write.c

linux 内核源代码

				BUG();
			if (test_set_page_writeback(page))
				BUG();
			unlock_page(page);
			put_page(page);
		}
		count += loop;
		if (loop < n) {
			for (; loop < n; loop++)
				put_page(pages[loop]);
			goto no_more;
		}

		start += loop;
	} while (start <= wb->last && count < 65536);

no_more:
	/* we now have a contiguous set of dirty pages, each with writeback set
	 * and the dirty mark cleared; the first page is locked and must remain
	 * so, all the rest are unlocked */
	first = primary_page->index;
	last = first + count - 1;

	offset = (first == wb->first) ? wb->offset_first : 0;
	to = (last == wb->last) ? wb->to_last : PAGE_SIZE;

	_debug("write back %lx[%u..] to %lx[..%u]", first, offset, last, to);

	ret = afs_vnode_store_data(wb, first, last, offset, to);
	if (ret < 0) {
		switch (ret) {
		case -EDQUOT:
		case -ENOSPC:
			set_bit(AS_ENOSPC,
				&wb->vnode->vfs_inode.i_mapping->flags);
			break;
		case -EROFS:
		case -EIO:
		case -EREMOTEIO:
		case -EFBIG:
		case -ENOENT:
		case -ENOMEDIUM:
		case -ENXIO:
			afs_kill_pages(wb->vnode, true, first, last);
			set_bit(AS_EIO, &wb->vnode->vfs_inode.i_mapping->flags);
			break;
		case -EACCES:
		case -EPERM:
		case -ENOKEY:
		case -EKEYEXPIRED:
		case -EKEYREJECTED:
		case -EKEYREVOKED:
			afs_kill_pages(wb->vnode, false, first, last);
			break;
		default:
			break;
		}
	} else {
		ret = count;
	}

	_leave(" = %d", ret);
	return ret;
}

/*
 * write a page back to the server
 * - the caller locked the page for us
 */
int afs_writepage(struct page *page, struct writeback_control *wbc)
{
	struct backing_dev_info *bdi = page->mapping->backing_dev_info;
	struct afs_writeback *wb;
	int ret;

	_enter("{%lx},", page->index);

	wb = (struct afs_writeback *) page_private(page);
	ASSERT(wb != NULL);

	ret = afs_write_back_from_locked_page(wb, page);
	unlock_page(page);
	if (ret < 0) {
		_leave(" = %d", ret);
		return 0;
	}

	wbc->nr_to_write -= ret;
	if (wbc->nonblocking && bdi_write_congested(bdi))
		wbc->encountered_congestion = 1;

	_leave(" = 0");
	return 0;
}

/*
 * write a region of pages back to the server
 */
static int afs_writepages_region(struct address_space *mapping,
				 struct writeback_control *wbc,
				 pgoff_t index, pgoff_t end, pgoff_t *_next)
{
	struct backing_dev_info *bdi = mapping->backing_dev_info;
	struct afs_writeback *wb;
	struct page *page;
	int ret, n;

	_enter(",,%lx,%lx,", index, end);

	do {
		n = find_get_pages_tag(mapping, &index, PAGECACHE_TAG_DIRTY,
				       1, &page);
		if (!n)
			break;

		_debug("wback %lx", page->index);

		if (page->index > end) {
			*_next = index;
			page_cache_release(page);
			_leave(" = 0 [%lx]", *_next);
			return 0;
		}

		/* at this point we hold neither mapping->tree_lock nor lock on
		 * the page itself: the page may be truncated or invalidated
		 * (changing page->mapping to NULL), or even swizzled back from
		 * swapper_space to tmpfs file mapping
		 */
		lock_page(page);

		if (page->mapping != mapping) {
			unlock_page(page);
			page_cache_release(page);
			continue;
		}

		if (wbc->sync_mode != WB_SYNC_NONE)
			wait_on_page_writeback(page);

		if (PageWriteback(page) || !PageDirty(page)) {
			unlock_page(page);
			continue;
		}

		wb = (struct afs_writeback *) page_private(page);
		ASSERT(wb != NULL);

		spin_lock(&wb->vnode->writeback_lock);
		wb->state = AFS_WBACK_WRITING;
		spin_unlock(&wb->vnode->writeback_lock);

		ret = afs_write_back_from_locked_page(wb, page);
		unlock_page(page);
		page_cache_release(page);
		if (ret < 0) {
			_leave(" = %d", ret);
			return ret;
		}

		wbc->nr_to_write -= ret;

		if (wbc->nonblocking && bdi_write_congested(bdi)) {
			wbc->encountered_congestion = 1;
			break;
		}

		cond_resched();
	} while (index < end && wbc->nr_to_write > 0);

	*_next = index;
	_leave(" = 0 [%lx]", *_next);
	return 0;
}

/*
 * write some of the pending data back to the server
 */
int afs_writepages(struct address_space *mapping,
		   struct writeback_control *wbc)
{
	struct backing_dev_info *bdi = mapping->backing_dev_info;
	pgoff_t start, end, next;
	int ret;

	_enter("");

	if (wbc->nonblocking && bdi_write_congested(bdi)) {
		wbc->encountered_congestion = 1;
		_leave(" = 0 [congest]");
		return 0;
	}

	if (wbc->range_cyclic) {
		start = mapping->writeback_index;
		end = -1;
		ret = afs_writepages_region(mapping, wbc, start, end, &next);
		if (start > 0 && wbc->nr_to_write > 0 && ret == 0 &&
		    !(wbc->nonblocking && wbc->encountered_congestion))
			ret = afs_writepages_region(mapping, wbc, 0, start,
						    &next);
		mapping->writeback_index = next;
	} else if (wbc->range_start == 0 && wbc->range_end == LLONG_MAX) {
		end = (pgoff_t)(LLONG_MAX >> PAGE_CACHE_SHIFT);
		ret = afs_writepages_region(mapping, wbc, 0, end, &next);
		if (wbc->nr_to_write > 0)
			mapping->writeback_index = next;
	} else {
		start = wbc->range_start >> PAGE_CACHE_SHIFT;
		end = wbc->range_end >> PAGE_CACHE_SHIFT;
		ret = afs_writepages_region(mapping, wbc, start, end, &next);
	}

	_leave(" = %d", ret);
	return ret;
}

/*
 * write an inode back
 */
int afs_write_inode(struct inode *inode, int sync)
{
	struct afs_vnode *vnode = AFS_FS_I(inode);
	int ret;

	_enter("{%x:%u},", vnode->fid.vid, vnode->fid.vnode);

	ret = 0;
	if (sync) {
		ret = filemap_fdatawait(inode->i_mapping);
		if (ret < 0)
			__mark_inode_dirty(inode, I_DIRTY_DATASYNC);
	}

	_leave(" = %d", ret);
	return ret;
}

/*
 * completion of write to server
 */
void afs_pages_written_back(struct afs_vnode *vnode, struct afs_call *call)
{
	struct afs_writeback *wb = call->wb;
	struct pagevec pv;
	unsigned count, loop;
	pgoff_t first = call->first, last = call->last;
	bool free_wb;

	_enter("{%x:%u},{%lx-%lx}",
	       vnode->fid.vid, vnode->fid.vnode, first, last);

	ASSERT(wb != NULL);

	pagevec_init(&pv, 0);

	do {
		_debug("done %lx-%lx", first, last);

		count = last - first + 1;
		if (count > PAGEVEC_SIZE)
			count = PAGEVEC_SIZE;
		pv.nr = find_get_pages_contig(call->mapping, first, count,
					      pv.pages);
		ASSERTCMP(pv.nr, ==, count);

		spin_lock(&vnode->writeback_lock);
		for (loop = 0; loop < count; loop++) {
			struct page *page = pv.pages[loop];
			end_page_writeback(page);
			if (page_private(page) == (unsigned long) wb) {
				set_page_private(page, 0);
				ClearPagePrivate(page);
				wb->usage--;
			}
		}
		free_wb = false;
		if (wb->usage == 0) {
			afs_unlink_writeback(wb);
			free_wb = true;
		}
		spin_unlock(&vnode->writeback_lock);
		first += count;
		if (free_wb) {
			afs_free_writeback(wb);
			wb = NULL;
		}

		__pagevec_release(&pv);
	} while (first <= last);

	_leave("");
}

/*
 * write to an AFS file
 */
ssize_t afs_file_write(struct kiocb *iocb, const struct iovec *iov,
		       unsigned long nr_segs, loff_t pos)
{
	struct dentry *dentry = iocb->ki_filp->f_path.dentry;
	struct afs_vnode *vnode = AFS_FS_I(dentry->d_inode);
	ssize_t result;
	size_t count = iov_length(iov, nr_segs);
	int ret;

	_enter("{%x.%u},{%zu},%lu,",
	       vnode->fid.vid, vnode->fid.vnode, count, nr_segs);

	if (IS_SWAPFILE(&vnode->vfs_inode)) {
		printk(KERN_INFO
		       "AFS: Attempt to write to active swap file!\n");
		return -EBUSY;
	}

	if (!count)
		return 0;

	result = generic_file_aio_write(iocb, iov, nr_segs, pos);
	if (IS_ERR_VALUE(result)) {
		_leave(" = %zd", result);
		return result;
	}

	/* return error values for O_SYNC and IS_SYNC() */
	if (IS_SYNC(&vnode->vfs_inode) || iocb->ki_filp->f_flags & O_SYNC) {
		ret = afs_fsync(iocb->ki_filp, dentry, 1);
		if (ret < 0)
			result = ret;
	}

	_leave(" = %zd", result);
	return result;
}

/*
 * flush the vnode to the fileserver
 */
int afs_writeback_all(struct afs_vnode *vnode)
{
	struct address_space *mapping = vnode->vfs_inode.i_mapping;
	struct writeback_control wbc = {
		.bdi		= mapping->backing_dev_info,
		.sync_mode	= WB_SYNC_ALL,
		.nr_to_write	= LONG_MAX,
		.for_writepages = 1,
		.range_cyclic	= 1,
	};
	int ret;

	_enter("");

	ret = mapping->a_ops->writepages(mapping, &wbc);
	__mark_inode_dirty(mapping->host, I_DIRTY_PAGES);

	_leave(" = %d", ret);
	return ret;
}

/*
 * flush any dirty pages for this process, and check for write errors.
 * - the return status from this call provides a reliable indication of
 *   whether any write errors occurred for this process.
 */
int afs_fsync(struct file *file, struct dentry *dentry, int datasync)
{
	struct afs_writeback *wb, *xwb;
	struct afs_vnode *vnode = AFS_FS_I(dentry->d_inode);
	int ret;

	_enter("{%x:%u},{n=%s},%d",
	       vnode->fid.vid, vnode->fid.vnode, dentry->d_name.name,
	       datasync);

	/* use a writeback record as a marker in the queue - when this reaches
	 * the front of the queue, all the outstanding writes are either
	 * completed or rejected */
	wb = kzalloc(sizeof(*wb), GFP_KERNEL);
	if (!wb)
		return -ENOMEM;
	wb->vnode = vnode;
	wb->first = 0;
	wb->last = -1;
	wb->offset_first = 0;
	wb->to_last = PAGE_SIZE;
	wb->usage = 1;
	wb->state = AFS_WBACK_SYNCING;
	init_waitqueue_head(&wb->waitq);

	spin_lock(&vnode->writeback_lock);
	list_for_each_entry(xwb, &vnode->writebacks, link) {
		if (xwb->state == AFS_WBACK_PENDING)
			xwb->state = AFS_WBACK_CONFLICTING;
	}
	list_add_tail(&wb->link, &vnode->writebacks);
	spin_unlock(&vnode->writeback_lock);

	/* push all the outstanding writebacks to the server */
	ret = afs_writeback_all(vnode);
	if (ret < 0) {
		afs_put_writeback(wb);
		_leave(" = %d [wb]", ret);
		return ret;
	}

	/* wait for the preceding writes to actually complete */
	ret = wait_event_interruptible(wb->waitq,
				       wb->state == AFS_WBACK_COMPLETE ||
				       vnode->writebacks.next == &wb->link);
	afs_put_writeback(wb);
	_leave(" = %d", ret);
	return ret;
}