llvm-project/lldb/source/Host/macosx/Host.mm

1518 lines
48 KiB
Plaintext
Raw Normal View History

//===-- Host.mm -------------------------------------------------*- C++ -*-===//
//
// The LLVM Compiler Infrastructure
//
// This file is distributed under the University of Illinois Open Source
// License. See LICENSE.TXT for details.
//
//===----------------------------------------------------------------------===//
#include "lldb/Host/Host.h"
Added a new Host call to find LLDB related paths: static bool Host::GetLLDBPath (lldb::PathType path_type, FileSpec &file_spec); This will fill in "file_spec" with an appropriate path that is appropriate for the current Host OS. MacOSX will return paths within the LLDB.framework, and other unixes will return the paths they want. The current PathType enums are: typedef enum PathType { ePathTypeLLDBShlibDir, // The directory where the lldb.so (unix) or LLDB mach-o file in LLDB.framework (MacOSX) exists ePathTypeSupportExecutableDir, // Find LLDB support executable directory (debugserver, etc) ePathTypeHeaderDir, // Find LLDB header file directory ePathTypePythonDir // Find Python modules (PYTHONPATH) directory } PathType; All places that were finding executables are and python paths are now updated to use this Host call. Added another new host call to launch the inferior in a terminal. This ability will be very host specific and doesn't need to be supported on all systems. MacOSX currently will create a new .command file and tell Terminal.app to open the .command file. It also uses the new "darwin-debug" app which is a small app that uses posix to exec (no fork) and stop at the entry point of the program. The GDB remote plug-in is almost able launch a process and attach to it, it currently will spawn the process, but it won't attach to it just yet. This will let LLDB not have to share the terminal with another process and a new terminal window will pop up when you launch. This won't get hooked up until we work out all of the kinks. The new Host function is: static lldb::pid_t Host::LaunchInNewTerminal ( const char **argv, // argv[0] is executable const char **envp, const ArchSpec *arch_spec, bool stop_at_entry, bool disable_aslr); Cleaned up FileSpec::GetPath to not use strncpy() as it was always zero filling the entire path buffer. Fixed an issue with the dynamic checker function where I missed a '$' prefix that should have been added. llvm-svn: 116690
2010-10-18 06:03:32 +08:00
#include <crt_externs.h>
Fixed a race condition that could cause ProcessGDBRemote::DoResume() to return an error saying the resume timed out. Previously the thread that was trying to resume the process would eventually call ProcessGDBRemote::DoResume() which would broadcast an event over to the async GDB remote thread which would sent the continue packet to the remote gdb server. Right after this was sent, it would set a predicate boolean value (protected by a mutex and condition) and then the thread that issued the ProcessGDBRemote::DoResume() would then wait for that condition variable to be set. If the async gdb thread was too quick though, the predicate boolean value could have been set to true and back to false by the time the thread that issued the ProcessGDBRemote::DoResume() checks the boolean value. So we can't use the predicate value as a handshake. I have changed the code over to using a Event by having the GDB remote communication object post an event: GDBRemoteCommunication::eBroadcastBitRunPacketSent This allows reliable handshaking between the two threads and avoids the erroneous ProcessGDBRemote::DoResume() errors. Added a host backtrace service to allow in process backtraces when trying to track down tricky issues. I need to see if LLVM has any backtracing abilities abstracted in it already, and if so, use that, but I needed something ASAP for the current issue I was working on. The static function is: void Host::Backtrace (Stream &strm, uint32_t max_frames); And it will backtrace at most "max_frames" frames for the current thread and can be used with any of the Stream subclasses for logging. llvm-svn: 120793
2010-12-03 14:02:24 +08:00
#include <execinfo.h>
#include <grp.h>
#include <libproc.h>
#include <pwd.h>
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
#include <spawn.h>
Fixed a race condition that could cause ProcessGDBRemote::DoResume() to return an error saying the resume timed out. Previously the thread that was trying to resume the process would eventually call ProcessGDBRemote::DoResume() which would broadcast an event over to the async GDB remote thread which would sent the continue packet to the remote gdb server. Right after this was sent, it would set a predicate boolean value (protected by a mutex and condition) and then the thread that issued the ProcessGDBRemote::DoResume() would then wait for that condition variable to be set. If the async gdb thread was too quick though, the predicate boolean value could have been set to true and back to false by the time the thread that issued the ProcessGDBRemote::DoResume() checks the boolean value. So we can't use the predicate value as a handshake. I have changed the code over to using a Event by having the GDB remote communication object post an event: GDBRemoteCommunication::eBroadcastBitRunPacketSent This allows reliable handshaking between the two threads and avoids the erroneous ProcessGDBRemote::DoResume() errors. Added a host backtrace service to allow in process backtraces when trying to track down tricky issues. I need to see if LLVM has any backtracing abilities abstracted in it already, and if so, use that, but I needed something ASAP for the current issue I was working on. The static function is: void Host::Backtrace (Stream &strm, uint32_t max_frames); And it will backtrace at most "max_frames" frames for the current thread and can be used with any of the Stream subclasses for logging. llvm-svn: 120793
2010-12-03 14:02:24 +08:00
#include <stdio.h>
#include <sys/proc.h>
Added a new Host call to find LLDB related paths: static bool Host::GetLLDBPath (lldb::PathType path_type, FileSpec &file_spec); This will fill in "file_spec" with an appropriate path that is appropriate for the current Host OS. MacOSX will return paths within the LLDB.framework, and other unixes will return the paths they want. The current PathType enums are: typedef enum PathType { ePathTypeLLDBShlibDir, // The directory where the lldb.so (unix) or LLDB mach-o file in LLDB.framework (MacOSX) exists ePathTypeSupportExecutableDir, // Find LLDB support executable directory (debugserver, etc) ePathTypeHeaderDir, // Find LLDB header file directory ePathTypePythonDir // Find Python modules (PYTHONPATH) directory } PathType; All places that were finding executables are and python paths are now updated to use this Host call. Added another new host call to launch the inferior in a terminal. This ability will be very host specific and doesn't need to be supported on all systems. MacOSX currently will create a new .command file and tell Terminal.app to open the .command file. It also uses the new "darwin-debug" app which is a small app that uses posix to exec (no fork) and stop at the entry point of the program. The GDB remote plug-in is almost able launch a process and attach to it, it currently will spawn the process, but it won't attach to it just yet. This will let LLDB not have to share the terminal with another process and a new terminal window will pop up when you launch. This won't get hooked up until we work out all of the kinks. The new Host function is: static lldb::pid_t Host::LaunchInNewTerminal ( const char **argv, // argv[0] is executable const char **envp, const ArchSpec *arch_spec, bool stop_at_entry, bool disable_aslr); Cleaned up FileSpec::GetPath to not use strncpy() as it was always zero filling the entire path buffer. Fixed an issue with the dynamic checker function where I missed a '$' prefix that should have been added. llvm-svn: 116690
2010-10-18 06:03:32 +08:00
#include <sys/stat.h>
#include <sys/sysctl.h>
#include <sys/types.h>
Fixed a race condition that could cause ProcessGDBRemote::DoResume() to return an error saying the resume timed out. Previously the thread that was trying to resume the process would eventually call ProcessGDBRemote::DoResume() which would broadcast an event over to the async GDB remote thread which would sent the continue packet to the remote gdb server. Right after this was sent, it would set a predicate boolean value (protected by a mutex and condition) and then the thread that issued the ProcessGDBRemote::DoResume() would then wait for that condition variable to be set. If the async gdb thread was too quick though, the predicate boolean value could have been set to true and back to false by the time the thread that issued the ProcessGDBRemote::DoResume() checks the boolean value. So we can't use the predicate value as a handshake. I have changed the code over to using a Event by having the GDB remote communication object post an event: GDBRemoteCommunication::eBroadcastBitRunPacketSent This allows reliable handshaking between the two threads and avoids the erroneous ProcessGDBRemote::DoResume() errors. Added a host backtrace service to allow in process backtraces when trying to track down tricky issues. I need to see if LLVM has any backtracing abilities abstracted in it already, and if so, use that, but I needed something ASAP for the current issue I was working on. The static function is: void Host::Backtrace (Stream &strm, uint32_t max_frames); And it will backtrace at most "max_frames" frames for the current thread and can be used with any of the Stream subclasses for logging. llvm-svn: 120793
2010-12-03 14:02:24 +08:00
#include <unistd.h>
Added a new Host call to find LLDB related paths: static bool Host::GetLLDBPath (lldb::PathType path_type, FileSpec &file_spec); This will fill in "file_spec" with an appropriate path that is appropriate for the current Host OS. MacOSX will return paths within the LLDB.framework, and other unixes will return the paths they want. The current PathType enums are: typedef enum PathType { ePathTypeLLDBShlibDir, // The directory where the lldb.so (unix) or LLDB mach-o file in LLDB.framework (MacOSX) exists ePathTypeSupportExecutableDir, // Find LLDB support executable directory (debugserver, etc) ePathTypeHeaderDir, // Find LLDB header file directory ePathTypePythonDir // Find Python modules (PYTHONPATH) directory } PathType; All places that were finding executables are and python paths are now updated to use this Host call. Added another new host call to launch the inferior in a terminal. This ability will be very host specific and doesn't need to be supported on all systems. MacOSX currently will create a new .command file and tell Terminal.app to open the .command file. It also uses the new "darwin-debug" app which is a small app that uses posix to exec (no fork) and stop at the entry point of the program. The GDB remote plug-in is almost able launch a process and attach to it, it currently will spawn the process, but it won't attach to it just yet. This will let LLDB not have to share the terminal with another process and a new terminal window will pop up when you launch. This won't get hooked up until we work out all of the kinks. The new Host function is: static lldb::pid_t Host::LaunchInNewTerminal ( const char **argv, // argv[0] is executable const char **envp, const ArchSpec *arch_spec, bool stop_at_entry, bool disable_aslr); Cleaned up FileSpec::GetPath to not use strncpy() as it was always zero filling the entire path buffer. Fixed an issue with the dynamic checker function where I missed a '$' prefix that should have been added. llvm-svn: 116690
2010-10-18 06:03:32 +08:00
#include "lldb/Core/ArchSpec.h"
#include "lldb/Core/Communication.h"
#include "lldb/Core/ConnectionFileDescriptor.h"
#include "lldb/Core/DataExtractor.h"
#include "lldb/Core/Log.h"
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
#include "lldb/Core/Module.h"
Added a new Host call to find LLDB related paths: static bool Host::GetLLDBPath (lldb::PathType path_type, FileSpec &file_spec); This will fill in "file_spec" with an appropriate path that is appropriate for the current Host OS. MacOSX will return paths within the LLDB.framework, and other unixes will return the paths they want. The current PathType enums are: typedef enum PathType { ePathTypeLLDBShlibDir, // The directory where the lldb.so (unix) or LLDB mach-o file in LLDB.framework (MacOSX) exists ePathTypeSupportExecutableDir, // Find LLDB support executable directory (debugserver, etc) ePathTypeHeaderDir, // Find LLDB header file directory ePathTypePythonDir // Find Python modules (PYTHONPATH) directory } PathType; All places that were finding executables are and python paths are now updated to use this Host call. Added another new host call to launch the inferior in a terminal. This ability will be very host specific and doesn't need to be supported on all systems. MacOSX currently will create a new .command file and tell Terminal.app to open the .command file. It also uses the new "darwin-debug" app which is a small app that uses posix to exec (no fork) and stop at the entry point of the program. The GDB remote plug-in is almost able launch a process and attach to it, it currently will spawn the process, but it won't attach to it just yet. This will let LLDB not have to share the terminal with another process and a new terminal window will pop up when you launch. This won't get hooked up until we work out all of the kinks. The new Host function is: static lldb::pid_t Host::LaunchInNewTerminal ( const char **argv, // argv[0] is executable const char **envp, const ArchSpec *arch_spec, bool stop_at_entry, bool disable_aslr); Cleaned up FileSpec::GetPath to not use strncpy() as it was always zero filling the entire path buffer. Fixed an issue with the dynamic checker function where I missed a '$' prefix that should have been added. llvm-svn: 116690
2010-10-18 06:03:32 +08:00
#include "lldb/Core/StreamFile.h"
#include "lldb/Core/StreamString.h"
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
#include "lldb/Host/Endian.h"
#include "lldb/Host/FileSpec.h"
#include "lldb/Target/Platform.h"
#include "lldb/Target/Process.h"
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
#include "lldb/Utility/CleanUp.h"
#include "cfcpp/CFCBundle.h"
Added a new Host call to find LLDB related paths: static bool Host::GetLLDBPath (lldb::PathType path_type, FileSpec &file_spec); This will fill in "file_spec" with an appropriate path that is appropriate for the current Host OS. MacOSX will return paths within the LLDB.framework, and other unixes will return the paths they want. The current PathType enums are: typedef enum PathType { ePathTypeLLDBShlibDir, // The directory where the lldb.so (unix) or LLDB mach-o file in LLDB.framework (MacOSX) exists ePathTypeSupportExecutableDir, // Find LLDB support executable directory (debugserver, etc) ePathTypeHeaderDir, // Find LLDB header file directory ePathTypePythonDir // Find Python modules (PYTHONPATH) directory } PathType; All places that were finding executables are and python paths are now updated to use this Host call. Added another new host call to launch the inferior in a terminal. This ability will be very host specific and doesn't need to be supported on all systems. MacOSX currently will create a new .command file and tell Terminal.app to open the .command file. It also uses the new "darwin-debug" app which is a small app that uses posix to exec (no fork) and stop at the entry point of the program. The GDB remote plug-in is almost able launch a process and attach to it, it currently will spawn the process, but it won't attach to it just yet. This will let LLDB not have to share the terminal with another process and a new terminal window will pop up when you launch. This won't get hooked up until we work out all of the kinks. The new Host function is: static lldb::pid_t Host::LaunchInNewTerminal ( const char **argv, // argv[0] is executable const char **envp, const ArchSpec *arch_spec, bool stop_at_entry, bool disable_aslr); Cleaned up FileSpec::GetPath to not use strncpy() as it was always zero filling the entire path buffer. Fixed an issue with the dynamic checker function where I missed a '$' prefix that should have been added. llvm-svn: 116690
2010-10-18 06:03:32 +08:00
#include "cfcpp/CFCMutableArray.h"
#include "cfcpp/CFCMutableDictionary.h"
#include "cfcpp/CFCReleaser.h"
#include "cfcpp/CFCString.h"
#include "llvm/Support/Host.h"
#include "llvm/Support/MachO.h"
#include <objc/objc-auto.h>
#if defined(__arm__)
#include <UIKit/UIKit.h>
#else
#include <ApplicationServices/ApplicationServices.h>
#include <Carbon/Carbon.h>
#endif
#include <Foundation/Foundation.h>
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
#ifndef _POSIX_SPAWN_DISABLE_ASLR
#define _POSIX_SPAWN_DISABLE_ASLR 0x0100
#endif
extern "C"
{
int __pthread_chdir(const char *path);
int __pthread_fchdir (int fildes);
}
using namespace lldb;
using namespace lldb_private;
static pthread_once_t g_thread_create_once = PTHREAD_ONCE_INIT;
static pthread_key_t g_thread_create_key = 0;
class MacOSXDarwinThread
{
public:
MacOSXDarwinThread(const char *thread_name) :
m_pool (nil)
{
// Register our thread with the collector if garbage collection is enabled.
if (objc_collectingEnabled())
{
#if MAC_OS_X_VERSION_MAX_ALLOWED <= MAC_OS_X_VERSION_10_5
// On Leopard and earlier there is no way objc_registerThreadWithCollector
// function, so we do it manually.
auto_zone_register_thread(auto_zone());
#else
// On SnowLeopard and later we just call the thread registration function.
objc_registerThreadWithCollector();
#endif
}
else
{
m_pool = [[NSAutoreleasePool alloc] init];
}
Host::SetThreadName (LLDB_INVALID_PROCESS_ID, LLDB_INVALID_THREAD_ID, thread_name);
}
~MacOSXDarwinThread()
{
if (m_pool)
{
[m_pool release];
m_pool = nil;
}
}
static void PThreadDestructor (void *v)
{
if (v)
delete static_cast<MacOSXDarwinThread*>(v);
::pthread_setspecific (g_thread_create_key, NULL);
}
protected:
NSAutoreleasePool * m_pool;
private:
DISALLOW_COPY_AND_ASSIGN (MacOSXDarwinThread);
};
static void
InitThreadCreated()
{
::pthread_key_create (&g_thread_create_key, MacOSXDarwinThread::PThreadDestructor);
}
void
Host::ThreadCreated (const char *thread_name)
{
::pthread_once (&g_thread_create_once, InitThreadCreated);
if (g_thread_create_key)
{
::pthread_setspecific (g_thread_create_key, new MacOSXDarwinThread(thread_name));
}
}
bool
Added a new Host call to find LLDB related paths: static bool Host::GetLLDBPath (lldb::PathType path_type, FileSpec &file_spec); This will fill in "file_spec" with an appropriate path that is appropriate for the current Host OS. MacOSX will return paths within the LLDB.framework, and other unixes will return the paths they want. The current PathType enums are: typedef enum PathType { ePathTypeLLDBShlibDir, // The directory where the lldb.so (unix) or LLDB mach-o file in LLDB.framework (MacOSX) exists ePathTypeSupportExecutableDir, // Find LLDB support executable directory (debugserver, etc) ePathTypeHeaderDir, // Find LLDB header file directory ePathTypePythonDir // Find Python modules (PYTHONPATH) directory } PathType; All places that were finding executables are and python paths are now updated to use this Host call. Added another new host call to launch the inferior in a terminal. This ability will be very host specific and doesn't need to be supported on all systems. MacOSX currently will create a new .command file and tell Terminal.app to open the .command file. It also uses the new "darwin-debug" app which is a small app that uses posix to exec (no fork) and stop at the entry point of the program. The GDB remote plug-in is almost able launch a process and attach to it, it currently will spawn the process, but it won't attach to it just yet. This will let LLDB not have to share the terminal with another process and a new terminal window will pop up when you launch. This won't get hooked up until we work out all of the kinks. The new Host function is: static lldb::pid_t Host::LaunchInNewTerminal ( const char **argv, // argv[0] is executable const char **envp, const ArchSpec *arch_spec, bool stop_at_entry, bool disable_aslr); Cleaned up FileSpec::GetPath to not use strncpy() as it was always zero filling the entire path buffer. Fixed an issue with the dynamic checker function where I missed a '$' prefix that should have been added. llvm-svn: 116690
2010-10-18 06:03:32 +08:00
Host::ResolveExecutableInBundle (FileSpec &file)
{
#if defined (__APPLE__)
Added a new Host call to find LLDB related paths: static bool Host::GetLLDBPath (lldb::PathType path_type, FileSpec &file_spec); This will fill in "file_spec" with an appropriate path that is appropriate for the current Host OS. MacOSX will return paths within the LLDB.framework, and other unixes will return the paths they want. The current PathType enums are: typedef enum PathType { ePathTypeLLDBShlibDir, // The directory where the lldb.so (unix) or LLDB mach-o file in LLDB.framework (MacOSX) exists ePathTypeSupportExecutableDir, // Find LLDB support executable directory (debugserver, etc) ePathTypeHeaderDir, // Find LLDB header file directory ePathTypePythonDir // Find Python modules (PYTHONPATH) directory } PathType; All places that were finding executables are and python paths are now updated to use this Host call. Added another new host call to launch the inferior in a terminal. This ability will be very host specific and doesn't need to be supported on all systems. MacOSX currently will create a new .command file and tell Terminal.app to open the .command file. It also uses the new "darwin-debug" app which is a small app that uses posix to exec (no fork) and stop at the entry point of the program. The GDB remote plug-in is almost able launch a process and attach to it, it currently will spawn the process, but it won't attach to it just yet. This will let LLDB not have to share the terminal with another process and a new terminal window will pop up when you launch. This won't get hooked up until we work out all of the kinks. The new Host function is: static lldb::pid_t Host::LaunchInNewTerminal ( const char **argv, // argv[0] is executable const char **envp, const ArchSpec *arch_spec, bool stop_at_entry, bool disable_aslr); Cleaned up FileSpec::GetPath to not use strncpy() as it was always zero filling the entire path buffer. Fixed an issue with the dynamic checker function where I missed a '$' prefix that should have been added. llvm-svn: 116690
2010-10-18 06:03:32 +08:00
if (file.GetFileType () == FileSpec::eFileTypeDirectory)
{
Added a new Host call to find LLDB related paths: static bool Host::GetLLDBPath (lldb::PathType path_type, FileSpec &file_spec); This will fill in "file_spec" with an appropriate path that is appropriate for the current Host OS. MacOSX will return paths within the LLDB.framework, and other unixes will return the paths they want. The current PathType enums are: typedef enum PathType { ePathTypeLLDBShlibDir, // The directory where the lldb.so (unix) or LLDB mach-o file in LLDB.framework (MacOSX) exists ePathTypeSupportExecutableDir, // Find LLDB support executable directory (debugserver, etc) ePathTypeHeaderDir, // Find LLDB header file directory ePathTypePythonDir // Find Python modules (PYTHONPATH) directory } PathType; All places that were finding executables are and python paths are now updated to use this Host call. Added another new host call to launch the inferior in a terminal. This ability will be very host specific and doesn't need to be supported on all systems. MacOSX currently will create a new .command file and tell Terminal.app to open the .command file. It also uses the new "darwin-debug" app which is a small app that uses posix to exec (no fork) and stop at the entry point of the program. The GDB remote plug-in is almost able launch a process and attach to it, it currently will spawn the process, but it won't attach to it just yet. This will let LLDB not have to share the terminal with another process and a new terminal window will pop up when you launch. This won't get hooked up until we work out all of the kinks. The new Host function is: static lldb::pid_t Host::LaunchInNewTerminal ( const char **argv, // argv[0] is executable const char **envp, const ArchSpec *arch_spec, bool stop_at_entry, bool disable_aslr); Cleaned up FileSpec::GetPath to not use strncpy() as it was always zero filling the entire path buffer. Fixed an issue with the dynamic checker function where I missed a '$' prefix that should have been added. llvm-svn: 116690
2010-10-18 06:03:32 +08:00
char path[PATH_MAX];
if (file.GetPath(path, sizeof(path)))
{
Added a new Host call to find LLDB related paths: static bool Host::GetLLDBPath (lldb::PathType path_type, FileSpec &file_spec); This will fill in "file_spec" with an appropriate path that is appropriate for the current Host OS. MacOSX will return paths within the LLDB.framework, and other unixes will return the paths they want. The current PathType enums are: typedef enum PathType { ePathTypeLLDBShlibDir, // The directory where the lldb.so (unix) or LLDB mach-o file in LLDB.framework (MacOSX) exists ePathTypeSupportExecutableDir, // Find LLDB support executable directory (debugserver, etc) ePathTypeHeaderDir, // Find LLDB header file directory ePathTypePythonDir // Find Python modules (PYTHONPATH) directory } PathType; All places that were finding executables are and python paths are now updated to use this Host call. Added another new host call to launch the inferior in a terminal. This ability will be very host specific and doesn't need to be supported on all systems. MacOSX currently will create a new .command file and tell Terminal.app to open the .command file. It also uses the new "darwin-debug" app which is a small app that uses posix to exec (no fork) and stop at the entry point of the program. The GDB remote plug-in is almost able launch a process and attach to it, it currently will spawn the process, but it won't attach to it just yet. This will let LLDB not have to share the terminal with another process and a new terminal window will pop up when you launch. This won't get hooked up until we work out all of the kinks. The new Host function is: static lldb::pid_t Host::LaunchInNewTerminal ( const char **argv, // argv[0] is executable const char **envp, const ArchSpec *arch_spec, bool stop_at_entry, bool disable_aslr); Cleaned up FileSpec::GetPath to not use strncpy() as it was always zero filling the entire path buffer. Fixed an issue with the dynamic checker function where I missed a '$' prefix that should have been added. llvm-svn: 116690
2010-10-18 06:03:32 +08:00
CFCBundle bundle (path);
CFCReleaser<CFURLRef> url(bundle.CopyExecutableURL ());
if (url.get())
{
if (::CFURLGetFileSystemRepresentation (url.get(), YES, (UInt8*)path, sizeof(path)))
{
file.SetFile(path, false);
Added a new Host call to find LLDB related paths: static bool Host::GetLLDBPath (lldb::PathType path_type, FileSpec &file_spec); This will fill in "file_spec" with an appropriate path that is appropriate for the current Host OS. MacOSX will return paths within the LLDB.framework, and other unixes will return the paths they want. The current PathType enums are: typedef enum PathType { ePathTypeLLDBShlibDir, // The directory where the lldb.so (unix) or LLDB mach-o file in LLDB.framework (MacOSX) exists ePathTypeSupportExecutableDir, // Find LLDB support executable directory (debugserver, etc) ePathTypeHeaderDir, // Find LLDB header file directory ePathTypePythonDir // Find Python modules (PYTHONPATH) directory } PathType; All places that were finding executables are and python paths are now updated to use this Host call. Added another new host call to launch the inferior in a terminal. This ability will be very host specific and doesn't need to be supported on all systems. MacOSX currently will create a new .command file and tell Terminal.app to open the .command file. It also uses the new "darwin-debug" app which is a small app that uses posix to exec (no fork) and stop at the entry point of the program. The GDB remote plug-in is almost able launch a process and attach to it, it currently will spawn the process, but it won't attach to it just yet. This will let LLDB not have to share the terminal with another process and a new terminal window will pop up when you launch. This won't get hooked up until we work out all of the kinks. The new Host function is: static lldb::pid_t Host::LaunchInNewTerminal ( const char **argv, // argv[0] is executable const char **envp, const ArchSpec *arch_spec, bool stop_at_entry, bool disable_aslr); Cleaned up FileSpec::GetPath to not use strncpy() as it was always zero filling the entire path buffer. Fixed an issue with the dynamic checker function where I missed a '$' prefix that should have been added. llvm-svn: 116690
2010-10-18 06:03:32 +08:00
return true;
}
}
}
}
#endif
return false;
}
lldb::pid_t
Host::LaunchApplication (const FileSpec &app_file_spec)
{
#if defined (__arm__)
return LLDB_INVALID_PROCESS_ID;
#else
char app_path[PATH_MAX];
app_file_spec.GetPath(app_path, sizeof(app_path));
LSApplicationParameters app_params;
::memset (&app_params, 0, sizeof (app_params));
app_params.flags = kLSLaunchDefaults |
kLSLaunchDontAddToRecents |
Added a new Host call to find LLDB related paths: static bool Host::GetLLDBPath (lldb::PathType path_type, FileSpec &file_spec); This will fill in "file_spec" with an appropriate path that is appropriate for the current Host OS. MacOSX will return paths within the LLDB.framework, and other unixes will return the paths they want. The current PathType enums are: typedef enum PathType { ePathTypeLLDBShlibDir, // The directory where the lldb.so (unix) or LLDB mach-o file in LLDB.framework (MacOSX) exists ePathTypeSupportExecutableDir, // Find LLDB support executable directory (debugserver, etc) ePathTypeHeaderDir, // Find LLDB header file directory ePathTypePythonDir // Find Python modules (PYTHONPATH) directory } PathType; All places that were finding executables are and python paths are now updated to use this Host call. Added another new host call to launch the inferior in a terminal. This ability will be very host specific and doesn't need to be supported on all systems. MacOSX currently will create a new .command file and tell Terminal.app to open the .command file. It also uses the new "darwin-debug" app which is a small app that uses posix to exec (no fork) and stop at the entry point of the program. The GDB remote plug-in is almost able launch a process and attach to it, it currently will spawn the process, but it won't attach to it just yet. This will let LLDB not have to share the terminal with another process and a new terminal window will pop up when you launch. This won't get hooked up until we work out all of the kinks. The new Host function is: static lldb::pid_t Host::LaunchInNewTerminal ( const char **argv, // argv[0] is executable const char **envp, const ArchSpec *arch_spec, bool stop_at_entry, bool disable_aslr); Cleaned up FileSpec::GetPath to not use strncpy() as it was always zero filling the entire path buffer. Fixed an issue with the dynamic checker function where I missed a '$' prefix that should have been added. llvm-svn: 116690
2010-10-18 06:03:32 +08:00
kLSLaunchNewInstance;
FSRef app_fsref;
CFCString app_cfstr (app_path, kCFStringEncodingUTF8);
OSStatus error = ::FSPathMakeRef ((const UInt8 *)app_path, &app_fsref, NULL);
// If we found the app, then store away the name so we don't have to re-look it up.
if (error != noErr)
return LLDB_INVALID_PROCESS_ID;
app_params.application = &app_fsref;
ProcessSerialNumber psn;
error = ::LSOpenApplication (&app_params, &psn);
if (error != noErr)
return LLDB_INVALID_PROCESS_ID;
Added a new Host call to find LLDB related paths: static bool Host::GetLLDBPath (lldb::PathType path_type, FileSpec &file_spec); This will fill in "file_spec" with an appropriate path that is appropriate for the current Host OS. MacOSX will return paths within the LLDB.framework, and other unixes will return the paths they want. The current PathType enums are: typedef enum PathType { ePathTypeLLDBShlibDir, // The directory where the lldb.so (unix) or LLDB mach-o file in LLDB.framework (MacOSX) exists ePathTypeSupportExecutableDir, // Find LLDB support executable directory (debugserver, etc) ePathTypeHeaderDir, // Find LLDB header file directory ePathTypePythonDir // Find Python modules (PYTHONPATH) directory } PathType; All places that were finding executables are and python paths are now updated to use this Host call. Added another new host call to launch the inferior in a terminal. This ability will be very host specific and doesn't need to be supported on all systems. MacOSX currently will create a new .command file and tell Terminal.app to open the .command file. It also uses the new "darwin-debug" app which is a small app that uses posix to exec (no fork) and stop at the entry point of the program. The GDB remote plug-in is almost able launch a process and attach to it, it currently will spawn the process, but it won't attach to it just yet. This will let LLDB not have to share the terminal with another process and a new terminal window will pop up when you launch. This won't get hooked up until we work out all of the kinks. The new Host function is: static lldb::pid_t Host::LaunchInNewTerminal ( const char **argv, // argv[0] is executable const char **envp, const ArchSpec *arch_spec, bool stop_at_entry, bool disable_aslr); Cleaned up FileSpec::GetPath to not use strncpy() as it was always zero filling the entire path buffer. Fixed an issue with the dynamic checker function where I missed a '$' prefix that should have been added. llvm-svn: 116690
2010-10-18 06:03:32 +08:00
::pid_t pid = LLDB_INVALID_PROCESS_ID;
error = ::GetProcessPID(&psn, &pid);
return pid;
#endif
Added a new Host call to find LLDB related paths: static bool Host::GetLLDBPath (lldb::PathType path_type, FileSpec &file_spec); This will fill in "file_spec" with an appropriate path that is appropriate for the current Host OS. MacOSX will return paths within the LLDB.framework, and other unixes will return the paths they want. The current PathType enums are: typedef enum PathType { ePathTypeLLDBShlibDir, // The directory where the lldb.so (unix) or LLDB mach-o file in LLDB.framework (MacOSX) exists ePathTypeSupportExecutableDir, // Find LLDB support executable directory (debugserver, etc) ePathTypeHeaderDir, // Find LLDB header file directory ePathTypePythonDir // Find Python modules (PYTHONPATH) directory } PathType; All places that were finding executables are and python paths are now updated to use this Host call. Added another new host call to launch the inferior in a terminal. This ability will be very host specific and doesn't need to be supported on all systems. MacOSX currently will create a new .command file and tell Terminal.app to open the .command file. It also uses the new "darwin-debug" app which is a small app that uses posix to exec (no fork) and stop at the entry point of the program. The GDB remote plug-in is almost able launch a process and attach to it, it currently will spawn the process, but it won't attach to it just yet. This will let LLDB not have to share the terminal with another process and a new terminal window will pop up when you launch. This won't get hooked up until we work out all of the kinks. The new Host function is: static lldb::pid_t Host::LaunchInNewTerminal ( const char **argv, // argv[0] is executable const char **envp, const ArchSpec *arch_spec, bool stop_at_entry, bool disable_aslr); Cleaned up FileSpec::GetPath to not use strncpy() as it was always zero filling the entire path buffer. Fixed an issue with the dynamic checker function where I missed a '$' prefix that should have been added. llvm-svn: 116690
2010-10-18 06:03:32 +08:00
}
static void *
AcceptPIDFromInferior (void *arg)
{
const char *connect_url = (const char *)arg;
ConnectionFileDescriptor file_conn;
Error error;
if (file_conn.Connect (connect_url, &error) == eConnectionStatusSuccess)
{
char pid_str[256];
::memset (pid_str, 0, sizeof(pid_str));
ConnectionStatus status;
const size_t pid_str_len = file_conn.Read (pid_str, sizeof(pid_str), NULL, status, NULL);
if (pid_str_len > 0)
{
int pid = atoi (pid_str);
return (void *)(intptr_t)pid;
}
}
return NULL;
}
static bool
WaitForProcessToSIGSTOP (const lldb::pid_t pid, const int timeout_in_seconds)
{
const int time_delta_usecs = 100000;
const int num_retries = timeout_in_seconds/time_delta_usecs;
for (int i=0; i<num_retries; i++)
{
struct proc_bsdinfo bsd_info;
int error = ::proc_pidinfo (pid, PROC_PIDTBSDINFO,
(uint64_t) 0,
&bsd_info,
PROC_PIDTBSDINFO_SIZE);
switch (error)
{
case EINVAL:
case ENOTSUP:
case ESRCH:
case EPERM:
return false;
default:
break;
case 0:
if (bsd_info.pbi_status == SSTOP)
return true;
}
::usleep (time_delta_usecs);
}
return false;
}
#if !defined(__arm__)
static lldb::pid_t
LaunchInNewTerminalWithCommandFile
Added a new Host call to find LLDB related paths: static bool Host::GetLLDBPath (lldb::PathType path_type, FileSpec &file_spec); This will fill in "file_spec" with an appropriate path that is appropriate for the current Host OS. MacOSX will return paths within the LLDB.framework, and other unixes will return the paths they want. The current PathType enums are: typedef enum PathType { ePathTypeLLDBShlibDir, // The directory where the lldb.so (unix) or LLDB mach-o file in LLDB.framework (MacOSX) exists ePathTypeSupportExecutableDir, // Find LLDB support executable directory (debugserver, etc) ePathTypeHeaderDir, // Find LLDB header file directory ePathTypePythonDir // Find Python modules (PYTHONPATH) directory } PathType; All places that were finding executables are and python paths are now updated to use this Host call. Added another new host call to launch the inferior in a terminal. This ability will be very host specific and doesn't need to be supported on all systems. MacOSX currently will create a new .command file and tell Terminal.app to open the .command file. It also uses the new "darwin-debug" app which is a small app that uses posix to exec (no fork) and stop at the entry point of the program. The GDB remote plug-in is almost able launch a process and attach to it, it currently will spawn the process, but it won't attach to it just yet. This will let LLDB not have to share the terminal with another process and a new terminal window will pop up when you launch. This won't get hooked up until we work out all of the kinks. The new Host function is: static lldb::pid_t Host::LaunchInNewTerminal ( const char **argv, // argv[0] is executable const char **envp, const ArchSpec *arch_spec, bool stop_at_entry, bool disable_aslr); Cleaned up FileSpec::GetPath to not use strncpy() as it was always zero filling the entire path buffer. Fixed an issue with the dynamic checker function where I missed a '$' prefix that should have been added. llvm-svn: 116690
2010-10-18 06:03:32 +08:00
(
const char **argv,
const char **envp,
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,
Added a new Host call to find LLDB related paths: static bool Host::GetLLDBPath (lldb::PathType path_type, FileSpec &file_spec); This will fill in "file_spec" with an appropriate path that is appropriate for the current Host OS. MacOSX will return paths within the LLDB.framework, and other unixes will return the paths they want. The current PathType enums are: typedef enum PathType { ePathTypeLLDBShlibDir, // The directory where the lldb.so (unix) or LLDB mach-o file in LLDB.framework (MacOSX) exists ePathTypeSupportExecutableDir, // Find LLDB support executable directory (debugserver, etc) ePathTypeHeaderDir, // Find LLDB header file directory ePathTypePythonDir // Find Python modules (PYTHONPATH) directory } PathType; All places that were finding executables are and python paths are now updated to use this Host call. Added another new host call to launch the inferior in a terminal. This ability will be very host specific and doesn't need to be supported on all systems. MacOSX currently will create a new .command file and tell Terminal.app to open the .command file. It also uses the new "darwin-debug" app which is a small app that uses posix to exec (no fork) and stop at the entry point of the program. The GDB remote plug-in is almost able launch a process and attach to it, it currently will spawn the process, but it won't attach to it just yet. This will let LLDB not have to share the terminal with another process and a new terminal window will pop up when you launch. This won't get hooked up until we work out all of the kinks. The new Host function is: static lldb::pid_t Host::LaunchInNewTerminal ( const char **argv, // argv[0] is executable const char **envp, const ArchSpec *arch_spec, bool stop_at_entry, bool disable_aslr); Cleaned up FileSpec::GetPath to not use strncpy() as it was always zero filling the entire path buffer. Fixed an issue with the dynamic checker function where I missed a '$' prefix that should have been added. llvm-svn: 116690
2010-10-18 06:03:32 +08:00
const ArchSpec *arch_spec,
bool stop_at_entry,
bool disable_aslr
)
{
if (!argv || !argv[0])
return LLDB_INVALID_PROCESS_ID;
OSStatus error = 0;
FileSpec program (argv[0], false);
Added a new Host call to find LLDB related paths: static bool Host::GetLLDBPath (lldb::PathType path_type, FileSpec &file_spec); This will fill in "file_spec" with an appropriate path that is appropriate for the current Host OS. MacOSX will return paths within the LLDB.framework, and other unixes will return the paths they want. The current PathType enums are: typedef enum PathType { ePathTypeLLDBShlibDir, // The directory where the lldb.so (unix) or LLDB mach-o file in LLDB.framework (MacOSX) exists ePathTypeSupportExecutableDir, // Find LLDB support executable directory (debugserver, etc) ePathTypeHeaderDir, // Find LLDB header file directory ePathTypePythonDir // Find Python modules (PYTHONPATH) directory } PathType; All places that were finding executables are and python paths are now updated to use this Host call. Added another new host call to launch the inferior in a terminal. This ability will be very host specific and doesn't need to be supported on all systems. MacOSX currently will create a new .command file and tell Terminal.app to open the .command file. It also uses the new "darwin-debug" app which is a small app that uses posix to exec (no fork) and stop at the entry point of the program. The GDB remote plug-in is almost able launch a process and attach to it, it currently will spawn the process, but it won't attach to it just yet. This will let LLDB not have to share the terminal with another process and a new terminal window will pop up when you launch. This won't get hooked up until we work out all of the kinks. The new Host function is: static lldb::pid_t Host::LaunchInNewTerminal ( const char **argv, // argv[0] is executable const char **envp, const ArchSpec *arch_spec, bool stop_at_entry, bool disable_aslr); Cleaned up FileSpec::GetPath to not use strncpy() as it was always zero filling the entire path buffer. Fixed an issue with the dynamic checker function where I missed a '$' prefix that should have been added. llvm-svn: 116690
2010-10-18 06:03:32 +08:00
std::string unix_socket_name;
Added a new Host call to find LLDB related paths: static bool Host::GetLLDBPath (lldb::PathType path_type, FileSpec &file_spec); This will fill in "file_spec" with an appropriate path that is appropriate for the current Host OS. MacOSX will return paths within the LLDB.framework, and other unixes will return the paths they want. The current PathType enums are: typedef enum PathType { ePathTypeLLDBShlibDir, // The directory where the lldb.so (unix) or LLDB mach-o file in LLDB.framework (MacOSX) exists ePathTypeSupportExecutableDir, // Find LLDB support executable directory (debugserver, etc) ePathTypeHeaderDir, // Find LLDB header file directory ePathTypePythonDir // Find Python modules (PYTHONPATH) directory } PathType; All places that were finding executables are and python paths are now updated to use this Host call. Added another new host call to launch the inferior in a terminal. This ability will be very host specific and doesn't need to be supported on all systems. MacOSX currently will create a new .command file and tell Terminal.app to open the .command file. It also uses the new "darwin-debug" app which is a small app that uses posix to exec (no fork) and stop at the entry point of the program. The GDB remote plug-in is almost able launch a process and attach to it, it currently will spawn the process, but it won't attach to it just yet. This will let LLDB not have to share the terminal with another process and a new terminal window will pop up when you launch. This won't get hooked up until we work out all of the kinks. The new Host function is: static lldb::pid_t Host::LaunchInNewTerminal ( const char **argv, // argv[0] is executable const char **envp, const ArchSpec *arch_spec, bool stop_at_entry, bool disable_aslr); Cleaned up FileSpec::GetPath to not use strncpy() as it was always zero filling the entire path buffer. Fixed an issue with the dynamic checker function where I missed a '$' prefix that should have been added. llvm-svn: 116690
2010-10-18 06:03:32 +08:00
char temp_file_path[PATH_MAX];
const char *tmpdir = ::getenv ("TMPDIR");
if (tmpdir == NULL)
tmpdir = "/tmp/";
::snprintf (temp_file_path, sizeof(temp_file_path), "%s%s-XXXXXX", tmpdir, program.GetFilename().AsCString());
if (::mktemp (temp_file_path) == NULL)
return LLDB_INVALID_PROCESS_ID;
unix_socket_name.assign (temp_file_path);
Added a new Host call to find LLDB related paths: static bool Host::GetLLDBPath (lldb::PathType path_type, FileSpec &file_spec); This will fill in "file_spec" with an appropriate path that is appropriate for the current Host OS. MacOSX will return paths within the LLDB.framework, and other unixes will return the paths they want. The current PathType enums are: typedef enum PathType { ePathTypeLLDBShlibDir, // The directory where the lldb.so (unix) or LLDB mach-o file in LLDB.framework (MacOSX) exists ePathTypeSupportExecutableDir, // Find LLDB support executable directory (debugserver, etc) ePathTypeHeaderDir, // Find LLDB header file directory ePathTypePythonDir // Find Python modules (PYTHONPATH) directory } PathType; All places that were finding executables are and python paths are now updated to use this Host call. Added another new host call to launch the inferior in a terminal. This ability will be very host specific and doesn't need to be supported on all systems. MacOSX currently will create a new .command file and tell Terminal.app to open the .command file. It also uses the new "darwin-debug" app which is a small app that uses posix to exec (no fork) and stop at the entry point of the program. The GDB remote plug-in is almost able launch a process and attach to it, it currently will spawn the process, but it won't attach to it just yet. This will let LLDB not have to share the terminal with another process and a new terminal window will pop up when you launch. This won't get hooked up until we work out all of the kinks. The new Host function is: static lldb::pid_t Host::LaunchInNewTerminal ( const char **argv, // argv[0] is executable const char **envp, const ArchSpec *arch_spec, bool stop_at_entry, bool disable_aslr); Cleaned up FileSpec::GetPath to not use strncpy() as it was always zero filling the entire path buffer. Fixed an issue with the dynamic checker function where I missed a '$' prefix that should have been added. llvm-svn: 116690
2010-10-18 06:03:32 +08:00
::strncat (temp_file_path, ".command", sizeof (temp_file_path));
StreamFile command_file;
command_file.GetFile().Open (temp_file_path,
File::eOpenOptionWrite | File::eOpenOptionCanCreate,
File::ePermissionsDefault);
if (!command_file.GetFile().IsValid())
return LLDB_INVALID_PROCESS_ID;
Added a new Host call to find LLDB related paths: static bool Host::GetLLDBPath (lldb::PathType path_type, FileSpec &file_spec); This will fill in "file_spec" with an appropriate path that is appropriate for the current Host OS. MacOSX will return paths within the LLDB.framework, and other unixes will return the paths they want. The current PathType enums are: typedef enum PathType { ePathTypeLLDBShlibDir, // The directory where the lldb.so (unix) or LLDB mach-o file in LLDB.framework (MacOSX) exists ePathTypeSupportExecutableDir, // Find LLDB support executable directory (debugserver, etc) ePathTypeHeaderDir, // Find LLDB header file directory ePathTypePythonDir // Find Python modules (PYTHONPATH) directory } PathType; All places that were finding executables are and python paths are now updated to use this Host call. Added another new host call to launch the inferior in a terminal. This ability will be very host specific and doesn't need to be supported on all systems. MacOSX currently will create a new .command file and tell Terminal.app to open the .command file. It also uses the new "darwin-debug" app which is a small app that uses posix to exec (no fork) and stop at the entry point of the program. The GDB remote plug-in is almost able launch a process and attach to it, it currently will spawn the process, but it won't attach to it just yet. This will let LLDB not have to share the terminal with another process and a new terminal window will pop up when you launch. This won't get hooked up until we work out all of the kinks. The new Host function is: static lldb::pid_t Host::LaunchInNewTerminal ( const char **argv, // argv[0] is executable const char **envp, const ArchSpec *arch_spec, bool stop_at_entry, bool disable_aslr); Cleaned up FileSpec::GetPath to not use strncpy() as it was always zero filling the entire path buffer. Fixed an issue with the dynamic checker function where I missed a '$' prefix that should have been added. llvm-svn: 116690
2010-10-18 06:03:32 +08:00
FileSpec darwin_debug_file_spec;
if (!Host::GetLLDBPath (ePathTypeSupportExecutableDir, darwin_debug_file_spec))
return LLDB_INVALID_PROCESS_ID;
darwin_debug_file_spec.GetFilename().SetCString("darwin-debug");
if (!darwin_debug_file_spec.Exists())
return LLDB_INVALID_PROCESS_ID;
char launcher_path[PATH_MAX];
darwin_debug_file_spec.GetPath(launcher_path, sizeof(launcher_path));
command_file.Printf("\"%s\" ", launcher_path);
command_file.Printf("--unix-socket=%s ", unix_socket_name.c_str());
Added a new Host call to find LLDB related paths: static bool Host::GetLLDBPath (lldb::PathType path_type, FileSpec &file_spec); This will fill in "file_spec" with an appropriate path that is appropriate for the current Host OS. MacOSX will return paths within the LLDB.framework, and other unixes will return the paths they want. The current PathType enums are: typedef enum PathType { ePathTypeLLDBShlibDir, // The directory where the lldb.so (unix) or LLDB mach-o file in LLDB.framework (MacOSX) exists ePathTypeSupportExecutableDir, // Find LLDB support executable directory (debugserver, etc) ePathTypeHeaderDir, // Find LLDB header file directory ePathTypePythonDir // Find Python modules (PYTHONPATH) directory } PathType; All places that were finding executables are and python paths are now updated to use this Host call. Added another new host call to launch the inferior in a terminal. This ability will be very host specific and doesn't need to be supported on all systems. MacOSX currently will create a new .command file and tell Terminal.app to open the .command file. It also uses the new "darwin-debug" app which is a small app that uses posix to exec (no fork) and stop at the entry point of the program. The GDB remote plug-in is almost able launch a process and attach to it, it currently will spawn the process, but it won't attach to it just yet. This will let LLDB not have to share the terminal with another process and a new terminal window will pop up when you launch. This won't get hooked up until we work out all of the kinks. The new Host function is: static lldb::pid_t Host::LaunchInNewTerminal ( const char **argv, // argv[0] is executable const char **envp, const ArchSpec *arch_spec, bool stop_at_entry, bool disable_aslr); Cleaned up FileSpec::GetPath to not use strncpy() as it was always zero filling the entire path buffer. Fixed an issue with the dynamic checker function where I missed a '$' prefix that should have been added. llvm-svn: 116690
2010-10-18 06:03:32 +08:00
if (arch_spec && 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
command_file.Printf("--arch=%s ", arch_spec->GetArchitectureName());
Added a new Host call to find LLDB related paths: static bool Host::GetLLDBPath (lldb::PathType path_type, FileSpec &file_spec); This will fill in "file_spec" with an appropriate path that is appropriate for the current Host OS. MacOSX will return paths within the LLDB.framework, and other unixes will return the paths they want. The current PathType enums are: typedef enum PathType { ePathTypeLLDBShlibDir, // The directory where the lldb.so (unix) or LLDB mach-o file in LLDB.framework (MacOSX) exists ePathTypeSupportExecutableDir, // Find LLDB support executable directory (debugserver, etc) ePathTypeHeaderDir, // Find LLDB header file directory ePathTypePythonDir // Find Python modules (PYTHONPATH) directory } PathType; All places that were finding executables are and python paths are now updated to use this Host call. Added another new host call to launch the inferior in a terminal. This ability will be very host specific and doesn't need to be supported on all systems. MacOSX currently will create a new .command file and tell Terminal.app to open the .command file. It also uses the new "darwin-debug" app which is a small app that uses posix to exec (no fork) and stop at the entry point of the program. The GDB remote plug-in is almost able launch a process and attach to it, it currently will spawn the process, but it won't attach to it just yet. This will let LLDB not have to share the terminal with another process and a new terminal window will pop up when you launch. This won't get hooked up until we work out all of the kinks. The new Host function is: static lldb::pid_t Host::LaunchInNewTerminal ( const char **argv, // argv[0] is executable const char **envp, const ArchSpec *arch_spec, bool stop_at_entry, bool disable_aslr); Cleaned up FileSpec::GetPath to not use strncpy() as it was always zero filling the entire path buffer. Fixed an issue with the dynamic checker function where I missed a '$' prefix that should have been added. llvm-svn: 116690
2010-10-18 06:03:32 +08:00
}
if (disable_aslr)
{
command_file.PutCString("--disable-aslr ");
}
command_file.PutCString("-- ");
if (argv)
{
for (size_t i=0; argv[i] != NULL; ++i)
{
command_file.Printf("\"%s\" ", argv[i]);
}
}
command_file.PutCString("\necho Process exited with status $?\n");
command_file.GetFile().Close();
Added a new Host call to find LLDB related paths: static bool Host::GetLLDBPath (lldb::PathType path_type, FileSpec &file_spec); This will fill in "file_spec" with an appropriate path that is appropriate for the current Host OS. MacOSX will return paths within the LLDB.framework, and other unixes will return the paths they want. The current PathType enums are: typedef enum PathType { ePathTypeLLDBShlibDir, // The directory where the lldb.so (unix) or LLDB mach-o file in LLDB.framework (MacOSX) exists ePathTypeSupportExecutableDir, // Find LLDB support executable directory (debugserver, etc) ePathTypeHeaderDir, // Find LLDB header file directory ePathTypePythonDir // Find Python modules (PYTHONPATH) directory } PathType; All places that were finding executables are and python paths are now updated to use this Host call. Added another new host call to launch the inferior in a terminal. This ability will be very host specific and doesn't need to be supported on all systems. MacOSX currently will create a new .command file and tell Terminal.app to open the .command file. It also uses the new "darwin-debug" app which is a small app that uses posix to exec (no fork) and stop at the entry point of the program. The GDB remote plug-in is almost able launch a process and attach to it, it currently will spawn the process, but it won't attach to it just yet. This will let LLDB not have to share the terminal with another process and a new terminal window will pop up when you launch. This won't get hooked up until we work out all of the kinks. The new Host function is: static lldb::pid_t Host::LaunchInNewTerminal ( const char **argv, // argv[0] is executable const char **envp, const ArchSpec *arch_spec, bool stop_at_entry, bool disable_aslr); Cleaned up FileSpec::GetPath to not use strncpy() as it was always zero filling the entire path buffer. Fixed an issue with the dynamic checker function where I missed a '$' prefix that should have been added. llvm-svn: 116690
2010-10-18 06:03:32 +08:00
if (::chmod (temp_file_path, S_IRWXU | S_IRWXG) != 0)
return LLDB_INVALID_PROCESS_ID;
CFCMutableDictionary cf_env_dict;
const bool can_create = true;
if (envp)
{
for (size_t i=0; envp[i] != NULL; ++i)
{
const char *env_entry = envp[i];
const char *equal_pos = strchr(env_entry, '=');
if (equal_pos)
{
std::string env_key (env_entry, equal_pos);
std::string env_val (equal_pos + 1);
CFCString cf_env_key (env_key.c_str(), kCFStringEncodingUTF8);
CFCString cf_env_val (env_val.c_str(), kCFStringEncodingUTF8);
cf_env_dict.AddValue (cf_env_key.get(), cf_env_val.get(), can_create);
}
}
}
LSApplicationParameters app_params;
::memset (&app_params, 0, sizeof (app_params));
Added a new Host call to find LLDB related paths: static bool Host::GetLLDBPath (lldb::PathType path_type, FileSpec &file_spec); This will fill in "file_spec" with an appropriate path that is appropriate for the current Host OS. MacOSX will return paths within the LLDB.framework, and other unixes will return the paths they want. The current PathType enums are: typedef enum PathType { ePathTypeLLDBShlibDir, // The directory where the lldb.so (unix) or LLDB mach-o file in LLDB.framework (MacOSX) exists ePathTypeSupportExecutableDir, // Find LLDB support executable directory (debugserver, etc) ePathTypeHeaderDir, // Find LLDB header file directory ePathTypePythonDir // Find Python modules (PYTHONPATH) directory } PathType; All places that were finding executables are and python paths are now updated to use this Host call. Added another new host call to launch the inferior in a terminal. This ability will be very host specific and doesn't need to be supported on all systems. MacOSX currently will create a new .command file and tell Terminal.app to open the .command file. It also uses the new "darwin-debug" app which is a small app that uses posix to exec (no fork) and stop at the entry point of the program. The GDB remote plug-in is almost able launch a process and attach to it, it currently will spawn the process, but it won't attach to it just yet. This will let LLDB not have to share the terminal with another process and a new terminal window will pop up when you launch. This won't get hooked up until we work out all of the kinks. The new Host function is: static lldb::pid_t Host::LaunchInNewTerminal ( const char **argv, // argv[0] is executable const char **envp, const ArchSpec *arch_spec, bool stop_at_entry, bool disable_aslr); Cleaned up FileSpec::GetPath to not use strncpy() as it was always zero filling the entire path buffer. Fixed an issue with the dynamic checker function where I missed a '$' prefix that should have been added. llvm-svn: 116690
2010-10-18 06:03:32 +08:00
app_params.flags = kLSLaunchDontAddToRecents | kLSLaunchAsync;
app_params.argv = NULL;
app_params.environment = (CFDictionaryRef)cf_env_dict.get();
CFCReleaser<CFURLRef> command_file_url (::CFURLCreateFromFileSystemRepresentation (NULL,
(const UInt8 *)temp_file_path,
strlen(temp_file_path),
Added a new Host call to find LLDB related paths: static bool Host::GetLLDBPath (lldb::PathType path_type, FileSpec &file_spec); This will fill in "file_spec" with an appropriate path that is appropriate for the current Host OS. MacOSX will return paths within the LLDB.framework, and other unixes will return the paths they want. The current PathType enums are: typedef enum PathType { ePathTypeLLDBShlibDir, // The directory where the lldb.so (unix) or LLDB mach-o file in LLDB.framework (MacOSX) exists ePathTypeSupportExecutableDir, // Find LLDB support executable directory (debugserver, etc) ePathTypeHeaderDir, // Find LLDB header file directory ePathTypePythonDir // Find Python modules (PYTHONPATH) directory } PathType; All places that were finding executables are and python paths are now updated to use this Host call. Added another new host call to launch the inferior in a terminal. This ability will be very host specific and doesn't need to be supported on all systems. MacOSX currently will create a new .command file and tell Terminal.app to open the .command file. It also uses the new "darwin-debug" app which is a small app that uses posix to exec (no fork) and stop at the entry point of the program. The GDB remote plug-in is almost able launch a process and attach to it, it currently will spawn the process, but it won't attach to it just yet. This will let LLDB not have to share the terminal with another process and a new terminal window will pop up when you launch. This won't get hooked up until we work out all of the kinks. The new Host function is: static lldb::pid_t Host::LaunchInNewTerminal ( const char **argv, // argv[0] is executable const char **envp, const ArchSpec *arch_spec, bool stop_at_entry, bool disable_aslr); Cleaned up FileSpec::GetPath to not use strncpy() as it was always zero filling the entire path buffer. Fixed an issue with the dynamic checker function where I missed a '$' prefix that should have been added. llvm-svn: 116690
2010-10-18 06:03:32 +08:00
false));
CFCMutableArray urls;
// Terminal.app will open the ".command" file we have created
// and run our process inside it which will wait at the entry point
// for us to attach.
urls.AppendValue(command_file_url.get());
lldb::pid_t pid = LLDB_INVALID_PROCESS_ID;
Error lldb_error;
// Sleep and wait a bit for debugserver to start to listen...
char connect_url[128];
::snprintf (connect_url, sizeof(connect_url), "unix-accept://%s", unix_socket_name.c_str());
// Spawn a new thread to accept incoming connection on the connect_url
// so we can grab the pid from the inferior
lldb::thread_t accept_thread = Host::ThreadCreate (unix_socket_name.c_str(),
AcceptPIDFromInferior,
connect_url,
&lldb_error);
ProcessSerialNumber psn;
error = LSOpenURLsWithRole(urls.get(), kLSRolesShell, NULL, &app_params, &psn, 1);
if (error == noErr)
{
thread_result_t accept_thread_result = NULL;
if (Host::ThreadJoin (accept_thread, &accept_thread_result, &lldb_error))
{
if (accept_thread_result)
{
pid = (intptr_t)accept_thread_result;
// Wait for process to be stopped the the entry point by watching
// for the process status to be set to SSTOP which indicates it it
// SIGSTOP'ed at the entry point
WaitForProcessToSIGSTOP (pid, 5);
}
}
}
else
{
Host::ThreadCancel (accept_thread, &lldb_error);
}
return pid;
}
const char *applscript_in_new_tty =
"tell application \"Terminal\"\n"
" do script \"%s\"\n"
"end tell\n";
const char *applscript_in_existing_tty = "\
set the_shell_script to \"%s\"\n\
tell application \"Terminal\"\n\
repeat with the_window in (get windows)\n\
repeat with the_tab in tabs of the_window\n\
set the_tty to tty in the_tab\n\
if the_tty contains \"%s\" then\n\
if the_tab is not busy then\n\
set selected of the_tab to true\n\
set frontmost of the_window to true\n\
do script the_shell_script in the_tab\n\
return\n\
end if\n\
end if\n\
end repeat\n\
end repeat\n\
do script the_shell_script\n\
end tell\n";
static Error
LaunchInNewTerminalWithAppleScript (const char *exe_path, ProcessLaunchInfo &launch_info)
{
Error error;
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
char unix_socket_name[PATH_MAX] = "/tmp/XXXXXX";
if (::mktemp (unix_socket_name) == NULL)
{
error.SetErrorString ("failed to make temporary path for a unix socket");
return error;
}
StreamString command;
FileSpec darwin_debug_file_spec;
if (!Host::GetLLDBPath (ePathTypeSupportExecutableDir, darwin_debug_file_spec))
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.SetErrorString ("can't locate the 'darwin-debug' executable");
return error;
}
darwin_debug_file_spec.GetFilename().SetCString("darwin-debug");
if (!darwin_debug_file_spec.Exists())
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.SetErrorStringWithFormat ("the 'darwin-debug' executable doesn't exists at %s/%s",
darwin_debug_file_spec.GetDirectory().GetCString(),
darwin_debug_file_spec.GetFilename().GetCString());
return error;
}
char launcher_path[PATH_MAX];
darwin_debug_file_spec.GetPath(launcher_path, sizeof(launcher_path));
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 ArchSpec &arch_spec = launch_info.GetArchitecture();
if (arch_spec.IsValid())
command.Printf("arch -arch %s ", arch_spec.GetArchitectureName());
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
command.Printf("'%s' --unix-socket=%s", launcher_path, unix_socket_name);
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 (arch_spec.IsValid())
command.Printf(" --arch=%s", arch_spec.GetArchitectureName());
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 *working_dir = launch_info.GetWorkingDirectory();
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
if (working_dir)
command.Printf(" --working-dir '%s'", working_dir);
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 (launch_info.GetFlags().Test (eLaunchFlagDisableASLR))
command.PutCString(" --disable-aslr");
command.PutCString(" -- ");
const char **argv = launch_info.GetArguments().GetConstArgumentVector ();
if (argv)
{
for (size_t i=0; argv[i] != NULL; ++i)
{
if (i==0)
command.Printf(" '%s'", exe_path);
else
command.Printf(" '%s'", argv[i]);
}
}
else
{
command.Printf(" '%s'", exe_path);
}
command.PutCString (" ; echo Process exited with status $?");
StreamString applescript_source;
const char *tty_command = command.GetString().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
// if (tty_name && tty_name[0])
// {
// applescript_source.Printf (applscript_in_existing_tty,
// tty_command,
// tty_name);
// }
// else
// {
applescript_source.Printf (applscript_in_new_tty,
tty_command);
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 *script_source = applescript_source.GetString().c_str();
//puts (script_source);
NSAppleScript* applescript = [[NSAppleScript alloc] initWithSource:[NSString stringWithCString:script_source encoding:NSUTF8StringEncoding]];
lldb::pid_t pid = LLDB_INVALID_PROCESS_ID;
Error lldb_error;
// Sleep and wait a bit for debugserver to start to listen...
ConnectionFileDescriptor file_conn;
char connect_url[128];
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
::snprintf (connect_url, sizeof(connect_url), "unix-accept://%s", unix_socket_name);
// Spawn a new thread to accept incoming connection on the connect_url
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
// so we can grab the pid from the inferior. We have to do this because we
// are sending an AppleScript that will launch a process in Terminal.app,
// in a shell and the shell will fork/exec a couple of times before we get
// to the process that we wanted to launch. So when our process actually
// gets launched, we will handshake with it and get the process ID for it.
lldb::thread_t accept_thread = Host::ThreadCreate (unix_socket_name,
AcceptPIDFromInferior,
connect_url,
&lldb_error);
[applescript executeAndReturnError:nil];
thread_result_t accept_thread_result = NULL;
if (Host::ThreadJoin (accept_thread, &accept_thread_result, &lldb_error))
{
if (accept_thread_result)
{
pid = (intptr_t)accept_thread_result;
// Wait for process to be stopped the the entry point by watching
// for the process status to be set to SSTOP which indicates it it
// SIGSTOP'ed at the entry point
WaitForProcessToSIGSTOP (pid, 5);
}
}
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
::unlink (unix_socket_name);
[applescript release];
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 (pid != LLDB_INVALID_PROCESS_ID)
launch_info.SetProcessID (pid);
return error;
}
#endif // #if !defined(__arm__)
// On MacOSX CrashReporter will display a string for each shared library if
// the shared library has an exported symbol named "__crashreporter_info__".
static Mutex&
GetCrashReporterMutex ()
{
static Mutex g_mutex;
return g_mutex;
}
extern "C" {
const char *__crashreporter_info__ = NULL;
}
asm(".desc ___crashreporter_info__, 0x10");
void
Host::SetCrashDescriptionWithFormat (const char *format, ...)
{
static StreamString g_crash_description;
Mutex::Locker locker (GetCrashReporterMutex ());
if (format)
{
va_list args;
va_start (args, format);
g_crash_description.GetString().clear();
g_crash_description.PrintfVarArg(format, args);
va_end (args);
__crashreporter_info__ = g_crash_description.GetData();
}
else
{
__crashreporter_info__ = NULL;
}
}
void
Host::SetCrashDescription (const char *cstr)
{
Mutex::Locker locker (GetCrashReporterMutex ());
static std::string g_crash_description;
if (cstr)
{
g_crash_description.assign (cstr);
__crashreporter_info__ = g_crash_description.c_str();
}
else
{
__crashreporter_info__ = NULL;
}
}
bool
Host::OpenFileInExternalEditor (const FileSpec &file_spec, uint32_t line_no)
{
#if defined(__arm__)
return false;
#else
// We attach this to an 'odoc' event to specify a particular selection
typedef struct {
int16_t reserved0; // must be zero
int16_t fLineNumber;
int32_t fSelStart;
int32_t fSelEnd;
uint32_t reserved1; // must be zero
uint32_t reserved2; // must be zero
} BabelAESelInfo;
LogSP log(lldb_private::GetLogIfAnyCategoriesSet (LIBLLDB_LOG_HOST));
char file_path[PATH_MAX];
file_spec.GetPath(file_path, PATH_MAX);
CFCString file_cfstr (file_path, kCFStringEncodingUTF8);
CFCReleaser<CFURLRef> file_URL (::CFURLCreateWithFileSystemPath (NULL,
file_cfstr.get(),
kCFURLPOSIXPathStyle,
false));
if (log)
log->Printf("Sending source file: \"%s\" and line: %d to external editor.\n", file_path, line_no);
long error;
BabelAESelInfo file_and_line_info =
{
0, // reserved0
line_no - 1, // fLineNumber (zero based line number)
1, // fSelStart
1024, // fSelEnd
0, // reserved1
0 // reserved2
};
AEKeyDesc file_and_line_desc;
error = ::AECreateDesc (typeUTF8Text,
&file_and_line_info,
sizeof (file_and_line_info),
&(file_and_line_desc.descContent));
if (error != noErr)
{
if (log)
log->Printf("Error creating AEDesc: %ld.\n", error);
return false;
}
file_and_line_desc.descKey = keyAEPosition;
static std::string g_app_name;
static FSRef g_app_fsref;
LSApplicationParameters app_params;
::memset (&app_params, 0, sizeof (app_params));
app_params.flags = kLSLaunchDefaults |
kLSLaunchDontAddToRecents |
kLSLaunchDontSwitch;
char *external_editor = ::getenv ("LLDB_EXTERNAL_EDITOR");
if (external_editor)
{
if (log)
log->Printf("Looking for external editor \"%s\".\n", external_editor);
if (g_app_name.empty() || strcmp (g_app_name.c_str(), external_editor) != 0)
{
CFCString editor_name (external_editor, kCFStringEncodingUTF8);
error = ::LSFindApplicationForInfo (kLSUnknownCreator,
NULL,
editor_name.get(),
&g_app_fsref,
NULL);
// If we found the app, then store away the name so we don't have to re-look it up.
if (error != noErr)
{
if (log)
log->Printf("Could not find External Editor application, error: %ld.\n", error);
return false;
}
}
app_params.application = &g_app_fsref;
}
ProcessSerialNumber psn;
CFCReleaser<CFArrayRef> file_array(CFArrayCreate (NULL, (const void **) file_URL.ptr_address(false), 1, NULL));
error = ::LSOpenURLsWithRole (file_array.get(),
kLSRolesAll,
&file_and_line_desc,
&app_params,
&psn,
1);
AEDisposeDesc (&(file_and_line_desc.descContent));
if (error != noErr)
{
if (log)
log->Printf("LSOpenURLsWithRole failed, error: %ld.\n", error);
return false;
}
ProcessInfoRec which_process;
::memset(&which_process, 0, sizeof(which_process));
unsigned char ap_name[PATH_MAX];
which_process.processName = ap_name;
error = ::GetProcessInformation (&psn, &which_process);
bool using_xcode;
if (error != noErr)
{
if (log)
log->Printf("GetProcessInformation failed, error: %ld.\n", error);
using_xcode = false;
}
else
using_xcode = strncmp((char *) ap_name+1, "Xcode", (int) ap_name[0]) == 0;
// Xcode doesn't obey the line number in the Open Apple Event. So I have to send
// it an AppleScript to focus on the right line.
if (using_xcode)
{
static ComponentInstance osa_component = NULL;
static const char *as_template = "tell application \"Xcode\"\n"
"set doc to the first document whose path is \"%s\"\n"
"set the selection to paragraph %d of doc\n"
"--- set the selected paragraph range to {%d, %d} of doc\n"
"end tell\n";
const int chars_for_int = 32;
static int as_template_len = strlen (as_template);
char *as_str;
AEDesc as_desc;
if (osa_component == NULL)
{
osa_component = ::OpenDefaultComponent (kOSAComponentType,
kAppleScriptSubtype);
}
if (osa_component == NULL)
{
if (log)
log->Printf("Could not get default AppleScript component.\n");
return false;
}
uint32_t as_str_size = as_template_len + strlen (file_path) + 3 * chars_for_int + 1;
as_str = (char *) malloc (as_str_size);
::snprintf (as_str,
as_str_size - 1,
as_template,
file_path,
line_no,
line_no,
line_no);
error = ::AECreateDesc (typeChar,
as_str,
strlen (as_str),
&as_desc);
::free (as_str);
if (error != noErr)
{
if (log)
log->Printf("Failed to create AEDesc for Xcode AppleEvent: %ld.\n", error);
return false;
}
OSAID ret_OSAID;
error = ::OSACompileExecute (osa_component,
&as_desc,
kOSANullScript,
kOSAModeNeverInteract,
&ret_OSAID);
::OSADispose (osa_component, ret_OSAID);
::AEDisposeDesc (&as_desc);
if (error != noErr)
{
if (log)
log->Printf("Sending AppleEvent to Xcode failed, error: %ld.\n", error);
return false;
}
}
return true;
#endif // #if !defined(__arm__)
}
Fixed a race condition that could cause ProcessGDBRemote::DoResume() to return an error saying the resume timed out. Previously the thread that was trying to resume the process would eventually call ProcessGDBRemote::DoResume() which would broadcast an event over to the async GDB remote thread which would sent the continue packet to the remote gdb server. Right after this was sent, it would set a predicate boolean value (protected by a mutex and condition) and then the thread that issued the ProcessGDBRemote::DoResume() would then wait for that condition variable to be set. If the async gdb thread was too quick though, the predicate boolean value could have been set to true and back to false by the time the thread that issued the ProcessGDBRemote::DoResume() checks the boolean value. So we can't use the predicate value as a handshake. I have changed the code over to using a Event by having the GDB remote communication object post an event: GDBRemoteCommunication::eBroadcastBitRunPacketSent This allows reliable handshaking between the two threads and avoids the erroneous ProcessGDBRemote::DoResume() errors. Added a host backtrace service to allow in process backtraces when trying to track down tricky issues. I need to see if LLVM has any backtracing abilities abstracted in it already, and if so, use that, but I needed something ASAP for the current issue I was working on. The static function is: void Host::Backtrace (Stream &strm, uint32_t max_frames); And it will backtrace at most "max_frames" frames for the current thread and can be used with any of the Stream subclasses for logging. llvm-svn: 120793
2010-12-03 14:02:24 +08:00
void
Host::Backtrace (Stream &strm, uint32_t max_frames)
{
char backtrace_path[] = "/tmp/lldb-backtrace-tmp-XXXXXX";
int backtrace_fd = ::mkstemp (backtrace_path);
if (backtrace_fd != -1)
{
std::vector<void *> frame_buffer (max_frames, NULL);
int count = ::backtrace (&frame_buffer[0], frame_buffer.size());
::backtrace_symbols_fd (&frame_buffer[0], count, backtrace_fd);
const off_t buffer_size = ::lseek(backtrace_fd, 0, SEEK_CUR);
if (::lseek(backtrace_fd, 0, SEEK_SET) == 0)
{
char *buffer = (char *)::malloc (buffer_size);
if (buffer)
{
ssize_t bytes_read = ::read (backtrace_fd, buffer, buffer_size);
if (bytes_read > 0)
strm.Write(buffer, bytes_read);
::free (buffer);
}
}
::close (backtrace_fd);
::unlink (backtrace_path);
}
}
size_t
Host::GetEnvironment (StringList &env)
{
char **host_env = *_NSGetEnviron();
char *env_entry;
size_t i;
for (i=0; (env_entry = host_env[i]) != NULL; ++i)
env.AppendString(env_entry);
return i;
}
Added more platform support. There are now some new commands: platform status -- gets status information for the selected platform platform create <platform-name> -- creates a new instance of a remote platform platform list -- list all available platforms platform select -- select a platform instance as the current platform (not working yet) When using "platform create" it will create a remote platform and make it the selected platform. For instances for iPhone OS debugging on Mac OS X one can do: (lldb) platform create remote-ios --sdk-version=4.0 Remote platform: iOS platform SDK version: 4.0 SDK path: "/Developer/Platforms/iPhoneOS.platform/DeviceSupport/4.0" Not connected to a remote device. (lldb) file ~/Documents/a.out Current executable set to '~/Documents/a.out' (armv6). (lldb) image list [ 0] /Volumes/work/gclayton/Documents/devb/attach/a.out [ 1] /Developer/Platforms/iPhoneOS.platform/DeviceSupport/4.0/Symbols/usr/lib/dyld [ 2] /Developer/Platforms/iPhoneOS.platform/DeviceSupport/4.0/Symbols/usr/lib/libSystem.B.dylib Note that this is all happening prior to running _or_ connecting to a remote platform. Once connected to a remote platform the OS version might change which means we will need to update our dependecies. Also once we run, we will need to match up the actualy binaries with the actualy UUID's to files in the SDK, or download and cache them locally. This is just the start of the remote platforms, but this modification is the first iteration in getting the platforms really doing something. llvm-svn: 127934
2011-03-19 09:12:21 +08:00
bool
Host::GetOSBuildString (std::string &s)
{
int mib[2] = { CTL_KERN, KERN_OSVERSION };
char cstr[PATH_MAX];
size_t cstr_len = sizeof(cstr);
if (::sysctl (mib, 2, cstr, &cstr_len, NULL, 0) == 0)
{
s.assign (cstr, cstr_len);
return true;
}
s.clear();
return false;
}
bool
Host::GetOSKernelDescription (std::string &s)
{
int mib[2] = { CTL_KERN, KERN_VERSION };
char cstr[PATH_MAX];
size_t cstr_len = sizeof(cstr);
if (::sysctl (mib, 2, cstr, &cstr_len, NULL, 0) == 0)
{
s.assign (cstr, cstr_len);
return true;
}
s.clear();
return false;
}
Added more platform support. There are now some new commands: platform status -- gets status information for the selected platform platform create <platform-name> -- creates a new instance of a remote platform platform list -- list all available platforms platform select -- select a platform instance as the current platform (not working yet) When using "platform create" it will create a remote platform and make it the selected platform. For instances for iPhone OS debugging on Mac OS X one can do: (lldb) platform create remote-ios --sdk-version=4.0 Remote platform: iOS platform SDK version: 4.0 SDK path: "/Developer/Platforms/iPhoneOS.platform/DeviceSupport/4.0" Not connected to a remote device. (lldb) file ~/Documents/a.out Current executable set to '~/Documents/a.out' (armv6). (lldb) image list [ 0] /Volumes/work/gclayton/Documents/devb/attach/a.out [ 1] /Developer/Platforms/iPhoneOS.platform/DeviceSupport/4.0/Symbols/usr/lib/dyld [ 2] /Developer/Platforms/iPhoneOS.platform/DeviceSupport/4.0/Symbols/usr/lib/libSystem.B.dylib Note that this is all happening prior to running _or_ connecting to a remote platform. Once connected to a remote platform the OS version might change which means we will need to update our dependecies. Also once we run, we will need to match up the actualy binaries with the actualy UUID's to files in the SDK, or download and cache them locally. This is just the start of the remote platforms, but this modification is the first iteration in getting the platforms really doing something. llvm-svn: 127934
2011-03-19 09:12:21 +08:00
bool
Host::GetOSVersion
(
uint32_t &major,
uint32_t &minor,
uint32_t &update
)
{
#if defined (__arm__)
major = UINT32_MAX;
minor = UINT32_MAX;
update = UINT32_MAX;
NSString *system_version_nstr = [[UIDevice currentDevice] systemVersion];
if (system_version_nstr)
{
const char *system_version_cstr = system_version_nstr.UTF8String;
Args::StringToVersion(system_version_cstr, major, minor, update);
}
return major != UINT32_MAX;
#else
Added more platform support. There are now some new commands: platform status -- gets status information for the selected platform platform create <platform-name> -- creates a new instance of a remote platform platform list -- list all available platforms platform select -- select a platform instance as the current platform (not working yet) When using "platform create" it will create a remote platform and make it the selected platform. For instances for iPhone OS debugging on Mac OS X one can do: (lldb) platform create remote-ios --sdk-version=4.0 Remote platform: iOS platform SDK version: 4.0 SDK path: "/Developer/Platforms/iPhoneOS.platform/DeviceSupport/4.0" Not connected to a remote device. (lldb) file ~/Documents/a.out Current executable set to '~/Documents/a.out' (armv6). (lldb) image list [ 0] /Volumes/work/gclayton/Documents/devb/attach/a.out [ 1] /Developer/Platforms/iPhoneOS.platform/DeviceSupport/4.0/Symbols/usr/lib/dyld [ 2] /Developer/Platforms/iPhoneOS.platform/DeviceSupport/4.0/Symbols/usr/lib/libSystem.B.dylib Note that this is all happening prior to running _or_ connecting to a remote platform. Once connected to a remote platform the OS version might change which means we will need to update our dependecies. Also once we run, we will need to match up the actualy binaries with the actualy UUID's to files in the SDK, or download and cache them locally. This is just the start of the remote platforms, but this modification is the first iteration in getting the platforms really doing something. llvm-svn: 127934
2011-03-19 09:12:21 +08:00
SInt32 version;
OSErr err = ::Gestalt (gestaltSystemVersion, &version);
if (err != noErr)
return false;
if (version < 0x1040)
{
major = ((version & 0xF000) >> 12) * 10 + ((version & 0x0F00) >> 8);
minor = (version & 0x00F0) >> 4;
update = (version & 0x000F);
}
else
{
if (::Gestalt (gestaltSystemVersionMajor, &version) != noErr)
return false;
major = version;
if (::Gestalt (gestaltSystemVersionMinor, &version) == noErr)
minor = version;
else
minor = 0;
if (::Gestalt (gestaltSystemVersionBugFix, &version) == noErr)
update = version;
else
update = 0;
}
return true;
#endif
Added more platform support. There are now some new commands: platform status -- gets status information for the selected platform platform create <platform-name> -- creates a new instance of a remote platform platform list -- list all available platforms platform select -- select a platform instance as the current platform (not working yet) When using "platform create" it will create a remote platform and make it the selected platform. For instances for iPhone OS debugging on Mac OS X one can do: (lldb) platform create remote-ios --sdk-version=4.0 Remote platform: iOS platform SDK version: 4.0 SDK path: "/Developer/Platforms/iPhoneOS.platform/DeviceSupport/4.0" Not connected to a remote device. (lldb) file ~/Documents/a.out Current executable set to '~/Documents/a.out' (armv6). (lldb) image list [ 0] /Volumes/work/gclayton/Documents/devb/attach/a.out [ 1] /Developer/Platforms/iPhoneOS.platform/DeviceSupport/4.0/Symbols/usr/lib/dyld [ 2] /Developer/Platforms/iPhoneOS.platform/DeviceSupport/4.0/Symbols/usr/lib/libSystem.B.dylib Note that this is all happening prior to running _or_ connecting to a remote platform. Once connected to a remote platform the OS version might change which means we will need to update our dependecies. Also once we run, we will need to match up the actualy binaries with the actualy UUID's to files in the SDK, or download and cache them locally. This is just the start of the remote platforms, but this modification is the first iteration in getting the platforms really doing something. llvm-svn: 127934
2011-03-19 09:12:21 +08:00
}
static bool
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
GetMacOSXProcessName (const ProcessInstanceInfoMatch *match_info_ptr,
ProcessInstanceInfo &process_info)
{
if (process_info.ProcessIDIsValid())
{
char process_name[MAXCOMLEN * 2 + 1];
int name_len = ::proc_name(process_info.GetProcessID(), process_name, MAXCOMLEN * 2);
if (name_len == 0)
return false;
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 (match_info_ptr == NULL || NameMatches(process_name,
match_info_ptr->GetNameMatchType(),
match_info_ptr->GetProcessInfo().GetName()))
{
process_info.GetExecutableFile().SetFile (process_name, false);
return true;
}
}
process_info.GetExecutableFile().Clear();
return false;
}
static bool
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
GetMacOSXProcessCPUType (ProcessInstanceInfo &process_info)
{
if (process_info.ProcessIDIsValid())
{
// Make a new mib to stay thread safe
int mib[CTL_MAXNAME]={0,};
size_t mib_len = CTL_MAXNAME;
if (::sysctlnametomib("sysctl.proc_cputype", mib, &mib_len))
return false;
mib[mib_len] = process_info.GetProcessID();
mib_len++;
cpu_type_t cpu, sub = 0;
size_t cpu_len = sizeof(cpu);
if (::sysctl (mib, mib_len, &cpu, &cpu_len, 0, 0) == 0)
{
switch (cpu)
{
case llvm::MachO::CPUTypeI386: sub = llvm::MachO::CPUSubType_I386_ALL; break;
case llvm::MachO::CPUTypeX86_64: sub = llvm::MachO::CPUSubType_X86_64_ALL; break;
default: break;
}
process_info.GetArchitecture ().SetArchitecture (eArchTypeMachO, cpu, sub);
return true;
}
}
process_info.GetArchitecture().Clear();
return false;
}
static bool
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
GetMacOSXProcessArgs (const ProcessInstanceInfoMatch *match_info_ptr,
ProcessInstanceInfo &process_info)
{
if (process_info.ProcessIDIsValid())
{
int proc_args_mib[3] = { CTL_KERN, KERN_PROCARGS2, process_info.GetProcessID() };
char arg_data[8192];
size_t arg_data_size = sizeof(arg_data);
if (::sysctl (proc_args_mib, 3, arg_data, &arg_data_size , NULL, 0) == 0)
{
DataExtractor data (arg_data, arg_data_size, lldb::endian::InlHostByteOrder(), sizeof(void *));
uint32_t offset = 0;
uint32_t start_offset;
uint32_t argc = data.GetU32 (&offset);
const char *cstr;
cstr = data.GetCStr (&offset);
if (cstr)
{
process_info.GetExecutableFile().SetFile(cstr, false);
if (match_info_ptr == NULL ||
NameMatches (process_info.GetExecutableFile().GetFilename().GetCString(),
match_info_ptr->GetNameMatchType(),
match_info_ptr->GetProcessInfo().GetName()))
{
// Skip NULLs
while (1)
{
const uint8_t *p = data.PeekData(offset, 1);
if ((p == NULL) || (*p != '\0'))
break;
++offset;
}
// Now extract all arguments
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
Args &proc_args = process_info.GetArguments();
for (int i=0; i<argc; ++i)
{
start_offset = offset;
cstr = data.GetCStr(&offset);
if (cstr)
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
proc_args.AppendArgument(cstr);
}
return true;
}
}
}
}
return false;
}
static bool
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
GetMacOSXProcessUserAndGroup (ProcessInstanceInfo &process_info)
{
if (process_info.ProcessIDIsValid())
{
int mib[4];
mib[0] = CTL_KERN;
mib[1] = KERN_PROC;
mib[2] = KERN_PROC_PID;
mib[3] = process_info.GetProcessID();
struct kinfo_proc proc_kinfo;
size_t proc_kinfo_size = sizeof(struct kinfo_proc);
if (::sysctl (mib, 4, &proc_kinfo, &proc_kinfo_size, NULL, 0) == 0)
{
if (proc_kinfo_size > 0)
{
process_info.SetParentProcessID (proc_kinfo.kp_eproc.e_ppid);
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
process_info.SetUserID (proc_kinfo.kp_eproc.e_pcred.p_ruid);
process_info.SetGroupID (proc_kinfo.kp_eproc.e_pcred.p_rgid);
process_info.SetEffectiveUserID (proc_kinfo.kp_eproc.e_ucred.cr_uid);
if (proc_kinfo.kp_eproc.e_ucred.cr_ngroups > 0)
process_info.SetEffectiveGroupID (proc_kinfo.kp_eproc.e_ucred.cr_groups[0]);
else
process_info.SetEffectiveGroupID (UINT32_MAX);
return true;
}
}
}
process_info.SetParentProcessID (LLDB_INVALID_PROCESS_ID);
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
process_info.SetUserID (UINT32_MAX);
process_info.SetGroupID (UINT32_MAX);
process_info.SetEffectiveUserID (UINT32_MAX);
process_info.SetEffectiveGroupID (UINT32_MAX);
return false;
}
uint32_t
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
Host::FindProcesses (const ProcessInstanceInfoMatch &match_info, ProcessInstanceInfoList &process_infos)
{
std::vector<struct kinfo_proc> kinfos;
int mib[3] = { CTL_KERN, KERN_PROC, KERN_PROC_ALL };
size_t pid_data_size = 0;
if (::sysctl (mib, 4, NULL, &pid_data_size, NULL, 0) != 0)
return 0;
// Add a few extra in case a few more show up
const size_t estimated_pid_count = (pid_data_size / sizeof(struct kinfo_proc)) + 10;
kinfos.resize (estimated_pid_count);
pid_data_size = kinfos.size() * sizeof(struct kinfo_proc);
if (::sysctl (mib, 4, &kinfos[0], &pid_data_size, NULL, 0) != 0)
return 0;
const size_t actual_pid_count = (pid_data_size / sizeof(struct kinfo_proc));
bool all_users = match_info.GetMatchAllUsers();
const lldb::pid_t our_pid = getpid();
const uid_t our_uid = getuid();
for (int i = 0; i < actual_pid_count; i++)
{
const struct kinfo_proc &kinfo = kinfos[i];
bool kinfo_user_matches = false;
if (all_users)
kinfo_user_matches = true;
else
kinfo_user_matches = kinfo.kp_eproc.e_pcred.p_ruid == our_uid;
// Special case, if lldb is being run as root we can attach to anything.
if (our_uid == 0)
kinfo_user_matches = true;
if (kinfo_user_matches == false || // Make sure the user is acceptable
kinfo.kp_proc.p_pid == our_pid || // Skip this process
kinfo.kp_proc.p_pid == 0 || // Skip kernel (kernel pid is zero)
kinfo.kp_proc.p_stat == SZOMB || // Zombies are bad, they like brains...
kinfo.kp_proc.p_flag & P_TRACED || // Being debugged?
kinfo.kp_proc.p_flag & P_WEXIT || // Working on exiting?
kinfo.kp_proc.p_flag & P_TRANSLATED) // Skip translated ppc (Rosetta)
continue;
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
ProcessInstanceInfo process_info;
process_info.SetProcessID (kinfo.kp_proc.p_pid);
process_info.SetParentProcessID (kinfo.kp_eproc.e_ppid);
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
process_info.SetUserID (kinfo.kp_eproc.e_pcred.p_ruid);
process_info.SetGroupID (kinfo.kp_eproc.e_pcred.p_rgid);
process_info.SetEffectiveUserID (kinfo.kp_eproc.e_ucred.cr_uid);
if (kinfo.kp_eproc.e_ucred.cr_ngroups > 0)
process_info.SetEffectiveGroupID (kinfo.kp_eproc.e_ucred.cr_groups[0]);
else
process_info.SetEffectiveGroupID (UINT32_MAX);
// Make sure our info matches before we go fetch the name and cpu type
if (match_info.Matches (process_info))
{
if (GetMacOSXProcessArgs (&match_info, process_info))
{
GetMacOSXProcessCPUType (process_info);
if (match_info.Matches (process_info))
process_infos.Append (process_info);
}
}
}
return process_infos.GetSize();
}
bool
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
Host::GetProcessInfo (lldb::pid_t pid, ProcessInstanceInfo &process_info)
{
process_info.SetProcessID(pid);
if (GetMacOSXProcessArgs (NULL, process_info))
{
GetMacOSXProcessCPUType (process_info);
GetMacOSXProcessUserAndGroup (process_info);
return true;
}
process_info.Clear();
return false;
}
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
Host::LaunchProcess (ProcessLaunchInfo &launch_info)
{
Error error;
LogSP log(lldb_private::GetLogIfAllCategoriesSet (LIBLLDB_LOG_HOST | LIBLLDB_LOG_PROCESS));
char exe_path[PATH_MAX];
PlatformSP host_platform_sp (Platform::GetDefaultPlatform ());
const ArchSpec &arch_spec = launch_info.GetArchitecture();
FileSpec exe_spec(launch_info.GetExecutableFile());
FileSpec::FileType file_type = exe_spec.GetFileType();
if (file_type != FileSpec::eFileTypeRegular)
{
lldb::ModuleSP exe_module_sp;
error = host_platform_sp->ResolveExecutable (exe_spec,
arch_spec,
exe_module_sp);
if (error.Fail())
return error;
if (exe_module_sp)
exe_spec = exe_module_sp->GetFileSpec();
}
if (exe_spec.Exists())
{
exe_spec.GetPath (exe_path, sizeof(exe_path));
}
else
{
launch_info.GetExecutableFile().GetPath (exe_path, sizeof(exe_path));
error.SetErrorStringWithFormat ("executable doesn't exist: '%s'", exe_path);
return error;
}
if (launch_info.GetFlags().Test (eLaunchFlagLaunchInTTY))
{
#if !defined(__arm__)
return LaunchInNewTerminalWithAppleScript (exe_path, launch_info);
#else
error.SetErrorString ("launching a processs in a new terminal is not supported on iOS devices");
return error;
#endif
}
Error local_err; // Errors that don't affect the spawning.
posix_spawnattr_t attr;
error.SetError( ::posix_spawnattr_init (&attr), eErrorTypePOSIX);
if (error.Fail() || log)
error.PutToLog(log.get(), "::posix_spawnattr_init ( &attr )");
if (error.Fail())
return error;
// Make a quick class that will cleanup the posix spawn attributes in case
// we return in the middle of this function.
lldb_utility::CleanUp <posix_spawnattr_t *, int> posix_spawnattr_cleanup(&attr, posix_spawnattr_destroy);
sigset_t no_signals;
sigset_t all_signals;
sigemptyset (&no_signals);
sigfillset (&all_signals);
::posix_spawnattr_setsigmask(&attr, &no_signals);
::posix_spawnattr_setsigdefault(&attr, &all_signals);
short flags = POSIX_SPAWN_SETSIGDEF | POSIX_SPAWN_SETSIGMASK;
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 (launch_info.GetFlags().Test (eLaunchFlagExec))
flags |= POSIX_SPAWN_SETEXEC; // Darwin specific posix_spawn flag
if (launch_info.GetFlags().Test (eLaunchFlagDebug))
flags |= POSIX_SPAWN_START_SUSPENDED; // Darwin specific posix_spawn flag
if (launch_info.GetFlags().Test (eLaunchFlagDisableASLR))
flags |= _POSIX_SPAWN_DISABLE_ASLR; // Darwin specific posix_spawn flag
//#ifdef POSIX_SPAWN_CLOEXEC_DEFAULT
// // Close all files exception those with file actions if this is supported.
// flags |= POSIX_SPAWN_CLOEXEC_DEFAULT;
//#endif
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.SetError( ::posix_spawnattr_setflags (&attr, flags), eErrorTypePOSIX);
if (error.Fail() || log)
error.PutToLog(log.get(), "::posix_spawnattr_setflags ( &attr, flags=0x%8.8x )", flags);
if (error.Fail())
return error;
#if !defined(__arm__)
// We don't need to do this for ARM, and we really shouldn't now that we
// have multiple CPU subtypes and no posix_spawnattr call that allows us
// to set which CPU subtype to launch...
cpu_type_t cpu = arch_spec.GetMachOCPUType();
if (cpu != 0 &&
cpu != UINT32_MAX &&
cpu != LLDB_INVALID_CPUTYPE)
{
size_t ocount = 0;
error.SetError( ::posix_spawnattr_setbinpref_np (&attr, 1, &cpu, &ocount), eErrorTypePOSIX);
if (error.Fail() || log)
error.PutToLog(log.get(), "::posix_spawnattr_setbinpref_np ( &attr, 1, cpu_type = 0x%8.8x, count => %zu )", cpu, ocount);
if (error.Fail() || ocount != 1)
return error;
}
#endif
::pid_t pid = LLDB_INVALID_PROCESS_ID;
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 *tmp_argv[2];
char * const *argv = (char * const*)launch_info.GetArguments().GetConstArgumentVector();
char * const *envp = (char * const*)launch_info.GetEnvironmentEntries().GetConstArgumentVector();
if (argv == NULL)
{
// posix_spawn gets very unhappy if it doesn't have at least the program
// name in argv[0]. One of the side affects I have noticed is the environment
// variables don't make it into the child process if "argv == NULL"!!!
tmp_argv[0] = exe_path;
tmp_argv[1] = NULL;
argv = (char * const*)tmp_argv;
}
const char *working_dir = launch_info.GetWorkingDirectory();
if (working_dir)
{
// No more thread specific current working directory
__pthread_chdir (working_dir);
}
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 size_t num_file_actions = launch_info.GetNumFileActions ();
if (num_file_actions > 0)
{
posix_spawn_file_actions_t file_actions;
error.SetError( ::posix_spawn_file_actions_init (&file_actions), eErrorTypePOSIX);
if (error.Fail() || log)
error.PutToLog(log.get(), "::posix_spawn_file_actions_init ( &file_actions )");
if (error.Fail())
return error;
// Make a quick class that will cleanup the posix spawn attributes in case
// we return in the middle of this function.
lldb_utility::CleanUp <posix_spawn_file_actions_t *, int> posix_spawn_file_actions_cleanup (&file_actions, posix_spawn_file_actions_destroy);
for (size_t i=0; i<num_file_actions; ++i)
{
const ProcessLaunchInfo::FileAction *launch_file_action = launch_info.GetFileActionAtIndex(i);
if (launch_file_action)
{
if (!ProcessLaunchInfo::FileAction::AddPosixSpawnFileAction (&file_actions,
launch_file_action,
log.get(),
error))
return error;
}
}
error.SetError (::posix_spawnp (&pid,
exe_path,
&file_actions,
&attr,
argv,
envp),
eErrorTypePOSIX);
if (error.Fail() || log)
error.PutToLog(log.get(), "::posix_spawnp ( pid => %i, path = '%s', file_actions = %p, attr = %p, argv = %p, envp = %p )",
pid,
exe_path,
&file_actions,
&attr,
argv,
envp);
}
else
{
error.SetError (::posix_spawnp (&pid,
exe_path,
NULL,
&attr,
argv,
envp),
eErrorTypePOSIX);
if (error.Fail() || log)
error.PutToLog(log.get(), "::posix_spawnp ( pid => %i, path = '%s', file_actions = NULL, attr = %p, argv = %p, envp = %p )",
pid,
exe_path,
&attr,
argv,
envp);
}
if (working_dir)
{
// No more thread specific current working directory
__pthread_fchdir (-1);
}
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 (pid != LLDB_INVALID_PROCESS_ID)
{
// If all went well, then set the process ID into the launch info
launch_info.SetProcessID(pid);
// Make sure we reap any processes we spawn or we will have zombies.
if (!launch_info.MonitorProcess())
{
const bool monitor_signals = false;
StartMonitoringChildProcess (Process::SetProcessExitStatus,
NULL,
pid,
monitor_signals);
}
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
}
else
{
// Invalid process ID, something didn't go well
if (error.Success())
error.SetErrorString ("process launch failed for unknown reasons");
}
return error;
}
lldb::thread_t
Host::StartMonitoringChildProcess (Host::MonitorChildProcessCallback callback,
void *callback_baton,
lldb::pid_t pid,
bool monitor_signals)
{
lldb::thread_t thread = LLDB_INVALID_HOST_THREAD;
unsigned long mask = DISPATCH_PROC_EXIT;
if (monitor_signals)
mask |= DISPATCH_PROC_SIGNAL;
LogSP log(lldb_private::GetLogIfAnyCategoriesSet (LIBLLDB_LOG_HOST | LIBLLDB_LOG_PROCESS));
dispatch_source_t source = ::dispatch_source_create (DISPATCH_SOURCE_TYPE_PROC,
pid,
mask,
::dispatch_get_global_queue (DISPATCH_QUEUE_PRIORITY_DEFAULT,0));
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 (log)
log->Printf ("Host::StartMonitoringChildProcess (callback=%p, baton=%p, pid=%i, monitor_signals=%i) source = %p\n",
callback,
callback_baton,
(int)pid,
monitor_signals,
source);
if (source)
{
::dispatch_source_set_cancel_handler (source, ^{
::dispatch_release (source);
});
::dispatch_source_set_event_handler (source, ^{
int status= 0;
int wait_pid = 0;
bool cancel = false;
bool exited = false;
do
{
wait_pid = ::waitpid (pid, &status, 0);
} while (wait_pid < 0 && errno == EINTR);
if (wait_pid >= 0)
{
int signal = 0;
int exit_status = 0;
const char *status_cstr = NULL;
if (WIFSTOPPED(status))
{
signal = WSTOPSIG(status);
status_cstr = "STOPPED";
}
else if (WIFEXITED(status))
{
exit_status = WEXITSTATUS(status);
status_cstr = "EXITED";
exited = true;
}
else if (WIFSIGNALED(status))
{
signal = WTERMSIG(status);
status_cstr = "SIGNALED";
exited = true;
exit_status = -1;
}
else
{
status_cstr = "???";
}
if (log)
log->Printf ("::waitpid (pid = %llu, &status, 0) => pid = %i, status = 0x%8.8x (%s), signal = %i, exit_status = %i",
pid,
wait_pid,
status,
status_cstr,
signal,
exit_status);
if (callback)
cancel = callback (callback_baton, pid, exited, signal, exit_status);
if (exited)
{
::dispatch_source_cancel(source);
}
}
});
::dispatch_resume (source);
}
return thread;
}