that the OptionGroupOptions class can be used. It aggregates one or more OptionGroup objects and directs the option setting to the appropriate OptionGroup class. For an example of this, take a look at the CommandObjectFile and how it uses its "m_option_group" object shown below to be able to set values in both the FileOptionGroup and PlatformOptionGroup classes. The members used in CommandObjectFile are: OptionGroupOptions m_option_group; FileOptionGroup m_file_options; PlatformOptionGroup m_platform_options; Then in the constructor for CommandObjectFile you can combine the option settings. The code below shows a simplified version of the constructor: CommandObjectFile::CommandObjectFile(CommandInterpreter &interpreter) : CommandObject (...), m_option_group (interpreter), m_file_options (), m_platform_options(true) { m_option_group.Append (&m_file_options); m_option_group.Append (&m_platform_options); m_option_group.Finalize(); } We append the m_file_options and then the m_platform_options and then tell the option group the finalize the results. This allows the m_option_group to become the organizer of our prefs and after option parsing we end up with valid preference settings in both the m_file_options and m_platform_options objects. This also allows any other commands to use the FileOptionGroup and PlatformOptionGroup classes to implement options for their commands. Renamed: virtual void Options::ResetOptionValues(); to: virtual void Options::OptionParsingStarting(); And implemented a new callback named: virtual Error Options::OptionParsingFinished(); This allows Options subclasses to verify that the options all go together after all of the options have been specified and gives the chance for the command object to return an error. It also gives a chance to take all of the option values and produce or initialize objects after all options have completed parsing. Modfied: virtual Error SetOptionValue (int option_idx, const char *option_arg) = 0; to be: virtual Error SetOptionValue (uint32_t option_idx, const char *option_arg) = 0; (option_idx is now unsigned). llvm-svn: 129415
2011-04-13 08:18:08 +08:00
// Keep default values of all options in one place: OptionParsingStarting ()
OptionParsingStarting ();
}
~CommandOptions ()
{
}
Error
Added two new classes for command options: lldb_private::OptionGroup lldb_private::OptionGroupOptions OptionGroup lets you define a class that encapsulates settings that you want to reuse in multiple commands. It contains only the option definitions and the ability to set the option values, but it doesn't directly interface with the lldb_private::Options class that is the front end to all of the CommandObject option parsing. For that the OptionGroupOptions class can be used. It aggregates one or more OptionGroup objects and directs the option setting to the appropriate OptionGroup class. For an example of this, take a look at the CommandObjectFile and how it uses its "m_option_group" object shown below to be able to set values in both the FileOptionGroup and PlatformOptionGroup classes. The members used in CommandObjectFile are: OptionGroupOptions m_option_group; FileOptionGroup m_file_options; PlatformOptionGroup m_platform_options; Then in the constructor for CommandObjectFile you can combine the option settings. The code below shows a simplified version of the constructor: CommandObjectFile::CommandObjectFile(CommandInterpreter &interpreter) : CommandObject (...), m_option_group (interpreter), m_file_options (), m_platform_options(true) { m_option_group.Append (&m_file_options); m_option_group.Append (&m_platform_options); m_option_group.Finalize(); } We append the m_file_options and then the m_platform_options and then tell the option group the finalize the results. This allows the m_option_group to become the organizer of our prefs and after option parsing we end up with valid preference settings in both the m_file_options and m_platform_options objects. This also allows any other commands to use the FileOptionGroup and PlatformOptionGroup classes to implement options for their commands. Renamed: virtual void Options::ResetOptionValues(); to: virtual void Options::OptionParsingStarting(); And implemented a new callback named: virtual Error Options::OptionParsingFinished(); This allows Options subclasses to verify that the options all go together after all of the options have been specified and gives the chance for the command object to return an error. It also gives a chance to take all of the option values and produce or initialize objects after all options have completed parsing. Modfied: virtual Error SetOptionValue (int option_idx, const char *option_arg) = 0; to be: virtual Error SetOptionValue (uint32_t option_idx, const char *option_arg) = 0; (option_idx is now unsigned). llvm-svn: 129415
2011-04-13 08:18:08 +08:00
SetOptionValue (uint32_t option_idx, const char *option_arg)
{
Error error;
char short_option = (char) m_getopt_table[option_idx].val;
switch (short_option)
{
Added a new variant of SBTarget::Launch() that deprectates the old one that takes separate file handles for stdin, stdout, and stder and also allows for the working directory to be specified. Added support to "process launch" to a new option: --working-dir=PATH. We can now set the working directory. If this is not set, it defaults to that of the process that has LLDB loaded. Added the working directory to the host LaunchInNewTerminal function to allows the current working directory to be set in processes that are spawned in their own terminal. Also hooked this up to the lldb_private::Process and all mac plug-ins. The linux plug-in had its API changed, but nothing is making use of it yet. Modfied "debugserver" and "darwin-debug" to also handle the current working directory options and modified the code in LLDB that spawns these tools to pass the info along. Fixed ProcessGDBRemote to properly pass along all file handles for stdin, stdout and stderr. After clearing the default values for the stdin/out/err file handles for process to be NULL, we had a crasher in UserSettingsController::UpdateStringVariable which is now fixed. Also fixed the setting of boolean values to be able to be set as "true", "yes", "on", "1" for true (case insensitive) and "false", "no", "off", or "0" for false. Fixed debugserver to properly handle files for STDIN, STDOUT and STDERR that are not already opened. Previous to this fix debugserver would only correctly open and dupe file handles for the slave side of a pseudo terminal. It now correctly handles getting STDIN for the inferior from a file, and spitting STDOUT and STDERR out to files. Also made sure the file handles were correctly opened with the NOCTTY flag for terminals. llvm-svn: 124060
2011-01-23 13:56:20 +08:00
case 's': stop_at_entry = true; break;
case 'e': stderr_path.assign (option_arg); break;
case 'i': stdin_path.assign (option_arg); break;
case 'o': stdout_path.assign (option_arg); break;
case 'p': plugin_name.assign (option_arg); break;
case 'n': no_stdio = true; break;
case 'w': working_dir.assign (option_arg); break;
case 't':
if (option_arg && option_arg[0])
tty_name.assign (option_arg);
in_new_tty = true;
break;
default:
error.SetErrorStringWithFormat("invalid short option character '%c'", short_option);
break;
}
return error;
}
void
Added two new classes for command options: lldb_private::OptionGroup lldb_private::OptionGroupOptions OptionGroup lets you define a class that encapsulates settings that you want to reuse in multiple commands. It contains only the option definitions and the ability to set the option values, but it doesn't directly interface with the lldb_private::Options class that is the front end to all of the CommandObject option parsing. For that the OptionGroupOptions class can be used. It aggregates one or more OptionGroup objects and directs the option setting to the appropriate OptionGroup class. For an example of this, take a look at the CommandObjectFile and how it uses its "m_option_group" object shown below to be able to set values in both the FileOptionGroup and PlatformOptionGroup classes. The members used in CommandObjectFile are: OptionGroupOptions m_option_group; FileOptionGroup m_file_options; PlatformOptionGroup m_platform_options; Then in the constructor for CommandObjectFile you can combine the option settings. The code below shows a simplified version of the constructor: CommandObjectFile::CommandObjectFile(CommandInterpreter &interpreter) : CommandObject (...), m_option_group (interpreter), m_file_options (), m_platform_options(true) { m_option_group.Append (&m_file_options); m_option_group.Append (&m_platform_options); m_option_group.Finalize(); } We append the m_file_options and then the m_platform_options and then tell the option group the finalize the results. This allows the m_option_group to become the organizer of our prefs and after option parsing we end up with valid preference settings in both the m_file_options and m_platform_options objects. This also allows any other commands to use the FileOptionGroup and PlatformOptionGroup classes to implement options for their commands. Renamed: virtual void Options::ResetOptionValues(); to: virtual void Options::OptionParsingStarting(); And implemented a new callback named: virtual Error Options::OptionParsingFinished(); This allows Options subclasses to verify that the options all go together after all of the options have been specified and gives the chance for the command object to return an error. It also gives a chance to take all of the option values and produce or initialize objects after all options have completed parsing. Modfied: virtual Error SetOptionValue (int option_idx, const char *option_arg) = 0; to be: virtual Error SetOptionValue (uint32_t option_idx, const char *option_arg) = 0; (option_idx is now unsigned). llvm-svn: 129415
2011-04-13 08:18:08 +08:00
OptionParsingStarting ()
{
stop_at_entry = false;
in_new_tty = false;
tty_name.clear();
stdin_path.clear();
stdout_path.clear();
stderr_path.clear();
plugin_name.clear();
Added a new variant of SBTarget::Launch() that deprectates the old one that takes separate file handles for stdin, stdout, and stder and also allows for the working directory to be specified. Added support to "process launch" to a new option: --working-dir=PATH. We can now set the working directory. If this is not set, it defaults to that of the process that has LLDB loaded. Added the working directory to the host LaunchInNewTerminal function to allows the current working directory to be set in processes that are spawned in their own terminal. Also hooked this up to the lldb_private::Process and all mac plug-ins. The linux plug-in had its API changed, but nothing is making use of it yet. Modfied "debugserver" and "darwin-debug" to also handle the current working directory options and modified the code in LLDB that spawns these tools to pass the info along. Fixed ProcessGDBRemote to properly pass along all file handles for stdin, stdout and stderr. After clearing the default values for the stdin/out/err file handles for process to be NULL, we had a crasher in UserSettingsController::UpdateStringVariable which is now fixed. Also fixed the setting of boolean values to be able to be set as "true", "yes", "on", "1" for true (case insensitive) and "false", "no", "off", or "0" for false. Fixed debugserver to properly handle files for STDIN, STDOUT and STDERR that are not already opened. Previous to this fix debugserver would only correctly open and dupe file handles for the slave side of a pseudo terminal. It now correctly handles getting STDIN for the inferior from a file, and spitting STDOUT and STDERR out to files. Also made sure the file handles were correctly opened with the NOCTTY flag for terminals. llvm-svn: 124060
2011-01-23 13:56:20 +08:00
working_dir.clear();
no_stdio = false;
}
const OptionDefinition*
GetDefinitions ()
{
return g_option_table;
}
// Options table: Required for subclasses of Options.
static OptionDefinition g_option_table[];
// Instance variables to hold the values for command options.
bool stop_at_entry;
bool in_new_tty;
bool no_stdio;
std::string tty_name;
std::string stderr_path;
std::string stdin_path;
std::string stdout_path;
std::string plugin_name;
Added a new variant of SBTarget::Launch() that deprectates the old one that takes separate file handles for stdin, stdout, and stder and also allows for the working directory to be specified. Added support to "process launch" to a new option: --working-dir=PATH. We can now set the working directory. If this is not set, it defaults to that of the process that has LLDB loaded. Added the working directory to the host LaunchInNewTerminal function to allows the current working directory to be set in processes that are spawned in their own terminal. Also hooked this up to the lldb_private::Process and all mac plug-ins. The linux plug-in had its API changed, but nothing is making use of it yet. Modfied "debugserver" and "darwin-debug" to also handle the current working directory options and modified the code in LLDB that spawns these tools to pass the info along. Fixed ProcessGDBRemote to properly pass along all file handles for stdin, stdout and stderr. After clearing the default values for the stdin/out/err file handles for process to be NULL, we had a crasher in UserSettingsController::UpdateStringVariable which is now fixed. Also fixed the setting of boolean values to be able to be set as "true", "yes", "on", "1" for true (case insensitive) and "false", "no", "off", or "0" for false. Fixed debugserver to properly handle files for STDIN, STDOUT and STDERR that are not already opened. Previous to this fix debugserver would only correctly open and dupe file handles for the slave side of a pseudo terminal. It now correctly handles getting STDIN for the inferior from a file, and spitting STDOUT and STDERR out to files. Also made sure the file handles were correctly opened with the NOCTTY flag for terminals. llvm-svn: 124060
2011-01-23 13:56:20 +08:00
std::string working_dir;
};
CommandObjectProcessLaunch (CommandInterpreter &interpreter) :
CommandObject (interpreter,
"process launch",
"Launch the executable in the debugger.",
NULL),
m_options (interpreter)
{
CommandArgumentEntry arg;
CommandArgumentData run_args_arg;
// Define the first (and only) variant of this arg.
run_args_arg.arg_type = eArgTypeRunArgs;
run_args_arg.arg_repetition = eArgRepeatOptional;
// There is only one variant this argument could be; put it into the argument entry.
arg.push_back (run_args_arg);
// Push the data for the first argument into the m_arguments vector.
m_arguments.push_back (arg);
}
~CommandObjectProcessLaunch ()
{
}
Options *
GetOptions ()
{
return &m_options;
}
bool
Execute (Args& launch_args, CommandReturnObject &result)
{
Target *target = m_interpreter.GetDebugger().GetSelectedTarget().get();
if (target == NULL)
{
result.AppendError ("invalid target, create a debug target using the 'target create' command");
result.SetStatus (eReturnStatusFailed);
return false;
}
// If our listener is NULL, users aren't allows to launch
char filename[PATH_MAX];
const Module *exe_module = target->GetExecutableModulePointer();
if (exe_module == NULL)
{
result.AppendError ("no file in target, create a debug target using the 'target create' command");
result.SetStatus (eReturnStatusFailed);
return false;
}
exe_module->GetFileSpec().GetPath(filename, sizeof(filename));
StateType state = eStateInvalid;
Process *process = m_interpreter.GetExecutionContext().GetProcessPtr();
if (process)
{
state = process->GetState();
if (process->IsAlive() && state != eStateConnected)
{
char message[1024];
if (process->GetState() == eStateAttaching)
::strncpy (message, "There is a pending attach, abort it and launch a new process?", sizeof(message));
else
::strncpy (message, "There is a running process, kill it and restart?", sizeof(message));
if (!m_interpreter.Confirm (message, true))
{
result.SetStatus (eReturnStatusFailed);
return false;
}
else
{
Error error (process->Destroy());
if (error.Success())
{
result.SetStatus (eReturnStatusSuccessFinishResult);
}
else
{
result.AppendErrorWithFormat ("Failed to kill process: %s\n", error.AsCString());
result.SetStatus (eReturnStatusFailed);
}
}
}
}
if (state != eStateConnected)
{
const char *plugin_name;
if (!m_options.plugin_name.empty())
plugin_name = m_options.plugin_name.c_str();
else
plugin_name = NULL;
process = target->CreateProcess (m_interpreter.GetDebugger().GetListener(), plugin_name).get();
if (process == NULL)
{
result.AppendErrorWithFormat ("Failed to find a process plugin for executable.\n");
result.SetStatus (eReturnStatusFailed);
return false;
}
}
// If no launch args were given on the command line, then use any that
// might have been set using the "run-args" set variable.
if (launch_args.GetArgumentCount() == 0)
{
if (process->GetRunArguments().GetArgumentCount() > 0)
launch_args = process->GetRunArguments();
}
if (m_options.in_new_tty)
{
if (state == eStateConnected)
{
result.AppendWarning("launch in tty option is ignored when launching through a remote connection");
m_options.in_new_tty = false;
}
else
{
char exec_file_path[PATH_MAX];
if (exe_module->GetFileSpec().GetPath(exec_file_path, sizeof(exec_file_path)))
{
launch_args.InsertArgumentAtIndex(0, exec_file_path);
}
else
{
result.AppendError("invalid executable");
result.SetStatus (eReturnStatusFailed);
return false;
}
}
}
Args environment;
process->GetEnvironmentAsArgs (environment);
uint32_t launch_flags = eLaunchFlagNone;
if (process->GetDisableASLR())
launch_flags |= eLaunchFlagDisableASLR;
Moved the execution context that was in the Debugger into the CommandInterpreter where it was always being used. Make sure that Modules can track their object file offsets correctly to allow opening of sub object files (like the "__commpage" on darwin). Modified the Platforms to be able to launch processes. The first part of this move is the platform soon will become the entity that launches your program and when it does, it uses a new ProcessLaunchInfo class which encapsulates all process launching settings. This simplifies the internal APIs needed for launching. I want to slowly phase out process launching from the process classes, so for now we can still launch just as we used to, but eventually the platform is the object that should do the launching. Modified the Host::LaunchProcess in the MacOSX Host.mm to correctly be able to launch processes with all of the new eLaunchFlag settings. Modified any code that was manually launching processes to use the Host::LaunchProcess functions. Fixed an issue where lldb_private::Args had implicitly defined copy constructors that could do the wrong thing. This has now been fixed by adding an appropriate copy constructor and assignment operator. Make sure we don't add empty ModuleSP entries to a module list. Fixed the commpage module creation on MacOSX, but we still need to train the MacOSX dynamic loader to not get rid of it when it doesn't have an entry in the all image infos. Abstracted many more calls from in ProcessGDBRemote down into the GDBRemoteCommunicationClient subclass to make the classes cleaner and more efficient. Fixed the default iOS ARM register context to be correct and also added support for targets that don't support the qThreadStopInfo packet by selecting the current thread (only if needed) and then sending a stop reply packet. Debugserver can now start up with a --unix-socket (-u for short) and can then bind to port zero and send the port it bound to to a listening process on the other end. This allows the GDB remote platform to spawn new GDB server instances (debugserver) to allow platform debugging. llvm-svn: 129351
2011-04-12 13:54:46 +08:00
if (m_options.in_new_tty)
launch_flags |= eLaunchFlagLaunchInTTY;
if (m_options.no_stdio)
launch_flags |= eLaunchFlagDisableSTDIO;
else if (!m_options.in_new_tty
&& m_options.stdin_path.empty()
&& m_options.stdout_path.empty()
&& m_options.stderr_path.empty())
{
// Only use the settings value if the user hasn't specified any options that would override it.
if (process->GetDisableSTDIO())
launch_flags |= eLaunchFlagDisableSTDIO;
}
const char **inferior_argv = launch_args.GetArgumentCount() ? launch_args.GetConstArgumentVector() : NULL;
const char **inferior_envp = environment.GetArgumentCount() ? environment.GetConstArgumentVector() : NULL;
Error error;
Added a new variant of SBTarget::Launch() that deprectates the old one that takes separate file handles for stdin, stdout, and stder and also allows for the working directory to be specified. Added support to "process launch" to a new option: --working-dir=PATH. We can now set the working directory. If this is not set, it defaults to that of the process that has LLDB loaded. Added the working directory to the host LaunchInNewTerminal function to allows the current working directory to be set in processes that are spawned in their own terminal. Also hooked this up to the lldb_private::Process and all mac plug-ins. The linux plug-in had its API changed, but nothing is making use of it yet. Modfied "debugserver" and "darwin-debug" to also handle the current working directory options and modified the code in LLDB that spawns these tools to pass the info along. Fixed ProcessGDBRemote to properly pass along all file handles for stdin, stdout and stderr. After clearing the default values for the stdin/out/err file handles for process to be NULL, we had a crasher in UserSettingsController::UpdateStringVariable which is now fixed. Also fixed the setting of boolean values to be able to be set as "true", "yes", "on", "1" for true (case insensitive) and "false", "no", "off", or "0" for false. Fixed debugserver to properly handle files for STDIN, STDOUT and STDERR that are not already opened. Previous to this fix debugserver would only correctly open and dupe file handles for the slave side of a pseudo terminal. It now correctly handles getting STDIN for the inferior from a file, and spitting STDOUT and STDERR out to files. Also made sure the file handles were correctly opened with the NOCTTY flag for terminals. llvm-svn: 124060
2011-01-23 13:56:20 +08:00
const char *working_dir = NULL;
if (!m_options.working_dir.empty())
working_dir = m_options.working_dir.c_str();
Moved the execution context that was in the Debugger into the CommandInterpreter where it was always being used. Make sure that Modules can track their object file offsets correctly to allow opening of sub object files (like the "__commpage" on darwin). Modified the Platforms to be able to launch processes. The first part of this move is the platform soon will become the entity that launches your program and when it does, it uses a new ProcessLaunchInfo class which encapsulates all process launching settings. This simplifies the internal APIs needed for launching. I want to slowly phase out process launching from the process classes, so for now we can still launch just as we used to, but eventually the platform is the object that should do the launching. Modified the Host::LaunchProcess in the MacOSX Host.mm to correctly be able to launch processes with all of the new eLaunchFlag settings. Modified any code that was manually launching processes to use the Host::LaunchProcess functions. Fixed an issue where lldb_private::Args had implicitly defined copy constructors that could do the wrong thing. This has now been fixed by adding an appropriate copy constructor and assignment operator. Make sure we don't add empty ModuleSP entries to a module list. Fixed the commpage module creation on MacOSX, but we still need to train the MacOSX dynamic loader to not get rid of it when it doesn't have an entry in the all image infos. Abstracted many more calls from in ProcessGDBRemote down into the GDBRemoteCommunicationClient subclass to make the classes cleaner and more efficient. Fixed the default iOS ARM register context to be correct and also added support for targets that don't support the qThreadStopInfo packet by selecting the current thread (only if needed) and then sending a stop reply packet. Debugserver can now start up with a --unix-socket (-u for short) and can then bind to port zero and send the port it bound to to a listening process on the other end. This allows the GDB remote platform to spawn new GDB server instances (debugserver) to allow platform debugging. llvm-svn: 129351
2011-04-12 13:54:46 +08:00
const char * stdin_path = NULL;
const char * stdout_path = NULL;
const char * stderr_path = NULL;
// Were any standard input/output/error paths given on the command line?
if (m_options.stdin_path.empty() &&
m_options.stdout_path.empty() &&
m_options.stderr_path.empty())
{
Moved the execution context that was in the Debugger into the CommandInterpreter where it was always being used. Make sure that Modules can track their object file offsets correctly to allow opening of sub object files (like the "__commpage" on darwin). Modified the Platforms to be able to launch processes. The first part of this move is the platform soon will become the entity that launches your program and when it does, it uses a new ProcessLaunchInfo class which encapsulates all process launching settings. This simplifies the internal APIs needed for launching. I want to slowly phase out process launching from the process classes, so for now we can still launch just as we used to, but eventually the platform is the object that should do the launching. Modified the Host::LaunchProcess in the MacOSX Host.mm to correctly be able to launch processes with all of the new eLaunchFlag settings. Modified any code that was manually launching processes to use the Host::LaunchProcess functions. Fixed an issue where lldb_private::Args had implicitly defined copy constructors that could do the wrong thing. This has now been fixed by adding an appropriate copy constructor and assignment operator. Make sure we don't add empty ModuleSP entries to a module list. Fixed the commpage module creation on MacOSX, but we still need to train the MacOSX dynamic loader to not get rid of it when it doesn't have an entry in the all image infos. Abstracted many more calls from in ProcessGDBRemote down into the GDBRemoteCommunicationClient subclass to make the classes cleaner and more efficient. Fixed the default iOS ARM register context to be correct and also added support for targets that don't support the qThreadStopInfo packet by selecting the current thread (only if needed) and then sending a stop reply packet. Debugserver can now start up with a --unix-socket (-u for short) and can then bind to port zero and send the port it bound to to a listening process on the other end. This allows the GDB remote platform to spawn new GDB server instances (debugserver) to allow platform debugging. llvm-svn: 129351
2011-04-12 13:54:46 +08:00
// No standard file handles were given on the command line, check
// with the process object in case they were give using "set settings"
stdin_path = process->GetStandardInputPath();
stdout_path = process->GetStandardOutputPath();
stderr_path = process->GetStandardErrorPath();
}
else
{
Moved the execution context that was in the Debugger into the CommandInterpreter where it was always being used. Make sure that Modules can track their object file offsets correctly to allow opening of sub object files (like the "__commpage" on darwin). Modified the Platforms to be able to launch processes. The first part of this move is the platform soon will become the entity that launches your program and when it does, it uses a new ProcessLaunchInfo class which encapsulates all process launching settings. This simplifies the internal APIs needed for launching. I want to slowly phase out process launching from the process classes, so for now we can still launch just as we used to, but eventually the platform is the object that should do the launching. Modified the Host::LaunchProcess in the MacOSX Host.mm to correctly be able to launch processes with all of the new eLaunchFlag settings. Modified any code that was manually launching processes to use the Host::LaunchProcess functions. Fixed an issue where lldb_private::Args had implicitly defined copy constructors that could do the wrong thing. This has now been fixed by adding an appropriate copy constructor and assignment operator. Make sure we don't add empty ModuleSP entries to a module list. Fixed the commpage module creation on MacOSX, but we still need to train the MacOSX dynamic loader to not get rid of it when it doesn't have an entry in the all image infos. Abstracted many more calls from in ProcessGDBRemote down into the GDBRemoteCommunicationClient subclass to make the classes cleaner and more efficient. Fixed the default iOS ARM register context to be correct and also added support for targets that don't support the qThreadStopInfo packet by selecting the current thread (only if needed) and then sending a stop reply packet. Debugserver can now start up with a --unix-socket (-u for short) and can then bind to port zero and send the port it bound to to a listening process on the other end. This allows the GDB remote platform to spawn new GDB server instances (debugserver) to allow platform debugging. llvm-svn: 129351
2011-04-12 13:54:46 +08:00
stdin_path = m_options.stdin_path.empty() ? NULL : m_options.stdin_path.c_str();
stdout_path = m_options.stdout_path.empty() ? NULL : m_options.stdout_path.c_str();
stderr_path = m_options.stderr_path.empty() ? NULL : m_options.stderr_path.c_str();
}
Moved the execution context that was in the Debugger into the CommandInterpreter where it was always being used. Make sure that Modules can track their object file offsets correctly to allow opening of sub object files (like the "__commpage" on darwin). Modified the Platforms to be able to launch processes. The first part of this move is the platform soon will become the entity that launches your program and when it does, it uses a new ProcessLaunchInfo class which encapsulates all process launching settings. This simplifies the internal APIs needed for launching. I want to slowly phase out process launching from the process classes, so for now we can still launch just as we used to, but eventually the platform is the object that should do the launching. Modified the Host::LaunchProcess in the MacOSX Host.mm to correctly be able to launch processes with all of the new eLaunchFlag settings. Modified any code that was manually launching processes to use the Host::LaunchProcess functions. Fixed an issue where lldb_private::Args had implicitly defined copy constructors that could do the wrong thing. This has now been fixed by adding an appropriate copy constructor and assignment operator. Make sure we don't add empty ModuleSP entries to a module list. Fixed the commpage module creation on MacOSX, but we still need to train the MacOSX dynamic loader to not get rid of it when it doesn't have an entry in the all image infos. Abstracted many more calls from in ProcessGDBRemote down into the GDBRemoteCommunicationClient subclass to make the classes cleaner and more efficient. Fixed the default iOS ARM register context to be correct and also added support for targets that don't support the qThreadStopInfo packet by selecting the current thread (only if needed) and then sending a stop reply packet. Debugserver can now start up with a --unix-socket (-u for short) and can then bind to port zero and send the port it bound to to a listening process on the other end. This allows the GDB remote platform to spawn new GDB server instances (debugserver) to allow platform debugging. llvm-svn: 129351
2011-04-12 13:54:46 +08:00
error = process->Launch (inferior_argv,
inferior_envp,
launch_flags,
stdin_path,
stdout_path,
stderr_path,
working_dir);
if (error.Success())
{
Abtracted all mach-o and ELF out of ArchSpec. This patch is a modified form of Stephen Wilson's idea (thanks for the input Stephen!). What I ended up doing was: - Got rid of ArchSpec::CPU (which was a generic CPU enumeration that mimics the contents of llvm::Triple::ArchType). We now rely upon the llvm::Triple to give us the machine type from llvm::Triple::ArchType. - There is a new ArchSpec::Core definition which further qualifies the CPU core we are dealing with into a single enumeration. If you need support for a new Core and want to debug it in LLDB, it must be added to this list. In the future we can allow for dynamic core registration, but for now it is hard coded. - The ArchSpec can now be initialized with a llvm::Triple or with a C string that represents the triple (it can just be an arch still like "i386"). - The ArchSpec can still initialize itself with a architecture type -- mach-o with cpu type and subtype, or ELF with e_machine + e_flags -- and this will then get translated into the internal llvm::Triple::ArchSpec + ArchSpec::Core. The mach-o cpu type and subtype can be accessed using the getter functions: uint32_t ArchSpec::GetMachOCPUType () const; uint32_t ArchSpec::GetMachOCPUSubType () const; But these functions are just converting out internal llvm::Triple::ArchSpec + ArchSpec::Core back into mach-o. Same goes for ELF. All code has been updated to deal with the changes. This should abstract us until later when the llvm::TargetSpec stuff gets finalized and we can then adopt it. llvm-svn: 126278
2011-02-23 08:35:02 +08:00
const char *archname = exe_module->GetArchitecture().GetArchitectureName();
result.AppendMessageWithFormat ("Process %llu launched: '%s' (%s)\n", process->GetID(), filename, archname);
result.SetDidChangeProcessState (true);
if (m_options.stop_at_entry == false)
{
result.SetStatus (eReturnStatusSuccessContinuingNoResult);
StateType state = process->WaitForProcessToStop (NULL);
if (state == eStateStopped)
{
error = process->Resume();
if (error.Success())
{
bool synchronous_execution = m_interpreter.GetSynchronous ();
if (synchronous_execution)
{
state = process->WaitForProcessToStop (NULL);
Abtracted all mach-o and ELF out of ArchSpec. This patch is a modified form of Stephen Wilson's idea (thanks for the input Stephen!). What I ended up doing was: - Got rid of ArchSpec::CPU (which was a generic CPU enumeration that mimics the contents of llvm::Triple::ArchType). We now rely upon the llvm::Triple to give us the machine type from llvm::Triple::ArchType. - There is a new ArchSpec::Core definition which further qualifies the CPU core we are dealing with into a single enumeration. If you need support for a new Core and want to debug it in LLDB, it must be added to this list. In the future we can allow for dynamic core registration, but for now it is hard coded. - The ArchSpec can now be initialized with a llvm::Triple or with a C string that represents the triple (it can just be an arch still like "i386"). - The ArchSpec can still initialize itself with a architecture type -- mach-o with cpu type and subtype, or ELF with e_machine + e_flags -- and this will then get translated into the internal llvm::Triple::ArchSpec + ArchSpec::Core. The mach-o cpu type and subtype can be accessed using the getter functions: uint32_t ArchSpec::GetMachOCPUType () const; uint32_t ArchSpec::GetMachOCPUSubType () const; But these functions are just converting out internal llvm::Triple::ArchSpec + ArchSpec::Core back into mach-o. Same goes for ELF. All code has been updated to deal with the changes. This should abstract us until later when the llvm::TargetSpec stuff gets finalized and we can then adopt it. llvm-svn: 126278
2011-02-23 08:35:02 +08:00
if (!StateIsStoppedState(state))
{
result.AppendErrorWithFormat ("Process isn't stopped: %s", StateAsCString(state));
}
result.SetDidChangeProcessState (true);
result.SetStatus (eReturnStatusSuccessFinishResult);
}
else
{
result.SetStatus (eReturnStatusSuccessContinuingNoResult);
}
}
else
{
result.AppendErrorWithFormat ("Process resume at entry point failed: %s", error.AsCString());
result.SetStatus (eReturnStatusFailed);
}
}
else
{
result.AppendErrorWithFormat ("Initial process state wasn't stopped: %s", StateAsCString(state));
result.SetStatus (eReturnStatusFailed);
}
}
}
else
{
result.AppendErrorWithFormat ("process launch failed: %s", error.AsCString());
result.SetStatus (eReturnStatusFailed);
}
return result.Succeeded();
}
virtual const char *GetRepeatCommand (Args &current_command_args, uint32_t index)
{
// No repeat for "process launch"...
return "";
}
protected:
CommandOptions m_options;
};
#define SET1 LLDB_OPT_SET_1
#define SET2 LLDB_OPT_SET_2
#define SET3 LLDB_OPT_SET_3
OptionDefinition
CommandObjectProcessLaunch::CommandOptions::g_option_table[] =
{
{ SET1 | SET2 | SET3, false, "stop-at-entry", 's', no_argument, NULL, 0, eArgTypeNone, "Stop at the entry point of the program when launching a process."},
{ SET1 , false, "stdin", 'i', required_argument, NULL, 0, eArgTypePath, "Redirect stdin for the process to <path>."},
{ SET1 , false, "stdout", 'o', required_argument, NULL, 0, eArgTypePath, "Redirect stdout for the process to <path>."},
{ SET1 , false, "stderr", 'e', required_argument, NULL, 0, eArgTypePath, "Redirect stderr for the process to <path>."},
{ SET1 | SET2 | SET3, false, "plugin", 'p', required_argument, NULL, 0, eArgTypePlugin, "Name of the process plugin you want to use."},
{ SET2 , false, "tty", 't', optional_argument, NULL, 0, eArgTypePath, "Start the process in a terminal. If <path> is specified, look for a terminal whose name contains <path>, else start the process in a new terminal."},
{ SET3, false, "no-stdio", 'n', no_argument, NULL, 0, eArgTypeNone, "Do not set up for terminal I/O to go to running process."},
Added a new variant of SBTarget::Launch() that deprectates the old one that takes separate file handles for stdin, stdout, and stder and also allows for the working directory to be specified. Added support to "process launch" to a new option: --working-dir=PATH. We can now set the working directory. If this is not set, it defaults to that of the process that has LLDB loaded. Added the working directory to the host LaunchInNewTerminal function to allows the current working directory to be set in processes that are spawned in their own terminal. Also hooked this up to the lldb_private::Process and all mac plug-ins. The linux plug-in had its API changed, but nothing is making use of it yet. Modfied "debugserver" and "darwin-debug" to also handle the current working directory options and modified the code in LLDB that spawns these tools to pass the info along. Fixed ProcessGDBRemote to properly pass along all file handles for stdin, stdout and stderr. After clearing the default values for the stdin/out/err file handles for process to be NULL, we had a crasher in UserSettingsController::UpdateStringVariable which is now fixed. Also fixed the setting of boolean values to be able to be set as "true", "yes", "on", "1" for true (case insensitive) and "false", "no", "off", or "0" for false. Fixed debugserver to properly handle files for STDIN, STDOUT and STDERR that are not already opened. Previous to this fix debugserver would only correctly open and dupe file handles for the slave side of a pseudo terminal. It now correctly handles getting STDIN for the inferior from a file, and spitting STDOUT and STDERR out to files. Also made sure the file handles were correctly opened with the NOCTTY flag for terminals. llvm-svn: 124060
2011-01-23 13:56:20 +08:00
{ SET1 | SET2 | SET3, false, "working-dir", 'w', required_argument, NULL, 0, eArgTypePath, "Set the current working directory to <path> when running the inferior."},
{ 0, false, NULL, 0, 0, NULL, 0, eArgTypeNone, NULL }
};
#undef SET1
#undef SET2
#undef SET3
//-------------------------------------------------------------------------
// CommandObjectProcessAttach
//-------------------------------------------------------------------------
#pragma mark CommandObjectProcessAttach
class CommandObjectProcessAttach : public CommandObject
{
public:
class CommandOptions : public Options
{
public:
CommandOptions (CommandInterpreter &interpreter) :
Options(interpreter)
{
Added two new classes for command options: lldb_private::OptionGroup lldb_private::OptionGroupOptions OptionGroup lets you define a class that encapsulates settings that you want to reuse in multiple commands. It contains only the option definitions and the ability to set the option values, but it doesn't directly interface with the lldb_private::Options class that is the front end to all of the CommandObject option parsing. For that the OptionGroupOptions class can be used. It aggregates one or more OptionGroup objects and directs the option setting to the appropriate OptionGroup class. For an example of this, take a look at the CommandObjectFile and how it uses its "m_option_group" object shown below to be able to set values in both the FileOptionGroup and PlatformOptionGroup classes. The members used in CommandObjectFile are: OptionGroupOptions m_option_group; FileOptionGroup m_file_options; PlatformOptionGroup m_platform_options; Then in the constructor for CommandObjectFile you can combine the option settings. The code below shows a simplified version of the constructor: CommandObjectFile::CommandObjectFile(CommandInterpreter &interpreter) : CommandObject (...), m_option_group (interpreter), m_file_options (), m_platform_options(true) { m_option_group.Append (&m_file_options); m_option_group.Append (&m_platform_options); m_option_group.Finalize(); } We append the m_file_options and then the m_platform_options and then tell the option group the finalize the results. This allows the m_option_group to become the organizer of our prefs and after option parsing we end up with valid preference settings in both the m_file_options and m_platform_options objects. This also allows any other commands to use the FileOptionGroup and PlatformOptionGroup classes to implement options for their commands. Renamed: virtual void Options::ResetOptionValues(); to: virtual void Options::OptionParsingStarting(); And implemented a new callback named: virtual Error Options::OptionParsingFinished(); This allows Options subclasses to verify that the options all go together after all of the options have been specified and gives the chance for the command object to return an error. It also gives a chance to take all of the option values and produce or initialize objects after all options have completed parsing. Modfied: virtual Error SetOptionValue (int option_idx, const char *option_arg) = 0; to be: virtual Error SetOptionValue (uint32_t option_idx, const char *option_arg) = 0; (option_idx is now unsigned). llvm-svn: 129415
2011-04-13 08:18:08 +08:00
// Keep default values of all options in one place: OptionParsingStarting ()
OptionParsingStarting ();
}
~CommandOptions ()
{
}
Error
Added two new classes for command options: lldb_private::OptionGroup lldb_private::OptionGroupOptions OptionGroup lets you define a class that encapsulates settings that you want to reuse in multiple commands. It contains only the option definitions and the ability to set the option values, but it doesn't directly interface with the lldb_private::Options class that is the front end to all of the CommandObject option parsing. For that the OptionGroupOptions class can be used. It aggregates one or more OptionGroup objects and directs the option setting to the appropriate OptionGroup class. For an example of this, take a look at the CommandObjectFile and how it uses its "m_option_group" object shown below to be able to set values in both the FileOptionGroup and PlatformOptionGroup classes. The members used in CommandObjectFile are: OptionGroupOptions m_option_group; FileOptionGroup m_file_options; PlatformOptionGroup m_platform_options; Then in the constructor for CommandObjectFile you can combine the option settings. The code below shows a simplified version of the constructor: CommandObjectFile::CommandObjectFile(CommandInterpreter &interpreter) : CommandObject (...), m_option_group (interpreter), m_file_options (), m_platform_options(true) { m_option_group.Append (&m_file_options); m_option_group.Append (&m_platform_options); m_option_group.Finalize(); } We append the m_file_options and then the m_platform_options and then tell the option group the finalize the results. This allows the m_option_group to become the organizer of our prefs and after option parsing we end up with valid preference settings in both the m_file_options and m_platform_options objects. This also allows any other commands to use the FileOptionGroup and PlatformOptionGroup classes to implement options for their commands. Renamed: virtual void Options::ResetOptionValues(); to: virtual void Options::OptionParsingStarting(); And implemented a new callback named: virtual Error Options::OptionParsingFinished(); This allows Options subclasses to verify that the options all go together after all of the options have been specified and gives the chance for the command object to return an error. It also gives a chance to take all of the option values and produce or initialize objects after all options have completed parsing. Modfied: virtual Error SetOptionValue (int option_idx, const char *option_arg) = 0; to be: virtual Error SetOptionValue (uint32_t option_idx, const char *option_arg) = 0; (option_idx is now unsigned). llvm-svn: 129415
2011-04-13 08:18:08 +08:00
SetOptionValue (uint32_t option_idx, const char *option_arg)
{
Error error;
char short_option = (char) m_getopt_table[option_idx].val;
bool success = false;
switch (short_option)
{
case 'p':
pid = Args::StringToUInt32 (option_arg, LLDB_INVALID_PROCESS_ID, 0, &success);
if (!success || pid == LLDB_INVALID_PROCESS_ID)
{
error.SetErrorStringWithFormat("invalid process ID '%s'", option_arg);
}
break;
case 'P':
plugin_name = option_arg;
break;
case 'n':
name.assign(option_arg);
break;
case 'w':
waitfor = true;
break;
default:
error.SetErrorStringWithFormat("invalid short option character '%c'", short_option);
break;
}
return error;
}
void
Added two new classes for command options: lldb_private::OptionGroup lldb_private::OptionGroupOptions OptionGroup lets you define a class that encapsulates settings that you want to reuse in multiple commands. It contains only the option definitions and the ability to set the option values, but it doesn't directly interface with the lldb_private::Options class that is the front end to all of the CommandObject option parsing. For that the OptionGroupOptions class can be used. It aggregates one or more OptionGroup objects and directs the option setting to the appropriate OptionGroup class. For an example of this, take a look at the CommandObjectFile and how it uses its "m_option_group" object shown below to be able to set values in both the FileOptionGroup and PlatformOptionGroup classes. The members used in CommandObjectFile are: OptionGroupOptions m_option_group; FileOptionGroup m_file_options; PlatformOptionGroup m_platform_options; Then in the constructor for CommandObjectFile you can combine the option settings. The code below shows a simplified version of the constructor: CommandObjectFile::CommandObjectFile(CommandInterpreter &interpreter) : CommandObject (...), m_option_group (interpreter), m_file_options (), m_platform_options(true) { m_option_group.Append (&m_file_options); m_option_group.Append (&m_platform_options); m_option_group.Finalize(); } We append the m_file_options and then the m_platform_options and then tell the option group the finalize the results. This allows the m_option_group to become the organizer of our prefs and after option parsing we end up with valid preference settings in both the m_file_options and m_platform_options objects. This also allows any other commands to use the FileOptionGroup and PlatformOptionGroup classes to implement options for their commands. Renamed: virtual void Options::ResetOptionValues(); to: virtual void Options::OptionParsingStarting(); And implemented a new callback named: virtual Error Options::OptionParsingFinished(); This allows Options subclasses to verify that the options all go together after all of the options have been specified and gives the chance for the command object to return an error. It also gives a chance to take all of the option values and produce or initialize objects after all options have completed parsing. Modfied: virtual Error SetOptionValue (int option_idx, const char *option_arg) = 0; to be: virtual Error SetOptionValue (uint32_t option_idx, const char *option_arg) = 0; (option_idx is now unsigned). llvm-svn: 129415
2011-04-13 08:18:08 +08:00
OptionParsingStarting ()
{
pid = LLDB_INVALID_PROCESS_ID;
name.clear();
waitfor = false;
}
const OptionDefinition*
GetDefinitions ()
{
return g_option_table;
}
virtual bool
HandleOptionArgumentCompletion (Args &input,
int cursor_index,
int char_pos,
OptionElementVector &opt_element_vector,
int opt_element_index,
int match_start_point,
int max_return_elements,
bool &word_complete,
StringList &matches)
{
int opt_arg_pos = opt_element_vector[opt_element_index].opt_arg_pos;
int opt_defs_index = opt_element_vector[opt_element_index].opt_defs_index;
// We are only completing the name option for now...
const OptionDefinition *opt_defs = GetDefinitions();
if (opt_defs[opt_defs_index].short_option == 'n')
{
// Are we in the name?
// Look to see if there is a -P argument provided, and if so use that plugin, otherwise
// use the default plugin.
const char *partial_name = NULL;
partial_name = input.GetArgumentAtIndex(opt_arg_pos);
LLDB now has "Platform" plug-ins. Platform plug-ins are plug-ins that provide an interface to a local or remote debugging platform. By default each host OS that supports LLDB should be registering a "default" platform that will be used unless a new platform is selected. Platforms are responsible for things such as: - getting process information by name or by processs ID - finding platform files. This is useful for remote debugging where there is an SDK with files that might already or need to be cached for debug access. - getting a list of platform supported architectures in the exact order they should be selected. This helps the native x86 platform on MacOSX select the correct x86_64/i386 slice from universal binaries. - Connect to remote platforms for remote debugging - Resolving an executable including finding an executable inside platform specific bundles (macosx uses .app bundles that contain files) and also selecting the appropriate slice of universal files for a given platform. So by default there is always a local platform, but remote platforms can be connected to. I will soon be adding a new "platform" command that will support the following commands: (lldb) platform connect --name machine1 macosx connect://host:port Connected to "machine1" platform. (lldb) platform disconnect macosx This allows LLDB to be well setup to do remote debugging and also once connected process listing and finding for things like: (lldb) process attach --name x<TAB> The currently selected platform plug-in can now auto complete any available processes that start with "x". The responsibilities for the platform plug-in will soon grow and expand. llvm-svn: 127286
2011-03-09 06:40:15 +08:00
Moved the execution context that was in the Debugger into the CommandInterpreter where it was always being used. Make sure that Modules can track their object file offsets correctly to allow opening of sub object files (like the "__commpage" on darwin). Modified the Platforms to be able to launch processes. The first part of this move is the platform soon will become the entity that launches your program and when it does, it uses a new ProcessLaunchInfo class which encapsulates all process launching settings. This simplifies the internal APIs needed for launching. I want to slowly phase out process launching from the process classes, so for now we can still launch just as we used to, but eventually the platform is the object that should do the launching. Modified the Host::LaunchProcess in the MacOSX Host.mm to correctly be able to launch processes with all of the new eLaunchFlag settings. Modified any code that was manually launching processes to use the Host::LaunchProcess functions. Fixed an issue where lldb_private::Args had implicitly defined copy constructors that could do the wrong thing. This has now been fixed by adding an appropriate copy constructor and assignment operator. Make sure we don't add empty ModuleSP entries to a module list. Fixed the commpage module creation on MacOSX, but we still need to train the MacOSX dynamic loader to not get rid of it when it doesn't have an entry in the all image infos. Abstracted many more calls from in ProcessGDBRemote down into the GDBRemoteCommunicationClient subclass to make the classes cleaner and more efficient. Fixed the default iOS ARM register context to be correct and also added support for targets that don't support the qThreadStopInfo packet by selecting the current thread (only if needed) and then sending a stop reply packet. Debugserver can now start up with a --unix-socket (-u for short) and can then bind to port zero and send the port it bound to to a listening process on the other end. This allows the GDB remote platform to spawn new GDB server instances (debugserver) to allow platform debugging. llvm-svn: 129351
2011-04-12 13:54:46 +08:00
PlatformSP platform_sp (m_interpreter.GetPlatform (true));
LLDB now has "Platform" plug-ins. Platform plug-ins are plug-ins that provide an interface to a local or remote debugging platform. By default each host OS that supports LLDB should be registering a "default" platform that will be used unless a new platform is selected. Platforms are responsible for things such as: - getting process information by name or by processs ID - finding platform files. This is useful for remote debugging where there is an SDK with files that might already or need to be cached for debug access. - getting a list of platform supported architectures in the exact order they should be selected. This helps the native x86 platform on MacOSX select the correct x86_64/i386 slice from universal binaries. - Connect to remote platforms for remote debugging - Resolving an executable including finding an executable inside platform specific bundles (macosx uses .app bundles that contain files) and also selecting the appropriate slice of universal files for a given platform. So by default there is always a local platform, but remote platforms can be connected to. I will soon be adding a new "platform" command that will support the following commands: (lldb) platform connect --name machine1 macosx connect://host:port Connected to "machine1" platform. (lldb) platform disconnect macosx This allows LLDB to be well setup to do remote debugging and also once connected process listing and finding for things like: (lldb) process attach --name x<TAB> The currently selected platform plug-in can now auto complete any available processes that start with "x". The responsibilities for the platform plug-in will soon grow and expand. llvm-svn: 127286
2011-03-09 06:40:15 +08:00
if (platform_sp)
{
Moved the execution context that was in the Debugger into the CommandInterpreter where it was always being used. Make sure that Modules can track their object file offsets correctly to allow opening of sub object files (like the "__commpage" on darwin). Modified the Platforms to be able to launch processes. The first part of this move is the platform soon will become the entity that launches your program and when it does, it uses a new ProcessLaunchInfo class which encapsulates all process launching settings. This simplifies the internal APIs needed for launching. I want to slowly phase out process launching from the process classes, so for now we can still launch just as we used to, but eventually the platform is the object that should do the launching. Modified the Host::LaunchProcess in the MacOSX Host.mm to correctly be able to launch processes with all of the new eLaunchFlag settings. Modified any code that was manually launching processes to use the Host::LaunchProcess functions. Fixed an issue where lldb_private::Args had implicitly defined copy constructors that could do the wrong thing. This has now been fixed by adding an appropriate copy constructor and assignment operator. Make sure we don't add empty ModuleSP entries to a module list. Fixed the commpage module creation on MacOSX, but we still need to train the MacOSX dynamic loader to not get rid of it when it doesn't have an entry in the all image infos. Abstracted many more calls from in ProcessGDBRemote down into the GDBRemoteCommunicationClient subclass to make the classes cleaner and more efficient. Fixed the default iOS ARM register context to be correct and also added support for targets that don't support the qThreadStopInfo packet by selecting the current thread (only if needed) and then sending a stop reply packet. Debugserver can now start up with a --unix-socket (-u for short) and can then bind to port zero and send the port it bound to to a listening process on the other end. This allows the GDB remote platform to spawn new GDB server instances (debugserver) to allow platform debugging. llvm-svn: 129351
2011-04-12 13:54:46 +08:00
ProcessInstanceInfoList process_infos;
ProcessInstanceInfoMatch match_info;
Many improvements to the Platform base class and subclasses. The base Platform class now implements the Host functionality for a lot of things that make sense by default so that subclasses can check: int PlatformSubclass::Foo () { if (IsHost()) return Platform::Foo (); // Let the platform base class do the host specific stuff // Platform subclass specific code... int result = ... return result; } Added new functions to the platform: virtual const char *Platform::GetUserName (uint32_t uid); virtual const char *Platform::GetGroupName (uint32_t gid); The user and group names are cached locally so that remote platforms can avoid sending packets multiple times to resolve this information. Added the parent process ID to the ProcessInfo class. Added a new ProcessInfoMatch class which helps us to match processes up and changed the Host layer over to using this new class. The new class allows us to search for processs: 1 - by name (equal to, starts with, ends with, contains, and regex) 2 - by pid 3 - And further check for parent pid == value, uid == value, gid == value, euid == value, egid == value, arch == value, parent == value. This is all hookup up to the "platform process list" command which required adding dumping routines to dump process information. If the Host class implements the process lookup routines, you can now lists processes on your local machine: machine1.foo.com % lldb (lldb) platform process list PID PARENT USER GROUP EFF USER EFF GROUP TRIPLE NAME ====== ====== ========== ========== ========== ========== ======================== ============================ 99538 1 username usergroup username usergroup x86_64-apple-darwin FileMerge 94943 1 username usergroup username usergroup x86_64-apple-darwin mdworker 94852 244 username usergroup username usergroup x86_64-apple-darwin Safari 94727 244 username usergroup username usergroup x86_64-apple-darwin Xcode 92742 92710 username usergroup username usergroup i386-apple-darwin debugserver This of course also works remotely with the lldb-platform: machine1.foo.com % lldb-platform --listen 1234 machine2.foo.com % lldb (lldb) platform create remote-macosx Platform: remote-macosx Connected: no (lldb) platform connect connect://localhost:1444 Platform: remote-macosx Triple: x86_64-apple-darwin OS Version: 10.6.7 (10J869) Kernel: Darwin Kernel Version 10.7.0: Sat Jan 29 15:17:16 PST 2011; root:xnu-1504.9.37~1/RELEASE_I386 Hostname: machine1.foo.com Connected: yes (lldb) platform process list PID PARENT USER GROUP EFF USER EFF GROUP TRIPLE NAME ====== ====== ========== ========== ========== ========== ======================== ============================ 99556 244 username usergroup username usergroup x86_64-apple-darwin trustevaluation 99548 65539 username usergroup username usergroup x86_64-apple-darwin lldb 99538 1 username usergroup username usergroup x86_64-apple-darwin FileMerge 94943 1 username usergroup username usergroup x86_64-apple-darwin mdworker 94852 244 username usergroup username usergroup x86_64-apple-darwin Safari The lldb-platform implements everything with the Host:: layer, so this should "just work" for linux. I will probably be adding more stuff to the Host layer for launching processes and attaching to processes so that this support should eventually just work as well. Modified the target to be able to be created with an architecture that differs from the main executable. This is needed for iOS debugging since we can have an "armv6" binary which can run on an "armv7" machine, so we want to be able to do: % lldb (lldb) platform create remote-ios (lldb) file --arch armv7 a.out Where "a.out" is an armv6 executable. The platform then can correctly decide to open all "armv7" images for all dependent shared libraries. Modified the disassembly to show the current PC value. Example output: (lldb) disassemble --frame a.out`main: 0x1eb7: pushl %ebp 0x1eb8: movl %esp, %ebp 0x1eba: pushl %ebx 0x1ebb: subl $20, %esp 0x1ebe: calll 0x1ec3 ; main + 12 at test.c:18 0x1ec3: popl %ebx -> 0x1ec4: calll 0x1f12 ; getpid 0x1ec9: movl %eax, 4(%esp) 0x1ecd: leal 199(%ebx), %eax 0x1ed3: movl %eax, (%esp) 0x1ed6: calll 0x1f18 ; printf 0x1edb: leal 213(%ebx), %eax 0x1ee1: movl %eax, (%esp) 0x1ee4: calll 0x1f1e ; puts 0x1ee9: calll 0x1f0c ; getchar 0x1eee: movl $20, (%esp) 0x1ef5: calll 0x1e6a ; sleep_loop at test.c:6 0x1efa: movl $12, %eax 0x1eff: addl $20, %esp 0x1f02: popl %ebx 0x1f03: leave 0x1f04: ret This can be handy when dealing with the new --line options that was recently added: (lldb) disassemble --line a.out`main + 13 at test.c:19 18 { -> 19 printf("Process: %i\n\n", getpid()); 20 puts("Press any key to continue..."); getchar(); -> 0x1ec4: calll 0x1f12 ; getpid 0x1ec9: movl %eax, 4(%esp) 0x1ecd: leal 199(%ebx), %eax 0x1ed3: movl %eax, (%esp) 0x1ed6: calll 0x1f18 ; printf Modified the ModuleList to have a lookup based solely on a UUID. Since the UUID is typically the MD5 checksum of a binary image, there is no need to give the path and architecture when searching for a pre-existing image in an image list. Now that we support remote debugging a bit better, our lldb_private::Module needs to be able to track what the original path for file was as the platform knows it, as well as where the file is locally. The module has the two following functions to retrieve both paths: const FileSpec &Module::GetFileSpec () const; const FileSpec &Module::GetPlatformFileSpec () const; llvm-svn: 128563
2011-03-31 02:16:51 +08:00
if (partial_name)
{
match_info.GetProcessInfo().SetName(partial_name);
match_info.SetNameMatchType(eNameMatchStartsWith);
}
platform_sp->FindProcesses (match_info, process_infos);
LLDB now has "Platform" plug-ins. Platform plug-ins are plug-ins that provide an interface to a local or remote debugging platform. By default each host OS that supports LLDB should be registering a "default" platform that will be used unless a new platform is selected. Platforms are responsible for things such as: - getting process information by name or by processs ID - finding platform files. This is useful for remote debugging where there is an SDK with files that might already or need to be cached for debug access. - getting a list of platform supported architectures in the exact order they should be selected. This helps the native x86 platform on MacOSX select the correct x86_64/i386 slice from universal binaries. - Connect to remote platforms for remote debugging - Resolving an executable including finding an executable inside platform specific bundles (macosx uses .app bundles that contain files) and also selecting the appropriate slice of universal files for a given platform. So by default there is always a local platform, but remote platforms can be connected to. I will soon be adding a new "platform" command that will support the following commands: (lldb) platform connect --name machine1 macosx connect://host:port Connected to "machine1" platform. (lldb) platform disconnect macosx This allows LLDB to be well setup to do remote debugging and also once connected process listing and finding for things like: (lldb) process attach --name x<TAB> The currently selected platform plug-in can now auto complete any available processes that start with "x". The responsibilities for the platform plug-in will soon grow and expand. llvm-svn: 127286
2011-03-09 06:40:15 +08:00
const uint32_t num_matches = process_infos.GetSize();
if (num_matches > 0)
{
for (uint32_t i=0; i<num_matches; ++i)
{
matches.AppendString (process_infos.GetProcessNameAtIndex(i),
process_infos.GetProcessNameLengthAtIndex(i));
}
}
}
}
return false;
}
// Options table: Required for subclasses of Options.
static OptionDefinition g_option_table[];
// Instance variables to hold the values for command options.
lldb::pid_t pid;
std::string plugin_name;
std::string name;
bool waitfor;
};
CommandObjectProcessAttach (CommandInterpreter &interpreter) :
CommandObject (interpreter,
"process attach",
"Attach to a process.",
"process attach <cmd-options>"),
m_options (interpreter)
{
}
~CommandObjectProcessAttach ()
{
}
bool
Execute (Args& command,
CommandReturnObject &result)
{
Target *target = m_interpreter.GetDebugger().GetSelectedTarget().get();
// N.B. The attach should be synchronous. It doesn't help much to get the prompt back between initiating the attach
// and the target actually stopping. So even if the interpreter is set to be asynchronous, we wait for the stop
// ourselves here.
Process *process = m_interpreter.GetExecutionContext().GetProcessPtr();
StateType state = eStateInvalid;
if (process)
{
state = process->GetState();
if (process->IsAlive() && state != eStateConnected)
{
result.AppendErrorWithFormat ("Process %llu is currently being debugged, kill the process before attaching.\n",
process->GetID());
result.SetStatus (eReturnStatusFailed);
return false;
}
}
if (target == NULL)
{
// If there isn't a current target create one.
TargetSP new_target_sp;
FileSpec emptyFileSpec;
Error error;
error = m_interpreter.GetDebugger().GetTargetList().CreateTarget (m_interpreter.GetDebugger(),
emptyFileSpec,
NULL,
false,
NULL, // No platform options
new_target_sp);
target = new_target_sp.get();
if (target == NULL || error.Fail())
{
Added support for attaching to a remote debug server with the new command: (lldb) process connect <remote-url> Currently when you specify a file with the file command it helps us to find a process plug-in that is suitable for debugging. If you specify a file you can rely upon this to find the correct debugger plug-in: % lldb a.out Current executable set to 'a.out' (x86_64). (lldb) process connect connect://localhost:2345 ... If you don't specify a file, you will need to specify the plug-in name that you wish to use: % lldb (lldb) process connect --plugin process.gdb-remote connect://localhost:2345 Other connection URL examples: (lldb) process connect connect://localhost:2345 (lldb) process connect tcp://127.0.0.1 (lldb) process connect file:///dev/ttyS1 We are currently treating the "connect://host:port" as a way to do raw socket connections. If there is a URL for this already, please let me know and we will adopt it. So now you can connect to a remote debug server with the ProcessGDBRemote plug-in. After connection, it will ask for the pid info using the "qC" packet and if it responds with a valid process ID, it will be equivalent to attaching. If it response with an error or invalid process ID, the LLDB process will be in a new state: eStateConnected. This allows us to then download a program or specify the program to run (using the 'A' packet), or specify a process to attach to (using the "vAttach" packets), or query info about the processes that might be available. llvm-svn: 124846
2011-02-04 09:58:07 +08:00
result.AppendError(error.AsCString("Error creating target"));
return false;
}
m_interpreter.GetDebugger().GetTargetList().SetSelectedTarget(target);
}
// Record the old executable module, we want to issue a warning if the process of attaching changed the
// current executable (like somebody said "file foo" then attached to a PID whose executable was bar.)
ModuleSP old_exec_module_sp = target->GetExecutableModule();
ArchSpec old_arch_spec = target->GetArchitecture();
if (command.GetArgumentCount())
{
result.AppendErrorWithFormat("Invalid arguments for '%s'.\nUsage: %s\n", m_cmd_name.c_str(), m_cmd_syntax.c_str());
result.SetStatus (eReturnStatusFailed);
}
else
{
if (state != eStateConnected)
{
const char *plugin_name = NULL;
if (!m_options.plugin_name.empty())
plugin_name = m_options.plugin_name.c_str();
process = target->CreateProcess (m_interpreter.GetDebugger().GetListener(), plugin_name).get();
}
if (process)
{
Error error;
int attach_pid = m_options.pid;
const char *wait_name = NULL;
if (m_options.name.empty())
{
if (old_exec_module_sp)
{
wait_name = old_exec_module_sp->GetFileSpec().GetFilename().AsCString();
}
}
else
{
wait_name = m_options.name.c_str();
}
// If we are waiting for a process with this name to show up, do that first.
if (m_options.waitfor)
{
if (wait_name == NULL)
{
result.AppendError("Invalid arguments: must have a file loaded or supply a process name with the waitfor option.\n");
result.SetStatus (eReturnStatusFailed);
return false;
}
result.AppendMessageWithFormat("Waiting to attach to a process named \"%s\".\n", wait_name);
error = process->Attach (wait_name, m_options.waitfor);
if (error.Success())
{
result.SetStatus (eReturnStatusSuccessContinuingNoResult);
}
else
{
result.AppendErrorWithFormat ("Waiting for a process to launch named '%s': %s\n",
wait_name,
error.AsCString());
result.SetStatus (eReturnStatusFailed);
return false;
}
// If we're synchronous, wait for the stopped event and report that.
// Otherwise just return.
// FIXME: in the async case it will now be possible to get to the command
// interpreter with a state eStateAttaching. Make sure we handle that correctly.
StateType state = process->WaitForProcessToStop (NULL);
result.SetDidChangeProcessState (true);
result.AppendMessageWithFormat ("Process %llu %s\n", process->GetID(), StateAsCString (state));
result.SetStatus (eReturnStatusSuccessFinishNoResult);
}
else
{
// If the process was specified by name look it up, so we can warn if there are multiple
// processes with this pid.
if (attach_pid == LLDB_INVALID_PROCESS_ID && wait_name != NULL)
{
Moved the execution context that was in the Debugger into the CommandInterpreter where it was always being used. Make sure that Modules can track their object file offsets correctly to allow opening of sub object files (like the "__commpage" on darwin). Modified the Platforms to be able to launch processes. The first part of this move is the platform soon will become the entity that launches your program and when it does, it uses a new ProcessLaunchInfo class which encapsulates all process launching settings. This simplifies the internal APIs needed for launching. I want to slowly phase out process launching from the process classes, so for now we can still launch just as we used to, but eventually the platform is the object that should do the launching. Modified the Host::LaunchProcess in the MacOSX Host.mm to correctly be able to launch processes with all of the new eLaunchFlag settings. Modified any code that was manually launching processes to use the Host::LaunchProcess functions. Fixed an issue where lldb_private::Args had implicitly defined copy constructors that could do the wrong thing. This has now been fixed by adding an appropriate copy constructor and assignment operator. Make sure we don't add empty ModuleSP entries to a module list. Fixed the commpage module creation on MacOSX, but we still need to train the MacOSX dynamic loader to not get rid of it when it doesn't have an entry in the all image infos. Abstracted many more calls from in ProcessGDBRemote down into the GDBRemoteCommunicationClient subclass to make the classes cleaner and more efficient. Fixed the default iOS ARM register context to be correct and also added support for targets that don't support the qThreadStopInfo packet by selecting the current thread (only if needed) and then sending a stop reply packet. Debugserver can now start up with a --unix-socket (-u for short) and can then bind to port zero and send the port it bound to to a listening process on the other end. This allows the GDB remote platform to spawn new GDB server instances (debugserver) to allow platform debugging. llvm-svn: 129351
2011-04-12 13:54:46 +08:00
ProcessInstanceInfoList process_infos;
PlatformSP platform_sp (m_interpreter.GetPlatform (true));
LLDB now has "Platform" plug-ins. Platform plug-ins are plug-ins that provide an interface to a local or remote debugging platform. By default each host OS that supports LLDB should be registering a "default" platform that will be used unless a new platform is selected. Platforms are responsible for things such as: - getting process information by name or by processs ID - finding platform files. This is useful for remote debugging where there is an SDK with files that might already or need to be cached for debug access. - getting a list of platform supported architectures in the exact order they should be selected. This helps the native x86 platform on MacOSX select the correct x86_64/i386 slice from universal binaries. - Connect to remote platforms for remote debugging - Resolving an executable including finding an executable inside platform specific bundles (macosx uses .app bundles that contain files) and also selecting the appropriate slice of universal files for a given platform. So by default there is always a local platform, but remote platforms can be connected to. I will soon be adding a new "platform" command that will support the following commands: (lldb) platform connect --name machine1 macosx connect://host:port Connected to "machine1" platform. (lldb) platform disconnect macosx This allows LLDB to be well setup to do remote debugging and also once connected process listing and finding for things like: (lldb) process attach --name x<TAB> The currently selected platform plug-in can now auto complete any available processes that start with "x". The responsibilities for the platform plug-in will soon grow and expand. llvm-svn: 127286
2011-03-09 06:40:15 +08:00
if (platform_sp)
{
Moved the execution context that was in the Debugger into the CommandInterpreter where it was always being used. Make sure that Modules can track their object file offsets correctly to allow opening of sub object files (like the "__commpage" on darwin). Modified the Platforms to be able to launch processes. The first part of this move is the platform soon will become the entity that launches your program and when it does, it uses a new ProcessLaunchInfo class which encapsulates all process launching settings. This simplifies the internal APIs needed for launching. I want to slowly phase out process launching from the process classes, so for now we can still launch just as we used to, but eventually the platform is the object that should do the launching. Modified the Host::LaunchProcess in the MacOSX Host.mm to correctly be able to launch processes with all of the new eLaunchFlag settings. Modified any code that was manually launching processes to use the Host::LaunchProcess functions. Fixed an issue where lldb_private::Args had implicitly defined copy constructors that could do the wrong thing. This has now been fixed by adding an appropriate copy constructor and assignment operator. Make sure we don't add empty ModuleSP entries to a module list. Fixed the commpage module creation on MacOSX, but we still need to train the MacOSX dynamic loader to not get rid of it when it doesn't have an entry in the all image infos. Abstracted many more calls from in ProcessGDBRemote down into the GDBRemoteCommunicationClient subclass to make the classes cleaner and more efficient. Fixed the default iOS ARM register context to be correct and also added support for targets that don't support the qThreadStopInfo packet by selecting the current thread (only if needed) and then sending a stop reply packet. Debugserver can now start up with a --unix-socket (-u for short) and can then bind to port zero and send the port it bound to to a listening process on the other end. This allows the GDB remote platform to spawn new GDB server instances (debugserver) to allow platform debugging. llvm-svn: 129351
2011-04-12 13:54:46 +08:00
ProcessInstanceInfoMatch match_info (wait_name, eNameMatchEquals);
Many improvements to the Platform base class and subclasses. The base Platform class now implements the Host functionality for a lot of things that make sense by default so that subclasses can check: int PlatformSubclass::Foo () { if (IsHost()) return Platform::Foo (); // Let the platform base class do the host specific stuff // Platform subclass specific code... int result = ... return result; } Added new functions to the platform: virtual const char *Platform::GetUserName (uint32_t uid); virtual const char *Platform::GetGroupName (uint32_t gid); The user and group names are cached locally so that remote platforms can avoid sending packets multiple times to resolve this information. Added the parent process ID to the ProcessInfo class. Added a new ProcessInfoMatch class which helps us to match processes up and changed the Host layer over to using this new class. The new class allows us to search for processs: 1 - by name (equal to, starts with, ends with, contains, and regex) 2 - by pid 3 - And further check for parent pid == value, uid == value, gid == value, euid == value, egid == value, arch == value, parent == value. This is all hookup up to the "platform process list" command which required adding dumping routines to dump process information. If the Host class implements the process lookup routines, you can now lists processes on your local machine: machine1.foo.com % lldb (lldb) platform process list PID PARENT USER GROUP EFF USER EFF GROUP TRIPLE NAME ====== ====== ========== ========== ========== ========== ======================== ============================ 99538 1 username usergroup username usergroup x86_64-apple-darwin FileMerge 94943 1 username usergroup username usergroup x86_64-apple-darwin mdworker 94852 244 username usergroup username usergroup x86_64-apple-darwin Safari 94727 244 username usergroup username usergroup x86_64-apple-darwin Xcode 92742 92710 username usergroup username usergroup i386-apple-darwin debugserver This of course also works remotely with the lldb-platform: machine1.foo.com % lldb-platform --listen 1234 machine2.foo.com % lldb (lldb) platform create remote-macosx Platform: remote-macosx Connected: no (lldb) platform connect connect://localhost:1444 Platform: remote-macosx Triple: x86_64-apple-darwin OS Version: 10.6.7 (10J869) Kernel: Darwin Kernel Version 10.7.0: Sat Jan 29 15:17:16 PST 2011; root:xnu-1504.9.37~1/RELEASE_I386 Hostname: machine1.foo.com Connected: yes (lldb) platform process list PID PARENT USER GROUP EFF USER EFF GROUP TRIPLE NAME ====== ====== ========== ========== ========== ========== ======================== ============================ 99556 244 username usergroup username usergroup x86_64-apple-darwin trustevaluation 99548 65539 username usergroup username usergroup x86_64-apple-darwin lldb 99538 1 username usergroup username usergroup x86_64-apple-darwin FileMerge 94943 1 username usergroup username usergroup x86_64-apple-darwin mdworker 94852 244 username usergroup username usergroup x86_64-apple-darwin Safari The lldb-platform implements everything with the Host:: layer, so this should "just work" for linux. I will probably be adding more stuff to the Host layer for launching processes and attaching to processes so that this support should eventually just work as well. Modified the target to be able to be created with an architecture that differs from the main executable. This is needed for iOS debugging since we can have an "armv6" binary which can run on an "armv7" machine, so we want to be able to do: % lldb (lldb) platform create remote-ios (lldb) file --arch armv7 a.out Where "a.out" is an armv6 executable. The platform then can correctly decide to open all "armv7" images for all dependent shared libraries. Modified the disassembly to show the current PC value. Example output: (lldb) disassemble --frame a.out`main: 0x1eb7: pushl %ebp 0x1eb8: movl %esp, %ebp 0x1eba: pushl %ebx 0x1ebb: subl $20, %esp 0x1ebe: calll 0x1ec3 ; main + 12 at test.c:18 0x1ec3: popl %ebx -> 0x1ec4: calll 0x1f12 ; getpid 0x1ec9: movl %eax, 4(%esp) 0x1ecd: leal 199(%ebx), %eax 0x1ed3: movl %eax, (%esp) 0x1ed6: calll 0x1f18 ; printf 0x1edb: leal 213(%ebx), %eax 0x1ee1: movl %eax, (%esp) 0x1ee4: calll 0x1f1e ; puts 0x1ee9: calll 0x1f0c ; getchar 0x1eee: movl $20, (%esp) 0x1ef5: calll 0x1e6a ; sleep_loop at test.c:6 0x1efa: movl $12, %eax 0x1eff: addl $20, %esp 0x1f02: popl %ebx 0x1f03: leave 0x1f04: ret This can be handy when dealing with the new --line options that was recently added: (lldb) disassemble --line a.out`main + 13 at test.c:19 18 { -> 19 printf("Process: %i\n\n", getpid()); 20 puts("Press any key to continue..."); getchar(); -> 0x1ec4: calll 0x1f12 ; getpid 0x1ec9: movl %eax, 4(%esp) 0x1ecd: leal 199(%ebx), %eax 0x1ed3: movl %eax, (%esp) 0x1ed6: calll 0x1f18 ; printf Modified the ModuleList to have a lookup based solely on a UUID. Since the UUID is typically the MD5 checksum of a binary image, there is no need to give the path and architecture when searching for a pre-existing image in an image list. Now that we support remote debugging a bit better, our lldb_private::Module needs to be able to track what the original path for file was as the platform knows it, as well as where the file is locally. The module has the two following functions to retrieve both paths: const FileSpec &Module::GetFileSpec () const; const FileSpec &Module::GetPlatformFileSpec () const; llvm-svn: 128563
2011-03-31 02:16:51 +08:00
platform_sp->FindProcesses (match_info, process_infos);
LLDB now has "Platform" plug-ins. Platform plug-ins are plug-ins that provide an interface to a local or remote debugging platform. By default each host OS that supports LLDB should be registering a "default" platform that will be used unless a new platform is selected. Platforms are responsible for things such as: - getting process information by name or by processs ID - finding platform files. This is useful for remote debugging where there is an SDK with files that might already or need to be cached for debug access. - getting a list of platform supported architectures in the exact order they should be selected. This helps the native x86 platform on MacOSX select the correct x86_64/i386 slice from universal binaries. - Connect to remote platforms for remote debugging - Resolving an executable including finding an executable inside platform specific bundles (macosx uses .app bundles that contain files) and also selecting the appropriate slice of universal files for a given platform. So by default there is always a local platform, but remote platforms can be connected to. I will soon be adding a new "platform" command that will support the following commands: (lldb) platform connect --name machine1 macosx connect://host:port Connected to "machine1" platform. (lldb) platform disconnect macosx This allows LLDB to be well setup to do remote debugging and also once connected process listing and finding for things like: (lldb) process attach --name x<TAB> The currently selected platform plug-in can now auto complete any available processes that start with "x". The responsibilities for the platform plug-in will soon grow and expand. llvm-svn: 127286
2011-03-09 06:40:15 +08:00
}
if (process_infos.GetSize() > 1)
{
result.AppendErrorWithFormat("More than one process named %s\n", wait_name);
result.SetStatus (eReturnStatusFailed);
return false;
}
LLDB now has "Platform" plug-ins. Platform plug-ins are plug-ins that provide an interface to a local or remote debugging platform. By default each host OS that supports LLDB should be registering a "default" platform that will be used unless a new platform is selected. Platforms are responsible for things such as: - getting process information by name or by processs ID - finding platform files. This is useful for remote debugging where there is an SDK with files that might already or need to be cached for debug access. - getting a list of platform supported architectures in the exact order they should be selected. This helps the native x86 platform on MacOSX select the correct x86_64/i386 slice from universal binaries. - Connect to remote platforms for remote debugging - Resolving an executable including finding an executable inside platform specific bundles (macosx uses .app bundles that contain files) and also selecting the appropriate slice of universal files for a given platform. So by default there is always a local platform, but remote platforms can be connected to. I will soon be adding a new "platform" command that will support the following commands: (lldb) platform connect --name machine1 macosx connect://host:port Connected to "machine1" platform. (lldb) platform disconnect macosx This allows LLDB to be well setup to do remote debugging and also once connected process listing and finding for things like: (lldb) process attach --name x<TAB> The currently selected platform plug-in can now auto complete any available processes that start with "x". The responsibilities for the platform plug-in will soon grow and expand. llvm-svn: 127286
2011-03-09 06:40:15 +08:00
else if (process_infos.GetSize() == 0)
{
result.AppendErrorWithFormat("Could not find a process named %s\n", wait_name);
result.SetStatus (eReturnStatusFailed);
return false;
}
else
{
LLDB now has "Platform" plug-ins. Platform plug-ins are plug-ins that provide an interface to a local or remote debugging platform. By default each host OS that supports LLDB should be registering a "default" platform that will be used unless a new platform is selected. Platforms are responsible for things such as: - getting process information by name or by processs ID - finding platform files. This is useful for remote debugging where there is an SDK with files that might already or need to be cached for debug access. - getting a list of platform supported architectures in the exact order they should be selected. This helps the native x86 platform on MacOSX select the correct x86_64/i386 slice from universal binaries. - Connect to remote platforms for remote debugging - Resolving an executable including finding an executable inside platform specific bundles (macosx uses .app bundles that contain files) and also selecting the appropriate slice of universal files for a given platform. So by default there is always a local platform, but remote platforms can be connected to. I will soon be adding a new "platform" command that will support the following commands: (lldb) platform connect --name machine1 macosx connect://host:port Connected to "machine1" platform. (lldb) platform disconnect macosx This allows LLDB to be well setup to do remote debugging and also once connected process listing and finding for things like: (lldb) process attach --name x<TAB> The currently selected platform plug-in can now auto complete any available processes that start with "x". The responsibilities for the platform plug-in will soon grow and expand. llvm-svn: 127286
2011-03-09 06:40:15 +08:00
attach_pid = process_infos.GetProcessIDAtIndex (0);
}
}
if (attach_pid != LLDB_INVALID_PROCESS_ID)
{
error = process->Attach (attach_pid);
if (error.Success())
{
result.SetStatus (eReturnStatusSuccessContinuingNoResult);
}
else
{
result.AppendErrorWithFormat ("Attaching to process %i failed: %s.\n",
attach_pid,
error.AsCString());
result.SetStatus (eReturnStatusFailed);
}
StateType state = process->WaitForProcessToStop (NULL);
result.SetDidChangeProcessState (true);
result.AppendMessageWithFormat ("Process %llu %s\n", process->GetID(), StateAsCString (state));
result.SetStatus (eReturnStatusSuccessFinishNoResult);
}
else
{
result.AppendErrorWithFormat ("No PID specified for attach\n");
result.SetStatus (eReturnStatusFailed);
}
}
}
}
if (result.Succeeded())
{
// Okay, we're done. Last step is to warn if the executable module has changed:
char new_path[PATH_MAX];
ModuleSP new_exec_module_sp (target->GetExecutableModule());
if (!old_exec_module_sp)
{
// We might not have a module if we attached to a raw pid...
if (new_exec_module_sp)
{
new_exec_module_sp->GetFileSpec().GetPath(new_path, PATH_MAX);
result.AppendMessageWithFormat("Executable module set to \"%s\".\n", new_path);
}
}
else if (old_exec_module_sp->GetFileSpec() != new_exec_module_sp->GetFileSpec())
{
char old_path[PATH_MAX];
old_exec_module_sp->GetFileSpec().GetPath (old_path, PATH_MAX);
new_exec_module_sp->GetFileSpec().GetPath (new_path, PATH_MAX);
result.AppendWarningWithFormat("Executable module changed from \"%s\" to \"%s\".\n",
old_path, new_path);
}
if (!old_arch_spec.IsValid())
{
Abtracted all mach-o and ELF out of ArchSpec. This patch is a modified form of Stephen Wilson's idea (thanks for the input Stephen!). What I ended up doing was: - Got rid of ArchSpec::CPU (which was a generic CPU enumeration that mimics the contents of llvm::Triple::ArchType). We now rely upon the llvm::Triple to give us the machine type from llvm::Triple::ArchType. - There is a new ArchSpec::Core definition which further qualifies the CPU core we are dealing with into a single enumeration. If you need support for a new Core and want to debug it in LLDB, it must be added to this list. In the future we can allow for dynamic core registration, but for now it is hard coded. - The ArchSpec can now be initialized with a llvm::Triple or with a C string that represents the triple (it can just be an arch still like "i386"). - The ArchSpec can still initialize itself with a architecture type -- mach-o with cpu type and subtype, or ELF with e_machine + e_flags -- and this will then get translated into the internal llvm::Triple::ArchSpec + ArchSpec::Core. The mach-o cpu type and subtype can be accessed using the getter functions: uint32_t ArchSpec::GetMachOCPUType () const; uint32_t ArchSpec::GetMachOCPUSubType () const; But these functions are just converting out internal llvm::Triple::ArchSpec + ArchSpec::Core back into mach-o. Same goes for ELF. All code has been updated to deal with the changes. This should abstract us until later when the llvm::TargetSpec stuff gets finalized and we can then adopt it. llvm-svn: 126278
2011-02-23 08:35:02 +08:00
result.AppendMessageWithFormat ("Architecture set to: %s.\n", target->GetArchitecture().GetArchitectureName());
}
else if (old_arch_spec != target->GetArchitecture())
{
result.AppendWarningWithFormat("Architecture changed from %s to %s.\n",
Abtracted all mach-o and ELF out of ArchSpec. This patch is a modified form of Stephen Wilson's idea (thanks for the input Stephen!). What I ended up doing was: - Got rid of ArchSpec::CPU (which was a generic CPU enumeration that mimics the contents of llvm::Triple::ArchType). We now rely upon the llvm::Triple to give us the machine type from llvm::Triple::ArchType. - There is a new ArchSpec::Core definition which further qualifies the CPU core we are dealing with into a single enumeration. If you need support for a new Core and want to debug it in LLDB, it must be added to this list. In the future we can allow for dynamic core registration, but for now it is hard coded. - The ArchSpec can now be initialized with a llvm::Triple or with a C string that represents the triple (it can just be an arch still like "i386"). - The ArchSpec can still initialize itself with a architecture type -- mach-o with cpu type and subtype, or ELF with e_machine + e_flags -- and this will then get translated into the internal llvm::Triple::ArchSpec + ArchSpec::Core. The mach-o cpu type and subtype can be accessed using the getter functions: uint32_t ArchSpec::GetMachOCPUType () const; uint32_t ArchSpec::GetMachOCPUSubType () const; But these functions are just converting out internal llvm::Triple::ArchSpec + ArchSpec::Core back into mach-o. Same goes for ELF. All code has been updated to deal with the changes. This should abstract us until later when the llvm::TargetSpec stuff gets finalized and we can then adopt it. llvm-svn: 126278
2011-02-23 08:35:02 +08:00
old_arch_spec.GetArchitectureName(), target->GetArchitecture().GetArchitectureName());
}
}
return result.Succeeded();
}
Options *
GetOptions ()
{
return &m_options;
}
protected:
CommandOptions m_options;
};
OptionDefinition
CommandObjectProcessAttach::CommandOptions::g_option_table[] =
{
{ LLDB_OPT_SET_ALL, false, "plugin", 'P', required_argument, NULL, 0, eArgTypePlugin, "Name of the process plugin you want to use."},
{ LLDB_OPT_SET_1, false, "pid", 'p', required_argument, NULL, 0, eArgTypePid, "The process ID of an existing process to attach to."},
{ LLDB_OPT_SET_2, false, "name", 'n', required_argument, NULL, 0, eArgTypeProcessName, "The name of the process to attach to."},
{ LLDB_OPT_SET_2, false, "waitfor",'w', no_argument, NULL, 0, eArgTypeNone, "Wait for the the process with <process-name> to launch."},
{ 0, false, NULL, 0, 0, NULL, 0, eArgTypeNone, NULL }
};
//-------------------------------------------------------------------------
// CommandObjectProcessContinue
//-------------------------------------------------------------------------
#pragma mark CommandObjectProcessContinue
class CommandObjectProcessContinue : public CommandObject
{
public:
CommandObjectProcessContinue (CommandInterpreter &interpreter) :
CommandObject (interpreter,
"process continue",
"Continue execution of all threads in the current process.",
"process continue",
eFlagProcessMustBeLaunched | eFlagProcessMustBePaused)
{
}
~CommandObjectProcessContinue ()
{
}
bool
Execute (Args& command,
CommandReturnObject &result)
{
Process *process = m_interpreter.GetExecutionContext().GetProcessPtr();
bool synchronous_execution = m_interpreter.GetSynchronous ();
if (process == NULL)
{
result.AppendError ("no process to continue");
result.SetStatus (eReturnStatusFailed);
return false;
}
StateType state = process->GetState();
if (state == eStateStopped)
{
if (command.GetArgumentCount() != 0)
{
result.AppendErrorWithFormat ("The '%s' command does not take any arguments.\n", m_cmd_name.c_str());
result.SetStatus (eReturnStatusFailed);
return false;
}
const uint32_t num_threads = process->GetThreadList().GetSize();
// Set the actions that the threads should each take when resuming
for (uint32_t idx=0; idx<num_threads; ++idx)
{
process->GetThreadList().GetThreadAtIndex(idx)->SetResumeState (eStateRunning);
}
Error error(process->Resume());
if (error.Success())
{
result.AppendMessageWithFormat ("Process %llu resuming\n", process->GetID());
if (synchronous_execution)
{
state = process->WaitForProcessToStop (NULL);
result.SetDidChangeProcessState (true);
result.AppendMessageWithFormat ("Process %llu %s\n", process->GetID(), StateAsCString (state));
result.SetStatus (eReturnStatusSuccessFinishNoResult);
}
else
{
result.SetStatus (eReturnStatusSuccessContinuingNoResult);
}
}
else
{
result.AppendErrorWithFormat("Failed to resume process: %s.\n", error.AsCString());
result.SetStatus (eReturnStatusFailed);
}
}
else
{
result.AppendErrorWithFormat ("Process cannot be continued from its current state (%s).\n",
StateAsCString(state));
result.SetStatus (eReturnStatusFailed);
}
return result.Succeeded();
}
};
//-------------------------------------------------------------------------
// CommandObjectProcessDetach
//-------------------------------------------------------------------------
#pragma mark CommandObjectProcessDetach
class CommandObjectProcessDetach : public CommandObject
{
public:
CommandObjectProcessDetach (CommandInterpreter &interpreter) :
CommandObject (interpreter,
"process detach",
"Detach from the current process being debugged.",
"process detach",
eFlagProcessMustBeLaunched)
{
}
~CommandObjectProcessDetach ()
{
}
bool
Execute (Args& command,
CommandReturnObject &result)
{
Process *process = m_interpreter.GetExecutionContext().GetProcessPtr();
if (process == NULL)
{
result.AppendError ("must have a valid process in order to detach");
result.SetStatus (eReturnStatusFailed);
return false;
}
result.AppendMessageWithFormat ("Detaching from process %llu\n", process->GetID());
Error error (process->Detach());
if (error.Success())
{
result.SetStatus (eReturnStatusSuccessFinishResult);
}
else
{
result.AppendErrorWithFormat ("Detach failed: %s\n", error.AsCString());
result.SetStatus (eReturnStatusFailed);
return false;
}
return result.Succeeded();
}
};
Added support for attaching to a remote debug server with the new command: (lldb) process connect <remote-url> Currently when you specify a file with the file command it helps us to find a process plug-in that is suitable for debugging. If you specify a file you can rely upon this to find the correct debugger plug-in: % lldb a.out Current executable set to 'a.out' (x86_64). (lldb) process connect connect://localhost:2345 ... If you don't specify a file, you will need to specify the plug-in name that you wish to use: % lldb (lldb) process connect --plugin process.gdb-remote connect://localhost:2345 Other connection URL examples: (lldb) process connect connect://localhost:2345 (lldb) process connect tcp://127.0.0.1 (lldb) process connect file:///dev/ttyS1 We are currently treating the "connect://host:port" as a way to do raw socket connections. If there is a URL for this already, please let me know and we will adopt it. So now you can connect to a remote debug server with the ProcessGDBRemote plug-in. After connection, it will ask for the pid info using the "qC" packet and if it responds with a valid process ID, it will be equivalent to attaching. If it response with an error or invalid process ID, the LLDB process will be in a new state: eStateConnected. This allows us to then download a program or specify the program to run (using the 'A' packet), or specify a process to attach to (using the "vAttach" packets), or query info about the processes that might be available. llvm-svn: 124846
2011-02-04 09:58:07 +08:00
//-------------------------------------------------------------------------
// CommandObjectProcessConnect
//-------------------------------------------------------------------------
#pragma mark CommandObjectProcessConnect
class CommandObjectProcessConnect : public CommandObject
{
public:
class CommandOptions : public Options
{
public:
CommandOptions (CommandInterpreter &interpreter) :
Options(interpreter)
Added support for attaching to a remote debug server with the new command: (lldb) process connect <remote-url> Currently when you specify a file with the file command it helps us to find a process plug-in that is suitable for debugging. If you specify a file you can rely upon this to find the correct debugger plug-in: % lldb a.out Current executable set to 'a.out' (x86_64). (lldb) process connect connect://localhost:2345 ... If you don't specify a file, you will need to specify the plug-in name that you wish to use: % lldb (lldb) process connect --plugin process.gdb-remote connect://localhost:2345 Other connection URL examples: (lldb) process connect connect://localhost:2345 (lldb) process connect tcp://127.0.0.1 (lldb) process connect file:///dev/ttyS1 We are currently treating the "connect://host:port" as a way to do raw socket connections. If there is a URL for this already, please let me know and we will adopt it. So now you can connect to a remote debug server with the ProcessGDBRemote plug-in. After connection, it will ask for the pid info using the "qC" packet and if it responds with a valid process ID, it will be equivalent to attaching. If it response with an error or invalid process ID, the LLDB process will be in a new state: eStateConnected. This allows us to then download a program or specify the program to run (using the 'A' packet), or specify a process to attach to (using the "vAttach" packets), or query info about the processes that might be available. llvm-svn: 124846
2011-02-04 09:58:07 +08:00
{
Added two new classes for command options: lldb_private::OptionGroup lldb_private::OptionGroupOptions OptionGroup lets you define a class that encapsulates settings that you want to reuse in multiple commands. It contains only the option definitions and the ability to set the option values, but it doesn't directly interface with the lldb_private::Options class that is the front end to all of the CommandObject option parsing. For that the OptionGroupOptions class can be used. It aggregates one or more OptionGroup objects and directs the option setting to the appropriate OptionGroup class. For an example of this, take a look at the CommandObjectFile and how it uses its "m_option_group" object shown below to be able to set values in both the FileOptionGroup and PlatformOptionGroup classes. The members used in CommandObjectFile are: OptionGroupOptions m_option_group; FileOptionGroup m_file_options; PlatformOptionGroup m_platform_options; Then in the constructor for CommandObjectFile you can combine the option settings. The code below shows a simplified version of the constructor: CommandObjectFile::CommandObjectFile(CommandInterpreter &interpreter) : CommandObject (...), m_option_group (interpreter), m_file_options (), m_platform_options(true) { m_option_group.Append (&m_file_options); m_option_group.Append (&m_platform_options); m_option_group.Finalize(); } We append the m_file_options and then the m_platform_options and then tell the option group the finalize the results. This allows the m_option_group to become the organizer of our prefs and after option parsing we end up with valid preference settings in both the m_file_options and m_platform_options objects. This also allows any other commands to use the FileOptionGroup and PlatformOptionGroup classes to implement options for their commands. Renamed: virtual void Options::ResetOptionValues(); to: virtual void Options::OptionParsingStarting(); And implemented a new callback named: virtual Error Options::OptionParsingFinished(); This allows Options subclasses to verify that the options all go together after all of the options have been specified and gives the chance for the command object to return an error. It also gives a chance to take all of the option values and produce or initialize objects after all options have completed parsing. Modfied: virtual Error SetOptionValue (int option_idx, const char *option_arg) = 0; to be: virtual Error SetOptionValue (uint32_t option_idx, const char *option_arg) = 0; (option_idx is now unsigned). llvm-svn: 129415
2011-04-13 08:18:08 +08:00
// Keep default values of all options in one place: OptionParsingStarting ()
OptionParsingStarting ();
Added support for attaching to a remote debug server with the new command: (lldb) process connect <remote-url> Currently when you specify a file with the file command it helps us to find a process plug-in that is suitable for debugging. If you specify a file you can rely upon this to find the correct debugger plug-in: % lldb a.out Current executable set to 'a.out' (x86_64). (lldb) process connect connect://localhost:2345 ... If you don't specify a file, you will need to specify the plug-in name that you wish to use: % lldb (lldb) process connect --plugin process.gdb-remote connect://localhost:2345 Other connection URL examples: (lldb) process connect connect://localhost:2345 (lldb) process connect tcp://127.0.0.1 (lldb) process connect file:///dev/ttyS1 We are currently treating the "connect://host:port" as a way to do raw socket connections. If there is a URL for this already, please let me know and we will adopt it. So now you can connect to a remote debug server with the ProcessGDBRemote plug-in. After connection, it will ask for the pid info using the "qC" packet and if it responds with a valid process ID, it will be equivalent to attaching. If it response with an error or invalid process ID, the LLDB process will be in a new state: eStateConnected. This allows us to then download a program or specify the program to run (using the 'A' packet), or specify a process to attach to (using the "vAttach" packets), or query info about the processes that might be available. llvm-svn: 124846
2011-02-04 09:58:07 +08:00
}
~CommandOptions ()
{
}
Error
Added two new classes for command options: lldb_private::OptionGroup lldb_private::OptionGroupOptions OptionGroup lets you define a class that encapsulates settings that you want to reuse in multiple commands. It contains only the option definitions and the ability to set the option values, but it doesn't directly interface with the lldb_private::Options class that is the front end to all of the CommandObject option parsing. For that the OptionGroupOptions class can be used. It aggregates one or more OptionGroup objects and directs the option setting to the appropriate OptionGroup class. For an example of this, take a look at the CommandObjectFile and how it uses its "m_option_group" object shown below to be able to set values in both the FileOptionGroup and PlatformOptionGroup classes. The members used in CommandObjectFile are: OptionGroupOptions m_option_group; FileOptionGroup m_file_options; PlatformOptionGroup m_platform_options; Then in the constructor for CommandObjectFile you can combine the option settings. The code below shows a simplified version of the constructor: CommandObjectFile::CommandObjectFile(CommandInterpreter &interpreter) : CommandObject (...), m_option_group (interpreter), m_file_options (), m_platform_options(true) { m_option_group.Append (&m_file_options); m_option_group.Append (&m_platform_options); m_option_group.Finalize(); } We append the m_file_options and then the m_platform_options and then tell the option group the finalize the results. This allows the m_option_group to become the organizer of our prefs and after option parsing we end up with valid preference settings in both the m_file_options and m_platform_options objects. This also allows any other commands to use the FileOptionGroup and PlatformOptionGroup classes to implement options for their commands. Renamed: virtual void Options::ResetOptionValues(); to: virtual void Options::OptionParsingStarting(); And implemented a new callback named: virtual Error Options::OptionParsingFinished(); This allows Options subclasses to verify that the options all go together after all of the options have been specified and gives the chance for the command object to return an error. It also gives a chance to take all of the option values and produce or initialize objects after all options have completed parsing. Modfied: virtual Error SetOptionValue (int option_idx, const char *option_arg) = 0; to be: virtual Error SetOptionValue (uint32_t option_idx, const char *option_arg) = 0; (option_idx is now unsigned). llvm-svn: 129415
2011-04-13 08:18:08 +08:00
SetOptionValue (uint32_t option_idx, const char *option_arg)
Added support for attaching to a remote debug server with the new command: (lldb) process connect <remote-url> Currently when you specify a file with the file command it helps us to find a process plug-in that is suitable for debugging. If you specify a file you can rely upon this to find the correct debugger plug-in: % lldb a.out Current executable set to 'a.out' (x86_64). (lldb) process connect connect://localhost:2345 ... If you don't specify a file, you will need to specify the plug-in name that you wish to use: % lldb (lldb) process connect --plugin process.gdb-remote connect://localhost:2345 Other connection URL examples: (lldb) process connect connect://localhost:2345 (lldb) process connect tcp://127.0.0.1 (lldb) process connect file:///dev/ttyS1 We are currently treating the "connect://host:port" as a way to do raw socket connections. If there is a URL for this already, please let me know and we will adopt it. So now you can connect to a remote debug server with the ProcessGDBRemote plug-in. After connection, it will ask for the pid info using the "qC" packet and if it responds with a valid process ID, it will be equivalent to attaching. If it response with an error or invalid process ID, the LLDB process will be in a new state: eStateConnected. This allows us to then download a program or specify the program to run (using the 'A' packet), or specify a process to attach to (using the "vAttach" packets), or query info about the processes that might be available. llvm-svn: 124846
2011-02-04 09:58:07 +08:00
{
Error error;
char short_option = (char) m_getopt_table[option_idx].val;
switch (short_option)
{
case 'p':
plugin_name.assign (option_arg);
break;
default:
error.SetErrorStringWithFormat("invalid short option character '%c'", short_option);
Added support for attaching to a remote debug server with the new command: (lldb) process connect <remote-url> Currently when you specify a file with the file command it helps us to find a process plug-in that is suitable for debugging. If you specify a file you can rely upon this to find the correct debugger plug-in: % lldb a.out Current executable set to 'a.out' (x86_64). (lldb) process connect connect://localhost:2345 ... If you don't specify a file, you will need to specify the plug-in name that you wish to use: % lldb (lldb) process connect --plugin process.gdb-remote connect://localhost:2345 Other connection URL examples: (lldb) process connect connect://localhost:2345 (lldb) process connect tcp://127.0.0.1 (lldb) process connect file:///dev/ttyS1 We are currently treating the "connect://host:port" as a way to do raw socket connections. If there is a URL for this already, please let me know and we will adopt it. So now you can connect to a remote debug server with the ProcessGDBRemote plug-in. After connection, it will ask for the pid info using the "qC" packet and if it responds with a valid process ID, it will be equivalent to attaching. If it response with an error or invalid process ID, the LLDB process will be in a new state: eStateConnected. This allows us to then download a program or specify the program to run (using the 'A' packet), or specify a process to attach to (using the "vAttach" packets), or query info about the processes that might be available. llvm-svn: 124846
2011-02-04 09:58:07 +08:00
break;
}
return error;
}
void
Added two new classes for command options: lldb_private::OptionGroup lldb_private::OptionGroupOptions OptionGroup lets you define a class that encapsulates settings that you want to reuse in multiple commands. It contains only the option definitions and the ability to set the option values, but it doesn't directly interface with the lldb_private::Options class that is the front end to all of the CommandObject option parsing. For that the OptionGroupOptions class can be used. It aggregates one or more OptionGroup objects and directs the option setting to the appropriate OptionGroup class. For an example of this, take a look at the CommandObjectFile and how it uses its "m_option_group" object shown below to be able to set values in both the FileOptionGroup and PlatformOptionGroup classes. The members used in CommandObjectFile are: OptionGroupOptions m_option_group; FileOptionGroup m_file_options; PlatformOptionGroup m_platform_options; Then in the constructor for CommandObjectFile you can combine the option settings. The code below shows a simplified version of the constructor: CommandObjectFile::CommandObjectFile(CommandInterpreter &interpreter) : CommandObject (...), m_option_group (interpreter), m_file_options (), m_platform_options(true) { m_option_group.Append (&m_file_options); m_option_group.Append (&m_platform_options); m_option_group.Finalize(); } We append the m_file_options and then the m_platform_options and then tell the option group the finalize the results. This allows the m_option_group to become the organizer of our prefs and after option parsing we end up with valid preference settings in both the m_file_options and m_platform_options objects. This also allows any other commands to use the FileOptionGroup and PlatformOptionGroup classes to implement options for their commands. Renamed: virtual void Options::ResetOptionValues(); to: virtual void Options::OptionParsingStarting(); And implemented a new callback named: virtual Error Options::OptionParsingFinished(); This allows Options subclasses to verify that the options all go together after all of the options have been specified and gives the chance for the command object to return an error. It also gives a chance to take all of the option values and produce or initialize objects after all options have completed parsing. Modfied: virtual Error SetOptionValue (int option_idx, const char *option_arg) = 0; to be: virtual Error SetOptionValue (uint32_t option_idx, const char *option_arg) = 0; (option_idx is now unsigned). llvm-svn: 129415
2011-04-13 08:18:08 +08:00
OptionParsingStarting ()
Added support for attaching to a remote debug server with the new command: (lldb) process connect <remote-url> Currently when you specify a file with the file command it helps us to find a process plug-in that is suitable for debugging. If you specify a file you can rely upon this to find the correct debugger plug-in: % lldb a.out Current executable set to 'a.out' (x86_64). (lldb) process connect connect://localhost:2345 ... If you don't specify a file, you will need to specify the plug-in name that you wish to use: % lldb (lldb) process connect --plugin process.gdb-remote connect://localhost:2345 Other connection URL examples: (lldb) process connect connect://localhost:2345 (lldb) process connect tcp://127.0.0.1 (lldb) process connect file:///dev/ttyS1 We are currently treating the "connect://host:port" as a way to do raw socket connections. If there is a URL for this already, please let me know and we will adopt it. So now you can connect to a remote debug server with the ProcessGDBRemote plug-in. After connection, it will ask for the pid info using the "qC" packet and if it responds with a valid process ID, it will be equivalent to attaching. If it response with an error or invalid process ID, the LLDB process will be in a new state: eStateConnected. This allows us to then download a program or specify the program to run (using the 'A' packet), or specify a process to attach to (using the "vAttach" packets), or query info about the processes that might be available. llvm-svn: 124846
2011-02-04 09:58:07 +08:00
{
plugin_name.clear();
}
const OptionDefinition*
Added support for attaching to a remote debug server with the new command: (lldb) process connect <remote-url> Currently when you specify a file with the file command it helps us to find a process plug-in that is suitable for debugging. If you specify a file you can rely upon this to find the correct debugger plug-in: % lldb a.out Current executable set to 'a.out' (x86_64). (lldb) process connect connect://localhost:2345 ... If you don't specify a file, you will need to specify the plug-in name that you wish to use: % lldb (lldb) process connect --plugin process.gdb-remote connect://localhost:2345 Other connection URL examples: (lldb) process connect connect://localhost:2345 (lldb) process connect tcp://127.0.0.1 (lldb) process connect file:///dev/ttyS1 We are currently treating the "connect://host:port" as a way to do raw socket connections. If there is a URL for this already, please let me know and we will adopt it. So now you can connect to a remote debug server with the ProcessGDBRemote plug-in. After connection, it will ask for the pid info using the "qC" packet and if it responds with a valid process ID, it will be equivalent to attaching. If it response with an error or invalid process ID, the LLDB process will be in a new state: eStateConnected. This allows us to then download a program or specify the program to run (using the 'A' packet), or specify a process to attach to (using the "vAttach" packets), or query info about the processes that might be available. llvm-svn: 124846
2011-02-04 09:58:07 +08:00
GetDefinitions ()
{
return g_option_table;
}
// Options table: Required for subclasses of Options.
static OptionDefinition g_option_table[];
Added support for attaching to a remote debug server with the new command: (lldb) process connect <remote-url> Currently when you specify a file with the file command it helps us to find a process plug-in that is suitable for debugging. If you specify a file you can rely upon this to find the correct debugger plug-in: % lldb a.out Current executable set to 'a.out' (x86_64). (lldb) process connect connect://localhost:2345 ... If you don't specify a file, you will need to specify the plug-in name that you wish to use: % lldb (lldb) process connect --plugin process.gdb-remote connect://localhost:2345 Other connection URL examples: (lldb) process connect connect://localhost:2345 (lldb) process connect tcp://127.0.0.1 (lldb) process connect file:///dev/ttyS1 We are currently treating the "connect://host:port" as a way to do raw socket connections. If there is a URL for this already, please let me know and we will adopt it. So now you can connect to a remote debug server with the ProcessGDBRemote plug-in. After connection, it will ask for the pid info using the "qC" packet and if it responds with a valid process ID, it will be equivalent to attaching. If it response with an error or invalid process ID, the LLDB process will be in a new state: eStateConnected. This allows us to then download a program or specify the program to run (using the 'A' packet), or specify a process to attach to (using the "vAttach" packets), or query info about the processes that might be available. llvm-svn: 124846
2011-02-04 09:58:07 +08:00
// Instance variables to hold the values for command options.
std::string plugin_name;
};
CommandObjectProcessConnect (CommandInterpreter &interpreter) :
CommandObject (interpreter,
"process connect",
"Connect to a remote debug service.",
"process connect <remote-url>",
0),
m_options (interpreter)
Added support for attaching to a remote debug server with the new command: (lldb) process connect <remote-url> Currently when you specify a file with the file command it helps us to find a process plug-in that is suitable for debugging. If you specify a file you can rely upon this to find the correct debugger plug-in: % lldb a.out Current executable set to 'a.out' (x86_64). (lldb) process connect connect://localhost:2345 ... If you don't specify a file, you will need to specify the plug-in name that you wish to use: % lldb (lldb) process connect --plugin process.gdb-remote connect://localhost:2345 Other connection URL examples: (lldb) process connect connect://localhost:2345 (lldb) process connect tcp://127.0.0.1 (lldb) process connect file:///dev/ttyS1 We are currently treating the "connect://host:port" as a way to do raw socket connections. If there is a URL for this already, please let me know and we will adopt it. So now you can connect to a remote debug server with the ProcessGDBRemote plug-in. After connection, it will ask for the pid info using the "qC" packet and if it responds with a valid process ID, it will be equivalent to attaching. If it response with an error or invalid process ID, the LLDB process will be in a new state: eStateConnected. This allows us to then download a program or specify the program to run (using the 'A' packet), or specify a process to attach to (using the "vAttach" packets), or query info about the processes that might be available. llvm-svn: 124846
2011-02-04 09:58:07 +08:00
{
}
~CommandObjectProcessConnect ()
{
}
bool
Execute (Args& command,
CommandReturnObject &result)
{
TargetSP target_sp (m_interpreter.GetDebugger().GetSelectedTarget());
Error error;
Process *process = m_interpreter.GetExecutionContext().GetProcessPtr();
Added support for attaching to a remote debug server with the new command: (lldb) process connect <remote-url> Currently when you specify a file with the file command it helps us to find a process plug-in that is suitable for debugging. If you specify a file you can rely upon this to find the correct debugger plug-in: % lldb a.out Current executable set to 'a.out' (x86_64). (lldb) process connect connect://localhost:2345 ... If you don't specify a file, you will need to specify the plug-in name that you wish to use: % lldb (lldb) process connect --plugin process.gdb-remote connect://localhost:2345 Other connection URL examples: (lldb) process connect connect://localhost:2345 (lldb) process connect tcp://127.0.0.1 (lldb) process connect file:///dev/ttyS1 We are currently treating the "connect://host:port" as a way to do raw socket connections. If there is a URL for this already, please let me know and we will adopt it. So now you can connect to a remote debug server with the ProcessGDBRemote plug-in. After connection, it will ask for the pid info using the "qC" packet and if it responds with a valid process ID, it will be equivalent to attaching. If it response with an error or invalid process ID, the LLDB process will be in a new state: eStateConnected. This allows us to then download a program or specify the program to run (using the 'A' packet), or specify a process to attach to (using the "vAttach" packets), or query info about the processes that might be available. llvm-svn: 124846
2011-02-04 09:58:07 +08:00
if (process)
{
if (process->IsAlive())
{
result.AppendErrorWithFormat ("Process %llu is currently being debugged, kill the process before connecting.\n",
Added support for attaching to a remote debug server with the new command: (lldb) process connect <remote-url> Currently when you specify a file with the file command it helps us to find a process plug-in that is suitable for debugging. If you specify a file you can rely upon this to find the correct debugger plug-in: % lldb a.out Current executable set to 'a.out' (x86_64). (lldb) process connect connect://localhost:2345 ... If you don't specify a file, you will need to specify the plug-in name that you wish to use: % lldb (lldb) process connect --plugin process.gdb-remote connect://localhost:2345 Other connection URL examples: (lldb) process connect connect://localhost:2345 (lldb) process connect tcp://127.0.0.1 (lldb) process connect file:///dev/ttyS1 We are currently treating the "connect://host:port" as a way to do raw socket connections. If there is a URL for this already, please let me know and we will adopt it. So now you can connect to a remote debug server with the ProcessGDBRemote plug-in. After connection, it will ask for the pid info using the "qC" packet and if it responds with a valid process ID, it will be equivalent to attaching. If it response with an error or invalid process ID, the LLDB process will be in a new state: eStateConnected. This allows us to then download a program or specify the program to run (using the 'A' packet), or specify a process to attach to (using the "vAttach" packets), or query info about the processes that might be available. llvm-svn: 124846
2011-02-04 09:58:07 +08:00
process->GetID());
result.SetStatus (eReturnStatusFailed);
return false;
}
}
if (!target_sp)
{
// If there isn't a current target create one.
FileSpec emptyFileSpec;
error = m_interpreter.GetDebugger().GetTargetList().CreateTarget (m_interpreter.GetDebugger(),
emptyFileSpec,
NULL,
Added support for attaching to a remote debug server with the new command: (lldb) process connect <remote-url> Currently when you specify a file with the file command it helps us to find a process plug-in that is suitable for debugging. If you specify a file you can rely upon this to find the correct debugger plug-in: % lldb a.out Current executable set to 'a.out' (x86_64). (lldb) process connect connect://localhost:2345 ... If you don't specify a file, you will need to specify the plug-in name that you wish to use: % lldb (lldb) process connect --plugin process.gdb-remote connect://localhost:2345 Other connection URL examples: (lldb) process connect connect://localhost:2345 (lldb) process connect tcp://127.0.0.1 (lldb) process connect file:///dev/ttyS1 We are currently treating the "connect://host:port" as a way to do raw socket connections. If there is a URL for this already, please let me know and we will adopt it. So now you can connect to a remote debug server with the ProcessGDBRemote plug-in. After connection, it will ask for the pid info using the "qC" packet and if it responds with a valid process ID, it will be equivalent to attaching. If it response with an error or invalid process ID, the LLDB process will be in a new state: eStateConnected. This allows us to then download a program or specify the program to run (using the 'A' packet), or specify a process to attach to (using the "vAttach" packets), or query info about the processes that might be available. llvm-svn: 124846
2011-02-04 09:58:07 +08:00
false,
NULL, // No platform options
Added support for attaching to a remote debug server with the new command: (lldb) process connect <remote-url> Currently when you specify a file with the file command it helps us to find a process plug-in that is suitable for debugging. If you specify a file you can rely upon this to find the correct debugger plug-in: % lldb a.out Current executable set to 'a.out' (x86_64). (lldb) process connect connect://localhost:2345 ... If you don't specify a file, you will need to specify the plug-in name that you wish to use: % lldb (lldb) process connect --plugin process.gdb-remote connect://localhost:2345 Other connection URL examples: (lldb) process connect connect://localhost:2345 (lldb) process connect tcp://127.0.0.1 (lldb) process connect file:///dev/ttyS1 We are currently treating the "connect://host:port" as a way to do raw socket connections. If there is a URL for this already, please let me know and we will adopt it. So now you can connect to a remote debug server with the ProcessGDBRemote plug-in. After connection, it will ask for the pid info using the "qC" packet and if it responds with a valid process ID, it will be equivalent to attaching. If it response with an error or invalid process ID, the LLDB process will be in a new state: eStateConnected. This allows us to then download a program or specify the program to run (using the 'A' packet), or specify a process to attach to (using the "vAttach" packets), or query info about the processes that might be available. llvm-svn: 124846
2011-02-04 09:58:07 +08:00
target_sp);
if (!target_sp || error.Fail())
{
result.AppendError(error.AsCString("Error creating target"));
result.SetStatus (eReturnStatusFailed);
return false;
}
m_interpreter.GetDebugger().GetTargetList().SetSelectedTarget(target_sp.get());
}
if (command.GetArgumentCount() == 1)
{
const char *plugin_name = NULL;
if (!m_options.plugin_name.empty())
plugin_name = m_options.plugin_name.c_str();
const char *remote_url = command.GetArgumentAtIndex(0);
process = target_sp->CreateProcess (m_interpreter.GetDebugger().GetListener(), plugin_name).get();
if (process)
{
error = process->ConnectRemote (remote_url);
if (error.Fail())
{
result.AppendError(error.AsCString("Remote connect failed"));
result.SetStatus (eReturnStatusFailed);
return false;
}
}
else
{
result.AppendErrorWithFormat ("Unable to find process plug-in for remote URL '%s'.\nPlease specify a process plug-in name with the --plugin option, or specify an object file using the \"file\" command.\n",
m_cmd_name.c_str());
Added support for attaching to a remote debug server with the new command: (lldb) process connect <remote-url> Currently when you specify a file with the file command it helps us to find a process plug-in that is suitable for debugging. If you specify a file you can rely upon this to find the correct debugger plug-in: % lldb a.out Current executable set to 'a.out' (x86_64). (lldb) process connect connect://localhost:2345 ... If you don't specify a file, you will need to specify the plug-in name that you wish to use: % lldb (lldb) process connect --plugin process.gdb-remote connect://localhost:2345 Other connection URL examples: (lldb) process connect connect://localhost:2345 (lldb) process connect tcp://127.0.0.1 (lldb) process connect file:///dev/ttyS1 We are currently treating the "connect://host:port" as a way to do raw socket connections. If there is a URL for this already, please let me know and we will adopt it. So now you can connect to a remote debug server with the ProcessGDBRemote plug-in. After connection, it will ask for the pid info using the "qC" packet and if it responds with a valid process ID, it will be equivalent to attaching. If it response with an error or invalid process ID, the LLDB process will be in a new state: eStateConnected. This allows us to then download a program or specify the program to run (using the 'A' packet), or specify a process to attach to (using the "vAttach" packets), or query info about the processes that might be available. llvm-svn: 124846
2011-02-04 09:58:07 +08:00
result.SetStatus (eReturnStatusFailed);
}
}
else
{
result.AppendErrorWithFormat ("'%s' takes exactly one argument:\nUsage: %s\n",
Added support for attaching to a remote debug server with the new command: (lldb) process connect <remote-url> Currently when you specify a file with the file command it helps us to find a process plug-in that is suitable for debugging. If you specify a file you can rely upon this to find the correct debugger plug-in: % lldb a.out Current executable set to 'a.out' (x86_64). (lldb) process connect connect://localhost:2345 ... If you don't specify a file, you will need to specify the plug-in name that you wish to use: % lldb (lldb) process connect --plugin process.gdb-remote connect://localhost:2345 Other connection URL examples: (lldb) process connect connect://localhost:2345 (lldb) process connect tcp://127.0.0.1 (lldb) process connect file:///dev/ttyS1 We are currently treating the "connect://host:port" as a way to do raw socket connections. If there is a URL for this already, please let me know and we will adopt it. So now you can connect to a remote debug server with the ProcessGDBRemote plug-in. After connection, it will ask for the pid info using the "qC" packet and if it responds with a valid process ID, it will be equivalent to attaching. If it response with an error or invalid process ID, the LLDB process will be in a new state: eStateConnected. This allows us to then download a program or specify the program to run (using the 'A' packet), or specify a process to attach to (using the "vAttach" packets), or query info about the processes that might be available. llvm-svn: 124846
2011-02-04 09:58:07 +08:00
m_cmd_name.c_str(),
m_cmd_syntax.c_str());
result.SetStatus (eReturnStatusFailed);
}
return result.Succeeded();
}
Options *
GetOptions ()
{
return &m_options;
}
protected:
CommandOptions m_options;
};
OptionDefinition
Added support for attaching to a remote debug server with the new command: (lldb) process connect <remote-url> Currently when you specify a file with the file command it helps us to find a process plug-in that is suitable for debugging. If you specify a file you can rely upon this to find the correct debugger plug-in: % lldb a.out Current executable set to 'a.out' (x86_64). (lldb) process connect connect://localhost:2345 ... If you don't specify a file, you will need to specify the plug-in name that you wish to use: % lldb (lldb) process connect --plugin process.gdb-remote connect://localhost:2345 Other connection URL examples: (lldb) process connect connect://localhost:2345 (lldb) process connect tcp://127.0.0.1 (lldb) process connect file:///dev/ttyS1 We are currently treating the "connect://host:port" as a way to do raw socket connections. If there is a URL for this already, please let me know and we will adopt it. So now you can connect to a remote debug server with the ProcessGDBRemote plug-in. After connection, it will ask for the pid info using the "qC" packet and if it responds with a valid process ID, it will be equivalent to attaching. If it response with an error or invalid process ID, the LLDB process will be in a new state: eStateConnected. This allows us to then download a program or specify the program to run (using the 'A' packet), or specify a process to attach to (using the "vAttach" packets), or query info about the processes that might be available. llvm-svn: 124846
2011-02-04 09:58:07 +08:00
CommandObjectProcessConnect::CommandOptions::g_option_table[] =
{
{ LLDB_OPT_SET_ALL, false, "plugin", 'p', required_argument, NULL, 0, eArgTypePlugin, "Name of the process plugin you want to use."},
{ 0, false, NULL, 0 , 0, NULL, 0, eArgTypeNone, NULL }
};
//-------------------------------------------------------------------------
// CommandObjectProcessLoad
//-------------------------------------------------------------------------
#pragma mark CommandObjectProcessLoad
class CommandObjectProcessLoad : public CommandObject
{
public:
CommandObjectProcessLoad (CommandInterpreter &interpreter) :
CommandObject (interpreter,
"process load",
"Load a shared library into the current process.",
"process load <filename> [<filename> ...]",
eFlagProcessMustBeLaunched | eFlagProcessMustBePaused)
{
}
~CommandObjectProcessLoad ()
{
}
bool
Execute (Args& command,
CommandReturnObject &result)
{
Process *process = m_interpreter.GetExecutionContext().GetProcessPtr();
if (process == NULL)
{
result.AppendError ("must have a valid process in order to load a shared library");
result.SetStatus (eReturnStatusFailed);
return false;
}
const uint32_t argc = command.GetArgumentCount();
for (uint32_t i=0; i<argc; ++i)
{
Error error;
const char *image_path = command.GetArgumentAtIndex(i);
FileSpec image_spec (image_path, false);
process->GetTarget().GetPlatform()->ResolveRemotePath(image_spec, image_spec);
uint32_t image_token = process->LoadImage(image_spec, error);
if (image_token != LLDB_INVALID_IMAGE_TOKEN)
{
result.AppendMessageWithFormat ("Loading \"%s\"...ok\nImage %u loaded.\n", image_path, image_token);
result.SetStatus (eReturnStatusSuccessFinishResult);
}
else
{
result.AppendErrorWithFormat ("failed to load '%s': %s", image_path, error.AsCString());
result.SetStatus (eReturnStatusFailed);
}
}
return result.Succeeded();
}
};
//-------------------------------------------------------------------------
// CommandObjectProcessUnload
//-------------------------------------------------------------------------
#pragma mark CommandObjectProcessUnload
class CommandObjectProcessUnload : public CommandObject
{
public:
CommandObjectProcessUnload (CommandInterpreter &interpreter) :
CommandObject (interpreter,
"process unload",
"Unload a shared library from the current process using the index returned by a previous call to \"process load\".",
"process unload <index>",
eFlagProcessMustBeLaunched | eFlagProcessMustBePaused)
{
}
~CommandObjectProcessUnload ()
{
}
bool
Execute (Args& command,
CommandReturnObject &result)
{
Process *process = m_interpreter.GetExecutionContext().GetProcessPtr();
if (process == NULL)
{
result.AppendError ("must have a valid process in order to load a shared library");
result.SetStatus (eReturnStatusFailed);
return false;
}
const uint32_t argc = command.GetArgumentCount();
for (uint32_t i=0; i<argc; ++i)
{
const char *image_token_cstr = command.GetArgumentAtIndex(i);
uint32_t image_token = Args::StringToUInt32(image_token_cstr, LLDB_INVALID_IMAGE_TOKEN, 0);
if (image_token == LLDB_INVALID_IMAGE_TOKEN)
{
result.AppendErrorWithFormat ("invalid image index argument '%s'", image_token_cstr);
result.SetStatus (eReturnStatusFailed);
break;
}
else
{
Error error (process->UnloadImage(image_token));
if (error.Success())
{
result.AppendMessageWithFormat ("Unloading shared library with index %u...ok\n", image_token);
result.SetStatus (eReturnStatusSuccessFinishResult);
}
else
{
result.AppendErrorWithFormat ("failed to unload image: %s", error.AsCString());
result.SetStatus (eReturnStatusFailed);
break;
}
}
}
return result.Succeeded();
}
};
//-------------------------------------------------------------------------
// CommandObjectProcessSignal
//-------------------------------------------------------------------------
#pragma mark CommandObjectProcessSignal
class CommandObjectProcessSignal : public CommandObject
{
public:
CommandObjectProcessSignal (CommandInterpreter &interpreter) :
CommandObject (interpreter,
"process signal",
"Send a UNIX signal to the current process being debugged.",
NULL)
{
CommandArgumentEntry arg;
CommandArgumentData signal_arg;
// Define the first (and only) variant of this arg.
signal_arg.arg_type = eArgTypeUnixSignal;
signal_arg.arg_repetition = eArgRepeatPlain;
// There is only one variant this argument could be; put it into the argument entry.
arg.push_back (signal_arg);
// Push the data for the first argument into the m_arguments vector.
m_arguments.push_back (arg);
}
~CommandObjectProcessSignal ()
{
}
bool
Execute (Args& command,
CommandReturnObject &result)
{
Process *process = m_interpreter.GetExecutionContext().GetProcessPtr();
if (process == NULL)
{
result.AppendError ("no process to signal");
result.SetStatus (eReturnStatusFailed);
return false;
}
if (command.GetArgumentCount() == 1)
{
int signo = LLDB_INVALID_SIGNAL_NUMBER;
const char *signal_name = command.GetArgumentAtIndex(0);
if (::isxdigit (signal_name[0]))
signo = Args::StringToSInt32(signal_name, LLDB_INVALID_SIGNAL_NUMBER, 0);
else
signo = process->GetUnixSignals().GetSignalNumberFromName (signal_name);
if (signo == LLDB_INVALID_SIGNAL_NUMBER)
{
result.AppendErrorWithFormat ("Invalid signal argument '%s'.\n", command.GetArgumentAtIndex(0));
result.SetStatus (eReturnStatusFailed);
}
else
{
Error error (process->Signal (signo));
if (error.Success())
{
result.SetStatus (eReturnStatusSuccessFinishResult);
}
else
{
result.AppendErrorWithFormat ("Failed to send signal %i: %s\n", signo, error.AsCString());
result.SetStatus (eReturnStatusFailed);
}
}
}
else
{
result.AppendErrorWithFormat("'%s' takes exactly one signal number argument:\nUsage: %s\n", m_cmd_name.c_str(),
m_cmd_syntax.c_str());
result.SetStatus (eReturnStatusFailed);
}
return result.Succeeded();
}
};
//-------------------------------------------------------------------------
// CommandObjectProcessInterrupt
//-------------------------------------------------------------------------
#pragma mark CommandObjectProcessInterrupt
class CommandObjectProcessInterrupt : public CommandObject
{
public:
CommandObjectProcessInterrupt (CommandInterpreter &interpreter) :
CommandObject (interpreter,
"process interrupt",
"Interrupt the current process being debugged.",
"process interrupt",
eFlagProcessMustBeLaunched)
{
}
~CommandObjectProcessInterrupt ()
{
}
bool
Execute (Args& command,
CommandReturnObject &result)
{
Process *process = m_interpreter.GetExecutionContext().GetProcessPtr();
if (process == NULL)
{
result.AppendError ("no process to halt");
result.SetStatus (eReturnStatusFailed);
return false;
}
if (command.GetArgumentCount() == 0)
{
Error error(process->Halt ());
if (error.Success())
{
result.SetStatus (eReturnStatusSuccessFinishResult);
// Maybe we should add a "SuspendThreadPlans so we
// can halt, and keep in place all the current thread plans.
process->GetThreadList().DiscardThreadPlans();
}
else
{
result.AppendErrorWithFormat ("Failed to halt process: %s\n", error.AsCString());
result.SetStatus (eReturnStatusFailed);
}
}
else
{
result.AppendErrorWithFormat("'%s' takes no arguments:\nUsage: %s\n",
m_cmd_name.c_str(),
m_cmd_syntax.c_str());
result.SetStatus (eReturnStatusFailed);
}
return result.Succeeded();
}
};
//-------------------------------------------------------------------------
// CommandObjectProcessKill
//-------------------------------------------------------------------------
#pragma mark CommandObjectProcessKill
class CommandObjectProcessKill : public CommandObject
{
public:
CommandObjectProcessKill (CommandInterpreter &interpreter) :
CommandObject (interpreter,
"process kill",
"Terminate the current process being debugged.",
"process kill",
eFlagProcessMustBeLaunched)
{
}
~CommandObjectProcessKill ()
{
}
bool
Execute (Args& command,
CommandReturnObject &result)
{
Process *process = m_interpreter.GetExecutionContext().GetProcessPtr();
if (process == NULL)
{
result.AppendError ("no process to kill");
result.SetStatus (eReturnStatusFailed);
return false;
}
if (command.GetArgumentCount() == 0)
{
Error error (process->Destroy());
if (error.Success())
{
result.SetStatus (eReturnStatusSuccessFinishResult);
}
else
{
result.AppendErrorWithFormat ("Failed to kill process: %s\n", error.AsCString());
result.SetStatus (eReturnStatusFailed);
}
}
else
{
result.AppendErrorWithFormat("'%s' takes no arguments:\nUsage: %s\n",
m_cmd_name.c_str(),
m_cmd_syntax.c_str());
result.SetStatus (eReturnStatusFailed);
}
return result.Succeeded();
}
};
//-------------------------------------------------------------------------
// CommandObjectProcessStatus
//-------------------------------------------------------------------------
#pragma mark CommandObjectProcessStatus
class CommandObjectProcessStatus : public CommandObject
{
public:
CommandObjectProcessStatus (CommandInterpreter &interpreter) :
CommandObject (interpreter,
"process status",
"Show the current status and location of executing process.",
"process status",
0)
{
}
~CommandObjectProcessStatus()
{
}
bool
Execute
(
Args& command,
CommandReturnObject &result
)
{
Centralized a lot of the status information for processes, threads, and stack frame down in the lldb_private::Process, lldb_private::Thread, lldb_private::StackFrameList and the lldb_private::StackFrame classes. We had some command line commands that had duplicate versions of the process status output ("thread list" and "process status" for example). Removed the "file" command and placed it where it should have been: "target create". Made an alias for "file" to "target create" so we stay compatible with GDB commands. We can now have multple usable targets in lldb at the same time. This is nice for comparing two runs of a program or debugging more than one binary at the same time. The new command is "target select <target-idx>" and also to see a list of the current targets you can use the new "target list" command. The flow in a debug session can be: (lldb) target create /path/to/exe/a.out (lldb) breakpoint set --name main (lldb) run ... hit breakpoint (lldb) target create /bin/ls (lldb) run /tmp Process 36001 exited with status = 0 (0x00000000) (lldb) target list Current targets: target #0: /tmp/args/a.out ( arch=x86_64-apple-darwin, platform=localhost, pid=35999, state=stopped ) * target #1: /bin/ls ( arch=x86_64-apple-darwin, platform=localhost, pid=36001, state=exited ) (lldb) target select 0 Current targets: * target #0: /tmp/args/a.out ( arch=x86_64-apple-darwin, platform=localhost, pid=35999, state=stopped ) target #1: /bin/ls ( arch=x86_64-apple-darwin, platform=localhost, pid=36001, state=exited ) (lldb) bt * thread #1: tid = 0x2d03, 0x0000000100000b9a a.out`main + 42 at main.c:16, stop reason = breakpoint 1.1 frame #0: 0x0000000100000b9a a.out`main + 42 at main.c:16 frame #1: 0x0000000100000b64 a.out`start + 52 Above we created a target for "a.out" and ran and hit a breakpoint at "main". Then we created a new target for /bin/ls and ran it. Then we listed the targest and selected our original "a.out" program, so we showed two concurent debug sessions going on at the same time. llvm-svn: 129695
2011-04-18 16:33:37 +08:00
Stream &strm = result.GetOutputStream();
result.SetStatus (eReturnStatusSuccessFinishNoResult);
Moved the execution context that was in the Debugger into the CommandInterpreter where it was always being used. Make sure that Modules can track their object file offsets correctly to allow opening of sub object files (like the "__commpage" on darwin). Modified the Platforms to be able to launch processes. The first part of this move is the platform soon will become the entity that launches your program and when it does, it uses a new ProcessLaunchInfo class which encapsulates all process launching settings. This simplifies the internal APIs needed for launching. I want to slowly phase out process launching from the process classes, so for now we can still launch just as we used to, but eventually the platform is the object that should do the launching. Modified the Host::LaunchProcess in the MacOSX Host.mm to correctly be able to launch processes with all of the new eLaunchFlag settings. Modified any code that was manually launching processes to use the Host::LaunchProcess functions. Fixed an issue where lldb_private::Args had implicitly defined copy constructors that could do the wrong thing. This has now been fixed by adding an appropriate copy constructor and assignment operator. Make sure we don't add empty ModuleSP entries to a module list. Fixed the commpage module creation on MacOSX, but we still need to train the MacOSX dynamic loader to not get rid of it when it doesn't have an entry in the all image infos. Abstracted many more calls from in ProcessGDBRemote down into the GDBRemoteCommunicationClient subclass to make the classes cleaner and more efficient. Fixed the default iOS ARM register context to be correct and also added support for targets that don't support the qThreadStopInfo packet by selecting the current thread (only if needed) and then sending a stop reply packet. Debugserver can now start up with a --unix-socket (-u for short) and can then bind to port zero and send the port it bound to to a listening process on the other end. This allows the GDB remote platform to spawn new GDB server instances (debugserver) to allow platform debugging. llvm-svn: 129351
2011-04-12 13:54:46 +08:00
ExecutionContext exe_ctx(m_interpreter.GetExecutionContext());
Process *process = exe_ctx.GetProcessPtr();
if (process)
{
Centralized a lot of the status information for processes, threads, and stack frame down in the lldb_private::Process, lldb_private::Thread, lldb_private::StackFrameList and the lldb_private::StackFrame classes. We had some command line commands that had duplicate versions of the process status output ("thread list" and "process status" for example). Removed the "file" command and placed it where it should have been: "target create". Made an alias for "file" to "target create" so we stay compatible with GDB commands. We can now have multple usable targets in lldb at the same time. This is nice for comparing two runs of a program or debugging more than one binary at the same time. The new command is "target select <target-idx>" and also to see a list of the current targets you can use the new "target list" command. The flow in a debug session can be: (lldb) target create /path/to/exe/a.out (lldb) breakpoint set --name main (lldb) run ... hit breakpoint (lldb) target create /bin/ls (lldb) run /tmp Process 36001 exited with status = 0 (0x00000000) (lldb) target list Current targets: target #0: /tmp/args/a.out ( arch=x86_64-apple-darwin, platform=localhost, pid=35999, state=stopped ) * target #1: /bin/ls ( arch=x86_64-apple-darwin, platform=localhost, pid=36001, state=exited ) (lldb) target select 0 Current targets: * target #0: /tmp/args/a.out ( arch=x86_64-apple-darwin, platform=localhost, pid=35999, state=stopped ) target #1: /bin/ls ( arch=x86_64-apple-darwin, platform=localhost, pid=36001, state=exited ) (lldb) bt * thread #1: tid = 0x2d03, 0x0000000100000b9a a.out`main + 42 at main.c:16, stop reason = breakpoint 1.1 frame #0: 0x0000000100000b9a a.out`main + 42 at main.c:16 frame #1: 0x0000000100000b64 a.out`start + 52 Above we created a target for "a.out" and ran and hit a breakpoint at "main". Then we created a new target for /bin/ls and ran it. Then we listed the targest and selected our original "a.out" program, so we showed two concurent debug sessions going on at the same time. llvm-svn: 129695
2011-04-18 16:33:37 +08:00
const bool only_threads_with_stop_reason = true;
const uint32_t start_frame = 0;
const uint32_t num_frames = 1;
const uint32_t num_frames_with_source = 1;
process->GetStatus(strm);
process->GetThreadStatus (strm,
only_threads_with_stop_reason,
start_frame,
num_frames,
num_frames_with_source);
Centralized a lot of the status information for processes, threads, and stack frame down in the lldb_private::Process, lldb_private::Thread, lldb_private::StackFrameList and the lldb_private::StackFrame classes. We had some command line commands that had duplicate versions of the process status output ("thread list" and "process status" for example). Removed the "file" command and placed it where it should have been: "target create". Made an alias for "file" to "target create" so we stay compatible with GDB commands. We can now have multple usable targets in lldb at the same time. This is nice for comparing two runs of a program or debugging more than one binary at the same time. The new command is "target select <target-idx>" and also to see a list of the current targets you can use the new "target list" command. The flow in a debug session can be: (lldb) target create /path/to/exe/a.out (lldb) breakpoint set --name main (lldb) run ... hit breakpoint (lldb) target create /bin/ls (lldb) run /tmp Process 36001 exited with status = 0 (0x00000000) (lldb) target list Current targets: target #0: /tmp/args/a.out ( arch=x86_64-apple-darwin, platform=localhost, pid=35999, state=stopped ) * target #1: /bin/ls ( arch=x86_64-apple-darwin, platform=localhost, pid=36001, state=exited ) (lldb) target select 0 Current targets: * target #0: /tmp/args/a.out ( arch=x86_64-apple-darwin, platform=localhost, pid=35999, state=stopped ) target #1: /bin/ls ( arch=x86_64-apple-darwin, platform=localhost, pid=36001, state=exited ) (lldb) bt * thread #1: tid = 0x2d03, 0x0000000100000b9a a.out`main + 42 at main.c:16, stop reason = breakpoint 1.1 frame #0: 0x0000000100000b9a a.out`main + 42 at main.c:16 frame #1: 0x0000000100000b64 a.out`start + 52 Above we created a target for "a.out" and ran and hit a breakpoint at "main". Then we created a new target for /bin/ls and ran it. Then we listed the targest and selected our original "a.out" program, so we showed two concurent debug sessions going on at the same time. llvm-svn: 129695
2011-04-18 16:33:37 +08:00
}
else
{
Centralized a lot of the status information for processes, threads, and stack frame down in the lldb_private::Process, lldb_private::Thread, lldb_private::StackFrameList and the lldb_private::StackFrame classes. We had some command line commands that had duplicate versions of the process status output ("thread list" and "process status" for example). Removed the "file" command and placed it where it should have been: "target create". Made an alias for "file" to "target create" so we stay compatible with GDB commands. We can now have multple usable targets in lldb at the same time. This is nice for comparing two runs of a program or debugging more than one binary at the same time. The new command is "target select <target-idx>" and also to see a list of the current targets you can use the new "target list" command. The flow in a debug session can be: (lldb) target create /path/to/exe/a.out (lldb) breakpoint set --name main (lldb) run ... hit breakpoint (lldb) target create /bin/ls (lldb) run /tmp Process 36001 exited with status = 0 (0x00000000) (lldb) target list Current targets: target #0: /tmp/args/a.out ( arch=x86_64-apple-darwin, platform=localhost, pid=35999, state=stopped ) * target #1: /bin/ls ( arch=x86_64-apple-darwin, platform=localhost, pid=36001, state=exited ) (lldb) target select 0 Current targets: * target #0: /tmp/args/a.out ( arch=x86_64-apple-darwin, platform=localhost, pid=35999, state=stopped ) target #1: /bin/ls ( arch=x86_64-apple-darwin, platform=localhost, pid=36001, state=exited ) (lldb) bt * thread #1: tid = 0x2d03, 0x0000000100000b9a a.out`main + 42 at main.c:16, stop reason = breakpoint 1.1 frame #0: 0x0000000100000b9a a.out`main + 42 at main.c:16 frame #1: 0x0000000100000b64 a.out`start + 52 Above we created a target for "a.out" and ran and hit a breakpoint at "main". Then we created a new target for /bin/ls and ran it. Then we listed the targest and selected our original "a.out" program, so we showed two concurent debug sessions going on at the same time. llvm-svn: 129695
2011-04-18 16:33:37 +08:00
result.AppendError ("No process.");
result.SetStatus (eReturnStatusFailed);
}
return result.Succeeded();
}
};
//-------------------------------------------------------------------------
// CommandObjectProcessHandle
//-------------------------------------------------------------------------
#pragma mark CommandObjectProcessHandle
class CommandObjectProcessHandle : public CommandObject
{
public:
class CommandOptions : public Options
{
public:
CommandOptions (CommandInterpreter &interpreter) :
Options (interpreter)
{
Added two new classes for command options: lldb_private::OptionGroup lldb_private::OptionGroupOptions OptionGroup lets you define a class that encapsulates settings that you want to reuse in multiple commands. It contains only the option definitions and the ability to set the option values, but it doesn't directly interface with the lldb_private::Options class that is the front end to all of the CommandObject option parsing. For that the OptionGroupOptions class can be used. It aggregates one or more OptionGroup objects and directs the option setting to the appropriate OptionGroup class. For an example of this, take a look at the CommandObjectFile and how it uses its "m_option_group" object shown below to be able to set values in both the FileOptionGroup and PlatformOptionGroup classes. The members used in CommandObjectFile are: OptionGroupOptions m_option_group; FileOptionGroup m_file_options; PlatformOptionGroup m_platform_options; Then in the constructor for CommandObjectFile you can combine the option settings. The code below shows a simplified version of the constructor: CommandObjectFile::CommandObjectFile(CommandInterpreter &interpreter) : CommandObject (...), m_option_group (interpreter), m_file_options (), m_platform_options(true) { m_option_group.Append (&m_file_options); m_option_group.Append (&m_platform_options); m_option_group.Finalize(); } We append the m_file_options and then the m_platform_options and then tell the option group the finalize the results. This allows the m_option_group to become the organizer of our prefs and after option parsing we end up with valid preference settings in both the m_file_options and m_platform_options objects. This also allows any other commands to use the FileOptionGroup and PlatformOptionGroup classes to implement options for their commands. Renamed: virtual void Options::ResetOptionValues(); to: virtual void Options::OptionParsingStarting(); And implemented a new callback named: virtual Error Options::OptionParsingFinished(); This allows Options subclasses to verify that the options all go together after all of the options have been specified and gives the chance for the command object to return an error. It also gives a chance to take all of the option values and produce or initialize objects after all options have completed parsing. Modfied: virtual Error SetOptionValue (int option_idx, const char *option_arg) = 0; to be: virtual Error SetOptionValue (uint32_t option_idx, const char *option_arg) = 0; (option_idx is now unsigned). llvm-svn: 129415
2011-04-13 08:18:08 +08:00
OptionParsingStarting ();
}
~CommandOptions ()
{
}
Error
Added two new classes for command options: lldb_private::OptionGroup lldb_private::OptionGroupOptions OptionGroup lets you define a class that encapsulates settings that you want to reuse in multiple commands. It contains only the option definitions and the ability to set the option values, but it doesn't directly interface with the lldb_private::Options class that is the front end to all of the CommandObject option parsing. For that the OptionGroupOptions class can be used. It aggregates one or more OptionGroup objects and directs the option setting to the appropriate OptionGroup class. For an example of this, take a look at the CommandObjectFile and how it uses its "m_option_group" object shown below to be able to set values in both the FileOptionGroup and PlatformOptionGroup classes. The members used in CommandObjectFile are: OptionGroupOptions m_option_group; FileOptionGroup m_file_options; PlatformOptionGroup m_platform_options; Then in the constructor for CommandObjectFile you can combine the option settings. The code below shows a simplified version of the constructor: CommandObjectFile::CommandObjectFile(CommandInterpreter &interpreter) : CommandObject (...), m_option_group (interpreter), m_file_options (), m_platform_options(true) { m_option_group.Append (&m_file_options); m_option_group.Append (&m_platform_options); m_option_group.Finalize(); } We append the m_file_options and then the m_platform_options and then tell the option group the finalize the results. This allows the m_option_group to become the organizer of our prefs and after option parsing we end up with valid preference settings in both the m_file_options and m_platform_options objects. This also allows any other commands to use the FileOptionGroup and PlatformOptionGroup classes to implement options for their commands. Renamed: virtual void Options::ResetOptionValues(); to: virtual void Options::OptionParsingStarting(); And implemented a new callback named: virtual Error Options::OptionParsingFinished(); This allows Options subclasses to verify that the options all go together after all of the options have been specified and gives the chance for the command object to return an error. It also gives a chance to take all of the option values and produce or initialize objects after all options have completed parsing. Modfied: virtual Error SetOptionValue (int option_idx, const char *option_arg) = 0; to be: virtual Error SetOptionValue (uint32_t option_idx, const char *option_arg) = 0; (option_idx is now unsigned). llvm-svn: 129415
2011-04-13 08:18:08 +08:00
SetOptionValue (uint32_t option_idx, const char *option_arg)
{
Error error;
char short_option = (char) m_getopt_table[option_idx].val;
switch (short_option)
{
case 's':
stop = option_arg;
break;
case 'n':
notify = option_arg;
break;
case 'p':
pass = option_arg;
break;
default:
error.SetErrorStringWithFormat("invalid short option character '%c'", short_option);
break;
}
return error;
}
void
Added two new classes for command options: lldb_private::OptionGroup lldb_private::OptionGroupOptions OptionGroup lets you define a class that encapsulates settings that you want to reuse in multiple commands. It contains only the option definitions and the ability to set the option values, but it doesn't directly interface with the lldb_private::Options class that is the front end to all of the CommandObject option parsing. For that the OptionGroupOptions class can be used. It aggregates one or more OptionGroup objects and directs the option setting to the appropriate OptionGroup class. For an example of this, take a look at the CommandObjectFile and how it uses its "m_option_group" object shown below to be able to set values in both the FileOptionGroup and PlatformOptionGroup classes. The members used in CommandObjectFile are: OptionGroupOptions m_option_group; FileOptionGroup m_file_options; PlatformOptionGroup m_platform_options; Then in the constructor for CommandObjectFile you can combine the option settings. The code below shows a simplified version of the constructor: CommandObjectFile::CommandObjectFile(CommandInterpreter &interpreter) : CommandObject (...), m_option_group (interpreter), m_file_options (), m_platform_options(true) { m_option_group.Append (&m_file_options); m_option_group.Append (&m_platform_options); m_option_group.Finalize(); } We append the m_file_options and then the m_platform_options and then tell the option group the finalize the results. This allows the m_option_group to become the organizer of our prefs and after option parsing we end up with valid preference settings in both the m_file_options and m_platform_options objects. This also allows any other commands to use the FileOptionGroup and PlatformOptionGroup classes to implement options for their commands. Renamed: virtual void Options::ResetOptionValues(); to: virtual void Options::OptionParsingStarting(); And implemented a new callback named: virtual Error Options::OptionParsingFinished(); This allows Options subclasses to verify that the options all go together after all of the options have been specified and gives the chance for the command object to return an error. It also gives a chance to take all of the option values and produce or initialize objects after all options have completed parsing. Modfied: virtual Error SetOptionValue (int option_idx, const char *option_arg) = 0; to be: virtual Error SetOptionValue (uint32_t option_idx, const char *option_arg) = 0; (option_idx is now unsigned). llvm-svn: 129415
2011-04-13 08:18:08 +08:00
OptionParsingStarting ()
{
stop.clear();
notify.clear();
pass.clear();
}
const OptionDefinition*
GetDefinitions ()
{
return g_option_table;
}
// Options table: Required for subclasses of Options.
static OptionDefinition g_option_table[];
// Instance variables to hold the values for command options.
std::string stop;
std::string notify;
std::string pass;
};
CommandObjectProcessHandle (CommandInterpreter &interpreter) :
CommandObject (interpreter,
"process handle",
"Show or update what the process and debugger should do with various signals received from the OS.",
NULL),
m_options (interpreter)
{
SetHelpLong ("If no signals are specified, update them all. If no update option is specified, list the current values.\n");
CommandArgumentEntry arg;
CommandArgumentData signal_arg;
signal_arg.arg_type = eArgTypeUnixSignal;
signal_arg.arg_repetition = eArgRepeatStar;
arg.push_back (signal_arg);
m_arguments.push_back (arg);
}
~CommandObjectProcessHandle ()
{
}
Options *
GetOptions ()
{
return &m_options;
}
bool
VerifyCommandOptionValue (const std::string &option, int &real_value)
{
bool okay = true;
bool success = false;
bool tmp_value = Args::StringToBoolean (option.c_str(), false, &success);
if (success && tmp_value)
real_value = 1;
else if (success && !tmp_value)
real_value = 0;
else
{
// If the value isn't 'true' or 'false', it had better be 0 or 1.
real_value = Args::StringToUInt32 (option.c_str(), 3);
if (real_value != 0 && real_value != 1)
okay = false;
}
return okay;
}
void
PrintSignalHeader (Stream &str)
{
str.Printf ("NAME PASS STOP NOTIFY\n");
str.Printf ("========== ===== ===== ======\n");
}
void
PrintSignal (Stream &str, int32_t signo, const char *sig_name, UnixSignals &signals)
{
bool stop;
bool suppress;
bool notify;
str.Printf ("%-10s ", sig_name);
if (signals.GetSignalInfo (signo, suppress, stop, notify))
{
bool pass = !suppress;
str.Printf ("%s %s %s",
(pass ? "true " : "false"),
(stop ? "true " : "false"),
(notify ? "true " : "false"));
}
str.Printf ("\n");
}
void
PrintSignalInformation (Stream &str, Args &signal_args, int num_valid_signals, UnixSignals &signals)
{
PrintSignalHeader (str);
if (num_valid_signals > 0)
{
size_t num_args = signal_args.GetArgumentCount();
for (size_t i = 0; i < num_args; ++i)
{
int32_t signo = signals.GetSignalNumberFromName (signal_args.GetArgumentAtIndex (i));
if (signo != LLDB_INVALID_SIGNAL_NUMBER)
PrintSignal (str, signo, signal_args.GetArgumentAtIndex (i), signals);
}
}
else // Print info for ALL signals
{
int32_t signo = signals.GetFirstSignalNumber();
while (signo != LLDB_INVALID_SIGNAL_NUMBER)
{
PrintSignal (str, signo, signals.GetSignalAsCString (signo), signals);
signo = signals.GetNextSignalNumber (signo);
}
}
}
bool
Execute (Args &signal_args, CommandReturnObject &result)
{
TargetSP target_sp = m_interpreter.GetDebugger().GetSelectedTarget();
if (!target_sp)
{
result.AppendError ("No current target;"
" cannot handle signals until you have a valid target and process.\n");
result.SetStatus (eReturnStatusFailed);
return false;
}
ProcessSP process_sp = target_sp->GetProcessSP();
if (!process_sp)
{
result.AppendError ("No current process; cannot handle signals until you have a valid process.\n");
result.SetStatus (eReturnStatusFailed);
return false;
}
int stop_action = -1; // -1 means leave the current setting alone
int pass_action = -1; // -1 means leave the current setting alone
int notify_action = -1; // -1 means leave the current setting alone
if (! m_options.stop.empty()
&& ! VerifyCommandOptionValue (m_options.stop, stop_action))
{
result.AppendError ("Invalid argument for command option --stop; must be true or false.\n");
result.SetStatus (eReturnStatusFailed);
return false;
}
if (! m_options.notify.empty()
&& ! VerifyCommandOptionValue (m_options.notify, notify_action))
{
result.AppendError ("Invalid argument for command option --notify; must be true or false.\n");
result.SetStatus (eReturnStatusFailed);
return false;
}
if (! m_options.pass.empty()
&& ! VerifyCommandOptionValue (m_options.pass, pass_action))
{
result.AppendError ("Invalid argument for command option --pass; must be true or false.\n");
result.SetStatus (eReturnStatusFailed);
return false;
}
size_t num_args = signal_args.GetArgumentCount();
UnixSignals &signals = process_sp->GetUnixSignals();
int num_signals_set = 0;
if (num_args > 0)
{
for (size_t i = 0; i < num_args; ++i)
{
int32_t signo = signals.GetSignalNumberFromName (signal_args.GetArgumentAtIndex (i));
if (signo != LLDB_INVALID_SIGNAL_NUMBER)
{
// Casting the actions as bools here should be okay, because VerifyCommandOptionValue guarantees
// the value is either 0 or 1.
if (stop_action != -1)
signals.SetShouldStop (signo, (bool) stop_action);
if (pass_action != -1)
{
bool suppress = ! ((bool) pass_action);
signals.SetShouldSuppress (signo, suppress);
}
if (notify_action != -1)
signals.SetShouldNotify (signo, (bool) notify_action);
++num_signals_set;
}
else
{
result.AppendErrorWithFormat ("Invalid signal name '%s'\n", signal_args.GetArgumentAtIndex (i));
}
}
}
else
{
// No signal specified, if any command options were specified, update ALL signals.
if ((notify_action != -1) || (stop_action != -1) || (pass_action != -1))
{
if (m_interpreter.Confirm ("Do you really want to update all the signals?", false))
{
int32_t signo = signals.GetFirstSignalNumber();
while (signo != LLDB_INVALID_SIGNAL_NUMBER)
{
if (notify_action != -1)
signals.SetShouldNotify (signo, (bool) notify_action);
if (stop_action != -1)
signals.SetShouldStop (signo, (bool) stop_action);
if (pass_action != -1)
{
bool suppress = ! ((bool) pass_action);
signals.SetShouldSuppress (signo, suppress);
}
signo = signals.GetNextSignalNumber (signo);
}
}
}
}
PrintSignalInformation (result.GetOutputStream(), signal_args, num_signals_set, signals);
if (num_signals_set > 0)
result.SetStatus (eReturnStatusSuccessFinishNoResult);
else
result.SetStatus (eReturnStatusFailed);
return result.Succeeded();
}
protected:
CommandOptions m_options;
};
OptionDefinition
CommandObjectProcessHandle::CommandOptions::g_option_table[] =
{
{ LLDB_OPT_SET_1, false, "stop", 's', required_argument, NULL, 0, eArgTypeBoolean, "Whether or not the process should be stopped if the signal is received." },
{ LLDB_OPT_SET_1, false, "notify", 'n', required_argument, NULL, 0, eArgTypeBoolean, "Whether or not the debugger should notify the user if the signal is received." },
{ LLDB_OPT_SET_1, false, "pass", 'p', required_argument, NULL, 0, eArgTypeBoolean, "Whether or not the signal should be passed to the process." },
{ 0, false, NULL, 0, 0, NULL, 0, eArgTypeNone, NULL }
};
//-------------------------------------------------------------------------
// CommandObjectMultiwordProcess
//-------------------------------------------------------------------------
CommandObjectMultiwordProcess::CommandObjectMultiwordProcess (CommandInterpreter &interpreter) :
CommandObjectMultiword (interpreter,
"process",
"A set of commands for operating on a process.",
"process <subcommand> [<subcommand-options>]")
{
LoadSubCommand ("attach", CommandObjectSP (new CommandObjectProcessAttach (interpreter)));
LoadSubCommand ("launch", CommandObjectSP (new CommandObjectProcessLaunch (interpreter)));
LoadSubCommand ("continue", CommandObjectSP (new CommandObjectProcessContinue (interpreter)));
LoadSubCommand ("connect", CommandObjectSP (new CommandObjectProcessConnect (interpreter)));
LoadSubCommand ("detach", CommandObjectSP (new CommandObjectProcessDetach (interpreter)));
LoadSubCommand ("load", CommandObjectSP (new CommandObjectProcessLoad (interpreter)));
LoadSubCommand ("unload", CommandObjectSP (new CommandObjectProcessUnload (interpreter)));
LoadSubCommand ("signal", CommandObjectSP (new CommandObjectProcessSignal (interpreter)));
LoadSubCommand ("handle", CommandObjectSP (new CommandObjectProcessHandle (interpreter)));
LoadSubCommand ("status", CommandObjectSP (new CommandObjectProcessStatus (interpreter)));
LoadSubCommand ("interrupt", CommandObjectSP (new CommandObjectProcessInterrupt (interpreter)));
LoadSubCommand ("kill", CommandObjectSP (new CommandObjectProcessKill (interpreter)));
}
CommandObjectMultiwordProcess::~CommandObjectMultiwordProcess ()
{
}