#include "xfs_error.h"
#include "xfs_rw.h"
#include "xfs_iomap.h"
+#include "xfs_vnodeops.h"
#include <linux/mpage.h>
#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;
#define xfs_page_trace(tag, inode, page, pgoff)
#endif
-/*
- * 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)
+STATIC struct block_device *
+xfs_find_bdev_for_inode(
+ struct xfs_inode *ip)
{
- if (atomic_dec_and_test(&ioend->io_remaining)) {
- queue_work(xfsdatad_workqueue, &ioend->io_work);
- if (wait)
- flush_workqueue(xfsdatad_workqueue);
- }
+ struct xfs_mount *mp = ip->i_mount;
+
+ if (XFS_IS_REALTIME_INODE(ip))
+ return mp->m_rtdev_targp->bt_bdev;
+ else
+ return mp->m_ddev_targp->bt_bdev;
}
/*
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(ioend->io_vnode, ioend->io_error, __FILE__,__LINE__);
- vn_iowake(ioend->io_vnode);
+
+ /*
+ * 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__);
+ }
+
+ 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 io_new_size will be the intended file size until i_size is
+ * 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_inode_t *ip = XFS_I(ioend->io_inode);
xfs_fsize_t isize;
- xfs_fsize_t bsize;
-
- ip = xfs_vtoi(ioend->io_vnode);
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_iocore.io_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;
+ xfs_mark_inode_dirty_sync(ip);
}
xfs_iunlock(ip, XFS_ILOCK_EXCL);
{
xfs_ioend_t *ioend =
container_of(work, xfs_ioend_t, io_work);
- bhv_vnode_t *vp = ioend->io_vnode;
+ struct xfs_inode *ip = XFS_I(ioend->io_inode);
xfs_off_t offset = ioend->io_offset;
size_t size = ioend->io_size;
if (likely(!ioend->io_error)) {
- bhv_vop_bmap(vp, offset, size, BMAPI_UNWRITTEN, NULL, NULL);
+ if (!XFS_FORCED_SHUTDOWN(ip->i_mount)) {
+ int error;
+ error = xfs_iomap_write_unwritten(ip, offset, size);
+ if (error)
+ ioend->io_error = error;
+ }
xfs_setfilesize(ioend);
}
xfs_destroy_ioend(ioend);
}
/*
+ * 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
ioend->io_error = 0;
ioend->io_list = NULL;
ioend->io_type = type;
- ioend->io_vnode = vn_from_inode(inode);
+ ioend->io_inode = inode;
ioend->io_buffer_head = NULL;
ioend->io_buffer_tail = NULL;
- atomic_inc(&ioend->io_vnode->v_iocount);
+ atomic_inc(&XFS_I(ioend->io_inode)->i_iocount);
ioend->io_offset = 0;
ioend->io_size = 0;
xfs_iomap_t *mapp,
int flags)
{
- bhv_vnode_t *vp = vn_from_inode(inode);
- int error, nmaps = 1;
+ int nmaps = 1;
- error = bhv_vop_bmap(vp, offset, count, flags, mapp, &nmaps);
- if (!error && (flags & (BMAPI_WRITE|BMAPI_ALLOCATE)))
- VMODIFY(vp);
- return -error;
+ return -xfs_iomap(XFS_I(inode), offset, count, flags, mapp, &nmaps);
}
STATIC_INLINE int
/*
* BIO completion handler for buffered IO.
*/
-STATIC int
+STATIC void
xfs_end_bio(
struct bio *bio,
- unsigned int bytes_done,
int error)
{
xfs_ioend_t *ioend = bio->bi_private;
- if (bio->bi_size)
- return 1;
-
ASSERT(atomic_read(&bio->bi_cnt) >= 1);
ioend->io_error = test_bit(BIO_UPTODATE, &bio->bi_flags) ? 0 : error;
bio_put(bio);
xfs_finish_ioend(ioend, 0);
- return 0;
}
STATIC void
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);
STATIC void
xfs_start_page_writeback(
struct page *page,
- struct writeback_control *wbc,
int clear_dirty,
int buffers)
{
clear_page_dirty_for_io(page);
set_page_writeback(page);
unlock_page(page);
- if (!buffers) {
+ /* If no buffers on the page are to be written, finish it here */
+ if (!buffers)
end_page_writeback(page);
- wbc->pages_skipped++; /* We didn't write this page */
- }
}
static inline int bio_add_buffer(struct bio *bio, struct buffer_head *bh)
unlock_buffer(bh);
} while ((bh = next_bh) != NULL);
- vn_iowake(ioend->io_vnode);
+ xfs_ioend_wake(XFS_I(ioend->io_inode));
mempool_free(ioend, xfs_ioend_pool);
} while ((ioend = next) != NULL);
}
for (i = 0; i < pagevec_count(&pvec); i++) {
struct page *page = pvec.pages[i];
- size_t pg_offset, len = 0;
+ size_t pg_offset, pg_len = 0;
if (tindex == tlast) {
pg_offset =
} else
pg_offset = PAGE_CACHE_SIZE;
- if (page->index == tindex && !TestSetPageLocked(page)) {
- len = xfs_probe_page(page, pg_offset, mapped);
+ if (page->index == tindex && trylock_page(page)) {
+ pg_len = xfs_probe_page(page, pg_offset, mapped);
unlock_page(page);
}
- if (!len) {
+ if (!pg_len) {
done = 1;
break;
}
- total += len;
+ total += pg_len;
tindex++;
}
if (page->index != tindex)
goto fail;
- if (TestSetPageLocked(page))
+ if (!trylock_page(page))
goto fail;
if (PageWriteback(page))
goto fail_unlock_page;
done = 1;
}
}
- xfs_start_page_writeback(page, wbc, !page_dirty, count);
+ xfs_start_page_writeback(page, !page_dirty, count);
}
return done;
if (buffer_unwritten(bh) || buffer_delay(bh) ||
((buffer_uptodate(bh) || PageUptodate(page)) &&
!buffer_mapped(bh) && (unmapped || startio))) {
+ int new_ioend = 0;
+
/*
* Make sure we don't use a read-only iomap
*/
}
if (!iomap_valid) {
+ /*
+ * if we didn't have a valid mapping then we
+ * need to ensure that we put the new mapping
+ * in a new ioend structure. This needs to be
+ * done to ensure that the ioends correctly
+ * reflect the block mappings at io completion
+ * for unwritten extent conversion.
+ */
+ new_ioend = 1;
if (type == IOMAP_NEW) {
size = xfs_probe_cluster(inode,
page, bh, head, 0);
if (startio) {
xfs_add_to_ioend(inode, bh, offset,
type, &ioend,
- !iomap_valid);
+ new_ioend);
} else {
set_buffer_dirty(bh);
unlock_buffer(bh);
* 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;
SetPageUptodate(page);
if (startio)
- xfs_start_page_writeback(page, wbc, 1, count);
+ xfs_start_page_writeback(page, 1, count);
if (ioend && iomap_valid) {
offset = (iomap.iomap_offset + iomap.iomap_bsize - 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.
struct address_space *mapping,
struct writeback_control *wbc)
{
- struct bhv_vnode *vp = vn_from_inode(mapping->host);
-
- if (VN_TRUNC(vp))
- VUNTRUNCATE(vp);
+ xfs_iflags_clear(XFS_I(mapping->host), XFS_ITRUNCATED);
return generic_writepages(mapping, wbc);
}
int direct,
bmapi_flags_t flags)
{
- bhv_vnode_t *vp = vn_from_inode(inode);
xfs_iomap_t iomap;
xfs_off_t offset;
ssize_t size;
offset = (xfs_off_t)iblock << inode->i_blkbits;
ASSERT(bh_result->b_size >= (1 << inode->i_blkbits));
size = bh_result->b_size;
- error = bhv_vop_bmap(vp, offset, 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)
return -error;
{
struct file *file = iocb->ki_filp;
struct inode *inode = file->f_mapping->host;
- bhv_vnode_t *vp = vn_from_inode(inode);
- xfs_iomap_t iomap;
- int maps = 1;
- int error;
+ struct block_device *bdev;
ssize_t ret;
- error = bhv_vop_bmap(vp, offset, 0, BMAPI_DEVICE, &iomap, &maps);
- if (error)
- return -error;
+ bdev = xfs_find_bdev_for_inode(XFS_I(inode));
if (rw == WRITE) {
iocb->private = xfs_alloc_ioend(inode, IOMAP_UNWRITTEN);
ret = blockdev_direct_IO_own_locking(rw, iocb, inode,
- iomap.iomap_target->bt_bdev,
- iov, offset, nr_segs,
+ bdev, iov, offset, nr_segs,
xfs_get_blocks_direct,
xfs_end_io_direct);
} else {
iocb->private = xfs_alloc_ioend(inode, IOMAP_READ);
ret = blockdev_direct_IO_no_locking(rw, iocb, inode,
- iomap.iomap_target->bt_bdev,
- iov, offset, nr_segs,
+ bdev, iov, offset, nr_segs,
xfs_get_blocks_direct,
xfs_end_io_direct);
}
}
STATIC int
-xfs_vm_prepare_write(
+xfs_vm_write_begin(
struct file *file,
- struct page *page,
- unsigned int from,
- unsigned int to)
+ struct address_space *mapping,
+ loff_t pos,
+ unsigned len,
+ unsigned flags,
+ struct page **pagep,
+ void **fsdata)
{
- return block_prepare_write(page, from, to, xfs_get_blocks);
+ *pagep = NULL;
+ return block_write_begin(file, mapping, pos, len, flags, pagep, fsdata,
+ xfs_get_blocks);
}
STATIC sector_t
sector_t block)
{
struct inode *inode = (struct inode *)mapping->host;
- bhv_vnode_t *vp = vn_from_inode(inode);
+ struct xfs_inode *ip = XFS_I(inode);
- vn_trace_entry(vp, __FUNCTION__, (inst_t *)__return_address);
- bhv_vop_rwlock(vp, VRWLOCK_READ);
- bhv_vop_flush_pages(vp, (xfs_off_t)0, -1, 0, FI_REMAPF);
- bhv_vop_rwunlock(vp, VRWLOCK_READ);
+ xfs_itrace_entry(XFS_I(inode));
+ xfs_ilock(ip, XFS_IOLOCK_SHARED);
+ xfs_flush_pages(ip, (xfs_off_t)0, -1, 0, FI_REMAPF);
+ xfs_iunlock(ip, XFS_IOLOCK_SHARED);
return generic_block_bmap(mapping, block, xfs_get_blocks);
}
.sync_page = block_sync_page,
.releasepage = xfs_vm_releasepage,
.invalidatepage = xfs_vm_invalidatepage,
- .prepare_write = xfs_vm_prepare_write,
- .commit_write = generic_commit_write,
+ .write_begin = xfs_vm_write_begin,
+ .write_end = generic_write_end,
.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,
};