From: Mark Harmstone <[email protected]>
To: <[email protected]>, <[email protected]>
Cc: Mark Harmstone <[email protected]>
Subject: [PATCH 2/6] btrfs: store encoded read state in struct btrfs_encoded_read_private
Date: Fri, 23 Aug 2024 17:27:44 +0100 [thread overview]
Message-ID: <[email protected]> (raw)
In-Reply-To: <[email protected]>
Move the various stack variables needed for encoded reads into struct
btrfs_encoded_read_private, so that we can split it into several
functions.
Signed-off-by: Mark Harmstone <[email protected]>
---
fs/btrfs/btrfs_inode.h | 20 ++++-
fs/btrfs/inode.c | 170 +++++++++++++++++++++--------------------
fs/btrfs/ioctl.c | 60 ++++++++-------
3 files changed, 135 insertions(+), 115 deletions(-)
diff --git a/fs/btrfs/btrfs_inode.h b/fs/btrfs/btrfs_inode.h
index affe70929234..5cd4308bd337 100644
--- a/fs/btrfs/btrfs_inode.h
+++ b/fs/btrfs/btrfs_inode.h
@@ -605,9 +605,23 @@ int btrfs_encoded_read_regular_fill_pages(struct btrfs_inode *inode,
u64 file_offset, u64 disk_bytenr,
u64 disk_io_size,
struct page **pages);
-ssize_t btrfs_encoded_read(struct file *file, loff_t offset,
- struct iov_iter *iter,
- struct btrfs_ioctl_encoded_io_args *encoded);
+
+struct btrfs_encoded_read_private {
+ wait_queue_head_t wait;
+ atomic_t pending;
+ blk_status_t status;
+ unsigned long nr_pages;
+ struct page **pages;
+ struct extent_state *cached_state;
+ size_t count;
+ struct iovec iovstack[UIO_FASTIOV];
+ struct iovec *iov;
+ struct iov_iter iter;
+ struct btrfs_ioctl_encoded_io_args args;
+ struct file *file;
+};
+
+ssize_t btrfs_encoded_read(struct btrfs_encoded_read_private *priv);
ssize_t btrfs_do_encoded_write(struct kiocb *iocb, struct iov_iter *from,
const struct btrfs_ioctl_encoded_io_args *encoded);
diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c
index a0cc029d95ed..c1292e58366a 100644
--- a/fs/btrfs/inode.c
+++ b/fs/btrfs/inode.c
@@ -9078,12 +9078,6 @@ static ssize_t btrfs_encoded_read_inline(
return ret;
}
-struct btrfs_encoded_read_private {
- wait_queue_head_t wait;
- atomic_t pending;
- blk_status_t status;
-};
-
static void btrfs_encoded_read_endio(struct btrfs_bio *bbio)
{
struct btrfs_encoded_read_private *priv = bbio->private;
@@ -9104,33 +9098,31 @@ static void btrfs_encoded_read_endio(struct btrfs_bio *bbio)
bio_put(&bbio->bio);
}
-int btrfs_encoded_read_regular_fill_pages(struct btrfs_inode *inode,
- u64 file_offset, u64 disk_bytenr,
- u64 disk_io_size, struct page **pages)
+static void _btrfs_encoded_read_regular_fill_pages(struct btrfs_inode *inode,
+ u64 file_offset, u64 disk_bytenr,
+ u64 disk_io_size,
+ struct btrfs_encoded_read_private *priv)
{
struct btrfs_fs_info *fs_info = inode->root->fs_info;
- struct btrfs_encoded_read_private priv = {
- .pending = ATOMIC_INIT(1),
- };
unsigned long i = 0;
struct btrfs_bio *bbio;
- init_waitqueue_head(&priv.wait);
+ init_waitqueue_head(&priv->wait);
bbio = btrfs_bio_alloc(BIO_MAX_VECS, REQ_OP_READ, fs_info,
- btrfs_encoded_read_endio, &priv);
+ btrfs_encoded_read_endio, priv);
bbio->bio.bi_iter.bi_sector = disk_bytenr >> SECTOR_SHIFT;
bbio->inode = inode;
do {
size_t bytes = min_t(u64, disk_io_size, PAGE_SIZE);
- if (bio_add_page(&bbio->bio, pages[i], bytes, 0) < bytes) {
- atomic_inc(&priv.pending);
+ if (bio_add_page(&bbio->bio, priv->pages[i], bytes, 0) < bytes) {
+ atomic_inc(&priv->pending);
btrfs_submit_bio(bbio, 0);
bbio = btrfs_bio_alloc(BIO_MAX_VECS, REQ_OP_READ, fs_info,
- btrfs_encoded_read_endio, &priv);
+ btrfs_encoded_read_endio, priv);
bbio->bio.bi_iter.bi_sector = disk_bytenr >> SECTOR_SHIFT;
bbio->inode = inode;
continue;
@@ -9141,8 +9133,21 @@ int btrfs_encoded_read_regular_fill_pages(struct btrfs_inode *inode,
disk_io_size -= bytes;
} while (disk_io_size);
- atomic_inc(&priv.pending);
+ atomic_inc(&priv->pending);
btrfs_submit_bio(bbio, 0);
+}
+
+int btrfs_encoded_read_regular_fill_pages(struct btrfs_inode *inode,
+ u64 file_offset, u64 disk_bytenr,
+ u64 disk_io_size, struct page **pages)
+{
+ struct btrfs_encoded_read_private priv = {
+ .pending = ATOMIC_INIT(1),
+ .pages = pages,
+ };
+
+ _btrfs_encoded_read_regular_fill_pages(inode, file_offset, disk_bytenr,
+ disk_io_size, &priv);
if (atomic_dec_return(&priv.pending))
io_wait_event(priv.wait, !atomic_read(&priv.pending));
@@ -9150,54 +9155,56 @@ int btrfs_encoded_read_regular_fill_pages(struct btrfs_inode *inode,
return blk_status_to_errno(READ_ONCE(priv.status));
}
-static ssize_t btrfs_encoded_read_regular(struct btrfs_inode *inode,
- loff_t offset, struct iov_iter *iter,
+static ssize_t btrfs_encoded_read_regular(struct btrfs_encoded_read_private *priv,
u64 start, u64 lockend,
- struct extent_state **cached_state,
u64 disk_bytenr, u64 disk_io_size,
- size_t count, bool compressed,
bool *unlocked)
{
+ struct btrfs_inode *inode = BTRFS_I(file_inode(priv->file));
struct extent_io_tree *io_tree = &inode->io_tree;
- struct page **pages;
- unsigned long nr_pages, i;
+ unsigned long i;
u64 cur;
size_t page_offset;
ssize_t ret;
- nr_pages = DIV_ROUND_UP(disk_io_size, PAGE_SIZE);
- pages = kcalloc(nr_pages, sizeof(struct page *), GFP_NOFS);
- if (!pages)
+ priv->nr_pages = DIV_ROUND_UP(disk_io_size, PAGE_SIZE);
+ priv->pages = kcalloc(priv->nr_pages, sizeof(struct page *), GFP_NOFS);
+ if (!priv->pages)
return -ENOMEM;
- ret = btrfs_alloc_page_array(nr_pages, pages, false);
+ ret = btrfs_alloc_page_array(priv->nr_pages, priv->pages, false);
if (ret) {
ret = -ENOMEM;
goto out;
}
- ret = btrfs_encoded_read_regular_fill_pages(inode, start, disk_bytenr,
- disk_io_size, pages);
+ _btrfs_encoded_read_regular_fill_pages(inode, start, disk_bytenr,
+ disk_io_size, priv);
+
+ if (atomic_dec_return(&priv->pending))
+ io_wait_event(priv->wait, !atomic_read(&priv->pending));
+
+ ret = blk_status_to_errno(READ_ONCE(priv->status));
if (ret)
goto out;
- unlock_extent(io_tree, start, lockend, cached_state);
+ unlock_extent(io_tree, start, lockend, &priv->cached_state);
btrfs_inode_unlock(inode, BTRFS_ILOCK_SHARED);
*unlocked = true;
- if (compressed) {
+ if (priv->args.compression) {
i = 0;
page_offset = 0;
} else {
- i = (offset - start) >> PAGE_SHIFT;
- page_offset = (offset - start) & (PAGE_SIZE - 1);
+ i = (priv->args.offset - start) >> PAGE_SHIFT;
+ page_offset = (priv->args.offset - start) & (PAGE_SIZE - 1);
}
cur = 0;
- while (cur < count) {
- size_t bytes = min_t(size_t, count - cur,
+ while (cur < priv->count) {
+ size_t bytes = min_t(size_t, priv->count - cur,
PAGE_SIZE - page_offset);
- if (copy_page_to_iter(pages[i], page_offset, bytes,
- iter) != bytes) {
+ if (copy_page_to_iter(priv->pages[i], page_offset, bytes,
+ &priv->iter) != bytes) {
ret = -EFAULT;
goto out;
}
@@ -9205,42 +9212,40 @@ static ssize_t btrfs_encoded_read_regular(struct btrfs_inode *inode,
cur += bytes;
page_offset = 0;
}
- ret = count;
+ ret = priv->count;
out:
- for (i = 0; i < nr_pages; i++) {
- if (pages[i])
- __free_page(pages[i]);
+ for (i = 0; i < priv->nr_pages; i++) {
+ if (priv->pages[i])
+ __free_page(priv->pages[i]);
}
- kfree(pages);
+ kfree(priv->pages);
return ret;
}
-ssize_t btrfs_encoded_read(struct file *file, loff_t offset,
- struct iov_iter *iter,
- struct btrfs_ioctl_encoded_io_args *encoded)
+ssize_t btrfs_encoded_read(struct btrfs_encoded_read_private *priv)
{
- struct btrfs_inode *inode = BTRFS_I(file_inode(file));
+ struct btrfs_inode *inode = BTRFS_I(file_inode(priv->file));
struct btrfs_fs_info *fs_info = inode->root->fs_info;
struct extent_io_tree *io_tree = &inode->io_tree;
ssize_t ret;
- size_t count = iov_iter_count(iter);
u64 start, lockend, disk_bytenr, disk_io_size;
- struct extent_state *cached_state = NULL;
struct extent_map *em;
bool unlocked = false;
- file_accessed(file);
+ priv->count = iov_iter_count(&priv->iter);
+
+ file_accessed(priv->file);
btrfs_inode_lock(inode, BTRFS_ILOCK_SHARED);
- if (offset >= inode->vfs_inode.i_size) {
+ if (priv->args.offset >= inode->vfs_inode.i_size) {
btrfs_inode_unlock(inode, BTRFS_ILOCK_SHARED);
return 0;
}
- start = ALIGN_DOWN(offset, fs_info->sectorsize);
+ start = ALIGN_DOWN(priv->args.offset, fs_info->sectorsize);
/*
- * We don't know how long the extent containing offset is, but if
- * it's compressed we know that it won't be longer than this.
+ * We don't know how long the extent containing priv->args.offset is,
+ * but if it's compressed we know that it won't be longer than this.
*/
lockend = start + BTRFS_MAX_UNCOMPRESSED - 1;
@@ -9251,13 +9256,13 @@ ssize_t btrfs_encoded_read(struct file *file, loff_t offset,
lockend - start + 1);
if (ret)
goto out_unlock_inode;
- lock_extent(io_tree, start, lockend, &cached_state);
+ lock_extent(io_tree, start, lockend, &priv->cached_state);
ordered = btrfs_lookup_ordered_range(inode, start,
lockend - start + 1);
if (!ordered)
break;
btrfs_put_ordered_extent(ordered);
- unlock_extent(io_tree, start, lockend, &cached_state);
+ unlock_extent(io_tree, start, lockend, &priv->cached_state);
cond_resched();
}
@@ -9276,10 +9281,11 @@ ssize_t btrfs_encoded_read(struct file *file, loff_t offset,
*/
free_extent_map(em);
em = NULL;
- ret = btrfs_encoded_read_inline(inode, offset, iter, start,
- lockend, &cached_state,
- extent_start, count, encoded,
- &unlocked);
+ ret = btrfs_encoded_read_inline(inode, priv->args.offset,
+ &priv->iter, start,
+ lockend, &priv->cached_state,
+ extent_start, priv->count,
+ &priv->args, &unlocked);
goto out_em;
}
@@ -9287,62 +9293,60 @@ ssize_t btrfs_encoded_read(struct file *file, loff_t offset,
* We only want to return up to EOF even if the extent extends beyond
* that.
*/
- encoded->len = min_t(u64, extent_map_end(em),
- inode->vfs_inode.i_size) - offset;
+ priv->args.len = min_t(u64, extent_map_end(em),
+ inode->vfs_inode.i_size) - priv->args.offset;
if (em->disk_bytenr == EXTENT_MAP_HOLE ||
(em->flags & EXTENT_FLAG_PREALLOC)) {
disk_bytenr = EXTENT_MAP_HOLE;
- count = min_t(u64, count, encoded->len);
- encoded->len = count;
- encoded->unencoded_len = count;
+ priv->count = min_t(u64, priv->count, priv->args.len);
+ priv->args.len = priv->count;
+ priv->args.unencoded_len = priv->count;
} else if (extent_map_is_compressed(em)) {
disk_bytenr = em->disk_bytenr;
/*
* Bail if the buffer isn't large enough to return the whole
* compressed extent.
*/
- if (em->disk_num_bytes > count) {
+ if (em->disk_num_bytes > priv->count) {
ret = -ENOBUFS;
goto out_em;
}
disk_io_size = em->disk_num_bytes;
- count = em->disk_num_bytes;
- encoded->unencoded_len = em->ram_bytes;
- encoded->unencoded_offset = offset - (em->start - em->offset);
+ priv->count = em->disk_num_bytes;
+ priv->args.unencoded_len = em->ram_bytes;
+ priv->args.unencoded_offset = priv->args.offset - (em->start - em->offset);
ret = btrfs_encoded_io_compression_from_extent(fs_info,
extent_map_compression(em));
if (ret < 0)
goto out_em;
- encoded->compression = ret;
+ priv->args.compression = ret;
} else {
disk_bytenr = extent_map_block_start(em) + (start - em->start);
- if (encoded->len > count)
- encoded->len = count;
+ if (priv->args.len > priv->count)
+ priv->args.len = priv->count;
/*
* Don't read beyond what we locked. This also limits the page
* allocations that we'll do.
*/
- disk_io_size = min(lockend + 1, offset + encoded->len) - start;
- count = start + disk_io_size - offset;
- encoded->len = count;
- encoded->unencoded_len = count;
+ disk_io_size = min(lockend + 1, priv->args.offset + priv->args.len) - start;
+ priv->count = start + disk_io_size - priv->args.offset;
+ priv->args.len = priv->count;
+ priv->args.unencoded_len = priv->count;
disk_io_size = ALIGN(disk_io_size, fs_info->sectorsize);
}
free_extent_map(em);
em = NULL;
if (disk_bytenr == EXTENT_MAP_HOLE) {
- unlock_extent(io_tree, start, lockend, &cached_state);
+ unlock_extent(io_tree, start, lockend, &priv->cached_state);
btrfs_inode_unlock(inode, BTRFS_ILOCK_SHARED);
unlocked = true;
- ret = iov_iter_zero(count, iter);
- if (ret != count)
+ ret = iov_iter_zero(priv->count, &priv->iter);
+ if (ret != priv->count)
ret = -EFAULT;
} else {
- ret = btrfs_encoded_read_regular(inode, offset, iter, start,
- lockend, &cached_state,
+ ret = btrfs_encoded_read_regular(priv, start, lockend,
disk_bytenr, disk_io_size,
- count, encoded->compression,
&unlocked);
}
@@ -9350,7 +9354,7 @@ ssize_t btrfs_encoded_read(struct file *file, loff_t offset,
free_extent_map(em);
out_unlock_extent:
if (!unlocked)
- unlock_extent(io_tree, start, lockend, &cached_state);
+ unlock_extent(io_tree, start, lockend, &priv->cached_state);
out_unlock_inode:
if (!unlocked)
btrfs_inode_unlock(inode, BTRFS_ILOCK_SHARED);
diff --git a/fs/btrfs/ioctl.c b/fs/btrfs/ioctl.c
index 406ed70814f5..770bd609f386 100644
--- a/fs/btrfs/ioctl.c
+++ b/fs/btrfs/ioctl.c
@@ -4512,19 +4512,19 @@ static int _btrfs_ioctl_send(struct btrfs_inode *inode, void __user *argp, bool
static int btrfs_ioctl_encoded_read(struct file *file, void __user *argp,
bool compat)
{
- struct btrfs_ioctl_encoded_io_args args = { 0 };
size_t copy_end_kernel = offsetofend(struct btrfs_ioctl_encoded_io_args,
flags);
size_t copy_end;
- struct iovec iovstack[UIO_FASTIOV];
- struct iovec *iov = iovstack;
- struct iov_iter iter;
loff_t pos;
ssize_t ret;
+ struct btrfs_encoded_read_private priv = {
+ .pending = ATOMIC_INIT(1),
+ .file = file,
+ };
if (!capable(CAP_SYS_ADMIN)) {
ret = -EPERM;
- goto out_acct;
+ goto out;
}
if (compat) {
@@ -4535,53 +4535,55 @@ static int btrfs_ioctl_encoded_read(struct file *file, void __user *argp,
flags);
if (copy_from_user(&args32, argp, copy_end)) {
ret = -EFAULT;
- goto out_acct;
+ goto out;
}
- args.iov = compat_ptr(args32.iov);
- args.iovcnt = args32.iovcnt;
- args.offset = args32.offset;
- args.flags = args32.flags;
+ priv.args.iov = compat_ptr(args32.iov);
+ priv.args.iovcnt = args32.iovcnt;
+ priv.args.offset = args32.offset;
+ priv.args.flags = args32.flags;
#else
return -ENOTTY;
#endif
} else {
copy_end = copy_end_kernel;
- if (copy_from_user(&args, argp, copy_end)) {
+ if (copy_from_user(&priv.args, argp, copy_end)) {
ret = -EFAULT;
- goto out_acct;
+ goto out;
}
}
- if (args.flags != 0) {
+ if (priv.args.flags != 0) {
ret = -EINVAL;
- goto out_acct;
+ goto out;
}
- ret = import_iovec(ITER_DEST, args.iov, args.iovcnt, ARRAY_SIZE(iovstack),
- &iov, &iter);
- if (ret < 0)
- goto out_acct;
+ priv.iov = priv.iovstack;
+ ret = import_iovec(ITER_DEST, priv.args.iov, priv.args.iovcnt,
+ ARRAY_SIZE(priv.iovstack), &priv.iov, &priv.iter);
+ if (ret < 0) {
+ priv.iov = NULL;
+ goto out;
+ }
- if (iov_iter_count(&iter) == 0) {
+ if (iov_iter_count(&priv.iter) == 0) {
ret = 0;
- goto out_iov;
+ goto out;
}
- pos = args.offset;
- ret = rw_verify_area(READ, file, &pos, args.len);
+ pos = priv.args.offset;
+ ret = rw_verify_area(READ, file, &pos, priv.args.len);
if (ret < 0)
- goto out_iov;
+ goto out;
- ret = btrfs_encoded_read(file, pos, &iter, &args);
+ ret = btrfs_encoded_read(&priv);
if (ret >= 0) {
fsnotify_access(file);
if (copy_to_user(argp + copy_end,
- (char *)&args + copy_end_kernel,
- sizeof(args) - copy_end_kernel))
+ (char *)&priv.args + copy_end_kernel,
+ sizeof(priv.args) - copy_end_kernel))
ret = -EFAULT;
}
-out_iov:
- kfree(iov);
-out_acct:
+out:
+ kfree(priv.iov);
if (ret > 0)
add_rchar(current, ret);
inc_syscr(current);
--
2.44.2
next prev parent reply other threads:[~2024-08-23 16:28 UTC|newest]
Thread overview: 14+ messages / expand[flat|nested] mbox.gz Atom feed top
2024-08-23 16:27 [PATCH 0/6] btrfs: add io_uring for encoded reads Mark Harmstone
2024-08-23 16:27 ` [PATCH 1/6] btrfs: remove iocb from btrfs_encoded_read Mark Harmstone
2024-08-27 1:12 ` David Sterba
2024-08-23 16:27 ` Mark Harmstone [this message]
2024-08-26 15:22 ` [PATCH 2/6] btrfs: store encoded read state in struct btrfs_encoded_read_private David Sterba
2024-08-27 1:03 ` David Sterba
2024-09-06 15:19 ` Pavel Begunkov
2024-08-23 16:27 ` [PATCH 3/6] btrfs: add btrfs_encoded_read_finish Mark Harmstone
2024-08-23 16:27 ` [PATCH 4/6] btrfs: add btrfs_prepare_encoded_read Mark Harmstone
2024-08-23 16:27 ` [PATCH 5/6] btrfs: move wait out of btrfs_encoded_read Mark Harmstone
2024-09-06 15:11 ` Pavel Begunkov
2024-08-23 16:27 ` [PATCH 6/6] btrfs: add io_uring interface for encoded reads Mark Harmstone
2024-09-06 14:41 ` Pavel Begunkov
2024-09-06 15:33 ` Pavel Begunkov
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
[email protected] \
[email protected] \
[email protected] \
[email protected] \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox