llvm-project/lldb/source/Core/UserSettingsController.cpp

2490 lines
88 KiB
C++
Raw Normal View History

//====-- UserSettingsController.cpp ------------------------------*- C++ -*-===//
//
// The LLVM Compiler Infrastructure
//
// This file is distributed under the University of Illinois Open Source
// License. See LICENSE.TXT for details.
//
//===----------------------------------------------------------------------===//
#include <string.h>
#include <algorithm>
#include "lldb/Core/UserSettingsController.h"
#include "lldb/Core/Error.h"
#include "lldb/Core/RegularExpression.h"
#include "lldb/Core/Stream.h"
#include "lldb/Core/StreamString.h"
#include "lldb/Interpreter/CommandInterpreter.h"
Added new target instance settings for execution settings: Targets can now specify some additional parameters for when we debug executables that can help with plug-in selection: target.execution-level = auto | user | kernel target.execution-mode = auto | dynamic | static target.execution-os-type = auto | none | halted | live On some systems, the binaries that are created are the same wether you use them to debug a kernel, or a user space program. Many times inspecting an object file can reveal what an executable should be. For these cases we can now be a little more complete by specifying wether to detect all of these things automatically (inspect the main executable file and select a plug-in accordingly), or manually to force the selection of certain plug-ins. To do this we now allow the specficifation of wether one is debugging a user space program (target.execution-level = user) or a kernel program (target.execution-level = kernel). We can also specify if we want to debug a program where shared libraries are dynamically loaded using a DynamicLoader plug-in (target.execution-mode = dynamic), or wether we will treat all symbol files as already linked at the correct address (target.execution-mode = static). We can also specify if the inferior we are debugging is being debugged on a bare board (target.execution-os-type = none), or debugging an OS where we have a JTAG or other direct connection to the inferior stops the entire OS (target.execution-os-type = halted), or if we are debugging a program on something that has live debug services (target.execution-os-type = live). For the "target.execution-os-type = halted" mode, we will need to create ProcessHelper plug-ins that allow us to extract the process/thread and other OS information by reading/writing memory. This should allow LLDB to be used for a wide variety of debugging tasks and handle them all correctly. llvm-svn: 125815
2011-02-18 09:44:25 +08:00
using namespace lldb;
using namespace lldb_private;
Added new target instance settings for execution settings: Targets can now specify some additional parameters for when we debug executables that can help with plug-in selection: target.execution-level = auto | user | kernel target.execution-mode = auto | dynamic | static target.execution-os-type = auto | none | halted | live On some systems, the binaries that are created are the same wether you use them to debug a kernel, or a user space program. Many times inspecting an object file can reveal what an executable should be. For these cases we can now be a little more complete by specifying wether to detect all of these things automatically (inspect the main executable file and select a plug-in accordingly), or manually to force the selection of certain plug-ins. To do this we now allow the specficifation of wether one is debugging a user space program (target.execution-level = user) or a kernel program (target.execution-level = kernel). We can also specify if we want to debug a program where shared libraries are dynamically loaded using a DynamicLoader plug-in (target.execution-mode = dynamic), or wether we will treat all symbol files as already linked at the correct address (target.execution-mode = static). We can also specify if the inferior we are debugging is being debugged on a bare board (target.execution-os-type = none), or debugging an OS where we have a JTAG or other direct connection to the inferior stops the entire OS (target.execution-os-type = halted), or if we are debugging a program on something that has live debug services (target.execution-os-type = live). For the "target.execution-os-type = halted" mode, we will need to create ProcessHelper plug-ins that allow us to extract the process/thread and other OS information by reading/writing memory. This should allow LLDB to be used for a wide variety of debugging tasks and handle them all correctly. llvm-svn: 125815
2011-02-18 09:44:25 +08:00
static void
DumpSettingEntry (CommandInterpreter &interpreter,
Stream &strm,
Added new target instance settings for execution settings: Targets can now specify some additional parameters for when we debug executables that can help with plug-in selection: target.execution-level = auto | user | kernel target.execution-mode = auto | dynamic | static target.execution-os-type = auto | none | halted | live On some systems, the binaries that are created are the same wether you use them to debug a kernel, or a user space program. Many times inspecting an object file can reveal what an executable should be. For these cases we can now be a little more complete by specifying wether to detect all of these things automatically (inspect the main executable file and select a plug-in accordingly), or manually to force the selection of certain plug-ins. To do this we now allow the specficifation of wether one is debugging a user space program (target.execution-level = user) or a kernel program (target.execution-level = kernel). We can also specify if we want to debug a program where shared libraries are dynamically loaded using a DynamicLoader plug-in (target.execution-mode = dynamic), or wether we will treat all symbol files as already linked at the correct address (target.execution-mode = static). We can also specify if the inferior we are debugging is being debugged on a bare board (target.execution-os-type = none), or debugging an OS where we have a JTAG or other direct connection to the inferior stops the entire OS (target.execution-os-type = halted), or if we are debugging a program on something that has live debug services (target.execution-os-type = live). For the "target.execution-os-type = halted" mode, we will need to create ProcessHelper plug-ins that allow us to extract the process/thread and other OS information by reading/writing memory. This should allow LLDB to be used for a wide variety of debugging tasks and handle them all correctly. llvm-svn: 125815
2011-02-18 09:44:25 +08:00
const uint32_t max_len,
const SettingEntry &entry)
{
StreamString description;
if (entry.description)
description.Printf ("%s", entry.description);
if (entry.default_value && entry.default_value[0])
description.Printf (" (default: %s)", entry.default_value);
interpreter.OutputFormattedHelpText (strm,
Added new target instance settings for execution settings: Targets can now specify some additional parameters for when we debug executables that can help with plug-in selection: target.execution-level = auto | user | kernel target.execution-mode = auto | dynamic | static target.execution-os-type = auto | none | halted | live On some systems, the binaries that are created are the same wether you use them to debug a kernel, or a user space program. Many times inspecting an object file can reveal what an executable should be. For these cases we can now be a little more complete by specifying wether to detect all of these things automatically (inspect the main executable file and select a plug-in accordingly), or manually to force the selection of certain plug-ins. To do this we now allow the specficifation of wether one is debugging a user space program (target.execution-level = user) or a kernel program (target.execution-level = kernel). We can also specify if we want to debug a program where shared libraries are dynamically loaded using a DynamicLoader plug-in (target.execution-mode = dynamic), or wether we will treat all symbol files as already linked at the correct address (target.execution-mode = static). We can also specify if the inferior we are debugging is being debugged on a bare board (target.execution-os-type = none), or debugging an OS where we have a JTAG or other direct connection to the inferior stops the entire OS (target.execution-os-type = halted), or if we are debugging a program on something that has live debug services (target.execution-os-type = live). For the "target.execution-os-type = halted" mode, we will need to create ProcessHelper plug-ins that allow us to extract the process/thread and other OS information by reading/writing memory. This should allow LLDB to be used for a wide variety of debugging tasks and handle them all correctly. llvm-svn: 125815
2011-02-18 09:44:25 +08:00
entry.var_name,
"--",
description.GetData(),
max_len);
if (entry.enum_values && entry.enum_values[0].string_value)
{
interpreter.OutputFormattedHelpText (strm,
Added new target instance settings for execution settings: Targets can now specify some additional parameters for when we debug executables that can help with plug-in selection: target.execution-level = auto | user | kernel target.execution-mode = auto | dynamic | static target.execution-os-type = auto | none | halted | live On some systems, the binaries that are created are the same wether you use them to debug a kernel, or a user space program. Many times inspecting an object file can reveal what an executable should be. For these cases we can now be a little more complete by specifying wether to detect all of these things automatically (inspect the main executable file and select a plug-in accordingly), or manually to force the selection of certain plug-ins. To do this we now allow the specficifation of wether one is debugging a user space program (target.execution-level = user) or a kernel program (target.execution-level = kernel). We can also specify if we want to debug a program where shared libraries are dynamically loaded using a DynamicLoader plug-in (target.execution-mode = dynamic), or wether we will treat all symbol files as already linked at the correct address (target.execution-mode = static). We can also specify if the inferior we are debugging is being debugged on a bare board (target.execution-os-type = none), or debugging an OS where we have a JTAG or other direct connection to the inferior stops the entire OS (target.execution-os-type = halted), or if we are debugging a program on something that has live debug services (target.execution-os-type = live). For the "target.execution-os-type = halted" mode, we will need to create ProcessHelper plug-ins that allow us to extract the process/thread and other OS information by reading/writing memory. This should allow LLDB to be used for a wide variety of debugging tasks and handle them all correctly. llvm-svn: 125815
2011-02-18 09:44:25 +08:00
"",
" ",
"Enumeration values:",
max_len);
for (uint32_t enum_idx=0; entry.enum_values[enum_idx].string_value != NULL; ++enum_idx)
{
description.Clear();
if (entry.enum_values[enum_idx].usage)
description.Printf ("%s = %s",
entry.enum_values[enum_idx].string_value,
entry.enum_values[enum_idx].usage);
else
description.Printf ("%s", entry.enum_values[enum_idx].string_value);
interpreter.OutputFormattedHelpText (strm,
Added new target instance settings for execution settings: Targets can now specify some additional parameters for when we debug executables that can help with plug-in selection: target.execution-level = auto | user | kernel target.execution-mode = auto | dynamic | static target.execution-os-type = auto | none | halted | live On some systems, the binaries that are created are the same wether you use them to debug a kernel, or a user space program. Many times inspecting an object file can reveal what an executable should be. For these cases we can now be a little more complete by specifying wether to detect all of these things automatically (inspect the main executable file and select a plug-in accordingly), or manually to force the selection of certain plug-ins. To do this we now allow the specficifation of wether one is debugging a user space program (target.execution-level = user) or a kernel program (target.execution-level = kernel). We can also specify if we want to debug a program where shared libraries are dynamically loaded using a DynamicLoader plug-in (target.execution-mode = dynamic), or wether we will treat all symbol files as already linked at the correct address (target.execution-mode = static). We can also specify if the inferior we are debugging is being debugged on a bare board (target.execution-os-type = none), or debugging an OS where we have a JTAG or other direct connection to the inferior stops the entire OS (target.execution-os-type = halted), or if we are debugging a program on something that has live debug services (target.execution-os-type = live). For the "target.execution-os-type = halted" mode, we will need to create ProcessHelper plug-ins that allow us to extract the process/thread and other OS information by reading/writing memory. This should allow LLDB to be used for a wide variety of debugging tasks and handle them all correctly. llvm-svn: 125815
2011-02-18 09:44:25 +08:00
"",
" ",
description.GetData(),
max_len);
}
}
}
UserSettingsController::UserSettingsController (const char *level_name,
Added new target instance settings for execution settings: Targets can now specify some additional parameters for when we debug executables that can help with plug-in selection: target.execution-level = auto | user | kernel target.execution-mode = auto | dynamic | static target.execution-os-type = auto | none | halted | live On some systems, the binaries that are created are the same wether you use them to debug a kernel, or a user space program. Many times inspecting an object file can reveal what an executable should be. For these cases we can now be a little more complete by specifying wether to detect all of these things automatically (inspect the main executable file and select a plug-in accordingly), or manually to force the selection of certain plug-ins. To do this we now allow the specficifation of wether one is debugging a user space program (target.execution-level = user) or a kernel program (target.execution-level = kernel). We can also specify if we want to debug a program where shared libraries are dynamically loaded using a DynamicLoader plug-in (target.execution-mode = dynamic), or wether we will treat all symbol files as already linked at the correct address (target.execution-mode = static). We can also specify if the inferior we are debugging is being debugged on a bare board (target.execution-os-type = none), or debugging an OS where we have a JTAG or other direct connection to the inferior stops the entire OS (target.execution-os-type = halted), or if we are debugging a program on something that has live debug services (target.execution-os-type = live). For the "target.execution-os-type = halted" mode, we will need to create ProcessHelper plug-ins that allow us to extract the process/thread and other OS information by reading/writing memory. This should allow LLDB to be used for a wide variety of debugging tasks and handle them all correctly. llvm-svn: 125815
2011-02-18 09:44:25 +08:00
const UserSettingsControllerSP &parent) :
m_default_settings (),
m_settings (),
m_children (),
m_pending_settings (),
m_live_settings (),
m_children_mutex (Mutex::eMutexTypeNormal),
m_pending_settings_mutex (Mutex::eMutexTypeRecursive),
m_live_settings_mutex (Mutex::eMutexTypeRecursive)
{
m_settings.parent = parent;
m_settings.level_name.SetCString (level_name);
}
UserSettingsController::~UserSettingsController ()
{
Mutex::Locker locker (m_live_settings_mutex);
m_live_settings.clear();
}
Added code that will allow completely customizable prompts for use in replacing the "(lldb)" prompt, the "frame #1..." displays when doing stack backtracing and the "thread #1....". This will allow you to see exactly the information that you want to see where you want to see it. This currently isn't hookup up to the prompts yet, but it will be soon. So what is the format of the prompts? Prompts can contain variables that have access to the current program state. Variables are text that appears in between a prefix of "${" and ends with a "}". Some of the interesting variables include: // The frame index (0, 1, 2, 3...) ${frame.index} // common frame registers with generic names ${frame.pc} ${frame.sp} ${frame.fp} ${frame.ra} ${frame.flags} // Access to any frame registers by name where REGNAME is any register name: ${frame.reg.REGNAME} // The current compile unit file where the frame is located ${file.basename} ${file.fullpath} // Function information ${function.name} ${function.pc-offset} // Process info ${process.file.basename} ${process.file.fullpath} ${process.id} ${process.name} // Thread info ${thread.id} ${thread.index} ${thread.name} ${thread.queue} ${thread.stop-reason} // Target information ${target.arch} // The current module for the current frame (the shared library or executable // that contains the current frame PC value): ${module.file.basename} ${module.file.fullpath} // Access to the line entry for where the current frame is when your thread // is stopped: ${line.file.basename} ${line.file.fullpath} ${line.number} ${line.start-addr} ${line.end-addr} Many times the information that you might have in your prompt might not be available and you won't want it to print out if it isn't valid. To take care of this you can enclose everything that must resolve into a scope. A scope is starts with '{' and ends with '}'. For example in order to only display the current file and line number when the information is available the format would be: "{ at {$line.file.basename}:${line.number}}" Broken down this is: start the scope: "{" format whose content will only be displayed if all information is available: "at {$line.file.basename}:${line.number}" end the scope: "}" We currently can represent the infomration we see when stopped at a frame: frame #0: 0x0000000100000e85 a.out`main + 4 at test.c:19 with the following format: "frame #${frame.index}: ${frame.pc} {${module.file.basename}`}{${function.name}{${function.pc-offset}}{ at ${line.file.basename}:${line.number}}\n" This breaks down to always print: "frame #${frame.index}: ${frame.pc} " only print the module followed by a tick if we have a valid module: "{${module.file.basename}`}" print the function name with optional offset: "{${function.name}{${function.pc-offset}}" print the line info if it is available: "{ at ${line.file.basename}:${line.number}}" then finish off with a newline: "\n" Notice you can also put newlines ("\n") and tabs and everything else you are used to putting in a format string when desensitized with the \ character. Cleaned up some of the user settings controller subclasses. All of them do not have any global settings variables and were all implementing stubs for the get/set global settings variable. Now there is a default version in UserSettingsController that will do nothing. llvm-svn: 114306
2010-09-19 10:33:57 +08:00
bool
UserSettingsController::SetGlobalVariable
(
const ConstString &var_name,
const char *index_value,
const char *value,
const SettingEntry &entry,
Added new target instance settings for execution settings: Targets can now specify some additional parameters for when we debug executables that can help with plug-in selection: target.execution-level = auto | user | kernel target.execution-mode = auto | dynamic | static target.execution-os-type = auto | none | halted | live On some systems, the binaries that are created are the same wether you use them to debug a kernel, or a user space program. Many times inspecting an object file can reveal what an executable should be. For these cases we can now be a little more complete by specifying wether to detect all of these things automatically (inspect the main executable file and select a plug-in accordingly), or manually to force the selection of certain plug-ins. To do this we now allow the specficifation of wether one is debugging a user space program (target.execution-level = user) or a kernel program (target.execution-level = kernel). We can also specify if we want to debug a program where shared libraries are dynamically loaded using a DynamicLoader plug-in (target.execution-mode = dynamic), or wether we will treat all symbol files as already linked at the correct address (target.execution-mode = static). We can also specify if the inferior we are debugging is being debugged on a bare board (target.execution-os-type = none), or debugging an OS where we have a JTAG or other direct connection to the inferior stops the entire OS (target.execution-os-type = halted), or if we are debugging a program on something that has live debug services (target.execution-os-type = live). For the "target.execution-os-type = halted" mode, we will need to create ProcessHelper plug-ins that allow us to extract the process/thread and other OS information by reading/writing memory. This should allow LLDB to be used for a wide variety of debugging tasks and handle them all correctly. llvm-svn: 125815
2011-02-18 09:44:25 +08:00
const VarSetOperationType op,
Added code that will allow completely customizable prompts for use in replacing the "(lldb)" prompt, the "frame #1..." displays when doing stack backtracing and the "thread #1....". This will allow you to see exactly the information that you want to see where you want to see it. This currently isn't hookup up to the prompts yet, but it will be soon. So what is the format of the prompts? Prompts can contain variables that have access to the current program state. Variables are text that appears in between a prefix of "${" and ends with a "}". Some of the interesting variables include: // The frame index (0, 1, 2, 3...) ${frame.index} // common frame registers with generic names ${frame.pc} ${frame.sp} ${frame.fp} ${frame.ra} ${frame.flags} // Access to any frame registers by name where REGNAME is any register name: ${frame.reg.REGNAME} // The current compile unit file where the frame is located ${file.basename} ${file.fullpath} // Function information ${function.name} ${function.pc-offset} // Process info ${process.file.basename} ${process.file.fullpath} ${process.id} ${process.name} // Thread info ${thread.id} ${thread.index} ${thread.name} ${thread.queue} ${thread.stop-reason} // Target information ${target.arch} // The current module for the current frame (the shared library or executable // that contains the current frame PC value): ${module.file.basename} ${module.file.fullpath} // Access to the line entry for where the current frame is when your thread // is stopped: ${line.file.basename} ${line.file.fullpath} ${line.number} ${line.start-addr} ${line.end-addr} Many times the information that you might have in your prompt might not be available and you won't want it to print out if it isn't valid. To take care of this you can enclose everything that must resolve into a scope. A scope is starts with '{' and ends with '}'. For example in order to only display the current file and line number when the information is available the format would be: "{ at {$line.file.basename}:${line.number}}" Broken down this is: start the scope: "{" format whose content will only be displayed if all information is available: "at {$line.file.basename}:${line.number}" end the scope: "}" We currently can represent the infomration we see when stopped at a frame: frame #0: 0x0000000100000e85 a.out`main + 4 at test.c:19 with the following format: "frame #${frame.index}: ${frame.pc} {${module.file.basename}`}{${function.name}{${function.pc-offset}}{ at ${line.file.basename}:${line.number}}\n" This breaks down to always print: "frame #${frame.index}: ${frame.pc} " only print the module followed by a tick if we have a valid module: "{${module.file.basename}`}" print the function name with optional offset: "{${function.name}{${function.pc-offset}}" print the line info if it is available: "{ at ${line.file.basename}:${line.number}}" then finish off with a newline: "\n" Notice you can also put newlines ("\n") and tabs and everything else you are used to putting in a format string when desensitized with the \ character. Cleaned up some of the user settings controller subclasses. All of them do not have any global settings variables and were all implementing stubs for the get/set global settings variable. Now there is a default version in UserSettingsController that will do nothing. llvm-svn: 114306
2010-09-19 10:33:57 +08:00
Error &err
)
{
err.SetErrorString ("UserSettingsController has no global settings");
return false;
}
bool
UserSettingsController::GetGlobalVariable
(
const ConstString &var_name,
StringList &value,
Error &err
Added code that will allow completely customizable prompts for use in replacing the "(lldb)" prompt, the "frame #1..." displays when doing stack backtracing and the "thread #1....". This will allow you to see exactly the information that you want to see where you want to see it. This currently isn't hookup up to the prompts yet, but it will be soon. So what is the format of the prompts? Prompts can contain variables that have access to the current program state. Variables are text that appears in between a prefix of "${" and ends with a "}". Some of the interesting variables include: // The frame index (0, 1, 2, 3...) ${frame.index} // common frame registers with generic names ${frame.pc} ${frame.sp} ${frame.fp} ${frame.ra} ${frame.flags} // Access to any frame registers by name where REGNAME is any register name: ${frame.reg.REGNAME} // The current compile unit file where the frame is located ${file.basename} ${file.fullpath} // Function information ${function.name} ${function.pc-offset} // Process info ${process.file.basename} ${process.file.fullpath} ${process.id} ${process.name} // Thread info ${thread.id} ${thread.index} ${thread.name} ${thread.queue} ${thread.stop-reason} // Target information ${target.arch} // The current module for the current frame (the shared library or executable // that contains the current frame PC value): ${module.file.basename} ${module.file.fullpath} // Access to the line entry for where the current frame is when your thread // is stopped: ${line.file.basename} ${line.file.fullpath} ${line.number} ${line.start-addr} ${line.end-addr} Many times the information that you might have in your prompt might not be available and you won't want it to print out if it isn't valid. To take care of this you can enclose everything that must resolve into a scope. A scope is starts with '{' and ends with '}'. For example in order to only display the current file and line number when the information is available the format would be: "{ at {$line.file.basename}:${line.number}}" Broken down this is: start the scope: "{" format whose content will only be displayed if all information is available: "at {$line.file.basename}:${line.number}" end the scope: "}" We currently can represent the infomration we see when stopped at a frame: frame #0: 0x0000000100000e85 a.out`main + 4 at test.c:19 with the following format: "frame #${frame.index}: ${frame.pc} {${module.file.basename}`}{${function.name}{${function.pc-offset}}{ at ${line.file.basename}:${line.number}}\n" This breaks down to always print: "frame #${frame.index}: ${frame.pc} " only print the module followed by a tick if we have a valid module: "{${module.file.basename}`}" print the function name with optional offset: "{${function.name}{${function.pc-offset}}" print the line info if it is available: "{ at ${line.file.basename}:${line.number}}" then finish off with a newline: "\n" Notice you can also put newlines ("\n") and tabs and everything else you are used to putting in a format string when desensitized with the \ character. Cleaned up some of the user settings controller subclasses. All of them do not have any global settings variables and were all implementing stubs for the get/set global settings variable. Now there is a default version in UserSettingsController that will do nothing. llvm-svn: 114306
2010-09-19 10:33:57 +08:00
)
{
return false;
}
bool
Added new target instance settings for execution settings: Targets can now specify some additional parameters for when we debug executables that can help with plug-in selection: target.execution-level = auto | user | kernel target.execution-mode = auto | dynamic | static target.execution-os-type = auto | none | halted | live On some systems, the binaries that are created are the same wether you use them to debug a kernel, or a user space program. Many times inspecting an object file can reveal what an executable should be. For these cases we can now be a little more complete by specifying wether to detect all of these things automatically (inspect the main executable file and select a plug-in accordingly), or manually to force the selection of certain plug-ins. To do this we now allow the specficifation of wether one is debugging a user space program (target.execution-level = user) or a kernel program (target.execution-level = kernel). We can also specify if we want to debug a program where shared libraries are dynamically loaded using a DynamicLoader plug-in (target.execution-mode = dynamic), or wether we will treat all symbol files as already linked at the correct address (target.execution-mode = static). We can also specify if the inferior we are debugging is being debugged on a bare board (target.execution-os-type = none), or debugging an OS where we have a JTAG or other direct connection to the inferior stops the entire OS (target.execution-os-type = halted), or if we are debugging a program on something that has live debug services (target.execution-os-type = live). For the "target.execution-os-type = halted" mode, we will need to create ProcessHelper plug-ins that allow us to extract the process/thread and other OS information by reading/writing memory. This should allow LLDB to be used for a wide variety of debugging tasks and handle them all correctly. llvm-svn: 125815
2011-02-18 09:44:25 +08:00
UserSettingsController::InitializeSettingsController (UserSettingsControllerSP &controller_sp,
SettingEntry *global_settings,
SettingEntry *instance_settings)
{
Added new target instance settings for execution settings: Targets can now specify some additional parameters for when we debug executables that can help with plug-in selection: target.execution-level = auto | user | kernel target.execution-mode = auto | dynamic | static target.execution-os-type = auto | none | halted | live On some systems, the binaries that are created are the same wether you use them to debug a kernel, or a user space program. Many times inspecting an object file can reveal what an executable should be. For these cases we can now be a little more complete by specifying wether to detect all of these things automatically (inspect the main executable file and select a plug-in accordingly), or manually to force the selection of certain plug-ins. To do this we now allow the specficifation of wether one is debugging a user space program (target.execution-level = user) or a kernel program (target.execution-level = kernel). We can also specify if we want to debug a program where shared libraries are dynamically loaded using a DynamicLoader plug-in (target.execution-mode = dynamic), or wether we will treat all symbol files as already linked at the correct address (target.execution-mode = static). We can also specify if the inferior we are debugging is being debugged on a bare board (target.execution-os-type = none), or debugging an OS where we have a JTAG or other direct connection to the inferior stops the entire OS (target.execution-os-type = halted), or if we are debugging a program on something that has live debug services (target.execution-os-type = live). For the "target.execution-os-type = halted" mode, we will need to create ProcessHelper plug-ins that allow us to extract the process/thread and other OS information by reading/writing memory. This should allow LLDB to be used for a wide variety of debugging tasks and handle them all correctly. llvm-svn: 125815
2011-02-18 09:44:25 +08:00
const UserSettingsControllerSP &parent = controller_sp->GetParent ();
if (parent)
parent->RegisterChild (controller_sp);
controller_sp->CreateSettingsVector (global_settings, true);
controller_sp->CreateSettingsVector (instance_settings, false);
controller_sp->InitializeGlobalVariables ();
controller_sp->CreateDefaultInstanceSettings ();
return true;
}
void
Added new target instance settings for execution settings: Targets can now specify some additional parameters for when we debug executables that can help with plug-in selection: target.execution-level = auto | user | kernel target.execution-mode = auto | dynamic | static target.execution-os-type = auto | none | halted | live On some systems, the binaries that are created are the same wether you use them to debug a kernel, or a user space program. Many times inspecting an object file can reveal what an executable should be. For these cases we can now be a little more complete by specifying wether to detect all of these things automatically (inspect the main executable file and select a plug-in accordingly), or manually to force the selection of certain plug-ins. To do this we now allow the specficifation of wether one is debugging a user space program (target.execution-level = user) or a kernel program (target.execution-level = kernel). We can also specify if we want to debug a program where shared libraries are dynamically loaded using a DynamicLoader plug-in (target.execution-mode = dynamic), or wether we will treat all symbol files as already linked at the correct address (target.execution-mode = static). We can also specify if the inferior we are debugging is being debugged on a bare board (target.execution-os-type = none), or debugging an OS where we have a JTAG or other direct connection to the inferior stops the entire OS (target.execution-os-type = halted), or if we are debugging a program on something that has live debug services (target.execution-os-type = live). For the "target.execution-os-type = halted" mode, we will need to create ProcessHelper plug-ins that allow us to extract the process/thread and other OS information by reading/writing memory. This should allow LLDB to be used for a wide variety of debugging tasks and handle them all correctly. llvm-svn: 125815
2011-02-18 09:44:25 +08:00
UserSettingsController::FinalizeSettingsController (UserSettingsControllerSP &controller_sp)
{
Added new target instance settings for execution settings: Targets can now specify some additional parameters for when we debug executables that can help with plug-in selection: target.execution-level = auto | user | kernel target.execution-mode = auto | dynamic | static target.execution-os-type = auto | none | halted | live On some systems, the binaries that are created are the same wether you use them to debug a kernel, or a user space program. Many times inspecting an object file can reveal what an executable should be. For these cases we can now be a little more complete by specifying wether to detect all of these things automatically (inspect the main executable file and select a plug-in accordingly), or manually to force the selection of certain plug-ins. To do this we now allow the specficifation of wether one is debugging a user space program (target.execution-level = user) or a kernel program (target.execution-level = kernel). We can also specify if we want to debug a program where shared libraries are dynamically loaded using a DynamicLoader plug-in (target.execution-mode = dynamic), or wether we will treat all symbol files as already linked at the correct address (target.execution-mode = static). We can also specify if the inferior we are debugging is being debugged on a bare board (target.execution-os-type = none), or debugging an OS where we have a JTAG or other direct connection to the inferior stops the entire OS (target.execution-os-type = halted), or if we are debugging a program on something that has live debug services (target.execution-os-type = live). For the "target.execution-os-type = halted" mode, we will need to create ProcessHelper plug-ins that allow us to extract the process/thread and other OS information by reading/writing memory. This should allow LLDB to be used for a wide variety of debugging tasks and handle them all correctly. llvm-svn: 125815
2011-02-18 09:44:25 +08:00
const UserSettingsControllerSP &parent = controller_sp->GetParent ();
if (parent)
parent->RemoveChild (controller_sp);
}
void
UserSettingsController::InitializeGlobalVariables ()
{
int num_entries;
const char *prefix = GetLevelName().GetCString();
num_entries = m_settings.global_settings.size();
for (int i = 0; i < num_entries; ++i)
{
Added new target instance settings for execution settings: Targets can now specify some additional parameters for when we debug executables that can help with plug-in selection: target.execution-level = auto | user | kernel target.execution-mode = auto | dynamic | static target.execution-os-type = auto | none | halted | live On some systems, the binaries that are created are the same wether you use them to debug a kernel, or a user space program. Many times inspecting an object file can reveal what an executable should be. For these cases we can now be a little more complete by specifying wether to detect all of these things automatically (inspect the main executable file and select a plug-in accordingly), or manually to force the selection of certain plug-ins. To do this we now allow the specficifation of wether one is debugging a user space program (target.execution-level = user) or a kernel program (target.execution-level = kernel). We can also specify if we want to debug a program where shared libraries are dynamically loaded using a DynamicLoader plug-in (target.execution-mode = dynamic), or wether we will treat all symbol files as already linked at the correct address (target.execution-mode = static). We can also specify if the inferior we are debugging is being debugged on a bare board (target.execution-os-type = none), or debugging an OS where we have a JTAG or other direct connection to the inferior stops the entire OS (target.execution-os-type = halted), or if we are debugging a program on something that has live debug services (target.execution-os-type = live). For the "target.execution-os-type = halted" mode, we will need to create ProcessHelper plug-ins that allow us to extract the process/thread and other OS information by reading/writing memory. This should allow LLDB to be used for a wide variety of debugging tasks and handle them all correctly. llvm-svn: 125815
2011-02-18 09:44:25 +08:00
const SettingEntry &entry = m_settings.global_settings[i];
if (entry.default_value != NULL)
{
StreamString full_name;
if (prefix[0] != '\0')
full_name.Printf ("%s.%s", prefix, entry.var_name);
else
full_name.Printf ("%s", entry.var_name);
Added new target instance settings for execution settings: Targets can now specify some additional parameters for when we debug executables that can help with plug-in selection: target.execution-level = auto | user | kernel target.execution-mode = auto | dynamic | static target.execution-os-type = auto | none | halted | live On some systems, the binaries that are created are the same wether you use them to debug a kernel, or a user space program. Many times inspecting an object file can reveal what an executable should be. For these cases we can now be a little more complete by specifying wether to detect all of these things automatically (inspect the main executable file and select a plug-in accordingly), or manually to force the selection of certain plug-ins. To do this we now allow the specficifation of wether one is debugging a user space program (target.execution-level = user) or a kernel program (target.execution-level = kernel). We can also specify if we want to debug a program where shared libraries are dynamically loaded using a DynamicLoader plug-in (target.execution-mode = dynamic), or wether we will treat all symbol files as already linked at the correct address (target.execution-mode = static). We can also specify if the inferior we are debugging is being debugged on a bare board (target.execution-os-type = none), or debugging an OS where we have a JTAG or other direct connection to the inferior stops the entire OS (target.execution-os-type = halted), or if we are debugging a program on something that has live debug services (target.execution-os-type = live). For the "target.execution-os-type = halted" mode, we will need to create ProcessHelper plug-ins that allow us to extract the process/thread and other OS information by reading/writing memory. This should allow LLDB to be used for a wide variety of debugging tasks and handle them all correctly. llvm-svn: 125815
2011-02-18 09:44:25 +08:00
SetVariable (full_name.GetData(), entry.default_value, eVarSetOperationAssign, false, "");
}
}
}
Added new target instance settings for execution settings: Targets can now specify some additional parameters for when we debug executables that can help with plug-in selection: target.execution-level = auto | user | kernel target.execution-mode = auto | dynamic | static target.execution-os-type = auto | none | halted | live On some systems, the binaries that are created are the same wether you use them to debug a kernel, or a user space program. Many times inspecting an object file can reveal what an executable should be. For these cases we can now be a little more complete by specifying wether to detect all of these things automatically (inspect the main executable file and select a plug-in accordingly), or manually to force the selection of certain plug-ins. To do this we now allow the specficifation of wether one is debugging a user space program (target.execution-level = user) or a kernel program (target.execution-level = kernel). We can also specify if we want to debug a program where shared libraries are dynamically loaded using a DynamicLoader plug-in (target.execution-mode = dynamic), or wether we will treat all symbol files as already linked at the correct address (target.execution-mode = static). We can also specify if the inferior we are debugging is being debugged on a bare board (target.execution-os-type = none), or debugging an OS where we have a JTAG or other direct connection to the inferior stops the entire OS (target.execution-os-type = halted), or if we are debugging a program on something that has live debug services (target.execution-os-type = live). For the "target.execution-os-type = halted" mode, we will need to create ProcessHelper plug-ins that allow us to extract the process/thread and other OS information by reading/writing memory. This should allow LLDB to be used for a wide variety of debugging tasks and handle them all correctly. llvm-svn: 125815
2011-02-18 09:44:25 +08:00
const UserSettingsControllerSP &
UserSettingsController::GetParent ()
{
return m_settings.parent;
}
void
Added new target instance settings for execution settings: Targets can now specify some additional parameters for when we debug executables that can help with plug-in selection: target.execution-level = auto | user | kernel target.execution-mode = auto | dynamic | static target.execution-os-type = auto | none | halted | live On some systems, the binaries that are created are the same wether you use them to debug a kernel, or a user space program. Many times inspecting an object file can reveal what an executable should be. For these cases we can now be a little more complete by specifying wether to detect all of these things automatically (inspect the main executable file and select a plug-in accordingly), or manually to force the selection of certain plug-ins. To do this we now allow the specficifation of wether one is debugging a user space program (target.execution-level = user) or a kernel program (target.execution-level = kernel). We can also specify if we want to debug a program where shared libraries are dynamically loaded using a DynamicLoader plug-in (target.execution-mode = dynamic), or wether we will treat all symbol files as already linked at the correct address (target.execution-mode = static). We can also specify if the inferior we are debugging is being debugged on a bare board (target.execution-os-type = none), or debugging an OS where we have a JTAG or other direct connection to the inferior stops the entire OS (target.execution-os-type = halted), or if we are debugging a program on something that has live debug services (target.execution-os-type = live). For the "target.execution-os-type = halted" mode, we will need to create ProcessHelper plug-ins that allow us to extract the process/thread and other OS information by reading/writing memory. This should allow LLDB to be used for a wide variety of debugging tasks and handle them all correctly. llvm-svn: 125815
2011-02-18 09:44:25 +08:00
UserSettingsController::RegisterChild (const UserSettingsControllerSP &child)
{
Mutex::Locker locker (m_children_mutex);
// Verify child is not already in m_children.
size_t num_children = m_children.size();
bool found = false;
for (size_t i = 0; i < num_children; ++i)
{
if (m_children[i].get() == child.get())
found = true;
}
// Add child to m_children.
if (! found)
m_children.push_back (child);
}
const ConstString &
UserSettingsController::GetLevelName ()
{
return m_settings.level_name;
}
size_t
UserSettingsController::GetNumChildren ()
{
return m_children.size();
}
Added new target instance settings for execution settings: Targets can now specify some additional parameters for when we debug executables that can help with plug-in selection: target.execution-level = auto | user | kernel target.execution-mode = auto | dynamic | static target.execution-os-type = auto | none | halted | live On some systems, the binaries that are created are the same wether you use them to debug a kernel, or a user space program. Many times inspecting an object file can reveal what an executable should be. For these cases we can now be a little more complete by specifying wether to detect all of these things automatically (inspect the main executable file and select a plug-in accordingly), or manually to force the selection of certain plug-ins. To do this we now allow the specficifation of wether one is debugging a user space program (target.execution-level = user) or a kernel program (target.execution-level = kernel). We can also specify if we want to debug a program where shared libraries are dynamically loaded using a DynamicLoader plug-in (target.execution-mode = dynamic), or wether we will treat all symbol files as already linked at the correct address (target.execution-mode = static). We can also specify if the inferior we are debugging is being debugged on a bare board (target.execution-os-type = none), or debugging an OS where we have a JTAG or other direct connection to the inferior stops the entire OS (target.execution-os-type = halted), or if we are debugging a program on something that has live debug services (target.execution-os-type = live). For the "target.execution-os-type = halted" mode, we will need to create ProcessHelper plug-ins that allow us to extract the process/thread and other OS information by reading/writing memory. This should allow LLDB to be used for a wide variety of debugging tasks and handle them all correctly. llvm-svn: 125815
2011-02-18 09:44:25 +08:00
const UserSettingsControllerSP
UserSettingsController::GetChildAtIndex (size_t index)
{
if (index < m_children.size())
return m_children[index];
Added new target instance settings for execution settings: Targets can now specify some additional parameters for when we debug executables that can help with plug-in selection: target.execution-level = auto | user | kernel target.execution-mode = auto | dynamic | static target.execution-os-type = auto | none | halted | live On some systems, the binaries that are created are the same wether you use them to debug a kernel, or a user space program. Many times inspecting an object file can reveal what an executable should be. For these cases we can now be a little more complete by specifying wether to detect all of these things automatically (inspect the main executable file and select a plug-in accordingly), or manually to force the selection of certain plug-ins. To do this we now allow the specficifation of wether one is debugging a user space program (target.execution-level = user) or a kernel program (target.execution-level = kernel). We can also specify if we want to debug a program where shared libraries are dynamically loaded using a DynamicLoader plug-in (target.execution-mode = dynamic), or wether we will treat all symbol files as already linked at the correct address (target.execution-mode = static). We can also specify if the inferior we are debugging is being debugged on a bare board (target.execution-os-type = none), or debugging an OS where we have a JTAG or other direct connection to the inferior stops the entire OS (target.execution-os-type = halted), or if we are debugging a program on something that has live debug services (target.execution-os-type = live). For the "target.execution-os-type = halted" mode, we will need to create ProcessHelper plug-ins that allow us to extract the process/thread and other OS information by reading/writing memory. This should allow LLDB to be used for a wide variety of debugging tasks and handle them all correctly. llvm-svn: 125815
2011-02-18 09:44:25 +08:00
UserSettingsControllerSP dummy_value;
return dummy_value;
}
const SettingEntry *
UserSettingsController::GetGlobalEntry (const ConstString &var_name)
{
for (int i = 0; i < m_settings.global_settings.size(); ++i)
{
Added new target instance settings for execution settings: Targets can now specify some additional parameters for when we debug executables that can help with plug-in selection: target.execution-level = auto | user | kernel target.execution-mode = auto | dynamic | static target.execution-os-type = auto | none | halted | live On some systems, the binaries that are created are the same wether you use them to debug a kernel, or a user space program. Many times inspecting an object file can reveal what an executable should be. For these cases we can now be a little more complete by specifying wether to detect all of these things automatically (inspect the main executable file and select a plug-in accordingly), or manually to force the selection of certain plug-ins. To do this we now allow the specficifation of wether one is debugging a user space program (target.execution-level = user) or a kernel program (target.execution-level = kernel). We can also specify if we want to debug a program where shared libraries are dynamically loaded using a DynamicLoader plug-in (target.execution-mode = dynamic), or wether we will treat all symbol files as already linked at the correct address (target.execution-mode = static). We can also specify if the inferior we are debugging is being debugged on a bare board (target.execution-os-type = none), or debugging an OS where we have a JTAG or other direct connection to the inferior stops the entire OS (target.execution-os-type = halted), or if we are debugging a program on something that has live debug services (target.execution-os-type = live). For the "target.execution-os-type = halted" mode, we will need to create ProcessHelper plug-ins that allow us to extract the process/thread and other OS information by reading/writing memory. This should allow LLDB to be used for a wide variety of debugging tasks and handle them all correctly. llvm-svn: 125815
2011-02-18 09:44:25 +08:00
const SettingEntry &entry = m_settings.global_settings[i];
ConstString entry_name (entry.var_name);
if (entry_name == var_name)
return &entry;
}
return NULL;
}
const SettingEntry *
UserSettingsController::GetInstanceEntry (const ConstString &const_var_name)
{
for (int i = 0; i < m_settings.instance_settings.size(); ++i)
{
SettingEntry &entry = m_settings.instance_settings[i];
ConstString entry_name (entry.var_name);
if (entry_name == const_var_name)
return &entry;
}
return NULL;
}
void
UserSettingsController::BuildParentPrefix (std::string &parent_prefix)
{
Added new target instance settings for execution settings: Targets can now specify some additional parameters for when we debug executables that can help with plug-in selection: target.execution-level = auto | user | kernel target.execution-mode = auto | dynamic | static target.execution-os-type = auto | none | halted | live On some systems, the binaries that are created are the same wether you use them to debug a kernel, or a user space program. Many times inspecting an object file can reveal what an executable should be. For these cases we can now be a little more complete by specifying wether to detect all of these things automatically (inspect the main executable file and select a plug-in accordingly), or manually to force the selection of certain plug-ins. To do this we now allow the specficifation of wether one is debugging a user space program (target.execution-level = user) or a kernel program (target.execution-level = kernel). We can also specify if we want to debug a program where shared libraries are dynamically loaded using a DynamicLoader plug-in (target.execution-mode = dynamic), or wether we will treat all symbol files as already linked at the correct address (target.execution-mode = static). We can also specify if the inferior we are debugging is being debugged on a bare board (target.execution-os-type = none), or debugging an OS where we have a JTAG or other direct connection to the inferior stops the entire OS (target.execution-os-type = halted), or if we are debugging a program on something that has live debug services (target.execution-os-type = live). For the "target.execution-os-type = halted" mode, we will need to create ProcessHelper plug-ins that allow us to extract the process/thread and other OS information by reading/writing memory. This should allow LLDB to be used for a wide variety of debugging tasks and handle them all correctly. llvm-svn: 125815
2011-02-18 09:44:25 +08:00
UserSettingsControllerSP parent = GetParent();
if (parent.get() != NULL)
{
parent->BuildParentPrefix (parent_prefix);
if (parent_prefix.length() > 0)
parent_prefix.append (".");
}
parent_prefix.append (GetLevelName().GetCString());
}
void
Added new target instance settings for execution settings: Targets can now specify some additional parameters for when we debug executables that can help with plug-in selection: target.execution-level = auto | user | kernel target.execution-mode = auto | dynamic | static target.execution-os-type = auto | none | halted | live On some systems, the binaries that are created are the same wether you use them to debug a kernel, or a user space program. Many times inspecting an object file can reveal what an executable should be. For these cases we can now be a little more complete by specifying wether to detect all of these things automatically (inspect the main executable file and select a plug-in accordingly), or manually to force the selection of certain plug-ins. To do this we now allow the specficifation of wether one is debugging a user space program (target.execution-level = user) or a kernel program (target.execution-level = kernel). We can also specify if we want to debug a program where shared libraries are dynamically loaded using a DynamicLoader plug-in (target.execution-mode = dynamic), or wether we will treat all symbol files as already linked at the correct address (target.execution-mode = static). We can also specify if the inferior we are debugging is being debugged on a bare board (target.execution-os-type = none), or debugging an OS where we have a JTAG or other direct connection to the inferior stops the entire OS (target.execution-os-type = halted), or if we are debugging a program on something that has live debug services (target.execution-os-type = live). For the "target.execution-os-type = halted" mode, we will need to create ProcessHelper plug-ins that allow us to extract the process/thread and other OS information by reading/writing memory. This should allow LLDB to be used for a wide variety of debugging tasks and handle them all correctly. llvm-svn: 125815
2011-02-18 09:44:25 +08:00
UserSettingsController::RemoveChild (const UserSettingsControllerSP &child)
{
Mutex::Locker locker (m_children_mutex);
Added new target instance settings for execution settings: Targets can now specify some additional parameters for when we debug executables that can help with plug-in selection: target.execution-level = auto | user | kernel target.execution-mode = auto | dynamic | static target.execution-os-type = auto | none | halted | live On some systems, the binaries that are created are the same wether you use them to debug a kernel, or a user space program. Many times inspecting an object file can reveal what an executable should be. For these cases we can now be a little more complete by specifying wether to detect all of these things automatically (inspect the main executable file and select a plug-in accordingly), or manually to force the selection of certain plug-ins. To do this we now allow the specficifation of wether one is debugging a user space program (target.execution-level = user) or a kernel program (target.execution-level = kernel). We can also specify if we want to debug a program where shared libraries are dynamically loaded using a DynamicLoader plug-in (target.execution-mode = dynamic), or wether we will treat all symbol files as already linked at the correct address (target.execution-mode = static). We can also specify if the inferior we are debugging is being debugged on a bare board (target.execution-os-type = none), or debugging an OS where we have a JTAG or other direct connection to the inferior stops the entire OS (target.execution-os-type = halted), or if we are debugging a program on something that has live debug services (target.execution-os-type = live). For the "target.execution-os-type = halted" mode, we will need to create ProcessHelper plug-ins that allow us to extract the process/thread and other OS information by reading/writing memory. This should allow LLDB to be used for a wide variety of debugging tasks and handle them all correctly. llvm-svn: 125815
2011-02-18 09:44:25 +08:00
std::vector<UserSettingsControllerSP>::iterator pos, end = m_children.end();
for (pos = m_children.begin(); pos != end; ++pos)
{
Added new target instance settings for execution settings: Targets can now specify some additional parameters for when we debug executables that can help with plug-in selection: target.execution-level = auto | user | kernel target.execution-mode = auto | dynamic | static target.execution-os-type = auto | none | halted | live On some systems, the binaries that are created are the same wether you use them to debug a kernel, or a user space program. Many times inspecting an object file can reveal what an executable should be. For these cases we can now be a little more complete by specifying wether to detect all of these things automatically (inspect the main executable file and select a plug-in accordingly), or manually to force the selection of certain plug-ins. To do this we now allow the specficifation of wether one is debugging a user space program (target.execution-level = user) or a kernel program (target.execution-level = kernel). We can also specify if we want to debug a program where shared libraries are dynamically loaded using a DynamicLoader plug-in (target.execution-mode = dynamic), or wether we will treat all symbol files as already linked at the correct address (target.execution-mode = static). We can also specify if the inferior we are debugging is being debugged on a bare board (target.execution-os-type = none), or debugging an OS where we have a JTAG or other direct connection to the inferior stops the entire OS (target.execution-os-type = halted), or if we are debugging a program on something that has live debug services (target.execution-os-type = live). For the "target.execution-os-type = halted" mode, we will need to create ProcessHelper plug-ins that allow us to extract the process/thread and other OS information by reading/writing memory. This should allow LLDB to be used for a wide variety of debugging tasks and handle them all correctly. llvm-svn: 125815
2011-02-18 09:44:25 +08:00
UserSettingsControllerSP entry = *pos;
if (entry == child)
{
m_children.erase (pos);
break;
}
}
}
Error
UserSettingsController::SetVariable (const char *full_dot_name,
const char *value,
Added new target instance settings for execution settings: Targets can now specify some additional parameters for when we debug executables that can help with plug-in selection: target.execution-level = auto | user | kernel target.execution-mode = auto | dynamic | static target.execution-os-type = auto | none | halted | live On some systems, the binaries that are created are the same wether you use them to debug a kernel, or a user space program. Many times inspecting an object file can reveal what an executable should be. For these cases we can now be a little more complete by specifying wether to detect all of these things automatically (inspect the main executable file and select a plug-in accordingly), or manually to force the selection of certain plug-ins. To do this we now allow the specficifation of wether one is debugging a user space program (target.execution-level = user) or a kernel program (target.execution-level = kernel). We can also specify if we want to debug a program where shared libraries are dynamically loaded using a DynamicLoader plug-in (target.execution-mode = dynamic), or wether we will treat all symbol files as already linked at the correct address (target.execution-mode = static). We can also specify if the inferior we are debugging is being debugged on a bare board (target.execution-os-type = none), or debugging an OS where we have a JTAG or other direct connection to the inferior stops the entire OS (target.execution-os-type = halted), or if we are debugging a program on something that has live debug services (target.execution-os-type = live). For the "target.execution-os-type = halted" mode, we will need to create ProcessHelper plug-ins that allow us to extract the process/thread and other OS information by reading/writing memory. This should allow LLDB to be used for a wide variety of debugging tasks and handle them all correctly. llvm-svn: 125815
2011-02-18 09:44:25 +08:00
const VarSetOperationType op,
const bool override,
const char *debugger_instance_name,
const char *index_value)
{
Error err;
ConstString const_var_name;
const ConstString &default_name = InstanceSettings::GetDefaultName();
Args names = UserSettingsController::BreakNameIntoPieces (full_dot_name);
int num_pieces = names.GetArgumentCount();
if (num_pieces < 1)
{
err.SetErrorStringWithFormat ("'%s' is not a valid variable name; cannot assign value.\n", full_dot_name);
return err;
}
ConstString prefix (names.GetArgumentAtIndex (0));
if ((prefix == m_settings.level_name)
|| (m_settings.level_name.GetLength() == 0))
{
if (prefix == m_settings.level_name)
{
names.Shift ();
num_pieces = names.GetArgumentCount();
}
if (num_pieces == 0)
{
err.SetErrorString ("No variable name specified; cannot assign value.\n");
return err;
}
else if (num_pieces == 1)
{
// Must be one of the class-wide settings.
const_var_name.SetCString (names.GetArgumentAtIndex (0));
const SettingEntry *entry = GetGlobalEntry (const_var_name);
if (entry)
{
UserSettingsController::VerifyOperationForType (entry->var_type, op, const_var_name, err);
if (err.Fail())
return err;
if ((value == NULL || value[0] == '\0')
Added new target instance settings for execution settings: Targets can now specify some additional parameters for when we debug executables that can help with plug-in selection: target.execution-level = auto | user | kernel target.execution-mode = auto | dynamic | static target.execution-os-type = auto | none | halted | live On some systems, the binaries that are created are the same wether you use them to debug a kernel, or a user space program. Many times inspecting an object file can reveal what an executable should be. For these cases we can now be a little more complete by specifying wether to detect all of these things automatically (inspect the main executable file and select a plug-in accordingly), or manually to force the selection of certain plug-ins. To do this we now allow the specficifation of wether one is debugging a user space program (target.execution-level = user) or a kernel program (target.execution-level = kernel). We can also specify if we want to debug a program where shared libraries are dynamically loaded using a DynamicLoader plug-in (target.execution-mode = dynamic), or wether we will treat all symbol files as already linked at the correct address (target.execution-mode = static). We can also specify if the inferior we are debugging is being debugged on a bare board (target.execution-os-type = none), or debugging an OS where we have a JTAG or other direct connection to the inferior stops the entire OS (target.execution-os-type = halted), or if we are debugging a program on something that has live debug services (target.execution-os-type = live). For the "target.execution-os-type = halted" mode, we will need to create ProcessHelper plug-ins that allow us to extract the process/thread and other OS information by reading/writing memory. This should allow LLDB to be used for a wide variety of debugging tasks and handle them all correctly. llvm-svn: 125815
2011-02-18 09:44:25 +08:00
&& (op == eVarSetOperationAssign))
{
Added new target instance settings for execution settings: Targets can now specify some additional parameters for when we debug executables that can help with plug-in selection: target.execution-level = auto | user | kernel target.execution-mode = auto | dynamic | static target.execution-os-type = auto | none | halted | live On some systems, the binaries that are created are the same wether you use them to debug a kernel, or a user space program. Many times inspecting an object file can reveal what an executable should be. For these cases we can now be a little more complete by specifying wether to detect all of these things automatically (inspect the main executable file and select a plug-in accordingly), or manually to force the selection of certain plug-ins. To do this we now allow the specficifation of wether one is debugging a user space program (target.execution-level = user) or a kernel program (target.execution-level = kernel). We can also specify if we want to debug a program where shared libraries are dynamically loaded using a DynamicLoader plug-in (target.execution-mode = dynamic), or wether we will treat all symbol files as already linked at the correct address (target.execution-mode = static). We can also specify if the inferior we are debugging is being debugged on a bare board (target.execution-os-type = none), or debugging an OS where we have a JTAG or other direct connection to the inferior stops the entire OS (target.execution-os-type = halted), or if we are debugging a program on something that has live debug services (target.execution-os-type = live). For the "target.execution-os-type = halted" mode, we will need to create ProcessHelper plug-ins that allow us to extract the process/thread and other OS information by reading/writing memory. This should allow LLDB to be used for a wide variety of debugging tasks and handle them all correctly. llvm-svn: 125815
2011-02-18 09:44:25 +08:00
if (entry->var_type != eSetVarTypeEnum)
value = entry->default_value;
else
value = entry->enum_values[0].string_value;
}
Added code that will allow completely customizable prompts for use in replacing the "(lldb)" prompt, the "frame #1..." displays when doing stack backtracing and the "thread #1....". This will allow you to see exactly the information that you want to see where you want to see it. This currently isn't hookup up to the prompts yet, but it will be soon. So what is the format of the prompts? Prompts can contain variables that have access to the current program state. Variables are text that appears in between a prefix of "${" and ends with a "}". Some of the interesting variables include: // The frame index (0, 1, 2, 3...) ${frame.index} // common frame registers with generic names ${frame.pc} ${frame.sp} ${frame.fp} ${frame.ra} ${frame.flags} // Access to any frame registers by name where REGNAME is any register name: ${frame.reg.REGNAME} // The current compile unit file where the frame is located ${file.basename} ${file.fullpath} // Function information ${function.name} ${function.pc-offset} // Process info ${process.file.basename} ${process.file.fullpath} ${process.id} ${process.name} // Thread info ${thread.id} ${thread.index} ${thread.name} ${thread.queue} ${thread.stop-reason} // Target information ${target.arch} // The current module for the current frame (the shared library or executable // that contains the current frame PC value): ${module.file.basename} ${module.file.fullpath} // Access to the line entry for where the current frame is when your thread // is stopped: ${line.file.basename} ${line.file.fullpath} ${line.number} ${line.start-addr} ${line.end-addr} Many times the information that you might have in your prompt might not be available and you won't want it to print out if it isn't valid. To take care of this you can enclose everything that must resolve into a scope. A scope is starts with '{' and ends with '}'. For example in order to only display the current file and line number when the information is available the format would be: "{ at {$line.file.basename}:${line.number}}" Broken down this is: start the scope: "{" format whose content will only be displayed if all information is available: "at {$line.file.basename}:${line.number}" end the scope: "}" We currently can represent the infomration we see when stopped at a frame: frame #0: 0x0000000100000e85 a.out`main + 4 at test.c:19 with the following format: "frame #${frame.index}: ${frame.pc} {${module.file.basename}`}{${function.name}{${function.pc-offset}}{ at ${line.file.basename}:${line.number}}\n" This breaks down to always print: "frame #${frame.index}: ${frame.pc} " only print the module followed by a tick if we have a valid module: "{${module.file.basename}`}" print the function name with optional offset: "{${function.name}{${function.pc-offset}}" print the line info if it is available: "{ at ${line.file.basename}:${line.number}}" then finish off with a newline: "\n" Notice you can also put newlines ("\n") and tabs and everything else you are used to putting in a format string when desensitized with the \ character. Cleaned up some of the user settings controller subclasses. All of them do not have any global settings variables and were all implementing stubs for the get/set global settings variable. Now there is a default version in UserSettingsController that will do nothing. llvm-svn: 114306
2010-09-19 10:33:57 +08:00
SetGlobalVariable (const_var_name, index_value, value, *entry, op, err);
}
else
{
// MIGHT be instance variable, to be for ALL instances.
entry = GetInstanceEntry (const_var_name);
if (entry == NULL)
{
err.SetErrorStringWithFormat ("Unable to find variable '%s.%s'; cannot assign value.\n",
prefix.GetCString(), const_var_name.GetCString());
return err;
}
else
{
UserSettingsController::VerifyOperationForType (entry->var_type, op, const_var_name, err);
if (err.Fail())
return err;
if ((value == NULL || value[0] == '\0')
Added new target instance settings for execution settings: Targets can now specify some additional parameters for when we debug executables that can help with plug-in selection: target.execution-level = auto | user | kernel target.execution-mode = auto | dynamic | static target.execution-os-type = auto | none | halted | live On some systems, the binaries that are created are the same wether you use them to debug a kernel, or a user space program. Many times inspecting an object file can reveal what an executable should be. For these cases we can now be a little more complete by specifying wether to detect all of these things automatically (inspect the main executable file and select a plug-in accordingly), or manually to force the selection of certain plug-ins. To do this we now allow the specficifation of wether one is debugging a user space program (target.execution-level = user) or a kernel program (target.execution-level = kernel). We can also specify if we want to debug a program where shared libraries are dynamically loaded using a DynamicLoader plug-in (target.execution-mode = dynamic), or wether we will treat all symbol files as already linked at the correct address (target.execution-mode = static). We can also specify if the inferior we are debugging is being debugged on a bare board (target.execution-os-type = none), or debugging an OS where we have a JTAG or other direct connection to the inferior stops the entire OS (target.execution-os-type = halted), or if we are debugging a program on something that has live debug services (target.execution-os-type = live). For the "target.execution-os-type = halted" mode, we will need to create ProcessHelper plug-ins that allow us to extract the process/thread and other OS information by reading/writing memory. This should allow LLDB to be used for a wide variety of debugging tasks and handle them all correctly. llvm-svn: 125815
2011-02-18 09:44:25 +08:00
&& (op == eVarSetOperationAssign))
{
Added new target instance settings for execution settings: Targets can now specify some additional parameters for when we debug executables that can help with plug-in selection: target.execution-level = auto | user | kernel target.execution-mode = auto | dynamic | static target.execution-os-type = auto | none | halted | live On some systems, the binaries that are created are the same wether you use them to debug a kernel, or a user space program. Many times inspecting an object file can reveal what an executable should be. For these cases we can now be a little more complete by specifying wether to detect all of these things automatically (inspect the main executable file and select a plug-in accordingly), or manually to force the selection of certain plug-ins. To do this we now allow the specficifation of wether one is debugging a user space program (target.execution-level = user) or a kernel program (target.execution-level = kernel). We can also specify if we want to debug a program where shared libraries are dynamically loaded using a DynamicLoader plug-in (target.execution-mode = dynamic), or wether we will treat all symbol files as already linked at the correct address (target.execution-mode = static). We can also specify if the inferior we are debugging is being debugged on a bare board (target.execution-os-type = none), or debugging an OS where we have a JTAG or other direct connection to the inferior stops the entire OS (target.execution-os-type = halted), or if we are debugging a program on something that has live debug services (target.execution-os-type = live). For the "target.execution-os-type = halted" mode, we will need to create ProcessHelper plug-ins that allow us to extract the process/thread and other OS information by reading/writing memory. This should allow LLDB to be used for a wide variety of debugging tasks and handle them all correctly. llvm-svn: 125815
2011-02-18 09:44:25 +08:00
if (entry->var_type != eSetVarTypeEnum)
value = entry->default_value;
else
value = entry->enum_values[0].string_value;
}
if ((m_settings.level_name.GetLength() > 0)
|| strlen (debugger_instance_name) == 0)
{
// Set the default settings
m_default_settings->UpdateInstanceSettingsVariable (const_var_name, index_value, value,
default_name, *entry, op, err, true);
}
else
{
// We're at the Debugger level; find the correct debugger instance and set those settings
StreamString tmp_name;
if (debugger_instance_name[0] != '[')
tmp_name.Printf ("[%s]", debugger_instance_name);
else
tmp_name.Printf ("%s", debugger_instance_name);
ConstString dbg_name (tmp_name.GetData());
InstanceSettings *dbg_settings = FindSettingsForInstance (dbg_name);
if (dbg_settings)
dbg_settings->UpdateInstanceSettingsVariable (const_var_name, index_value, value, dbg_name,
*entry, op, err, false);
}
if (override)
{
OverrideAllInstances (const_var_name, value, op, index_value, err);
// Update all pending records as well.
Added new target instance settings for execution settings: Targets can now specify some additional parameters for when we debug executables that can help with plug-in selection: target.execution-level = auto | user | kernel target.execution-mode = auto | dynamic | static target.execution-os-type = auto | none | halted | live On some systems, the binaries that are created are the same wether you use them to debug a kernel, or a user space program. Many times inspecting an object file can reveal what an executable should be. For these cases we can now be a little more complete by specifying wether to detect all of these things automatically (inspect the main executable file and select a plug-in accordingly), or manually to force the selection of certain plug-ins. To do this we now allow the specficifation of wether one is debugging a user space program (target.execution-level = user) or a kernel program (target.execution-level = kernel). We can also specify if we want to debug a program where shared libraries are dynamically loaded using a DynamicLoader plug-in (target.execution-mode = dynamic), or wether we will treat all symbol files as already linked at the correct address (target.execution-mode = static). We can also specify if the inferior we are debugging is being debugged on a bare board (target.execution-os-type = none), or debugging an OS where we have a JTAG or other direct connection to the inferior stops the entire OS (target.execution-os-type = halted), or if we are debugging a program on something that has live debug services (target.execution-os-type = live). For the "target.execution-os-type = halted" mode, we will need to create ProcessHelper plug-ins that allow us to extract the process/thread and other OS information by reading/writing memory. This should allow LLDB to be used for a wide variety of debugging tasks and handle them all correctly. llvm-svn: 125815
2011-02-18 09:44:25 +08:00
// std::map<std::string, InstanceSettingsSP>::iterator pos, end = m_pending_settings.end();
// for (pos = m_pending_settings.begin(); pos != end; end++)
// {
// const ConstString instance_name (pos->first.c_str());
Added new target instance settings for execution settings: Targets can now specify some additional parameters for when we debug executables that can help with plug-in selection: target.execution-level = auto | user | kernel target.execution-mode = auto | dynamic | static target.execution-os-type = auto | none | halted | live On some systems, the binaries that are created are the same wether you use them to debug a kernel, or a user space program. Many times inspecting an object file can reveal what an executable should be. For these cases we can now be a little more complete by specifying wether to detect all of these things automatically (inspect the main executable file and select a plug-in accordingly), or manually to force the selection of certain plug-ins. To do this we now allow the specficifation of wether one is debugging a user space program (target.execution-level = user) or a kernel program (target.execution-level = kernel). We can also specify if we want to debug a program where shared libraries are dynamically loaded using a DynamicLoader plug-in (target.execution-mode = dynamic), or wether we will treat all symbol files as already linked at the correct address (target.execution-mode = static). We can also specify if the inferior we are debugging is being debugged on a bare board (target.execution-os-type = none), or debugging an OS where we have a JTAG or other direct connection to the inferior stops the entire OS (target.execution-os-type = halted), or if we are debugging a program on something that has live debug services (target.execution-os-type = live). For the "target.execution-os-type = halted" mode, we will need to create ProcessHelper plug-ins that allow us to extract the process/thread and other OS information by reading/writing memory. This should allow LLDB to be used for a wide variety of debugging tasks and handle them all correctly. llvm-svn: 125815
2011-02-18 09:44:25 +08:00
// InstanceSettingsSP setting_sp = pos->second;
// setting_sp->UpdateInstanceSettingsVariable (const_var_name, index_value, value,
// instance_name, *entry, op, err, true);
// }
}
}
}
}
else
{
// Either a child's setting or an instance setting.
if (names.GetArgumentAtIndex(0)[0] == '[')
{
// An instance setting. Supposedly.
ConstString instance_name (names.GetArgumentAtIndex (0));
// First verify that there is only one more name.
names.Shift();
if (names.GetArgumentCount() != 1)
{
err.SetErrorStringWithFormat ("Invalid variable name format '%s'; cannot assign value.\n",
full_dot_name);
return err;
}
// Next verify that it is a valid instance setting name.
const_var_name.SetCString (names.GetArgumentAtIndex (0));
const SettingEntry *entry = GetInstanceEntry (const_var_name);
if (entry == NULL)
{
err.SetErrorStringWithFormat ("Unknown instance variable '%s'; cannot assign value.\n",
const_var_name.GetCString());
return err;
}
UserSettingsController::VerifyOperationForType (entry->var_type, op, const_var_name, err);
if (err.Fail())
return err;
if ((value == NULL || value[0] == '\0')
Added new target instance settings for execution settings: Targets can now specify some additional parameters for when we debug executables that can help with plug-in selection: target.execution-level = auto | user | kernel target.execution-mode = auto | dynamic | static target.execution-os-type = auto | none | halted | live On some systems, the binaries that are created are the same wether you use them to debug a kernel, or a user space program. Many times inspecting an object file can reveal what an executable should be. For these cases we can now be a little more complete by specifying wether to detect all of these things automatically (inspect the main executable file and select a plug-in accordingly), or manually to force the selection of certain plug-ins. To do this we now allow the specficifation of wether one is debugging a user space program (target.execution-level = user) or a kernel program (target.execution-level = kernel). We can also specify if we want to debug a program where shared libraries are dynamically loaded using a DynamicLoader plug-in (target.execution-mode = dynamic), or wether we will treat all symbol files as already linked at the correct address (target.execution-mode = static). We can also specify if the inferior we are debugging is being debugged on a bare board (target.execution-os-type = none), or debugging an OS where we have a JTAG or other direct connection to the inferior stops the entire OS (target.execution-os-type = halted), or if we are debugging a program on something that has live debug services (target.execution-os-type = live). For the "target.execution-os-type = halted" mode, we will need to create ProcessHelper plug-ins that allow us to extract the process/thread and other OS information by reading/writing memory. This should allow LLDB to be used for a wide variety of debugging tasks and handle them all correctly. llvm-svn: 125815
2011-02-18 09:44:25 +08:00
&& (op == eVarSetOperationAssign))
{
Added new target instance settings for execution settings: Targets can now specify some additional parameters for when we debug executables that can help with plug-in selection: target.execution-level = auto | user | kernel target.execution-mode = auto | dynamic | static target.execution-os-type = auto | none | halted | live On some systems, the binaries that are created are the same wether you use them to debug a kernel, or a user space program. Many times inspecting an object file can reveal what an executable should be. For these cases we can now be a little more complete by specifying wether to detect all of these things automatically (inspect the main executable file and select a plug-in accordingly), or manually to force the selection of certain plug-ins. To do this we now allow the specficifation of wether one is debugging a user space program (target.execution-level = user) or a kernel program (target.execution-level = kernel). We can also specify if we want to debug a program where shared libraries are dynamically loaded using a DynamicLoader plug-in (target.execution-mode = dynamic), or wether we will treat all symbol files as already linked at the correct address (target.execution-mode = static). We can also specify if the inferior we are debugging is being debugged on a bare board (target.execution-os-type = none), or debugging an OS where we have a JTAG or other direct connection to the inferior stops the entire OS (target.execution-os-type = halted), or if we are debugging a program on something that has live debug services (target.execution-os-type = live). For the "target.execution-os-type = halted" mode, we will need to create ProcessHelper plug-ins that allow us to extract the process/thread and other OS information by reading/writing memory. This should allow LLDB to be used for a wide variety of debugging tasks and handle them all correctly. llvm-svn: 125815
2011-02-18 09:44:25 +08:00
if (entry->var_type != eSetVarTypeEnum)
value = entry->default_value;
else
value = entry->enum_values[0].string_value;
}
// Now look for existing instance with given instance name; if not found, find or create pending
// setting for instance with given name.
InstanceSettings *current_settings = FindSettingsForInstance (instance_name);
if (current_settings != NULL)
{
current_settings->UpdateInstanceSettingsVariable (const_var_name, index_value, value,
instance_name, *entry, op, err, false);
}
else
{
// Instance does not currently exist; make or update a pending setting for it.
Added new target instance settings for execution settings: Targets can now specify some additional parameters for when we debug executables that can help with plug-in selection: target.execution-level = auto | user | kernel target.execution-mode = auto | dynamic | static target.execution-os-type = auto | none | halted | live On some systems, the binaries that are created are the same wether you use them to debug a kernel, or a user space program. Many times inspecting an object file can reveal what an executable should be. For these cases we can now be a little more complete by specifying wether to detect all of these things automatically (inspect the main executable file and select a plug-in accordingly), or manually to force the selection of certain plug-ins. To do this we now allow the specficifation of wether one is debugging a user space program (target.execution-level = user) or a kernel program (target.execution-level = kernel). We can also specify if we want to debug a program where shared libraries are dynamically loaded using a DynamicLoader plug-in (target.execution-mode = dynamic), or wether we will treat all symbol files as already linked at the correct address (target.execution-mode = static). We can also specify if the inferior we are debugging is being debugged on a bare board (target.execution-os-type = none), or debugging an OS where we have a JTAG or other direct connection to the inferior stops the entire OS (target.execution-os-type = halted), or if we are debugging a program on something that has live debug services (target.execution-os-type = live). For the "target.execution-os-type = halted" mode, we will need to create ProcessHelper plug-ins that allow us to extract the process/thread and other OS information by reading/writing memory. This should allow LLDB to be used for a wide variety of debugging tasks and handle them all correctly. llvm-svn: 125815
2011-02-18 09:44:25 +08:00
InstanceSettingsSP current_settings_sp = PendingSettingsForInstance (instance_name);
// Now we have a settings record, update it appropriately.
current_settings_sp->UpdateInstanceSettingsVariable (const_var_name, index_value, value,
instance_name, *entry, op, err, true);
{ // Scope for mutex.
Mutex::Locker locker (m_pending_settings_mutex);
m_pending_settings[instance_name.GetCString()] = current_settings_sp;
}
if (override)
{
OverrideAllInstances (const_var_name, value, op, index_value, err);
// Update all pending records as well.
Added new target instance settings for execution settings: Targets can now specify some additional parameters for when we debug executables that can help with plug-in selection: target.execution-level = auto | user | kernel target.execution-mode = auto | dynamic | static target.execution-os-type = auto | none | halted | live On some systems, the binaries that are created are the same wether you use them to debug a kernel, or a user space program. Many times inspecting an object file can reveal what an executable should be. For these cases we can now be a little more complete by specifying wether to detect all of these things automatically (inspect the main executable file and select a plug-in accordingly), or manually to force the selection of certain plug-ins. To do this we now allow the specficifation of wether one is debugging a user space program (target.execution-level = user) or a kernel program (target.execution-level = kernel). We can also specify if we want to debug a program where shared libraries are dynamically loaded using a DynamicLoader plug-in (target.execution-mode = dynamic), or wether we will treat all symbol files as already linked at the correct address (target.execution-mode = static). We can also specify if the inferior we are debugging is being debugged on a bare board (target.execution-os-type = none), or debugging an OS where we have a JTAG or other direct connection to the inferior stops the entire OS (target.execution-os-type = halted), or if we are debugging a program on something that has live debug services (target.execution-os-type = live). For the "target.execution-os-type = halted" mode, we will need to create ProcessHelper plug-ins that allow us to extract the process/thread and other OS information by reading/writing memory. This should allow LLDB to be used for a wide variety of debugging tasks and handle them all correctly. llvm-svn: 125815
2011-02-18 09:44:25 +08:00
std::map<std::string, InstanceSettingsSP>::iterator pos;
std::map<std::string, InstanceSettingsSP>::iterator end = m_pending_settings.end();
for (pos = m_pending_settings.begin(); pos != end; end++)
{
const ConstString tmp_inst_name (pos->first.c_str());
Added new target instance settings for execution settings: Targets can now specify some additional parameters for when we debug executables that can help with plug-in selection: target.execution-level = auto | user | kernel target.execution-mode = auto | dynamic | static target.execution-os-type = auto | none | halted | live On some systems, the binaries that are created are the same wether you use them to debug a kernel, or a user space program. Many times inspecting an object file can reveal what an executable should be. For these cases we can now be a little more complete by specifying wether to detect all of these things automatically (inspect the main executable file and select a plug-in accordingly), or manually to force the selection of certain plug-ins. To do this we now allow the specficifation of wether one is debugging a user space program (target.execution-level = user) or a kernel program (target.execution-level = kernel). We can also specify if we want to debug a program where shared libraries are dynamically loaded using a DynamicLoader plug-in (target.execution-mode = dynamic), or wether we will treat all symbol files as already linked at the correct address (target.execution-mode = static). We can also specify if the inferior we are debugging is being debugged on a bare board (target.execution-os-type = none), or debugging an OS where we have a JTAG or other direct connection to the inferior stops the entire OS (target.execution-os-type = halted), or if we are debugging a program on something that has live debug services (target.execution-os-type = live). For the "target.execution-os-type = halted" mode, we will need to create ProcessHelper plug-ins that allow us to extract the process/thread and other OS information by reading/writing memory. This should allow LLDB to be used for a wide variety of debugging tasks and handle them all correctly. llvm-svn: 125815
2011-02-18 09:44:25 +08:00
InstanceSettingsSP setting_sp = pos->second;
setting_sp->UpdateInstanceSettingsVariable (const_var_name, index_value, value,
tmp_inst_name, *entry, op, err, true);
}
}
}
}
else
{
// A child setting.
Added new target instance settings for execution settings: Targets can now specify some additional parameters for when we debug executables that can help with plug-in selection: target.execution-level = auto | user | kernel target.execution-mode = auto | dynamic | static target.execution-os-type = auto | none | halted | live On some systems, the binaries that are created are the same wether you use them to debug a kernel, or a user space program. Many times inspecting an object file can reveal what an executable should be. For these cases we can now be a little more complete by specifying wether to detect all of these things automatically (inspect the main executable file and select a plug-in accordingly), or manually to force the selection of certain plug-ins. To do this we now allow the specficifation of wether one is debugging a user space program (target.execution-level = user) or a kernel program (target.execution-level = kernel). We can also specify if we want to debug a program where shared libraries are dynamically loaded using a DynamicLoader plug-in (target.execution-mode = dynamic), or wether we will treat all symbol files as already linked at the correct address (target.execution-mode = static). We can also specify if the inferior we are debugging is being debugged on a bare board (target.execution-os-type = none), or debugging an OS where we have a JTAG or other direct connection to the inferior stops the entire OS (target.execution-os-type = halted), or if we are debugging a program on something that has live debug services (target.execution-os-type = live). For the "target.execution-os-type = halted" mode, we will need to create ProcessHelper plug-ins that allow us to extract the process/thread and other OS information by reading/writing memory. This should allow LLDB to be used for a wide variety of debugging tasks and handle them all correctly. llvm-svn: 125815
2011-02-18 09:44:25 +08:00
UserSettingsControllerSP child;
ConstString child_prefix (names.GetArgumentAtIndex (0));
int num_children = GetNumChildren();
bool found = false;
for (int i = 0; i < num_children && !found; ++i)
{
child = GetChildAtIndex (i);
ConstString current_prefix = child->GetLevelName();
if (current_prefix == child_prefix)
{
found = true;
std::string new_name;
for (int j = 0; j < names.GetArgumentCount(); ++j)
{
if (j > 0)
new_name += '.';
new_name += names.GetArgumentAtIndex (j);
}
return child->SetVariable (new_name.c_str(), value, op, override, debugger_instance_name,
index_value);
}
}
if (!found)
{
err.SetErrorStringWithFormat ("Unable to find variable '%s'; cannot assign value.\n",
full_dot_name);
return err;
}
}
}
}
else
{
err.SetErrorStringWithFormat ("'%s' is not a valid level name; was expecting '%s'. Cannot assign value.\n",
prefix.GetCString(), m_settings.level_name.GetCString());
}
return err;
}
StringList
Added code that will allow completely customizable prompts for use in replacing the "(lldb)" prompt, the "frame #1..." displays when doing stack backtracing and the "thread #1....". This will allow you to see exactly the information that you want to see where you want to see it. This currently isn't hookup up to the prompts yet, but it will be soon. So what is the format of the prompts? Prompts can contain variables that have access to the current program state. Variables are text that appears in between a prefix of "${" and ends with a "}". Some of the interesting variables include: // The frame index (0, 1, 2, 3...) ${frame.index} // common frame registers with generic names ${frame.pc} ${frame.sp} ${frame.fp} ${frame.ra} ${frame.flags} // Access to any frame registers by name where REGNAME is any register name: ${frame.reg.REGNAME} // The current compile unit file where the frame is located ${file.basename} ${file.fullpath} // Function information ${function.name} ${function.pc-offset} // Process info ${process.file.basename} ${process.file.fullpath} ${process.id} ${process.name} // Thread info ${thread.id} ${thread.index} ${thread.name} ${thread.queue} ${thread.stop-reason} // Target information ${target.arch} // The current module for the current frame (the shared library or executable // that contains the current frame PC value): ${module.file.basename} ${module.file.fullpath} // Access to the line entry for where the current frame is when your thread // is stopped: ${line.file.basename} ${line.file.fullpath} ${line.number} ${line.start-addr} ${line.end-addr} Many times the information that you might have in your prompt might not be available and you won't want it to print out if it isn't valid. To take care of this you can enclose everything that must resolve into a scope. A scope is starts with '{' and ends with '}'. For example in order to only display the current file and line number when the information is available the format would be: "{ at {$line.file.basename}:${line.number}}" Broken down this is: start the scope: "{" format whose content will only be displayed if all information is available: "at {$line.file.basename}:${line.number}" end the scope: "}" We currently can represent the infomration we see when stopped at a frame: frame #0: 0x0000000100000e85 a.out`main + 4 at test.c:19 with the following format: "frame #${frame.index}: ${frame.pc} {${module.file.basename}`}{${function.name}{${function.pc-offset}}{ at ${line.file.basename}:${line.number}}\n" This breaks down to always print: "frame #${frame.index}: ${frame.pc} " only print the module followed by a tick if we have a valid module: "{${module.file.basename}`}" print the function name with optional offset: "{${function.name}{${function.pc-offset}}" print the line info if it is available: "{ at ${line.file.basename}:${line.number}}" then finish off with a newline: "\n" Notice you can also put newlines ("\n") and tabs and everything else you are used to putting in a format string when desensitized with the \ character. Cleaned up some of the user settings controller subclasses. All of them do not have any global settings variables and were all implementing stubs for the get/set global settings variable. Now there is a default version in UserSettingsController that will do nothing. llvm-svn: 114306
2010-09-19 10:33:57 +08:00
UserSettingsController::GetVariable
(
const char *full_dot_name,
Added new target instance settings for execution settings: Targets can now specify some additional parameters for when we debug executables that can help with plug-in selection: target.execution-level = auto | user | kernel target.execution-mode = auto | dynamic | static target.execution-os-type = auto | none | halted | live On some systems, the binaries that are created are the same wether you use them to debug a kernel, or a user space program. Many times inspecting an object file can reveal what an executable should be. For these cases we can now be a little more complete by specifying wether to detect all of these things automatically (inspect the main executable file and select a plug-in accordingly), or manually to force the selection of certain plug-ins. To do this we now allow the specficifation of wether one is debugging a user space program (target.execution-level = user) or a kernel program (target.execution-level = kernel). We can also specify if we want to debug a program where shared libraries are dynamically loaded using a DynamicLoader plug-in (target.execution-mode = dynamic), or wether we will treat all symbol files as already linked at the correct address (target.execution-mode = static). We can also specify if the inferior we are debugging is being debugged on a bare board (target.execution-os-type = none), or debugging an OS where we have a JTAG or other direct connection to the inferior stops the entire OS (target.execution-os-type = halted), or if we are debugging a program on something that has live debug services (target.execution-os-type = live). For the "target.execution-os-type = halted" mode, we will need to create ProcessHelper plug-ins that allow us to extract the process/thread and other OS information by reading/writing memory. This should allow LLDB to be used for a wide variety of debugging tasks and handle them all correctly. llvm-svn: 125815
2011-02-18 09:44:25 +08:00
SettableVariableType &var_type,
const char *debugger_instance_name,
Error &err
Added code that will allow completely customizable prompts for use in replacing the "(lldb)" prompt, the "frame #1..." displays when doing stack backtracing and the "thread #1....". This will allow you to see exactly the information that you want to see where you want to see it. This currently isn't hookup up to the prompts yet, but it will be soon. So what is the format of the prompts? Prompts can contain variables that have access to the current program state. Variables are text that appears in between a prefix of "${" and ends with a "}". Some of the interesting variables include: // The frame index (0, 1, 2, 3...) ${frame.index} // common frame registers with generic names ${frame.pc} ${frame.sp} ${frame.fp} ${frame.ra} ${frame.flags} // Access to any frame registers by name where REGNAME is any register name: ${frame.reg.REGNAME} // The current compile unit file where the frame is located ${file.basename} ${file.fullpath} // Function information ${function.name} ${function.pc-offset} // Process info ${process.file.basename} ${process.file.fullpath} ${process.id} ${process.name} // Thread info ${thread.id} ${thread.index} ${thread.name} ${thread.queue} ${thread.stop-reason} // Target information ${target.arch} // The current module for the current frame (the shared library or executable // that contains the current frame PC value): ${module.file.basename} ${module.file.fullpath} // Access to the line entry for where the current frame is when your thread // is stopped: ${line.file.basename} ${line.file.fullpath} ${line.number} ${line.start-addr} ${line.end-addr} Many times the information that you might have in your prompt might not be available and you won't want it to print out if it isn't valid. To take care of this you can enclose everything that must resolve into a scope. A scope is starts with '{' and ends with '}'. For example in order to only display the current file and line number when the information is available the format would be: "{ at {$line.file.basename}:${line.number}}" Broken down this is: start the scope: "{" format whose content will only be displayed if all information is available: "at {$line.file.basename}:${line.number}" end the scope: "}" We currently can represent the infomration we see when stopped at a frame: frame #0: 0x0000000100000e85 a.out`main + 4 at test.c:19 with the following format: "frame #${frame.index}: ${frame.pc} {${module.file.basename}`}{${function.name}{${function.pc-offset}}{ at ${line.file.basename}:${line.number}}\n" This breaks down to always print: "frame #${frame.index}: ${frame.pc} " only print the module followed by a tick if we have a valid module: "{${module.file.basename}`}" print the function name with optional offset: "{${function.name}{${function.pc-offset}}" print the line info if it is available: "{ at ${line.file.basename}:${line.number}}" then finish off with a newline: "\n" Notice you can also put newlines ("\n") and tabs and everything else you are used to putting in a format string when desensitized with the \ character. Cleaned up some of the user settings controller subclasses. All of them do not have any global settings variables and were all implementing stubs for the get/set global settings variable. Now there is a default version in UserSettingsController that will do nothing. llvm-svn: 114306
2010-09-19 10:33:57 +08:00
)
{
Args names = UserSettingsController::BreakNameIntoPieces (full_dot_name);
ConstString const_var_name;
StringList value;
int num_pieces = names.GetArgumentCount();
ConstString prefix (names.GetArgumentAtIndex (0));
const_var_name.SetCString (names.GetArgumentAtIndex (num_pieces - 1));
const SettingEntry *global_entry = GetGlobalEntry (const_var_name);
const SettingEntry *instance_entry = GetInstanceEntry (const_var_name);
if ((prefix != m_settings.level_name)
&& (m_settings.level_name.GetLength () > 0))
{
err.SetErrorString ("Invalid variable name");
return value;
}
// prefix name matched; remove it from names.
if (m_settings.level_name.GetLength() > 0)
names.Shift();
// Should we pass this off to a child? If there is more than one name piece left, and the next name piece
// matches a child prefix, then yes.
Added new target instance settings for execution settings: Targets can now specify some additional parameters for when we debug executables that can help with plug-in selection: target.execution-level = auto | user | kernel target.execution-mode = auto | dynamic | static target.execution-os-type = auto | none | halted | live On some systems, the binaries that are created are the same wether you use them to debug a kernel, or a user space program. Many times inspecting an object file can reveal what an executable should be. For these cases we can now be a little more complete by specifying wether to detect all of these things automatically (inspect the main executable file and select a plug-in accordingly), or manually to force the selection of certain plug-ins. To do this we now allow the specficifation of wether one is debugging a user space program (target.execution-level = user) or a kernel program (target.execution-level = kernel). We can also specify if we want to debug a program where shared libraries are dynamically loaded using a DynamicLoader plug-in (target.execution-mode = dynamic), or wether we will treat all symbol files as already linked at the correct address (target.execution-mode = static). We can also specify if the inferior we are debugging is being debugged on a bare board (target.execution-os-type = none), or debugging an OS where we have a JTAG or other direct connection to the inferior stops the entire OS (target.execution-os-type = halted), or if we are debugging a program on something that has live debug services (target.execution-os-type = live). For the "target.execution-os-type = halted" mode, we will need to create ProcessHelper plug-ins that allow us to extract the process/thread and other OS information by reading/writing memory. This should allow LLDB to be used for a wide variety of debugging tasks and handle them all correctly. llvm-svn: 125815
2011-02-18 09:44:25 +08:00
UserSettingsControllerSP child;
if (names.GetArgumentCount() > 1)
{
ConstString child_prefix (names.GetArgumentAtIndex (0));
bool found = false;
for (int i = 0; i < m_children.size() && !found; ++i)
{
if (child_prefix == m_children[i]->GetLevelName())
{
found = true;
child = m_children[i];
std::string new_name;
for (int j = 0; j < names.GetArgumentCount(); ++j)
{
if (j > 0)
new_name += '.';
new_name += names.GetArgumentAtIndex (j);
}
return child->GetVariable (new_name.c_str(), var_type, debugger_instance_name, err);
}
}
if (!found)
{
// Cannot be handled by a child, because name did not match any child prefixes.
// Cannot be a class-wide variable because there are too many name pieces.
if (instance_entry != NULL)
{
var_type = instance_entry->var_type;
ConstString instance_name (names.GetArgumentAtIndex (0));
InstanceSettings *current_settings = FindSettingsForInstance (instance_name);
if (current_settings != NULL)
{
current_settings->GetInstanceSettingsValue (*instance_entry, const_var_name, value, &err);
}
else
{
// Look for instance name setting in pending settings.
std::string inst_name_str = instance_name.GetCString();
Added new target instance settings for execution settings: Targets can now specify some additional parameters for when we debug executables that can help with plug-in selection: target.execution-level = auto | user | kernel target.execution-mode = auto | dynamic | static target.execution-os-type = auto | none | halted | live On some systems, the binaries that are created are the same wether you use them to debug a kernel, or a user space program. Many times inspecting an object file can reveal what an executable should be. For these cases we can now be a little more complete by specifying wether to detect all of these things automatically (inspect the main executable file and select a plug-in accordingly), or manually to force the selection of certain plug-ins. To do this we now allow the specficifation of wether one is debugging a user space program (target.execution-level = user) or a kernel program (target.execution-level = kernel). We can also specify if we want to debug a program where shared libraries are dynamically loaded using a DynamicLoader plug-in (target.execution-mode = dynamic), or wether we will treat all symbol files as already linked at the correct address (target.execution-mode = static). We can also specify if the inferior we are debugging is being debugged on a bare board (target.execution-os-type = none), or debugging an OS where we have a JTAG or other direct connection to the inferior stops the entire OS (target.execution-os-type = halted), or if we are debugging a program on something that has live debug services (target.execution-os-type = live). For the "target.execution-os-type = halted" mode, we will need to create ProcessHelper plug-ins that allow us to extract the process/thread and other OS information by reading/writing memory. This should allow LLDB to be used for a wide variety of debugging tasks and handle them all correctly. llvm-svn: 125815
2011-02-18 09:44:25 +08:00
std::map<std::string, InstanceSettingsSP>::iterator pos;
pos = m_pending_settings.find (inst_name_str);
if (pos != m_pending_settings.end())
{
Added new target instance settings for execution settings: Targets can now specify some additional parameters for when we debug executables that can help with plug-in selection: target.execution-level = auto | user | kernel target.execution-mode = auto | dynamic | static target.execution-os-type = auto | none | halted | live On some systems, the binaries that are created are the same wether you use them to debug a kernel, or a user space program. Many times inspecting an object file can reveal what an executable should be. For these cases we can now be a little more complete by specifying wether to detect all of these things automatically (inspect the main executable file and select a plug-in accordingly), or manually to force the selection of certain plug-ins. To do this we now allow the specficifation of wether one is debugging a user space program (target.execution-level = user) or a kernel program (target.execution-level = kernel). We can also specify if we want to debug a program where shared libraries are dynamically loaded using a DynamicLoader plug-in (target.execution-mode = dynamic), or wether we will treat all symbol files as already linked at the correct address (target.execution-mode = static). We can also specify if the inferior we are debugging is being debugged on a bare board (target.execution-os-type = none), or debugging an OS where we have a JTAG or other direct connection to the inferior stops the entire OS (target.execution-os-type = halted), or if we are debugging a program on something that has live debug services (target.execution-os-type = live). For the "target.execution-os-type = halted" mode, we will need to create ProcessHelper plug-ins that allow us to extract the process/thread and other OS information by reading/writing memory. This should allow LLDB to be used for a wide variety of debugging tasks and handle them all correctly. llvm-svn: 125815
2011-02-18 09:44:25 +08:00
InstanceSettingsSP settings_sp = pos->second;
settings_sp->GetInstanceSettingsValue (*instance_entry, const_var_name, value, &err);
}
else
{
if (m_settings.level_name.GetLength() > 0)
{
// No valid instance name; assume they want the default settings.
m_default_settings->GetInstanceSettingsValue (*instance_entry, const_var_name, value, &err);
}
else
{
// We're at the Debugger level; use the debugger's instance settings.
StreamString tmp_name;
if (debugger_instance_name[0] != '[')
tmp_name.Printf ("[%s]", debugger_instance_name);
else
tmp_name.Printf ("%s", debugger_instance_name);
ConstString dbg_name (debugger_instance_name);
InstanceSettings *dbg_settings = FindSettingsForInstance (dbg_name);
if (dbg_settings)
dbg_settings->GetInstanceSettingsValue (*instance_entry, const_var_name, value, &err);
}
}
}
}
else
err.SetErrorString ("Invalid variable name");
}
}
else
{
// Only one name left. It must belong to the current level, or be an error.
if ((global_entry == NULL)
&& (instance_entry == NULL))
{
err.SetErrorString ("Invalid variable name");
}
else if (global_entry)
{
var_type = global_entry->var_type;
GetGlobalVariable (const_var_name, value, err);
}
else if (instance_entry)
{
var_type = instance_entry->var_type;
if (m_settings.level_name.GetLength() > 0)
m_default_settings->GetInstanceSettingsValue (*instance_entry, const_var_name, value, &err);
else
{
// We're at the Debugger level; use the debugger's instance settings.
StreamString tmp_name;
if (debugger_instance_name[0] != '[')
tmp_name.Printf ("[%s]", debugger_instance_name);
else
tmp_name.Printf ("%s", debugger_instance_name);
ConstString dbg_name (tmp_name.GetData());
InstanceSettings *dbg_settings = FindSettingsForInstance (dbg_name);
if (dbg_settings)
dbg_settings->GetInstanceSettingsValue (*instance_entry, const_var_name, value, &err);
}
}
}
return value;
}
void
UserSettingsController::RemovePendingSettings (const ConstString &instance_name)
{
StreamString tmp_name;
// Add surrounding brackets to instance name if not already present.
if (instance_name.GetCString()[0] != '[')
tmp_name.Printf ("[%s]", instance_name.GetCString());
else
tmp_name.Printf ("%s", instance_name.GetCString());
std::string instance_name_str (tmp_name.GetData());
Added new target instance settings for execution settings: Targets can now specify some additional parameters for when we debug executables that can help with plug-in selection: target.execution-level = auto | user | kernel target.execution-mode = auto | dynamic | static target.execution-os-type = auto | none | halted | live On some systems, the binaries that are created are the same wether you use them to debug a kernel, or a user space program. Many times inspecting an object file can reveal what an executable should be. For these cases we can now be a little more complete by specifying wether to detect all of these things automatically (inspect the main executable file and select a plug-in accordingly), or manually to force the selection of certain plug-ins. To do this we now allow the specficifation of wether one is debugging a user space program (target.execution-level = user) or a kernel program (target.execution-level = kernel). We can also specify if we want to debug a program where shared libraries are dynamically loaded using a DynamicLoader plug-in (target.execution-mode = dynamic), or wether we will treat all symbol files as already linked at the correct address (target.execution-mode = static). We can also specify if the inferior we are debugging is being debugged on a bare board (target.execution-os-type = none), or debugging an OS where we have a JTAG or other direct connection to the inferior stops the entire OS (target.execution-os-type = halted), or if we are debugging a program on something that has live debug services (target.execution-os-type = live). For the "target.execution-os-type = halted" mode, we will need to create ProcessHelper plug-ins that allow us to extract the process/thread and other OS information by reading/writing memory. This should allow LLDB to be used for a wide variety of debugging tasks and handle them all correctly. llvm-svn: 125815
2011-02-18 09:44:25 +08:00
std::map<std::string, InstanceSettingsSP>::iterator pos;
Mutex::Locker locker (m_pending_settings_mutex);
m_pending_settings.erase (instance_name_str);
}
Added new target instance settings for execution settings: Targets can now specify some additional parameters for when we debug executables that can help with plug-in selection: target.execution-level = auto | user | kernel target.execution-mode = auto | dynamic | static target.execution-os-type = auto | none | halted | live On some systems, the binaries that are created are the same wether you use them to debug a kernel, or a user space program. Many times inspecting an object file can reveal what an executable should be. For these cases we can now be a little more complete by specifying wether to detect all of these things automatically (inspect the main executable file and select a plug-in accordingly), or manually to force the selection of certain plug-ins. To do this we now allow the specficifation of wether one is debugging a user space program (target.execution-level = user) or a kernel program (target.execution-level = kernel). We can also specify if we want to debug a program where shared libraries are dynamically loaded using a DynamicLoader plug-in (target.execution-mode = dynamic), or wether we will treat all symbol files as already linked at the correct address (target.execution-mode = static). We can also specify if the inferior we are debugging is being debugged on a bare board (target.execution-os-type = none), or debugging an OS where we have a JTAG or other direct connection to the inferior stops the entire OS (target.execution-os-type = halted), or if we are debugging a program on something that has live debug services (target.execution-os-type = live). For the "target.execution-os-type = halted" mode, we will need to create ProcessHelper plug-ins that allow us to extract the process/thread and other OS information by reading/writing memory. This should allow LLDB to be used for a wide variety of debugging tasks and handle them all correctly. llvm-svn: 125815
2011-02-18 09:44:25 +08:00
const InstanceSettingsSP &
UserSettingsController::FindPendingSettings (const ConstString &instance_name)
{
Added new target instance settings for execution settings: Targets can now specify some additional parameters for when we debug executables that can help with plug-in selection: target.execution-level = auto | user | kernel target.execution-mode = auto | dynamic | static target.execution-os-type = auto | none | halted | live On some systems, the binaries that are created are the same wether you use them to debug a kernel, or a user space program. Many times inspecting an object file can reveal what an executable should be. For these cases we can now be a little more complete by specifying wether to detect all of these things automatically (inspect the main executable file and select a plug-in accordingly), or manually to force the selection of certain plug-ins. To do this we now allow the specficifation of wether one is debugging a user space program (target.execution-level = user) or a kernel program (target.execution-level = kernel). We can also specify if we want to debug a program where shared libraries are dynamically loaded using a DynamicLoader plug-in (target.execution-mode = dynamic), or wether we will treat all symbol files as already linked at the correct address (target.execution-mode = static). We can also specify if the inferior we are debugging is being debugged on a bare board (target.execution-os-type = none), or debugging an OS where we have a JTAG or other direct connection to the inferior stops the entire OS (target.execution-os-type = halted), or if we are debugging a program on something that has live debug services (target.execution-os-type = live). For the "target.execution-os-type = halted" mode, we will need to create ProcessHelper plug-ins that allow us to extract the process/thread and other OS information by reading/writing memory. This should allow LLDB to be used for a wide variety of debugging tasks and handle them all correctly. llvm-svn: 125815
2011-02-18 09:44:25 +08:00
std::map<std::string, InstanceSettingsSP>::iterator pos;
StreamString tmp_name;
// Add surrounding brackets to instance name if not already present.
if (instance_name.GetCString()[0] != '[')
tmp_name.Printf ("[%s]", instance_name.GetCString());
else
tmp_name.Printf ("%s", instance_name.GetCString());
std::string instance_name_str (tmp_name.GetData()); // Need std::string for std::map look-up
{ // Scope for mutex.
Mutex::Locker locker (m_pending_settings_mutex);
pos = m_pending_settings.find (instance_name_str);
if (pos != m_pending_settings.end())
return pos->second;
}
return m_default_settings;
}
void
UserSettingsController::CreateDefaultInstanceSettings ()
{
Error err;
const ConstString &default_instance_name = InstanceSettings::GetDefaultName();
for (int i = 0; i < m_settings.instance_settings.size(); ++i)
{
SettingEntry &entry = m_settings.instance_settings[i];
ConstString var_name (entry.var_name);
const char *default_value = entry.default_value;
// If there is no default value, then use the first enumeration value
// as the default value
if (default_value == NULL && entry.var_type == eSetVarTypeEnum)
default_value = entry.enum_values[0].string_value;
if (default_value != NULL)
m_default_settings->UpdateInstanceSettingsVariable (var_name,
NULL,
default_value,
default_instance_name,
entry,
eVarSetOperationAssign,
err,
true);
}
}
void
Added new target instance settings for execution settings: Targets can now specify some additional parameters for when we debug executables that can help with plug-in selection: target.execution-level = auto | user | kernel target.execution-mode = auto | dynamic | static target.execution-os-type = auto | none | halted | live On some systems, the binaries that are created are the same wether you use them to debug a kernel, or a user space program. Many times inspecting an object file can reveal what an executable should be. For these cases we can now be a little more complete by specifying wether to detect all of these things automatically (inspect the main executable file and select a plug-in accordingly), or manually to force the selection of certain plug-ins. To do this we now allow the specficifation of wether one is debugging a user space program (target.execution-level = user) or a kernel program (target.execution-level = kernel). We can also specify if we want to debug a program where shared libraries are dynamically loaded using a DynamicLoader plug-in (target.execution-mode = dynamic), or wether we will treat all symbol files as already linked at the correct address (target.execution-mode = static). We can also specify if the inferior we are debugging is being debugged on a bare board (target.execution-os-type = none), or debugging an OS where we have a JTAG or other direct connection to the inferior stops the entire OS (target.execution-os-type = halted), or if we are debugging a program on something that has live debug services (target.execution-os-type = live). For the "target.execution-os-type = halted" mode, we will need to create ProcessHelper plug-ins that allow us to extract the process/thread and other OS information by reading/writing memory. This should allow LLDB to be used for a wide variety of debugging tasks and handle them all correctly. llvm-svn: 125815
2011-02-18 09:44:25 +08:00
UserSettingsController::CopyDefaultSettings (const InstanceSettingsSP &actual_settings,
const ConstString &instance_name,
bool pending)
{
Error err;
for (int i = 0; i < m_settings.instance_settings.size(); ++i)
{
SettingEntry &entry = m_settings.instance_settings[i];
ConstString var_name (entry.var_name);
StringList value;
m_default_settings->GetInstanceSettingsValue (entry, var_name, value, NULL);
std::string value_str;
if (value.GetSize() == 1)
value_str.append (value.GetStringAtIndex (0));
else if (value.GetSize() > 1)
{
for (int j = 0; j < value.GetSize(); ++j)
{
if (j > 0)
value_str.append (" ");
value_str.append (value.GetStringAtIndex (j));
}
}
actual_settings->UpdateInstanceSettingsVariable (var_name, NULL, value_str.c_str(), instance_name, entry,
Added new target instance settings for execution settings: Targets can now specify some additional parameters for when we debug executables that can help with plug-in selection: target.execution-level = auto | user | kernel target.execution-mode = auto | dynamic | static target.execution-os-type = auto | none | halted | live On some systems, the binaries that are created are the same wether you use them to debug a kernel, or a user space program. Many times inspecting an object file can reveal what an executable should be. For these cases we can now be a little more complete by specifying wether to detect all of these things automatically (inspect the main executable file and select a plug-in accordingly), or manually to force the selection of certain plug-ins. To do this we now allow the specficifation of wether one is debugging a user space program (target.execution-level = user) or a kernel program (target.execution-level = kernel). We can also specify if we want to debug a program where shared libraries are dynamically loaded using a DynamicLoader plug-in (target.execution-mode = dynamic), or wether we will treat all symbol files as already linked at the correct address (target.execution-mode = static). We can also specify if the inferior we are debugging is being debugged on a bare board (target.execution-os-type = none), or debugging an OS where we have a JTAG or other direct connection to the inferior stops the entire OS (target.execution-os-type = halted), or if we are debugging a program on something that has live debug services (target.execution-os-type = live). For the "target.execution-os-type = halted" mode, we will need to create ProcessHelper plug-ins that allow us to extract the process/thread and other OS information by reading/writing memory. This should allow LLDB to be used for a wide variety of debugging tasks and handle them all correctly. llvm-svn: 125815
2011-02-18 09:44:25 +08:00
eVarSetOperationAssign, err, pending);
}
}
Added new target instance settings for execution settings: Targets can now specify some additional parameters for when we debug executables that can help with plug-in selection: target.execution-level = auto | user | kernel target.execution-mode = auto | dynamic | static target.execution-os-type = auto | none | halted | live On some systems, the binaries that are created are the same wether you use them to debug a kernel, or a user space program. Many times inspecting an object file can reveal what an executable should be. For these cases we can now be a little more complete by specifying wether to detect all of these things automatically (inspect the main executable file and select a plug-in accordingly), or manually to force the selection of certain plug-ins. To do this we now allow the specficifation of wether one is debugging a user space program (target.execution-level = user) or a kernel program (target.execution-level = kernel). We can also specify if we want to debug a program where shared libraries are dynamically loaded using a DynamicLoader plug-in (target.execution-mode = dynamic), or wether we will treat all symbol files as already linked at the correct address (target.execution-mode = static). We can also specify if the inferior we are debugging is being debugged on a bare board (target.execution-os-type = none), or debugging an OS where we have a JTAG or other direct connection to the inferior stops the entire OS (target.execution-os-type = halted), or if we are debugging a program on something that has live debug services (target.execution-os-type = live). For the "target.execution-os-type = halted" mode, we will need to create ProcessHelper plug-ins that allow us to extract the process/thread and other OS information by reading/writing memory. This should allow LLDB to be used for a wide variety of debugging tasks and handle them all correctly. llvm-svn: 125815
2011-02-18 09:44:25 +08:00
InstanceSettingsSP
UserSettingsController::PendingSettingsForInstance (const ConstString &instance_name)
{
std::string name_str (instance_name.GetCString());
Added new target instance settings for execution settings: Targets can now specify some additional parameters for when we debug executables that can help with plug-in selection: target.execution-level = auto | user | kernel target.execution-mode = auto | dynamic | static target.execution-os-type = auto | none | halted | live On some systems, the binaries that are created are the same wether you use them to debug a kernel, or a user space program. Many times inspecting an object file can reveal what an executable should be. For these cases we can now be a little more complete by specifying wether to detect all of these things automatically (inspect the main executable file and select a plug-in accordingly), or manually to force the selection of certain plug-ins. To do this we now allow the specficifation of wether one is debugging a user space program (target.execution-level = user) or a kernel program (target.execution-level = kernel). We can also specify if we want to debug a program where shared libraries are dynamically loaded using a DynamicLoader plug-in (target.execution-mode = dynamic), or wether we will treat all symbol files as already linked at the correct address (target.execution-mode = static). We can also specify if the inferior we are debugging is being debugged on a bare board (target.execution-os-type = none), or debugging an OS where we have a JTAG or other direct connection to the inferior stops the entire OS (target.execution-os-type = halted), or if we are debugging a program on something that has live debug services (target.execution-os-type = live). For the "target.execution-os-type = halted" mode, we will need to create ProcessHelper plug-ins that allow us to extract the process/thread and other OS information by reading/writing memory. This should allow LLDB to be used for a wide variety of debugging tasks and handle them all correctly. llvm-svn: 125815
2011-02-18 09:44:25 +08:00
std::map<std::string, InstanceSettingsSP>::iterator pos;
Mutex::Locker locker (m_pending_settings_mutex);
pos = m_pending_settings.find (name_str);
if (pos != m_pending_settings.end())
Added code that will allow completely customizable prompts for use in replacing the "(lldb)" prompt, the "frame #1..." displays when doing stack backtracing and the "thread #1....". This will allow you to see exactly the information that you want to see where you want to see it. This currently isn't hookup up to the prompts yet, but it will be soon. So what is the format of the prompts? Prompts can contain variables that have access to the current program state. Variables are text that appears in between a prefix of "${" and ends with a "}". Some of the interesting variables include: // The frame index (0, 1, 2, 3...) ${frame.index} // common frame registers with generic names ${frame.pc} ${frame.sp} ${frame.fp} ${frame.ra} ${frame.flags} // Access to any frame registers by name where REGNAME is any register name: ${frame.reg.REGNAME} // The current compile unit file where the frame is located ${file.basename} ${file.fullpath} // Function information ${function.name} ${function.pc-offset} // Process info ${process.file.basename} ${process.file.fullpath} ${process.id} ${process.name} // Thread info ${thread.id} ${thread.index} ${thread.name} ${thread.queue} ${thread.stop-reason} // Target information ${target.arch} // The current module for the current frame (the shared library or executable // that contains the current frame PC value): ${module.file.basename} ${module.file.fullpath} // Access to the line entry for where the current frame is when your thread // is stopped: ${line.file.basename} ${line.file.fullpath} ${line.number} ${line.start-addr} ${line.end-addr} Many times the information that you might have in your prompt might not be available and you won't want it to print out if it isn't valid. To take care of this you can enclose everything that must resolve into a scope. A scope is starts with '{' and ends with '}'. For example in order to only display the current file and line number when the information is available the format would be: "{ at {$line.file.basename}:${line.number}}" Broken down this is: start the scope: "{" format whose content will only be displayed if all information is available: "at {$line.file.basename}:${line.number}" end the scope: "}" We currently can represent the infomration we see when stopped at a frame: frame #0: 0x0000000100000e85 a.out`main + 4 at test.c:19 with the following format: "frame #${frame.index}: ${frame.pc} {${module.file.basename}`}{${function.name}{${function.pc-offset}}{ at ${line.file.basename}:${line.number}}\n" This breaks down to always print: "frame #${frame.index}: ${frame.pc} " only print the module followed by a tick if we have a valid module: "{${module.file.basename}`}" print the function name with optional offset: "{${function.name}{${function.pc-offset}}" print the line info if it is available: "{ at ${line.file.basename}:${line.number}}" then finish off with a newline: "\n" Notice you can also put newlines ("\n") and tabs and everything else you are used to putting in a format string when desensitized with the \ character. Cleaned up some of the user settings controller subclasses. All of them do not have any global settings variables and were all implementing stubs for the get/set global settings variable. Now there is a default version in UserSettingsController that will do nothing. llvm-svn: 114306
2010-09-19 10:33:57 +08:00
{
Added new target instance settings for execution settings: Targets can now specify some additional parameters for when we debug executables that can help with plug-in selection: target.execution-level = auto | user | kernel target.execution-mode = auto | dynamic | static target.execution-os-type = auto | none | halted | live On some systems, the binaries that are created are the same wether you use them to debug a kernel, or a user space program. Many times inspecting an object file can reveal what an executable should be. For these cases we can now be a little more complete by specifying wether to detect all of these things automatically (inspect the main executable file and select a plug-in accordingly), or manually to force the selection of certain plug-ins. To do this we now allow the specficifation of wether one is debugging a user space program (target.execution-level = user) or a kernel program (target.execution-level = kernel). We can also specify if we want to debug a program where shared libraries are dynamically loaded using a DynamicLoader plug-in (target.execution-mode = dynamic), or wether we will treat all symbol files as already linked at the correct address (target.execution-mode = static). We can also specify if the inferior we are debugging is being debugged on a bare board (target.execution-os-type = none), or debugging an OS where we have a JTAG or other direct connection to the inferior stops the entire OS (target.execution-os-type = halted), or if we are debugging a program on something that has live debug services (target.execution-os-type = live). For the "target.execution-os-type = halted" mode, we will need to create ProcessHelper plug-ins that allow us to extract the process/thread and other OS information by reading/writing memory. This should allow LLDB to be used for a wide variety of debugging tasks and handle them all correctly. llvm-svn: 125815
2011-02-18 09:44:25 +08:00
InstanceSettingsSP settings_sp = pos->second;
return settings_sp;
Added code that will allow completely customizable prompts for use in replacing the "(lldb)" prompt, the "frame #1..." displays when doing stack backtracing and the "thread #1....". This will allow you to see exactly the information that you want to see where you want to see it. This currently isn't hookup up to the prompts yet, but it will be soon. So what is the format of the prompts? Prompts can contain variables that have access to the current program state. Variables are text that appears in between a prefix of "${" and ends with a "}". Some of the interesting variables include: // The frame index (0, 1, 2, 3...) ${frame.index} // common frame registers with generic names ${frame.pc} ${frame.sp} ${frame.fp} ${frame.ra} ${frame.flags} // Access to any frame registers by name where REGNAME is any register name: ${frame.reg.REGNAME} // The current compile unit file where the frame is located ${file.basename} ${file.fullpath} // Function information ${function.name} ${function.pc-offset} // Process info ${process.file.basename} ${process.file.fullpath} ${process.id} ${process.name} // Thread info ${thread.id} ${thread.index} ${thread.name} ${thread.queue} ${thread.stop-reason} // Target information ${target.arch} // The current module for the current frame (the shared library or executable // that contains the current frame PC value): ${module.file.basename} ${module.file.fullpath} // Access to the line entry for where the current frame is when your thread // is stopped: ${line.file.basename} ${line.file.fullpath} ${line.number} ${line.start-addr} ${line.end-addr} Many times the information that you might have in your prompt might not be available and you won't want it to print out if it isn't valid. To take care of this you can enclose everything that must resolve into a scope. A scope is starts with '{' and ends with '}'. For example in order to only display the current file and line number when the information is available the format would be: "{ at {$line.file.basename}:${line.number}}" Broken down this is: start the scope: "{" format whose content will only be displayed if all information is available: "at {$line.file.basename}:${line.number}" end the scope: "}" We currently can represent the infomration we see when stopped at a frame: frame #0: 0x0000000100000e85 a.out`main + 4 at test.c:19 with the following format: "frame #${frame.index}: ${frame.pc} {${module.file.basename}`}{${function.name}{${function.pc-offset}}{ at ${line.file.basename}:${line.number}}\n" This breaks down to always print: "frame #${frame.index}: ${frame.pc} " only print the module followed by a tick if we have a valid module: "{${module.file.basename}`}" print the function name with optional offset: "{${function.name}{${function.pc-offset}}" print the line info if it is available: "{ at ${line.file.basename}:${line.number}}" then finish off with a newline: "\n" Notice you can also put newlines ("\n") and tabs and everything else you are used to putting in a format string when desensitized with the \ character. Cleaned up some of the user settings controller subclasses. All of them do not have any global settings variables and were all implementing stubs for the get/set global settings variable. Now there is a default version in UserSettingsController that will do nothing. llvm-svn: 114306
2010-09-19 10:33:57 +08:00
}
else
Added code that will allow completely customizable prompts for use in replacing the "(lldb)" prompt, the "frame #1..." displays when doing stack backtracing and the "thread #1....". This will allow you to see exactly the information that you want to see where you want to see it. This currently isn't hookup up to the prompts yet, but it will be soon. So what is the format of the prompts? Prompts can contain variables that have access to the current program state. Variables are text that appears in between a prefix of "${" and ends with a "}". Some of the interesting variables include: // The frame index (0, 1, 2, 3...) ${frame.index} // common frame registers with generic names ${frame.pc} ${frame.sp} ${frame.fp} ${frame.ra} ${frame.flags} // Access to any frame registers by name where REGNAME is any register name: ${frame.reg.REGNAME} // The current compile unit file where the frame is located ${file.basename} ${file.fullpath} // Function information ${function.name} ${function.pc-offset} // Process info ${process.file.basename} ${process.file.fullpath} ${process.id} ${process.name} // Thread info ${thread.id} ${thread.index} ${thread.name} ${thread.queue} ${thread.stop-reason} // Target information ${target.arch} // The current module for the current frame (the shared library or executable // that contains the current frame PC value): ${module.file.basename} ${module.file.fullpath} // Access to the line entry for where the current frame is when your thread // is stopped: ${line.file.basename} ${line.file.fullpath} ${line.number} ${line.start-addr} ${line.end-addr} Many times the information that you might have in your prompt might not be available and you won't want it to print out if it isn't valid. To take care of this you can enclose everything that must resolve into a scope. A scope is starts with '{' and ends with '}'. For example in order to only display the current file and line number when the information is available the format would be: "{ at {$line.file.basename}:${line.number}}" Broken down this is: start the scope: "{" format whose content will only be displayed if all information is available: "at {$line.file.basename}:${line.number}" end the scope: "}" We currently can represent the infomration we see when stopped at a frame: frame #0: 0x0000000100000e85 a.out`main + 4 at test.c:19 with the following format: "frame #${frame.index}: ${frame.pc} {${module.file.basename}`}{${function.name}{${function.pc-offset}}{ at ${line.file.basename}:${line.number}}\n" This breaks down to always print: "frame #${frame.index}: ${frame.pc} " only print the module followed by a tick if we have a valid module: "{${module.file.basename}`}" print the function name with optional offset: "{${function.name}{${function.pc-offset}}" print the line info if it is available: "{ at ${line.file.basename}:${line.number}}" then finish off with a newline: "\n" Notice you can also put newlines ("\n") and tabs and everything else you are used to putting in a format string when desensitized with the \ character. Cleaned up some of the user settings controller subclasses. All of them do not have any global settings variables and were all implementing stubs for the get/set global settings variable. Now there is a default version in UserSettingsController that will do nothing. llvm-svn: 114306
2010-09-19 10:33:57 +08:00
{
InstanceSettingsSP new_settings_sp = CreateInstanceSettings (instance_name.GetCString());
CopyDefaultSettings (new_settings_sp, instance_name, true);
m_pending_settings[name_str] = new_settings_sp;
return new_settings_sp;
Added code that will allow completely customizable prompts for use in replacing the "(lldb)" prompt, the "frame #1..." displays when doing stack backtracing and the "thread #1....". This will allow you to see exactly the information that you want to see where you want to see it. This currently isn't hookup up to the prompts yet, but it will be soon. So what is the format of the prompts? Prompts can contain variables that have access to the current program state. Variables are text that appears in between a prefix of "${" and ends with a "}". Some of the interesting variables include: // The frame index (0, 1, 2, 3...) ${frame.index} // common frame registers with generic names ${frame.pc} ${frame.sp} ${frame.fp} ${frame.ra} ${frame.flags} // Access to any frame registers by name where REGNAME is any register name: ${frame.reg.REGNAME} // The current compile unit file where the frame is located ${file.basename} ${file.fullpath} // Function information ${function.name} ${function.pc-offset} // Process info ${process.file.basename} ${process.file.fullpath} ${process.id} ${process.name} // Thread info ${thread.id} ${thread.index} ${thread.name} ${thread.queue} ${thread.stop-reason} // Target information ${target.arch} // The current module for the current frame (the shared library or executable // that contains the current frame PC value): ${module.file.basename} ${module.file.fullpath} // Access to the line entry for where the current frame is when your thread // is stopped: ${line.file.basename} ${line.file.fullpath} ${line.number} ${line.start-addr} ${line.end-addr} Many times the information that you might have in your prompt might not be available and you won't want it to print out if it isn't valid. To take care of this you can enclose everything that must resolve into a scope. A scope is starts with '{' and ends with '}'. For example in order to only display the current file and line number when the information is available the format would be: "{ at {$line.file.basename}:${line.number}}" Broken down this is: start the scope: "{" format whose content will only be displayed if all information is available: "at {$line.file.basename}:${line.number}" end the scope: "}" We currently can represent the infomration we see when stopped at a frame: frame #0: 0x0000000100000e85 a.out`main + 4 at test.c:19 with the following format: "frame #${frame.index}: ${frame.pc} {${module.file.basename}`}{${function.name}{${function.pc-offset}}{ at ${line.file.basename}:${line.number}}\n" This breaks down to always print: "frame #${frame.index}: ${frame.pc} " only print the module followed by a tick if we have a valid module: "{${module.file.basename}`}" print the function name with optional offset: "{${function.name}{${function.pc-offset}}" print the line info if it is available: "{ at ${line.file.basename}:${line.number}}" then finish off with a newline: "\n" Notice you can also put newlines ("\n") and tabs and everything else you are used to putting in a format string when desensitized with the \ character. Cleaned up some of the user settings controller subclasses. All of them do not have any global settings variables and were all implementing stubs for the get/set global settings variable. Now there is a default version in UserSettingsController that will do nothing. llvm-svn: 114306
2010-09-19 10:33:57 +08:00
}
// Should never reach this line.
Added new target instance settings for execution settings: Targets can now specify some additional parameters for when we debug executables that can help with plug-in selection: target.execution-level = auto | user | kernel target.execution-mode = auto | dynamic | static target.execution-os-type = auto | none | halted | live On some systems, the binaries that are created are the same wether you use them to debug a kernel, or a user space program. Many times inspecting an object file can reveal what an executable should be. For these cases we can now be a little more complete by specifying wether to detect all of these things automatically (inspect the main executable file and select a plug-in accordingly), or manually to force the selection of certain plug-ins. To do this we now allow the specficifation of wether one is debugging a user space program (target.execution-level = user) or a kernel program (target.execution-level = kernel). We can also specify if we want to debug a program where shared libraries are dynamically loaded using a DynamicLoader plug-in (target.execution-mode = dynamic), or wether we will treat all symbol files as already linked at the correct address (target.execution-mode = static). We can also specify if the inferior we are debugging is being debugged on a bare board (target.execution-os-type = none), or debugging an OS where we have a JTAG or other direct connection to the inferior stops the entire OS (target.execution-os-type = halted), or if we are debugging a program on something that has live debug services (target.execution-os-type = live). For the "target.execution-os-type = halted" mode, we will need to create ProcessHelper plug-ins that allow us to extract the process/thread and other OS information by reading/writing memory. This should allow LLDB to be used for a wide variety of debugging tasks and handle them all correctly. llvm-svn: 125815
2011-02-18 09:44:25 +08:00
InstanceSettingsSP dummy;
return dummy;
}
void
UserSettingsController::GetAllDefaultSettingValues (Stream &strm)
{
std::string parent_prefix;
BuildParentPrefix (parent_prefix);
for (int i = 0; i < m_settings.instance_settings.size(); ++i)
{
SettingEntry &entry = m_settings.instance_settings[i];
ConstString var_name (entry.var_name);
StringList value;
m_default_settings->GetInstanceSettingsValue (entry, var_name, value, NULL);
if (!parent_prefix.empty())
strm.Printf ("%s.", parent_prefix.c_str());
DumpValue (var_name.GetCString(),
entry.var_type,
value,
strm);
}
}
void
UserSettingsController::GetAllPendingSettingValues (Stream &strm)
{
Added new target instance settings for execution settings: Targets can now specify some additional parameters for when we debug executables that can help with plug-in selection: target.execution-level = auto | user | kernel target.execution-mode = auto | dynamic | static target.execution-os-type = auto | none | halted | live On some systems, the binaries that are created are the same wether you use them to debug a kernel, or a user space program. Many times inspecting an object file can reveal what an executable should be. For these cases we can now be a little more complete by specifying wether to detect all of these things automatically (inspect the main executable file and select a plug-in accordingly), or manually to force the selection of certain plug-ins. To do this we now allow the specficifation of wether one is debugging a user space program (target.execution-level = user) or a kernel program (target.execution-level = kernel). We can also specify if we want to debug a program where shared libraries are dynamically loaded using a DynamicLoader plug-in (target.execution-mode = dynamic), or wether we will treat all symbol files as already linked at the correct address (target.execution-mode = static). We can also specify if the inferior we are debugging is being debugged on a bare board (target.execution-os-type = none), or debugging an OS where we have a JTAG or other direct connection to the inferior stops the entire OS (target.execution-os-type = halted), or if we are debugging a program on something that has live debug services (target.execution-os-type = live). For the "target.execution-os-type = halted" mode, we will need to create ProcessHelper plug-ins that allow us to extract the process/thread and other OS information by reading/writing memory. This should allow LLDB to be used for a wide variety of debugging tasks and handle them all correctly. llvm-svn: 125815
2011-02-18 09:44:25 +08:00
std::map<std::string, InstanceSettingsSP>::iterator pos;
std::string parent_prefix;
BuildParentPrefix (parent_prefix);
const char *prefix = parent_prefix.c_str();
for (pos = m_pending_settings.begin(); pos != m_pending_settings.end(); ++pos)
{
std::string tmp_name = pos->first;
Added new target instance settings for execution settings: Targets can now specify some additional parameters for when we debug executables that can help with plug-in selection: target.execution-level = auto | user | kernel target.execution-mode = auto | dynamic | static target.execution-os-type = auto | none | halted | live On some systems, the binaries that are created are the same wether you use them to debug a kernel, or a user space program. Many times inspecting an object file can reveal what an executable should be. For these cases we can now be a little more complete by specifying wether to detect all of these things automatically (inspect the main executable file and select a plug-in accordingly), or manually to force the selection of certain plug-ins. To do this we now allow the specficifation of wether one is debugging a user space program (target.execution-level = user) or a kernel program (target.execution-level = kernel). We can also specify if we want to debug a program where shared libraries are dynamically loaded using a DynamicLoader plug-in (target.execution-mode = dynamic), or wether we will treat all symbol files as already linked at the correct address (target.execution-mode = static). We can also specify if the inferior we are debugging is being debugged on a bare board (target.execution-os-type = none), or debugging an OS where we have a JTAG or other direct connection to the inferior stops the entire OS (target.execution-os-type = halted), or if we are debugging a program on something that has live debug services (target.execution-os-type = live). For the "target.execution-os-type = halted" mode, we will need to create ProcessHelper plug-ins that allow us to extract the process/thread and other OS information by reading/writing memory. This should allow LLDB to be used for a wide variety of debugging tasks and handle them all correctly. llvm-svn: 125815
2011-02-18 09:44:25 +08:00
InstanceSettingsSP settings_sp = pos->second;
const ConstString instance_name (tmp_name.c_str());
for (int i = 0; i < m_settings.instance_settings.size(); ++i)
{
SettingEntry &entry = m_settings.instance_settings[i];
ConstString var_name (entry.var_name);
StringList tmp_value;
settings_sp->GetInstanceSettingsValue (entry, var_name, tmp_value, NULL);
StreamString value_str;
if (tmp_value.GetSize() == 1)
value_str.Printf ("%s", tmp_value.GetStringAtIndex (0));
else
{
for (int j = 0; j < tmp_value.GetSize(); ++j)
value_str.Printf ("%s ", tmp_value.GetStringAtIndex (j));
}
if (parent_prefix.length() > 0)
{
strm.Printf ("%s.%s.%s (%s) = '%s' [pending]\n", prefix, instance_name.GetCString(),
var_name.GetCString(), UserSettingsController::GetTypeString (entry.var_type),
value_str.GetData());
}
else
{
strm.Printf ("%s (%s) = '%s' [pending]\n", var_name.GetCString(),
UserSettingsController::GetTypeString (entry.var_type),
value_str.GetData());
}
}
}
}
InstanceSettings *
UserSettingsController::FindSettingsForInstance (const ConstString &instance_name)
{
std::string instance_name_str (instance_name.GetCString());
Mutex::Locker locker (m_live_settings_mutex);
InstanceSettingsMap::iterator pos = m_live_settings.find (instance_name_str);
if (pos != m_live_settings.end ())
return pos->second;
return NULL;
}
void
UserSettingsController::GetAllInstanceVariableValues (CommandInterpreter &interpreter,
Stream &strm)
{
std::string parent_prefix;
BuildParentPrefix (parent_prefix);
StreamString description;
Mutex::Locker locker (m_live_settings_mutex);
for (InstanceSettingsMap::iterator pos = m_live_settings.begin(); pos != m_live_settings.end(); ++pos)
{
std::string instance_name = pos->first;
InstanceSettings *settings = pos->second;
for (int i = 0; i < m_settings.instance_settings.size(); ++i)
{
SettingEntry &entry = m_settings.instance_settings[i];
const ConstString var_name (entry.var_name);
StringList tmp_value;
settings->GetInstanceSettingsValue (entry, var_name, tmp_value, NULL);
if (!parent_prefix.empty())
strm.Printf ("%s.", parent_prefix.c_str());
DumpValue(var_name.GetCString(), entry.var_type, tmp_value, strm);
}
}
}
void
UserSettingsController::OverrideAllInstances (const ConstString &var_name,
const char *value,
Added new target instance settings for execution settings: Targets can now specify some additional parameters for when we debug executables that can help with plug-in selection: target.execution-level = auto | user | kernel target.execution-mode = auto | dynamic | static target.execution-os-type = auto | none | halted | live On some systems, the binaries that are created are the same wether you use them to debug a kernel, or a user space program. Many times inspecting an object file can reveal what an executable should be. For these cases we can now be a little more complete by specifying wether to detect all of these things automatically (inspect the main executable file and select a plug-in accordingly), or manually to force the selection of certain plug-ins. To do this we now allow the specficifation of wether one is debugging a user space program (target.execution-level = user) or a kernel program (target.execution-level = kernel). We can also specify if we want to debug a program where shared libraries are dynamically loaded using a DynamicLoader plug-in (target.execution-mode = dynamic), or wether we will treat all symbol files as already linked at the correct address (target.execution-mode = static). We can also specify if the inferior we are debugging is being debugged on a bare board (target.execution-os-type = none), or debugging an OS where we have a JTAG or other direct connection to the inferior stops the entire OS (target.execution-os-type = halted), or if we are debugging a program on something that has live debug services (target.execution-os-type = live). For the "target.execution-os-type = halted" mode, we will need to create ProcessHelper plug-ins that allow us to extract the process/thread and other OS information by reading/writing memory. This should allow LLDB to be used for a wide variety of debugging tasks and handle them all correctly. llvm-svn: 125815
2011-02-18 09:44:25 +08:00
VarSetOperationType op,
const char *index_value,
Error &err)
{
StreamString description;
Mutex::Locker locker (m_live_settings_mutex);
for (InstanceSettingsMap::iterator pos = m_live_settings.begin(); pos != m_live_settings.end(); ++pos)
{
InstanceSettings *settings = pos->second;
StreamString tmp_name;
tmp_name.Printf ("[%s]", settings->GetInstanceName().GetCString());
const ConstString instance_name (tmp_name.GetData());
const SettingEntry *entry = GetInstanceEntry (var_name);
settings->UpdateInstanceSettingsVariable (var_name, index_value, value, instance_name, *entry, op, err, false);
}
}
void
UserSettingsController::RegisterInstanceSettings (InstanceSettings *instance_settings)
{
Mutex::Locker locker (m_live_settings_mutex);
StreamString tmp_name;
tmp_name.Printf ("[%s]", instance_settings->GetInstanceName().GetCString());
const ConstString instance_name (tmp_name.GetData());
std::string instance_name_str (instance_name.GetCString());
if (instance_name_str.compare (InstanceSettings::GetDefaultName().GetCString()) != 0)
m_live_settings[instance_name_str] = instance_settings;
}
void
UserSettingsController::UnregisterInstanceSettings (InstanceSettings *instance)
{
Mutex::Locker locker (m_live_settings_mutex);
StreamString tmp_name;
tmp_name.Printf ("[%s]", instance->GetInstanceName().GetCString());
std::string instance_name (tmp_name.GetData());
InstanceSettingsMap::iterator pos = m_live_settings.find (instance_name);
if (pos != m_live_settings.end())
m_live_settings.erase (pos);
}
void
UserSettingsController::CreateSettingsVector (const SettingEntry *table,
bool global)
{
int i = 0;
while (table[i].var_name != NULL)
{
const SettingEntry &table_entry = table[i];
ConstString const_var_name (table_entry.var_name);
SettingEntry new_entry;
new_entry = table_entry;
new_entry.var_name = const_var_name.GetCString();
if (global)
m_settings.global_settings.push_back (new_entry);
else
m_settings.instance_settings.push_back (new_entry);
++i;
}
}
//----------------------------------------------------------------------
// UserSettingsController static methods
//----------------------------------------------------------------------
int
FindMaxNameLength (std::vector<SettingEntry> table)
{
int max_length = 1;
for (int i = 0; i < table.size(); ++i)
{
int len = strlen (table[i].var_name);
if (len > max_length)
max_length = len;
}
return max_length;
}
const char *
Added new target instance settings for execution settings: Targets can now specify some additional parameters for when we debug executables that can help with plug-in selection: target.execution-level = auto | user | kernel target.execution-mode = auto | dynamic | static target.execution-os-type = auto | none | halted | live On some systems, the binaries that are created are the same wether you use them to debug a kernel, or a user space program. Many times inspecting an object file can reveal what an executable should be. For these cases we can now be a little more complete by specifying wether to detect all of these things automatically (inspect the main executable file and select a plug-in accordingly), or manually to force the selection of certain plug-ins. To do this we now allow the specficifation of wether one is debugging a user space program (target.execution-level = user) or a kernel program (target.execution-level = kernel). We can also specify if we want to debug a program where shared libraries are dynamically loaded using a DynamicLoader plug-in (target.execution-mode = dynamic), or wether we will treat all symbol files as already linked at the correct address (target.execution-mode = static). We can also specify if the inferior we are debugging is being debugged on a bare board (target.execution-os-type = none), or debugging an OS where we have a JTAG or other direct connection to the inferior stops the entire OS (target.execution-os-type = halted), or if we are debugging a program on something that has live debug services (target.execution-os-type = live). For the "target.execution-os-type = halted" mode, we will need to create ProcessHelper plug-ins that allow us to extract the process/thread and other OS information by reading/writing memory. This should allow LLDB to be used for a wide variety of debugging tasks and handle them all correctly. llvm-svn: 125815
2011-02-18 09:44:25 +08:00
UserSettingsController::GetTypeString (SettableVariableType var_type)
{
switch (var_type)
{
Added new target instance settings for execution settings: Targets can now specify some additional parameters for when we debug executables that can help with plug-in selection: target.execution-level = auto | user | kernel target.execution-mode = auto | dynamic | static target.execution-os-type = auto | none | halted | live On some systems, the binaries that are created are the same wether you use them to debug a kernel, or a user space program. Many times inspecting an object file can reveal what an executable should be. For these cases we can now be a little more complete by specifying wether to detect all of these things automatically (inspect the main executable file and select a plug-in accordingly), or manually to force the selection of certain plug-ins. To do this we now allow the specficifation of wether one is debugging a user space program (target.execution-level = user) or a kernel program (target.execution-level = kernel). We can also specify if we want to debug a program where shared libraries are dynamically loaded using a DynamicLoader plug-in (target.execution-mode = dynamic), or wether we will treat all symbol files as already linked at the correct address (target.execution-mode = static). We can also specify if the inferior we are debugging is being debugged on a bare board (target.execution-os-type = none), or debugging an OS where we have a JTAG or other direct connection to the inferior stops the entire OS (target.execution-os-type = halted), or if we are debugging a program on something that has live debug services (target.execution-os-type = live). For the "target.execution-os-type = halted" mode, we will need to create ProcessHelper plug-ins that allow us to extract the process/thread and other OS information by reading/writing memory. This should allow LLDB to be used for a wide variety of debugging tasks and handle them all correctly. llvm-svn: 125815
2011-02-18 09:44:25 +08:00
case eSetVarTypeInt:
return "int";
Added new target instance settings for execution settings: Targets can now specify some additional parameters for when we debug executables that can help with plug-in selection: target.execution-level = auto | user | kernel target.execution-mode = auto | dynamic | static target.execution-os-type = auto | none | halted | live On some systems, the binaries that are created are the same wether you use them to debug a kernel, or a user space program. Many times inspecting an object file can reveal what an executable should be. For these cases we can now be a little more complete by specifying wether to detect all of these things automatically (inspect the main executable file and select a plug-in accordingly), or manually to force the selection of certain plug-ins. To do this we now allow the specficifation of wether one is debugging a user space program (target.execution-level = user) or a kernel program (target.execution-level = kernel). We can also specify if we want to debug a program where shared libraries are dynamically loaded using a DynamicLoader plug-in (target.execution-mode = dynamic), or wether we will treat all symbol files as already linked at the correct address (target.execution-mode = static). We can also specify if the inferior we are debugging is being debugged on a bare board (target.execution-os-type = none), or debugging an OS where we have a JTAG or other direct connection to the inferior stops the entire OS (target.execution-os-type = halted), or if we are debugging a program on something that has live debug services (target.execution-os-type = live). For the "target.execution-os-type = halted" mode, we will need to create ProcessHelper plug-ins that allow us to extract the process/thread and other OS information by reading/writing memory. This should allow LLDB to be used for a wide variety of debugging tasks and handle them all correctly. llvm-svn: 125815
2011-02-18 09:44:25 +08:00
case eSetVarTypeBoolean:
return "boolean";
Added new target instance settings for execution settings: Targets can now specify some additional parameters for when we debug executables that can help with plug-in selection: target.execution-level = auto | user | kernel target.execution-mode = auto | dynamic | static target.execution-os-type = auto | none | halted | live On some systems, the binaries that are created are the same wether you use them to debug a kernel, or a user space program. Many times inspecting an object file can reveal what an executable should be. For these cases we can now be a little more complete by specifying wether to detect all of these things automatically (inspect the main executable file and select a plug-in accordingly), or manually to force the selection of certain plug-ins. To do this we now allow the specficifation of wether one is debugging a user space program (target.execution-level = user) or a kernel program (target.execution-level = kernel). We can also specify if we want to debug a program where shared libraries are dynamically loaded using a DynamicLoader plug-in (target.execution-mode = dynamic), or wether we will treat all symbol files as already linked at the correct address (target.execution-mode = static). We can also specify if the inferior we are debugging is being debugged on a bare board (target.execution-os-type = none), or debugging an OS where we have a JTAG or other direct connection to the inferior stops the entire OS (target.execution-os-type = halted), or if we are debugging a program on something that has live debug services (target.execution-os-type = live). For the "target.execution-os-type = halted" mode, we will need to create ProcessHelper plug-ins that allow us to extract the process/thread and other OS information by reading/writing memory. This should allow LLDB to be used for a wide variety of debugging tasks and handle them all correctly. llvm-svn: 125815
2011-02-18 09:44:25 +08:00
case eSetVarTypeString:
return "string";
Added new target instance settings for execution settings: Targets can now specify some additional parameters for when we debug executables that can help with plug-in selection: target.execution-level = auto | user | kernel target.execution-mode = auto | dynamic | static target.execution-os-type = auto | none | halted | live On some systems, the binaries that are created are the same wether you use them to debug a kernel, or a user space program. Many times inspecting an object file can reveal what an executable should be. For these cases we can now be a little more complete by specifying wether to detect all of these things automatically (inspect the main executable file and select a plug-in accordingly), or manually to force the selection of certain plug-ins. To do this we now allow the specficifation of wether one is debugging a user space program (target.execution-level = user) or a kernel program (target.execution-level = kernel). We can also specify if we want to debug a program where shared libraries are dynamically loaded using a DynamicLoader plug-in (target.execution-mode = dynamic), or wether we will treat all symbol files as already linked at the correct address (target.execution-mode = static). We can also specify if the inferior we are debugging is being debugged on a bare board (target.execution-os-type = none), or debugging an OS where we have a JTAG or other direct connection to the inferior stops the entire OS (target.execution-os-type = halted), or if we are debugging a program on something that has live debug services (target.execution-os-type = live). For the "target.execution-os-type = halted" mode, we will need to create ProcessHelper plug-ins that allow us to extract the process/thread and other OS information by reading/writing memory. This should allow LLDB to be used for a wide variety of debugging tasks and handle them all correctly. llvm-svn: 125815
2011-02-18 09:44:25 +08:00
case eSetVarTypeArray:
return "array";
Added new target instance settings for execution settings: Targets can now specify some additional parameters for when we debug executables that can help with plug-in selection: target.execution-level = auto | user | kernel target.execution-mode = auto | dynamic | static target.execution-os-type = auto | none | halted | live On some systems, the binaries that are created are the same wether you use them to debug a kernel, or a user space program. Many times inspecting an object file can reveal what an executable should be. For these cases we can now be a little more complete by specifying wether to detect all of these things automatically (inspect the main executable file and select a plug-in accordingly), or manually to force the selection of certain plug-ins. To do this we now allow the specficifation of wether one is debugging a user space program (target.execution-level = user) or a kernel program (target.execution-level = kernel). We can also specify if we want to debug a program where shared libraries are dynamically loaded using a DynamicLoader plug-in (target.execution-mode = dynamic), or wether we will treat all symbol files as already linked at the correct address (target.execution-mode = static). We can also specify if the inferior we are debugging is being debugged on a bare board (target.execution-os-type = none), or debugging an OS where we have a JTAG or other direct connection to the inferior stops the entire OS (target.execution-os-type = halted), or if we are debugging a program on something that has live debug services (target.execution-os-type = live). For the "target.execution-os-type = halted" mode, we will need to create ProcessHelper plug-ins that allow us to extract the process/thread and other OS information by reading/writing memory. This should allow LLDB to be used for a wide variety of debugging tasks and handle them all correctly. llvm-svn: 125815
2011-02-18 09:44:25 +08:00
case eSetVarTypeDictionary:
return "dictionary";
Added new target instance settings for execution settings: Targets can now specify some additional parameters for when we debug executables that can help with plug-in selection: target.execution-level = auto | user | kernel target.execution-mode = auto | dynamic | static target.execution-os-type = auto | none | halted | live On some systems, the binaries that are created are the same wether you use them to debug a kernel, or a user space program. Many times inspecting an object file can reveal what an executable should be. For these cases we can now be a little more complete by specifying wether to detect all of these things automatically (inspect the main executable file and select a plug-in accordingly), or manually to force the selection of certain plug-ins. To do this we now allow the specficifation of wether one is debugging a user space program (target.execution-level = user) or a kernel program (target.execution-level = kernel). We can also specify if we want to debug a program where shared libraries are dynamically loaded using a DynamicLoader plug-in (target.execution-mode = dynamic), or wether we will treat all symbol files as already linked at the correct address (target.execution-mode = static). We can also specify if the inferior we are debugging is being debugged on a bare board (target.execution-os-type = none), or debugging an OS where we have a JTAG or other direct connection to the inferior stops the entire OS (target.execution-os-type = halted), or if we are debugging a program on something that has live debug services (target.execution-os-type = live). For the "target.execution-os-type = halted" mode, we will need to create ProcessHelper plug-ins that allow us to extract the process/thread and other OS information by reading/writing memory. This should allow LLDB to be used for a wide variety of debugging tasks and handle them all correctly. llvm-svn: 125815
2011-02-18 09:44:25 +08:00
case eSetVarTypeEnum:
return "enum";
Added new target instance settings for execution settings: Targets can now specify some additional parameters for when we debug executables that can help with plug-in selection: target.execution-level = auto | user | kernel target.execution-mode = auto | dynamic | static target.execution-os-type = auto | none | halted | live On some systems, the binaries that are created are the same wether you use them to debug a kernel, or a user space program. Many times inspecting an object file can reveal what an executable should be. For these cases we can now be a little more complete by specifying wether to detect all of these things automatically (inspect the main executable file and select a plug-in accordingly), or manually to force the selection of certain plug-ins. To do this we now allow the specficifation of wether one is debugging a user space program (target.execution-level = user) or a kernel program (target.execution-level = kernel). We can also specify if we want to debug a program where shared libraries are dynamically loaded using a DynamicLoader plug-in (target.execution-mode = dynamic), or wether we will treat all symbol files as already linked at the correct address (target.execution-mode = static). We can also specify if the inferior we are debugging is being debugged on a bare board (target.execution-os-type = none), or debugging an OS where we have a JTAG or other direct connection to the inferior stops the entire OS (target.execution-os-type = halted), or if we are debugging a program on something that has live debug services (target.execution-os-type = live). For the "target.execution-os-type = halted" mode, we will need to create ProcessHelper plug-ins that allow us to extract the process/thread and other OS information by reading/writing memory. This should allow LLDB to be used for a wide variety of debugging tasks and handle them all correctly. llvm-svn: 125815
2011-02-18 09:44:25 +08:00
case eSetVarTypeNone:
return "no type";
}
return "";
}
void
Added new target instance settings for execution settings: Targets can now specify some additional parameters for when we debug executables that can help with plug-in selection: target.execution-level = auto | user | kernel target.execution-mode = auto | dynamic | static target.execution-os-type = auto | none | halted | live On some systems, the binaries that are created are the same wether you use them to debug a kernel, or a user space program. Many times inspecting an object file can reveal what an executable should be. For these cases we can now be a little more complete by specifying wether to detect all of these things automatically (inspect the main executable file and select a plug-in accordingly), or manually to force the selection of certain plug-ins. To do this we now allow the specficifation of wether one is debugging a user space program (target.execution-level = user) or a kernel program (target.execution-level = kernel). We can also specify if we want to debug a program where shared libraries are dynamically loaded using a DynamicLoader plug-in (target.execution-mode = dynamic), or wether we will treat all symbol files as already linked at the correct address (target.execution-mode = static). We can also specify if the inferior we are debugging is being debugged on a bare board (target.execution-os-type = none), or debugging an OS where we have a JTAG or other direct connection to the inferior stops the entire OS (target.execution-os-type = halted), or if we are debugging a program on something that has live debug services (target.execution-os-type = live). For the "target.execution-os-type = halted" mode, we will need to create ProcessHelper plug-ins that allow us to extract the process/thread and other OS information by reading/writing memory. This should allow LLDB to be used for a wide variety of debugging tasks and handle them all correctly. llvm-svn: 125815
2011-02-18 09:44:25 +08:00
UserSettingsController::PrintEnumValues (const OptionEnumValueElement *enum_values, Stream &str)
{
int i = 0;
while (enum_values[i].string_value != NULL)
{
str.Printf ("%s ", enum_values[i].string_value);
++i;
}
}
void
UserSettingsController::FindAllSettingsDescriptions (CommandInterpreter &interpreter,
const UserSettingsControllerSP& usc_sp,
const char *current_prefix,
Stream &strm,
Error &err)
{
// Write out current prefix line.
StreamString prefix_line;
StreamString description;
uint32_t max_len = FindMaxNameLength (usc_sp->m_settings.global_settings);
int num_entries = usc_sp->m_settings.global_settings.size();
if (current_prefix && current_prefix[0])
strm.Printf ("\n'%s' variables:\n\n", current_prefix);
else
strm.Printf ("\nTop level variables:\n\n");
if (num_entries > 0)
{
// Write out all "global" variables.
for (int i = 0; i < num_entries; ++i)
{
DumpSettingEntry (interpreter, strm, max_len, usc_sp->m_settings.global_settings[i]);
}
}
num_entries = usc_sp->m_settings.instance_settings.size();
max_len = FindMaxNameLength (usc_sp->m_settings.instance_settings);
if (num_entries > 0)
{
// Write out all instance variables.
for (int i = 0; i < num_entries; ++i)
{
DumpSettingEntry (interpreter, strm, max_len, usc_sp->m_settings.instance_settings[i]);
}
}
// Now, recurse across all children.
int num_children = usc_sp->GetNumChildren();
for (int i = 0; i < num_children; ++i)
{
UserSettingsControllerSP child = usc_sp->GetChildAtIndex (i);
if (child)
{
ConstString child_prefix = child->GetLevelName();
if (current_prefix && current_prefix[0])
{
StreamString new_prefix;
new_prefix.Printf ("%s.%s", current_prefix, child_prefix.GetCString());
UserSettingsController::FindAllSettingsDescriptions (interpreter,
child,
new_prefix.GetData(),
strm,
err);
}
else
{
UserSettingsController::FindAllSettingsDescriptions (interpreter,
child,
child_prefix.GetCString(),
strm,
err);
}
}
}
}
void
UserSettingsController::FindSettingsDescriptions (CommandInterpreter &interpreter,
const UserSettingsControllerSP& usc_sp,
const char *current_prefix,
const char *search_name,
Stream &strm,
Error &err)
{
Args names = UserSettingsController::BreakNameIntoPieces (search_name);
int num_pieces = names.GetArgumentCount ();
if (num_pieces == 0)
return;
if (usc_sp->GetLevelName().GetLength() > 0)
{
ConstString prefix (names.GetArgumentAtIndex (0));
if (prefix != usc_sp->GetLevelName())
{
std::string parent_prefix;
usc_sp->BuildParentPrefix (parent_prefix);
err.SetErrorStringWithFormat ("Cannot find match for '%s.%s'\n", parent_prefix.c_str(),
prefix.GetCString());
return;
}
else
{
names.Shift();
--num_pieces;
}
}
// If there's nothing left then dump all global and instance descriptions for this root.
if (num_pieces == 0)
{
StreamString prefix_line;
StreamString description;
uint32_t max_len;
int num_entries = usc_sp->m_settings.global_settings.size();
max_len = FindMaxNameLength (usc_sp->m_settings.global_settings);
strm.Printf ("\n'%s' variables:\n\n", search_name);
if (num_entries > 0)
{
// Write out all "global" variables.
for (int i = 0; i < num_entries; ++i)
{
DumpSettingEntry (interpreter, strm, max_len, usc_sp->m_settings.global_settings[i]);
}
}
num_entries = usc_sp->m_settings.instance_settings.size();
max_len = FindMaxNameLength (usc_sp->m_settings.instance_settings);
if (num_entries > 0)
{
// Write out all instance variables.
for (int i = 0; i < num_entries; ++i)
{
DumpSettingEntry (interpreter, strm, max_len, usc_sp->m_settings.instance_settings[i]);
}
}
}
else if (num_pieces == 1)
{
ConstString var_name (names.GetArgumentAtIndex (0));
bool is_global = false;
const SettingEntry *setting_entry = usc_sp->GetGlobalEntry (var_name);
if (setting_entry == NULL)
setting_entry = usc_sp->GetInstanceEntry (var_name);
else
is_global = true;
// Check to see if it is a global or instance variable name.
if (setting_entry != NULL)
{
DumpSettingEntry (interpreter, strm, var_name.GetLength(), *setting_entry);
}
else
{
// It must be a child name.
int num_children = usc_sp->GetNumChildren();
bool found = false;
for (int i = 0; i < num_children && !found; ++i)
{
UserSettingsControllerSP child = usc_sp->GetChildAtIndex (i);
if (child)
{
ConstString child_prefix = child->GetLevelName();
if (child_prefix == var_name)
{
found = true;
Added new target instance settings for execution settings: Targets can now specify some additional parameters for when we debug executables that can help with plug-in selection: target.execution-level = auto | user | kernel target.execution-mode = auto | dynamic | static target.execution-os-type = auto | none | halted | live On some systems, the binaries that are created are the same wether you use them to debug a kernel, or a user space program. Many times inspecting an object file can reveal what an executable should be. For these cases we can now be a little more complete by specifying wether to detect all of these things automatically (inspect the main executable file and select a plug-in accordingly), or manually to force the selection of certain plug-ins. To do this we now allow the specficifation of wether one is debugging a user space program (target.execution-level = user) or a kernel program (target.execution-level = kernel). We can also specify if we want to debug a program where shared libraries are dynamically loaded using a DynamicLoader plug-in (target.execution-mode = dynamic), or wether we will treat all symbol files as already linked at the correct address (target.execution-mode = static). We can also specify if the inferior we are debugging is being debugged on a bare board (target.execution-os-type = none), or debugging an OS where we have a JTAG or other direct connection to the inferior stops the entire OS (target.execution-os-type = halted), or if we are debugging a program on something that has live debug services (target.execution-os-type = live). For the "target.execution-os-type = halted" mode, we will need to create ProcessHelper plug-ins that allow us to extract the process/thread and other OS information by reading/writing memory. This should allow LLDB to be used for a wide variety of debugging tasks and handle them all correctly. llvm-svn: 125815
2011-02-18 09:44:25 +08:00
UserSettingsController::FindSettingsDescriptions (interpreter,
child,
current_prefix,
var_name.GetCString(),
strm,
Added new target instance settings for execution settings: Targets can now specify some additional parameters for when we debug executables that can help with plug-in selection: target.execution-level = auto | user | kernel target.execution-mode = auto | dynamic | static target.execution-os-type = auto | none | halted | live On some systems, the binaries that are created are the same wether you use them to debug a kernel, or a user space program. Many times inspecting an object file can reveal what an executable should be. For these cases we can now be a little more complete by specifying wether to detect all of these things automatically (inspect the main executable file and select a plug-in accordingly), or manually to force the selection of certain plug-ins. To do this we now allow the specficifation of wether one is debugging a user space program (target.execution-level = user) or a kernel program (target.execution-level = kernel). We can also specify if we want to debug a program where shared libraries are dynamically loaded using a DynamicLoader plug-in (target.execution-mode = dynamic), or wether we will treat all symbol files as already linked at the correct address (target.execution-mode = static). We can also specify if the inferior we are debugging is being debugged on a bare board (target.execution-os-type = none), or debugging an OS where we have a JTAG or other direct connection to the inferior stops the entire OS (target.execution-os-type = halted), or if we are debugging a program on something that has live debug services (target.execution-os-type = live). For the "target.execution-os-type = halted" mode, we will need to create ProcessHelper plug-ins that allow us to extract the process/thread and other OS information by reading/writing memory. This should allow LLDB to be used for a wide variety of debugging tasks and handle them all correctly. llvm-svn: 125815
2011-02-18 09:44:25 +08:00
err);
}
}
}
if (!found)
{
std::string parent_prefix;
usc_sp->BuildParentPrefix (parent_prefix);
err.SetErrorStringWithFormat ("Cannot find match for '%s.%s'\n", parent_prefix.c_str(), search_name);
return;
}
}
}
else
{
// It must be a child name; find the child and call this function recursively on child.
ConstString child_name (names.GetArgumentAtIndex (0));
StreamString rest_of_search_name;
for (int i = 0; i < num_pieces; ++i)
{
rest_of_search_name.Printf ("%s", names.GetArgumentAtIndex (i));
if ((i + 1) < num_pieces)
rest_of_search_name.Printf (".");
}
int num_children = usc_sp->GetNumChildren();
bool found = false;
for (int i = 0; i < num_children && !found; ++i)
{
UserSettingsControllerSP child = usc_sp->GetChildAtIndex (i);
if (child)
{
ConstString child_prefix = child->GetLevelName();
if (child_prefix == child_name)
{
found = true;
UserSettingsController::FindSettingsDescriptions (interpreter, child, current_prefix,
rest_of_search_name.GetData(), strm,
err);
}
}
}
if (!found)
{
std::string parent_prefix;
usc_sp->BuildParentPrefix (parent_prefix);
err.SetErrorStringWithFormat ("Cannot find match for '%s.%s'\n", parent_prefix.c_str(), search_name);
return;
}
}
}
void
UserSettingsController::SearchAllSettingsDescriptions (CommandInterpreter &interpreter,
const UserSettingsControllerSP& usc_sp,
const char *current_prefix,
const char *search_word,
Stream &strm)
{
if ((search_word == NULL) || (strlen (search_word) == 0))
return;
int num_entries = usc_sp->m_settings.global_settings.size();
if (num_entries > 0)
{
for (int i = 0; i < num_entries; ++i)
{
const SettingEntry &entry = usc_sp->m_settings.global_settings[i];
if (strcasestr (entry.description, search_word) != NULL)
{
StreamString var_name;
if (current_prefix && current_prefix[0])
var_name.Printf ("%s.%s", current_prefix, entry.var_name);
else
var_name.Printf ("%s", entry.var_name);
interpreter.OutputFormattedHelpText (strm, var_name.GetData(), "--", entry.description,
var_name.GetSize());
}
}
}
num_entries = usc_sp->m_settings.instance_settings.size();
if (num_entries > 0)
{
for (int i = 0; i < num_entries; ++i)
{
SettingEntry &entry = usc_sp->m_settings.instance_settings[i];
if (strcasestr (entry.description, search_word) != NULL)
{
StreamString var_name;
if (current_prefix && current_prefix[0])
var_name.Printf ("%s.%s", current_prefix, entry.var_name);
else
var_name.Printf ("%s", entry.var_name);
interpreter.OutputFormattedHelpText (strm,
var_name.GetData(),
"--",
entry.description,
var_name.GetSize());
}
}
}
int num_children = usc_sp->GetNumChildren ();
for (int i = 0; i < num_children; ++i)
{
UserSettingsControllerSP child = usc_sp->GetChildAtIndex (i);
if (child)
{
ConstString child_prefix = child->GetLevelName();
if (current_prefix && current_prefix[0])
{
StreamString new_prefix;
new_prefix.Printf ("%s.%s", current_prefix, child_prefix.GetCString());
UserSettingsController::SearchAllSettingsDescriptions (interpreter,
child,
new_prefix.GetData(),
search_word,
strm);
}
else
{
UserSettingsController::SearchAllSettingsDescriptions (interpreter,
child,
child_prefix.GetCString(),
search_word,
strm);
}
}
}
}
bool
UserSettingsController::DumpValue (CommandInterpreter &interpreter,
const UserSettingsControllerSP& usc_sp,
const char *variable_dot_name,
Stream &strm)
{
SettableVariableType var_type;
Error err;
StringList value = usc_sp->GetVariable (variable_dot_name,
var_type,
interpreter.GetDebugger().GetInstanceName().GetCString(),
err);
if (err.Success())
return DumpValue (variable_dot_name, var_type, value, strm);
return false;
}
bool
UserSettingsController::DumpValue (const char *variable_dot_name,
SettableVariableType var_type,
const StringList &value,
Stream &strm)
{
const char *type_name = UserSettingsController::GetTypeString (var_type);
strm.Printf ("%s (%s) = ", variable_dot_name, type_name);
if (value.GetSize() == 0)
{
strm.EOL();
}
else
{
switch (var_type)
{
case eSetVarTypeNone:
case eSetVarTypeEnum:
case eSetVarTypeInt:
case eSetVarTypeBoolean:
strm.Printf ("%s\n", value.GetStringAtIndex (0));
break;
case eSetVarTypeString:
strm.Printf ("\"%s\"\n", value.GetStringAtIndex (0));
break;
case eSetVarTypeArray:
{
strm.EOL();
for (unsigned i = 0, e = value.GetSize(); i != e; ++i)
strm.Printf (" [%u]: \"%s\"\n", i, value.GetStringAtIndex (i));
}
break;
case eSetVarTypeDictionary:
{
strm.EOL();
for (unsigned i = 0, e = value.GetSize(); i != e; ++i)
strm.Printf (" %s\n", value.GetStringAtIndex (i));
}
break;
default:
return false;
}
}
return true;
}
void
UserSettingsController::GetAllVariableValues (CommandInterpreter &interpreter,
const UserSettingsControllerSP& usc_sp,
const char *current_prefix,
Stream &strm,
Error &err)
{
StreamString description;
int num_entries = usc_sp->m_settings.global_settings.size();
for (int i = 0; i < num_entries; ++i)
{
StreamString full_var_name;
const SettingEntry &entry = usc_sp->m_settings.global_settings[i];
if (current_prefix && current_prefix[0])
full_var_name.Printf ("%s.%s", current_prefix, entry.var_name);
else
full_var_name.Printf ("%s", entry.var_name);
DumpValue (interpreter, usc_sp, full_var_name.GetData(), strm);
}
usc_sp->GetAllInstanceVariableValues (interpreter, strm);
usc_sp->GetAllPendingSettingValues (strm);
if (usc_sp->GetLevelName().GetLength() > 0) // Don't bother with default values for Debugger level.
usc_sp->GetAllDefaultSettingValues (strm);
// Now, recurse across all children.
int num_children = usc_sp->GetNumChildren();
for (int i = 0; i < num_children; ++i)
{
UserSettingsControllerSP child = usc_sp->GetChildAtIndex (i);
if (child)
{
ConstString child_prefix = child->GetLevelName();
if (current_prefix && current_prefix[0])
{
StreamString new_prefix;
new_prefix.Printf ("%s.%s", current_prefix, child_prefix.GetCString());
UserSettingsController::GetAllVariableValues (interpreter,
child,
new_prefix.GetData(),
strm,
err);
}
else
{
UserSettingsController::GetAllVariableValues (interpreter,
child,
child_prefix.GetCString(),
strm,
err);
}
}
}
}
Args
UserSettingsController::BreakNameIntoPieces (const char *full_dot_name)
{
Args return_value;
std::string name_string (full_dot_name);
bool done = false;
std::string piece;
std::string remainder (full_dot_name);
while (!done)
{
size_t idx = remainder.find_first_of ('.');
piece = remainder.substr (0, idx);
return_value.AppendArgument (piece.c_str());
if (idx != std::string::npos)
remainder = remainder.substr (idx+1);
else
done = true;
}
return return_value;
}
bool
UserSettingsController::IsLiveInstance (const std::string &instance_name)
{
Mutex::Locker locker (m_live_settings_mutex);
InstanceSettingsMap::iterator pos = m_live_settings.find (instance_name);
if (pos != m_live_settings.end())
return true;
return false;
}
int
UserSettingsController::CompleteSettingsValue (const UserSettingsControllerSP& usc_sp,
const char *full_dot_name,
const char *partial_value,
bool &word_complete,
StringList &matches)
{
Args names = UserSettingsController::BreakNameIntoPieces (full_dot_name);
int num_pieces = names.GetArgumentCount();
word_complete = true;
ConstString root_level = usc_sp->GetLevelName();
int num_extra_levels = num_pieces - 2;
if ((num_extra_levels > 0)
&& root_level.GetLength() > 0)
{
ConstString current_level (names.GetArgumentAtIndex (0));
if (current_level == root_level)
{
names.Shift();
--num_extra_levels;
}
else
return 0;
}
for (int i = 0; i < num_extra_levels; ++i)
{
ConstString child_level (names.GetArgumentAtIndex (0));
bool found = false;
int num_children = usc_sp->GetNumChildren();
UserSettingsControllerSP child_usc_sp = usc_sp;
for (int j = 0; j < num_children && !found; ++j)
{
if (child_usc_sp->GetChildAtIndex (j)->GetLevelName() == child_level)
{
found = true;
child_usc_sp = child_usc_sp->GetChildAtIndex (j);
names.Shift();
}
}
if (!found)
return 0;
}
if (names.GetArgumentCount() != 2)
return 0;
std::string next_name (names.GetArgumentAtIndex (0));
int len = next_name.length();
names.Shift();
if ((next_name[0] == '[') && (next_name[len-1] == ']'))
{
// 'next_name' is instance name. Instance names are irrelevent here.
}
else
{
// 'next_name' is child name.
bool found = false;
int num_children = usc_sp->GetNumChildren();
ConstString child_level (next_name.c_str());
UserSettingsControllerSP child_usc_sp = usc_sp;
for (int j = 0; j < num_children && !found; ++j)
{
if (child_usc_sp->GetChildAtIndex (j)->GetLevelName() == child_level)
{
found = true;
child_usc_sp = child_usc_sp->GetChildAtIndex (j);
}
}
if (!found)
return 0;
}
ConstString var_name (names.GetArgumentAtIndex(0));
const SettingEntry *entry = usc_sp->GetGlobalEntry (var_name);
if (entry == NULL)
entry = usc_sp->GetInstanceEntry (var_name);
if (entry == NULL)
return 0;
Added new target instance settings for execution settings: Targets can now specify some additional parameters for when we debug executables that can help with plug-in selection: target.execution-level = auto | user | kernel target.execution-mode = auto | dynamic | static target.execution-os-type = auto | none | halted | live On some systems, the binaries that are created are the same wether you use them to debug a kernel, or a user space program. Many times inspecting an object file can reveal what an executable should be. For these cases we can now be a little more complete by specifying wether to detect all of these things automatically (inspect the main executable file and select a plug-in accordingly), or manually to force the selection of certain plug-ins. To do this we now allow the specficifation of wether one is debugging a user space program (target.execution-level = user) or a kernel program (target.execution-level = kernel). We can also specify if we want to debug a program where shared libraries are dynamically loaded using a DynamicLoader plug-in (target.execution-mode = dynamic), or wether we will treat all symbol files as already linked at the correct address (target.execution-mode = static). We can also specify if the inferior we are debugging is being debugged on a bare board (target.execution-os-type = none), or debugging an OS where we have a JTAG or other direct connection to the inferior stops the entire OS (target.execution-os-type = halted), or if we are debugging a program on something that has live debug services (target.execution-os-type = live). For the "target.execution-os-type = halted" mode, we will need to create ProcessHelper plug-ins that allow us to extract the process/thread and other OS information by reading/writing memory. This should allow LLDB to be used for a wide variety of debugging tasks and handle them all correctly. llvm-svn: 125815
2011-02-18 09:44:25 +08:00
if (entry->var_type == eSetVarTypeBoolean)
return UserSettingsController::BooleanMatches (partial_value, word_complete, matches);
Added new target instance settings for execution settings: Targets can now specify some additional parameters for when we debug executables that can help with plug-in selection: target.execution-level = auto | user | kernel target.execution-mode = auto | dynamic | static target.execution-os-type = auto | none | halted | live On some systems, the binaries that are created are the same wether you use them to debug a kernel, or a user space program. Many times inspecting an object file can reveal what an executable should be. For these cases we can now be a little more complete by specifying wether to detect all of these things automatically (inspect the main executable file and select a plug-in accordingly), or manually to force the selection of certain plug-ins. To do this we now allow the specficifation of wether one is debugging a user space program (target.execution-level = user) or a kernel program (target.execution-level = kernel). We can also specify if we want to debug a program where shared libraries are dynamically loaded using a DynamicLoader plug-in (target.execution-mode = dynamic), or wether we will treat all symbol files as already linked at the correct address (target.execution-mode = static). We can also specify if the inferior we are debugging is being debugged on a bare board (target.execution-os-type = none), or debugging an OS where we have a JTAG or other direct connection to the inferior stops the entire OS (target.execution-os-type = halted), or if we are debugging a program on something that has live debug services (target.execution-os-type = live). For the "target.execution-os-type = halted" mode, we will need to create ProcessHelper plug-ins that allow us to extract the process/thread and other OS information by reading/writing memory. This should allow LLDB to be used for a wide variety of debugging tasks and handle them all correctly. llvm-svn: 125815
2011-02-18 09:44:25 +08:00
else if (entry->var_type == eSetVarTypeEnum)
return UserSettingsController::EnumMatches (partial_value, entry->enum_values, word_complete, matches);
else
return 0;
}
int
UserSettingsController::BooleanMatches (const char *partial_value,
bool &word_complete,
StringList &matches)
{
static const std::string true_string ("true");
static const std::string false_string ("false");
if (partial_value == NULL)
{
matches.AppendString ("true");
matches.AppendString ("false");
}
else
{
int partial_len = strlen (partial_value);
if ((partial_len <= true_string.length())
&& (true_string.find (partial_value) == 0))
matches.AppendString ("true");
else if ((partial_len <= false_string.length())
&& (false_string.find (partial_value) == 0))
matches.AppendString ("false");
}
word_complete = false;
if (matches.GetSize() == 1)
word_complete = true;
return matches.GetSize();
}
int
UserSettingsController::EnumMatches (const char *partial_value,
Added new target instance settings for execution settings: Targets can now specify some additional parameters for when we debug executables that can help with plug-in selection: target.execution-level = auto | user | kernel target.execution-mode = auto | dynamic | static target.execution-os-type = auto | none | halted | live On some systems, the binaries that are created are the same wether you use them to debug a kernel, or a user space program. Many times inspecting an object file can reveal what an executable should be. For these cases we can now be a little more complete by specifying wether to detect all of these things automatically (inspect the main executable file and select a plug-in accordingly), or manually to force the selection of certain plug-ins. To do this we now allow the specficifation of wether one is debugging a user space program (target.execution-level = user) or a kernel program (target.execution-level = kernel). We can also specify if we want to debug a program where shared libraries are dynamically loaded using a DynamicLoader plug-in (target.execution-mode = dynamic), or wether we will treat all symbol files as already linked at the correct address (target.execution-mode = static). We can also specify if the inferior we are debugging is being debugged on a bare board (target.execution-os-type = none), or debugging an OS where we have a JTAG or other direct connection to the inferior stops the entire OS (target.execution-os-type = halted), or if we are debugging a program on something that has live debug services (target.execution-os-type = live). For the "target.execution-os-type = halted" mode, we will need to create ProcessHelper plug-ins that allow us to extract the process/thread and other OS information by reading/writing memory. This should allow LLDB to be used for a wide variety of debugging tasks and handle them all correctly. llvm-svn: 125815
2011-02-18 09:44:25 +08:00
OptionEnumValueElement *enum_values,
bool &word_complete,
StringList &matches)
{
int len = (partial_value != NULL) ? strlen (partial_value) : 0;
int i = 0;
while (enum_values[i].string_value != NULL)
{
if (len == 0)
matches.AppendString (enum_values[i].string_value);
else
{
std::string tmp_value (enum_values[i].string_value);
if ((len <= tmp_value.length())
&& tmp_value.find (partial_value) == 0)
matches.AppendString (enum_values[i].string_value);
}
++i;
}
word_complete = false;
if (matches.GetSize() == 1)
word_complete = true;
return matches.GetSize();
}
int
UserSettingsController::CompleteSettingsNames (const UserSettingsControllerSP& usc_sp,
Args &partial_setting_name_pieces,
bool &word_complete,
StringList &matches)
{
int num_matches = 0;
int num_name_pieces = partial_setting_name_pieces.GetArgumentCount();
if (num_name_pieces > 1)
{
// There are at least two pieces, perhaps with multiple level names preceding them.
// First traverse all the extra levels, until we have exactly two pieces left.
int num_extra_levels = num_name_pieces - 2;
// Deal with current level first.
ConstString root_level = usc_sp->GetLevelName();
if ((num_extra_levels > 0)
&& (root_level.GetLength() > 0))
{
ConstString current_level (partial_setting_name_pieces.GetArgumentAtIndex (0));
if (current_level == root_level)
{
partial_setting_name_pieces.Shift();
--num_extra_levels;
}
else
return 0; // The current level did not match the name pieces; something is wrong, so return immediately
}
for (int i = 0; i < num_extra_levels; ++i)
{
ConstString child_level (partial_setting_name_pieces.GetArgumentAtIndex (0));
bool found = false;
int num_children = usc_sp->GetNumChildren();
UserSettingsControllerSP child_usc_sp = usc_sp;
for (int j = 0; j < num_children && !found; ++j)
{
if (child_usc_sp->GetChildAtIndex (j)->GetLevelName() == child_level)
{
found = true;
child_usc_sp = child_usc_sp->GetChildAtIndex (j);
partial_setting_name_pieces.Shift();
}
}
if (! found)
{
return 0; // Unable to find a matching child level name; something is wrong, so return immediately.
}
}
// Now there should be exactly two name pieces left. If not there is an error, so return immediately
if (partial_setting_name_pieces.GetArgumentCount() != 2)
return 0;
std::string next_name (partial_setting_name_pieces.GetArgumentAtIndex (0));
int len = next_name.length();
partial_setting_name_pieces.Shift();
if ((next_name[0] == '[') && (next_name[len-1] == ']'))
{
// 'next_name' is an instance name. The last name piece must be a non-empty partial match against an
// instance_name, assuming 'next_name' is valid.
if (usc_sp->IsLiveInstance (next_name))
{
std::string complete_prefix;
usc_sp->BuildParentPrefix (complete_prefix);
num_matches = usc_sp->InstanceVariableMatches(partial_setting_name_pieces.GetArgumentAtIndex(0),
complete_prefix,
next_name.c_str(),
matches);
word_complete = true;
if (num_matches > 1)
word_complete = false;
return num_matches;
}
else
return 0; // Invalid instance_name
}
else
{
// 'next_name' must be a child name. Find the correct child and pass the remaining piece to be resolved.
bool found = false;
int num_children = usc_sp->GetNumChildren();
ConstString child_level (next_name.c_str());
for (int i = 0; i < num_children; ++i)
{
if (usc_sp->GetChildAtIndex (i)->GetLevelName() == child_level)
{
found = true;
return UserSettingsController::CompleteSettingsNames (usc_sp->GetChildAtIndex (i),
partial_setting_name_pieces,
word_complete, matches);
}
}
if (!found)
return 0;
}
}
else if (num_name_pieces == 1)
{
std::string complete_prefix;
usc_sp->BuildParentPrefix (complete_prefix);
word_complete = true;
std::string name (partial_setting_name_pieces.GetArgumentAtIndex (0));
if (name[0] == '[')
{
// It's a partial instance name.
num_matches = usc_sp->LiveInstanceMatches (name.c_str(), complete_prefix, word_complete, matches);
}
else
{
// It could be anything *except* an instance name...
num_matches = usc_sp->GlobalVariableMatches (name.c_str(), complete_prefix, matches);
num_matches += usc_sp->InstanceVariableMatches (name.c_str(), complete_prefix, NULL, matches);
num_matches += usc_sp->ChildMatches (name.c_str(), complete_prefix, word_complete, matches);
}
if (num_matches > 1)
word_complete = false;
return num_matches;
}
else
{
// We have a user settings controller with a blank partial string. Return everything possible at this level.
std::string complete_prefix;
usc_sp->BuildParentPrefix (complete_prefix);
num_matches = usc_sp->GlobalVariableMatches (NULL, complete_prefix, matches);
num_matches += usc_sp->InstanceVariableMatches (NULL, complete_prefix, NULL, matches);
num_matches += usc_sp->LiveInstanceMatches (NULL, complete_prefix, word_complete, matches);
num_matches += usc_sp->ChildMatches (NULL, complete_prefix, word_complete, matches);
word_complete = false;
return num_matches;
}
return num_matches;
}
int
UserSettingsController::GlobalVariableMatches (const char *partial_name,
const std::string &complete_prefix,
StringList &matches)
{
int partial_len = (partial_name != NULL) ? strlen (partial_name) : 0;
int num_matches = 0;
for (size_t i = 0; i < m_settings.global_settings.size(); ++i)
{
Added new target instance settings for execution settings: Targets can now specify some additional parameters for when we debug executables that can help with plug-in selection: target.execution-level = auto | user | kernel target.execution-mode = auto | dynamic | static target.execution-os-type = auto | none | halted | live On some systems, the binaries that are created are the same wether you use them to debug a kernel, or a user space program. Many times inspecting an object file can reveal what an executable should be. For these cases we can now be a little more complete by specifying wether to detect all of these things automatically (inspect the main executable file and select a plug-in accordingly), or manually to force the selection of certain plug-ins. To do this we now allow the specficifation of wether one is debugging a user space program (target.execution-level = user) or a kernel program (target.execution-level = kernel). We can also specify if we want to debug a program where shared libraries are dynamically loaded using a DynamicLoader plug-in (target.execution-mode = dynamic), or wether we will treat all symbol files as already linked at the correct address (target.execution-mode = static). We can also specify if the inferior we are debugging is being debugged on a bare board (target.execution-os-type = none), or debugging an OS where we have a JTAG or other direct connection to the inferior stops the entire OS (target.execution-os-type = halted), or if we are debugging a program on something that has live debug services (target.execution-os-type = live). For the "target.execution-os-type = halted" mode, we will need to create ProcessHelper plug-ins that allow us to extract the process/thread and other OS information by reading/writing memory. This should allow LLDB to be used for a wide variety of debugging tasks and handle them all correctly. llvm-svn: 125815
2011-02-18 09:44:25 +08:00
const SettingEntry &entry = m_settings.global_settings[i];
std::string var_name (entry.var_name);
if ((partial_len == 0)
|| ((partial_len <= var_name.length())
&& (var_name.find (partial_name) == 0)))
{
StreamString match_name;
if (complete_prefix.length() > 0)
{
match_name.Printf ("%s.%s", complete_prefix.c_str(), var_name.c_str());
matches.AppendString (match_name.GetData());
}
else
matches.AppendString (var_name.c_str());
++num_matches;
}
}
return num_matches;
}
int
UserSettingsController::InstanceVariableMatches (const char *partial_name,
const std::string &complete_prefix,
const char *instance_name,
StringList &matches)
{
int partial_len = (partial_name != NULL) ? strlen (partial_name) : 0;
int num_matches = 0;
for (size_t i = 0; i < m_settings.instance_settings.size(); ++i)
{
SettingEntry &entry = m_settings.instance_settings[i];
std::string var_name (entry.var_name);
if ((partial_len == 0)
|| ((partial_len <= var_name.length())
&& (var_name.find (partial_name) == 0)))
{
StreamString match_name;
if (complete_prefix.length() > 0)
{
if (instance_name != NULL)
match_name.Printf ("%s.%s.%s", complete_prefix.c_str(), instance_name, var_name.c_str());
else
match_name.Printf ("%s.%s", complete_prefix.c_str(), var_name.c_str());
matches.AppendString (match_name.GetData());
}
else
{
if (instance_name != NULL)
{
match_name.Printf ("%s.%s", instance_name, var_name.c_str());
matches.AppendString (match_name.GetData());
}
else
matches.AppendString (var_name.c_str());
}
++num_matches;
}
}
return num_matches;
}
int
UserSettingsController::LiveInstanceMatches (const char *partial_name,
const std::string &complete_prefix,
bool &word_complete,
StringList &matches)
{
int partial_len = (partial_name != NULL) ? strlen (partial_name) : 0;
int num_matches = 0;
InstanceSettingsMap::iterator pos;
Mutex::Locker locker (m_live_settings_mutex);
for (pos = m_live_settings.begin(); pos != m_live_settings.end(); ++pos)
{
std::string instance_name = pos->first;
if ((partial_len == 0)
|| ((partial_len <= instance_name.length())
&& (instance_name.find (partial_name) == 0)))
{
StreamString match_name;
if (complete_prefix.length() > 0)
match_name.Printf ("%s.%s.", complete_prefix.c_str(), instance_name.c_str());
else
match_name.Printf ("%s.", instance_name.c_str());
matches.AppendString (match_name.GetData());
++num_matches;
}
}
if (num_matches > 0)
word_complete = false;
return num_matches;
}
int
UserSettingsController::ChildMatches (const char *partial_name,
const std::string &complete_prefix,
bool &word_complete,
StringList &matches)
{
int partial_len = (partial_name != NULL) ? strlen (partial_name) : 0;
int num_children = GetNumChildren();
int num_matches = 0;
for (int i = 0; i < num_children; ++i)
{
std::string child_name (GetChildAtIndex(i)->GetLevelName().GetCString());
StreamString match_name;
if ((partial_len == 0)
|| ((partial_len <= child_name.length())
&& (child_name.find (partial_name) == 0)))
{
if (complete_prefix.length() > 0)
match_name.Printf ("%s.%s.", complete_prefix.c_str(), child_name.c_str());
else
match_name.Printf ("%s.", child_name.c_str());
matches.AppendString (match_name.GetData());
++num_matches;
}
}
if (num_matches > 0)
word_complete = false;
return num_matches;
}
void
Added new target instance settings for execution settings: Targets can now specify some additional parameters for when we debug executables that can help with plug-in selection: target.execution-level = auto | user | kernel target.execution-mode = auto | dynamic | static target.execution-os-type = auto | none | halted | live On some systems, the binaries that are created are the same wether you use them to debug a kernel, or a user space program. Many times inspecting an object file can reveal what an executable should be. For these cases we can now be a little more complete by specifying wether to detect all of these things automatically (inspect the main executable file and select a plug-in accordingly), or manually to force the selection of certain plug-ins. To do this we now allow the specficifation of wether one is debugging a user space program (target.execution-level = user) or a kernel program (target.execution-level = kernel). We can also specify if we want to debug a program where shared libraries are dynamically loaded using a DynamicLoader plug-in (target.execution-mode = dynamic), or wether we will treat all symbol files as already linked at the correct address (target.execution-mode = static). We can also specify if the inferior we are debugging is being debugged on a bare board (target.execution-os-type = none), or debugging an OS where we have a JTAG or other direct connection to the inferior stops the entire OS (target.execution-os-type = halted), or if we are debugging a program on something that has live debug services (target.execution-os-type = live). For the "target.execution-os-type = halted" mode, we will need to create ProcessHelper plug-ins that allow us to extract the process/thread and other OS information by reading/writing memory. This should allow LLDB to be used for a wide variety of debugging tasks and handle them all correctly. llvm-svn: 125815
2011-02-18 09:44:25 +08:00
UserSettingsController::VerifyOperationForType (SettableVariableType var_type,
VarSetOperationType op,
const ConstString &var_name,
Error &err)
{
Added new target instance settings for execution settings: Targets can now specify some additional parameters for when we debug executables that can help with plug-in selection: target.execution-level = auto | user | kernel target.execution-mode = auto | dynamic | static target.execution-os-type = auto | none | halted | live On some systems, the binaries that are created are the same wether you use them to debug a kernel, or a user space program. Many times inspecting an object file can reveal what an executable should be. For these cases we can now be a little more complete by specifying wether to detect all of these things automatically (inspect the main executable file and select a plug-in accordingly), or manually to force the selection of certain plug-ins. To do this we now allow the specficifation of wether one is debugging a user space program (target.execution-level = user) or a kernel program (target.execution-level = kernel). We can also specify if we want to debug a program where shared libraries are dynamically loaded using a DynamicLoader plug-in (target.execution-mode = dynamic), or wether we will treat all symbol files as already linked at the correct address (target.execution-mode = static). We can also specify if the inferior we are debugging is being debugged on a bare board (target.execution-os-type = none), or debugging an OS where we have a JTAG or other direct connection to the inferior stops the entire OS (target.execution-os-type = halted), or if we are debugging a program on something that has live debug services (target.execution-os-type = live). For the "target.execution-os-type = halted" mode, we will need to create ProcessHelper plug-ins that allow us to extract the process/thread and other OS information by reading/writing memory. This should allow LLDB to be used for a wide variety of debugging tasks and handle them all correctly. llvm-svn: 125815
2011-02-18 09:44:25 +08:00
if (op == eVarSetOperationAssign)
return;
Added new target instance settings for execution settings: Targets can now specify some additional parameters for when we debug executables that can help with plug-in selection: target.execution-level = auto | user | kernel target.execution-mode = auto | dynamic | static target.execution-os-type = auto | none | halted | live On some systems, the binaries that are created are the same wether you use them to debug a kernel, or a user space program. Many times inspecting an object file can reveal what an executable should be. For these cases we can now be a little more complete by specifying wether to detect all of these things automatically (inspect the main executable file and select a plug-in accordingly), or manually to force the selection of certain plug-ins. To do this we now allow the specficifation of wether one is debugging a user space program (target.execution-level = user) or a kernel program (target.execution-level = kernel). We can also specify if we want to debug a program where shared libraries are dynamically loaded using a DynamicLoader plug-in (target.execution-mode = dynamic), or wether we will treat all symbol files as already linked at the correct address (target.execution-mode = static). We can also specify if the inferior we are debugging is being debugged on a bare board (target.execution-os-type = none), or debugging an OS where we have a JTAG or other direct connection to the inferior stops the entire OS (target.execution-os-type = halted), or if we are debugging a program on something that has live debug services (target.execution-os-type = live). For the "target.execution-os-type = halted" mode, we will need to create ProcessHelper plug-ins that allow us to extract the process/thread and other OS information by reading/writing memory. This should allow LLDB to be used for a wide variety of debugging tasks and handle them all correctly. llvm-svn: 125815
2011-02-18 09:44:25 +08:00
if (op == eVarSetOperationInvalid)
{
err.SetErrorString ("Invalid 'settings ' subcommand operation.\n");
return;
}
switch (op)
{
Added new target instance settings for execution settings: Targets can now specify some additional parameters for when we debug executables that can help with plug-in selection: target.execution-level = auto | user | kernel target.execution-mode = auto | dynamic | static target.execution-os-type = auto | none | halted | live On some systems, the binaries that are created are the same wether you use them to debug a kernel, or a user space program. Many times inspecting an object file can reveal what an executable should be. For these cases we can now be a little more complete by specifying wether to detect all of these things automatically (inspect the main executable file and select a plug-in accordingly), or manually to force the selection of certain plug-ins. To do this we now allow the specficifation of wether one is debugging a user space program (target.execution-level = user) or a kernel program (target.execution-level = kernel). We can also specify if we want to debug a program where shared libraries are dynamically loaded using a DynamicLoader plug-in (target.execution-mode = dynamic), or wether we will treat all symbol files as already linked at the correct address (target.execution-mode = static). We can also specify if the inferior we are debugging is being debugged on a bare board (target.execution-os-type = none), or debugging an OS where we have a JTAG or other direct connection to the inferior stops the entire OS (target.execution-os-type = halted), or if we are debugging a program on something that has live debug services (target.execution-os-type = live). For the "target.execution-os-type = halted" mode, we will need to create ProcessHelper plug-ins that allow us to extract the process/thread and other OS information by reading/writing memory. This should allow LLDB to be used for a wide variety of debugging tasks and handle them all correctly. llvm-svn: 125815
2011-02-18 09:44:25 +08:00
case eVarSetOperationInsertBefore:
case eVarSetOperationInsertAfter:
if (var_type != eSetVarTypeArray)
err.SetErrorString ("Invalid operation: This operation can only be performed on array variables.\n");
break;
Added new target instance settings for execution settings: Targets can now specify some additional parameters for when we debug executables that can help with plug-in selection: target.execution-level = auto | user | kernel target.execution-mode = auto | dynamic | static target.execution-os-type = auto | none | halted | live On some systems, the binaries that are created are the same wether you use them to debug a kernel, or a user space program. Many times inspecting an object file can reveal what an executable should be. For these cases we can now be a little more complete by specifying wether to detect all of these things automatically (inspect the main executable file and select a plug-in accordingly), or manually to force the selection of certain plug-ins. To do this we now allow the specficifation of wether one is debugging a user space program (target.execution-level = user) or a kernel program (target.execution-level = kernel). We can also specify if we want to debug a program where shared libraries are dynamically loaded using a DynamicLoader plug-in (target.execution-mode = dynamic), or wether we will treat all symbol files as already linked at the correct address (target.execution-mode = static). We can also specify if the inferior we are debugging is being debugged on a bare board (target.execution-os-type = none), or debugging an OS where we have a JTAG or other direct connection to the inferior stops the entire OS (target.execution-os-type = halted), or if we are debugging a program on something that has live debug services (target.execution-os-type = live). For the "target.execution-os-type = halted" mode, we will need to create ProcessHelper plug-ins that allow us to extract the process/thread and other OS information by reading/writing memory. This should allow LLDB to be used for a wide variety of debugging tasks and handle them all correctly. llvm-svn: 125815
2011-02-18 09:44:25 +08:00
case eVarSetOperationReplace:
case eVarSetOperationRemove:
if ((var_type != eSetVarTypeArray)
&& (var_type != eSetVarTypeDictionary))
err.SetErrorString ("Invalid operation: This operation can only be performed on array or dictionary"
" variables.\n");
break;
Added new target instance settings for execution settings: Targets can now specify some additional parameters for when we debug executables that can help with plug-in selection: target.execution-level = auto | user | kernel target.execution-mode = auto | dynamic | static target.execution-os-type = auto | none | halted | live On some systems, the binaries that are created are the same wether you use them to debug a kernel, or a user space program. Many times inspecting an object file can reveal what an executable should be. For these cases we can now be a little more complete by specifying wether to detect all of these things automatically (inspect the main executable file and select a plug-in accordingly), or manually to force the selection of certain plug-ins. To do this we now allow the specficifation of wether one is debugging a user space program (target.execution-level = user) or a kernel program (target.execution-level = kernel). We can also specify if we want to debug a program where shared libraries are dynamically loaded using a DynamicLoader plug-in (target.execution-mode = dynamic), or wether we will treat all symbol files as already linked at the correct address (target.execution-mode = static). We can also specify if the inferior we are debugging is being debugged on a bare board (target.execution-os-type = none), or debugging an OS where we have a JTAG or other direct connection to the inferior stops the entire OS (target.execution-os-type = halted), or if we are debugging a program on something that has live debug services (target.execution-os-type = live). For the "target.execution-os-type = halted" mode, we will need to create ProcessHelper plug-ins that allow us to extract the process/thread and other OS information by reading/writing memory. This should allow LLDB to be used for a wide variety of debugging tasks and handle them all correctly. llvm-svn: 125815
2011-02-18 09:44:25 +08:00
case eVarSetOperationAppend:
case eVarSetOperationClear:
if ((var_type != eSetVarTypeArray)
&& (var_type != eSetVarTypeDictionary)
&& (var_type != eSetVarTypeString))
err.SetErrorString ("Invalid operation: This operation can only be performed on array, dictionary "
"or string variables.\n");
break;
default:
break;
}
return;
}
void
Added new target instance settings for execution settings: Targets can now specify some additional parameters for when we debug executables that can help with plug-in selection: target.execution-level = auto | user | kernel target.execution-mode = auto | dynamic | static target.execution-os-type = auto | none | halted | live On some systems, the binaries that are created are the same wether you use them to debug a kernel, or a user space program. Many times inspecting an object file can reveal what an executable should be. For these cases we can now be a little more complete by specifying wether to detect all of these things automatically (inspect the main executable file and select a plug-in accordingly), or manually to force the selection of certain plug-ins. To do this we now allow the specficifation of wether one is debugging a user space program (target.execution-level = user) or a kernel program (target.execution-level = kernel). We can also specify if we want to debug a program where shared libraries are dynamically loaded using a DynamicLoader plug-in (target.execution-mode = dynamic), or wether we will treat all symbol files as already linked at the correct address (target.execution-mode = static). We can also specify if the inferior we are debugging is being debugged on a bare board (target.execution-os-type = none), or debugging an OS where we have a JTAG or other direct connection to the inferior stops the entire OS (target.execution-os-type = halted), or if we are debugging a program on something that has live debug services (target.execution-os-type = live). For the "target.execution-os-type = halted" mode, we will need to create ProcessHelper plug-ins that allow us to extract the process/thread and other OS information by reading/writing memory. This should allow LLDB to be used for a wide variety of debugging tasks and handle them all correctly. llvm-svn: 125815
2011-02-18 09:44:25 +08:00
UserSettingsController::UpdateStringVariable (VarSetOperationType op,
std::string &string_var,
const char *new_value,
Error &err)
{
Added new target instance settings for execution settings: Targets can now specify some additional parameters for when we debug executables that can help with plug-in selection: target.execution-level = auto | user | kernel target.execution-mode = auto | dynamic | static target.execution-os-type = auto | none | halted | live On some systems, the binaries that are created are the same wether you use them to debug a kernel, or a user space program. Many times inspecting an object file can reveal what an executable should be. For these cases we can now be a little more complete by specifying wether to detect all of these things automatically (inspect the main executable file and select a plug-in accordingly), or manually to force the selection of certain plug-ins. To do this we now allow the specficifation of wether one is debugging a user space program (target.execution-level = user) or a kernel program (target.execution-level = kernel). We can also specify if we want to debug a program where shared libraries are dynamically loaded using a DynamicLoader plug-in (target.execution-mode = dynamic), or wether we will treat all symbol files as already linked at the correct address (target.execution-mode = static). We can also specify if the inferior we are debugging is being debugged on a bare board (target.execution-os-type = none), or debugging an OS where we have a JTAG or other direct connection to the inferior stops the entire OS (target.execution-os-type = halted), or if we are debugging a program on something that has live debug services (target.execution-os-type = live). For the "target.execution-os-type = halted" mode, we will need to create ProcessHelper plug-ins that allow us to extract the process/thread and other OS information by reading/writing memory. This should allow LLDB to be used for a wide variety of debugging tasks and handle them all correctly. llvm-svn: 125815
2011-02-18 09:44:25 +08:00
if (op == eVarSetOperationAssign)
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 (new_value && new_value[0])
string_var.assign (new_value);
else
string_var.clear();
}
Added new target instance settings for execution settings: Targets can now specify some additional parameters for when we debug executables that can help with plug-in selection: target.execution-level = auto | user | kernel target.execution-mode = auto | dynamic | static target.execution-os-type = auto | none | halted | live On some systems, the binaries that are created are the same wether you use them to debug a kernel, or a user space program. Many times inspecting an object file can reveal what an executable should be. For these cases we can now be a little more complete by specifying wether to detect all of these things automatically (inspect the main executable file and select a plug-in accordingly), or manually to force the selection of certain plug-ins. To do this we now allow the specficifation of wether one is debugging a user space program (target.execution-level = user) or a kernel program (target.execution-level = kernel). We can also specify if we want to debug a program where shared libraries are dynamically loaded using a DynamicLoader plug-in (target.execution-mode = dynamic), or wether we will treat all symbol files as already linked at the correct address (target.execution-mode = static). We can also specify if the inferior we are debugging is being debugged on a bare board (target.execution-os-type = none), or debugging an OS where we have a JTAG or other direct connection to the inferior stops the entire OS (target.execution-os-type = halted), or if we are debugging a program on something that has live debug services (target.execution-os-type = live). For the "target.execution-os-type = halted" mode, we will need to create ProcessHelper plug-ins that allow us to extract the process/thread and other OS information by reading/writing memory. This should allow LLDB to be used for a wide variety of debugging tasks and handle them all correctly. llvm-svn: 125815
2011-02-18 09:44:25 +08:00
else if (op == eVarSetOperationAppend)
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 (new_value && new_value[0])
string_var.append (new_value);
}
Added new target instance settings for execution settings: Targets can now specify some additional parameters for when we debug executables that can help with plug-in selection: target.execution-level = auto | user | kernel target.execution-mode = auto | dynamic | static target.execution-os-type = auto | none | halted | live On some systems, the binaries that are created are the same wether you use them to debug a kernel, or a user space program. Many times inspecting an object file can reveal what an executable should be. For these cases we can now be a little more complete by specifying wether to detect all of these things automatically (inspect the main executable file and select a plug-in accordingly), or manually to force the selection of certain plug-ins. To do this we now allow the specficifation of wether one is debugging a user space program (target.execution-level = user) or a kernel program (target.execution-level = kernel). We can also specify if we want to debug a program where shared libraries are dynamically loaded using a DynamicLoader plug-in (target.execution-mode = dynamic), or wether we will treat all symbol files as already linked at the correct address (target.execution-mode = static). We can also specify if the inferior we are debugging is being debugged on a bare board (target.execution-os-type = none), or debugging an OS where we have a JTAG or other direct connection to the inferior stops the entire OS (target.execution-os-type = halted), or if we are debugging a program on something that has live debug services (target.execution-os-type = live). For the "target.execution-os-type = halted" mode, we will need to create ProcessHelper plug-ins that allow us to extract the process/thread and other OS information by reading/writing memory. This should allow LLDB to be used for a wide variety of debugging tasks and handle them all correctly. llvm-svn: 125815
2011-02-18 09:44:25 +08:00
else if (op == eVarSetOperationClear)
string_var.clear();
else
err.SetErrorString ("Unrecognized operation. Cannot update value.\n");
}
Error
UserSettingsController::UpdateStringOptionValue (const char *value,
VarSetOperationType op,
OptionValueString &option_value)
{
Error error;
if (op == eVarSetOperationAssign)
{
option_value.SetCurrentValue (value);
}
else if (op == eVarSetOperationAppend)
{
option_value.AppendToCurrentValue (value);
}
else if (op == eVarSetOperationClear)
{
option_value.Clear();
}
else
{
error.SetErrorString ("Unrecognized operation. Cannot update value.\n");
}
return error;
}
Error
UserSettingsController::UpdateFileSpecOptionValue (const char *value,
VarSetOperationType op,
OptionValueFileSpec &option_value)
{
Error error;
if (op == eVarSetOperationAssign)
{
option_value.GetCurrentValue().SetFile (value, false);
}
else if (op == eVarSetOperationAppend)
{
char path[PATH_MAX];
if (option_value.GetCurrentValue().GetPath (path, sizeof(path)))
{
int path_len = ::strlen (path);
int value_len = ::strlen (value);
if (value_len + 1 > sizeof(path) - path_len)
{
error.SetErrorString("path too long.");
}
else
{
::strncat (path, value, sizeof(path) - path_len - 1);
option_value.GetCurrentValue().SetFile (path, false);
}
}
else
{
error.SetErrorString("path too long.");
}
}
else if (op == eVarSetOperationClear)
{
option_value.Clear();
}
else
{
error.SetErrorString ("operation not supported for FileSpec option value type.");
}
return error;
}
void
Added new target instance settings for execution settings: Targets can now specify some additional parameters for when we debug executables that can help with plug-in selection: target.execution-level = auto | user | kernel target.execution-mode = auto | dynamic | static target.execution-os-type = auto | none | halted | live On some systems, the binaries that are created are the same wether you use them to debug a kernel, or a user space program. Many times inspecting an object file can reveal what an executable should be. For these cases we can now be a little more complete by specifying wether to detect all of these things automatically (inspect the main executable file and select a plug-in accordingly), or manually to force the selection of certain plug-ins. To do this we now allow the specficifation of wether one is debugging a user space program (target.execution-level = user) or a kernel program (target.execution-level = kernel). We can also specify if we want to debug a program where shared libraries are dynamically loaded using a DynamicLoader plug-in (target.execution-mode = dynamic), or wether we will treat all symbol files as already linked at the correct address (target.execution-mode = static). We can also specify if the inferior we are debugging is being debugged on a bare board (target.execution-os-type = none), or debugging an OS where we have a JTAG or other direct connection to the inferior stops the entire OS (target.execution-os-type = halted), or if we are debugging a program on something that has live debug services (target.execution-os-type = live). For the "target.execution-os-type = halted" mode, we will need to create ProcessHelper plug-ins that allow us to extract the process/thread and other OS information by reading/writing memory. This should allow LLDB to be used for a wide variety of debugging tasks and handle them all correctly. llvm-svn: 125815
2011-02-18 09:44:25 +08:00
UserSettingsController::UpdateBooleanVariable (VarSetOperationType op,
Did some work on the "register read" command to only show the first register set by default when dumping registers. If you want to see all of the register sets you can use the "--all" option: (lldb) register read --all If you want to just see some register sets, you can currently specify them by index: (lldb) register read --set 0 --set 2 We need to get shorter register set names soon so we can specify the register sets by name without having to type too much. I will make this change soon. You can also have any integer encoded registers resolve the address values back to any code or data from the object files using the "--lookup" option. Below is sample output when stopped in the libc function "puts" with some const strings in registers: Process 8973 stopped * thread #1: tid = 0x2c03, 0x00007fff828fa30f libSystem.B.dylib`puts + 1, stop reason = instruction step into frame #0: 0x00007fff828fa30f libSystem.B.dylib`puts + 1 (lldb) register read --lookup General Purpose Registers: rax = 0x0000000100000e98 "----------------------------------------------------------------------" rbx = 0x0000000000000000 rcx = 0x0000000000000001 rdx = 0x0000000000000000 rdi = 0x0000000100000e98 "----------------------------------------------------------------------" rsi = 0x0000000100800000 rbp = 0x00007fff5fbff710 rsp = 0x00007fff5fbff280 r8 = 0x0000000000000040 r9 = 0x0000000000000000 r10 = 0x0000000000000000 r11 = 0x0000000000000246 r12 = 0x0000000000000000 r13 = 0x0000000000000000 r14 = 0x0000000000000000 r15 = 0x0000000000000000 rip = 0x00007fff828fa30f libSystem.B.dylib`puts + 1 rflags = 0x0000000000000246 cs = 0x0000000000000027 fs = 0x0000000000000000 gs = 0x0000000000000000 As we can see, we see two constant strings and the PC (register "rip") is showing the code it resolves to. I fixed the register "--format" option to work as expected. Added a setting to disable skipping the function prologue when setting breakpoints as a target settings variable: (lldb) settings set target.skip-prologue false Updated the user settings controller boolean value handler funciton to be able to take the default value so it can correctly respond to the eVarSetOperationClear operation. Did some usability work on the OptionValue classes. Fixed the "image lookup" command to correctly respond to the "--verbose" option and display the detailed symbol context information when looking up line table entries and functions by name. This previously was only working for address lookups. llvm-svn: 129977
2011-04-22 11:55:06 +08:00
bool &bool_value,
const char *value_cstr,
bool clear_value,
Error &err)
{
Did some work on the "register read" command to only show the first register set by default when dumping registers. If you want to see all of the register sets you can use the "--all" option: (lldb) register read --all If you want to just see some register sets, you can currently specify them by index: (lldb) register read --set 0 --set 2 We need to get shorter register set names soon so we can specify the register sets by name without having to type too much. I will make this change soon. You can also have any integer encoded registers resolve the address values back to any code or data from the object files using the "--lookup" option. Below is sample output when stopped in the libc function "puts" with some const strings in registers: Process 8973 stopped * thread #1: tid = 0x2c03, 0x00007fff828fa30f libSystem.B.dylib`puts + 1, stop reason = instruction step into frame #0: 0x00007fff828fa30f libSystem.B.dylib`puts + 1 (lldb) register read --lookup General Purpose Registers: rax = 0x0000000100000e98 "----------------------------------------------------------------------" rbx = 0x0000000000000000 rcx = 0x0000000000000001 rdx = 0x0000000000000000 rdi = 0x0000000100000e98 "----------------------------------------------------------------------" rsi = 0x0000000100800000 rbp = 0x00007fff5fbff710 rsp = 0x00007fff5fbff280 r8 = 0x0000000000000040 r9 = 0x0000000000000000 r10 = 0x0000000000000000 r11 = 0x0000000000000246 r12 = 0x0000000000000000 r13 = 0x0000000000000000 r14 = 0x0000000000000000 r15 = 0x0000000000000000 rip = 0x00007fff828fa30f libSystem.B.dylib`puts + 1 rflags = 0x0000000000000246 cs = 0x0000000000000027 fs = 0x0000000000000000 gs = 0x0000000000000000 As we can see, we see two constant strings and the PC (register "rip") is showing the code it resolves to. I fixed the register "--format" option to work as expected. Added a setting to disable skipping the function prologue when setting breakpoints as a target settings variable: (lldb) settings set target.skip-prologue false Updated the user settings controller boolean value handler funciton to be able to take the default value so it can correctly respond to the eVarSetOperationClear operation. Did some usability work on the OptionValue classes. Fixed the "image lookup" command to correctly respond to the "--verbose" option and display the detailed symbol context information when looking up line table entries and functions by name. This previously was only working for address lookups. llvm-svn: 129977
2011-04-22 11:55:06 +08:00
switch (op)
{
case eVarSetOperationReplace:
case eVarSetOperationInsertBefore:
case eVarSetOperationInsertAfter:
case eVarSetOperationRemove:
case eVarSetOperationAppend:
case eVarSetOperationInvalid:
default:
err.SetErrorString ("Invalid operation for Boolean variable. Cannot update value.\n");
Did some work on the "register read" command to only show the first register set by default when dumping registers. If you want to see all of the register sets you can use the "--all" option: (lldb) register read --all If you want to just see some register sets, you can currently specify them by index: (lldb) register read --set 0 --set 2 We need to get shorter register set names soon so we can specify the register sets by name without having to type too much. I will make this change soon. You can also have any integer encoded registers resolve the address values back to any code or data from the object files using the "--lookup" option. Below is sample output when stopped in the libc function "puts" with some const strings in registers: Process 8973 stopped * thread #1: tid = 0x2c03, 0x00007fff828fa30f libSystem.B.dylib`puts + 1, stop reason = instruction step into frame #0: 0x00007fff828fa30f libSystem.B.dylib`puts + 1 (lldb) register read --lookup General Purpose Registers: rax = 0x0000000100000e98 "----------------------------------------------------------------------" rbx = 0x0000000000000000 rcx = 0x0000000000000001 rdx = 0x0000000000000000 rdi = 0x0000000100000e98 "----------------------------------------------------------------------" rsi = 0x0000000100800000 rbp = 0x00007fff5fbff710 rsp = 0x00007fff5fbff280 r8 = 0x0000000000000040 r9 = 0x0000000000000000 r10 = 0x0000000000000000 r11 = 0x0000000000000246 r12 = 0x0000000000000000 r13 = 0x0000000000000000 r14 = 0x0000000000000000 r15 = 0x0000000000000000 rip = 0x00007fff828fa30f libSystem.B.dylib`puts + 1 rflags = 0x0000000000000246 cs = 0x0000000000000027 fs = 0x0000000000000000 gs = 0x0000000000000000 As we can see, we see two constant strings and the PC (register "rip") is showing the code it resolves to. I fixed the register "--format" option to work as expected. Added a setting to disable skipping the function prologue when setting breakpoints as a target settings variable: (lldb) settings set target.skip-prologue false Updated the user settings controller boolean value handler funciton to be able to take the default value so it can correctly respond to the eVarSetOperationClear operation. Did some usability work on the OptionValue classes. Fixed the "image lookup" command to correctly respond to the "--verbose" option and display the detailed symbol context information when looking up line table entries and functions by name. This previously was only working for address lookups. llvm-svn: 129977
2011-04-22 11:55:06 +08:00
break;
Did some work on the "register read" command to only show the first register set by default when dumping registers. If you want to see all of the register sets you can use the "--all" option: (lldb) register read --all If you want to just see some register sets, you can currently specify them by index: (lldb) register read --set 0 --set 2 We need to get shorter register set names soon so we can specify the register sets by name without having to type too much. I will make this change soon. You can also have any integer encoded registers resolve the address values back to any code or data from the object files using the "--lookup" option. Below is sample output when stopped in the libc function "puts" with some const strings in registers: Process 8973 stopped * thread #1: tid = 0x2c03, 0x00007fff828fa30f libSystem.B.dylib`puts + 1, stop reason = instruction step into frame #0: 0x00007fff828fa30f libSystem.B.dylib`puts + 1 (lldb) register read --lookup General Purpose Registers: rax = 0x0000000100000e98 "----------------------------------------------------------------------" rbx = 0x0000000000000000 rcx = 0x0000000000000001 rdx = 0x0000000000000000 rdi = 0x0000000100000e98 "----------------------------------------------------------------------" rsi = 0x0000000100800000 rbp = 0x00007fff5fbff710 rsp = 0x00007fff5fbff280 r8 = 0x0000000000000040 r9 = 0x0000000000000000 r10 = 0x0000000000000000 r11 = 0x0000000000000246 r12 = 0x0000000000000000 r13 = 0x0000000000000000 r14 = 0x0000000000000000 r15 = 0x0000000000000000 rip = 0x00007fff828fa30f libSystem.B.dylib`puts + 1 rflags = 0x0000000000000246 cs = 0x0000000000000027 fs = 0x0000000000000000 gs = 0x0000000000000000 As we can see, we see two constant strings and the PC (register "rip") is showing the code it resolves to. I fixed the register "--format" option to work as expected. Added a setting to disable skipping the function prologue when setting breakpoints as a target settings variable: (lldb) settings set target.skip-prologue false Updated the user settings controller boolean value handler funciton to be able to take the default value so it can correctly respond to the eVarSetOperationClear operation. Did some usability work on the OptionValue classes. Fixed the "image lookup" command to correctly respond to the "--verbose" option and display the detailed symbol context information when looking up line table entries and functions by name. This previously was only working for address lookups. llvm-svn: 129977
2011-04-22 11:55:06 +08:00
case eVarSetOperationClear:
err.Clear();
bool_value = clear_value;
break;
case eVarSetOperationAssign:
{
bool success = false;
if (value_cstr == NULL)
err.SetErrorStringWithFormat ("invalid boolean string value (NULL)\n");
Did some work on the "register read" command to only show the first register set by default when dumping registers. If you want to see all of the register sets you can use the "--all" option: (lldb) register read --all If you want to just see some register sets, you can currently specify them by index: (lldb) register read --set 0 --set 2 We need to get shorter register set names soon so we can specify the register sets by name without having to type too much. I will make this change soon. You can also have any integer encoded registers resolve the address values back to any code or data from the object files using the "--lookup" option. Below is sample output when stopped in the libc function "puts" with some const strings in registers: Process 8973 stopped * thread #1: tid = 0x2c03, 0x00007fff828fa30f libSystem.B.dylib`puts + 1, stop reason = instruction step into frame #0: 0x00007fff828fa30f libSystem.B.dylib`puts + 1 (lldb) register read --lookup General Purpose Registers: rax = 0x0000000100000e98 "----------------------------------------------------------------------" rbx = 0x0000000000000000 rcx = 0x0000000000000001 rdx = 0x0000000000000000 rdi = 0x0000000100000e98 "----------------------------------------------------------------------" rsi = 0x0000000100800000 rbp = 0x00007fff5fbff710 rsp = 0x00007fff5fbff280 r8 = 0x0000000000000040 r9 = 0x0000000000000000 r10 = 0x0000000000000000 r11 = 0x0000000000000246 r12 = 0x0000000000000000 r13 = 0x0000000000000000 r14 = 0x0000000000000000 r15 = 0x0000000000000000 rip = 0x00007fff828fa30f libSystem.B.dylib`puts + 1 rflags = 0x0000000000000246 cs = 0x0000000000000027 fs = 0x0000000000000000 gs = 0x0000000000000000 As we can see, we see two constant strings and the PC (register "rip") is showing the code it resolves to. I fixed the register "--format" option to work as expected. Added a setting to disable skipping the function prologue when setting breakpoints as a target settings variable: (lldb) settings set target.skip-prologue false Updated the user settings controller boolean value handler funciton to be able to take the default value so it can correctly respond to the eVarSetOperationClear operation. Did some usability work on the OptionValue classes. Fixed the "image lookup" command to correctly respond to the "--verbose" option and display the detailed symbol context information when looking up line table entries and functions by name. This previously was only working for address lookups. llvm-svn: 129977
2011-04-22 11:55:06 +08:00
else if (value_cstr[0] == '\0')
err.SetErrorStringWithFormat ("invalid boolean string value (empty)\n");
Did some work on the "register read" command to only show the first register set by default when dumping registers. If you want to see all of the register sets you can use the "--all" option: (lldb) register read --all If you want to just see some register sets, you can currently specify them by index: (lldb) register read --set 0 --set 2 We need to get shorter register set names soon so we can specify the register sets by name without having to type too much. I will make this change soon. You can also have any integer encoded registers resolve the address values back to any code or data from the object files using the "--lookup" option. Below is sample output when stopped in the libc function "puts" with some const strings in registers: Process 8973 stopped * thread #1: tid = 0x2c03, 0x00007fff828fa30f libSystem.B.dylib`puts + 1, stop reason = instruction step into frame #0: 0x00007fff828fa30f libSystem.B.dylib`puts + 1 (lldb) register read --lookup General Purpose Registers: rax = 0x0000000100000e98 "----------------------------------------------------------------------" rbx = 0x0000000000000000 rcx = 0x0000000000000001 rdx = 0x0000000000000000 rdi = 0x0000000100000e98 "----------------------------------------------------------------------" rsi = 0x0000000100800000 rbp = 0x00007fff5fbff710 rsp = 0x00007fff5fbff280 r8 = 0x0000000000000040 r9 = 0x0000000000000000 r10 = 0x0000000000000000 r11 = 0x0000000000000246 r12 = 0x0000000000000000 r13 = 0x0000000000000000 r14 = 0x0000000000000000 r15 = 0x0000000000000000 rip = 0x00007fff828fa30f libSystem.B.dylib`puts + 1 rflags = 0x0000000000000246 cs = 0x0000000000000027 fs = 0x0000000000000000 gs = 0x0000000000000000 As we can see, we see two constant strings and the PC (register "rip") is showing the code it resolves to. I fixed the register "--format" option to work as expected. Added a setting to disable skipping the function prologue when setting breakpoints as a target settings variable: (lldb) settings set target.skip-prologue false Updated the user settings controller boolean value handler funciton to be able to take the default value so it can correctly respond to the eVarSetOperationClear operation. Did some usability work on the OptionValue classes. Fixed the "image lookup" command to correctly respond to the "--verbose" option and display the detailed symbol context information when looking up line table entries and functions by name. This previously was only working for address lookups. llvm-svn: 129977
2011-04-22 11:55:06 +08:00
else
{
bool new_value = Args::StringToBoolean (value_cstr, false, &success);
if (success)
{
err.Clear();
bool_value = new_value;
}
else
err.SetErrorStringWithFormat ("invalid boolean string value: '%s'\n", value_cstr);
}
}
break;
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
}
}
Error
UserSettingsController::UpdateBooleanOptionValue (const char *value,
VarSetOperationType op,
OptionValueBoolean &option_value)
{
Error error;
switch (op)
{
case eVarSetOperationReplace:
case eVarSetOperationInsertBefore:
case eVarSetOperationInsertAfter:
case eVarSetOperationRemove:
case eVarSetOperationAppend:
case eVarSetOperationInvalid:
default:
error.SetErrorString ("Invalid operation for Boolean variable. Cannot update value.\n");
break;
case eVarSetOperationClear:
option_value.Clear();
break;
case eVarSetOperationAssign:
{
bool success = false;
error = option_value.SetValueFromCString(value);
if (value == NULL)
error.SetErrorStringWithFormat ("invalid boolean string value (NULL)\n");
else if (value[0] == '\0')
error.SetErrorStringWithFormat ("invalid boolean string value (empty)\n");
else
{
bool new_value = Args::StringToBoolean (value, false, &success);
if (success)
{
error.Clear();
option_value = new_value;
}
else
error.SetErrorStringWithFormat ("invalid boolean string value: '%s'\n", value);
}
}
break;
}
return error;
}
void
Added new target instance settings for execution settings: Targets can now specify some additional parameters for when we debug executables that can help with plug-in selection: target.execution-level = auto | user | kernel target.execution-mode = auto | dynamic | static target.execution-os-type = auto | none | halted | live On some systems, the binaries that are created are the same wether you use them to debug a kernel, or a user space program. Many times inspecting an object file can reveal what an executable should be. For these cases we can now be a little more complete by specifying wether to detect all of these things automatically (inspect the main executable file and select a plug-in accordingly), or manually to force the selection of certain plug-ins. To do this we now allow the specficifation of wether one is debugging a user space program (target.execution-level = user) or a kernel program (target.execution-level = kernel). We can also specify if we want to debug a program where shared libraries are dynamically loaded using a DynamicLoader plug-in (target.execution-mode = dynamic), or wether we will treat all symbol files as already linked at the correct address (target.execution-mode = static). We can also specify if the inferior we are debugging is being debugged on a bare board (target.execution-os-type = none), or debugging an OS where we have a JTAG or other direct connection to the inferior stops the entire OS (target.execution-os-type = halted), or if we are debugging a program on something that has live debug services (target.execution-os-type = live). For the "target.execution-os-type = halted" mode, we will need to create ProcessHelper plug-ins that allow us to extract the process/thread and other OS information by reading/writing memory. This should allow LLDB to be used for a wide variety of debugging tasks and handle them all correctly. llvm-svn: 125815
2011-02-18 09:44:25 +08:00
UserSettingsController::UpdateStringArrayVariable (VarSetOperationType op,
const char *index_value,
Args &array_var,
const char *new_value,
Error &err)
{
int index = -1;
bool valid_index = true;
if (index_value != NULL)
{
for (int i = 0; i < strlen(index_value); ++i)
if (!isdigit (index_value[i]))
{
valid_index = false;
err.SetErrorStringWithFormat ("'%s' is not a valid integer index. Cannot update array value.\n",
index_value);
}
if (valid_index)
index = atoi (index_value);
if (index < 0
|| index >= array_var.GetArgumentCount())
{
valid_index = false;
err.SetErrorStringWithFormat ("%d is outside the bounds of the specified array variable. "
"Cannot update array value.\n", index);
}
}
switch (op)
{
Added new target instance settings for execution settings: Targets can now specify some additional parameters for when we debug executables that can help with plug-in selection: target.execution-level = auto | user | kernel target.execution-mode = auto | dynamic | static target.execution-os-type = auto | none | halted | live On some systems, the binaries that are created are the same wether you use them to debug a kernel, or a user space program. Many times inspecting an object file can reveal what an executable should be. For these cases we can now be a little more complete by specifying wether to detect all of these things automatically (inspect the main executable file and select a plug-in accordingly), or manually to force the selection of certain plug-ins. To do this we now allow the specficifation of wether one is debugging a user space program (target.execution-level = user) or a kernel program (target.execution-level = kernel). We can also specify if we want to debug a program where shared libraries are dynamically loaded using a DynamicLoader plug-in (target.execution-mode = dynamic), or wether we will treat all symbol files as already linked at the correct address (target.execution-mode = static). We can also specify if the inferior we are debugging is being debugged on a bare board (target.execution-os-type = none), or debugging an OS where we have a JTAG or other direct connection to the inferior stops the entire OS (target.execution-os-type = halted), or if we are debugging a program on something that has live debug services (target.execution-os-type = live). For the "target.execution-os-type = halted" mode, we will need to create ProcessHelper plug-ins that allow us to extract the process/thread and other OS information by reading/writing memory. This should allow LLDB to be used for a wide variety of debugging tasks and handle them all correctly. llvm-svn: 125815
2011-02-18 09:44:25 +08:00
case eVarSetOperationAssign:
array_var.SetCommandString (new_value);
break;
Added new target instance settings for execution settings: Targets can now specify some additional parameters for when we debug executables that can help with plug-in selection: target.execution-level = auto | user | kernel target.execution-mode = auto | dynamic | static target.execution-os-type = auto | none | halted | live On some systems, the binaries that are created are the same wether you use them to debug a kernel, or a user space program. Many times inspecting an object file can reveal what an executable should be. For these cases we can now be a little more complete by specifying wether to detect all of these things automatically (inspect the main executable file and select a plug-in accordingly), or manually to force the selection of certain plug-ins. To do this we now allow the specficifation of wether one is debugging a user space program (target.execution-level = user) or a kernel program (target.execution-level = kernel). We can also specify if we want to debug a program where shared libraries are dynamically loaded using a DynamicLoader plug-in (target.execution-mode = dynamic), or wether we will treat all symbol files as already linked at the correct address (target.execution-mode = static). We can also specify if the inferior we are debugging is being debugged on a bare board (target.execution-os-type = none), or debugging an OS where we have a JTAG or other direct connection to the inferior stops the entire OS (target.execution-os-type = halted), or if we are debugging a program on something that has live debug services (target.execution-os-type = live). For the "target.execution-os-type = halted" mode, we will need to create ProcessHelper plug-ins that allow us to extract the process/thread and other OS information by reading/writing memory. This should allow LLDB to be used for a wide variety of debugging tasks and handle them all correctly. llvm-svn: 125815
2011-02-18 09:44:25 +08:00
case eVarSetOperationReplace:
{
if (valid_index)
array_var.ReplaceArgumentAtIndex (index, new_value);
break;
}
Added new target instance settings for execution settings: Targets can now specify some additional parameters for when we debug executables that can help with plug-in selection: target.execution-level = auto | user | kernel target.execution-mode = auto | dynamic | static target.execution-os-type = auto | none | halted | live On some systems, the binaries that are created are the same wether you use them to debug a kernel, or a user space program. Many times inspecting an object file can reveal what an executable should be. For these cases we can now be a little more complete by specifying wether to detect all of these things automatically (inspect the main executable file and select a plug-in accordingly), or manually to force the selection of certain plug-ins. To do this we now allow the specficifation of wether one is debugging a user space program (target.execution-level = user) or a kernel program (target.execution-level = kernel). We can also specify if we want to debug a program where shared libraries are dynamically loaded using a DynamicLoader plug-in (target.execution-mode = dynamic), or wether we will treat all symbol files as already linked at the correct address (target.execution-mode = static). We can also specify if the inferior we are debugging is being debugged on a bare board (target.execution-os-type = none), or debugging an OS where we have a JTAG or other direct connection to the inferior stops the entire OS (target.execution-os-type = halted), or if we are debugging a program on something that has live debug services (target.execution-os-type = live). For the "target.execution-os-type = halted" mode, we will need to create ProcessHelper plug-ins that allow us to extract the process/thread and other OS information by reading/writing memory. This should allow LLDB to be used for a wide variety of debugging tasks and handle them all correctly. llvm-svn: 125815
2011-02-18 09:44:25 +08:00
case eVarSetOperationInsertBefore:
case eVarSetOperationInsertAfter:
{
if (valid_index)
{
Args new_array (new_value);
Added new target instance settings for execution settings: Targets can now specify some additional parameters for when we debug executables that can help with plug-in selection: target.execution-level = auto | user | kernel target.execution-mode = auto | dynamic | static target.execution-os-type = auto | none | halted | live On some systems, the binaries that are created are the same wether you use them to debug a kernel, or a user space program. Many times inspecting an object file can reveal what an executable should be. For these cases we can now be a little more complete by specifying wether to detect all of these things automatically (inspect the main executable file and select a plug-in accordingly), or manually to force the selection of certain plug-ins. To do this we now allow the specficifation of wether one is debugging a user space program (target.execution-level = user) or a kernel program (target.execution-level = kernel). We can also specify if we want to debug a program where shared libraries are dynamically loaded using a DynamicLoader plug-in (target.execution-mode = dynamic), or wether we will treat all symbol files as already linked at the correct address (target.execution-mode = static). We can also specify if the inferior we are debugging is being debugged on a bare board (target.execution-os-type = none), or debugging an OS where we have a JTAG or other direct connection to the inferior stops the entire OS (target.execution-os-type = halted), or if we are debugging a program on something that has live debug services (target.execution-os-type = live). For the "target.execution-os-type = halted" mode, we will need to create ProcessHelper plug-ins that allow us to extract the process/thread and other OS information by reading/writing memory. This should allow LLDB to be used for a wide variety of debugging tasks and handle them all correctly. llvm-svn: 125815
2011-02-18 09:44:25 +08:00
if (op == eVarSetOperationInsertAfter)
++index;
for (int i = 0; i < new_array.GetArgumentCount(); ++i)
array_var.InsertArgumentAtIndex (index, new_array.GetArgumentAtIndex (i));
}
break;
}
Added new target instance settings for execution settings: Targets can now specify some additional parameters for when we debug executables that can help with plug-in selection: target.execution-level = auto | user | kernel target.execution-mode = auto | dynamic | static target.execution-os-type = auto | none | halted | live On some systems, the binaries that are created are the same wether you use them to debug a kernel, or a user space program. Many times inspecting an object file can reveal what an executable should be. For these cases we can now be a little more complete by specifying wether to detect all of these things automatically (inspect the main executable file and select a plug-in accordingly), or manually to force the selection of certain plug-ins. To do this we now allow the specficifation of wether one is debugging a user space program (target.execution-level = user) or a kernel program (target.execution-level = kernel). We can also specify if we want to debug a program where shared libraries are dynamically loaded using a DynamicLoader plug-in (target.execution-mode = dynamic), or wether we will treat all symbol files as already linked at the correct address (target.execution-mode = static). We can also specify if the inferior we are debugging is being debugged on a bare board (target.execution-os-type = none), or debugging an OS where we have a JTAG or other direct connection to the inferior stops the entire OS (target.execution-os-type = halted), or if we are debugging a program on something that has live debug services (target.execution-os-type = live). For the "target.execution-os-type = halted" mode, we will need to create ProcessHelper plug-ins that allow us to extract the process/thread and other OS information by reading/writing memory. This should allow LLDB to be used for a wide variety of debugging tasks and handle them all correctly. llvm-svn: 125815
2011-02-18 09:44:25 +08:00
case eVarSetOperationRemove:
{
if (valid_index)
array_var.DeleteArgumentAtIndex (index);
break;
}
Added new target instance settings for execution settings: Targets can now specify some additional parameters for when we debug executables that can help with plug-in selection: target.execution-level = auto | user | kernel target.execution-mode = auto | dynamic | static target.execution-os-type = auto | none | halted | live On some systems, the binaries that are created are the same wether you use them to debug a kernel, or a user space program. Many times inspecting an object file can reveal what an executable should be. For these cases we can now be a little more complete by specifying wether to detect all of these things automatically (inspect the main executable file and select a plug-in accordingly), or manually to force the selection of certain plug-ins. To do this we now allow the specficifation of wether one is debugging a user space program (target.execution-level = user) or a kernel program (target.execution-level = kernel). We can also specify if we want to debug a program where shared libraries are dynamically loaded using a DynamicLoader plug-in (target.execution-mode = dynamic), or wether we will treat all symbol files as already linked at the correct address (target.execution-mode = static). We can also specify if the inferior we are debugging is being debugged on a bare board (target.execution-os-type = none), or debugging an OS where we have a JTAG or other direct connection to the inferior stops the entire OS (target.execution-os-type = halted), or if we are debugging a program on something that has live debug services (target.execution-os-type = live). For the "target.execution-os-type = halted" mode, we will need to create ProcessHelper plug-ins that allow us to extract the process/thread and other OS information by reading/writing memory. This should allow LLDB to be used for a wide variety of debugging tasks and handle them all correctly. llvm-svn: 125815
2011-02-18 09:44:25 +08:00
case eVarSetOperationAppend:
{
Args new_array (new_value);
array_var.AppendArguments (new_array);
break;
}
Added new target instance settings for execution settings: Targets can now specify some additional parameters for when we debug executables that can help with plug-in selection: target.execution-level = auto | user | kernel target.execution-mode = auto | dynamic | static target.execution-os-type = auto | none | halted | live On some systems, the binaries that are created are the same wether you use them to debug a kernel, or a user space program. Many times inspecting an object file can reveal what an executable should be. For these cases we can now be a little more complete by specifying wether to detect all of these things automatically (inspect the main executable file and select a plug-in accordingly), or manually to force the selection of certain plug-ins. To do this we now allow the specficifation of wether one is debugging a user space program (target.execution-level = user) or a kernel program (target.execution-level = kernel). We can also specify if we want to debug a program where shared libraries are dynamically loaded using a DynamicLoader plug-in (target.execution-mode = dynamic), or wether we will treat all symbol files as already linked at the correct address (target.execution-mode = static). We can also specify if the inferior we are debugging is being debugged on a bare board (target.execution-os-type = none), or debugging an OS where we have a JTAG or other direct connection to the inferior stops the entire OS (target.execution-os-type = halted), or if we are debugging a program on something that has live debug services (target.execution-os-type = live). For the "target.execution-os-type = halted" mode, we will need to create ProcessHelper plug-ins that allow us to extract the process/thread and other OS information by reading/writing memory. This should allow LLDB to be used for a wide variety of debugging tasks and handle them all correctly. llvm-svn: 125815
2011-02-18 09:44:25 +08:00
case eVarSetOperationClear:
array_var.Clear();
break;
default:
err.SetErrorString ("Unrecognized operation. Cannot update value.\n");
break;
}
}
void
Added new target instance settings for execution settings: Targets can now specify some additional parameters for when we debug executables that can help with plug-in selection: target.execution-level = auto | user | kernel target.execution-mode = auto | dynamic | static target.execution-os-type = auto | none | halted | live On some systems, the binaries that are created are the same wether you use them to debug a kernel, or a user space program. Many times inspecting an object file can reveal what an executable should be. For these cases we can now be a little more complete by specifying wether to detect all of these things automatically (inspect the main executable file and select a plug-in accordingly), or manually to force the selection of certain plug-ins. To do this we now allow the specficifation of wether one is debugging a user space program (target.execution-level = user) or a kernel program (target.execution-level = kernel). We can also specify if we want to debug a program where shared libraries are dynamically loaded using a DynamicLoader plug-in (target.execution-mode = dynamic), or wether we will treat all symbol files as already linked at the correct address (target.execution-mode = static). We can also specify if the inferior we are debugging is being debugged on a bare board (target.execution-os-type = none), or debugging an OS where we have a JTAG or other direct connection to the inferior stops the entire OS (target.execution-os-type = halted), or if we are debugging a program on something that has live debug services (target.execution-os-type = live). For the "target.execution-os-type = halted" mode, we will need to create ProcessHelper plug-ins that allow us to extract the process/thread and other OS information by reading/writing memory. This should allow LLDB to be used for a wide variety of debugging tasks and handle them all correctly. llvm-svn: 125815
2011-02-18 09:44:25 +08:00
UserSettingsController::UpdateDictionaryVariable (VarSetOperationType op,
const char *index_value,
std::map<std::string, std::string> &dictionary,
const char *new_value,
Error &err)
{
switch (op)
{
Added new target instance settings for execution settings: Targets can now specify some additional parameters for when we debug executables that can help with plug-in selection: target.execution-level = auto | user | kernel target.execution-mode = auto | dynamic | static target.execution-os-type = auto | none | halted | live On some systems, the binaries that are created are the same wether you use them to debug a kernel, or a user space program. Many times inspecting an object file can reveal what an executable should be. For these cases we can now be a little more complete by specifying wether to detect all of these things automatically (inspect the main executable file and select a plug-in accordingly), or manually to force the selection of certain plug-ins. To do this we now allow the specficifation of wether one is debugging a user space program (target.execution-level = user) or a kernel program (target.execution-level = kernel). We can also specify if we want to debug a program where shared libraries are dynamically loaded using a DynamicLoader plug-in (target.execution-mode = dynamic), or wether we will treat all symbol files as already linked at the correct address (target.execution-mode = static). We can also specify if the inferior we are debugging is being debugged on a bare board (target.execution-os-type = none), or debugging an OS where we have a JTAG or other direct connection to the inferior stops the entire OS (target.execution-os-type = halted), or if we are debugging a program on something that has live debug services (target.execution-os-type = live). For the "target.execution-os-type = halted" mode, we will need to create ProcessHelper plug-ins that allow us to extract the process/thread and other OS information by reading/writing memory. This should allow LLDB to be used for a wide variety of debugging tasks and handle them all correctly. llvm-svn: 125815
2011-02-18 09:44:25 +08:00
case eVarSetOperationReplace:
if (index_value != NULL)
{
std::string key (index_value);
std::map<std::string, std::string>::iterator pos;
pos = dictionary.find (key);
if (pos != dictionary.end())
dictionary[key] = new_value;
else
err.SetErrorStringWithFormat ("'%s' is not an existing key; cannot replace value.\n", index_value);
}
else
err.SetErrorString ("'settings replace' requires a key for dictionary variables. No key supplied.\n");
break;
Added new target instance settings for execution settings: Targets can now specify some additional parameters for when we debug executables that can help with plug-in selection: target.execution-level = auto | user | kernel target.execution-mode = auto | dynamic | static target.execution-os-type = auto | none | halted | live On some systems, the binaries that are created are the same wether you use them to debug a kernel, or a user space program. Many times inspecting an object file can reveal what an executable should be. For these cases we can now be a little more complete by specifying wether to detect all of these things automatically (inspect the main executable file and select a plug-in accordingly), or manually to force the selection of certain plug-ins. To do this we now allow the specficifation of wether one is debugging a user space program (target.execution-level = user) or a kernel program (target.execution-level = kernel). We can also specify if we want to debug a program where shared libraries are dynamically loaded using a DynamicLoader plug-in (target.execution-mode = dynamic), or wether we will treat all symbol files as already linked at the correct address (target.execution-mode = static). We can also specify if the inferior we are debugging is being debugged on a bare board (target.execution-os-type = none), or debugging an OS where we have a JTAG or other direct connection to the inferior stops the entire OS (target.execution-os-type = halted), or if we are debugging a program on something that has live debug services (target.execution-os-type = live). For the "target.execution-os-type = halted" mode, we will need to create ProcessHelper plug-ins that allow us to extract the process/thread and other OS information by reading/writing memory. This should allow LLDB to be used for a wide variety of debugging tasks and handle them all correctly. llvm-svn: 125815
2011-02-18 09:44:25 +08:00
case eVarSetOperationRemove:
if (index_value != NULL)
{
std::string key (index_value);
dictionary.erase (key);
}
else
err.SetErrorString ("'settings remove' requires a key for dictionary variables. No key supplied.\n");
break;
Added new target instance settings for execution settings: Targets can now specify some additional parameters for when we debug executables that can help with plug-in selection: target.execution-level = auto | user | kernel target.execution-mode = auto | dynamic | static target.execution-os-type = auto | none | halted | live On some systems, the binaries that are created are the same wether you use them to debug a kernel, or a user space program. Many times inspecting an object file can reveal what an executable should be. For these cases we can now be a little more complete by specifying wether to detect all of these things automatically (inspect the main executable file and select a plug-in accordingly), or manually to force the selection of certain plug-ins. To do this we now allow the specficifation of wether one is debugging a user space program (target.execution-level = user) or a kernel program (target.execution-level = kernel). We can also specify if we want to debug a program where shared libraries are dynamically loaded using a DynamicLoader plug-in (target.execution-mode = dynamic), or wether we will treat all symbol files as already linked at the correct address (target.execution-mode = static). We can also specify if the inferior we are debugging is being debugged on a bare board (target.execution-os-type = none), or debugging an OS where we have a JTAG or other direct connection to the inferior stops the entire OS (target.execution-os-type = halted), or if we are debugging a program on something that has live debug services (target.execution-os-type = live). For the "target.execution-os-type = halted" mode, we will need to create ProcessHelper plug-ins that allow us to extract the process/thread and other OS information by reading/writing memory. This should allow LLDB to be used for a wide variety of debugging tasks and handle them all correctly. llvm-svn: 125815
2011-02-18 09:44:25 +08:00
case eVarSetOperationClear:
dictionary.clear ();
break;
Added new target instance settings for execution settings: Targets can now specify some additional parameters for when we debug executables that can help with plug-in selection: target.execution-level = auto | user | kernel target.execution-mode = auto | dynamic | static target.execution-os-type = auto | none | halted | live On some systems, the binaries that are created are the same wether you use them to debug a kernel, or a user space program. Many times inspecting an object file can reveal what an executable should be. For these cases we can now be a little more complete by specifying wether to detect all of these things automatically (inspect the main executable file and select a plug-in accordingly), or manually to force the selection of certain plug-ins. To do this we now allow the specficifation of wether one is debugging a user space program (target.execution-level = user) or a kernel program (target.execution-level = kernel). We can also specify if we want to debug a program where shared libraries are dynamically loaded using a DynamicLoader plug-in (target.execution-mode = dynamic), or wether we will treat all symbol files as already linked at the correct address (target.execution-mode = static). We can also specify if the inferior we are debugging is being debugged on a bare board (target.execution-os-type = none), or debugging an OS where we have a JTAG or other direct connection to the inferior stops the entire OS (target.execution-os-type = halted), or if we are debugging a program on something that has live debug services (target.execution-os-type = live). For the "target.execution-os-type = halted" mode, we will need to create ProcessHelper plug-ins that allow us to extract the process/thread and other OS information by reading/writing memory. This should allow LLDB to be used for a wide variety of debugging tasks and handle them all correctly. llvm-svn: 125815
2011-02-18 09:44:25 +08:00
case eVarSetOperationAppend:
case eVarSetOperationAssign:
{
// Clear the dictionary if it's an assign with new_value as NULL.
Added new target instance settings for execution settings: Targets can now specify some additional parameters for when we debug executables that can help with plug-in selection: target.execution-level = auto | user | kernel target.execution-mode = auto | dynamic | static target.execution-os-type = auto | none | halted | live On some systems, the binaries that are created are the same wether you use them to debug a kernel, or a user space program. Many times inspecting an object file can reveal what an executable should be. For these cases we can now be a little more complete by specifying wether to detect all of these things automatically (inspect the main executable file and select a plug-in accordingly), or manually to force the selection of certain plug-ins. To do this we now allow the specficifation of wether one is debugging a user space program (target.execution-level = user) or a kernel program (target.execution-level = kernel). We can also specify if we want to debug a program where shared libraries are dynamically loaded using a DynamicLoader plug-in (target.execution-mode = dynamic), or wether we will treat all symbol files as already linked at the correct address (target.execution-mode = static). We can also specify if the inferior we are debugging is being debugged on a bare board (target.execution-os-type = none), or debugging an OS where we have a JTAG or other direct connection to the inferior stops the entire OS (target.execution-os-type = halted), or if we are debugging a program on something that has live debug services (target.execution-os-type = live). For the "target.execution-os-type = halted" mode, we will need to create ProcessHelper plug-ins that allow us to extract the process/thread and other OS information by reading/writing memory. This should allow LLDB to be used for a wide variety of debugging tasks and handle them all correctly. llvm-svn: 125815
2011-02-18 09:44:25 +08:00
if (new_value == NULL && op == eVarSetOperationAssign)
{
dictionary.clear ();
break;
}
Args args (new_value);
size_t num_args = args.GetArgumentCount();
RegularExpression regex("(\\[\"?)?" // Regex match 1 (optional key prefix of '["' pr '[')
"([A-Za-z_][A-Za-z_0-9]*)" // Regex match 2 (key string)
"(\"?\\])?" // Regex match 3 (optional key suffix of '"]' pr ']')
"=" // The equal sign that is required
"(.*)"); // Regex match 4 (value string)
std::string key, value;
for (size_t i = 0; i < num_args; ++i)
{
const char *key_equal_value_arg = args.GetArgumentAtIndex (i);
// Execute the regular expression on each arg.
if (regex.Execute(key_equal_value_arg, 5))
{
// The regular expression succeeded. The match at index
// zero will be the entire string that matched the entire
// regular expression. The match at index 1 - 4 will be
// as mentioned above by the creation of the regex pattern.
// Match index 2 is the key, match index 4 is the value.
regex.GetMatchAtIndex (key_equal_value_arg, 2, key);
regex.GetMatchAtIndex (key_equal_value_arg, 4, value);
dictionary[key] = value;
}
else
{
err.SetErrorString ("Invalid format for dictionary value. Expected one of '[\"<key>\"]=<value>', '[<key>]=<value>', or '<key>=<value>'\n");
}
}
}
break;
Added new target instance settings for execution settings: Targets can now specify some additional parameters for when we debug executables that can help with plug-in selection: target.execution-level = auto | user | kernel target.execution-mode = auto | dynamic | static target.execution-os-type = auto | none | halted | live On some systems, the binaries that are created are the same wether you use them to debug a kernel, or a user space program. Many times inspecting an object file can reveal what an executable should be. For these cases we can now be a little more complete by specifying wether to detect all of these things automatically (inspect the main executable file and select a plug-in accordingly), or manually to force the selection of certain plug-ins. To do this we now allow the specficifation of wether one is debugging a user space program (target.execution-level = user) or a kernel program (target.execution-level = kernel). We can also specify if we want to debug a program where shared libraries are dynamically loaded using a DynamicLoader plug-in (target.execution-mode = dynamic), or wether we will treat all symbol files as already linked at the correct address (target.execution-mode = static). We can also specify if the inferior we are debugging is being debugged on a bare board (target.execution-os-type = none), or debugging an OS where we have a JTAG or other direct connection to the inferior stops the entire OS (target.execution-os-type = halted), or if we are debugging a program on something that has live debug services (target.execution-os-type = live). For the "target.execution-os-type = halted" mode, we will need to create ProcessHelper plug-ins that allow us to extract the process/thread and other OS information by reading/writing memory. This should allow LLDB to be used for a wide variety of debugging tasks and handle them all correctly. llvm-svn: 125815
2011-02-18 09:44:25 +08:00
case eVarSetOperationInsertBefore:
case eVarSetOperationInsertAfter:
err.SetErrorString ("Specified operation cannot be performed on dictionary variables.\n");
break;
default:
err.SetErrorString ("Unrecognized operation.\n");
break;
}
}
const char *
Added new target instance settings for execution settings: Targets can now specify some additional parameters for when we debug executables that can help with plug-in selection: target.execution-level = auto | user | kernel target.execution-mode = auto | dynamic | static target.execution-os-type = auto | none | halted | live On some systems, the binaries that are created are the same wether you use them to debug a kernel, or a user space program. Many times inspecting an object file can reveal what an executable should be. For these cases we can now be a little more complete by specifying wether to detect all of these things automatically (inspect the main executable file and select a plug-in accordingly), or manually to force the selection of certain plug-ins. To do this we now allow the specficifation of wether one is debugging a user space program (target.execution-level = user) or a kernel program (target.execution-level = kernel). We can also specify if we want to debug a program where shared libraries are dynamically loaded using a DynamicLoader plug-in (target.execution-mode = dynamic), or wether we will treat all symbol files as already linked at the correct address (target.execution-mode = static). We can also specify if the inferior we are debugging is being debugged on a bare board (target.execution-os-type = none), or debugging an OS where we have a JTAG or other direct connection to the inferior stops the entire OS (target.execution-os-type = halted), or if we are debugging a program on something that has live debug services (target.execution-os-type = live). For the "target.execution-os-type = halted" mode, we will need to create ProcessHelper plug-ins that allow us to extract the process/thread and other OS information by reading/writing memory. This should allow LLDB to be used for a wide variety of debugging tasks and handle them all correctly. llvm-svn: 125815
2011-02-18 09:44:25 +08:00
UserSettingsController::EnumToString (const OptionEnumValueElement *enum_values,
int value)
{
int i = 0;
while (enum_values[i].string_value != NULL)
{
if (enum_values[i].value == value)
return enum_values[i].string_value;
++i;
}
Added new target instance settings for execution settings: Targets can now specify some additional parameters for when we debug executables that can help with plug-in selection: target.execution-level = auto | user | kernel target.execution-mode = auto | dynamic | static target.execution-os-type = auto | none | halted | live On some systems, the binaries that are created are the same wether you use them to debug a kernel, or a user space program. Many times inspecting an object file can reveal what an executable should be. For these cases we can now be a little more complete by specifying wether to detect all of these things automatically (inspect the main executable file and select a plug-in accordingly), or manually to force the selection of certain plug-ins. To do this we now allow the specficifation of wether one is debugging a user space program (target.execution-level = user) or a kernel program (target.execution-level = kernel). We can also specify if we want to debug a program where shared libraries are dynamically loaded using a DynamicLoader plug-in (target.execution-mode = dynamic), or wether we will treat all symbol files as already linked at the correct address (target.execution-mode = static). We can also specify if the inferior we are debugging is being debugged on a bare board (target.execution-os-type = none), or debugging an OS where we have a JTAG or other direct connection to the inferior stops the entire OS (target.execution-os-type = halted), or if we are debugging a program on something that has live debug services (target.execution-os-type = live). For the "target.execution-os-type = halted" mode, we will need to create ProcessHelper plug-ins that allow us to extract the process/thread and other OS information by reading/writing memory. This should allow LLDB to be used for a wide variety of debugging tasks and handle them all correctly. llvm-svn: 125815
2011-02-18 09:44:25 +08:00
return "";
}
void
Added new target instance settings for execution settings: Targets can now specify some additional parameters for when we debug executables that can help with plug-in selection: target.execution-level = auto | user | kernel target.execution-mode = auto | dynamic | static target.execution-os-type = auto | none | halted | live On some systems, the binaries that are created are the same wether you use them to debug a kernel, or a user space program. Many times inspecting an object file can reveal what an executable should be. For these cases we can now be a little more complete by specifying wether to detect all of these things automatically (inspect the main executable file and select a plug-in accordingly), or manually to force the selection of certain plug-ins. To do this we now allow the specficifation of wether one is debugging a user space program (target.execution-level = user) or a kernel program (target.execution-level = kernel). We can also specify if we want to debug a program where shared libraries are dynamically loaded using a DynamicLoader plug-in (target.execution-mode = dynamic), or wether we will treat all symbol files as already linked at the correct address (target.execution-mode = static). We can also specify if the inferior we are debugging is being debugged on a bare board (target.execution-os-type = none), or debugging an OS where we have a JTAG or other direct connection to the inferior stops the entire OS (target.execution-os-type = halted), or if we are debugging a program on something that has live debug services (target.execution-os-type = live). For the "target.execution-os-type = halted" mode, we will need to create ProcessHelper plug-ins that allow us to extract the process/thread and other OS information by reading/writing memory. This should allow LLDB to be used for a wide variety of debugging tasks and handle them all correctly. llvm-svn: 125815
2011-02-18 09:44:25 +08:00
UserSettingsController::UpdateEnumVariable (OptionEnumValueElement *enum_values,
int *enum_var,
const char *new_value,
Error &error)
{
*enum_var = Args::StringToOptionEnum (new_value, enum_values, enum_values[0].value, error);
}
void
UserSettingsController::RenameInstanceSettings (const char *old_name, const char *new_name)
{
Mutex::Locker live_mutex (m_live_settings_mutex);
Mutex::Locker pending_mutex (m_pending_settings_mutex);
std::string old_name_key (old_name);
std::string new_name_key (new_name);
// First, find the live instance settings for the old_name. If they don't exist in the live settings
// list, then this is not a setting that can be renamed.
if ((old_name_key[0] != '[') || (old_name_key[old_name_key.size() -1] != ']'))
{
StreamString tmp_str;
tmp_str.Printf ("[%s]", old_name);
old_name_key = tmp_str.GetData();
}
if ((new_name_key[0] != '[') || (new_name_key[new_name_key.size() -1] != ']'))
{
StreamString tmp_str;
tmp_str.Printf ("[%s]", new_name);
new_name_key = tmp_str.GetData();
}
if (old_name_key.compare (new_name_key) == 0)
return;
size_t len = new_name_key.length();
std::string stripped_new_name = new_name_key.substr (1, len-2); // new name without the '[ ]'
InstanceSettingsMap::iterator pos;
pos = m_live_settings.find (old_name_key);
if (pos != m_live_settings.end())
{
InstanceSettings *live_settings = pos->second;
// Rename the settings.
live_settings->ChangeInstanceName (stripped_new_name);
// Now see if there are any pending settings for the new name; if so, copy them into live_settings.
Added new target instance settings for execution settings: Targets can now specify some additional parameters for when we debug executables that can help with plug-in selection: target.execution-level = auto | user | kernel target.execution-mode = auto | dynamic | static target.execution-os-type = auto | none | halted | live On some systems, the binaries that are created are the same wether you use them to debug a kernel, or a user space program. Many times inspecting an object file can reveal what an executable should be. For these cases we can now be a little more complete by specifying wether to detect all of these things automatically (inspect the main executable file and select a plug-in accordingly), or manually to force the selection of certain plug-ins. To do this we now allow the specficifation of wether one is debugging a user space program (target.execution-level = user) or a kernel program (target.execution-level = kernel). We can also specify if we want to debug a program where shared libraries are dynamically loaded using a DynamicLoader plug-in (target.execution-mode = dynamic), or wether we will treat all symbol files as already linked at the correct address (target.execution-mode = static). We can also specify if the inferior we are debugging is being debugged on a bare board (target.execution-os-type = none), or debugging an OS where we have a JTAG or other direct connection to the inferior stops the entire OS (target.execution-os-type = halted), or if we are debugging a program on something that has live debug services (target.execution-os-type = live). For the "target.execution-os-type = halted" mode, we will need to create ProcessHelper plug-ins that allow us to extract the process/thread and other OS information by reading/writing memory. This should allow LLDB to be used for a wide variety of debugging tasks and handle them all correctly. llvm-svn: 125815
2011-02-18 09:44:25 +08:00
std::map<std::string, InstanceSettingsSP>::iterator pending_pos;
pending_pos = m_pending_settings.find (new_name_key);
if (pending_pos != m_pending_settings.end())
{
Added new target instance settings for execution settings: Targets can now specify some additional parameters for when we debug executables that can help with plug-in selection: target.execution-level = auto | user | kernel target.execution-mode = auto | dynamic | static target.execution-os-type = auto | none | halted | live On some systems, the binaries that are created are the same wether you use them to debug a kernel, or a user space program. Many times inspecting an object file can reveal what an executable should be. For these cases we can now be a little more complete by specifying wether to detect all of these things automatically (inspect the main executable file and select a plug-in accordingly), or manually to force the selection of certain plug-ins. To do this we now allow the specficifation of wether one is debugging a user space program (target.execution-level = user) or a kernel program (target.execution-level = kernel). We can also specify if we want to debug a program where shared libraries are dynamically loaded using a DynamicLoader plug-in (target.execution-mode = dynamic), or wether we will treat all symbol files as already linked at the correct address (target.execution-mode = static). We can also specify if the inferior we are debugging is being debugged on a bare board (target.execution-os-type = none), or debugging an OS where we have a JTAG or other direct connection to the inferior stops the entire OS (target.execution-os-type = halted), or if we are debugging a program on something that has live debug services (target.execution-os-type = live). For the "target.execution-os-type = halted" mode, we will need to create ProcessHelper plug-ins that allow us to extract the process/thread and other OS information by reading/writing memory. This should allow LLDB to be used for a wide variety of debugging tasks and handle them all correctly. llvm-svn: 125815
2011-02-18 09:44:25 +08:00
InstanceSettingsSP pending_settings_sp = pending_pos->second;
live_settings->CopyInstanceSettings (pending_settings_sp, false);
}
// Erase the old entry (under the old name) from live settings.
m_live_settings.erase (pos);
// Add the new entry, with the new name, into live settings.
m_live_settings[new_name_key] = live_settings;
}
}
//----------------------------------------------------------------------
// class InstanceSettings
//----------------------------------------------------------------------
InstanceSettings::InstanceSettings (UserSettingsController &owner, const char *instance_name, bool live_instance) :
m_owner (owner),
m_instance_name (instance_name)
{
if ((m_instance_name != InstanceSettings::GetDefaultName())
&& (m_instance_name != InstanceSettings::InvalidName())
&& live_instance)
m_owner.RegisterInstanceSettings (this);
}
InstanceSettings::~InstanceSettings ()
{
if (m_instance_name != InstanceSettings::GetDefaultName())
m_owner.UnregisterInstanceSettings (this);
}
const ConstString &
InstanceSettings::GetDefaultName ()
{
static const ConstString g_default_settings_name ("[DEFAULT]");
return g_default_settings_name;
}
const ConstString &
InstanceSettings::InvalidName ()
{
static const ConstString g_invalid_name ("Invalid instance name");
return g_invalid_name;
}
void
InstanceSettings::ChangeInstanceName (const std::string &new_instance_name)
{
m_instance_name.SetCString (new_instance_name.c_str());
}