summaryrefslogtreecommitdiff
path: root/fs
diff options
context:
space:
mode:
authorJosef Bacik <josef@redhat.com>2010-05-23 11:07:21 -0400
committerChris Mason <chris.mason@oracle.com>2010-05-25 10:34:57 -0400
commit11c65dccf70be9ace5dbd3906778e1a099b1fee1 (patch)
tree1289f139ddf652e39672374b6f9051994c21ce57 /fs
parent4b46fce23349bfca781a32e2707a18328ca5ae22 (diff)
Btrfs: do aio_write instead of write
In order for AIO to work, we need to implement aio_write. This patch converts our btrfs_file_write to btrfs_aio_write. I've tested this with xfstests and nothing broke, and the AIO stuff magically started working. Thanks, Signed-off-by: Josef Bacik <josef@redhat.com> Signed-off-by: Chris Mason <chris.mason@oracle.com>
Diffstat (limited to 'fs')
-rw-r--r--fs/btrfs/extent_io.c11
-rw-r--r--fs/btrfs/file.c176
2 files changed, 104 insertions, 83 deletions
diff --git a/fs/btrfs/extent_io.c b/fs/btrfs/extent_io.c
index 1a57c17d4029..a53aca338c7f 100644
--- a/fs/btrfs/extent_io.c
+++ b/fs/btrfs/extent_io.c
@@ -2017,6 +2017,7 @@ static int __extent_read_full_page(struct extent_io_tree *tree,
sector_t sector;
struct extent_map *em;
struct block_device *bdev;
+ struct btrfs_ordered_extent *ordered;
int ret;
int nr = 0;
size_t page_offset = 0;
@@ -2028,7 +2029,15 @@ static int __extent_read_full_page(struct extent_io_tree *tree,
set_page_extent_mapped(page);
end = page_end;
- lock_extent(tree, start, end, GFP_NOFS);
+ while (1) {
+ lock_extent(tree, start, end, GFP_NOFS);
+ ordered = btrfs_lookup_ordered_extent(inode, start);
+ if (!ordered)
+ break;
+ unlock_extent(tree, start, end, GFP_NOFS);
+ btrfs_start_ordered_extent(inode, ordered, 1);
+ btrfs_put_ordered_extent(ordered);
+ }
if (page->index == last_byte >> PAGE_CACHE_SHIFT) {
char *userpage;
diff --git a/fs/btrfs/file.c b/fs/btrfs/file.c
index a28810abfb98..233aea2e5ef2 100644
--- a/fs/btrfs/file.c
+++ b/fs/btrfs/file.c
@@ -46,32 +46,42 @@
static noinline int btrfs_copy_from_user(loff_t pos, int num_pages,
int write_bytes,
struct page **prepared_pages,
- const char __user *buf)
+ struct iov_iter *i)
{
- long page_fault = 0;
- int i;
+ size_t copied;
+ int pg = 0;
int offset = pos & (PAGE_CACHE_SIZE - 1);
- for (i = 0; i < num_pages && write_bytes > 0; i++, offset = 0) {
+ while (write_bytes > 0) {
size_t count = min_t(size_t,
PAGE_CACHE_SIZE - offset, write_bytes);
- struct page *page = prepared_pages[i];
- fault_in_pages_readable(buf, count);
+ struct page *page = prepared_pages[pg];
+again:
+ if (unlikely(iov_iter_fault_in_readable(i, count)))
+ return -EFAULT;
/* Copy data from userspace to the current page */
- kmap(page);
- page_fault = __copy_from_user(page_address(page) + offset,
- buf, count);
+ copied = iov_iter_copy_from_user(page, i, offset, count);
+
/* Flush processor's dcache for this page */
flush_dcache_page(page);
- kunmap(page);
- buf += count;
- write_bytes -= count;
+ iov_iter_advance(i, copied);
+ write_bytes -= copied;
- if (page_fault)
- break;
+ if (unlikely(copied == 0)) {
+ count = min_t(size_t, PAGE_CACHE_SIZE - offset,
+ iov_iter_single_seg_count(i));
+ goto again;
+ }
+
+ if (unlikely(copied < PAGE_CACHE_SIZE - offset)) {
+ offset += copied;
+ } else {
+ pg++;
+ offset = 0;
+ }
}
- return page_fault ? -EFAULT : 0;
+ return 0;
}
/*
@@ -822,60 +832,24 @@ again:
return 0;
}
-/* Copied from read-write.c */
-static void wait_on_retry_sync_kiocb(struct kiocb *iocb)
-{
- set_current_state(TASK_UNINTERRUPTIBLE);
- if (!kiocbIsKicked(iocb))
- schedule();
- else
- kiocbClearKicked(iocb);
- __set_current_state(TASK_RUNNING);
-}
-
-/*
- * Just a copy of what do_sync_write does.
- */
-static ssize_t __btrfs_direct_write(struct file *file, const char __user *buf,
- size_t count, loff_t pos, loff_t *ppos)
+static ssize_t btrfs_file_aio_write(struct kiocb *iocb,
+ const struct iovec *iov,
+ unsigned long nr_segs, loff_t pos)
{
- struct iovec iov = { .iov_base = (void __user *)buf, .iov_len = count };
- unsigned long nr_segs = 1;
- struct kiocb kiocb;
- ssize_t ret;
-
- init_sync_kiocb(&kiocb, file);
- kiocb.ki_pos = pos;
- kiocb.ki_left = count;
- kiocb.ki_nbytes = count;
-
- while (1) {
- ret = generic_file_direct_write(&kiocb, &iov, &nr_segs, pos,
- ppos, count, count);
- if (ret != -EIOCBRETRY)
- break;
- wait_on_retry_sync_kiocb(&kiocb);
- }
-
- if (ret == -EIOCBQUEUED)
- ret = wait_on_sync_kiocb(&kiocb);
- *ppos = kiocb.ki_pos;
- return ret;
-}
-
-static ssize_t btrfs_file_write(struct file *file, const char __user *buf,
- size_t count, loff_t *ppos)
-{
- loff_t pos;
+ struct file *file = iocb->ki_filp;
+ struct inode *inode = fdentry(file)->d_inode;
+ struct btrfs_root *root = BTRFS_I(inode)->root;
+ struct page *pinned[2];
+ struct page **pages = NULL;
+ struct iov_iter i;
+ loff_t *ppos = &iocb->ki_pos;
loff_t start_pos;
ssize_t num_written = 0;
ssize_t err = 0;
+ size_t count;
+ size_t ocount;
int ret = 0;
- struct inode *inode = fdentry(file)->d_inode;
- struct btrfs_root *root = BTRFS_I(inode)->root;
- struct page **pages = NULL;
int nrptrs;
- struct page *pinned[2];
unsigned long first_index;
unsigned long last_index;
int will_write;
@@ -887,13 +861,17 @@ static ssize_t btrfs_file_write(struct file *file, const char __user *buf,
pinned[0] = NULL;
pinned[1] = NULL;
- pos = *ppos;
start_pos = pos;
vfs_check_frozen(inode->i_sb, SB_FREEZE_WRITE);
mutex_lock(&inode->i_mutex);
+ err = generic_segment_checks(iov, &nr_segs, &ocount, VERIFY_READ);
+ if (err)
+ goto out;
+ count = ocount;
+
current->backing_dev_info = inode->i_mapping->backing_dev_info;
err = generic_write_checks(file, &pos, &count, S_ISBLK(inode->i_mode));
if (err)
@@ -910,14 +888,48 @@ static ssize_t btrfs_file_write(struct file *file, const char __user *buf,
BTRFS_I(inode)->sequence++;
if (unlikely(file->f_flags & O_DIRECT)) {
- num_written = __btrfs_direct_write(file, buf, count, pos,
- ppos);
- pos += num_written;
- count -= num_written;
+ ret = btrfs_delalloc_reserve_space(inode, count);
+ if (ret)
+ goto out;
- /* We've written everything we wanted to, exit */
- if (num_written < 0 || !count)
+ num_written = generic_file_direct_write(iocb, iov, &nr_segs,
+ pos, ppos, count,
+ ocount);
+
+ /*
+ * the generic O_DIRECT will update in-memory i_size after the
+ * DIOs are done. But our endio handlers that update the on
+ * disk i_size never update past the in memory i_size. So we
+ * need one more update here to catch any additions to the
+ * file
+ */
+ if (inode->i_size != BTRFS_I(inode)->disk_i_size) {
+ btrfs_ordered_update_i_size(inode, inode->i_size, NULL);
+ mark_inode_dirty(inode);
+ }
+
+ if (num_written < 0) {
+ if (num_written != -EIOCBQUEUED) {
+ /*
+ * aio land will take care of releasing the
+ * delalloc
+ */
+ btrfs_delalloc_release_space(inode, count);
+ }
+ ret = num_written;
+ num_written = 0;
goto out;
+ } else if (num_written == count) {
+ /* pick up pos changes done by the generic code */
+ pos = *ppos;
+ goto out;
+ }
+
+ /*
+ * the buffered IO will reserve bytes for the rest of the
+ * range, don't double count them here
+ */
+ btrfs_delalloc_release_space(inode, count - num_written);
/*
* We are going to do buffered for the rest of the range, so we
@@ -925,18 +937,20 @@ static ssize_t btrfs_file_write(struct file *file, const char __user *buf,
* done.
*/
buffered = 1;
- buf += num_written;
+ pos += num_written;
}
- nrptrs = min((count + PAGE_CACHE_SIZE - 1) / PAGE_CACHE_SIZE,
- PAGE_CACHE_SIZE / (sizeof(struct page *)));
+ iov_iter_init(&i, iov, nr_segs, count, num_written);
+ nrptrs = min((iov_iter_count(&i) + PAGE_CACHE_SIZE - 1) /
+ PAGE_CACHE_SIZE, PAGE_CACHE_SIZE /
+ (sizeof(struct page *)));
pages = kmalloc(nrptrs * sizeof(struct page *), GFP_KERNEL);
/* generic_write_checks can change our pos */
start_pos = pos;
first_index = pos >> PAGE_CACHE_SHIFT;
- last_index = (pos + count) >> PAGE_CACHE_SHIFT;
+ last_index = (pos + iov_iter_count(&i)) >> PAGE_CACHE_SHIFT;
/*
* there are lots of better ways to do this, but this code
@@ -953,7 +967,7 @@ static ssize_t btrfs_file_write(struct file *file, const char __user *buf,
unlock_page(pinned[0]);
}
}
- if ((pos + count) & (PAGE_CACHE_SIZE - 1)) {
+ if ((pos + iov_iter_count(&i)) & (PAGE_CACHE_SIZE - 1)) {
pinned[1] = grab_cache_page(inode->i_mapping, last_index);
if (!PageUptodate(pinned[1])) {
ret = btrfs_readpage(NULL, pinned[1]);
@@ -964,10 +978,10 @@ static ssize_t btrfs_file_write(struct file *file, const char __user *buf,
}
}
- while (count > 0) {
+ while (iov_iter_count(&i) > 0) {
size_t offset = pos & (PAGE_CACHE_SIZE - 1);
- size_t write_bytes = min(count, nrptrs *
- (size_t)PAGE_CACHE_SIZE -
+ size_t write_bytes = min(iov_iter_count(&i),
+ nrptrs * (size_t)PAGE_CACHE_SIZE -
offset);
size_t num_pages = (write_bytes + PAGE_CACHE_SIZE - 1) >>
PAGE_CACHE_SHIFT;
@@ -988,7 +1002,7 @@ static ssize_t btrfs_file_write(struct file *file, const char __user *buf,
}
ret = btrfs_copy_from_user(pos, num_pages,
- write_bytes, pages, buf);
+ write_bytes, pages, &i);
if (ret == 0) {
dirty_and_release_pages(NULL, root, file, pages,
num_pages, pos, write_bytes);
@@ -1012,8 +1026,6 @@ static ssize_t btrfs_file_write(struct file *file, const char __user *buf,
btrfs_throttle(root);
}
- buf += write_bytes;
- count -= write_bytes;
pos += write_bytes;
num_written += write_bytes;
@@ -1206,7 +1218,7 @@ const struct file_operations btrfs_file_operations = {
.read = do_sync_read,
.aio_read = generic_file_aio_read,
.splice_read = generic_file_splice_read,
- .write = btrfs_file_write,
+ .aio_write = btrfs_file_aio_write,
.mmap = btrfs_file_mmap,
.open = generic_file_open,
.release = btrfs_release_file,