nbd.c

来自「linux 内核源代码」· C语言 代码 · 共 740 行 · 第 1/2 页

/*
 * Network block device - make block devices work over TCP
 *
 * Note that you can not swap over this thing, yet. Seems to work but
 * deadlocks sometimes - you can not swap over TCP in general.
 * 
 * Copyright 1997-2000 Pavel Machek <[e-mail]>
 * Parts copyright 2001 Steven Whitehouse <[e-mail]>
 *
 * This file is released under GPLv2 or later.
 *
 * (part of code stolen from loop.c)
 */

#include <linux/major.h>

#include <linux/blkdev.h>
#include <linux/module.h>
#include <linux/init.h>
#include <linux/sched.h>
#include <linux/fs.h>
#include <linux/bio.h>
#include <linux/stat.h>
#include <linux/errno.h>
#include <linux/file.h>
#include <linux/ioctl.h>
#include <linux/compiler.h>
#include <linux/err.h>
#include <linux/kernel.h>
#include <net/sock.h>
#include <linux/net.h>

#include <asm/uaccess.h>
#include <asm/system.h>
#include <asm/types.h>

#include <linux/nbd.h>

#define LO_MAGIC 0x68797548

#ifdef NDEBUG
#define dprintk(flags, fmt...)
#else /* NDEBUG */
#define dprintk(flags, fmt...) do { \
	if (debugflags & (flags)) printk(KERN_DEBUG fmt); \
} while (0)
#define DBG_IOCTL       0x0004
#define DBG_INIT        0x0010
#define DBG_EXIT        0x0020
#define DBG_BLKDEV      0x0100
#define DBG_RX          0x0200
#define DBG_TX          0x0400
static unsigned int debugflags;
#endif /* NDEBUG */

static unsigned int nbds_max = 16;
static struct nbd_device nbd_dev[MAX_NBD];

/*
 * Use just one lock (or at most 1 per NIC). Two arguments for this:
 * 1. Each NIC is essentially a synchronization point for all servers
 *    accessed through that NIC so there's no need to have more locks
 *    than NICs anyway.
 * 2. More locks lead to more "Dirty cache line bouncing" which will slow
 *    down each lock to the point where they're actually slower than just
 *    a single lock.
 * Thanks go to Jens Axboe and Al Viro for their LKML emails explaining this!
 */
static DEFINE_SPINLOCK(nbd_lock);

#ifndef NDEBUG
static const char *ioctl_cmd_to_ascii(int cmd)
{
	switch (cmd) {
	case NBD_SET_SOCK: return "set-sock";
	case NBD_SET_BLKSIZE: return "set-blksize";
	case NBD_SET_SIZE: return "set-size";
	case NBD_DO_IT: return "do-it";
	case NBD_CLEAR_SOCK: return "clear-sock";
	case NBD_CLEAR_QUE: return "clear-que";
	case NBD_PRINT_DEBUG: return "print-debug";
	case NBD_SET_SIZE_BLOCKS: return "set-size-blocks";
	case NBD_DISCONNECT: return "disconnect";
	case BLKROSET: return "set-read-only";
	case BLKFLSBUF: return "flush-buffer-cache";
	}
	return "unknown";
}

static const char *nbdcmd_to_ascii(int cmd)
{
	switch (cmd) {
	case  NBD_CMD_READ: return "read";
	case NBD_CMD_WRITE: return "write";
	case  NBD_CMD_DISC: return "disconnect";
	}
	return "invalid";
}
#endif /* NDEBUG */

static void nbd_end_request(struct request *req)
{
	int uptodate = (req->errors == 0) ? 1 : 0;
	struct request_queue *q = req->q;
	unsigned long flags;

	dprintk(DBG_BLKDEV, "%s: request %p: %s\n", req->rq_disk->disk_name,
			req, uptodate? "done": "failed");

	spin_lock_irqsave(q->queue_lock, flags);
	if (!end_that_request_first(req, uptodate, req->nr_sectors)) {
		end_that_request_last(req, uptodate);
	}
	spin_unlock_irqrestore(q->queue_lock, flags);
}

static void sock_shutdown(struct nbd_device *lo, int lock)
{
	/* Forcibly shutdown the socket causing all listeners
	 * to error
	 *
	 * FIXME: This code is duplicated from sys_shutdown, but
	 * there should be a more generic interface rather than
	 * calling socket ops directly here */
	if (lock)
		mutex_lock(&lo->tx_lock);
	if (lo->sock) {
		printk(KERN_WARNING "%s: shutting down socket\n",
			lo->disk->disk_name);
		kernel_sock_shutdown(lo->sock, SHUT_RDWR);
		lo->sock = NULL;
	}
	if (lock)
		mutex_unlock(&lo->tx_lock);
}

static void nbd_xmit_timeout(unsigned long arg)
{
	struct task_struct *task = (struct task_struct *)arg;

	printk(KERN_WARNING "nbd: killing hung xmit (%s, pid: %d)\n",
		task->comm, task->pid);
	force_sig(SIGKILL, task);
}

/*
 *  Send or receive packet.
 */
static int sock_xmit(struct nbd_device *lo, int send, void *buf, int size,
		int msg_flags)
{
	struct socket *sock = lo->sock;
	int result;
	struct msghdr msg;
	struct kvec iov;
	sigset_t blocked, oldset;

	/* Allow interception of SIGKILL only
	 * Don't allow other signals to interrupt the transmission */
	siginitsetinv(&blocked, sigmask(SIGKILL));
	sigprocmask(SIG_SETMASK, &blocked, &oldset);

	do {
		sock->sk->sk_allocation = GFP_NOIO;
		iov.iov_base = buf;
		iov.iov_len = size;
		msg.msg_name = NULL;
		msg.msg_namelen = 0;
		msg.msg_control = NULL;
		msg.msg_controllen = 0;
		msg.msg_flags = msg_flags | MSG_NOSIGNAL;

		if (send) {
			struct timer_list ti;

			if (lo->xmit_timeout) {
				init_timer(&ti);
				ti.function = nbd_xmit_timeout;
				ti.data = (unsigned long)current;
				ti.expires = jiffies + lo->xmit_timeout;
				add_timer(&ti);
			}
			result = kernel_sendmsg(sock, &msg, &iov, 1, size);
			if (lo->xmit_timeout)
				del_timer_sync(&ti);
		} else
			result = kernel_recvmsg(sock, &msg, &iov, 1, size, 0);

		if (signal_pending(current)) {
			siginfo_t info;
			printk(KERN_WARNING "nbd (pid %d: %s) got signal %d\n",
				task_pid_nr(current), current->comm,
				dequeue_signal_lock(current, &current->blocked, &info));
			result = -EINTR;
			sock_shutdown(lo, !send);
			break;
		}

		if (result <= 0) {
			if (result == 0)
				result = -EPIPE; /* short read */
			break;
		}
		size -= result;
		buf += result;
	} while (size > 0);

	sigprocmask(SIG_SETMASK, &oldset, NULL);

	return result;
}

static inline int sock_send_bvec(struct nbd_device *lo, struct bio_vec *bvec,
		int flags)
{
	int result;
	void *kaddr = kmap(bvec->bv_page);
	result = sock_xmit(lo, 1, kaddr + bvec->bv_offset, bvec->bv_len, flags);
	kunmap(bvec->bv_page);
	return result;
}

/* always call with the tx_lock held */
static int nbd_send_req(struct nbd_device *lo, struct request *req)
{
	int result, flags;
	struct nbd_request request;
	unsigned long size = req->nr_sectors << 9;

	request.magic = htonl(NBD_REQUEST_MAGIC);
	request.type = htonl(nbd_cmd(req));
	request.from = cpu_to_be64((u64) req->sector << 9);
	request.len = htonl(size);
	memcpy(request.handle, &req, sizeof(req));

	dprintk(DBG_TX, "%s: request %p: sending control (%s@%llu,%luB)\n",
			lo->disk->disk_name, req,
			nbdcmd_to_ascii(nbd_cmd(req)),
			(unsigned long long)req->sector << 9,
			req->nr_sectors << 9);
	result = sock_xmit(lo, 1, &request, sizeof(request),
			(nbd_cmd(req) == NBD_CMD_WRITE) ? MSG_MORE : 0);
	if (result <= 0) {
		printk(KERN_ERR "%s: Send control failed (result %d)\n",
				lo->disk->disk_name, result);
		goto error_out;
	}

	if (nbd_cmd(req) == NBD_CMD_WRITE) {
		struct req_iterator iter;
		struct bio_vec *bvec;
		/*
		 * we are really probing at internals to determine
		 * whether to set MSG_MORE or not...
		 */
		rq_for_each_segment(bvec, req, iter) {
			flags = 0;
			if (!rq_iter_last(req, iter))
				flags = MSG_MORE;
			dprintk(DBG_TX, "%s: request %p: sending %d bytes data\n",
					lo->disk->disk_name, req, bvec->bv_len);
			result = sock_send_bvec(lo, bvec, flags);
			if (result <= 0) {
				printk(KERN_ERR "%s: Send data failed (result %d)\n",
						lo->disk->disk_name, result);
				goto error_out;
			}
		}
	}
	return 0;

error_out:
	return 1;
}

static struct request *nbd_find_request(struct nbd_device *lo,
					struct request *xreq)
{
	struct request *req, *tmp;
	int err;

	err = wait_event_interruptible(lo->active_wq, lo->active_req != xreq);
	if (unlikely(err))
		goto out;

	spin_lock(&lo->queue_lock);
	list_for_each_entry_safe(req, tmp, &lo->queue_head, queuelist) {
		if (req != xreq)
			continue;
		list_del_init(&req->queuelist);
		spin_unlock(&lo->queue_lock);
		return req;
	}
	spin_unlock(&lo->queue_lock);

	err = -ENOENT;

out:
	return ERR_PTR(err);
}

static inline int sock_recv_bvec(struct nbd_device *lo, struct bio_vec *bvec)
{
	int result;
	void *kaddr = kmap(bvec->bv_page);
	result = sock_xmit(lo, 0, kaddr + bvec->bv_offset, bvec->bv_len,
			MSG_WAITALL);
	kunmap(bvec->bv_page);
	return result;
}

/* NULL returned = something went wrong, inform userspace */
static struct request *nbd_read_stat(struct nbd_device *lo)
{
	int result;
	struct nbd_reply reply;
	struct request *req;

	reply.magic = 0;
	result = sock_xmit(lo, 0, &reply, sizeof(reply), MSG_WAITALL);
	if (result <= 0) {
		printk(KERN_ERR "%s: Receive control failed (result %d)\n",
				lo->disk->disk_name, result);
		goto harderror;
	}

	if (ntohl(reply.magic) != NBD_REPLY_MAGIC) {
		printk(KERN_ERR "%s: Wrong magic (0x%lx)\n",
				lo->disk->disk_name,
				(unsigned long)ntohl(reply.magic));
		result = -EPROTO;
		goto harderror;
	}

	req = nbd_find_request(lo, *(struct request **)reply.handle);
	if (unlikely(IS_ERR(req))) {
		result = PTR_ERR(req);
		if (result != -ENOENT)
			goto harderror;

		printk(KERN_ERR "%s: Unexpected reply (%p)\n",
				lo->disk->disk_name, reply.handle);
		result = -EBADR;
		goto harderror;
	}

	if (ntohl(reply.error)) {
		printk(KERN_ERR "%s: Other side returned error (%d)\n",
				lo->disk->disk_name, ntohl(reply.error));
		req->errors++;
		return req;
	}

	dprintk(DBG_RX, "%s: request %p: got reply\n",
			lo->disk->disk_name, req);
	if (nbd_cmd(req) == NBD_CMD_READ) {
		struct req_iterator iter;
		struct bio_vec *bvec;

		rq_for_each_segment(bvec, req, iter) {
			result = sock_recv_bvec(lo, bvec);
			if (result <= 0) {
				printk(KERN_ERR "%s: Receive data failed (result %d)\n",
						lo->disk->disk_name, result);
				req->errors++;
				return req;
			}
			dprintk(DBG_RX, "%s: request %p: got %d bytes data\n",
				lo->disk->disk_name, req, bvec->bv_len);
		}