#include <linux/pagevec.h>
#include <linux/writeback.h>
+
+/*
+ * Prime number of hash buckets since address is used as the key.
+ */
+#define NVSYNC 37
+#define to_ioend_wq(v) (&xfs_ioend_wq[((unsigned long)v) % NVSYNC])
+static wait_queue_head_t xfs_ioend_wq[NVSYNC];
+
+void __init
+xfs_ioend_init(void)
+{
+ int i;
+
+ for (i = 0; i < NVSYNC; i++)
+ init_waitqueue_head(&xfs_ioend_wq[i]);
+}
+
+void
+xfs_ioend_wait(
+ xfs_inode_t *ip)
+{
+ wait_queue_head_t *wq = to_ioend_wq(ip);
+
+ wait_event(*wq, (atomic_read(&ip->i_iocount) == 0));
+}
+
+STATIC void
+xfs_ioend_wake(
+ xfs_inode_t *ip)
+{
+ if (atomic_dec_and_test(&ip->i_iocount))
+ wake_up(to_ioend_wq(ip));
+}
+
STATIC void
xfs_count_page_state(
struct page *page,
unsigned long pgoff)
{
xfs_inode_t *ip;
- bhv_vnode_t *vp = vn_from_inode(inode);
loff_t isize = i_size_read(inode);
loff_t offset = page_offset(page);
int delalloc = -1, unmapped = -1, unwritten = -1;
if (page_has_buffers(page))
xfs_count_page_state(page, &delalloc, &unmapped, &unwritten);
- ip = xfs_vtoi(vp);
+ ip = XFS_I(inode);
if (!ip->i_rwtrace)
return;
}
/*
- * Schedule IO completion handling on a xfsdatad if this was
- * the final hold on this ioend. If we are asked to wait,
- * flush the workqueue.
- */
-STATIC void
-xfs_finish_ioend(
- xfs_ioend_t *ioend,
- int wait)
-{
- if (atomic_dec_and_test(&ioend->io_remaining)) {
- queue_work(xfsdatad_workqueue, &ioend->io_work);
- if (wait)
- flush_workqueue(xfsdatad_workqueue);
- }
-}
-
-/*
* We're now finished for good with this ioend structure.
* Update the page state via the associated buffer_heads,
* release holds on the inode and bio, and finally free
xfs_ioend_t *ioend)
{
struct buffer_head *bh, *next;
+ struct xfs_inode *ip = XFS_I(ioend->io_inode);
for (bh = ioend->io_buffer_head; bh; bh = next) {
next = bh->b_private;
bh->b_end_io(bh, !ioend->io_error);
}
- if (unlikely(ioend->io_error)) {
- vn_ioerror(XFS_I(ioend->io_inode), ioend->io_error,
- __FILE__,__LINE__);
+
+ /*
+ * Volume managers supporting multiple paths can send back ENODEV
+ * when the final path disappears. In this case continuing to fill
+ * the page cache with dirty data which cannot be written out is
+ * evil, so prevent that.
+ */
+ if (unlikely(ioend->io_error == -ENODEV)) {
+ xfs_do_force_shutdown(ip->i_mount, SHUTDOWN_DEVICE_REQ,
+ __FILE__, __LINE__);
}
- vn_iowake(XFS_I(ioend->io_inode));
+
+ xfs_ioend_wake(ip);
mempool_free(ioend, xfs_ioend_pool);
}
/*
+ * If the end of the current ioend is beyond the current EOF,
+ * return the new EOF value, otherwise zero.
+ */
+STATIC xfs_fsize_t
+xfs_ioend_new_eof(
+ xfs_ioend_t *ioend)
+{
+ xfs_inode_t *ip = XFS_I(ioend->io_inode);
+ xfs_fsize_t isize;
+ xfs_fsize_t bsize;
+
+ bsize = ioend->io_offset + ioend->io_size;
+ isize = MAX(ip->i_size, ip->i_new_size);
+ isize = MIN(isize, bsize);
+ return isize > ip->i_d.di_size ? isize : 0;
+}
+
+/*
* Update on-disk file size now that data has been written to disk.
* The current in-memory file size is i_size. If a write is beyond
* eof i_new_size will be the intended file size until i_size is
* updated. If this write does not extend all the way to the valid
* file size then restrict this update to the end of the write.
*/
+
STATIC void
xfs_setfilesize(
xfs_ioend_t *ioend)
{
xfs_inode_t *ip = XFS_I(ioend->io_inode);
xfs_fsize_t isize;
- xfs_fsize_t bsize;
ASSERT((ip->i_d.di_mode & S_IFMT) == S_IFREG);
ASSERT(ioend->io_type != IOMAP_READ);
if (unlikely(ioend->io_error))
return;
- bsize = ioend->io_offset + ioend->io_size;
-
xfs_ilock(ip, XFS_ILOCK_EXCL);
-
- isize = MAX(ip->i_size, ip->i_new_size);
- isize = MIN(isize, bsize);
-
- if (ip->i_d.di_size < isize) {
+ isize = xfs_ioend_new_eof(ioend);
+ if (isize) {
ip->i_d.di_size = isize;
- ip->i_update_core = 1;
- ip->i_update_size = 1;
- mark_inode_dirty_sync(ioend->io_inode);
+ xfs_mark_inode_dirty_sync(ip);
}
xfs_iunlock(ip, XFS_ILOCK_EXCL);
}
/*
+ * Schedule IO completion handling on a xfsdatad if this was
+ * the final hold on this ioend. If we are asked to wait,
+ * flush the workqueue.
+ */
+STATIC void
+xfs_finish_ioend(
+ xfs_ioend_t *ioend,
+ int wait)
+{
+ if (atomic_dec_and_test(&ioend->io_remaining)) {
+ struct workqueue_struct *wq = xfsdatad_workqueue;
+ if (ioend->io_work.func == xfs_end_bio_unwritten)
+ wq = xfsconvertd_workqueue;
+
+ queue_work(wq, &ioend->io_work);
+ if (wait)
+ flush_workqueue(wq);
+ }
+}
+
+/*
* Allocate and initialise an IO completion structure.
* We need to track unwritten extent write completion here initially.
* We'll need to extend this for updating the ondisk inode size later
xfs_iomap_t *mapp,
int flags)
{
- xfs_inode_t *ip = XFS_I(inode);
- int error, nmaps = 1;
-
- error = xfs_iomap(ip, offset, count,
- flags, mapp, &nmaps);
- if (!error && (flags & (BMAPI_WRITE|BMAPI_ALLOCATE)))
- xfs_iflags_set(ip, XFS_IMODIFIED);
- return -error;
+ int nmaps = 1;
+
+ return -xfs_iomap(XFS_I(inode), offset, count, flags, mapp, &nmaps);
}
STATIC_INLINE int
struct bio *bio)
{
atomic_inc(&ioend->io_remaining);
-
bio->bi_private = ioend;
bio->bi_end_io = xfs_end_bio;
+ /*
+ * If the I/O is beyond EOF we mark the inode dirty immediately
+ * but don't update the inode size until I/O completion.
+ */
+ if (xfs_ioend_new_eof(ioend))
+ xfs_mark_inode_dirty_sync(XFS_I(ioend->io_inode));
+
submit_bio(WRITE, bio);
ASSERT(!bio_flagged(bio, BIO_EOPNOTSUPP));
bio_put(bio);
unlock_buffer(bh);
} while ((bh = next_bh) != NULL);
- vn_iowake(XFS_I(ioend->io_inode));
+ xfs_ioend_wake(XFS_I(ioend->io_inode));
mempool_free(ioend, xfs_ioend_pool);
} while ((ioend = next) != NULL);
}
} else
pg_offset = PAGE_CACHE_SIZE;
- if (page->index == tindex && !TestSetPageLocked(page)) {
+ if (page->index == tindex && trylock_page(page)) {
pg_len = xfs_probe_page(page, pg_offset, mapped);
unlock_page(page);
}
if (page->index != tindex)
goto fail;
- if (TestSetPageLocked(page))
+ if (!trylock_page(page))
goto fail;
if (PageWriteback(page))
goto fail_unlock_page;
* that we are writing into for the first time.
*/
type = IOMAP_NEW;
- if (!test_and_set_bit(BH_Lock, &bh->b_state)) {
+ if (trylock_buffer(bh)) {
ASSERT(buffer_mapped(bh));
if (iomap_valid)
all_bh = 1;
if (!page_has_buffers(page))
create_empty_buffers(page, 1 << inode->i_blkbits, 0);
+
+ /*
+ * VM calculation for nr_to_write seems off. Bump it way
+ * up, this gets simple streaming writes zippy again.
+ * To be reviewed again after Jens' writeback changes.
+ */
+ wbc->nr_to_write *= 4;
+
/*
* Convert delayed allocate, unwritten or unmapped space
* to real space and flush out to disk.
offset = (xfs_off_t)iblock << inode->i_blkbits;
ASSERT(bh_result->b_size >= (1 << inode->i_blkbits));
size = bh_result->b_size;
+
+ if (!create && direct && offset >= i_size_read(inode))
+ return 0;
+
error = xfs_iomap(XFS_I(inode), offset, size,
create ? flags : BMAPI_READ, &iomap, &niomap);
if (error)
.bmap = xfs_vm_bmap,
.direct_IO = xfs_vm_direct_IO,
.migratepage = buffer_migrate_page,
+ .is_partially_uptodate = block_is_partially_uptodate,
+ .error_remove_page = generic_error_remove_page,
};