/* * Copyright (c) 2018-2020, Andreas Kling * All rights reserved. * * Redistribution and use in source and binary forms, with or without * modification, are permitted provided that the following conditions are met: * * 1. Redistributions of source code must retain the above copyright notice, this * list of conditions and the following disclaimer. * * 2. Redistributions in binary form must reproduce the above copyright notice, * this list of conditions and the following disclaimer in the documentation * and/or other materials provided with the distribution. * * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE * DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR * SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER * CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, * OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. */ #include #include #include #include #include #include #include #include #include #include #include #include namespace Kernel { static SpinLock s_all_inodes_lock; static AK::Singleton> s_list; SpinLock& Inode::all_inodes_lock() { return s_all_inodes_lock; } InlineLinkedList& Inode::all_with_lock() { ASSERT(s_all_inodes_lock.is_locked()); return *s_list; } void Inode::sync() { NonnullRefPtrVector inodes; { ScopedSpinLock all_inodes_lock(s_all_inodes_lock); for (auto& inode : all_with_lock()) { if (inode.is_metadata_dirty()) inodes.append(inode); } } for (auto& inode : inodes) { ASSERT(inode.is_metadata_dirty()); inode.flush_metadata(); } } KResultOr> Inode::read_entire(FileDescription* description) const { KBufferBuilder builder; ssize_t nread; u8 buffer[4096]; off_t offset = 0; for (;;) { auto buf = UserOrKernelBuffer::for_kernel_buffer(buffer); nread = read_bytes(offset, sizeof(buffer), buf, description); if (nread < 0) return KResult(nread); ASSERT(nread <= (ssize_t)sizeof(buffer)); if (nread <= 0) break; builder.append((const char*)buffer, nread); offset += nread; if (nread < (ssize_t)sizeof(buffer)) break; } if (nread < 0) { klog() << "Inode::read_entire: ERROR: " << nread; return KResult(nread); } auto entire_file = builder.build(); if (!entire_file) return KResult(-ENOMEM); return entire_file.release_nonnull(); } KResultOr> Inode::resolve_as_link(Custody& base, RefPtr* out_parent, int options, int symlink_recursion_level) const { // The default implementation simply treats the stored // contents as a path and resolves that. That is, it // behaves exactly how you would expect a symlink to work. auto contents_or = read_entire(); if (contents_or.is_error()) return contents_or.error(); auto& contents = contents_or.value(); auto path = StringView(contents->data(), contents->size()); return VFS::the().resolve_path(path, base, out_parent, options, symlink_recursion_level); } Inode::Inode(FS& fs, unsigned index) : m_fs(fs) , m_index(index) { ScopedSpinLock all_inodes_lock(s_all_inodes_lock); all_with_lock().append(this); } Inode::~Inode() { ScopedSpinLock all_inodes_lock(s_all_inodes_lock); all_with_lock().remove(this); } void Inode::will_be_destroyed() { LOCKER(m_lock); if (m_metadata_dirty) flush_metadata(); } void Inode::inode_contents_changed(off_t offset, ssize_t size, const UserOrKernelBuffer& data) { LOCKER(m_lock); if (auto shared_vmobject = this->shared_vmobject()) shared_vmobject->inode_contents_changed({}, offset, size, data); } void Inode::inode_size_changed(size_t old_size, size_t new_size) { LOCKER(m_lock); if (auto shared_vmobject = this->shared_vmobject()) shared_vmobject->inode_size_changed({}, old_size, new_size); } int Inode::set_atime(time_t) { return -ENOTIMPL; } int Inode::set_ctime(time_t) { return -ENOTIMPL; } int Inode::set_mtime(time_t) { return -ENOTIMPL; } KResult Inode::increment_link_count() { return KResult(-ENOTIMPL); } KResult Inode::decrement_link_count() { return KResult(-ENOTIMPL); } void Inode::set_shared_vmobject(SharedInodeVMObject& vmobject) { LOCKER(m_lock); m_shared_vmobject = vmobject; } bool Inode::bind_socket(LocalSocket& socket) { LOCKER(m_lock); if (m_socket) return false; m_socket = socket; return true; } bool Inode::unbind_socket() { LOCKER(m_lock); if (!m_socket) return false; m_socket = nullptr; return true; } void Inode::register_watcher(Badge, InodeWatcher& watcher) { LOCKER(m_lock); ASSERT(!m_watchers.contains(&watcher)); m_watchers.set(&watcher); } void Inode::unregister_watcher(Badge, InodeWatcher& watcher) { LOCKER(m_lock); ASSERT(m_watchers.contains(&watcher)); m_watchers.remove(&watcher); } NonnullRefPtr Inode::fifo() { LOCKER(m_lock); ASSERT(metadata().is_fifo()); // FIXME: Release m_fifo when it is closed by all readers and writers if (!m_fifo) m_fifo = FIFO::create(metadata().uid); ASSERT(m_fifo); return *m_fifo; } void Inode::set_metadata_dirty(bool metadata_dirty) { LOCKER(m_lock); if (m_metadata_dirty == metadata_dirty) return; m_metadata_dirty = metadata_dirty; if (m_metadata_dirty) { // FIXME: Maybe we should hook into modification events somewhere else, I'm not sure where. // We don't always end up on this particular code path, for instance when writing to an ext2fs file. for (auto& watcher : m_watchers) { watcher->notify_inode_event({}, InodeWatcherEvent::Type::Modified); } } } void Inode::did_add_child(const InodeIdentifier& child_id) { LOCKER(m_lock); for (auto& watcher : m_watchers) { watcher->notify_child_added({}, child_id); } } void Inode::did_remove_child(const InodeIdentifier& child_id) { LOCKER(m_lock); for (auto& watcher : m_watchers) { watcher->notify_child_removed({}, child_id); } } KResult Inode::prepare_to_write_data() { // FIXME: It's a poor design that filesystems are expected to call this before writing out data. // We should funnel everything through an interface at the VFS layer so this can happen from a single place. LOCKER(m_lock); if (fs().is_readonly()) return KResult(-EROFS); auto metadata = this->metadata(); if (metadata.is_setuid() || metadata.is_setgid()) { dbg() << "Inode::prepare_to_write_data(): Stripping SUID/SGID bits from " << identifier(); return chmod(metadata.mode & ~(04000 | 02000)); } return KSuccess; } RefPtr Inode::shared_vmobject() const { LOCKER(m_lock); return m_shared_vmobject.strong_ref(); } bool Inode::is_shared_vmobject(const SharedInodeVMObject& other) const { LOCKER(m_lock); return m_shared_vmobject.unsafe_ptr() == &other; } }