relay.c

Kernel code of linux kernel

	if (filp->f_mode & FMODE_READ) {
		poll_wait(filp, &buf->read_wait, wait);
		if (!relay_buf_empty(buf))
			mask |= POLLIN | POLLRDNORM;
	}

	return mask;
}

/**
 *	relay_file_release - release file op for relay files
 *	@inode: the inode
 *	@filp: the file
 *
 *	Decrements the channel refcount, as the filesystem is
 *	no longer using it.
 */
static int relay_file_release(struct inode *inode, struct file *filp)
{
	struct rchan_buf *buf = filp->private_data;
	kref_put(&buf->kref, relay_remove_buf);

	return 0;
}

/*
 *	relay_file_read_consume - update the consumed count for the buffer
 */
static void relay_file_read_consume(struct rchan_buf *buf,
				    size_t read_pos,
				    size_t bytes_consumed)
{
	size_t subbuf_size = buf->chan->subbuf_size;
	size_t n_subbufs = buf->chan->n_subbufs;
	size_t read_subbuf;

	if (buf->subbufs_produced == buf->subbufs_consumed &&
	    buf->offset == buf->bytes_consumed)
		return;

	if (buf->bytes_consumed + bytes_consumed > subbuf_size) {
		relay_subbufs_consumed(buf->chan, buf->cpu, 1);
		buf->bytes_consumed = 0;
	}

	buf->bytes_consumed += bytes_consumed;
	if (!read_pos)
		read_subbuf = buf->subbufs_consumed % n_subbufs;
	else
		read_subbuf = read_pos / buf->chan->subbuf_size;
	if (buf->bytes_consumed + buf->padding[read_subbuf] == subbuf_size) {
		if ((read_subbuf == buf->subbufs_produced % n_subbufs) &&
		    (buf->offset == subbuf_size))
			return;
		relay_subbufs_consumed(buf->chan, buf->cpu, 1);
		buf->bytes_consumed = 0;
	}
}

/*
 *	relay_file_read_avail - boolean, are there unconsumed bytes available?
 */
static int relay_file_read_avail(struct rchan_buf *buf, size_t read_pos)
{
	size_t subbuf_size = buf->chan->subbuf_size;
	size_t n_subbufs = buf->chan->n_subbufs;
	size_t produced = buf->subbufs_produced;
	size_t consumed = buf->subbufs_consumed;

	relay_file_read_consume(buf, read_pos, 0);

	consumed = buf->subbufs_consumed;

	if (unlikely(buf->offset > subbuf_size)) {
		if (produced == consumed)
			return 0;
		return 1;
	}

	if (unlikely(produced - consumed >= n_subbufs)) {
		consumed = produced - n_subbufs + 1;
		buf->subbufs_consumed = consumed;
		buf->bytes_consumed = 0;
	}

	produced = (produced % n_subbufs) * subbuf_size + buf->offset;
	consumed = (consumed % n_subbufs) * subbuf_size + buf->bytes_consumed;

	if (consumed > produced)
		produced += n_subbufs * subbuf_size;

	if (consumed == produced) {
		if (buf->offset == subbuf_size &&
		    buf->subbufs_produced > buf->subbufs_consumed)
			return 1;
		return 0;
	}

	return 1;
}

/**
 *	relay_file_read_subbuf_avail - return bytes available in sub-buffer
 *	@read_pos: file read position
 *	@buf: relay channel buffer
 */
static size_t relay_file_read_subbuf_avail(size_t read_pos,
					   struct rchan_buf *buf)
{
	size_t padding, avail = 0;
	size_t read_subbuf, read_offset, write_subbuf, write_offset;
	size_t subbuf_size = buf->chan->subbuf_size;

	write_subbuf = (buf->data - buf->start) / subbuf_size;
	write_offset = buf->offset > subbuf_size ? subbuf_size : buf->offset;
	read_subbuf = read_pos / subbuf_size;
	read_offset = read_pos % subbuf_size;
	padding = buf->padding[read_subbuf];

	if (read_subbuf == write_subbuf) {
		if (read_offset + padding < write_offset)
			avail = write_offset - (read_offset + padding);
	} else
		avail = (subbuf_size - padding) - read_offset;

	return avail;
}

/**
 *	relay_file_read_start_pos - find the first available byte to read
 *	@read_pos: file read position
 *	@buf: relay channel buffer
 *
 *	If the @read_pos is in the middle of padding, return the
 *	position of the first actually available byte, otherwise
 *	return the original value.
 */
static size_t relay_file_read_start_pos(size_t read_pos,
					struct rchan_buf *buf)
{
	size_t read_subbuf, padding, padding_start, padding_end;
	size_t subbuf_size = buf->chan->subbuf_size;
	size_t n_subbufs = buf->chan->n_subbufs;
	size_t consumed = buf->subbufs_consumed % n_subbufs;

	if (!read_pos)
		read_pos = consumed * subbuf_size + buf->bytes_consumed;
	read_subbuf = read_pos / subbuf_size;
	padding = buf->padding[read_subbuf];
	padding_start = (read_subbuf + 1) * subbuf_size - padding;
	padding_end = (read_subbuf + 1) * subbuf_size;
	if (read_pos >= padding_start && read_pos < padding_end) {
		read_subbuf = (read_subbuf + 1) % n_subbufs;
		read_pos = read_subbuf * subbuf_size;
	}

	return read_pos;
}

/**
 *	relay_file_read_end_pos - return the new read position
 *	@read_pos: file read position
 *	@buf: relay channel buffer
 *	@count: number of bytes to be read
 */
static size_t relay_file_read_end_pos(struct rchan_buf *buf,
				      size_t read_pos,
				      size_t count)
{
	size_t read_subbuf, padding, end_pos;
	size_t subbuf_size = buf->chan->subbuf_size;
	size_t n_subbufs = buf->chan->n_subbufs;

	read_subbuf = read_pos / subbuf_size;
	padding = buf->padding[read_subbuf];
	if (read_pos % subbuf_size + count + padding == subbuf_size)
		end_pos = (read_subbuf + 1) * subbuf_size;
	else
		end_pos = read_pos + count;
	if (end_pos >= subbuf_size * n_subbufs)
		end_pos = 0;

	return end_pos;
}

/*
 *	subbuf_read_actor - read up to one subbuf's worth of data
 */
static int subbuf_read_actor(size_t read_start,
			     struct rchan_buf *buf,
			     size_t avail,
			     read_descriptor_t *desc,
			     read_actor_t actor)
{
	void *from;
	int ret = 0;

	from = buf->start + read_start;
	ret = avail;
	if (copy_to_user(desc->arg.buf, from, avail)) {
		desc->error = -EFAULT;
		ret = 0;
	}
	desc->arg.data += ret;
	desc->written += ret;
	desc->count -= ret;

	return ret;
}

typedef int (*subbuf_actor_t) (size_t read_start,
			       struct rchan_buf *buf,
			       size_t avail,
			       read_descriptor_t *desc,
			       read_actor_t actor);

/*
 *	relay_file_read_subbufs - read count bytes, bridging subbuf boundaries
 */
static ssize_t relay_file_read_subbufs(struct file *filp, loff_t *ppos,
					subbuf_actor_t subbuf_actor,
					read_actor_t actor,
					read_descriptor_t *desc)
{
	struct rchan_buf *buf = filp->private_data;
	size_t read_start, avail;
	int ret;

	if (!desc->count)
		return 0;

	mutex_lock(&filp->f_path.dentry->d_inode->i_mutex);
	do {
		if (!relay_file_read_avail(buf, *ppos))
			break;

		read_start = relay_file_read_start_pos(*ppos, buf);
		avail = relay_file_read_subbuf_avail(read_start, buf);
		if (!avail)
			break;

		avail = min(desc->count, avail);
		ret = subbuf_actor(read_start, buf, avail, desc, actor);
		if (desc->error < 0)
			break;

		if (ret) {
			relay_file_read_consume(buf, read_start, ret);
			*ppos = relay_file_read_end_pos(buf, read_start, ret);
		}
	} while (desc->count && ret);
	mutex_unlock(&filp->f_path.dentry->d_inode->i_mutex);

	return desc->written;
}

static ssize_t relay_file_read(struct file *filp,
			       char __user *buffer,
			       size_t count,
			       loff_t *ppos)
{
	read_descriptor_t desc;
	desc.written = 0;
	desc.count = count;
	desc.arg.buf = buffer;
	desc.error = 0;
	return relay_file_read_subbufs(filp, ppos, subbuf_read_actor,
				       NULL, &desc);
}

static void relay_consume_bytes(struct rchan_buf *rbuf, int bytes_consumed)
{
	rbuf->bytes_consumed += bytes_consumed;

	if (rbuf->bytes_consumed >= rbuf->chan->subbuf_size) {
		relay_subbufs_consumed(rbuf->chan, rbuf->cpu, 1);
		rbuf->bytes_consumed %= rbuf->chan->subbuf_size;
	}
}

static void relay_pipe_buf_release(struct pipe_inode_info *pipe,
				   struct pipe_buffer *buf)
{
	struct rchan_buf *rbuf;

	rbuf = (struct rchan_buf *)page_private(buf->page);
	relay_consume_bytes(rbuf, buf->private);
}

static struct pipe_buf_operations relay_pipe_buf_ops = {
	.can_merge = 0,
	.map = generic_pipe_buf_map,
	.unmap = generic_pipe_buf_unmap,
	.confirm = generic_pipe_buf_confirm,
	.release = relay_pipe_buf_release,
	.steal = generic_pipe_buf_steal,
	.get = generic_pipe_buf_get,
};

static void relay_page_release(struct splice_pipe_desc *spd, unsigned int i)
{
}

/*
 *	subbuf_splice_actor - splice up to one subbuf's worth of data
 */
static int subbuf_splice_actor(struct file *in,
			       loff_t *ppos,
			       struct pipe_inode_info *pipe,
			       size_t len,
			       unsigned int flags,
			       int *nonpad_ret)
{
	unsigned int pidx, poff, total_len, subbuf_pages, nr_pages, ret;
	struct rchan_buf *rbuf = in->private_data;
	unsigned int subbuf_size = rbuf->chan->subbuf_size;
	uint64_t pos = (uint64_t) *ppos;
	uint32_t alloc_size = (uint32_t) rbuf->chan->alloc_size;
	size_t read_start = (size_t) do_div(pos, alloc_size);
	size_t read_subbuf = read_start / subbuf_size;
	size_t padding = rbuf->padding[read_subbuf];
	size_t nonpad_end = read_subbuf * subbuf_size + subbuf_size - padding;
	struct page *pages[PIPE_BUFFERS];
	struct partial_page partial[PIPE_BUFFERS];
	struct splice_pipe_desc spd = {
		.pages = pages,
		.nr_pages = 0,
		.partial = partial,
		.flags = flags,
		.ops = &relay_pipe_buf_ops,
		.spd_release = relay_page_release,
	};

	if (rbuf->subbufs_produced == rbuf->subbufs_consumed)
		return 0;

	/*
	 * Adjust read len, if longer than what is available
	 */
	if (len > (subbuf_size - read_start % subbuf_size))
		len = subbuf_size - read_start % subbuf_size;

	subbuf_pages = rbuf->chan->alloc_size >> PAGE_SHIFT;
	pidx = (read_start / PAGE_SIZE) % subbuf_pages;
	poff = read_start & ~PAGE_MASK;
	nr_pages = min_t(unsigned int, subbuf_pages, PIPE_BUFFERS);

	for (total_len = 0; spd.nr_pages < nr_pages; spd.nr_pages++) {
		unsigned int this_len, this_end, private;
		unsigned int cur_pos = read_start + total_len;

		if (!len)
			break;

		this_len = min_t(unsigned long, len, PAGE_SIZE - poff);
		private = this_len;

		spd.pages[spd.nr_pages] = rbuf->page_array[pidx];
		spd.partial[spd.nr_pages].offset = poff;

		this_end = cur_pos + this_len;
		if (this_end >= nonpad_end) {
			this_len = nonpad_end - cur_pos;
			private = this_len + padding;
		}
		spd.partial[spd.nr_pages].len = this_len;
		spd.partial[spd.nr_pages].private = private;

		len -= this_len;
		total_len += this_len;
		poff = 0;
		pidx = (pidx + 1) % subbuf_pages;

		if (this_end >= nonpad_end) {
			spd.nr_pages++;
			break;
		}
	}

	if (!spd.nr_pages)
		return 0;

	ret = *nonpad_ret = splice_to_pipe(pipe, &spd);
	if (ret < 0 || ret < total_len)
		return ret;

        if (read_start + ret == nonpad_end)
                ret += padding;

        return ret;
}

static ssize_t relay_file_splice_read(struct file *in,
				      loff_t *ppos,
				      struct pipe_inode_info *pipe,
				      size_t len,
				      unsigned int flags)
{
	ssize_t spliced;
	int ret;
	int nonpad_ret = 0;

	ret = 0;
	spliced = 0;

	while (len && !spliced) {
		ret = subbuf_splice_actor(in, ppos, pipe, len, flags, &nonpad_ret);
		if (ret < 0)
			break;
		else if (!ret) {
			if (spliced)
				break;
			if (flags & SPLICE_F_NONBLOCK) {
				ret = -EAGAIN;
				break;
			}
		}

		*ppos += ret;
		if (ret > len)
			len = 0;
		else
			len -= ret;
		spliced += nonpad_ret;
		nonpad_ret = 0;
	}

	if (spliced)
		return spliced;

	return ret;
}

const struct file_operations relay_file_operations = {
	.open		= relay_file_open,
	.poll		= relay_file_poll,
	.mmap		= relay_file_mmap,
	.read		= relay_file_read,
	.llseek		= no_llseek,
	.release	= relay_file_release,
	.splice_read	= relay_file_splice_read,
};
EXPORT_SYMBOL_GPL(relay_file_operations);

static __init int relay_init(void)
{

	hotcpu_notifier(relay_hotcpu_callback, 0);
	return 0;
}

early_initcall(relay_init);