sortix-mirror/kernel/kthread.cpp

376 lines
10 KiB
C++
Raw Permalink Normal View History

/*
* Copyright (c) 2012, 2014, 2021, 2022 Jonas 'Sortie' Termansen.
*
* Permission to use, copy, modify, and distribute this software for any
* purpose with or without fee is hereby granted, provided that the above
* copyright notice and this permission notice appear in all copies.
*
* THE SOFTWARE IS PROVIDED "AS IS" AND THE AUTHOR DISCLAIMS ALL WARRANTIES
* WITH REGARD TO THIS SOFTWARE INCLUDING ALL IMPLIED WARRANTIES OF
* MERCHANTABILITY AND FITNESS. IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR
* ANY SPECIAL, DIRECT, INDIRECT, OR CONSEQUENTIAL DAMAGES OR ANY DAMAGES
* WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR PROFITS, WHETHER IN AN
* ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS ACTION, ARISING OUT OF
* OR IN CONNECTION WITH THE USE OR PERFORMANCE OF THIS SOFTWARE.
*
* kthread.cpp
* Utility and synchronization mechanisms for kernel threads.
*/
#include <limits.h>
2013-05-12 22:52:58 +00:00
#include <sortix/signal.h>
2013-10-27 00:42:10 +00:00
#include <sortix/kernel/kernel.h>
#include <sortix/kernel/kthread.h>
#include <sortix/kernel/process.h>
2013-01-09 09:47:22 +00:00
#include <sortix/kernel/scheduler.h>
2013-05-12 22:52:58 +00:00
#include <sortix/kernel/signal.h>
#include <sortix/kernel/thread.h>
2013-10-27 00:42:10 +00:00
#include <sortix/kernel/worker.h>
#include "uart.h"
namespace Sortix {
static kthread_mutex_t kutex_lock = KTHREAD_MUTEX_INITIALIZER;
static Thread* kutex_first_waiting;
static Thread* kutex_last_waiting;
void kthread_yield()
{
Thread* thread = CurrentThread();
thread->yield_operation = YIELD_OPERATION_NONE;
#if defined(__i386__) || defined(__x86_64__)
asm volatile ("int $129");
#else
#error "kthread_yield needs to be implemented"
#endif
}
void kthread_wait_futex()
{
Thread* thread = CurrentThread();
thread->yield_operation = YIELD_OPERATION_WAIT_FUTEX;
#if defined(__i386__) || defined(__x86_64__)
asm volatile ("int $129");
#else
#error "kthread_wait_futex needs to be implemented"
#endif
}
void kthread_wait_futex_signal()
{
Thread* thread = CurrentThread();
thread->yield_operation = YIELD_OPERATION_WAIT_FUTEX_SIGNAL;
#if defined(__i386__) || defined(__x86_64__)
asm volatile ("int $129");
#else
#error "kthread_wait_futex needs to be implemented"
#endif
}
static void kthread_wait_kutex()
{
Thread* thread = CurrentThread();
thread->yield_operation = YIELD_OPERATION_WAIT_KUTEX;
#if defined(__i386__) || defined(__x86_64__)
asm volatile ("int $129");
#else
#error "kthread_wait_kutex needs to be implemented"
#endif
}
static void kthread_wait_kutex_signal()
{
Thread* thread = CurrentThread();
thread->yield_operation = YIELD_OPERATION_WAIT_KUTEX_SIGNAL;
#if defined(__i386__) || defined(__x86_64__)
asm volatile ("int $129");
#else
#error "kthread_wait_kutex needs to be implemented"
#endif
}
void kthread_wake_futex(Thread* thread)
{
Scheduler::SetThreadState(thread, ThreadState::RUNNABLE, true);
}
static const int UNLOCKED = 0;
static const int LOCKED = 1;
static const int CONTENDED = 2;
void kthread_spinlock_lock(kthread_mutex_t* mutex)
{
while ( true )
{
int state = UNLOCKED;
int desired = LOCKED;
if ( __atomic_compare_exchange_n(mutex, &state, desired, false,
__ATOMIC_SEQ_CST, __ATOMIC_SEQ_CST) )
break;
}
}
void kthread_spinlock_unlock(kthread_mutex_t* mutex)
{
__atomic_store_n(mutex, UNLOCKED, __ATOMIC_SEQ_CST);
}
static bool kutex_wait(int* address, int value, bool signal)
{
// TODO: Use a per-mutex wait queue instead.
Thread* thread = CurrentThread();
bool was_enabled = Interrupt::SetEnabled(false);
kthread_spinlock_lock(&kutex_lock);
thread->kutex_address = (uintptr_t) address;
thread->kutex_woken = false;
thread->kutex_prev_waiting = kutex_last_waiting;
thread->kutex_next_waiting = NULL;
(kutex_last_waiting ?
kutex_last_waiting->kutex_next_waiting :
kutex_first_waiting) = thread;
kutex_last_waiting = thread;
kthread_spinlock_unlock(&kutex_lock);
Interrupt::SetEnabled(was_enabled);
thread->timer_woken = false;
if ( __atomic_load_n(address, __ATOMIC_SEQ_CST) == value )
{
if ( signal )
kthread_wait_kutex_signal();
else
kthread_wait_kutex();
}
Interrupt::SetEnabled(false);
kthread_spinlock_lock(&kutex_lock);
bool result = thread->kutex_woken || !signal || !Signal::IsPending();
thread->kutex_address = 0;
thread->kutex_woken = false;
(thread->kutex_prev_waiting ?
thread->kutex_prev_waiting->kutex_next_waiting :
kutex_first_waiting) = thread->kutex_next_waiting;
(thread->kutex_next_waiting ?
thread->kutex_next_waiting->kutex_prev_waiting :
kutex_last_waiting) = thread->kutex_prev_waiting;
thread->kutex_prev_waiting = NULL;
thread->kutex_next_waiting = NULL;
kthread_spinlock_unlock(&kutex_lock);
Interrupt::SetEnabled(was_enabled);
return result;
}
static void kutex_wake(int* address, int count)
{
bool was_enabled = Interrupt::SetEnabled(false);
kthread_spinlock_lock(&kutex_lock);
for ( Thread* waiter = kutex_first_waiting;
0 < count && waiter;
waiter = waiter->kutex_next_waiting )
{
if ( waiter->kutex_address == (uintptr_t) address )
{
waiter->kutex_woken = true;
kthread_wake_futex(waiter);
if ( count != INT_MAX )
count--;
}
}
kthread_spinlock_unlock(&kutex_lock);
Interrupt::SetEnabled(was_enabled);
}
bool kthread_mutex_trylock(kthread_mutex_t* mutex)
{
int state = UNLOCKED;
if ( !__atomic_compare_exchange_n(mutex, &state, LOCKED, false,
__ATOMIC_SEQ_CST, __ATOMIC_SEQ_CST) )
return false;
return true;
}
void kthread_mutex_lock(kthread_mutex_t* mutex)
{
int state = UNLOCKED;
int desired = LOCKED;
while ( !__atomic_compare_exchange_n(mutex, &state, desired, false,
__ATOMIC_SEQ_CST, __ATOMIC_SEQ_CST) )
{
if ( state == LOCKED &&
!__atomic_compare_exchange_n(mutex, &state, CONTENDED, false,
__ATOMIC_SEQ_CST, __ATOMIC_SEQ_CST) )
{
state = UNLOCKED;
continue;
}
desired = CONTENDED;
kutex_wait(mutex, CONTENDED, false);
state = UNLOCKED;
}
}
bool kthread_mutex_lock_signal(kthread_mutex_t* mutex)
{
int state = UNLOCKED;
int desired = LOCKED;
while ( !__atomic_compare_exchange_n(mutex, &state, desired, false,
__ATOMIC_SEQ_CST, __ATOMIC_SEQ_CST) )
{
if ( state == LOCKED &&
!__atomic_compare_exchange_n(mutex, &state, CONTENDED, false,
__ATOMIC_SEQ_CST, __ATOMIC_SEQ_CST) )
{
state = UNLOCKED;
continue;
}
desired = CONTENDED;
if ( !kutex_wait(mutex, CONTENDED, true) )
return false;
state = UNLOCKED;
}
return true;
}
void kthread_mutex_unlock(kthread_mutex_t* mutex)
{
// TODO: Multiple threads could have caused the contention and this wakes
// only one such thread, but then the mutex returns to normal and the
// subsequent unlocks doesn't wake the other contended threads. Work
// around this bug by waking all of them instead, but it's better to
// instead count the number of waiting threads.
if ( __atomic_exchange_n(mutex, UNLOCKED, __ATOMIC_SEQ_CST) == CONTENDED )
kutex_wake(mutex, INT_MAX);
}
Multithreaded kernel and improvement of signal handling. Pardon the big ass-commit, this took months to develop and debug and the refactoring got so far that a clean merge became impossible. The good news is that this commit does quite a bit of cleaning up and generally improves the kernel quality. This makes the kernel fully pre-emptive and multithreaded. This was done by rewriting the interrupt code, the scheduler, introducing new threading primitives, and rewriting large parts of the kernel. During the past few commits the kernel has had its device drivers thread secured; this commit thread secures large parts of the core kernel. There still remains some parts of the kernel that is _not_ thread secured, but this is not a problem at this point. Each user-space thread has an associated kernel stack that it uses when it goes into kernel mode. This stack is by default 8 KiB since that value works for me and is also used by Linux. Strange things tends to happen on x86 in case of a stack overflow - there is no ideal way to catch such a situation right now. The system call conventions were changed, too. The %edx register is now used to provide the errno value of the call, instead of the kernel writing it into a registered global variable. The system call code has also been updated to better reflect the native calling conventions: not all registers have to be preserved. This makes system calls faster and simplifies the assembly. In the kernel, there is no longer the event.h header or the hacky method of 'resuming system calls' that closely resembles cooperative multitasking. If a system call wants to block, it should just block. The signal handling was also improved significantly. At this point, signals cannot interrupt kernel threads (but can always interrupt user-space threads if enabled), which introduces some problems with how a SIGINT could interrupt a blocking read, for instance. This commit introduces and uses a number of new primitives such as kthread_lock_mutex_signal() that attempts to get the lock but fails if a signal is pending. In this manner, the kernel is safer as kernel threads cannot be shut down inconveniently, but in return for complexity as blocking operations must check they if they should fail. Process exiting has also been refactored significantly. The _exit(2) system call sets the exit code and sends SIGKILL to all the threads in the process. Once all the threads have cleaned themselves up and exited, a worker thread calls the process's LastPrayer() method that unmaps memory, deletes the address space, notifies the parent, etc. This provides a very robust way to terminate processes as even half-constructed processes (during a failing fork for instance) can be gracefully terminated. I have introduced a number of kernel threads to help avoid threading problems and simplify kernel design. For instance, there is now a functional generic kernel worker thread that any kernel thread can schedule jobs for. Interrupt handlers run with interrupts off (hence they cannot call kthread_ functions as it may deadlock the system if another thread holds the lock) therefore they cannot use the standard kernel worker threads. Instead, they use a special purpose interrupt worker thread that works much like the generic one expect that interrupt handlers can safely queue work with interrupts off. Note that this also means that interrupt handlers cannot allocate memory or print to the kernel log/screen as such mechanisms uses locks. I'll introduce a lock free algorithm for such cases later on. The boot process has also changed. The original kernel init thread in kernel.cpp creates a new bootstrap thread and becomes the system idle thread. Note that pid=0 now means the kernel, as there is no longer a system idle process. The bootstrap thread launches all the kernel worker threads and then creates a new process and loads /bin/init into it and then creates a thread in pid=1, which starts the system. The bootstrap thread then quietly waits for pid=1 to exit after which it shuts down/reboots/panics the system. In general, the introduction of race conditions and dead locks have forced me to revise a lot of the design and make sure it was thread secure. Since early parts of the kernel was quite hacky, I had to refactor such code. So it seems that the risk of dead locks forces me to write better code. Note that a real preemptive multithreaded kernel simplifies the construction of blocking system calls. My hope is that this will trigger a clean up of the filesystem code that current is almost beyond repair. Almost all of the kernel was modified during this refactoring. To the extent possible, these changes have been backported to older non-multithreaded kernel, but many changes were tightly coupled and went into this commit. Of interest is the implementation of the kthread_ api based on the design of pthreads; this library allows easy synchronization mechanisms and includes C++-style scoped locks. This commit also introduces new worker threads and tested mechanisms for interrupt handlers to schedule work in a kernel worker thread. A lot of code have been rewritten from scratch and has become a lot more stable and correct. Share and enjoy!
2012-08-01 15:30:34 +00:00
// The kernel thread needs another stack to delete its own stack.
static void kthread_do_kill_thread(void* user)
{
Multithreaded kernel and improvement of signal handling. Pardon the big ass-commit, this took months to develop and debug and the refactoring got so far that a clean merge became impossible. The good news is that this commit does quite a bit of cleaning up and generally improves the kernel quality. This makes the kernel fully pre-emptive and multithreaded. This was done by rewriting the interrupt code, the scheduler, introducing new threading primitives, and rewriting large parts of the kernel. During the past few commits the kernel has had its device drivers thread secured; this commit thread secures large parts of the core kernel. There still remains some parts of the kernel that is _not_ thread secured, but this is not a problem at this point. Each user-space thread has an associated kernel stack that it uses when it goes into kernel mode. This stack is by default 8 KiB since that value works for me and is also used by Linux. Strange things tends to happen on x86 in case of a stack overflow - there is no ideal way to catch such a situation right now. The system call conventions were changed, too. The %edx register is now used to provide the errno value of the call, instead of the kernel writing it into a registered global variable. The system call code has also been updated to better reflect the native calling conventions: not all registers have to be preserved. This makes system calls faster and simplifies the assembly. In the kernel, there is no longer the event.h header or the hacky method of 'resuming system calls' that closely resembles cooperative multitasking. If a system call wants to block, it should just block. The signal handling was also improved significantly. At this point, signals cannot interrupt kernel threads (but can always interrupt user-space threads if enabled), which introduces some problems with how a SIGINT could interrupt a blocking read, for instance. This commit introduces and uses a number of new primitives such as kthread_lock_mutex_signal() that attempts to get the lock but fails if a signal is pending. In this manner, the kernel is safer as kernel threads cannot be shut down inconveniently, but in return for complexity as blocking operations must check they if they should fail. Process exiting has also been refactored significantly. The _exit(2) system call sets the exit code and sends SIGKILL to all the threads in the process. Once all the threads have cleaned themselves up and exited, a worker thread calls the process's LastPrayer() method that unmaps memory, deletes the address space, notifies the parent, etc. This provides a very robust way to terminate processes as even half-constructed processes (during a failing fork for instance) can be gracefully terminated. I have introduced a number of kernel threads to help avoid threading problems and simplify kernel design. For instance, there is now a functional generic kernel worker thread that any kernel thread can schedule jobs for. Interrupt handlers run with interrupts off (hence they cannot call kthread_ functions as it may deadlock the system if another thread holds the lock) therefore they cannot use the standard kernel worker threads. Instead, they use a special purpose interrupt worker thread that works much like the generic one expect that interrupt handlers can safely queue work with interrupts off. Note that this also means that interrupt handlers cannot allocate memory or print to the kernel log/screen as such mechanisms uses locks. I'll introduce a lock free algorithm for such cases later on. The boot process has also changed. The original kernel init thread in kernel.cpp creates a new bootstrap thread and becomes the system idle thread. Note that pid=0 now means the kernel, as there is no longer a system idle process. The bootstrap thread launches all the kernel worker threads and then creates a new process and loads /bin/init into it and then creates a thread in pid=1, which starts the system. The bootstrap thread then quietly waits for pid=1 to exit after which it shuts down/reboots/panics the system. In general, the introduction of race conditions and dead locks have forced me to revise a lot of the design and make sure it was thread secure. Since early parts of the kernel was quite hacky, I had to refactor such code. So it seems that the risk of dead locks forces me to write better code. Note that a real preemptive multithreaded kernel simplifies the construction of blocking system calls. My hope is that this will trigger a clean up of the filesystem code that current is almost beyond repair. Almost all of the kernel was modified during this refactoring. To the extent possible, these changes have been backported to older non-multithreaded kernel, but many changes were tightly coupled and went into this commit. Of interest is the implementation of the kthread_ api based on the design of pthreads; this library allows easy synchronization mechanisms and includes C++-style scoped locks. This commit also introduces new worker threads and tested mechanisms for interrupt handlers to schedule work in a kernel worker thread. A lot of code have been rewritten from scratch and has become a lot more stable and correct. Share and enjoy!
2012-08-01 15:30:34 +00:00
Thread* thread = (Thread*) user;
2013-01-09 09:47:22 +00:00
while ( thread->state != ThreadState::DEAD )
Multithreaded kernel and improvement of signal handling. Pardon the big ass-commit, this took months to develop and debug and the refactoring got so far that a clean merge became impossible. The good news is that this commit does quite a bit of cleaning up and generally improves the kernel quality. This makes the kernel fully pre-emptive and multithreaded. This was done by rewriting the interrupt code, the scheduler, introducing new threading primitives, and rewriting large parts of the kernel. During the past few commits the kernel has had its device drivers thread secured; this commit thread secures large parts of the core kernel. There still remains some parts of the kernel that is _not_ thread secured, but this is not a problem at this point. Each user-space thread has an associated kernel stack that it uses when it goes into kernel mode. This stack is by default 8 KiB since that value works for me and is also used by Linux. Strange things tends to happen on x86 in case of a stack overflow - there is no ideal way to catch such a situation right now. The system call conventions were changed, too. The %edx register is now used to provide the errno value of the call, instead of the kernel writing it into a registered global variable. The system call code has also been updated to better reflect the native calling conventions: not all registers have to be preserved. This makes system calls faster and simplifies the assembly. In the kernel, there is no longer the event.h header or the hacky method of 'resuming system calls' that closely resembles cooperative multitasking. If a system call wants to block, it should just block. The signal handling was also improved significantly. At this point, signals cannot interrupt kernel threads (but can always interrupt user-space threads if enabled), which introduces some problems with how a SIGINT could interrupt a blocking read, for instance. This commit introduces and uses a number of new primitives such as kthread_lock_mutex_signal() that attempts to get the lock but fails if a signal is pending. In this manner, the kernel is safer as kernel threads cannot be shut down inconveniently, but in return for complexity as blocking operations must check they if they should fail. Process exiting has also been refactored significantly. The _exit(2) system call sets the exit code and sends SIGKILL to all the threads in the process. Once all the threads have cleaned themselves up and exited, a worker thread calls the process's LastPrayer() method that unmaps memory, deletes the address space, notifies the parent, etc. This provides a very robust way to terminate processes as even half-constructed processes (during a failing fork for instance) can be gracefully terminated. I have introduced a number of kernel threads to help avoid threading problems and simplify kernel design. For instance, there is now a functional generic kernel worker thread that any kernel thread can schedule jobs for. Interrupt handlers run with interrupts off (hence they cannot call kthread_ functions as it may deadlock the system if another thread holds the lock) therefore they cannot use the standard kernel worker threads. Instead, they use a special purpose interrupt worker thread that works much like the generic one expect that interrupt handlers can safely queue work with interrupts off. Note that this also means that interrupt handlers cannot allocate memory or print to the kernel log/screen as such mechanisms uses locks. I'll introduce a lock free algorithm for such cases later on. The boot process has also changed. The original kernel init thread in kernel.cpp creates a new bootstrap thread and becomes the system idle thread. Note that pid=0 now means the kernel, as there is no longer a system idle process. The bootstrap thread launches all the kernel worker threads and then creates a new process and loads /bin/init into it and then creates a thread in pid=1, which starts the system. The bootstrap thread then quietly waits for pid=1 to exit after which it shuts down/reboots/panics the system. In general, the introduction of race conditions and dead locks have forced me to revise a lot of the design and make sure it was thread secure. Since early parts of the kernel was quite hacky, I had to refactor such code. So it seems that the risk of dead locks forces me to write better code. Note that a real preemptive multithreaded kernel simplifies the construction of blocking system calls. My hope is that this will trigger a clean up of the filesystem code that current is almost beyond repair. Almost all of the kernel was modified during this refactoring. To the extent possible, these changes have been backported to older non-multithreaded kernel, but many changes were tightly coupled and went into this commit. Of interest is the implementation of the kthread_ api based on the design of pthreads; this library allows easy synchronization mechanisms and includes C++-style scoped locks. This commit also introduces new worker threads and tested mechanisms for interrupt handlers to schedule work in a kernel worker thread. A lot of code have been rewritten from scratch and has become a lot more stable and correct. Share and enjoy!
2012-08-01 15:30:34 +00:00
kthread_yield();
delete thread;
}
void kthread_exit()
{
Process* process = CurrentProcess();
// Note: This requires all threads in this process to have been made by
// only threads in this process, except the initial thread. Otherwise more
// threads may appear, and we can't conclude whether this is the last thread
// in the process to exit.
kthread_mutex_lock(&process->threadlock);
bool is_last_to_exit = --process->threads_not_exiting_count == 0;
kthread_mutex_unlock(&process->threadlock);
// All other threads in the process have committed to exiting, though they
// might not have exited yet. However, we know they are only running the
// below code that schedules thread termination. It's therefore safe to run
// a final process termination step without interference.
if ( is_last_to_exit )
process->OnLastThreadExit();
Multithreaded kernel and improvement of signal handling. Pardon the big ass-commit, this took months to develop and debug and the refactoring got so far that a clean merge became impossible. The good news is that this commit does quite a bit of cleaning up and generally improves the kernel quality. This makes the kernel fully pre-emptive and multithreaded. This was done by rewriting the interrupt code, the scheduler, introducing new threading primitives, and rewriting large parts of the kernel. During the past few commits the kernel has had its device drivers thread secured; this commit thread secures large parts of the core kernel. There still remains some parts of the kernel that is _not_ thread secured, but this is not a problem at this point. Each user-space thread has an associated kernel stack that it uses when it goes into kernel mode. This stack is by default 8 KiB since that value works for me and is also used by Linux. Strange things tends to happen on x86 in case of a stack overflow - there is no ideal way to catch such a situation right now. The system call conventions were changed, too. The %edx register is now used to provide the errno value of the call, instead of the kernel writing it into a registered global variable. The system call code has also been updated to better reflect the native calling conventions: not all registers have to be preserved. This makes system calls faster and simplifies the assembly. In the kernel, there is no longer the event.h header or the hacky method of 'resuming system calls' that closely resembles cooperative multitasking. If a system call wants to block, it should just block. The signal handling was also improved significantly. At this point, signals cannot interrupt kernel threads (but can always interrupt user-space threads if enabled), which introduces some problems with how a SIGINT could interrupt a blocking read, for instance. This commit introduces and uses a number of new primitives such as kthread_lock_mutex_signal() that attempts to get the lock but fails if a signal is pending. In this manner, the kernel is safer as kernel threads cannot be shut down inconveniently, but in return for complexity as blocking operations must check they if they should fail. Process exiting has also been refactored significantly. The _exit(2) system call sets the exit code and sends SIGKILL to all the threads in the process. Once all the threads have cleaned themselves up and exited, a worker thread calls the process's LastPrayer() method that unmaps memory, deletes the address space, notifies the parent, etc. This provides a very robust way to terminate processes as even half-constructed processes (during a failing fork for instance) can be gracefully terminated. I have introduced a number of kernel threads to help avoid threading problems and simplify kernel design. For instance, there is now a functional generic kernel worker thread that any kernel thread can schedule jobs for. Interrupt handlers run with interrupts off (hence they cannot call kthread_ functions as it may deadlock the system if another thread holds the lock) therefore they cannot use the standard kernel worker threads. Instead, they use a special purpose interrupt worker thread that works much like the generic one expect that interrupt handlers can safely queue work with interrupts off. Note that this also means that interrupt handlers cannot allocate memory or print to the kernel log/screen as such mechanisms uses locks. I'll introduce a lock free algorithm for such cases later on. The boot process has also changed. The original kernel init thread in kernel.cpp creates a new bootstrap thread and becomes the system idle thread. Note that pid=0 now means the kernel, as there is no longer a system idle process. The bootstrap thread launches all the kernel worker threads and then creates a new process and loads /bin/init into it and then creates a thread in pid=1, which starts the system. The bootstrap thread then quietly waits for pid=1 to exit after which it shuts down/reboots/panics the system. In general, the introduction of race conditions and dead locks have forced me to revise a lot of the design and make sure it was thread secure. Since early parts of the kernel was quite hacky, I had to refactor such code. So it seems that the risk of dead locks forces me to write better code. Note that a real preemptive multithreaded kernel simplifies the construction of blocking system calls. My hope is that this will trigger a clean up of the filesystem code that current is almost beyond repair. Almost all of the kernel was modified during this refactoring. To the extent possible, these changes have been backported to older non-multithreaded kernel, but many changes were tightly coupled and went into this commit. Of interest is the implementation of the kthread_ api based on the design of pthreads; this library allows easy synchronization mechanisms and includes C++-style scoped locks. This commit also introduces new worker threads and tested mechanisms for interrupt handlers to schedule work in a kernel worker thread. A lot of code have been rewritten from scratch and has become a lot more stable and correct. Share and enjoy!
2012-08-01 15:30:34 +00:00
Worker::Schedule(kthread_do_kill_thread, CurrentThread());
#if defined(__i386__) || defined(__x86_64__)
asm volatile ("int $132");
#else
#error "kthread_exit needs to be implemented"
#endif
2013-06-26 15:14:07 +00:00
__builtin_unreachable();
}
Multithreaded kernel and improvement of signal handling. Pardon the big ass-commit, this took months to develop and debug and the refactoring got so far that a clean merge became impossible. The good news is that this commit does quite a bit of cleaning up and generally improves the kernel quality. This makes the kernel fully pre-emptive and multithreaded. This was done by rewriting the interrupt code, the scheduler, introducing new threading primitives, and rewriting large parts of the kernel. During the past few commits the kernel has had its device drivers thread secured; this commit thread secures large parts of the core kernel. There still remains some parts of the kernel that is _not_ thread secured, but this is not a problem at this point. Each user-space thread has an associated kernel stack that it uses when it goes into kernel mode. This stack is by default 8 KiB since that value works for me and is also used by Linux. Strange things tends to happen on x86 in case of a stack overflow - there is no ideal way to catch such a situation right now. The system call conventions were changed, too. The %edx register is now used to provide the errno value of the call, instead of the kernel writing it into a registered global variable. The system call code has also been updated to better reflect the native calling conventions: not all registers have to be preserved. This makes system calls faster and simplifies the assembly. In the kernel, there is no longer the event.h header or the hacky method of 'resuming system calls' that closely resembles cooperative multitasking. If a system call wants to block, it should just block. The signal handling was also improved significantly. At this point, signals cannot interrupt kernel threads (but can always interrupt user-space threads if enabled), which introduces some problems with how a SIGINT could interrupt a blocking read, for instance. This commit introduces and uses a number of new primitives such as kthread_lock_mutex_signal() that attempts to get the lock but fails if a signal is pending. In this manner, the kernel is safer as kernel threads cannot be shut down inconveniently, but in return for complexity as blocking operations must check they if they should fail. Process exiting has also been refactored significantly. The _exit(2) system call sets the exit code and sends SIGKILL to all the threads in the process. Once all the threads have cleaned themselves up and exited, a worker thread calls the process's LastPrayer() method that unmaps memory, deletes the address space, notifies the parent, etc. This provides a very robust way to terminate processes as even half-constructed processes (during a failing fork for instance) can be gracefully terminated. I have introduced a number of kernel threads to help avoid threading problems and simplify kernel design. For instance, there is now a functional generic kernel worker thread that any kernel thread can schedule jobs for. Interrupt handlers run with interrupts off (hence they cannot call kthread_ functions as it may deadlock the system if another thread holds the lock) therefore they cannot use the standard kernel worker threads. Instead, they use a special purpose interrupt worker thread that works much like the generic one expect that interrupt handlers can safely queue work with interrupts off. Note that this also means that interrupt handlers cannot allocate memory or print to the kernel log/screen as such mechanisms uses locks. I'll introduce a lock free algorithm for such cases later on. The boot process has also changed. The original kernel init thread in kernel.cpp creates a new bootstrap thread and becomes the system idle thread. Note that pid=0 now means the kernel, as there is no longer a system idle process. The bootstrap thread launches all the kernel worker threads and then creates a new process and loads /bin/init into it and then creates a thread in pid=1, which starts the system. The bootstrap thread then quietly waits for pid=1 to exit after which it shuts down/reboots/panics the system. In general, the introduction of race conditions and dead locks have forced me to revise a lot of the design and make sure it was thread secure. Since early parts of the kernel was quite hacky, I had to refactor such code. So it seems that the risk of dead locks forces me to write better code. Note that a real preemptive multithreaded kernel simplifies the construction of blocking system calls. My hope is that this will trigger a clean up of the filesystem code that current is almost beyond repair. Almost all of the kernel was modified during this refactoring. To the extent possible, these changes have been backported to older non-multithreaded kernel, but many changes were tightly coupled and went into this commit. Of interest is the implementation of the kthread_ api based on the design of pthreads; this library allows easy synchronization mechanisms and includes C++-style scoped locks. This commit also introduces new worker threads and tested mechanisms for interrupt handlers to schedule work in a kernel worker thread. A lot of code have been rewritten from scratch and has become a lot more stable and correct. Share and enjoy!
2012-08-01 15:30:34 +00:00
struct kthread_cond_elem
{
Multithreaded kernel and improvement of signal handling. Pardon the big ass-commit, this took months to develop and debug and the refactoring got so far that a clean merge became impossible. The good news is that this commit does quite a bit of cleaning up and generally improves the kernel quality. This makes the kernel fully pre-emptive and multithreaded. This was done by rewriting the interrupt code, the scheduler, introducing new threading primitives, and rewriting large parts of the kernel. During the past few commits the kernel has had its device drivers thread secured; this commit thread secures large parts of the core kernel. There still remains some parts of the kernel that is _not_ thread secured, but this is not a problem at this point. Each user-space thread has an associated kernel stack that it uses when it goes into kernel mode. This stack is by default 8 KiB since that value works for me and is also used by Linux. Strange things tends to happen on x86 in case of a stack overflow - there is no ideal way to catch such a situation right now. The system call conventions were changed, too. The %edx register is now used to provide the errno value of the call, instead of the kernel writing it into a registered global variable. The system call code has also been updated to better reflect the native calling conventions: not all registers have to be preserved. This makes system calls faster and simplifies the assembly. In the kernel, there is no longer the event.h header or the hacky method of 'resuming system calls' that closely resembles cooperative multitasking. If a system call wants to block, it should just block. The signal handling was also improved significantly. At this point, signals cannot interrupt kernel threads (but can always interrupt user-space threads if enabled), which introduces some problems with how a SIGINT could interrupt a blocking read, for instance. This commit introduces and uses a number of new primitives such as kthread_lock_mutex_signal() that attempts to get the lock but fails if a signal is pending. In this manner, the kernel is safer as kernel threads cannot be shut down inconveniently, but in return for complexity as blocking operations must check they if they should fail. Process exiting has also been refactored significantly. The _exit(2) system call sets the exit code and sends SIGKILL to all the threads in the process. Once all the threads have cleaned themselves up and exited, a worker thread calls the process's LastPrayer() method that unmaps memory, deletes the address space, notifies the parent, etc. This provides a very robust way to terminate processes as even half-constructed processes (during a failing fork for instance) can be gracefully terminated. I have introduced a number of kernel threads to help avoid threading problems and simplify kernel design. For instance, there is now a functional generic kernel worker thread that any kernel thread can schedule jobs for. Interrupt handlers run with interrupts off (hence they cannot call kthread_ functions as it may deadlock the system if another thread holds the lock) therefore they cannot use the standard kernel worker threads. Instead, they use a special purpose interrupt worker thread that works much like the generic one expect that interrupt handlers can safely queue work with interrupts off. Note that this also means that interrupt handlers cannot allocate memory or print to the kernel log/screen as such mechanisms uses locks. I'll introduce a lock free algorithm for such cases later on. The boot process has also changed. The original kernel init thread in kernel.cpp creates a new bootstrap thread and becomes the system idle thread. Note that pid=0 now means the kernel, as there is no longer a system idle process. The bootstrap thread launches all the kernel worker threads and then creates a new process and loads /bin/init into it and then creates a thread in pid=1, which starts the system. The bootstrap thread then quietly waits for pid=1 to exit after which it shuts down/reboots/panics the system. In general, the introduction of race conditions and dead locks have forced me to revise a lot of the design and make sure it was thread secure. Since early parts of the kernel was quite hacky, I had to refactor such code. So it seems that the risk of dead locks forces me to write better code. Note that a real preemptive multithreaded kernel simplifies the construction of blocking system calls. My hope is that this will trigger a clean up of the filesystem code that current is almost beyond repair. Almost all of the kernel was modified during this refactoring. To the extent possible, these changes have been backported to older non-multithreaded kernel, but many changes were tightly coupled and went into this commit. Of interest is the implementation of the kthread_ api based on the design of pthreads; this library allows easy synchronization mechanisms and includes C++-style scoped locks. This commit also introduces new worker threads and tested mechanisms for interrupt handlers to schedule work in a kernel worker thread. A lot of code have been rewritten from scratch and has become a lot more stable and correct. Share and enjoy!
2012-08-01 15:30:34 +00:00
kthread_cond_elem_t* next;
kthread_cond_elem_t* prev;
int woken;
Multithreaded kernel and improvement of signal handling. Pardon the big ass-commit, this took months to develop and debug and the refactoring got so far that a clean merge became impossible. The good news is that this commit does quite a bit of cleaning up and generally improves the kernel quality. This makes the kernel fully pre-emptive and multithreaded. This was done by rewriting the interrupt code, the scheduler, introducing new threading primitives, and rewriting large parts of the kernel. During the past few commits the kernel has had its device drivers thread secured; this commit thread secures large parts of the core kernel. There still remains some parts of the kernel that is _not_ thread secured, but this is not a problem at this point. Each user-space thread has an associated kernel stack that it uses when it goes into kernel mode. This stack is by default 8 KiB since that value works for me and is also used by Linux. Strange things tends to happen on x86 in case of a stack overflow - there is no ideal way to catch such a situation right now. The system call conventions were changed, too. The %edx register is now used to provide the errno value of the call, instead of the kernel writing it into a registered global variable. The system call code has also been updated to better reflect the native calling conventions: not all registers have to be preserved. This makes system calls faster and simplifies the assembly. In the kernel, there is no longer the event.h header or the hacky method of 'resuming system calls' that closely resembles cooperative multitasking. If a system call wants to block, it should just block. The signal handling was also improved significantly. At this point, signals cannot interrupt kernel threads (but can always interrupt user-space threads if enabled), which introduces some problems with how a SIGINT could interrupt a blocking read, for instance. This commit introduces and uses a number of new primitives such as kthread_lock_mutex_signal() that attempts to get the lock but fails if a signal is pending. In this manner, the kernel is safer as kernel threads cannot be shut down inconveniently, but in return for complexity as blocking operations must check they if they should fail. Process exiting has also been refactored significantly. The _exit(2) system call sets the exit code and sends SIGKILL to all the threads in the process. Once all the threads have cleaned themselves up and exited, a worker thread calls the process's LastPrayer() method that unmaps memory, deletes the address space, notifies the parent, etc. This provides a very robust way to terminate processes as even half-constructed processes (during a failing fork for instance) can be gracefully terminated. I have introduced a number of kernel threads to help avoid threading problems and simplify kernel design. For instance, there is now a functional generic kernel worker thread that any kernel thread can schedule jobs for. Interrupt handlers run with interrupts off (hence they cannot call kthread_ functions as it may deadlock the system if another thread holds the lock) therefore they cannot use the standard kernel worker threads. Instead, they use a special purpose interrupt worker thread that works much like the generic one expect that interrupt handlers can safely queue work with interrupts off. Note that this also means that interrupt handlers cannot allocate memory or print to the kernel log/screen as such mechanisms uses locks. I'll introduce a lock free algorithm for such cases later on. The boot process has also changed. The original kernel init thread in kernel.cpp creates a new bootstrap thread and becomes the system idle thread. Note that pid=0 now means the kernel, as there is no longer a system idle process. The bootstrap thread launches all the kernel worker threads and then creates a new process and loads /bin/init into it and then creates a thread in pid=1, which starts the system. The bootstrap thread then quietly waits for pid=1 to exit after which it shuts down/reboots/panics the system. In general, the introduction of race conditions and dead locks have forced me to revise a lot of the design and make sure it was thread secure. Since early parts of the kernel was quite hacky, I had to refactor such code. So it seems that the risk of dead locks forces me to write better code. Note that a real preemptive multithreaded kernel simplifies the construction of blocking system calls. My hope is that this will trigger a clean up of the filesystem code that current is almost beyond repair. Almost all of the kernel was modified during this refactoring. To the extent possible, these changes have been backported to older non-multithreaded kernel, but many changes were tightly coupled and went into this commit. Of interest is the implementation of the kthread_ api based on the design of pthreads; this library allows easy synchronization mechanisms and includes C++-style scoped locks. This commit also introduces new worker threads and tested mechanisms for interrupt handlers to schedule work in a kernel worker thread. A lot of code have been rewritten from scratch and has become a lot more stable and correct. Share and enjoy!
2012-08-01 15:30:34 +00:00
};
void kthread_cond_wait(kthread_cond_t* cond, kthread_mutex_t* mutex)
{
Multithreaded kernel and improvement of signal handling. Pardon the big ass-commit, this took months to develop and debug and the refactoring got so far that a clean merge became impossible. The good news is that this commit does quite a bit of cleaning up and generally improves the kernel quality. This makes the kernel fully pre-emptive and multithreaded. This was done by rewriting the interrupt code, the scheduler, introducing new threading primitives, and rewriting large parts of the kernel. During the past few commits the kernel has had its device drivers thread secured; this commit thread secures large parts of the core kernel. There still remains some parts of the kernel that is _not_ thread secured, but this is not a problem at this point. Each user-space thread has an associated kernel stack that it uses when it goes into kernel mode. This stack is by default 8 KiB since that value works for me and is also used by Linux. Strange things tends to happen on x86 in case of a stack overflow - there is no ideal way to catch such a situation right now. The system call conventions were changed, too. The %edx register is now used to provide the errno value of the call, instead of the kernel writing it into a registered global variable. The system call code has also been updated to better reflect the native calling conventions: not all registers have to be preserved. This makes system calls faster and simplifies the assembly. In the kernel, there is no longer the event.h header or the hacky method of 'resuming system calls' that closely resembles cooperative multitasking. If a system call wants to block, it should just block. The signal handling was also improved significantly. At this point, signals cannot interrupt kernel threads (but can always interrupt user-space threads if enabled), which introduces some problems with how a SIGINT could interrupt a blocking read, for instance. This commit introduces and uses a number of new primitives such as kthread_lock_mutex_signal() that attempts to get the lock but fails if a signal is pending. In this manner, the kernel is safer as kernel threads cannot be shut down inconveniently, but in return for complexity as blocking operations must check they if they should fail. Process exiting has also been refactored significantly. The _exit(2) system call sets the exit code and sends SIGKILL to all the threads in the process. Once all the threads have cleaned themselves up and exited, a worker thread calls the process's LastPrayer() method that unmaps memory, deletes the address space, notifies the parent, etc. This provides a very robust way to terminate processes as even half-constructed processes (during a failing fork for instance) can be gracefully terminated. I have introduced a number of kernel threads to help avoid threading problems and simplify kernel design. For instance, there is now a functional generic kernel worker thread that any kernel thread can schedule jobs for. Interrupt handlers run with interrupts off (hence they cannot call kthread_ functions as it may deadlock the system if another thread holds the lock) therefore they cannot use the standard kernel worker threads. Instead, they use a special purpose interrupt worker thread that works much like the generic one expect that interrupt handlers can safely queue work with interrupts off. Note that this also means that interrupt handlers cannot allocate memory or print to the kernel log/screen as such mechanisms uses locks. I'll introduce a lock free algorithm for such cases later on. The boot process has also changed. The original kernel init thread in kernel.cpp creates a new bootstrap thread and becomes the system idle thread. Note that pid=0 now means the kernel, as there is no longer a system idle process. The bootstrap thread launches all the kernel worker threads and then creates a new process and loads /bin/init into it and then creates a thread in pid=1, which starts the system. The bootstrap thread then quietly waits for pid=1 to exit after which it shuts down/reboots/panics the system. In general, the introduction of race conditions and dead locks have forced me to revise a lot of the design and make sure it was thread secure. Since early parts of the kernel was quite hacky, I had to refactor such code. So it seems that the risk of dead locks forces me to write better code. Note that a real preemptive multithreaded kernel simplifies the construction of blocking system calls. My hope is that this will trigger a clean up of the filesystem code that current is almost beyond repair. Almost all of the kernel was modified during this refactoring. To the extent possible, these changes have been backported to older non-multithreaded kernel, but many changes were tightly coupled and went into this commit. Of interest is the implementation of the kthread_ api based on the design of pthreads; this library allows easy synchronization mechanisms and includes C++-style scoped locks. This commit also introduces new worker threads and tested mechanisms for interrupt handlers to schedule work in a kernel worker thread. A lot of code have been rewritten from scratch and has become a lot more stable and correct. Share and enjoy!
2012-08-01 15:30:34 +00:00
kthread_cond_elem_t elem;
elem.next = NULL;
elem.prev = cond->last;
Multithreaded kernel and improvement of signal handling. Pardon the big ass-commit, this took months to develop and debug and the refactoring got so far that a clean merge became impossible. The good news is that this commit does quite a bit of cleaning up and generally improves the kernel quality. This makes the kernel fully pre-emptive and multithreaded. This was done by rewriting the interrupt code, the scheduler, introducing new threading primitives, and rewriting large parts of the kernel. During the past few commits the kernel has had its device drivers thread secured; this commit thread secures large parts of the core kernel. There still remains some parts of the kernel that is _not_ thread secured, but this is not a problem at this point. Each user-space thread has an associated kernel stack that it uses when it goes into kernel mode. This stack is by default 8 KiB since that value works for me and is also used by Linux. Strange things tends to happen on x86 in case of a stack overflow - there is no ideal way to catch such a situation right now. The system call conventions were changed, too. The %edx register is now used to provide the errno value of the call, instead of the kernel writing it into a registered global variable. The system call code has also been updated to better reflect the native calling conventions: not all registers have to be preserved. This makes system calls faster and simplifies the assembly. In the kernel, there is no longer the event.h header or the hacky method of 'resuming system calls' that closely resembles cooperative multitasking. If a system call wants to block, it should just block. The signal handling was also improved significantly. At this point, signals cannot interrupt kernel threads (but can always interrupt user-space threads if enabled), which introduces some problems with how a SIGINT could interrupt a blocking read, for instance. This commit introduces and uses a number of new primitives such as kthread_lock_mutex_signal() that attempts to get the lock but fails if a signal is pending. In this manner, the kernel is safer as kernel threads cannot be shut down inconveniently, but in return for complexity as blocking operations must check they if they should fail. Process exiting has also been refactored significantly. The _exit(2) system call sets the exit code and sends SIGKILL to all the threads in the process. Once all the threads have cleaned themselves up and exited, a worker thread calls the process's LastPrayer() method that unmaps memory, deletes the address space, notifies the parent, etc. This provides a very robust way to terminate processes as even half-constructed processes (during a failing fork for instance) can be gracefully terminated. I have introduced a number of kernel threads to help avoid threading problems and simplify kernel design. For instance, there is now a functional generic kernel worker thread that any kernel thread can schedule jobs for. Interrupt handlers run with interrupts off (hence they cannot call kthread_ functions as it may deadlock the system if another thread holds the lock) therefore they cannot use the standard kernel worker threads. Instead, they use a special purpose interrupt worker thread that works much like the generic one expect that interrupt handlers can safely queue work with interrupts off. Note that this also means that interrupt handlers cannot allocate memory or print to the kernel log/screen as such mechanisms uses locks. I'll introduce a lock free algorithm for such cases later on. The boot process has also changed. The original kernel init thread in kernel.cpp creates a new bootstrap thread and becomes the system idle thread. Note that pid=0 now means the kernel, as there is no longer a system idle process. The bootstrap thread launches all the kernel worker threads and then creates a new process and loads /bin/init into it and then creates a thread in pid=1, which starts the system. The bootstrap thread then quietly waits for pid=1 to exit after which it shuts down/reboots/panics the system. In general, the introduction of race conditions and dead locks have forced me to revise a lot of the design and make sure it was thread secure. Since early parts of the kernel was quite hacky, I had to refactor such code. So it seems that the risk of dead locks forces me to write better code. Note that a real preemptive multithreaded kernel simplifies the construction of blocking system calls. My hope is that this will trigger a clean up of the filesystem code that current is almost beyond repair. Almost all of the kernel was modified during this refactoring. To the extent possible, these changes have been backported to older non-multithreaded kernel, but many changes were tightly coupled and went into this commit. Of interest is the implementation of the kthread_ api based on the design of pthreads; this library allows easy synchronization mechanisms and includes C++-style scoped locks. This commit also introduces new worker threads and tested mechanisms for interrupt handlers to schedule work in a kernel worker thread. A lot of code have been rewritten from scratch and has become a lot more stable and correct. Share and enjoy!
2012-08-01 15:30:34 +00:00
elem.woken = 0;
if ( cond->last )
cond->last->next = &elem;
if ( !cond->first )
cond->first = &elem;
2013-07-11 16:05:02 +00:00
cond->last = &elem;
kthread_mutex_unlock(mutex);
while ( !__atomic_load_n(&elem.woken, __ATOMIC_SEQ_CST) )
kutex_wait(&elem.woken, 0, false);
kthread_mutex_lock(mutex);
if ( !__atomic_load_n(&elem.woken, __ATOMIC_SEQ_CST) )
Multithreaded kernel and improvement of signal handling. Pardon the big ass-commit, this took months to develop and debug and the refactoring got so far that a clean merge became impossible. The good news is that this commit does quite a bit of cleaning up and generally improves the kernel quality. This makes the kernel fully pre-emptive and multithreaded. This was done by rewriting the interrupt code, the scheduler, introducing new threading primitives, and rewriting large parts of the kernel. During the past few commits the kernel has had its device drivers thread secured; this commit thread secures large parts of the core kernel. There still remains some parts of the kernel that is _not_ thread secured, but this is not a problem at this point. Each user-space thread has an associated kernel stack that it uses when it goes into kernel mode. This stack is by default 8 KiB since that value works for me and is also used by Linux. Strange things tends to happen on x86 in case of a stack overflow - there is no ideal way to catch such a situation right now. The system call conventions were changed, too. The %edx register is now used to provide the errno value of the call, instead of the kernel writing it into a registered global variable. The system call code has also been updated to better reflect the native calling conventions: not all registers have to be preserved. This makes system calls faster and simplifies the assembly. In the kernel, there is no longer the event.h header or the hacky method of 'resuming system calls' that closely resembles cooperative multitasking. If a system call wants to block, it should just block. The signal handling was also improved significantly. At this point, signals cannot interrupt kernel threads (but can always interrupt user-space threads if enabled), which introduces some problems with how a SIGINT could interrupt a blocking read, for instance. This commit introduces and uses a number of new primitives such as kthread_lock_mutex_signal() that attempts to get the lock but fails if a signal is pending. In this manner, the kernel is safer as kernel threads cannot be shut down inconveniently, but in return for complexity as blocking operations must check they if they should fail. Process exiting has also been refactored significantly. The _exit(2) system call sets the exit code and sends SIGKILL to all the threads in the process. Once all the threads have cleaned themselves up and exited, a worker thread calls the process's LastPrayer() method that unmaps memory, deletes the address space, notifies the parent, etc. This provides a very robust way to terminate processes as even half-constructed processes (during a failing fork for instance) can be gracefully terminated. I have introduced a number of kernel threads to help avoid threading problems and simplify kernel design. For instance, there is now a functional generic kernel worker thread that any kernel thread can schedule jobs for. Interrupt handlers run with interrupts off (hence they cannot call kthread_ functions as it may deadlock the system if another thread holds the lock) therefore they cannot use the standard kernel worker threads. Instead, they use a special purpose interrupt worker thread that works much like the generic one expect that interrupt handlers can safely queue work with interrupts off. Note that this also means that interrupt handlers cannot allocate memory or print to the kernel log/screen as such mechanisms uses locks. I'll introduce a lock free algorithm for such cases later on. The boot process has also changed. The original kernel init thread in kernel.cpp creates a new bootstrap thread and becomes the system idle thread. Note that pid=0 now means the kernel, as there is no longer a system idle process. The bootstrap thread launches all the kernel worker threads and then creates a new process and loads /bin/init into it and then creates a thread in pid=1, which starts the system. The bootstrap thread then quietly waits for pid=1 to exit after which it shuts down/reboots/panics the system. In general, the introduction of race conditions and dead locks have forced me to revise a lot of the design and make sure it was thread secure. Since early parts of the kernel was quite hacky, I had to refactor such code. So it seems that the risk of dead locks forces me to write better code. Note that a real preemptive multithreaded kernel simplifies the construction of blocking system calls. My hope is that this will trigger a clean up of the filesystem code that current is almost beyond repair. Almost all of the kernel was modified during this refactoring. To the extent possible, these changes have been backported to older non-multithreaded kernel, but many changes were tightly coupled and went into this commit. Of interest is the implementation of the kthread_ api based on the design of pthreads; this library allows easy synchronization mechanisms and includes C++-style scoped locks. This commit also introduces new worker threads and tested mechanisms for interrupt handlers to schedule work in a kernel worker thread. A lot of code have been rewritten from scratch and has become a lot more stable and correct. Share and enjoy!
2012-08-01 15:30:34 +00:00
{
if ( elem.next )
elem.next->prev = elem.prev;
else
cond->last = elem.prev;
if ( elem.prev )
elem.prev->next = elem.next;
else
cond->first = elem.next;
Multithreaded kernel and improvement of signal handling. Pardon the big ass-commit, this took months to develop and debug and the refactoring got so far that a clean merge became impossible. The good news is that this commit does quite a bit of cleaning up and generally improves the kernel quality. This makes the kernel fully pre-emptive and multithreaded. This was done by rewriting the interrupt code, the scheduler, introducing new threading primitives, and rewriting large parts of the kernel. During the past few commits the kernel has had its device drivers thread secured; this commit thread secures large parts of the core kernel. There still remains some parts of the kernel that is _not_ thread secured, but this is not a problem at this point. Each user-space thread has an associated kernel stack that it uses when it goes into kernel mode. This stack is by default 8 KiB since that value works for me and is also used by Linux. Strange things tends to happen on x86 in case of a stack overflow - there is no ideal way to catch such a situation right now. The system call conventions were changed, too. The %edx register is now used to provide the errno value of the call, instead of the kernel writing it into a registered global variable. The system call code has also been updated to better reflect the native calling conventions: not all registers have to be preserved. This makes system calls faster and simplifies the assembly. In the kernel, there is no longer the event.h header or the hacky method of 'resuming system calls' that closely resembles cooperative multitasking. If a system call wants to block, it should just block. The signal handling was also improved significantly. At this point, signals cannot interrupt kernel threads (but can always interrupt user-space threads if enabled), which introduces some problems with how a SIGINT could interrupt a blocking read, for instance. This commit introduces and uses a number of new primitives such as kthread_lock_mutex_signal() that attempts to get the lock but fails if a signal is pending. In this manner, the kernel is safer as kernel threads cannot be shut down inconveniently, but in return for complexity as blocking operations must check they if they should fail. Process exiting has also been refactored significantly. The _exit(2) system call sets the exit code and sends SIGKILL to all the threads in the process. Once all the threads have cleaned themselves up and exited, a worker thread calls the process's LastPrayer() method that unmaps memory, deletes the address space, notifies the parent, etc. This provides a very robust way to terminate processes as even half-constructed processes (during a failing fork for instance) can be gracefully terminated. I have introduced a number of kernel threads to help avoid threading problems and simplify kernel design. For instance, there is now a functional generic kernel worker thread that any kernel thread can schedule jobs for. Interrupt handlers run with interrupts off (hence they cannot call kthread_ functions as it may deadlock the system if another thread holds the lock) therefore they cannot use the standard kernel worker threads. Instead, they use a special purpose interrupt worker thread that works much like the generic one expect that interrupt handlers can safely queue work with interrupts off. Note that this also means that interrupt handlers cannot allocate memory or print to the kernel log/screen as such mechanisms uses locks. I'll introduce a lock free algorithm for such cases later on. The boot process has also changed. The original kernel init thread in kernel.cpp creates a new bootstrap thread and becomes the system idle thread. Note that pid=0 now means the kernel, as there is no longer a system idle process. The bootstrap thread launches all the kernel worker threads and then creates a new process and loads /bin/init into it and then creates a thread in pid=1, which starts the system. The bootstrap thread then quietly waits for pid=1 to exit after which it shuts down/reboots/panics the system. In general, the introduction of race conditions and dead locks have forced me to revise a lot of the design and make sure it was thread secure. Since early parts of the kernel was quite hacky, I had to refactor such code. So it seems that the risk of dead locks forces me to write better code. Note that a real preemptive multithreaded kernel simplifies the construction of blocking system calls. My hope is that this will trigger a clean up of the filesystem code that current is almost beyond repair. Almost all of the kernel was modified during this refactoring. To the extent possible, these changes have been backported to older non-multithreaded kernel, but many changes were tightly coupled and went into this commit. Of interest is the implementation of the kthread_ api based on the design of pthreads; this library allows easy synchronization mechanisms and includes C++-style scoped locks. This commit also introduces new worker threads and tested mechanisms for interrupt handlers to schedule work in a kernel worker thread. A lot of code have been rewritten from scratch and has become a lot more stable and correct. Share and enjoy!
2012-08-01 15:30:34 +00:00
}
}
bool kthread_cond_wait_signal(kthread_cond_t* cond, kthread_mutex_t* mutex)
{
Multithreaded kernel and improvement of signal handling. Pardon the big ass-commit, this took months to develop and debug and the refactoring got so far that a clean merge became impossible. The good news is that this commit does quite a bit of cleaning up and generally improves the kernel quality. This makes the kernel fully pre-emptive and multithreaded. This was done by rewriting the interrupt code, the scheduler, introducing new threading primitives, and rewriting large parts of the kernel. During the past few commits the kernel has had its device drivers thread secured; this commit thread secures large parts of the core kernel. There still remains some parts of the kernel that is _not_ thread secured, but this is not a problem at this point. Each user-space thread has an associated kernel stack that it uses when it goes into kernel mode. This stack is by default 8 KiB since that value works for me and is also used by Linux. Strange things tends to happen on x86 in case of a stack overflow - there is no ideal way to catch such a situation right now. The system call conventions were changed, too. The %edx register is now used to provide the errno value of the call, instead of the kernel writing it into a registered global variable. The system call code has also been updated to better reflect the native calling conventions: not all registers have to be preserved. This makes system calls faster and simplifies the assembly. In the kernel, there is no longer the event.h header or the hacky method of 'resuming system calls' that closely resembles cooperative multitasking. If a system call wants to block, it should just block. The signal handling was also improved significantly. At this point, signals cannot interrupt kernel threads (but can always interrupt user-space threads if enabled), which introduces some problems with how a SIGINT could interrupt a blocking read, for instance. This commit introduces and uses a number of new primitives such as kthread_lock_mutex_signal() that attempts to get the lock but fails if a signal is pending. In this manner, the kernel is safer as kernel threads cannot be shut down inconveniently, but in return for complexity as blocking operations must check they if they should fail. Process exiting has also been refactored significantly. The _exit(2) system call sets the exit code and sends SIGKILL to all the threads in the process. Once all the threads have cleaned themselves up and exited, a worker thread calls the process's LastPrayer() method that unmaps memory, deletes the address space, notifies the parent, etc. This provides a very robust way to terminate processes as even half-constructed processes (during a failing fork for instance) can be gracefully terminated. I have introduced a number of kernel threads to help avoid threading problems and simplify kernel design. For instance, there is now a functional generic kernel worker thread that any kernel thread can schedule jobs for. Interrupt handlers run with interrupts off (hence they cannot call kthread_ functions as it may deadlock the system if another thread holds the lock) therefore they cannot use the standard kernel worker threads. Instead, they use a special purpose interrupt worker thread that works much like the generic one expect that interrupt handlers can safely queue work with interrupts off. Note that this also means that interrupt handlers cannot allocate memory or print to the kernel log/screen as such mechanisms uses locks. I'll introduce a lock free algorithm for such cases later on. The boot process has also changed. The original kernel init thread in kernel.cpp creates a new bootstrap thread and becomes the system idle thread. Note that pid=0 now means the kernel, as there is no longer a system idle process. The bootstrap thread launches all the kernel worker threads and then creates a new process and loads /bin/init into it and then creates a thread in pid=1, which starts the system. The bootstrap thread then quietly waits for pid=1 to exit after which it shuts down/reboots/panics the system. In general, the introduction of race conditions and dead locks have forced me to revise a lot of the design and make sure it was thread secure. Since early parts of the kernel was quite hacky, I had to refactor such code. So it seems that the risk of dead locks forces me to write better code. Note that a real preemptive multithreaded kernel simplifies the construction of blocking system calls. My hope is that this will trigger a clean up of the filesystem code that current is almost beyond repair. Almost all of the kernel was modified during this refactoring. To the extent possible, these changes have been backported to older non-multithreaded kernel, but many changes were tightly coupled and went into this commit. Of interest is the implementation of the kthread_ api based on the design of pthreads; this library allows easy synchronization mechanisms and includes C++-style scoped locks. This commit also introduces new worker threads and tested mechanisms for interrupt handlers to schedule work in a kernel worker thread. A lot of code have been rewritten from scratch and has become a lot more stable and correct. Share and enjoy!
2012-08-01 15:30:34 +00:00
if ( Signal::IsPending() )
return false;
Multithreaded kernel and improvement of signal handling. Pardon the big ass-commit, this took months to develop and debug and the refactoring got so far that a clean merge became impossible. The good news is that this commit does quite a bit of cleaning up and generally improves the kernel quality. This makes the kernel fully pre-emptive and multithreaded. This was done by rewriting the interrupt code, the scheduler, introducing new threading primitives, and rewriting large parts of the kernel. During the past few commits the kernel has had its device drivers thread secured; this commit thread secures large parts of the core kernel. There still remains some parts of the kernel that is _not_ thread secured, but this is not a problem at this point. Each user-space thread has an associated kernel stack that it uses when it goes into kernel mode. This stack is by default 8 KiB since that value works for me and is also used by Linux. Strange things tends to happen on x86 in case of a stack overflow - there is no ideal way to catch such a situation right now. The system call conventions were changed, too. The %edx register is now used to provide the errno value of the call, instead of the kernel writing it into a registered global variable. The system call code has also been updated to better reflect the native calling conventions: not all registers have to be preserved. This makes system calls faster and simplifies the assembly. In the kernel, there is no longer the event.h header or the hacky method of 'resuming system calls' that closely resembles cooperative multitasking. If a system call wants to block, it should just block. The signal handling was also improved significantly. At this point, signals cannot interrupt kernel threads (but can always interrupt user-space threads if enabled), which introduces some problems with how a SIGINT could interrupt a blocking read, for instance. This commit introduces and uses a number of new primitives such as kthread_lock_mutex_signal() that attempts to get the lock but fails if a signal is pending. In this manner, the kernel is safer as kernel threads cannot be shut down inconveniently, but in return for complexity as blocking operations must check they if they should fail. Process exiting has also been refactored significantly. The _exit(2) system call sets the exit code and sends SIGKILL to all the threads in the process. Once all the threads have cleaned themselves up and exited, a worker thread calls the process's LastPrayer() method that unmaps memory, deletes the address space, notifies the parent, etc. This provides a very robust way to terminate processes as even half-constructed processes (during a failing fork for instance) can be gracefully terminated. I have introduced a number of kernel threads to help avoid threading problems and simplify kernel design. For instance, there is now a functional generic kernel worker thread that any kernel thread can schedule jobs for. Interrupt handlers run with interrupts off (hence they cannot call kthread_ functions as it may deadlock the system if another thread holds the lock) therefore they cannot use the standard kernel worker threads. Instead, they use a special purpose interrupt worker thread that works much like the generic one expect that interrupt handlers can safely queue work with interrupts off. Note that this also means that interrupt handlers cannot allocate memory or print to the kernel log/screen as such mechanisms uses locks. I'll introduce a lock free algorithm for such cases later on. The boot process has also changed. The original kernel init thread in kernel.cpp creates a new bootstrap thread and becomes the system idle thread. Note that pid=0 now means the kernel, as there is no longer a system idle process. The bootstrap thread launches all the kernel worker threads and then creates a new process and loads /bin/init into it and then creates a thread in pid=1, which starts the system. The bootstrap thread then quietly waits for pid=1 to exit after which it shuts down/reboots/panics the system. In general, the introduction of race conditions and dead locks have forced me to revise a lot of the design and make sure it was thread secure. Since early parts of the kernel was quite hacky, I had to refactor such code. So it seems that the risk of dead locks forces me to write better code. Note that a real preemptive multithreaded kernel simplifies the construction of blocking system calls. My hope is that this will trigger a clean up of the filesystem code that current is almost beyond repair. Almost all of the kernel was modified during this refactoring. To the extent possible, these changes have been backported to older non-multithreaded kernel, but many changes were tightly coupled and went into this commit. Of interest is the implementation of the kthread_ api based on the design of pthreads; this library allows easy synchronization mechanisms and includes C++-style scoped locks. This commit also introduces new worker threads and tested mechanisms for interrupt handlers to schedule work in a kernel worker thread. A lot of code have been rewritten from scratch and has become a lot more stable and correct. Share and enjoy!
2012-08-01 15:30:34 +00:00
kthread_cond_elem_t elem;
elem.next = NULL;
elem.prev = cond->last;
Multithreaded kernel and improvement of signal handling. Pardon the big ass-commit, this took months to develop and debug and the refactoring got so far that a clean merge became impossible. The good news is that this commit does quite a bit of cleaning up and generally improves the kernel quality. This makes the kernel fully pre-emptive and multithreaded. This was done by rewriting the interrupt code, the scheduler, introducing new threading primitives, and rewriting large parts of the kernel. During the past few commits the kernel has had its device drivers thread secured; this commit thread secures large parts of the core kernel. There still remains some parts of the kernel that is _not_ thread secured, but this is not a problem at this point. Each user-space thread has an associated kernel stack that it uses when it goes into kernel mode. This stack is by default 8 KiB since that value works for me and is also used by Linux. Strange things tends to happen on x86 in case of a stack overflow - there is no ideal way to catch such a situation right now. The system call conventions were changed, too. The %edx register is now used to provide the errno value of the call, instead of the kernel writing it into a registered global variable. The system call code has also been updated to better reflect the native calling conventions: not all registers have to be preserved. This makes system calls faster and simplifies the assembly. In the kernel, there is no longer the event.h header or the hacky method of 'resuming system calls' that closely resembles cooperative multitasking. If a system call wants to block, it should just block. The signal handling was also improved significantly. At this point, signals cannot interrupt kernel threads (but can always interrupt user-space threads if enabled), which introduces some problems with how a SIGINT could interrupt a blocking read, for instance. This commit introduces and uses a number of new primitives such as kthread_lock_mutex_signal() that attempts to get the lock but fails if a signal is pending. In this manner, the kernel is safer as kernel threads cannot be shut down inconveniently, but in return for complexity as blocking operations must check they if they should fail. Process exiting has also been refactored significantly. The _exit(2) system call sets the exit code and sends SIGKILL to all the threads in the process. Once all the threads have cleaned themselves up and exited, a worker thread calls the process's LastPrayer() method that unmaps memory, deletes the address space, notifies the parent, etc. This provides a very robust way to terminate processes as even half-constructed processes (during a failing fork for instance) can be gracefully terminated. I have introduced a number of kernel threads to help avoid threading problems and simplify kernel design. For instance, there is now a functional generic kernel worker thread that any kernel thread can schedule jobs for. Interrupt handlers run with interrupts off (hence they cannot call kthread_ functions as it may deadlock the system if another thread holds the lock) therefore they cannot use the standard kernel worker threads. Instead, they use a special purpose interrupt worker thread that works much like the generic one expect that interrupt handlers can safely queue work with interrupts off. Note that this also means that interrupt handlers cannot allocate memory or print to the kernel log/screen as such mechanisms uses locks. I'll introduce a lock free algorithm for such cases later on. The boot process has also changed. The original kernel init thread in kernel.cpp creates a new bootstrap thread and becomes the system idle thread. Note that pid=0 now means the kernel, as there is no longer a system idle process. The bootstrap thread launches all the kernel worker threads and then creates a new process and loads /bin/init into it and then creates a thread in pid=1, which starts the system. The bootstrap thread then quietly waits for pid=1 to exit after which it shuts down/reboots/panics the system. In general, the introduction of race conditions and dead locks have forced me to revise a lot of the design and make sure it was thread secure. Since early parts of the kernel was quite hacky, I had to refactor such code. So it seems that the risk of dead locks forces me to write better code. Note that a real preemptive multithreaded kernel simplifies the construction of blocking system calls. My hope is that this will trigger a clean up of the filesystem code that current is almost beyond repair. Almost all of the kernel was modified during this refactoring. To the extent possible, these changes have been backported to older non-multithreaded kernel, but many changes were tightly coupled and went into this commit. Of interest is the implementation of the kthread_ api based on the design of pthreads; this library allows easy synchronization mechanisms and includes C++-style scoped locks. This commit also introduces new worker threads and tested mechanisms for interrupt handlers to schedule work in a kernel worker thread. A lot of code have been rewritten from scratch and has become a lot more stable and correct. Share and enjoy!
2012-08-01 15:30:34 +00:00
elem.woken = 0;
if ( cond->last )
cond->last->next = &elem;
if ( !cond->first )
cond->first = &elem;
2013-07-11 16:05:02 +00:00
cond->last = &elem;
kthread_mutex_unlock(mutex);
bool result = true;
while ( !__atomic_load_n(&elem.woken, __ATOMIC_SEQ_CST) )
Multithreaded kernel and improvement of signal handling. Pardon the big ass-commit, this took months to develop and debug and the refactoring got so far that a clean merge became impossible. The good news is that this commit does quite a bit of cleaning up and generally improves the kernel quality. This makes the kernel fully pre-emptive and multithreaded. This was done by rewriting the interrupt code, the scheduler, introducing new threading primitives, and rewriting large parts of the kernel. During the past few commits the kernel has had its device drivers thread secured; this commit thread secures large parts of the core kernel. There still remains some parts of the kernel that is _not_ thread secured, but this is not a problem at this point. Each user-space thread has an associated kernel stack that it uses when it goes into kernel mode. This stack is by default 8 KiB since that value works for me and is also used by Linux. Strange things tends to happen on x86 in case of a stack overflow - there is no ideal way to catch such a situation right now. The system call conventions were changed, too. The %edx register is now used to provide the errno value of the call, instead of the kernel writing it into a registered global variable. The system call code has also been updated to better reflect the native calling conventions: not all registers have to be preserved. This makes system calls faster and simplifies the assembly. In the kernel, there is no longer the event.h header or the hacky method of 'resuming system calls' that closely resembles cooperative multitasking. If a system call wants to block, it should just block. The signal handling was also improved significantly. At this point, signals cannot interrupt kernel threads (but can always interrupt user-space threads if enabled), which introduces some problems with how a SIGINT could interrupt a blocking read, for instance. This commit introduces and uses a number of new primitives such as kthread_lock_mutex_signal() that attempts to get the lock but fails if a signal is pending. In this manner, the kernel is safer as kernel threads cannot be shut down inconveniently, but in return for complexity as blocking operations must check they if they should fail. Process exiting has also been refactored significantly. The _exit(2) system call sets the exit code and sends SIGKILL to all the threads in the process. Once all the threads have cleaned themselves up and exited, a worker thread calls the process's LastPrayer() method that unmaps memory, deletes the address space, notifies the parent, etc. This provides a very robust way to terminate processes as even half-constructed processes (during a failing fork for instance) can be gracefully terminated. I have introduced a number of kernel threads to help avoid threading problems and simplify kernel design. For instance, there is now a functional generic kernel worker thread that any kernel thread can schedule jobs for. Interrupt handlers run with interrupts off (hence they cannot call kthread_ functions as it may deadlock the system if another thread holds the lock) therefore they cannot use the standard kernel worker threads. Instead, they use a special purpose interrupt worker thread that works much like the generic one expect that interrupt handlers can safely queue work with interrupts off. Note that this also means that interrupt handlers cannot allocate memory or print to the kernel log/screen as such mechanisms uses locks. I'll introduce a lock free algorithm for such cases later on. The boot process has also changed. The original kernel init thread in kernel.cpp creates a new bootstrap thread and becomes the system idle thread. Note that pid=0 now means the kernel, as there is no longer a system idle process. The bootstrap thread launches all the kernel worker threads and then creates a new process and loads /bin/init into it and then creates a thread in pid=1, which starts the system. The bootstrap thread then quietly waits for pid=1 to exit after which it shuts down/reboots/panics the system. In general, the introduction of race conditions and dead locks have forced me to revise a lot of the design and make sure it was thread secure. Since early parts of the kernel was quite hacky, I had to refactor such code. So it seems that the risk of dead locks forces me to write better code. Note that a real preemptive multithreaded kernel simplifies the construction of blocking system calls. My hope is that this will trigger a clean up of the filesystem code that current is almost beyond repair. Almost all of the kernel was modified during this refactoring. To the extent possible, these changes have been backported to older non-multithreaded kernel, but many changes were tightly coupled and went into this commit. Of interest is the implementation of the kthread_ api based on the design of pthreads; this library allows easy synchronization mechanisms and includes C++-style scoped locks. This commit also introduces new worker threads and tested mechanisms for interrupt handlers to schedule work in a kernel worker thread. A lot of code have been rewritten from scratch and has become a lot more stable and correct. Share and enjoy!
2012-08-01 15:30:34 +00:00
{
if ( !kutex_wait(&elem.woken, 0, true) )
Multithreaded kernel and improvement of signal handling. Pardon the big ass-commit, this took months to develop and debug and the refactoring got so far that a clean merge became impossible. The good news is that this commit does quite a bit of cleaning up and generally improves the kernel quality. This makes the kernel fully pre-emptive and multithreaded. This was done by rewriting the interrupt code, the scheduler, introducing new threading primitives, and rewriting large parts of the kernel. During the past few commits the kernel has had its device drivers thread secured; this commit thread secures large parts of the core kernel. There still remains some parts of the kernel that is _not_ thread secured, but this is not a problem at this point. Each user-space thread has an associated kernel stack that it uses when it goes into kernel mode. This stack is by default 8 KiB since that value works for me and is also used by Linux. Strange things tends to happen on x86 in case of a stack overflow - there is no ideal way to catch such a situation right now. The system call conventions were changed, too. The %edx register is now used to provide the errno value of the call, instead of the kernel writing it into a registered global variable. The system call code has also been updated to better reflect the native calling conventions: not all registers have to be preserved. This makes system calls faster and simplifies the assembly. In the kernel, there is no longer the event.h header or the hacky method of 'resuming system calls' that closely resembles cooperative multitasking. If a system call wants to block, it should just block. The signal handling was also improved significantly. At this point, signals cannot interrupt kernel threads (but can always interrupt user-space threads if enabled), which introduces some problems with how a SIGINT could interrupt a blocking read, for instance. This commit introduces and uses a number of new primitives such as kthread_lock_mutex_signal() that attempts to get the lock but fails if a signal is pending. In this manner, the kernel is safer as kernel threads cannot be shut down inconveniently, but in return for complexity as blocking operations must check they if they should fail. Process exiting has also been refactored significantly. The _exit(2) system call sets the exit code and sends SIGKILL to all the threads in the process. Once all the threads have cleaned themselves up and exited, a worker thread calls the process's LastPrayer() method that unmaps memory, deletes the address space, notifies the parent, etc. This provides a very robust way to terminate processes as even half-constructed processes (during a failing fork for instance) can be gracefully terminated. I have introduced a number of kernel threads to help avoid threading problems and simplify kernel design. For instance, there is now a functional generic kernel worker thread that any kernel thread can schedule jobs for. Interrupt handlers run with interrupts off (hence they cannot call kthread_ functions as it may deadlock the system if another thread holds the lock) therefore they cannot use the standard kernel worker threads. Instead, they use a special purpose interrupt worker thread that works much like the generic one expect that interrupt handlers can safely queue work with interrupts off. Note that this also means that interrupt handlers cannot allocate memory or print to the kernel log/screen as such mechanisms uses locks. I'll introduce a lock free algorithm for such cases later on. The boot process has also changed. The original kernel init thread in kernel.cpp creates a new bootstrap thread and becomes the system idle thread. Note that pid=0 now means the kernel, as there is no longer a system idle process. The bootstrap thread launches all the kernel worker threads and then creates a new process and loads /bin/init into it and then creates a thread in pid=1, which starts the system. The bootstrap thread then quietly waits for pid=1 to exit after which it shuts down/reboots/panics the system. In general, the introduction of race conditions and dead locks have forced me to revise a lot of the design and make sure it was thread secure. Since early parts of the kernel was quite hacky, I had to refactor such code. So it seems that the risk of dead locks forces me to write better code. Note that a real preemptive multithreaded kernel simplifies the construction of blocking system calls. My hope is that this will trigger a clean up of the filesystem code that current is almost beyond repair. Almost all of the kernel was modified during this refactoring. To the extent possible, these changes have been backported to older non-multithreaded kernel, but many changes were tightly coupled and went into this commit. Of interest is the implementation of the kthread_ api based on the design of pthreads; this library allows easy synchronization mechanisms and includes C++-style scoped locks. This commit also introduces new worker threads and tested mechanisms for interrupt handlers to schedule work in a kernel worker thread. A lot of code have been rewritten from scratch and has become a lot more stable and correct. Share and enjoy!
2012-08-01 15:30:34 +00:00
{
result = false;
break;
Multithreaded kernel and improvement of signal handling. Pardon the big ass-commit, this took months to develop and debug and the refactoring got so far that a clean merge became impossible. The good news is that this commit does quite a bit of cleaning up and generally improves the kernel quality. This makes the kernel fully pre-emptive and multithreaded. This was done by rewriting the interrupt code, the scheduler, introducing new threading primitives, and rewriting large parts of the kernel. During the past few commits the kernel has had its device drivers thread secured; this commit thread secures large parts of the core kernel. There still remains some parts of the kernel that is _not_ thread secured, but this is not a problem at this point. Each user-space thread has an associated kernel stack that it uses when it goes into kernel mode. This stack is by default 8 KiB since that value works for me and is also used by Linux. Strange things tends to happen on x86 in case of a stack overflow - there is no ideal way to catch such a situation right now. The system call conventions were changed, too. The %edx register is now used to provide the errno value of the call, instead of the kernel writing it into a registered global variable. The system call code has also been updated to better reflect the native calling conventions: not all registers have to be preserved. This makes system calls faster and simplifies the assembly. In the kernel, there is no longer the event.h header or the hacky method of 'resuming system calls' that closely resembles cooperative multitasking. If a system call wants to block, it should just block. The signal handling was also improved significantly. At this point, signals cannot interrupt kernel threads (but can always interrupt user-space threads if enabled), which introduces some problems with how a SIGINT could interrupt a blocking read, for instance. This commit introduces and uses a number of new primitives such as kthread_lock_mutex_signal() that attempts to get the lock but fails if a signal is pending. In this manner, the kernel is safer as kernel threads cannot be shut down inconveniently, but in return for complexity as blocking operations must check they if they should fail. Process exiting has also been refactored significantly. The _exit(2) system call sets the exit code and sends SIGKILL to all the threads in the process. Once all the threads have cleaned themselves up and exited, a worker thread calls the process's LastPrayer() method that unmaps memory, deletes the address space, notifies the parent, etc. This provides a very robust way to terminate processes as even half-constructed processes (during a failing fork for instance) can be gracefully terminated. I have introduced a number of kernel threads to help avoid threading problems and simplify kernel design. For instance, there is now a functional generic kernel worker thread that any kernel thread can schedule jobs for. Interrupt handlers run with interrupts off (hence they cannot call kthread_ functions as it may deadlock the system if another thread holds the lock) therefore they cannot use the standard kernel worker threads. Instead, they use a special purpose interrupt worker thread that works much like the generic one expect that interrupt handlers can safely queue work with interrupts off. Note that this also means that interrupt handlers cannot allocate memory or print to the kernel log/screen as such mechanisms uses locks. I'll introduce a lock free algorithm for such cases later on. The boot process has also changed. The original kernel init thread in kernel.cpp creates a new bootstrap thread and becomes the system idle thread. Note that pid=0 now means the kernel, as there is no longer a system idle process. The bootstrap thread launches all the kernel worker threads and then creates a new process and loads /bin/init into it and then creates a thread in pid=1, which starts the system. The bootstrap thread then quietly waits for pid=1 to exit after which it shuts down/reboots/panics the system. In general, the introduction of race conditions and dead locks have forced me to revise a lot of the design and make sure it was thread secure. Since early parts of the kernel was quite hacky, I had to refactor such code. So it seems that the risk of dead locks forces me to write better code. Note that a real preemptive multithreaded kernel simplifies the construction of blocking system calls. My hope is that this will trigger a clean up of the filesystem code that current is almost beyond repair. Almost all of the kernel was modified during this refactoring. To the extent possible, these changes have been backported to older non-multithreaded kernel, but many changes were tightly coupled and went into this commit. Of interest is the implementation of the kthread_ api based on the design of pthreads; this library allows easy synchronization mechanisms and includes C++-style scoped locks. This commit also introduces new worker threads and tested mechanisms for interrupt handlers to schedule work in a kernel worker thread. A lot of code have been rewritten from scratch and has become a lot more stable and correct. Share and enjoy!
2012-08-01 15:30:34 +00:00
}
}
kthread_mutex_lock(mutex);
if ( !__atomic_load_n(&elem.woken, __ATOMIC_SEQ_CST) )
{
if ( elem.next )
elem.next->prev = elem.prev;
else
cond->last = elem.prev;
if ( elem.prev )
elem.prev->next = elem.next;
else
cond->first = elem.next;
}
return result;
}
void kthread_cond_signal(kthread_cond_t* cond)
{
if ( cond->first )
{
struct kthread_cond_elem* elem = cond->first;
if ( elem->next )
elem->next->prev = elem->prev;
else
cond->last = elem->prev;
cond->first = elem->next;
elem->next = NULL;
elem->prev = NULL;
__atomic_store_n(&elem->woken, 1, __ATOMIC_SEQ_CST);
kutex_wake(&elem->woken, 1);
}
}
void kthread_cond_broadcast(kthread_cond_t* cond)
{
while ( cond->first )
kthread_cond_signal(cond);
}
} // namespace Sortix