kernel/fs/fd.cpp

/*
 * Copyright 2009-2011, Ingo Weinhold, ingo_weinhold@gmx.de.
 * Copyright 2002-2018, Axel Dörfler, axeld@pinc-software.de.
 * Distributed under the terms of the MIT License.
 */


//! Operations on file descriptors


#include <fd.h>

#include <stdlib.h>
#include <string.h>

#include <OS.h>

#include <AutoDeleter.h>
#include <AutoDeleterDrivers.h>
#include <BytePointer.h>

#include <syscalls.h>
#include <syscall_restart.h>
#include <slab/Slab.h>
#include <util/AutoLock.h>
#include <util/iovec_support.h>
#include <vfs.h>
#include <wait_for_objects.h>

#include "vfs_tracing.h"


//#define TRACE_FD
#ifdef TRACE_FD
#	define TRACE(x) dprintf x
#else
#	define TRACE(x)
#endif


static const size_t kMaxReadDirBufferSize = 64 * 1024;

extern object_cache* sFileDescriptorCache;


static struct file_descriptor* get_fd_locked(struct io_context* context,
	int fd);
static struct file_descriptor* remove_fd(struct io_context* context, int fd);
static void deselect_select_infos(file_descriptor* descriptor,
	select_info* infos, bool putSyncObjects);


struct FDGetterLocking {
	inline bool Lock(file_descriptor* /*lockable*/)
	{
		return false;
	}

	inline void Unlock(file_descriptor* lockable)
	{
		put_fd(lockable);
	}
};

class FDGetter : public AutoLocker<file_descriptor, FDGetterLocking> {
public:
	inline FDGetter()
		: AutoLocker<file_descriptor, FDGetterLocking>()
	{
	}

	inline FDGetter(io_context* context, int fd, bool contextLocked = false)
		: AutoLocker<file_descriptor, FDGetterLocking>(
			contextLocked ? get_fd_locked(context, fd) : get_fd(context, fd))
	{
	}

	inline file_descriptor* SetTo(io_context* context, int fd,
		bool contextLocked = false)
	{
		file_descriptor* descriptor
			= contextLocked ? get_fd_locked(context, fd) : get_fd(context, fd);
		AutoLocker<file_descriptor, FDGetterLocking>::SetTo(descriptor, true);
		return descriptor;
	}

	inline file_descriptor* SetTo(int fd, bool kernel,
		bool contextLocked = false)
	{
		return SetTo(get_current_io_context(kernel), fd, contextLocked);
	}

	inline file_descriptor* FD() const
	{
		return fLockable;
	}
};


//	#pragma mark - General fd routines


#ifdef DEBUG
void dump_fd(int fd, struct file_descriptor* descriptor);

void
dump_fd(int fd,struct file_descriptor* descriptor)
{
	dprintf("fd[%d] = %p: type = %" B_PRId32 ", ref_count = %" B_PRId32 ", ops "
		"= %p, u.vnode = %p, u.mount = %p, cookie = %p, open_mode = %" B_PRIx32
		", pos = %" B_PRId64 "\n",
		fd, descriptor, descriptor->type, descriptor->ref_count,
		descriptor->ops, descriptor->u.vnode, descriptor->u.mount,
		descriptor->cookie, descriptor->open_mode, descriptor->pos);
}
#endif


/*! Allocates and initializes a new file_descriptor.
*/
struct file_descriptor*
alloc_fd(void)
{
	file_descriptor* descriptor
		= (file_descriptor*)object_cache_alloc(sFileDescriptorCache, 0);
	if (descriptor == NULL)
		return NULL;

	descriptor->u.vnode = NULL;
	descriptor->cookie = NULL;
	descriptor->ref_count = 1;
	descriptor->open_count = 0;
	descriptor->open_mode = 0;
	descriptor->pos = 0;

	return descriptor;
}


bool
fd_close_on_exec(struct io_context* context, int fd)
{
	return CHECK_BIT(context->fds_close_on_exec[fd / 8], fd & 7) ? true : false;
}


void
fd_set_close_on_exec(struct io_context* context, int fd, bool closeFD)
{
	if (closeFD)
		context->fds_close_on_exec[fd / 8] |= (1 << (fd & 7));
	else
		context->fds_close_on_exec[fd / 8] &= ~(1 << (fd & 7));
}


/*!	Searches a free slot in the FD table of the provided I/O context, and
	inserts the specified descriptor into it.
*/
int
new_fd_etc(struct io_context* context, struct file_descriptor* descriptor,
	int firstIndex)
{
	int fd = -1;
	uint32 i;

	if (firstIndex < 0 || (uint32)firstIndex >= context->table_size)
		return B_BAD_VALUE;

	mutex_lock(&context->io_mutex);

	for (i = firstIndex; i < context->table_size; i++) {
		if (!context->fds[i]) {
			fd = i;
			break;
		}
	}
	if (fd < 0) {
		fd = B_NO_MORE_FDS;
		goto err;
	}

	TFD(NewFD(context, fd, descriptor));

	context->fds[fd] = descriptor;
	context->num_used_fds++;
	atomic_add(&descriptor->open_count, 1);

err:
	mutex_unlock(&context->io_mutex);

	return fd;
}


int
new_fd(struct io_context* context, struct file_descriptor* descriptor)
{
	return new_fd_etc(context, descriptor, 0);
}


/*!	Reduces the descriptor's reference counter, and frees all resources
	when it's no longer used.
*/
void
put_fd(struct file_descriptor* descriptor)
{
	int32 previous = atomic_add(&descriptor->ref_count, -1);

	TFD(PutFD(descriptor));

	TRACE(("put_fd(descriptor = %p [ref = %" B_PRId32 ", cookie = %p])\n",
		descriptor, descriptor->ref_count, descriptor->cookie));

	// free the descriptor if we don't need it anymore
	if (previous == 1) {
		// free the underlying object
		if (descriptor->ops != NULL && descriptor->ops->fd_free != NULL)
			descriptor->ops->fd_free(descriptor);

		object_cache_free(sFileDescriptorCache, descriptor, 0);
	} else if ((descriptor->open_mode & O_DISCONNECTED) != 0
		&& previous - 1 == descriptor->open_count
		&& descriptor->ops != NULL) {
		// the descriptor has been disconnected - it cannot
		// be accessed anymore, let's close it (no one is
		// currently accessing this descriptor)

		if (descriptor->ops->fd_close)
			descriptor->ops->fd_close(descriptor);
		if (descriptor->ops->fd_free)
			descriptor->ops->fd_free(descriptor);

		// prevent this descriptor from being closed/freed again
		descriptor->ops = NULL;
		descriptor->u.vnode = NULL;

		// the file descriptor is kept intact, so that it's not
		// reused until someone explicitly closes it
	}
}


/*!	Decrements the open counter of the file descriptor and invokes
	its close hook when appropriate.
*/
void
close_fd(struct io_context* context, struct file_descriptor* descriptor)
{
	// POSIX advisory locks need to be released when any file descriptor closes
	if (descriptor->type == FDTYPE_FILE)
		vfs_release_posix_lock(context, descriptor);

	if (atomic_add(&descriptor->open_count, -1) == 1) {
		vfs_unlock_vnode_if_locked(descriptor);

		if (descriptor->ops != NULL && descriptor->ops->fd_close != NULL)
			descriptor->ops->fd_close(descriptor);
	}
}


status_t
close_fd_index(struct io_context* context, int fd)
{
	struct file_descriptor* descriptor = remove_fd(context, fd);

	if (descriptor == NULL)
		return B_FILE_ERROR;

	close_fd(context, descriptor);
	put_fd(descriptor);
		// the reference associated with the slot

	return B_OK;
}


/*!	This descriptor's underlying object will be closed and freed as soon as
	possible (in one of the next calls to put_fd() - get_fd() will no longer
	succeed on this descriptor).
	This is useful if the underlying object is gone, for instance when a
	(mounted) volume got removed unexpectedly.
*/
void
disconnect_fd(struct file_descriptor* descriptor)
{
	descriptor->open_mode |= O_DISCONNECTED;
}


void
inc_fd_ref_count(struct file_descriptor* descriptor)
{
	atomic_add(&descriptor->ref_count, 1);
}


static struct file_descriptor*
get_fd_locked(struct io_context* context, int fd)
{
	if (fd < 0 || (uint32)fd >= context->table_size)
		return NULL;

	struct file_descriptor* descriptor = context->fds[fd];

	if (descriptor != NULL) {
		// disconnected descriptors cannot be accessed anymore
		if (descriptor->open_mode & O_DISCONNECTED)
			return NULL;

		TFD(GetFD(context, fd, descriptor));
		inc_fd_ref_count(descriptor);
	}

	return descriptor;
}


struct file_descriptor*
get_fd(struct io_context* context, int fd)
{
	MutexLocker _(context->io_mutex);

	return get_fd_locked(context, fd);
}


struct file_descriptor*
get_open_fd(struct io_context* context, int fd)
{
	MutexLocker _(context->io_mutex);

	file_descriptor* descriptor = get_fd_locked(context, fd);
	if (descriptor == NULL)
		return NULL;

	atomic_add(&descriptor->open_count, 1);

	return descriptor;
}


/*!	Removes the file descriptor from the specified slot.
*/
static struct file_descriptor*
remove_fd(struct io_context* context, int fd)
{
	struct file_descriptor* descriptor = NULL;

	if (fd < 0)
		return NULL;

	mutex_lock(&context->io_mutex);

	if ((uint32)fd < context->table_size)
		descriptor = context->fds[fd];

	select_info* selectInfos = NULL;
	bool disconnected = false;

	if (descriptor != NULL)	{
		// fd is valid
		TFD(RemoveFD(context, fd, descriptor));

		context->fds[fd] = NULL;
		fd_set_close_on_exec(context, fd, false);
		context->num_used_fds--;

		selectInfos = context->select_infos[fd];
		context->select_infos[fd] = NULL;

		disconnected = (descriptor->open_mode & O_DISCONNECTED);
	}

	mutex_unlock(&context->io_mutex);

	if (selectInfos != NULL)
		deselect_select_infos(descriptor, selectInfos, true);

	return disconnected ? NULL : descriptor;
}


static int
dup_fd(int fd, bool kernel)
{
	struct io_context* context = get_current_io_context(kernel);
	struct file_descriptor* descriptor;
	int status;

	TRACE(("dup_fd: fd = %d\n", fd));

	// Try to get the fd structure
	descriptor = get_fd(context, fd);
	if (descriptor == NULL)
		return B_FILE_ERROR;

	// now put the fd in place
	status = new_fd(context, descriptor);
	if (status < 0)
		put_fd(descriptor);
	else {
		mutex_lock(&context->io_mutex);
		fd_set_close_on_exec(context, status, false);
		mutex_unlock(&context->io_mutex);
	}

	return status;
}


/*!	POSIX says this should be the same as:
		close(newfd);
		fcntl(oldfd, F_DUPFD, newfd);

	We do dup2() directly to be thread-safe.
*/
static int
dup2_fd(int oldfd, int newfd, bool kernel)
{
	struct file_descriptor* evicted = NULL;
	struct io_context* context;

	TRACE(("dup2_fd: ofd = %d, nfd = %d\n", oldfd, newfd));

	// quick check
	if (oldfd < 0 || newfd < 0)
		return B_FILE_ERROR;

	// Get current I/O context and lock it
	context = get_current_io_context(kernel);
	mutex_lock(&context->io_mutex);

	// Check if the fds are valid (mutex must be locked because
	// the table size could be changed)
	if ((uint32)oldfd >= context->table_size
		|| (uint32)newfd >= context->table_size
		|| context->fds[oldfd] == NULL
		|| (context->fds[oldfd]->open_mode & O_DISCONNECTED) != 0) {
		mutex_unlock(&context->io_mutex);
		return B_FILE_ERROR;
	}

	// Check for identity, note that it cannot be made above
	// because we always want to return an error on invalid
	// handles
	select_info* selectInfos = NULL;
	if (oldfd != newfd) {
		// Now do the work
		TFD(Dup2FD(context, oldfd, newfd));

		evicted = context->fds[newfd];
		selectInfos = context->select_infos[newfd];
		context->select_infos[newfd] = NULL;
		atomic_add(&context->fds[oldfd]->ref_count, 1);
		atomic_add(&context->fds[oldfd]->open_count, 1);
		context->fds[newfd] = context->fds[oldfd];

		if (evicted == NULL)
			context->num_used_fds++;
	}

	fd_set_close_on_exec(context, newfd, false);

	mutex_unlock(&context->io_mutex);

	// Say bye bye to the evicted fd
	if (evicted) {
		deselect_select_infos(evicted, selectInfos, true);
		close_fd(context, evicted);
		put_fd(evicted);
	}

	return newfd;
}


/*!	Duplicates an FD from another team to this/the kernel team.
	\param fromTeam The team which owns the FD.
	\param fd The FD to duplicate.
	\param kernel If \c true, the new FD will be created in the kernel team,
			the current userland team otherwise.
	\return The newly created FD or an error code, if something went wrong.
*/
int
dup_foreign_fd(team_id fromTeam, int fd, bool kernel)
{
	// get the I/O context for the team in question
	Team* team = Team::Get(fromTeam);
	if (team == NULL)
		return B_BAD_TEAM_ID;
	BReference<Team> teamReference(team, true);

	io_context* fromContext = team->io_context;

	// get the file descriptor
	file_descriptor* descriptor = get_fd(fromContext, fd);
	if (descriptor == NULL)
		return B_FILE_ERROR;
	DescriptorPutter descriptorPutter(descriptor);

	// create a new FD in the target I/O context
	int result = new_fd(get_current_io_context(kernel), descriptor);
	if (result >= 0) {
		// the descriptor reference belongs to the slot, now
		descriptorPutter.Detach();
	}

	return result;
}


static status_t
fd_ioctl(bool kernelFD, int fd, uint32 op, void* buffer, size_t length)
{
	struct file_descriptor* descriptor;
	int status;

	descriptor = get_fd(get_current_io_context(kernelFD), fd);
	if (descriptor == NULL)
		return B_FILE_ERROR;

	if (descriptor->ops->fd_ioctl)
		status = descriptor->ops->fd_ioctl(descriptor, op, buffer, length);
	else
		status = B_DEV_INVALID_IOCTL;

	if (status == B_DEV_INVALID_IOCTL)
		status = ENOTTY;

	put_fd(descriptor);
	return status;
}


static void
deselect_select_infos(file_descriptor* descriptor, select_info* infos,
	bool putSyncObjects)
{
	TRACE(("deselect_select_infos(%p, %p)\n", descriptor, infos));

	select_info* info = infos;
	while (info != NULL) {
		select_sync* sync = info->sync;

		// deselect the selected events
		uint16 eventsToDeselect = info->selected_events & ~B_EVENT_INVALID;
		if (descriptor->ops->fd_deselect != NULL && eventsToDeselect != 0) {
			for (uint16 event = 1; event < 16; event++) {
				if ((eventsToDeselect & SELECT_FLAG(event)) != 0) {
					descriptor->ops->fd_deselect(descriptor, event,
						(selectsync*)info);
				}
			}
		}

		notify_select_events(info, B_EVENT_INVALID);
		info = info->next;

		if (putSyncObjects)
			put_select_sync(sync);
	}
}


status_t
select_fd(int32 fd, struct select_info* info, bool kernel)
{
	TRACE(("select_fd(fd = %" B_PRId32 ", info = %p (%p), 0x%x)\n", fd, info,
		info->sync, info->selected_events));

	FDGetter fdGetter;
		// define before the context locker, so it will be destroyed after it

	io_context* context = get_current_io_context(kernel);
	MutexLocker locker(context->io_mutex);

	struct file_descriptor* descriptor = fdGetter.SetTo(context, fd, true);
	if (descriptor == NULL)
		return B_FILE_ERROR;

	uint16 eventsToSelect = info->selected_events & ~B_EVENT_INVALID;

	if (descriptor->ops->fd_select == NULL) {
		// if the I/O subsystem doesn't support select(), we will
		// immediately notify the select call
		eventsToSelect &= ~SELECT_OUTPUT_ONLY_FLAGS;
		if (eventsToSelect != 0)
			return notify_select_events(info, eventsToSelect);
		else
			return B_OK;
	}

	// We need the FD to stay open while we're doing this, so no select()/
	// deselect() will be called on it after it is closed.
	atomic_add(&descriptor->open_count, 1);

	locker.Unlock();

	// select any events asked for
	uint32 selectedEvents = 0;

	for (uint16 event = 1; event < 16; event++) {
		if ((eventsToSelect & SELECT_FLAG(event)) != 0
			&& descriptor->ops->fd_select(descriptor, event,
				(selectsync*)info) == B_OK) {
			selectedEvents |= SELECT_FLAG(event);
		}
	}
	info->selected_events = selectedEvents
		| (info->selected_events & B_EVENT_INVALID);

	// Add the info to the IO context. Even if nothing has been selected -- we
	// always support B_EVENT_INVALID.
	locker.Lock();
	if (context->fds[fd] != descriptor) {
		// Someone close()d the index in the meantime. deselect() all
		// events.
		info->next = NULL;
		deselect_select_infos(descriptor, info, false);

		// Release our open reference of the descriptor.
		close_fd(context, descriptor);
		return B_FILE_ERROR;
	}

	// The FD index hasn't changed, so we add the select info to the table.

	info->next = context->select_infos[fd];
	context->select_infos[fd] = info;

	// As long as the info is in the list, we keep a reference to the sync
	// object.
	atomic_add(&info->sync->ref_count, 1);

	// Finally release our open reference. It is safe just to decrement,
	// since as long as the descriptor is associated with the slot,
	// someone else still has it open.
	atomic_add(&descriptor->open_count, -1);

	return B_OK;
}


status_t
deselect_fd(int32 fd, struct select_info* info, bool kernel)
{
	TRACE(("deselect_fd(fd = %" B_PRId32 ", info = %p (%p), 0x%x)\n", fd, info,
		info->sync, info->selected_events));

	FDGetter fdGetter;
		// define before the context locker, so it will be destroyed after it

	io_context* context = get_current_io_context(kernel);
	MutexLocker locker(context->io_mutex);

	struct file_descriptor* descriptor = fdGetter.SetTo(context, fd, true);
	if (descriptor == NULL)
		return B_FILE_ERROR;

	// remove the info from the IO context

	select_info** infoLocation = &context->select_infos[fd];
	while (*infoLocation != NULL && *infoLocation != info)
		infoLocation = &(*infoLocation)->next;

	// If not found, someone else beat us to it.
	if (*infoLocation != info)
		return B_OK;

	*infoLocation = info->next;

	locker.Unlock();

	// deselect the selected events
	uint16 eventsToDeselect = info->selected_events & ~B_EVENT_INVALID;
	if (descriptor->ops->fd_deselect != NULL && eventsToDeselect != 0) {
		for (uint16 event = 1; event < 16; event++) {
			if ((eventsToDeselect & SELECT_FLAG(event)) != 0) {
				descriptor->ops->fd_deselect(descriptor, event,
					(selectsync*)info);
			}
		}
	}

	put_select_sync(info->sync);

	return B_OK;
}


/*!	This function checks if the specified fd is valid in the current
	context. It can be used for a quick check; the fd is not locked
	so it could become invalid immediately after this check.
*/
bool
fd_is_valid(int fd, bool kernel)
{
	struct file_descriptor* descriptor
		= get_fd(get_current_io_context(kernel), fd);
	if (descriptor == NULL)
		return false;

	put_fd(descriptor);
	return true;
}


struct vnode*
fd_vnode(struct file_descriptor* descriptor)
{
	switch (descriptor->type) {
		case FDTYPE_FILE:
		case FDTYPE_DIR:
		case FDTYPE_ATTR_DIR:
		case FDTYPE_ATTR:
			return descriptor->u.vnode;
	}

	return NULL;
}


static status_t
common_close(int fd, bool kernel)
{
	return close_fd_index(get_current_io_context(kernel), fd);
}


static ssize_t
common_user_io(int fd, off_t pos, void* buffer, size_t length, bool write)
{
	if (pos < -1)
		return B_BAD_VALUE;

	FDGetter fdGetter;
	struct file_descriptor* descriptor = fdGetter.SetTo(fd, false);
	if (!descriptor)
		return B_FILE_ERROR;

	if (write ? (descriptor->open_mode & O_RWMASK) == O_RDONLY
			: (descriptor->open_mode & O_RWMASK) == O_WRONLY) {
		return B_FILE_ERROR;
	}

	bool movePosition = false;
	if (pos == -1) {
		pos = descriptor->pos;
		movePosition = true;
	}

	if (write ? descriptor->ops->fd_write == NULL
			: descriptor->ops->fd_read == NULL) {
		return B_BAD_VALUE;
	}

	if (length == 0)
		return 0;

	if (!is_user_address_range(buffer, length))
		return B_BAD_ADDRESS;

	SyscallRestartWrapper<status_t> status;

	if (write)
		status = descriptor->ops->fd_write(descriptor, pos, buffer, &length);
	else
		status = descriptor->ops->fd_read(descriptor, pos, buffer, &length);

	if (status != B_OK)
		return status;

	if (movePosition) {
		descriptor->pos = write && (descriptor->open_mode & O_APPEND) != 0
			? descriptor->ops->fd_seek(descriptor, 0, SEEK_END) : pos + length;
	}

	return length <= SSIZE_MAX ? (ssize_t)length : SSIZE_MAX;
}


static ssize_t
common_user_vector_io(int fd, off_t pos, const iovec* userVecs, size_t count,
	bool write)
{
	if (pos < -1)
		return B_BAD_VALUE;

	iovec* vecs;
	status_t error = get_iovecs_from_user(userVecs, count, vecs, true);
	if (error != B_OK)
		return error;
	MemoryDeleter _(vecs);

	FDGetter fdGetter;
	struct file_descriptor* descriptor = fdGetter.SetTo(fd, false);
	if (!descriptor)
		return B_FILE_ERROR;

	if (write ? (descriptor->open_mode & O_RWMASK) == O_RDONLY
			: (descriptor->open_mode & O_RWMASK) == O_WRONLY) {
		return B_FILE_ERROR;
	}

	bool movePosition = false;
	if (pos == -1) {
		pos = descriptor->pos;
		movePosition = true;
	}

	if (write ? descriptor->ops->fd_write == NULL
			: descriptor->ops->fd_read == NULL) {
		return B_BAD_VALUE;
	}

	SyscallRestartWrapper<status_t> status;

	ssize_t bytesTransferred = 0;
	for (size_t i = 0; i < count; i++) {
		if (vecs[i].iov_base == NULL)
			continue;

		size_t length = vecs[i].iov_len;
		if (write) {
			status = descriptor->ops->fd_write(descriptor, pos,
				vecs[i].iov_base, &length);
		} else {
			status = descriptor->ops->fd_read(descriptor, pos, vecs[i].iov_base,
				&length);
		}

		if (status != B_OK) {
			if (bytesTransferred == 0)
				return status;
			status = B_OK;
			break;
		}

		if ((uint64)bytesTransferred + length > SSIZE_MAX)
			bytesTransferred = SSIZE_MAX;
		else
			bytesTransferred += (ssize_t)length;

		pos += length;

		if (length < vecs[i].iov_len)
			break;
	}

	if (movePosition) {
		descriptor->pos = write && (descriptor->open_mode & O_APPEND) != 0
			? descriptor->ops->fd_seek(descriptor, 0, SEEK_END) : pos;
	}

	return bytesTransferred;
}


status_t
user_fd_kernel_ioctl(int fd, uint32 op, void* buffer, size_t length)
{
	TRACE(("user_fd_kernel_ioctl: fd %d\n", fd));

	return fd_ioctl(false, fd, op, buffer, length);
}


//	#pragma mark - User syscalls


ssize_t
_user_read(int fd, off_t pos, void* buffer, size_t length)
{
	return common_user_io(fd, pos, buffer, length, false);
}


ssize_t
_user_readv(int fd, off_t pos, const iovec* userVecs, size_t count)
{
	return common_user_vector_io(fd, pos, userVecs, count, false);
}


ssize_t
_user_write(int fd, off_t pos, const void* buffer, size_t length)
{
	return common_user_io(fd, pos, (void*)buffer, length, true);
}


ssize_t
_user_writev(int fd, off_t pos, const iovec* userVecs, size_t count)
{
	return common_user_vector_io(fd, pos, userVecs, count, true);
}


off_t
_user_seek(int fd, off_t pos, int seekType)
{
	syscall_64_bit_return_value();

	struct file_descriptor* descriptor;

	descriptor = get_fd(get_current_io_context(false), fd);
	if (!descriptor)
		return B_FILE_ERROR;

	TRACE(("user_seek(descriptor = %p)\n", descriptor));

	if (descriptor->ops->fd_seek)
		pos = descriptor->ops->fd_seek(descriptor, pos, seekType);
	else
		pos = ESPIPE;

	put_fd(descriptor);
	return pos;
}


status_t
_user_ioctl(int fd, uint32 op, void* buffer, size_t length)
{
	TRACE(("user_ioctl: fd %d\n", fd));

	// "buffer" is not always a pointer depending on "op", so we cannot
	// check that it is a userland buffer here. Instead we check that
	// it is at least not within the bounds of kernel memory; as in
	// the cases where it is a numeric constant it is usually a low one.
	if (IS_KERNEL_ADDRESS(buffer))
		return B_BAD_ADDRESS;

	SyscallRestartWrapper<status_t> status;

	return status = fd_ioctl(false, fd, op, buffer, length);
}


ssize_t
_user_read_dir(int fd, struct dirent* userBuffer, size_t bufferSize,
	uint32 maxCount)
{
	TRACE(("user_read_dir(fd = %d, userBuffer = %p, bufferSize = %ld, count = "
		"%" B_PRIu32 ")\n", fd, userBuffer, bufferSize, maxCount));

	if (maxCount == 0)
		return 0;

	if (userBuffer == NULL || !IS_USER_ADDRESS(userBuffer))
		return B_BAD_ADDRESS;

	// get I/O context and FD
	io_context* ioContext = get_current_io_context(false);
	FDGetter fdGetter;
	struct file_descriptor* descriptor = fdGetter.SetTo(ioContext, fd, false);
	if (descriptor == NULL)
		return B_FILE_ERROR;

	if (descriptor->ops->fd_read_dir == NULL)
		return B_UNSUPPORTED;

	// restrict buffer size and allocate a heap buffer
	if (bufferSize > kMaxReadDirBufferSize)
		bufferSize = kMaxReadDirBufferSize;
	struct dirent* buffer = (struct dirent*)malloc(bufferSize);
	if (buffer == NULL)
		return B_NO_MEMORY;
	MemoryDeleter bufferDeleter(buffer);

	// read the directory
	uint32 count = maxCount;
	status_t status = descriptor->ops->fd_read_dir(ioContext, descriptor,
		buffer, bufferSize, &count);
	if (status != B_OK)
		return status;

	ASSERT(count <= maxCount);

	// copy the buffer back -- determine the total buffer size first
	size_t sizeToCopy = 0;
	BytePointer<struct dirent> entry = buffer;
	for (uint32 i = 0; i < count; i++) {
		size_t length = entry->d_reclen;
		sizeToCopy += length;
		entry += length;
	}

	ASSERT(sizeToCopy <= bufferSize);

	if (user_memcpy(userBuffer, buffer, sizeToCopy) != B_OK)
		return B_BAD_ADDRESS;

	return count;
}


status_t
_user_rewind_dir(int fd)
{
	struct file_descriptor* descriptor;
	status_t status;

	TRACE(("user_rewind_dir(fd = %d)\n", fd));

	descriptor = get_fd(get_current_io_context(false), fd);
	if (descriptor == NULL)
		return B_FILE_ERROR;

	if (descriptor->ops->fd_rewind_dir)
		status = descriptor->ops->fd_rewind_dir(descriptor);
	else
		status = B_UNSUPPORTED;

	put_fd(descriptor);
	return status;
}


status_t
_user_close(int fd)
{
	return common_close(fd, false);
}


int
_user_dup(int fd)
{
	return dup_fd(fd, false);
}


int
_user_dup2(int ofd, int nfd)
{
	return dup2_fd(ofd, nfd, false);
}


//	#pragma mark - Kernel calls


ssize_t
_kern_read(int fd, off_t pos, void* buffer, size_t length)
{
	if (pos < -1)
		return B_BAD_VALUE;

	FDGetter fdGetter;
	struct file_descriptor* descriptor = fdGetter.SetTo(fd, true);

	if (!descriptor)
		return B_FILE_ERROR;
	if ((descriptor->open_mode & O_RWMASK) == O_WRONLY)
		return B_FILE_ERROR;

	bool movePosition = false;
	if (pos == -1) {
		pos = descriptor->pos;
		movePosition = true;
	}

	SyscallFlagUnsetter _;

	if (descriptor->ops->fd_read == NULL)
		return B_BAD_VALUE;

	ssize_t bytesRead = descriptor->ops->fd_read(descriptor, pos, buffer,
		&length);
	if (bytesRead >= B_OK) {
		if (length > SSIZE_MAX)
			bytesRead = SSIZE_MAX;
		else
			bytesRead = (ssize_t)length;

		if (movePosition)
			descriptor->pos = pos + length;
	}

	return bytesRead;
}


ssize_t
_kern_readv(int fd, off_t pos, const iovec* vecs, size_t count)
{
	bool movePosition = false;
	status_t status;

	if (pos < -1)
		return B_BAD_VALUE;

	FDGetter fdGetter;
	struct file_descriptor* descriptor = fdGetter.SetTo(fd, true);

	if (!descriptor)
		return B_FILE_ERROR;
	if ((descriptor->open_mode & O_RWMASK) == O_WRONLY)
		return B_FILE_ERROR;

	if (pos == -1) {
		pos = descriptor->pos;
		movePosition = true;
	}

	if (descriptor->ops->fd_read == NULL)
		return B_BAD_VALUE;

	SyscallFlagUnsetter _;

	ssize_t bytesRead = 0;

	for (size_t i = 0; i < count; i++) {
		size_t length = vecs[i].iov_len;
		status = descriptor->ops->fd_read(descriptor, pos, vecs[i].iov_base,
			&length);
		if (status != B_OK) {
			bytesRead = status;
			break;
		}

		if ((uint64)bytesRead + length > SSIZE_MAX)
			bytesRead = SSIZE_MAX;
		else
			bytesRead += (ssize_t)length;

		pos += vecs[i].iov_len;
	}

	if (movePosition)
		descriptor->pos = pos;

	return bytesRead;
}


ssize_t
_kern_write(int fd, off_t pos, const void* buffer, size_t length)
{
	if (pos < -1)
		return B_BAD_VALUE;

	FDGetter fdGetter;
	struct file_descriptor* descriptor = fdGetter.SetTo(fd, true);

	if (descriptor == NULL)
		return B_FILE_ERROR;
	if ((descriptor->open_mode & O_RWMASK) == O_RDONLY)
		return B_FILE_ERROR;

	bool movePosition = false;
	if (pos == -1) {
		pos = descriptor->pos;
		movePosition = true;
	}

	if (descriptor->ops->fd_write == NULL)
		return B_BAD_VALUE;

	SyscallFlagUnsetter _;

	ssize_t bytesWritten = descriptor->ops->fd_write(descriptor, pos, buffer,
		&length);
	if (bytesWritten >= B_OK) {
		if (length > SSIZE_MAX)
			bytesWritten = SSIZE_MAX;
		else
			bytesWritten = (ssize_t)length;

		if (movePosition)
			descriptor->pos = pos + length;
	}

	return bytesWritten;
}


ssize_t
_kern_writev(int fd, off_t pos, const iovec* vecs, size_t count)
{
	bool movePosition = false;
	status_t status;

	if (pos < -1)
		return B_BAD_VALUE;

	FDGetter fdGetter;
	struct file_descriptor* descriptor = fdGetter.SetTo(fd, true);

	if (!descriptor)
		return B_FILE_ERROR;
	if ((descriptor->open_mode & O_RWMASK) == O_RDONLY)
		return B_FILE_ERROR;

	if (pos == -1) {
		pos = descriptor->pos;
		movePosition = true;
	}

	if (descriptor->ops->fd_write == NULL)
		return B_BAD_VALUE;

	SyscallFlagUnsetter _;

	ssize_t bytesWritten = 0;

	for (size_t i = 0; i < count; i++) {
		size_t length = vecs[i].iov_len;
		status = descriptor->ops->fd_write(descriptor, pos,
			vecs[i].iov_base, &length);
		if (status != B_OK) {
			bytesWritten = status;
			break;
		}

		if ((uint64)bytesWritten + length > SSIZE_MAX)
			bytesWritten = SSIZE_MAX;
		else
			bytesWritten += (ssize_t)length;

		pos += vecs[i].iov_len;
	}

	if (movePosition)
		descriptor->pos = pos;

	return bytesWritten;
}


off_t
_kern_seek(int fd, off_t pos, int seekType)
{
	struct file_descriptor* descriptor;

	descriptor = get_fd(get_current_io_context(true), fd);
	if (!descriptor)
		return B_FILE_ERROR;

	if (descriptor->ops->fd_seek)
		pos = descriptor->ops->fd_seek(descriptor, pos, seekType);
	else
		pos = ESPIPE;

	put_fd(descriptor);
	return pos;
}


status_t
_kern_ioctl(int fd, uint32 op, void* buffer, size_t length)
{
	TRACE(("kern_ioctl: fd %d\n", fd));

	SyscallFlagUnsetter _;

	return fd_ioctl(true, fd, op, buffer, length);
}


ssize_t
_kern_read_dir(int fd, struct dirent* buffer, size_t bufferSize,
	uint32 maxCount)
{
	struct file_descriptor* descriptor;
	ssize_t retval;

	TRACE(("sys_read_dir(fd = %d, buffer = %p, bufferSize = %ld, count = "
		"%" B_PRIu32 ")\n",fd, buffer, bufferSize, maxCount));

	struct io_context* ioContext = get_current_io_context(true);
	descriptor = get_fd(ioContext, fd);
	if (descriptor == NULL)
		return B_FILE_ERROR;

	if (descriptor->ops->fd_read_dir) {
		uint32 count = maxCount;
		retval = descriptor->ops->fd_read_dir(ioContext, descriptor, buffer,
			bufferSize, &count);
		if (retval >= 0)
			retval = count;
	} else
		retval = B_UNSUPPORTED;

	put_fd(descriptor);
	return retval;
}


status_t
_kern_rewind_dir(int fd)
{
	struct file_descriptor* descriptor;
	status_t status;

	TRACE(("sys_rewind_dir(fd = %d)\n",fd));

	descriptor = get_fd(get_current_io_context(true), fd);
	if (descriptor == NULL)
		return B_FILE_ERROR;

	if (descriptor->ops->fd_rewind_dir)
		status = descriptor->ops->fd_rewind_dir(descriptor);
	else
		status = B_UNSUPPORTED;

	put_fd(descriptor);
	return status;
}


status_t
_kern_close(int fd)
{
	return common_close(fd, true);
}


int
_kern_dup(int fd)
{
	return dup_fd(fd, true);
}


int
_kern_dup2(int ofd, int nfd)
{
	return dup2_fd(ofd, nfd, true);
}