#include "types.h" #include "Task.h" #include "kmalloc.h" #include "VGA.h" #include "StdLib.h" #include "i386.h" #include "system.h" #include #include #include #include "MemoryManager.h" #include "errno.h" #include "i8253.h" #include "RTC.h" #include "ProcFileSystem.h" #include //#define DEBUG_IO //#define TASK_DEBUG //#define SCHEDULER_DEBUG #define VALIDATE_USER_BUFFER(b, s) \ do { \ LinearAddress laddr((dword)(b)); \ if (!isValidAddressForUser(laddr) || !isValidAddressForUser(laddr.offset((s) - 1))) \ return -EFAULT; \ } while(0) static const DWORD defaultStackSize = 16384; Task* current; Task* s_kernelTask; static pid_t next_pid; static InlineLinkedList* s_tasks; static InlineLinkedList* s_deadTasks; static String* s_hostname; static String& hostnameStorage(InterruptDisabler&) { ASSERT(s_hostname); return *s_hostname; } static String getHostname() { InterruptDisabler disabler; return hostnameStorage(disabler).isolatedCopy(); } static bool contextSwitch(Task*); static void redoKernelTaskTSS() { if (!s_kernelTask->selector()) s_kernelTask->setSelector(allocateGDTEntry()); auto& tssDescriptor = getGDTEntry(s_kernelTask->selector()); tssDescriptor.setBase(&s_kernelTask->tss()); tssDescriptor.setLimit(0xffff); tssDescriptor.dpl = 0; tssDescriptor.segment_present = 1; tssDescriptor.granularity = 1; tssDescriptor.zero = 0; tssDescriptor.operation_size = 1; tssDescriptor.descriptor_type = 0; tssDescriptor.type = 9; flushGDT(); } void Task::prepForIRETToNewTask() { redoKernelTaskTSS(); s_kernelTask->tss().backlink = current->selector(); loadTaskRegister(s_kernelTask->selector()); } void Task::initialize() { current = nullptr; next_pid = 0; s_tasks = new InlineLinkedList; s_deadTasks = new InlineLinkedList; s_kernelTask = Task::createKernelTask(nullptr, "colonel"); s_hostname = new String("birx"); redoKernelTaskTSS(); loadTaskRegister(s_kernelTask->selector()); } #ifdef TASK_SANITY_CHECKS void Task::checkSanity(const char* msg) { char ch = current->name()[0]; kprintf("<%p> %s{%u}%b [%d] :%b: sanity check <%s>\n", current->name().characters(), current->name().characters(), current->name().length(), current->name()[current->name().length() - 1], current->pid(), ch, msg ? msg : ""); ASSERT((ch >= 'a' && ch <= 'z') || (ch >= 'A' && ch <= 'Z')); } #endif void Task::allocateLDT() { ASSERT(!m_tss.ldt); static const WORD numLDTEntries = 4; WORD newLDTSelector = allocateGDTEntry(); m_ldtEntries = new Descriptor[numLDTEntries]; #if 0 kprintf("new ldt selector = %x\n", newLDTSelector); kprintf("new ldt table at = %p\n", m_ldtEntries); kprintf("new ldt table size = %u\n", (numLDTEntries * 8) - 1); #endif Descriptor& ldt = getGDTEntry(newLDTSelector); ldt.setBase(m_ldtEntries); ldt.setLimit(numLDTEntries * 8 - 1); ldt.dpl = 0; ldt.segment_present = 1; ldt.granularity = 0; ldt.zero = 0; ldt.operation_size = 1; ldt.descriptor_type = 0; ldt.type = Descriptor::LDT; m_tss.ldt = newLDTSelector; } Vector Task::allTasks() { InterruptDisabler disabler; Vector tasks; tasks.ensureCapacity(s_tasks->sizeSlow()); for (auto* task = s_tasks->head(); task; task = task->next()) tasks.append(task); return tasks; } Task::Region* Task::allocateRegion(size_t size, String&& name) { // FIXME: This needs sanity checks. What if this overlaps existing regions? auto zone = MM.createZone(size); ASSERT(zone); m_regions.append(adopt(*new Region(m_nextRegion, size, move(zone), move(name)))); m_nextRegion = m_nextRegion.offset(size).offset(16384); return m_regions.last().ptr(); } bool Task::deallocateRegion(Region& region) { InterruptDisabler disabler; for (size_t i = 0; i < m_regions.size(); ++i) { if (m_regions[i].ptr() == ®ion) { MM.unmapRegion(*this, region); m_regions.remove(i); return true; } } return false; } Task::Region* Task::regionFromRange(LinearAddress laddr, size_t size) { for (auto& region : m_regions) { if (region->linearAddress == laddr && region->size == size) return region.ptr(); } return nullptr; } int Task::sys$set_mmap_name(void* addr, size_t size, const char* name) { VALIDATE_USER_BUFFER(name, strlen(name)); auto* region = regionFromRange(LinearAddress((dword)addr), size); if (!region) return -EINVAL; region->name = name; return 0; } void* Task::sys$mmap(void* addr, size_t size) { InterruptDisabler disabler; // FIXME: Implement mapping at a client-preferred address. ASSERT(addr == nullptr); auto* region = allocateRegion(size, "mmap"); if (!region) return (void*)-1; MM.mapRegion(*this, *region); return (void*)region->linearAddress.get(); } int Task::sys$munmap(void* addr, size_t size) { InterruptDisabler disabler; auto* region = regionFromRange(LinearAddress((dword)addr), size); if (!region) return -1; if (!deallocateRegion(*region)) return -1; return 0; } int Task::sys$gethostname(char* buffer, size_t size) { VALIDATE_USER_BUFFER(buffer, size); auto hostname = getHostname(); if (size < (hostname.length() + 1)) return -ENAMETOOLONG; memcpy(buffer, hostname.characters(), size); return 0; } int Task::sys$spawn(const char* path, const char** args) { if (args) { for (size_t i = 0; args[i]; ++i) { VALIDATE_USER_BUFFER(args[i], strlen(args[i])); } } int error = 0; auto* child = Task::createUserTask(path, m_uid, m_gid, m_pid, error, args, m_tty); if (child) return child->pid(); return error; } Task* Task::createUserTask(const String& path, uid_t uid, gid_t gid, pid_t parentPID, int& error, const char** args, TTY* tty) { auto parts = path.split('/'); if (parts.isEmpty()) { error = -ENOENT; return nullptr; } RetainPtr cwd; { InterruptDisabler disabler; if (auto* parentTask = Task::fromPID(parentPID)) cwd = parentTask->m_cwd.copyRef(); if (!cwd) cwd = VirtualFileSystem::the().root(); } auto handle = VirtualFileSystem::the().open(path, error, 0, cwd ? cwd->inode : InodeIdentifier()); if (!handle) return nullptr; if (!handle->metadata().mayExecute(uid, gid)) { error = -EACCES; return nullptr; } auto elfData = handle->readEntireFile(); if (!elfData) { error = -EIO; // FIXME: Get a more detailed error from VFS. return nullptr; } Vector taskArguments; if (args) { for (size_t i = 0; args[i]; ++i) { taskArguments.append(args[i]); } } else { taskArguments.append(parts.last()); } Vector taskEnvironment; taskEnvironment.append("PATH=/bin"); taskEnvironment.append("SHELL=/bin/sh"); taskEnvironment.append("TERM=console"); taskEnvironment.append("HOME=/"); InterruptDisabler disabler; // FIXME: Get rid of this, jesus christ. This "critical" section is HUGE. Task* t = new Task(parts.takeLast(), uid, gid, parentPID, Ring3, move(cwd), handle->vnode(), tty); t->m_arguments = move(taskArguments); t->m_initialEnvironment = move(taskEnvironment); ExecSpace space; Region* region = nullptr; space.hookableAlloc = [&] (const String& name, size_t size) { if (!size) return (void*)nullptr; size = ((size / 4096) + 1) * 4096; region = t->allocateRegion(size, String(name)); ASSERT(region); MM.mapRegion(*t, *region); return (void*)region->linearAddress.asPtr(); }; bool success = space.loadELF(move(elfData)); if (!success) { // FIXME: This is ugly. If we need to do this, it should be at a different level. MM.unmapRegionsForTask(*t); MM.mapRegionsForTask(*current); delete t; kprintf("Failure loading ELF %s\n", path.characters()); error = -ENOEXEC; return nullptr; } space.forEachArea([&] (const String& name, dword offset, size_t size, LinearAddress laddr) { if (laddr.isNull()) return; dword roundedOffset = offset & 0xfffff000; size_t roundedSize = 4096 * ceilDiv((offset - roundedOffset) + size, 4096u); LinearAddress roundedLaddr = laddr; roundedLaddr.mask(0xfffff000); t->m_subregions.append(make(*region, roundedOffset, roundedSize, roundedLaddr, String(name))); #ifdef SUBREGION_DEBUG kprintf(" req subregion %s (offset: %u, size: %u) @ %p\n", name.characters(), offset, size, laddr.get()); kprintf("actual subregion %s (offset: %u, size: %u) @ %p\n", name.characters(), roundedOffset, roundedSize, roundedLaddr.get()); #endif MM.mapSubregion(*t, *t->m_subregions.last()); }); t->m_tss.eip = (dword)space.symbolPtr("_start"); if (!t->m_tss.eip) { // FIXME: This is ugly. If we need to do this, it should be at a different level. MM.unmapRegionsForTask(*t); MM.mapRegionsForTask(*current); delete t; error = -ENOEXEC; return nullptr; } // FIXME: This is ugly. If we need to do this, it should be at a different level. MM.unmapRegionsForTask(*t); MM.mapRegionsForTask(*current); s_tasks->prepend(t); system.nprocess++; #ifdef TASK_DEBUG kprintf("Task %u (%s) spawned @ %p\n", t->pid(), t->name().characters(), t->m_tss.eip); #endif error = 0; return t; } int Task::sys$get_environment(char*** environ) { auto* region = allocateRegion(4096, "environ"); if (!region) return -ENOMEM; MM.mapRegion(*this, *region); char* envpage = (char*)region->linearAddress.get(); *environ = (char**)envpage; char* bufptr = envpage + (sizeof(char*) * (m_initialEnvironment.size() + 1)); for (size_t i = 0; i < m_initialEnvironment.size(); ++i) { (*environ)[i] = bufptr; memcpy(bufptr, m_initialEnvironment[i].characters(), m_initialEnvironment[i].length()); bufptr += m_initialEnvironment[i].length(); *(bufptr++) = '\0'; } (*environ)[m_initialEnvironment.size()] = nullptr; return 0; } int Task::sys$get_arguments(int* argc, char*** argv) { auto* region = allocateRegion(4096, "argv"); if (!region) return -ENOMEM; MM.mapRegion(*this, *region); char* argpage = (char*)region->linearAddress.get(); *argc = m_arguments.size(); *argv = (char**)argpage; char* bufptr = argpage + (sizeof(char*) * m_arguments.size()); for (size_t i = 0; i < m_arguments.size(); ++i) { (*argv)[i] = bufptr; memcpy(bufptr, m_arguments[i].characters(), m_arguments[i].length()); bufptr += m_arguments[i].length(); *(bufptr++) = '\0'; } return 0; } Task* Task::createKernelTask(void (*e)(), String&& name) { Task* task = new Task(move(name), (uid_t)0, (gid_t)0, (pid_t)0, Ring0); task->m_tss.eip = (dword)e; if (task->pid() != 0) { InterruptDisabler disabler; s_tasks->prepend(task); system.nprocess++; #ifdef TASK_DEBUG kprintf("Kernel task %u (%s) spawned @ %p\n", task->pid(), task->name().characters(), task->m_tss.eip); #endif } return task; } Task::Task(String&& name, uid_t uid, gid_t gid, pid_t parentPID, RingLevel ring, RetainPtr&& cwd, RetainPtr&& executable, TTY* tty) : m_name(move(name)) , m_pid(next_pid++) , m_uid(uid) , m_gid(gid) , m_state(Runnable) , m_ring(ring) , m_cwd(move(cwd)) , m_executable(move(executable)) , m_tty(tty) , m_parentPID(parentPID) { if (tty) { m_fileHandles.append(tty->open(O_RDONLY)); // stdin m_fileHandles.append(tty->open(O_WRONLY)); // stdout m_fileHandles.append(tty->open(O_WRONLY)); // stderr } else { m_fileHandles.append(nullptr); // stdin m_fileHandles.append(nullptr); // stdout m_fileHandles.append(nullptr); // stderr } m_nextRegion = LinearAddress(0x600000); memset(&m_tss, 0, sizeof(m_tss)); if (isRing3()) { memset(&m_ldtEntries, 0, sizeof(m_ldtEntries)); allocateLDT(); } // Only IF is set when a task boots. m_tss.eflags = 0x0202; word cs, ds, ss; if (isRing0()) { cs = 0x08; ds = 0x10; ss = 0x10; } else { cs = 0x1b; ds = 0x23; ss = 0x23; } m_tss.ds = ds; m_tss.es = ds; m_tss.fs = ds; m_tss.gs = ds; m_tss.ss = ss; m_tss.cs = cs; m_tss.cr3 = MM.pageDirectoryBase().get(); if (isRing0()) { // FIXME: This memory is leaked. // But uh, there's also no kernel task termination, so I guess it's not technically leaked... dword stackBottom = (dword)kmalloc_eternal(defaultStackSize); m_stackTop0 = (stackBottom + defaultStackSize) & 0xffffff8; m_tss.esp = m_stackTop0; } else { auto* region = allocateRegion(defaultStackSize, "stack"); ASSERT(region); m_stackTop3 = region->linearAddress.offset(defaultStackSize).get() & 0xfffffff8; m_tss.esp = m_stackTop3; } if (isRing3()) { // Ring3 tasks need a separate stack for Ring0. m_kernelStack = kmalloc(defaultStackSize); m_stackTop0 = ((DWORD)m_kernelStack + defaultStackSize) & 0xffffff8; m_tss.ss0 = 0x10; m_tss.esp0 = m_stackTop0; } // HACK: Ring2 SS in the TSS is the current PID. m_tss.ss2 = m_pid; m_farPtr.offset = 0x98765432; ProcFileSystem::the().addProcess(*this); } Task::~Task() { InterruptDisabler disabler; ProcFileSystem::the().removeProcess(*this); system.nprocess--; delete [] m_ldtEntries; m_ldtEntries = nullptr; if (m_kernelStack) { kfree(m_kernelStack); m_kernelStack = nullptr; } } void Task::dumpRegions() { kprintf("Task %s(%u) regions:\n", name().characters(), pid()); kprintf("BEGIN END SIZE NAME\n"); for (auto& region : m_regions) { kprintf("%x -- %x %x %s\n", region->linearAddress.get(), region->linearAddress.offset(region->size - 1).get(), region->size, region->name.characters()); } kprintf("Task %s(%u) subregions:\n", name().characters(), pid()); kprintf("REGION OFFSET BEGIN END SIZE NAME\n"); for (auto& subregion : m_subregions) { kprintf("%x %x %x -- %x %x %s\n", subregion->region->linearAddress.get(), subregion->offset, subregion->linearAddress.get(), subregion->linearAddress.offset(subregion->size - 1).get(), subregion->size, subregion->name.characters()); } } void Task::sys$exit(int status) { cli(); #ifdef TASK_DEBUG kprintf("sys$exit: %s(%u) exit with status %d\n", name().characters(), pid(), status); #endif setState(Exiting); MM.unmapRegionsForTask(*this); s_tasks->remove(this); for (auto* task = s_tasks->head(); task; task = task->next()) { if (task->waitee() == m_pid) task->m_waiteeStatus = status << 8; } if (!scheduleNewTask()) { kprintf("Task::sys$exit: Failed to schedule a new task :(\n"); HANG; } s_deadTasks->append(this); switchNow(); } void Task::murder() { ASSERT_INTERRUPTS_DISABLED(); bool wasCurrent = current == this; setState(Exiting); s_tasks->remove(this); if (wasCurrent) { MM.unmapRegionsForTask(*this); if (!scheduleNewTask()) { kprintf("Task::murder: Failed to schedule a new task :(\n"); HANG; } } s_deadTasks->append(this); if (wasCurrent) switchNow(); } void Task::taskDidCrash(Task* crashedTask) { ASSERT_INTERRUPTS_DISABLED(); if (crashedTask->state() == Crashing) { kprintf("Double crash :(\n"); HANG; } crashedTask->setState(Crashing); crashedTask->dumpRegions(); s_tasks->remove(crashedTask); MM.unmapRegionsForTask(*crashedTask); if (!scheduleNewTask()) { kprintf("Task::taskDidCrash: Failed to schedule a new task :(\n"); HANG; } s_deadTasks->append(crashedTask); switchNow(); } void Task::doHouseKeeping() { InterruptDisabler disabler; if (s_deadTasks->isEmpty()) return; Task* next = nullptr; for (auto* deadTask = s_deadTasks->head(); deadTask; deadTask = next) { next = deadTask->next(); delete deadTask; } s_deadTasks->clear(); } void yield() { if (!current) { kprintf( "PANIC: yield() with !current" ); HANG; } //kprintf("%s<%u> yield()\n", current->name().characters(), current->pid()); InterruptDisabler disabler; if (!scheduleNewTask()) return; //kprintf("yield() jumping to new task: %x (%s)\n", current->farPtr().selector, current->name().characters()); switchNow(); } void switchNow() { Descriptor& descriptor = getGDTEntry(current->selector()); descriptor.type = 9; flushGDT(); asm("sti\n" "ljmp *(%%eax)\n" ::"a"(¤t->farPtr()) ); } bool scheduleNewTask() { ASSERT_INTERRUPTS_DISABLED(); if (!current) { // XXX: The first ever context_switch() goes to the idle task. // This to setup a reliable place we can return to. return contextSwitch(Task::kernelTask()); } // Check and unblock tasks whose wait conditions have been met. for (auto* task = s_tasks->head(); task; task = task->next()) { if (task->state() == Task::BlockedSleep) { if (task->wakeupTime() <= system.uptime) { task->unblock(); continue; } } if (task->state() == Task::BlockedWait) { if (!Task::fromPID(task->waitee())) { task->unblock(); continue; } } if (task->state() == Task::BlockedRead) { ASSERT(task->m_fdBlockedOnRead != -1); if (task->m_fileHandles[task->m_fdBlockedOnRead]->hasDataAvailableForRead()) { task->unblock(); continue; } } } #ifdef SCHEDULER_DEBUG dbgprintf("Scheduler choices:\n"); for (auto* task = s_tasks->head(); task; task = task->next()) { //if (task->state() == Task::BlockedWait || task->state() == Task::BlockedSleep) // continue; dbgprintf("%w %s(%u)\n", task->state(), task->name().characters(), task->pid()); } #endif auto* prevHead = s_tasks->head(); for (;;) { // Move head to tail. s_tasks->append(s_tasks->removeHead()); auto* task = s_tasks->head(); if (task->state() == Task::Runnable || task->state() == Task::Running) { #ifdef SCHEDULER_DEBUG dbgprintf("switch to %s(%u) (%p vs %p)\n", task->name().characters(), task->pid(), task, current); #endif return contextSwitch(task); } if (task == prevHead) { // Back at task_head, nothing wants to run. kprintf("Nothing wants to run!\n"); kprintf("PID OWNER STATE NSCHED NAME\n"); for (auto* task = s_tasks->head(); task; task = task->next()) { kprintf("%w %w:%w %b %w %s\n", task->pid(), task->uid(), task->gid(), task->state(), task->timesScheduled(), task->name().characters()); } kprintf("Switch to kernel task\n"); return contextSwitch(Task::kernelTask()); } } } static bool contextSwitch(Task* t) { //kprintf("c_s to %s (same:%u)\n", t->name().characters(), current == t); t->setTicksLeft(5); t->didSchedule(); if (current == t) return false; // Some sanity checking to force a crash earlier. auto csRPL = t->tss().cs & 3; auto ssRPL = t->tss().ss & 3; if (csRPL != ssRPL) { kprintf("Fuckup! Switching from %s(%u) to %s(%u) has RPL mismatch\n", current->name().characters(), current->pid(), t->name().characters(), t->pid() ); kprintf("code: %w:%x\n", t->tss().cs, t->tss().eip); kprintf(" stk: %w:%x\n", t->tss().ss, t->tss().esp); ASSERT(csRPL == ssRPL); } if (current) { // If the last task hasn't blocked (still marked as running), // mark it as runnable for the next round. if (current->state() == Task::Running) current->setState(Task::Runnable); bool success = MM.unmapRegionsForTask(*current); ASSERT(success); } bool success = MM.mapRegionsForTask(*t); ASSERT(success); current = t; t->setState(Task::Running); if (!t->selector()) t->setSelector(allocateGDTEntry()); auto& tssDescriptor = getGDTEntry(t->selector()); tssDescriptor.limit_hi = 0; tssDescriptor.limit_lo = 0xFFFF; tssDescriptor.base_lo = (DWORD)(&t->tss()) & 0xFFFF; tssDescriptor.base_hi = ((DWORD)(&t->tss()) >> 16) & 0xFF; tssDescriptor.base_hi2 = ((DWORD)(&t->tss()) >> 24) & 0xFF; tssDescriptor.dpl = 0; tssDescriptor.segment_present = 1; tssDescriptor.granularity = 1; tssDescriptor.zero = 0; tssDescriptor.operation_size = 1; tssDescriptor.descriptor_type = 0; tssDescriptor.type = 11; // Busy TSS flushGDT(); return true; } Task* Task::fromPID(pid_t pid) { ASSERT_INTERRUPTS_DISABLED(); for (auto* task = s_tasks->head(); task; task = task->next()) { if (task->pid() == pid) return task; } return nullptr; } FileHandle* Task::fileHandleIfExists(int fd) { if (fd < 0) return nullptr; if ((unsigned)fd < m_fileHandles.size()) return m_fileHandles[fd].ptr(); return nullptr; } ssize_t Task::sys$get_dir_entries(int fd, void* buffer, size_t size) { VALIDATE_USER_BUFFER(buffer, size); auto* handle = fileHandleIfExists(fd); if (!handle) return -EBADF; return handle->get_dir_entries((byte*)buffer, size); } int Task::sys$lseek(int fd, off_t offset, int whence) { auto* handle = fileHandleIfExists(fd); if (!handle) return -EBADF; return handle->seek(offset, whence); } int Task::sys$ttyname_r(int fd, char* buffer, size_t size) { VALIDATE_USER_BUFFER(buffer, size); auto* handle = fileHandleIfExists(fd); if (!handle) return -EBADF; if (!handle->isTTY()) return -ENOTTY; auto ttyName = handle->tty()->ttyName(); if (size < ttyName.length() + 1) return -ERANGE; strcpy(buffer, ttyName.characters()); return 0; } ssize_t Task::sys$write(int fd, const void* data, size_t size) { VALIDATE_USER_BUFFER(data, size); #ifdef DEBUG_IO kprintf("Task::sys$write: called(%d, %p, %u)\n", fd, data, size); #endif auto* handle = fileHandleIfExists(fd); #ifdef DEBUG_IO kprintf("Task::sys$write: handle=%p\n", handle); #endif if (!handle) return -EBADF; auto nwritten = handle->write((const byte*)data, size); #ifdef DEBUG_IO kprintf("Task::sys$write: nwritten=%u\n", nwritten); #endif return nwritten; } ssize_t Task::sys$read(int fd, void* outbuf, size_t nread) { VALIDATE_USER_BUFFER(outbuf, nread); #ifdef DEBUG_IO kprintf("Task::sys$read: called(%d, %p, %u)\n", fd, outbuf, nread); #endif auto* handle = fileHandleIfExists(fd); #ifdef DEBUG_IO kprintf("Task::sys$read: handle=%p\n", handle); #endif if (!handle) return -EBADF; if (handle->isBlocking()) { if (!handle->hasDataAvailableForRead()) { m_fdBlockedOnRead = fd; block(BlockedRead); yield(); } } nread = handle->read((byte*)outbuf, nread); #ifdef DEBUG_IO kprintf("Task::sys$read: nread=%u\n", nread); #endif return nread; } int Task::sys$close(int fd) { auto* handle = fileHandleIfExists(fd); if (!handle) return -EBADF; // FIXME: Implement. return 0; } int Task::sys$lstat(const char* path, Unix::stat* statbuf) { VALIDATE_USER_BUFFER(statbuf, sizeof(Unix::stat)); int error; auto handle = VirtualFileSystem::the().open(move(path), error, O_NOFOLLOW_NOERROR, cwdInode()); if (!handle) return error; handle->stat(statbuf); return 0; } int Task::sys$stat(const char* path, Unix::stat* statbuf) { VALIDATE_USER_BUFFER(statbuf, sizeof(Unix::stat)); int error; auto handle = VirtualFileSystem::the().open(move(path), error, 0, cwdInode()); if (!handle) return error; handle->stat(statbuf); return 0; } int Task::sys$readlink(const char* path, char* buffer, size_t size) { VALIDATE_USER_BUFFER(path, strlen(path)); VALIDATE_USER_BUFFER(buffer, size); int error; auto handle = VirtualFileSystem::the().open(path, error, O_RDONLY | O_NOFOLLOW_NOERROR, cwdInode()); if (!handle) return error; if (!handle->metadata().isSymbolicLink()) return -EINVAL; auto contents = handle->readEntireFile(); if (!contents) return -EIO; // FIXME: Get a more detailed error from VFS. memcpy(buffer, contents.pointer(), min(size, contents.size())); if (contents.size() + 1 < size) buffer[contents.size()] = '\0'; return 0; } int Task::sys$chdir(const char* path) { VALIDATE_USER_BUFFER(path, strlen(path)); int error; auto handle = VirtualFileSystem::the().open(path, error, 0, cwdInode()); if (!handle) return error; if (!handle->isDirectory()) return -ENOTDIR; m_cwd = handle->vnode(); return 0; } int Task::sys$getcwd(char* buffer, size_t size) { VALIDATE_USER_BUFFER(buffer, size); auto path = VirtualFileSystem::the().absolutePath(cwdInode()); if (path.isNull()) return -EINVAL; if (size < path.length() + 1) return -ERANGE; strcpy(buffer, path.characters()); return -ENOTIMPL; } int Task::sys$open(const char* path, int options) { #ifdef DEBUG_IO kprintf("Task::sys$open(): PID=%u, path=%s {%u}\n", m_pid, path, pathLength); #endif VALIDATE_USER_BUFFER(path, strlen(path)); if (m_fileHandles.size() >= m_maxFileHandles) return -EMFILE; int error; auto handle = VirtualFileSystem::the().open(path, error, options, cwdInode()); if (!handle) return error; if (options & O_DIRECTORY && !handle->isDirectory()) return -ENOTDIR; // FIXME: This should be handled by VFS::open. int fd = m_fileHandles.size(); handle->setFD(fd); m_fileHandles.append(move(handle)); return fd; } int Task::sys$uname(utsname* buf) { VALIDATE_USER_BUFFER(buf, sizeof(utsname)); strcpy(buf->sysname, "Serenity"); strcpy(buf->release, "1.0-dev"); strcpy(buf->version, "FIXME"); strcpy(buf->machine, "i386"); strcpy(buf->nodename, getHostname().characters()); return 0; } int Task::sys$kill(pid_t pid, int sig) { (void) sig; if (pid == 0) { // FIXME: Send to same-group processes. ASSERT(pid != 0); } if (pid == -1) { // FIXME: Send to all processes. ASSERT(pid != -1); } ASSERT(pid != current->pid()); // FIXME: Support this scenario. InterruptDisabler disabler; auto* peer = Task::fromPID(pid); if (!peer) return -ESRCH; if (sig == 9) { peer->murder(); } else { ASSERT_NOT_REACHED(); } return -1; } int Task::sys$sleep(unsigned seconds) { if (!seconds) return 0; sleep(seconds * TICKS_PER_SECOND); return 0; } int Task::sys$gettimeofday(timeval* tv) { VALIDATE_USER_BUFFER(tv, sizeof(tv)); InterruptDisabler disabler; auto now = RTC::now(); tv->tv_sec = now; tv->tv_usec = 0; return 0; } uid_t Task::sys$getuid() { return m_uid; } gid_t Task::sys$getgid() { return m_gid; } pid_t Task::sys$getpid() { return m_pid; } pid_t Task::sys$waitpid(pid_t waitee, int* wstatus, int options) { if (wstatus) VALIDATE_USER_BUFFER(wstatus, sizeof(int)); InterruptDisabler disabler; if (!Task::fromPID(waitee)) return -1; m_waitee = waitee; m_waiteeStatus = 0; block(BlockedWait); yield(); if (wstatus) *wstatus = m_waiteeStatus; return m_waitee; } void Task::unblock() { ASSERT(m_state != Task::Runnable && m_state != Task::Running); system.nblocked--; m_state = Task::Runnable; } void Task::block(Task::State state) { ASSERT(current->state() == Task::Running); system.nblocked++; current->setState(state); } void block(Task::State state) { current->block(state); yield(); } void sleep(DWORD ticks) { ASSERT(current->state() == Task::Running); current->setWakeupTime(system.uptime + ticks); current->block(Task::BlockedSleep); yield(); } Task* Task::kernelTask() { ASSERT(s_kernelTask); return s_kernelTask; } Task::Region::Region(LinearAddress a, size_t s, RetainPtr&& z, String&& n) : linearAddress(a) , size(s) , zone(move(z)) , name(move(n)) { } Task::Region::~Region() { } Task::Subregion::Subregion(Region& r, dword o, size_t s, LinearAddress l, String&& n)\ : region(r) , offset(o) , size(s) , linearAddress(l) , name(move(n)) { } Task::Subregion::~Subregion() { } bool Task::isValidAddressForKernel(LinearAddress laddr) const { InterruptDisabler disabler; if (laddr.get() >= ksyms().first().address && laddr.get() <= ksyms().last().address) return true; if (is_kmalloc_address((void*)laddr.get())) return true; return isValidAddressForUser(laddr); } bool Task::isValidAddressForUser(LinearAddress laddr) const { InterruptDisabler disabler; for (auto& region: m_regions) { if (laddr >= region->linearAddress && laddr < region->linearAddress.offset(region->size)) return true; } for (auto& subregion: m_subregions) { if (laddr >= subregion->linearAddress && laddr < subregion->linearAddress.offset(subregion->size)) return true; } return false; }