write.c

linux 内核源代码

			req->wb_context->path.dentry->d_inode->i_sb->s_id,
			(long long)NFS_FILEID(req->wb_context->path.dentry->d_inode),
			req->wb_bytes,
			(long long)req_offset(req));

		if (task->tk_status < 0) {
			nfs_set_pageerror(page);
			nfs_context_set_write_error(req->wb_context, task->tk_status);
			dprintk(", error = %d\n", task->tk_status);
			goto remove_request;
		}

		if (nfs_write_need_commit(data)) {
			memcpy(&req->wb_verf, &data->verf, sizeof(req->wb_verf));
			nfs_mark_request_commit(req);
			nfs_end_page_writeback(page);
			dprintk(" marked for commit\n");
			goto next;
		}
		/* Set the PG_uptodate flag? */
		nfs_mark_uptodate(page, req->wb_pgbase, req->wb_bytes);
		dprintk(" OK\n");
remove_request:
		nfs_end_page_writeback(page);
		nfs_inode_remove_request(req);
	next:
		nfs_clear_page_tag_locked(req);
	}
}

static const struct rpc_call_ops nfs_write_full_ops = {
	.rpc_call_done = nfs_writeback_done_full,
	.rpc_release = nfs_writedata_release,
};


/*
 * This function is called when the WRITE call is complete.
 */
int nfs_writeback_done(struct rpc_task *task, struct nfs_write_data *data)
{
	struct nfs_writeargs	*argp = &data->args;
	struct nfs_writeres	*resp = &data->res;
	int status;

	dprintk("NFS: %5u nfs_writeback_done (status %d)\n",
		task->tk_pid, task->tk_status);

	/*
	 * ->write_done will attempt to use post-op attributes to detect
	 * conflicting writes by other clients.  A strict interpretation
	 * of close-to-open would allow us to continue caching even if
	 * another writer had changed the file, but some applications
	 * depend on tighter cache coherency when writing.
	 */
	status = NFS_PROTO(data->inode)->write_done(task, data);
	if (status != 0)
		return status;
	nfs_add_stats(data->inode, NFSIOS_SERVERWRITTENBYTES, resp->count);

#if defined(CONFIG_NFS_V3) || defined(CONFIG_NFS_V4)
	if (resp->verf->committed < argp->stable && task->tk_status >= 0) {
		/* We tried a write call, but the server did not
		 * commit data to stable storage even though we
		 * requested it.
		 * Note: There is a known bug in Tru64 < 5.0 in which
		 *	 the server reports NFS_DATA_SYNC, but performs
		 *	 NFS_FILE_SYNC. We therefore implement this checking
		 *	 as a dprintk() in order to avoid filling syslog.
		 */
		static unsigned long    complain;

		if (time_before(complain, jiffies)) {
			dprintk("NFS: faulty NFS server %s:"
				" (committed = %d) != (stable = %d)\n",
				NFS_SERVER(data->inode)->nfs_client->cl_hostname,
				resp->verf->committed, argp->stable);
			complain = jiffies + 300 * HZ;
		}
	}
#endif
	/* Is this a short write? */
	if (task->tk_status >= 0 && resp->count < argp->count) {
		static unsigned long    complain;

		nfs_inc_stats(data->inode, NFSIOS_SHORTWRITE);

		/* Has the server at least made some progress? */
		if (resp->count != 0) {
			/* Was this an NFSv2 write or an NFSv3 stable write? */
			if (resp->verf->committed != NFS_UNSTABLE) {
				/* Resend from where the server left off */
				argp->offset += resp->count;
				argp->pgbase += resp->count;
				argp->count -= resp->count;
			} else {
				/* Resend as a stable write in order to avoid
				 * headaches in the case of a server crash.
				 */
				argp->stable = NFS_FILE_SYNC;
			}
			rpc_restart_call(task);
			return -EAGAIN;
		}
		if (time_before(complain, jiffies)) {
			printk(KERN_WARNING
			       "NFS: Server wrote zero bytes, expected %u.\n",
					argp->count);
			complain = jiffies + 300 * HZ;
		}
		/* Can't do anything about it except throw an error. */
		task->tk_status = -EIO;
	}
	return 0;
}


#if defined(CONFIG_NFS_V3) || defined(CONFIG_NFS_V4)
void nfs_commit_release(void *wdata)
{
	nfs_commit_free(wdata);
}

/*
 * Set up the argument/result storage required for the RPC call.
 */
static void nfs_commit_rpcsetup(struct list_head *head,
		struct nfs_write_data *data,
		int how)
{
	struct nfs_page		*first;
	struct inode		*inode;
	int flags;

	/* Set up the RPC argument and reply structs
	 * NB: take care not to mess about with data->commit et al. */

	list_splice_init(head, &data->pages);
	first = nfs_list_entry(data->pages.next);
	inode = first->wb_context->path.dentry->d_inode;

	data->inode	  = inode;
	data->cred	  = first->wb_context->cred;

	data->args.fh     = NFS_FH(data->inode);
	/* Note: we always request a commit of the entire inode */
	data->args.offset = 0;
	data->args.count  = 0;
	data->res.count   = 0;
	data->res.fattr   = &data->fattr;
	data->res.verf    = &data->verf;
	nfs_fattr_init(&data->fattr);

	/* Set up the initial task struct.  */
	flags = (how & FLUSH_SYNC) ? 0 : RPC_TASK_ASYNC;
	rpc_init_task(&data->task, NFS_CLIENT(inode), flags, &nfs_commit_ops, data);
	NFS_PROTO(inode)->commit_setup(data, how);

	data->task.tk_priority = flush_task_priority(how);
	data->task.tk_cookie = (unsigned long)inode;
	
	dprintk("NFS: %5u initiated commit call\n", data->task.tk_pid);
}

/*
 * Commit dirty pages
 */
static int
nfs_commit_list(struct inode *inode, struct list_head *head, int how)
{
	struct nfs_write_data	*data;
	struct nfs_page         *req;

	data = nfs_commit_alloc();

	if (!data)
		goto out_bad;

	/* Set up the argument struct */
	nfs_commit_rpcsetup(head, data, how);

	nfs_execute_write(data);
	return 0;
 out_bad:
	while (!list_empty(head)) {
		req = nfs_list_entry(head->next);
		nfs_list_remove_request(req);
		nfs_mark_request_commit(req);
		dec_zone_page_state(req->wb_page, NR_UNSTABLE_NFS);
		dec_bdi_stat(req->wb_page->mapping->backing_dev_info,
				BDI_RECLAIMABLE);
		nfs_clear_page_tag_locked(req);
	}
	return -ENOMEM;
}

/*
 * COMMIT call returned
 */
static void nfs_commit_done(struct rpc_task *task, void *calldata)
{
	struct nfs_write_data	*data = calldata;
	struct nfs_page		*req;

        dprintk("NFS: %5u nfs_commit_done (status %d)\n",
                                task->tk_pid, task->tk_status);

	/* Call the NFS version-specific code */
	if (NFS_PROTO(data->inode)->commit_done(task, data) != 0)
		return;

	while (!list_empty(&data->pages)) {
		req = nfs_list_entry(data->pages.next);
		nfs_list_remove_request(req);
		clear_bit(PG_NEED_COMMIT, &(req)->wb_flags);
		dec_zone_page_state(req->wb_page, NR_UNSTABLE_NFS);
		dec_bdi_stat(req->wb_page->mapping->backing_dev_info,
				BDI_RECLAIMABLE);

		dprintk("NFS: commit (%s/%Ld %d@%Ld)",
			req->wb_context->path.dentry->d_inode->i_sb->s_id,
			(long long)NFS_FILEID(req->wb_context->path.dentry->d_inode),
			req->wb_bytes,
			(long long)req_offset(req));
		if (task->tk_status < 0) {
			nfs_context_set_write_error(req->wb_context, task->tk_status);
			nfs_inode_remove_request(req);
			dprintk(", error = %d\n", task->tk_status);
			goto next;
		}

		/* Okay, COMMIT succeeded, apparently. Check the verifier
		 * returned by the server against all stored verfs. */
		if (!memcmp(req->wb_verf.verifier, data->verf.verifier, sizeof(data->verf.verifier))) {
			/* We have a match */
			/* Set the PG_uptodate flag */
			nfs_mark_uptodate(req->wb_page, req->wb_pgbase,
					req->wb_bytes);
			nfs_inode_remove_request(req);
			dprintk(" OK\n");
			goto next;
		}
		/* We have a mismatch. Write the page again */
		dprintk(" mismatch\n");
		nfs_redirty_request(req);
	next:
		nfs_clear_page_tag_locked(req);
	}
}

static const struct rpc_call_ops nfs_commit_ops = {
	.rpc_call_done = nfs_commit_done,
	.rpc_release = nfs_commit_release,
};

int nfs_commit_inode(struct inode *inode, int how)
{
	LIST_HEAD(head);
	int res;

	spin_lock(&inode->i_lock);
	res = nfs_scan_commit(inode, &head, 0, 0);
	spin_unlock(&inode->i_lock);
	if (res) {
		int error = nfs_commit_list(inode, &head, how);
		if (error < 0)
			return error;
	}
	return res;
}
#else
static inline int nfs_commit_list(struct inode *inode, struct list_head *head, int how)
{
	return 0;
}
#endif

long nfs_sync_mapping_wait(struct address_space *mapping, struct writeback_control *wbc, int how)
{
	struct inode *inode = mapping->host;
	pgoff_t idx_start, idx_end;
	unsigned int npages = 0;
	LIST_HEAD(head);
	int nocommit = how & FLUSH_NOCOMMIT;
	long pages, ret;

	/* FIXME */
	if (wbc->range_cyclic)
		idx_start = 0;
	else {
		idx_start = wbc->range_start >> PAGE_CACHE_SHIFT;
		idx_end = wbc->range_end >> PAGE_CACHE_SHIFT;
		if (idx_end > idx_start) {
			pgoff_t l_npages = 1 + idx_end - idx_start;
			npages = l_npages;
			if (sizeof(npages) != sizeof(l_npages) &&
					(pgoff_t)npages != l_npages)
				npages = 0;
		}
	}
	how &= ~FLUSH_NOCOMMIT;
	spin_lock(&inode->i_lock);
	do {
		ret = nfs_wait_on_requests_locked(inode, idx_start, npages);
		if (ret != 0)
			continue;
		if (nocommit)
			break;
		pages = nfs_scan_commit(inode, &head, idx_start, npages);
		if (pages == 0)
			break;
		if (how & FLUSH_INVALIDATE) {
			spin_unlock(&inode->i_lock);
			nfs_cancel_commit_list(&head);
			ret = pages;
			spin_lock(&inode->i_lock);
			continue;
		}
		pages += nfs_scan_commit(inode, &head, 0, 0);
		spin_unlock(&inode->i_lock);
		ret = nfs_commit_list(inode, &head, how);
		spin_lock(&inode->i_lock);

	} while (ret >= 0);
	spin_unlock(&inode->i_lock);
	return ret;
}

static int __nfs_write_mapping(struct address_space *mapping, struct writeback_control *wbc, int how)
{
	int ret;

	ret = nfs_writepages(mapping, wbc);
	if (ret < 0)
		goto out;
	ret = nfs_sync_mapping_wait(mapping, wbc, how);
	if (ret < 0)
		goto out;
	return 0;
out:
	__mark_inode_dirty(mapping->host, I_DIRTY_PAGES);
	return ret;
}

/* Two pass sync: first using WB_SYNC_NONE, then WB_SYNC_ALL */
static int nfs_write_mapping(struct address_space *mapping, int how)
{
	struct writeback_control wbc = {
		.bdi = mapping->backing_dev_info,
		.sync_mode = WB_SYNC_NONE,
		.nr_to_write = LONG_MAX,
		.for_writepages = 1,
		.range_cyclic = 1,
	};
	int ret;

	ret = __nfs_write_mapping(mapping, &wbc, how);
	if (ret < 0)
		return ret;
	wbc.sync_mode = WB_SYNC_ALL;
	return __nfs_write_mapping(mapping, &wbc, how);
}

/*
 * flush the inode to disk.
 */
int nfs_wb_all(struct inode *inode)
{
	return nfs_write_mapping(inode->i_mapping, 0);
}

int nfs_wb_nocommit(struct inode *inode)
{
	return nfs_write_mapping(inode->i_mapping, FLUSH_NOCOMMIT);
}

int nfs_wb_page_cancel(struct inode *inode, struct page *page)
{
	struct nfs_page *req;
	loff_t range_start = page_offset(page);
	loff_t range_end = range_start + (loff_t)(PAGE_CACHE_SIZE - 1);
	struct writeback_control wbc = {
		.bdi = page->mapping->backing_dev_info,
		.sync_mode = WB_SYNC_ALL,
		.nr_to_write = LONG_MAX,
		.range_start = range_start,
		.range_end = range_end,
	};
	int ret = 0;

	BUG_ON(!PageLocked(page));
	for (;;) {
		req = nfs_page_find_request(page);
		if (req == NULL)
			goto out;
		if (test_bit(PG_NEED_COMMIT, &req->wb_flags)) {
			nfs_release_request(req);
			break;
		}
		if (nfs_lock_request_dontget(req)) {
			nfs_inode_remove_request(req);
			/*
			 * In case nfs_inode_remove_request has marked the
			 * page as being dirty
			 */
			cancel_dirty_page(page, PAGE_CACHE_SIZE);
			nfs_unlock_request(req);
			break;
		}
		ret = nfs_wait_on_request(req);
		if (ret < 0)
			goto out;
	}
	if (!PagePrivate(page))
		return 0;
	ret = nfs_sync_mapping_wait(page->mapping, &wbc, FLUSH_INVALIDATE);
out:
	return ret;
}

static int nfs_wb_page_priority(struct inode *inode, struct page *page,
				int how)
{
	loff_t range_start = page_offset(page);
	loff_t range_end = range_start + (loff_t)(PAGE_CACHE_SIZE - 1);
	struct writeback_control wbc = {
		.bdi = page->mapping->backing_dev_info,
		.sync_mode = WB_SYNC_ALL,
		.nr_to_write = LONG_MAX,
		.range_start = range_start,
		.range_end = range_end,
	};
	int ret;

	BUG_ON(!PageLocked(page));
	if (clear_page_dirty_for_io(page)) {
		ret = nfs_writepage_locked(page, &wbc);
		if (ret < 0)
			goto out;
	}
	if (!PagePrivate(page))
		return 0;
	ret = nfs_sync_mapping_wait(page->mapping, &wbc, how);
	if (ret >= 0)
		return 0;
out:
	__mark_inode_dirty(inode, I_DIRTY_PAGES);
	return ret;
}

/*
 * Write back all requests on one page - we do this before reading it.
 */
int nfs_wb_page(struct inode *inode, struct page* page)
{
	return nfs_wb_page_priority(inode, page, FLUSH_STABLE);
}

int __init nfs_init_writepagecache(void)
{
	nfs_wdata_cachep = kmem_cache_create("nfs_write_data",
					     sizeof(struct nfs_write_data),
					     0, SLAB_HWCACHE_ALIGN,
					     NULL);
	if (nfs_wdata_cachep == NULL)
		return -ENOMEM;

	nfs_wdata_mempool = mempool_create_slab_pool(MIN_POOL_WRITE,
						     nfs_wdata_cachep);
	if (nfs_wdata_mempool == NULL)
		return -ENOMEM;

	nfs_commit_mempool = mempool_create_slab_pool(MIN_POOL_COMMIT,
						      nfs_wdata_cachep);
	if (nfs_commit_mempool == NULL)
		return -ENOMEM;

	/*
	 * NFS congestion size, scale with available memory.
	 *
	 *  64MB:    8192k
	 * 128MB:   11585k
	 * 256MB:   16384k
	 * 512MB:   23170k
	 *   1GB:   32768k
	 *   2GB:   46340k
	 *   4GB:   65536k
	 *   8GB:   92681k
	 *  16GB:  131072k
	 *
	 * This allows larger machines to have larger/more transfers.
	 * Limit the default to 256M
	 */
	nfs_congestion_kb = (16*int_sqrt(totalram_pages)) << (PAGE_SHIFT-10);
	if (nfs_congestion_kb > 256*1024)
		nfs_congestion_kb = 256*1024;

	return 0;
}

void nfs_destroy_writepagecache(void)
{
	mempool_destroy(nfs_commit_mempool);
	mempool_destroy(nfs_wdata_mempool);
	kmem_cache_destroy(nfs_wdata_cachep);
}