aboutsummaryrefslogtreecommitdiffstats
path: root/fs/direct-io.c
diff options
context:
space:
mode:
Diffstat (limited to 'fs/direct-io.c')
-rw-r--r--fs/direct-io.c198
1 files changed, 70 insertions, 128 deletions
diff --git a/fs/direct-io.c b/fs/direct-io.c
index 00b4d15bb811..03d381377ae1 100644
--- a/fs/direct-io.c
+++ b/fs/direct-io.c
@@ -117,9 +117,7 @@ struct dio_submit {
/* dio_state communicated between submission path and end_io */
struct dio {
int flags; /* doesn't change */
- int op;
- int op_flags;
- blk_qc_t bio_cookie;
+ blk_opf_t opf; /* request operation type and flags */
struct gendisk *bio_disk;
struct inode *inode;
loff_t i_size; /* i_size when submitted */
@@ -168,12 +166,13 @@ static inline unsigned dio_pages_present(struct dio_submit *sdio)
*/
static inline int dio_refill_pages(struct dio *dio, struct dio_submit *sdio)
{
+ const enum req_op dio_op = dio->opf & REQ_OP_MASK;
ssize_t ret;
- ret = iov_iter_get_pages(sdio->iter, dio->pages, LONG_MAX, DIO_PAGES,
+ ret = iov_iter_get_pages2(sdio->iter, dio->pages, LONG_MAX, DIO_PAGES,
&sdio->from);
- if (ret < 0 && sdio->blocks_available && (dio->op == REQ_OP_WRITE)) {
+ if (ret < 0 && sdio->blocks_available && dio_op == REQ_OP_WRITE) {
struct page *page = ZERO_PAGE(0);
/*
* A memory fault, but the filesystem has some outstanding
@@ -192,7 +191,6 @@ static inline int dio_refill_pages(struct dio *dio, struct dio_submit *sdio)
}
if (ret >= 0) {
- iov_iter_advance(sdio->iter, ret);
ret += sdio->from;
sdio->head = 0;
sdio->tail = (ret + PAGE_SIZE - 1) / PAGE_SIZE;
@@ -235,6 +233,7 @@ static inline struct page *dio_get_page(struct dio *dio,
*/
static ssize_t dio_complete(struct dio *dio, ssize_t ret, unsigned int flags)
{
+ const enum req_op dio_op = dio->opf & REQ_OP_MASK;
loff_t offset = dio->iocb->ki_pos;
ssize_t transferred = 0;
int err;
@@ -252,7 +251,7 @@ static ssize_t dio_complete(struct dio *dio, ssize_t ret, unsigned int flags)
transferred = dio->result;
/* Check for short read case */
- if ((dio->op == REQ_OP_READ) &&
+ if (dio_op == REQ_OP_READ &&
((offset + transferred) > dio->i_size))
transferred = dio->i_size - offset;
/* ignore EFAULT if some IO has been done */
@@ -287,7 +286,7 @@ static ssize_t dio_complete(struct dio *dio, ssize_t ret, unsigned int flags)
* zeros from unwritten extents.
*/
if (flags & DIO_COMPLETE_INVALIDATE &&
- ret > 0 && dio->op == REQ_OP_WRITE &&
+ ret > 0 && dio_op == REQ_OP_WRITE &&
dio->inode->i_mapping->nrpages) {
err = invalidate_inode_pages2_range(dio->inode->i_mapping,
offset >> PAGE_SHIFT,
@@ -306,9 +305,9 @@ static ssize_t dio_complete(struct dio *dio, ssize_t ret, unsigned int flags)
*/
dio->iocb->ki_pos += transferred;
- if (ret > 0 && dio->op == REQ_OP_WRITE)
+ if (ret > 0 && dio_op == REQ_OP_WRITE)
ret = generic_write_sync(dio->iocb, ret);
- dio->iocb->ki_complete(dio->iocb, ret, 0);
+ dio->iocb->ki_complete(dio->iocb, ret);
}
kmem_cache_free(dio_cache, dio);
@@ -330,6 +329,7 @@ static blk_status_t dio_bio_complete(struct dio *dio, struct bio *bio);
static void dio_bio_end_aio(struct bio *bio)
{
struct dio *dio = bio->bi_private;
+ const enum req_op dio_op = dio->opf & REQ_OP_MASK;
unsigned long remaining;
unsigned long flags;
bool defer_completion = false;
@@ -354,7 +354,7 @@ static void dio_bio_end_aio(struct bio *bio)
*/
if (dio->result)
defer_completion = dio->defer_completion ||
- (dio->op == REQ_OP_WRITE &&
+ (dio_op == REQ_OP_WRITE &&
dio->inode->i_mapping->nrpages);
if (defer_completion) {
INIT_WORK(&dio->complete_work, dio_aio_complete_work);
@@ -386,25 +386,6 @@ static void dio_bio_end_io(struct bio *bio)
spin_unlock_irqrestore(&dio->bio_lock, flags);
}
-/**
- * dio_end_io - handle the end io action for the given bio
- * @bio: The direct io bio thats being completed
- *
- * This is meant to be called by any filesystem that uses their own dio_submit_t
- * so that the DIO specific endio actions are dealt with after the filesystem
- * has done it's completion work.
- */
-void dio_end_io(struct bio *bio)
-{
- struct dio *dio = bio->bi_private;
-
- if (dio->is_async)
- dio_bio_end_aio(bio);
- else
- dio_bio_end_io(bio);
-}
-EXPORT_SYMBOL_GPL(dio_end_io);
-
static inline void
dio_bio_alloc(struct dio *dio, struct dio_submit *sdio,
struct block_device *bdev,
@@ -416,18 +397,12 @@ dio_bio_alloc(struct dio *dio, struct dio_submit *sdio,
* bio_alloc() is guaranteed to return a bio when allowed to sleep and
* we request a valid number of vectors.
*/
- bio = bio_alloc(GFP_KERNEL, nr_vecs);
-
- bio_set_dev(bio, bdev);
+ bio = bio_alloc(bdev, nr_vecs, dio->opf, GFP_KERNEL);
bio->bi_iter.bi_sector = first_sector;
- bio_set_op_attrs(bio, dio->op, dio->op_flags);
if (dio->is_async)
bio->bi_end_io = dio_bio_end_aio;
else
bio->bi_end_io = dio_bio_end_io;
-
- bio->bi_write_hint = dio->iocb->ki_hint;
-
sdio->bio = bio;
sdio->logical_offset_in_bio = sdio->cur_page_fs_offset;
}
@@ -441,6 +416,7 @@ dio_bio_alloc(struct dio *dio, struct dio_submit *sdio,
*/
static inline void dio_bio_submit(struct dio *dio, struct dio_submit *sdio)
{
+ const enum req_op dio_op = dio->opf & REQ_OP_MASK;
struct bio *bio = sdio->bio;
unsigned long flags;
@@ -450,16 +426,15 @@ static inline void dio_bio_submit(struct dio *dio, struct dio_submit *sdio)
dio->refcount++;
spin_unlock_irqrestore(&dio->bio_lock, flags);
- if (dio->is_async && dio->op == REQ_OP_READ && dio->should_dirty)
+ if (dio->is_async && dio_op == REQ_OP_READ && dio->should_dirty)
bio_set_pages_dirty(bio);
- dio->bio_disk = bio->bi_disk;
+ dio->bio_disk = bio->bi_bdev->bd_disk;
- if (sdio->submit_io) {
+ if (sdio->submit_io)
sdio->submit_io(bio, dio->inode, sdio->logical_offset_in_bio);
- dio->bio_cookie = BLK_QC_T_NONE;
- } else
- dio->bio_cookie = submit_bio(bio);
+ else
+ submit_bio(bio);
sdio->bio = NULL;
sdio->boundary = 0;
@@ -479,7 +454,7 @@ static inline void dio_cleanup(struct dio *dio, struct dio_submit *sdio)
* Wait for the next BIO to complete. Remove it and return it. NULL is
* returned once all BIOs have been completed. This must only be called once
* all bios have been issued so that dio->refcount can only decrease. This
- * requires that that the caller hold a reference on the dio.
+ * requires that the caller hold a reference on the dio.
*/
static struct bio *dio_await_one(struct dio *dio)
{
@@ -498,9 +473,7 @@ static struct bio *dio_await_one(struct dio *dio)
__set_current_state(TASK_UNINTERRUPTIBLE);
dio->waiter = current;
spin_unlock_irqrestore(&dio->bio_lock, flags);
- if (!(dio->iocb->ki_flags & IOCB_HIPRI) ||
- !blk_poll(dio->bio_disk->queue, dio->bio_cookie, true))
- io_schedule();
+ blk_io_schedule();
/* wake up sets us TASK_RUNNING */
spin_lock_irqsave(&dio->bio_lock, flags);
dio->waiter = NULL;
@@ -519,7 +492,8 @@ static struct bio *dio_await_one(struct dio *dio)
static blk_status_t dio_bio_complete(struct dio *dio, struct bio *bio)
{
blk_status_t err = bio->bi_status;
- bool should_dirty = dio->op == REQ_OP_READ && dio->should_dirty;
+ const enum req_op dio_op = dio->opf & REQ_OP_MASK;
+ bool should_dirty = dio_op == REQ_OP_READ && dio->should_dirty;
if (err) {
if (err == BLK_STS_AGAIN && (bio->bi_opf & REQ_NOWAIT))
@@ -646,6 +620,7 @@ static int dio_set_defer_completion(struct dio *dio)
static int get_more_blocks(struct dio *dio, struct dio_submit *sdio,
struct buffer_head *map_bh)
{
+ const enum req_op dio_op = dio->opf & REQ_OP_MASK;
int ret;
sector_t fs_startblk; /* Into file, in filesystem-sized blocks */
sector_t fs_endblk; /* Into file, in filesystem-sized blocks */
@@ -680,7 +655,7 @@ static int get_more_blocks(struct dio *dio, struct dio_submit *sdio,
* which may decide to handle it or also return an unmapped
* buffer head.
*/
- create = dio->op == REQ_OP_WRITE;
+ create = dio_op == REQ_OP_WRITE;
if (dio->flags & DIO_SKIP_HOLES) {
i_size = i_size_read(dio->inode);
if (i_size && fs_startblk <= (i_size - 1) >> i_blkbits)
@@ -712,7 +687,7 @@ static inline int dio_new_bio(struct dio *dio, struct dio_submit *sdio,
if (ret)
goto out;
sector = start_sector << (sdio->blkbits - 9);
- nr_pages = min(sdio->pages_in_io, BIO_MAX_PAGES);
+ nr_pages = bio_max_segs(sdio->pages_in_io);
BUG_ON(nr_pages <= 0);
dio_bio_alloc(dio, sdio, map_bh->b_bdev, sector, nr_pages);
sdio->boundary = 0;
@@ -828,9 +803,11 @@ submit_page_section(struct dio *dio, struct dio_submit *sdio, struct page *page,
unsigned offset, unsigned len, sector_t blocknr,
struct buffer_head *map_bh)
{
+ const enum req_op dio_op = dio->opf & REQ_OP_MASK;
int ret = 0;
+ int boundary = sdio->boundary; /* dio_send_cur_page may clear it */
- if (dio->op == REQ_OP_WRITE) {
+ if (dio_op == REQ_OP_WRITE) {
/*
* Read accounting is performed in submit_bio()
*/
@@ -867,10 +844,10 @@ submit_page_section(struct dio *dio, struct dio_submit *sdio, struct page *page,
sdio->cur_page_fs_offset = sdio->block_in_file << sdio->blkbits;
out:
/*
- * If sdio->boundary then we want to schedule the IO now to
+ * If boundary then we want to schedule the IO now to
* avoid metadata seeks.
*/
- if (sdio->boundary) {
+ if (boundary) {
ret = dio_send_cur_page(dio, sdio, map_bh);
if (sdio->bio)
dio_bio_submit(dio, sdio);
@@ -943,6 +920,7 @@ static inline void dio_zero_block(struct dio *dio, struct dio_submit *sdio,
static int do_direct_IO(struct dio *dio, struct dio_submit *sdio,
struct buffer_head *map_bh)
{
+ const enum req_op dio_op = dio->opf & REQ_OP_MASK;
const unsigned blkbits = sdio->blkbits;
const unsigned i_blkbits = blkbits + sdio->blkfactor;
int ret = 0;
@@ -1018,7 +996,7 @@ do_holes:
loff_t i_size_aligned;
/* AKPM: eargh, -ENOTBLK is a hack */
- if (dio->op == REQ_OP_WRITE) {
+ if (dio_op == REQ_OP_WRITE) {
put_page(page);
return -ENOTBLK;
}
@@ -1141,11 +1119,10 @@ static inline int drop_refcount(struct dio *dio)
* individual fields and will generate much worse code. This is important
* for the whole file.
*/
-static inline ssize_t
-do_blockdev_direct_IO(struct kiocb *iocb, struct inode *inode,
- struct block_device *bdev, struct iov_iter *iter,
- get_block_t get_block, dio_iodone_t end_io,
- dio_submit_t submit_io, int flags)
+ssize_t __blockdev_direct_IO(struct kiocb *iocb, struct inode *inode,
+ struct block_device *bdev, struct iov_iter *iter,
+ get_block_t get_block, dio_iodone_t end_io,
+ dio_submit_t submit_io, int flags)
{
unsigned i_blkbits = READ_ONCE(inode->i_blkbits);
unsigned blkbits = i_blkbits;
@@ -1165,22 +1142,13 @@ do_blockdev_direct_IO(struct kiocb *iocb, struct inode *inode,
* the early prefetch in the caller enough time.
*/
- if (align & blocksize_mask) {
- if (bdev)
- blkbits = blksize_bits(bdev_logical_block_size(bdev));
- blocksize_mask = (1 << blkbits) - 1;
- if (align & blocksize_mask)
- goto out;
- }
-
/* watch out for a 0 len io from a tricksy fs */
if (iov_iter_rw(iter) == READ && !count)
return 0;
dio = kmem_cache_alloc(dio_cache, GFP_KERNEL);
- retval = -ENOMEM;
if (!dio)
- goto out;
+ return -ENOMEM;
/*
* Believe it or not, zeroing out the page array caused a .5%
* performance regression in a database benchmark. So, we take
@@ -1189,32 +1157,32 @@ do_blockdev_direct_IO(struct kiocb *iocb, struct inode *inode,
memset(dio, 0, offsetof(struct dio, pages));
dio->flags = flags;
- if (dio->flags & DIO_LOCKING) {
- if (iov_iter_rw(iter) == READ) {
- struct address_space *mapping =
- iocb->ki_filp->f_mapping;
-
- /* will be released by direct_io_worker */
- inode_lock(inode);
-
- retval = filemap_write_and_wait_range(mapping, offset,
- end - 1);
- if (retval) {
- inode_unlock(inode);
- kmem_cache_free(dio_cache, dio);
- goto out;
- }
- }
+ if (dio->flags & DIO_LOCKING && iov_iter_rw(iter) == READ) {
+ /* will be released by direct_io_worker */
+ inode_lock(inode);
}
/* Once we sampled i_size check for reads beyond EOF */
dio->i_size = i_size_read(inode);
if (iov_iter_rw(iter) == READ && offset >= dio->i_size) {
- if (dio->flags & DIO_LOCKING)
- inode_unlock(inode);
- kmem_cache_free(dio_cache, dio);
retval = 0;
- goto out;
+ goto fail_dio;
+ }
+
+ if (align & blocksize_mask) {
+ if (bdev)
+ blkbits = blksize_bits(bdev_logical_block_size(bdev));
+ blocksize_mask = (1 << blkbits) - 1;
+ if (align & blocksize_mask)
+ goto fail_dio;
+ }
+
+ if (dio->flags & DIO_LOCKING && iov_iter_rw(iter) == READ) {
+ struct address_space *mapping = iocb->ki_filp->f_mapping;
+
+ retval = filemap_write_and_wait_range(mapping, offset, end - 1);
+ if (retval)
+ goto fail_dio;
}
/*
@@ -1232,15 +1200,12 @@ do_blockdev_direct_IO(struct kiocb *iocb, struct inode *inode,
dio->inode = inode;
if (iov_iter_rw(iter) == WRITE) {
- dio->op = REQ_OP_WRITE;
- dio->op_flags = REQ_SYNC | REQ_IDLE;
+ dio->opf = REQ_OP_WRITE | REQ_SYNC | REQ_IDLE;
if (iocb->ki_flags & IOCB_NOWAIT)
- dio->op_flags |= REQ_NOWAIT;
+ dio->opf |= REQ_NOWAIT;
} else {
- dio->op = REQ_OP_READ;
+ dio->opf = REQ_OP_READ;
}
- if (iocb->ki_flags & IOCB_HIPRI)
- dio->op_flags |= REQ_HIPRI;
/*
* For AIO O_(D)SYNC writes we need to defer completions to a workqueue
@@ -1248,7 +1213,7 @@ do_blockdev_direct_IO(struct kiocb *iocb, struct inode *inode,
*/
if (dio->is_async && iov_iter_rw(iter) == WRITE) {
retval = 0;
- if (iocb->ki_flags & IOCB_DSYNC)
+ if (iocb_is_dsync(iocb))
retval = dio_set_defer_completion(dio);
else if (!dio->inode->i_sb->s_dio_done_wq) {
/*
@@ -1258,14 +1223,8 @@ do_blockdev_direct_IO(struct kiocb *iocb, struct inode *inode,
*/
retval = sb_init_dio_done_wq(dio->inode->i_sb);
}
- if (retval) {
- /*
- * We grab i_mutex only for reads so we don't have
- * to release it here
- */
- kmem_cache_free(dio_cache, dio);
- goto out;
- }
+ if (retval)
+ goto fail_dio;
}
/*
@@ -1289,7 +1248,7 @@ do_blockdev_direct_IO(struct kiocb *iocb, struct inode *inode,
spin_lock_init(&dio->bio_lock);
dio->refcount = 1;
- dio->should_dirty = iter_is_iovec(iter) && iov_iter_rw(iter) == READ;
+ dio->should_dirty = user_backed_iter(iter) && iov_iter_rw(iter) == READ;
sdio.iter = iter;
sdio.final_block_in_request = end >> blkbits;
@@ -1311,7 +1270,7 @@ do_blockdev_direct_IO(struct kiocb *iocb, struct inode *inode,
if (retval == -ENOTBLK) {
/*
* The remaining part of the request will be
- * be handled by buffered I/O when we return
+ * handled by buffered I/O when we return
*/
retval = 0;
}
@@ -1368,32 +1327,15 @@ do_blockdev_direct_IO(struct kiocb *iocb, struct inode *inode,
} else
BUG_ON(retval != -EIOCBQUEUED);
-out:
return retval;
-}
-ssize_t __blockdev_direct_IO(struct kiocb *iocb, struct inode *inode,
- struct block_device *bdev, struct iov_iter *iter,
- get_block_t get_block,
- dio_iodone_t end_io, dio_submit_t submit_io,
- int flags)
-{
- /*
- * The block device state is needed in the end to finally
- * submit everything. Since it's likely to be cache cold
- * prefetch it here as first thing to hide some of the
- * latency.
- *
- * Attempt to prefetch the pieces we likely need later.
- */
- prefetch(&bdev->bd_disk->part_tbl);
- prefetch(bdev->bd_queue);
- prefetch((char *)bdev->bd_queue + SMP_CACHE_BYTES);
+fail_dio:
+ if (dio->flags & DIO_LOCKING && iov_iter_rw(iter) == READ)
+ inode_unlock(inode);
- return do_blockdev_direct_IO(iocb, inode, bdev, iter, get_block,
- end_io, submit_io, flags);
+ kmem_cache_free(dio_cache, dio);
+ return retval;
}
-
EXPORT_SYMBOL(__blockdev_direct_IO);
static __init int dio_init(void)
Copyright © 1996 – 2023 Jason A. Donenfeld. All Rights Reverse Engineered.