2023-05-20 01:12:16

by David Howells

[permalink] [raw]
Subject: [PATCH v21 00/30] splice: Kill ITER_PIPE

Hi Jens, Al, Christoph,

I've split off splice patchset and moved the block patches to a separate
branch (though they are dependent on this one).

This patchset kills off ITER_PIPE to avoid a race between truncate,
iov_iter_revert() on the pipe and an as-yet incomplete DMA to a bio with
unpinned/unref'ed pages from an O_DIRECT splice read. This causes memory
corruption[2]. Instead, we use filemap_splice_read(), which invokes the
buffered file reading code and splices from the pagecache into the pipe;
copy_splice_read(), which bulk-allocates a buffer, reads into it and then
pushes the filled pages into the pipe; or handle it in filesystem-specific
code.

(1) Rename direct_splice_read() to copy_splice_read().

(2) Simplify the calculations for the number of pages to be reclaimed in
copy_splice_read().

(3) Turn do_splice_to() into a helper, vfs_splice_read(), so that it can
be used by overlayfs and coda to perform the checks on the lower fs.

(4) Make vfs_splice_read() jump to copy_splice_read() to handle direct-I/O
and DAX.

(5) Provide shmem with its own splice_read to handle non-existent pages
in the pagecache. We don't want a ->read_folio() as we don't want to
populate holes, but filemap_get_pages() requires it.

(6) Provide overlayfs with its own splice_read to call down to a lower
layer as overlayfs doesn't provide ->read_folio().

(7) Provide coda with its own splice_read to call down to a lower layer as
coda doesn't provide ->read_folio().

(8) Direct ->splice_read to copy_splice_read() in tty, procfs, kernfs
and random files as they just copy to the output buffer and don't
splice pages.

(9) Provide stubs for afs, ceph, ecryptfs, ext4, f2fs, nfs, ntfs3, ocfs2,
orangefs, xfs and zonefs to do locking and/or revalidation.

(10) Make cifs use filemap_splice_read().

(11) Replace pointers to generic_file_splice_read() with pointers to
filemap_splice_read() as DIO and DAX are handled in the caller;
filesystems can still provide their own alternate ->splice_read() op.

(12) Remove generic_file_splice_read().

(13) Remove ITER_PIPE and its paraphernalia as generic_file_splice_read()
was the only user.

I've pushed the patches here also:

https://git.kernel.org/pub/scm/linux/kernel/git/dhowells/linux-fs.git/log/?h=kill-iter-pipe

David

Changes:
========
ver #21)
- Split off the block-layer changes into a separate branch.
- Check for zero len in vfs_splice_read().
- Check s_maxbytes in filemap_splice_read().
- Rename direct_splice_read() to copy_splice_read().
- The direct I/O and DAX handling needs to be in vfs_splice_read(), not
generic_file_splice_read(), before ->splice_read() is called.
- Don't need #ifdef CONFIG_FS_DAX as IS_DAX() is false if !CONFIG_FS_DAX.
- Replace pointers to generic_file_splice_read() to filemap_splice_read().
- Remove generic_file_splice_read().
- In ceph, drop the caps ref.
- In NFS, Fix pos -> ppos in dprintk().

ver #20)
- Make direct_splice_read() limit the read to eof for regular files and
blockdevs.
- Check against s_maxbytes on the backing store, not a devnode inode.
- Provide stubs for afs, ceph, ecryptfs, ext4, f2fs, nfs, ntfs3, ocfs2,
orangefs, xfs and zonefs.
- Always use direct_splice_read() for 9p, trace and sockets.

ver #19)
- Remove a missed get_page() on the zeropage in shmem_splice_read().

ver #18)
- Split out the cifs bits from the patch the switches
generic_file_splice_read() over to using the non-ITER_PIPE splicing.
- Don't get/put refs on the zeropage in shmem_splice_read().

ver #17)
- Rename do_splice_to() to vfs_splice_read() and export it so that it can
be a helper and make overlayfs and coda use it, allowing duplicate
checks to be removed.

ver #16)
- The filemap_get_pages() changes are now upstream.
- filemap_splice_read() and direct_splice_read() are now upstream.
- iov_iter_extract_pages() is now upstream.

ver #15)
- Fixed up some errors in overlayfs_splice_read().

ver #14)
- Some changes to generic_file_buffered_splice_read():
- Rename to filemap_splice_read() and move to mm/filemap.c.
- Create a helper, pipe_head_buf().
- Use init_sync_kiocb().
- Some changes to generic_file_direct_splice_read():
- Use alloc_pages_bulk_array() rather than alloc_pages_bulk_list().
- Use release_pages() instead of __free_page() in a loop.
- Rename to direct_splice_read().
- Rearrange the patches to implement filemap_splice_read() and
direct_splice_read() separately to changing generic_file_splice_read().
- Don't call generic_file_splice_read() when there isn't a ->read_folio().
- Insert patches to fix read_folio-less cases:
- Make tty, procfs, kernfs and (u)random use direct_splice_read().
- Make overlayfs and coda call down to a lower layer.
- Give shmem its own splice-read that doesn't insert missing pages.
- Fixed a min() with mixed type args on some arches.

ver #13)
- Only use allocation in advance and ITER_BVEC for DIO read-splice.
- Make buffered read-splice get pages directly from the pagecache.
- Alter filemap_get_pages() & co. so that it doesn't need an iterator.

ver #12)
- Added the missing __bitwise on the iov_iter_extraction_t typedef.
- Rebased on -rc7.
- Don't specify FOLL_PIN to pin_user_pages_fast().
- Inserted patch at front to fix race between DIO read and truncation that
caused memory corruption when iov_iter_revert() got called on an
ITER_PIPE iterator[2].
- Inserted a patch after that to remove the now-unused ITER_PIPE and its
helper functions.
- Removed the ITER_PIPE bits from iov_iter_extract_pages().

ver #11)
- Fix iov_iter_extract_kvec_pages() to include the offset into the page in
the returned starting offset.
- Use __bitwise for the extraction flags

ver #10)
- Fix use of i->kvec in iov_iter_extract_bvec_pages() to be i->bvec.
- Drop bio_set_cleanup_mode(), open coding it instead.

ver #9)
- It's now not permitted to use FOLL_PIN outside of mm/, so:
- Change iov_iter_extract_mode() into iov_iter_extract_will_pin() and
return true/false instead of FOLL_PIN/0.
- Drop of folio_put_unpin() and page_put_unpin() and instead call
unpin_user_page() (and put_page()) directly as necessary.
- Make __bio_release_pages() call bio_release_page() instead of
unpin_user_page() as there's no BIO_* -> FOLL_* translation to do.
- Drop the FOLL_* renumbering patch.
- Change extract_flags to extraction_flags.

ver #8)
- Import Christoph Hellwig's changes.
- Split the conversion-to-extraction patch.
- Drop the extract_flags arg from iov_iter_extract_mode().
- Don't default bios to BIO_PAGE_REFFED, but set explicitly.
- Switch FOLL_PIN and FOLL_GET when renumbering so PIN is at bit 0.
- Switch BIO_PAGE_PINNED and BIO_PAGE_REFFED so PINNED is at bit 0.
- We should always be using FOLL_PIN (not FOLL_GET) for DIO, so adjust the
patches for that.

ver #7)
- For now, drop the parts to pass the I/O direction to iov_iter_*pages*()
as it turned out to be a lot more complicated, with places not setting
IOCB_WRITE when they should, for example.
- Drop all the patches that changed things other then the block layer's
bio handling. The netfslib and cifs changes can go into a separate
patchset.
- Add support for extracting pages from KVEC-type iterators.
- When extracting from BVEC/KVEC, skip over empty vecs at the front.

ver #6)
- Fix write() syscall and co. not setting IOCB_WRITE.
- Added iocb_is_read() and iocb_is_write() to check IOCB_WRITE.
- Use op_is_write() in bio_copy_user_iov().
- Drop the iterator direction checks from smbd_recv().
- Define FOLL_SOURCE_BUF and FOLL_DEST_BUF and pass them in as part of
gup_flags to iov_iter_get/extract_pages*().
- Replace iov_iter_get_pages*2() with iov_iter_get_pages*() and remove.
- Add back the function to indicate the cleanup mode.
- Drop the cleanup_mode return arg to iov_iter_extract_pages().
- Provide a helper to clean up a page.
- Renumbered FOLL_GET and FOLL_PIN and made BIO_PAGE_REFFED/PINNED have
the same numerical values, enforced with an assertion.
- Converted AF_ALG, SCSI vhost, generic DIO, FUSE, splice to pipe, 9P and
NFS.
- Added in the patches to make CIFS do top-to-bottom iterators and use
various of the added extraction functions.
- Added a pair of work-in-progess patches to make sk_buff fragments store
FOLL_GET and FOLL_PIN.

ver #5)
- Replace BIO_NO_PAGE_REF with BIO_PAGE_REFFED and split into own patch.
- Transcribe FOLL_GET/PIN into BIO_PAGE_REFFED/PINNED flags.
- Add patch to allow bio_flagged() to be combined by gcc.

ver #4)
- Drop the patch to move the FOLL_* flags to linux/mm_types.h as they're
no longer referenced by linux/uio.h.
- Add ITER_SOURCE/DEST cleanup patches.
- Make iov_iter/netfslib iter extraction patches use ITER_SOURCE/DEST.
- Allow additional gup_flags to be passed into iov_iter_extract_pages().
- Add struct bio patch.

ver #3)
- Switch to using EXPORT_SYMBOL_GPL to prevent indirect 3rd-party access
to get/pin_user_pages_fast()[1].

ver #2)
- Rolled the extraction cleanup mode query function into the extraction
function, returning the indication through the argument list.
- Fixed patch 4 (extract to scatterlist) to actually use the new
extraction API.

Link: https://lore.kernel.org/r/Y3zFzdWnWlEJ8X8/@infradead.org/ [1]
Link: https://lore.kernel.org/r/[email protected]/ [2]
Link: https://lore.kernel.org/r/166697254399.61150.1256557652599252121.stgit@warthog.procyon.org.uk/ # rfc
Link: https://lore.kernel.org/r/166722777223.2555743.162508599131141451.stgit@warthog.procyon.org.uk/ # rfc
Link: https://lore.kernel.org/r/166732024173.3186319.18204305072070871546.stgit@warthog.procyon.org.uk/ # rfc
Link: https://lore.kernel.org/r/166869687556.3723671.10061142538708346995.stgit@warthog.procyon.org.uk/ # rfc
Link: https://lore.kernel.org/r/166920902005.1461876.2786264600108839814.stgit@warthog.procyon.org.uk/ # v2
Link: https://lore.kernel.org/r/166997419665.9475.15014699817597102032.stgit@warthog.procyon.org.uk/ # v3
Link: https://lore.kernel.org/r/167305160937.1521586.133299343565358971.stgit@warthog.procyon.org.uk/ # v4
Link: https://lore.kernel.org/r/167344725490.2425628.13771289553670112965.stgit@warthog.procyon.org.uk/ # v5
Link: https://lore.kernel.org/r/167391047703.2311931.8115712773222260073.stgit@warthog.procyon.org.uk/ # v6
Link: https://lore.kernel.org/r/[email protected]/ # v7
Link: https://lore.kernel.org/r/[email protected]/ # v8
Link: https://lore.kernel.org/r/[email protected]/ # v9
Link: https://lore.kernel.org/r/[email protected]/ # v10
Link: https://lore.kernel.org/r/[email protected]/ # v11
Link: https://lore.kernel.org/r/[email protected]/ # v12
Link: https://lore.kernel.org/r/[email protected]/ # v13
Link: https://lore.kernel.org/r/[email protected]/ # v14
Link: https://lore.kernel.org/r/[email protected]/ # v16
Link: https://lore.kernel.org/r/[email protected]/ # v17
Link: https://lore.kernel.org/r/[email protected]/ # v18
Link: https://lore.kernel.org/r/[email protected]/ # v19
Link: https://lore.kernel.org/r/[email protected]/ #v20

Additional patches that got folded in:

Link: https://lore.kernel.org/r/[email protected]/ # v1
Link: https://lore.kernel.org/r/[email protected]/ # v2
Link: https://lore.kernel.org/r/[email protected]/ # v3

David Howells (30):
splice: Fix filemap of a blockdev
splice: Make filemap_splice_read() check s_maxbytes
splice: Rename direct_splice_read() to copy_splice_read()
splice: Clean up copy_splice_read() a bit
splice: Make do_splice_to() generic and export it
splice: Check for zero count in vfs_splice_read()
splice: Make splice from an O_DIRECT fd use copy_splice_read()
splice: Make splice from a DAX file use copy_splice_read()
shmem: Implement splice-read
overlayfs: Implement splice-read
coda: Implement splice-read
tty, proc, kernfs, random: Use copy_splice_read()
net: Make sock_splice_read() use copy_splice_read() by default
9p: Add splice_read stub
afs: Provide a splice-read stub
ceph: Provide a splice-read stub
ecryptfs: Provide a splice-read stub
ext4: Provide a splice-read stub
f2fs: Provide a splice-read stub
nfs: Provide a splice-read stub
ntfs3: Provide a splice-read stub
ocfs2: Provide a splice-read stub
orangefs: Provide a splice-read stub
xfs: Provide a splice-read stub
zonefs: Provide a splice-read stub
splice: Convert trace/seq to use copy_splice_read()
cifs: Use filemap_splice_read()
splice: Use filemap_splice_read() instead of
generic_file_splice_read()
splice: Remove generic_file_splice_read()
iov_iter: Kill ITER_PIPE

block/fops.c | 2 +-
drivers/char/random.c | 4 +-
drivers/tty/tty_io.c | 4 +-
fs/9p/vfs_file.c | 26 ++-
fs/adfs/file.c | 2 +-
fs/affs/file.c | 2 +-
fs/afs/file.c | 20 +-
fs/bfs/file.c | 2 +-
fs/btrfs/file.c | 2 +-
fs/ceph/file.c | 65 +++++-
fs/cifs/cifsfs.c | 12 +-
fs/cifs/cifsfs.h | 3 -
fs/cifs/file.c | 16 --
fs/coda/file.c | 29 ++-
fs/cramfs/inode.c | 2 +-
fs/ecryptfs/file.c | 27 ++-
fs/erofs/data.c | 2 +-
fs/exfat/file.c | 2 +-
fs/ext2/file.c | 2 +-
fs/ext4/file.c | 13 +-
fs/f2fs/file.c | 43 +++-
fs/fat/file.c | 2 +-
fs/fuse/file.c | 2 +-
fs/gfs2/file.c | 4 +-
fs/hfs/inode.c | 2 +-
fs/hfsplus/inode.c | 2 +-
fs/hostfs/hostfs_kern.c | 2 +-
fs/hpfs/file.c | 2 +-
fs/jffs2/file.c | 2 +-
fs/jfs/file.c | 2 +-
fs/kernfs/file.c | 2 +-
fs/minix/file.c | 2 +-
fs/nfs/file.c | 23 ++-
fs/nfs/internal.h | 2 +
fs/nfs/nfs4file.c | 2 +-
fs/nilfs2/file.c | 2 +-
fs/ntfs/file.c | 2 +-
fs/ntfs3/file.c | 31 ++-
fs/ocfs2/file.c | 41 +++-
fs/ocfs2/ocfs2_trace.h | 3 +
fs/omfs/file.c | 2 +-
fs/orangefs/file.c | 22 +-
fs/overlayfs/file.c | 23 ++-
fs/proc/inode.c | 4 +-
fs/proc/proc_sysctl.c | 2 +-
fs/proc_namespace.c | 6 +-
fs/ramfs/file-mmu.c | 2 +-
fs/ramfs/file-nommu.c | 2 +-
fs/read_write.c | 2 +-
fs/reiserfs/file.c | 2 +-
fs/romfs/mmap-nommu.c | 2 +-
fs/splice.c | 108 ++++------
fs/sysv/file.c | 2 +-
fs/ubifs/file.c | 2 +-
fs/udf/file.c | 2 +-
fs/ufs/file.c | 2 +-
fs/vboxsf/file.c | 2 +-
fs/xfs/xfs_file.c | 30 ++-
fs/xfs/xfs_trace.h | 2 +-
fs/zonefs/file.c | 40 +++-
include/linux/fs.h | 8 +-
include/linux/splice.h | 3 +
include/linux/uio.h | 14 --
kernel/trace/trace.c | 2 +-
lib/iov_iter.c | 431 +---------------------------------------
mm/filemap.c | 10 +-
mm/shmem.c | 134 ++++++++++++-
net/socket.c | 2 +-
68 files changed, 657 insertions(+), 616 deletions(-)



2023-05-20 01:14:19

by David Howells

[permalink] [raw]
Subject: [PATCH v21 29/30] splice: Remove generic_file_splice_read()

Remove generic_file_splice_read() as it has been replaced with calls to
filemap_splice_read() and copy_splice_read().

With this, ITER_PIPE is no longer used.

Signed-off-by: David Howells <[email protected]>
cc: Christoph Hellwig <[email protected]>
cc: Jens Axboe <[email protected]>
cc: Steve French <[email protected]>
cc: Al Viro <[email protected]>
cc: David Hildenbrand <[email protected]>
cc: John Hubbard <[email protected]>
cc: [email protected]
cc: [email protected]
cc: [email protected]
cc: [email protected]
---

Notes:
ver #21)
- Move zero-len check to vfs_splice_read().
- Move s_maxbytes check to filemap_splice_read().
- DIO (and DAX) are handled by vfs_splice_read().

ver #20)
- Use s_maxbytes from the backing store (in->f_mapping), not the front
inode (especially for a blockdev).

ver #18)
- Split out the change to cifs to make it use generic_file_splice_read().
- Split out the unexport of filemap_splice_read() (still needed by cifs).

fs/splice.c | 43 -------------------------------------------
include/linux/fs.h | 2 --
2 files changed, 45 deletions(-)

diff --git a/fs/splice.c b/fs/splice.c
index 8268248df3a9..9be4cb3b9879 100644
--- a/fs/splice.c
+++ b/fs/splice.c
@@ -386,49 +386,6 @@ ssize_t copy_splice_read(struct file *in, loff_t *ppos,
}
EXPORT_SYMBOL(copy_splice_read);

-/**
- * generic_file_splice_read - splice data from file to a pipe
- * @in: file to splice from
- * @ppos: position in @in
- * @pipe: pipe to splice to
- * @len: number of bytes to splice
- * @flags: splice modifier flags
- *
- * Description:
- * Will read pages from given file and fill them into a pipe. Can be
- * used as long as it has more or less sane ->read_iter().
- *
- */
-ssize_t generic_file_splice_read(struct file *in, loff_t *ppos,
- struct pipe_inode_info *pipe, size_t len,
- unsigned int flags)
-{
- struct iov_iter to;
- struct kiocb kiocb;
- int ret;
-
- iov_iter_pipe(&to, ITER_DEST, pipe, len);
- init_sync_kiocb(&kiocb, in);
- kiocb.ki_pos = *ppos;
- ret = call_read_iter(in, &kiocb, &to);
- if (ret > 0) {
- *ppos = kiocb.ki_pos;
- file_accessed(in);
- } else if (ret < 0) {
- /* free what was emitted */
- pipe_discard_from(pipe, to.start_head);
- /*
- * callers of ->splice_read() expect -EAGAIN on
- * "can't put anything in there", rather than -EFAULT.
- */
- if (ret == -EFAULT)
- ret = -EAGAIN;
- }
-
- return ret;
-}
-EXPORT_SYMBOL(generic_file_splice_read);
-
const struct pipe_buf_operations default_pipe_buf_ops = {
.release = generic_pipe_buf_release,
.try_steal = generic_pipe_buf_try_steal,
diff --git a/include/linux/fs.h b/include/linux/fs.h
index e3c22efa413e..08ba2ae1d3ce 100644
--- a/include/linux/fs.h
+++ b/include/linux/fs.h
@@ -2755,8 +2755,6 @@ ssize_t filemap_splice_read(struct file *in, loff_t *ppos,
ssize_t copy_splice_read(struct file *in, loff_t *ppos,
struct pipe_inode_info *pipe,
size_t len, unsigned int flags);
-extern ssize_t generic_file_splice_read(struct file *, loff_t *,
- struct pipe_inode_info *, size_t, unsigned int);
extern ssize_t iter_file_splice_write(struct pipe_inode_info *,
struct file *, loff_t *, size_t, unsigned int);
extern ssize_t generic_splice_sendpage(struct pipe_inode_info *pipe,


2023-05-20 01:16:37

by David Howells

[permalink] [raw]
Subject: [PATCH v21 26/30] splice: Convert trace/seq to use copy_splice_read()

For the splice from the trace seq buffer, just use copy_splice_read().

In the future, something better can probably be done by gifting pages from
seq->buf into the pipe, but that would require changing seq->buf into a
vmap over an array of pages.

Signed-off-by: David Howells <[email protected]>
cc: Christoph Hellwig <[email protected]>
cc: Al Viro <[email protected]>
cc: Jens Axboe <[email protected]>
cc: Steven Rostedt <[email protected]>
cc: Masami Hiramatsu <[email protected]>
cc: [email protected]
cc: [email protected]
cc: [email protected]
cc: [email protected]
cc: [email protected]
---
kernel/trace/trace.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/kernel/trace/trace.c b/kernel/trace/trace.c
index ebc59781456a..c210d02fac97 100644
--- a/kernel/trace/trace.c
+++ b/kernel/trace/trace.c
@@ -5171,7 +5171,7 @@ static const struct file_operations tracing_fops = {
.open = tracing_open,
.read = seq_read,
.read_iter = seq_read_iter,
- .splice_read = generic_file_splice_read,
+ .splice_read = copy_splice_read,
.write = tracing_write_stub,
.llseek = tracing_lseek,
.release = tracing_release,


2023-05-20 01:17:47

by David Howells

[permalink] [raw]
Subject: [PATCH v21 15/30] afs: Provide a splice-read stub

Provide a splice_read stub for AFS to call afs_validate() before going into
generic_file_splice_read() so that we're likely to have a callback promise
from the server.

Signed-off-by: David Howells <[email protected]>
cc: Christoph Hellwig <[email protected]>
cc: Al Viro <[email protected]>
cc: Jens Axboe <[email protected]>
cc: Marc Dionne <[email protected]>
cc: [email protected]
cc: [email protected]
cc: [email protected]
cc: [email protected]
---
fs/afs/file.c | 20 +++++++++++++++++++-
1 file changed, 19 insertions(+), 1 deletion(-)

diff --git a/fs/afs/file.c b/fs/afs/file.c
index 719b31374879..d8a6b09dadf7 100644
--- a/fs/afs/file.c
+++ b/fs/afs/file.c
@@ -25,6 +25,9 @@ static void afs_invalidate_folio(struct folio *folio, size_t offset,
static bool afs_release_folio(struct folio *folio, gfp_t gfp_flags);

static ssize_t afs_file_read_iter(struct kiocb *iocb, struct iov_iter *iter);
+static ssize_t afs_file_splice_read(struct file *in, loff_t *ppos,
+ struct pipe_inode_info *pipe,
+ size_t len, unsigned int flags);
static void afs_vm_open(struct vm_area_struct *area);
static void afs_vm_close(struct vm_area_struct *area);
static vm_fault_t afs_vm_map_pages(struct vm_fault *vmf, pgoff_t start_pgoff, pgoff_t end_pgoff);
@@ -36,7 +39,7 @@ const struct file_operations afs_file_operations = {
.read_iter = afs_file_read_iter,
.write_iter = afs_file_write,
.mmap = afs_file_mmap,
- .splice_read = generic_file_splice_read,
+ .splice_read = afs_file_splice_read,
.splice_write = iter_file_splice_write,
.fsync = afs_fsync,
.lock = afs_lock,
@@ -587,3 +590,18 @@ static ssize_t afs_file_read_iter(struct kiocb *iocb, struct iov_iter *iter)

return generic_file_read_iter(iocb, iter);
}
+
+static ssize_t afs_file_splice_read(struct file *in, loff_t *ppos,
+ struct pipe_inode_info *pipe,
+ size_t len, unsigned int flags)
+{
+ struct afs_vnode *vnode = AFS_FS_I(file_inode(in));
+ struct afs_file *af = in->private_data;
+ int ret;
+
+ ret = afs_validate(vnode, af->key);
+ if (ret < 0)
+ return ret;
+
+ return generic_file_splice_read(in, ppos, pipe, len, flags);
+}


2023-05-20 01:18:22

by David Howells

[permalink] [raw]
Subject: [PATCH v21 05/30] splice: Make do_splice_to() generic and export it

Rename do_splice_to() to vfs_splice_read() and export it so that it can be
used as a helper when calling down to a lower layer filesystem as it
performs all the necessary checks[1].

Signed-off-by: David Howells <[email protected]>
Reviewed-by: Christoph Hellwig <[email protected]>
cc: Miklos Szeredi <[email protected]>
cc: Jens Axboe <[email protected]>
cc: Al Viro <[email protected]>
cc: John Hubbard <[email protected]>
cc: David Hildenbrand <[email protected]>
cc: Matthew Wilcox <[email protected]>
cc: [email protected]
cc: [email protected]
cc: [email protected]
cc: [email protected]
Link: https://lore.kernel.org/r/CAJfpeguGksS3sCigmRi9hJdUec8qtM9f+_9jC1rJhsXT+dV01w@mail.gmail.com/ [1]
---
fs/splice.c | 27 ++++++++++++++++++++-------
include/linux/splice.h | 3 +++
2 files changed, 23 insertions(+), 7 deletions(-)

diff --git a/fs/splice.c b/fs/splice.c
index f9a9be797b0c..d815a69f6589 100644
--- a/fs/splice.c
+++ b/fs/splice.c
@@ -867,12 +867,24 @@ static long do_splice_from(struct pipe_inode_info *pipe, struct file *out,
return out->f_op->splice_write(pipe, out, ppos, len, flags);
}

-/*
- * Attempt to initiate a splice from a file to a pipe.
+/**
+ * vfs_splice_read - Read data from a file and splice it into a pipe
+ * @in: File to splice from
+ * @ppos: Input file offset
+ * @pipe: Pipe to splice to
+ * @len: Number of bytes to splice
+ * @flags: Splice modifier flags (SPLICE_F_*)
+ *
+ * Splice the requested amount of data from the input file to the pipe. This
+ * is synchronous as the caller must hold the pipe lock across the entire
+ * operation.
+ *
+ * If successful, it returns the amount of data spliced, 0 if it hit the EOF or
+ * a hole and a negative error code otherwise.
*/
-static long do_splice_to(struct file *in, loff_t *ppos,
- struct pipe_inode_info *pipe, size_t len,
- unsigned int flags)
+long vfs_splice_read(struct file *in, loff_t *ppos,
+ struct pipe_inode_info *pipe, size_t len,
+ unsigned int flags)
{
unsigned int p_space;
int ret;
@@ -895,6 +907,7 @@ static long do_splice_to(struct file *in, loff_t *ppos,
return warn_unsupported(in, "read");
return in->f_op->splice_read(in, ppos, pipe, len, flags);
}
+EXPORT_SYMBOL_GPL(vfs_splice_read);

/**
* splice_direct_to_actor - splices data directly between two non-pipes
@@ -964,7 +977,7 @@ ssize_t splice_direct_to_actor(struct file *in, struct splice_desc *sd,
size_t read_len;
loff_t pos = sd->pos, prev_pos = pos;

- ret = do_splice_to(in, &pos, pipe, len, flags);
+ ret = vfs_splice_read(in, &pos, pipe, len, flags);
if (unlikely(ret <= 0))
goto out_release;

@@ -1112,7 +1125,7 @@ long splice_file_to_pipe(struct file *in,
pipe_lock(opipe);
ret = wait_for_space(opipe, flags);
if (!ret)
- ret = do_splice_to(in, offset, opipe, len, flags);
+ ret = vfs_splice_read(in, offset, opipe, len, flags);
pipe_unlock(opipe);
if (ret > 0)
wakeup_pipe_readers(opipe);
diff --git a/include/linux/splice.h b/include/linux/splice.h
index a55179fd60fc..8f052c3dae95 100644
--- a/include/linux/splice.h
+++ b/include/linux/splice.h
@@ -76,6 +76,9 @@ extern ssize_t splice_to_pipe(struct pipe_inode_info *,
struct splice_pipe_desc *);
extern ssize_t add_to_pipe(struct pipe_inode_info *,
struct pipe_buffer *);
+long vfs_splice_read(struct file *in, loff_t *ppos,
+ struct pipe_inode_info *pipe, size_t len,
+ unsigned int flags);
extern ssize_t splice_direct_to_actor(struct file *, struct splice_desc *,
splice_direct_actor *);
extern long do_splice(struct file *in, loff_t *off_in,


2023-05-20 01:20:00

by David Howells

[permalink] [raw]
Subject: [PATCH v21 06/30] splice: Check for zero count in vfs_splice_read()

Make vfs_splice_read() return immediately if the length is 0.

Signed-off-by: David Howells <[email protected]>
cc: Christoph Hellwig <[email protected]>
cc: Jens Axboe <[email protected]>
cc: Al Viro <[email protected]>
cc: [email protected]
cc: [email protected]
cc: [email protected]
---
fs/splice.c | 2 ++
1 file changed, 2 insertions(+)

diff --git a/fs/splice.c b/fs/splice.c
index d815a69f6589..fe3309ffeb26 100644
--- a/fs/splice.c
+++ b/fs/splice.c
@@ -891,6 +891,8 @@ long vfs_splice_read(struct file *in, loff_t *ppos,

if (unlikely(!(in->f_mode & FMODE_READ)))
return -EBADF;
+ if (!len)
+ return 0;

/* Don't try to read more the pipe has space for. */
p_space = pipe->max_usage - pipe_occupancy(pipe->head, pipe->tail);


2023-05-20 01:21:41

by David Howells

[permalink] [raw]
Subject: [PATCH v21 12/30] tty, proc, kernfs, random: Use copy_splice_read()

Use copy_splice_read() for tty, procfs, kernfs and random files rather
than going through generic_file_splice_read() as they just copy the file
into the output buffer and don't splice pages. This avoids the need for
them to have a ->read_folio() to satisfy filemap_splice_read().

Signed-off-by: David Howells <[email protected]>
Acked-by: Greg Kroah-Hartman <[email protected]>
cc: Christoph Hellwig <[email protected]>
cc: Jens Axboe <[email protected]>
cc: Al Viro <[email protected]>
cc: John Hubbard <[email protected]>
cc: David Hildenbrand <[email protected]>
cc: Matthew Wilcox <[email protected]>
cc: Miklos Szeredi <[email protected]>
cc: Arnd Bergmann <[email protected]>
cc: [email protected]
cc: [email protected]
cc: [email protected]
---
drivers/char/random.c | 4 ++--
drivers/tty/tty_io.c | 4 ++--
fs/kernfs/file.c | 2 +-
fs/proc/inode.c | 4 ++--
fs/proc/proc_sysctl.c | 2 +-
fs/proc_namespace.c | 6 +++---
6 files changed, 11 insertions(+), 11 deletions(-)

diff --git a/drivers/char/random.c b/drivers/char/random.c
index 253f2ddb8913..3cb37760dfec 100644
--- a/drivers/char/random.c
+++ b/drivers/char/random.c
@@ -1546,7 +1546,7 @@ const struct file_operations random_fops = {
.compat_ioctl = compat_ptr_ioctl,
.fasync = random_fasync,
.llseek = noop_llseek,
- .splice_read = generic_file_splice_read,
+ .splice_read = copy_splice_read,
.splice_write = iter_file_splice_write,
};

@@ -1557,7 +1557,7 @@ const struct file_operations urandom_fops = {
.compat_ioctl = compat_ptr_ioctl,
.fasync = random_fasync,
.llseek = noop_llseek,
- .splice_read = generic_file_splice_read,
+ .splice_read = copy_splice_read,
.splice_write = iter_file_splice_write,
};

diff --git a/drivers/tty/tty_io.c b/drivers/tty/tty_io.c
index c84be40fb8df..4737a8f92c2e 100644
--- a/drivers/tty/tty_io.c
+++ b/drivers/tty/tty_io.c
@@ -466,7 +466,7 @@ static const struct file_operations tty_fops = {
.llseek = no_llseek,
.read_iter = tty_read,
.write_iter = tty_write,
- .splice_read = generic_file_splice_read,
+ .splice_read = copy_splice_read,
.splice_write = iter_file_splice_write,
.poll = tty_poll,
.unlocked_ioctl = tty_ioctl,
@@ -481,7 +481,7 @@ static const struct file_operations console_fops = {
.llseek = no_llseek,
.read_iter = tty_read,
.write_iter = redirected_tty_write,
- .splice_read = generic_file_splice_read,
+ .splice_read = copy_splice_read,
.splice_write = iter_file_splice_write,
.poll = tty_poll,
.unlocked_ioctl = tty_ioctl,
diff --git a/fs/kernfs/file.c b/fs/kernfs/file.c
index 40c4661f15b7..180906c36f51 100644
--- a/fs/kernfs/file.c
+++ b/fs/kernfs/file.c
@@ -1011,7 +1011,7 @@ const struct file_operations kernfs_file_fops = {
.release = kernfs_fop_release,
.poll = kernfs_fop_poll,
.fsync = noop_fsync,
- .splice_read = generic_file_splice_read,
+ .splice_read = copy_splice_read,
.splice_write = iter_file_splice_write,
};

diff --git a/fs/proc/inode.c b/fs/proc/inode.c
index f495fdb39151..67b09a1d9433 100644
--- a/fs/proc/inode.c
+++ b/fs/proc/inode.c
@@ -591,7 +591,7 @@ static const struct file_operations proc_iter_file_ops = {
.llseek = proc_reg_llseek,
.read_iter = proc_reg_read_iter,
.write = proc_reg_write,
- .splice_read = generic_file_splice_read,
+ .splice_read = copy_splice_read,
.poll = proc_reg_poll,
.unlocked_ioctl = proc_reg_unlocked_ioctl,
.mmap = proc_reg_mmap,
@@ -617,7 +617,7 @@ static const struct file_operations proc_reg_file_ops_compat = {
static const struct file_operations proc_iter_file_ops_compat = {
.llseek = proc_reg_llseek,
.read_iter = proc_reg_read_iter,
- .splice_read = generic_file_splice_read,
+ .splice_read = copy_splice_read,
.write = proc_reg_write,
.poll = proc_reg_poll,
.unlocked_ioctl = proc_reg_unlocked_ioctl,
diff --git a/fs/proc/proc_sysctl.c b/fs/proc/proc_sysctl.c
index 8038833ff5b0..ae832e982003 100644
--- a/fs/proc/proc_sysctl.c
+++ b/fs/proc/proc_sysctl.c
@@ -868,7 +868,7 @@ static const struct file_operations proc_sys_file_operations = {
.poll = proc_sys_poll,
.read_iter = proc_sys_read,
.write_iter = proc_sys_write,
- .splice_read = generic_file_splice_read,
+ .splice_read = copy_splice_read,
.splice_write = iter_file_splice_write,
.llseek = default_llseek,
};
diff --git a/fs/proc_namespace.c b/fs/proc_namespace.c
index 846f9455ae22..250eb5bf7b52 100644
--- a/fs/proc_namespace.c
+++ b/fs/proc_namespace.c
@@ -324,7 +324,7 @@ static int mountstats_open(struct inode *inode, struct file *file)
const struct file_operations proc_mounts_operations = {
.open = mounts_open,
.read_iter = seq_read_iter,
- .splice_read = generic_file_splice_read,
+ .splice_read = copy_splice_read,
.llseek = seq_lseek,
.release = mounts_release,
.poll = mounts_poll,
@@ -333,7 +333,7 @@ const struct file_operations proc_mounts_operations = {
const struct file_operations proc_mountinfo_operations = {
.open = mountinfo_open,
.read_iter = seq_read_iter,
- .splice_read = generic_file_splice_read,
+ .splice_read = copy_splice_read,
.llseek = seq_lseek,
.release = mounts_release,
.poll = mounts_poll,
@@ -342,7 +342,7 @@ const struct file_operations proc_mountinfo_operations = {
const struct file_operations proc_mountstats_operations = {
.open = mountstats_open,
.read_iter = seq_read_iter,
- .splice_read = generic_file_splice_read,
+ .splice_read = copy_splice_read,
.llseek = seq_lseek,
.release = mounts_release,
};


2023-05-20 01:22:54

by David Howells

[permalink] [raw]
Subject: [PATCH v21 28/30] splice: Use filemap_splice_read() instead of generic_file_splice_read()

Replace pointers to generic_file_splice_read() with calls to
filemap_splice_read().

Signed-off-by: David Howells <[email protected]>
cc: Christoph Hellwig <[email protected]>
cc: Jens Axboe <[email protected]>
cc: Al Viro <[email protected]>
cc: David Hildenbrand <[email protected]>
cc: John Hubbard <[email protected]>
cc: [email protected]
cc: [email protected]
cc: [email protected]
---
block/fops.c | 2 +-
fs/adfs/file.c | 2 +-
fs/affs/file.c | 2 +-
fs/afs/file.c | 2 +-
fs/bfs/file.c | 2 +-
fs/btrfs/file.c | 2 +-
fs/cramfs/inode.c | 2 +-
fs/ecryptfs/file.c | 4 ++--
fs/erofs/data.c | 2 +-
fs/exfat/file.c | 2 +-
fs/ext2/file.c | 2 +-
fs/ext4/file.c | 2 +-
fs/fat/file.c | 2 +-
fs/fuse/file.c | 2 +-
fs/gfs2/file.c | 4 ++--
fs/hfs/inode.c | 2 +-
fs/hfsplus/inode.c | 2 +-
fs/hostfs/hostfs_kern.c | 2 +-
fs/hpfs/file.c | 2 +-
fs/jffs2/file.c | 2 +-
fs/jfs/file.c | 2 +-
fs/minix/file.c | 2 +-
fs/nilfs2/file.c | 2 +-
fs/ntfs/file.c | 2 +-
fs/ntfs3/file.c | 2 +-
fs/ocfs2/file.c | 4 ++--
fs/omfs/file.c | 2 +-
fs/ramfs/file-mmu.c | 2 +-
fs/ramfs/file-nommu.c | 2 +-
fs/read_write.c | 2 +-
fs/reiserfs/file.c | 2 +-
fs/romfs/mmap-nommu.c | 2 +-
fs/sysv/file.c | 2 +-
fs/ubifs/file.c | 2 +-
fs/udf/file.c | 2 +-
fs/ufs/file.c | 2 +-
fs/vboxsf/file.c | 2 +-
37 files changed, 40 insertions(+), 40 deletions(-)

diff --git a/block/fops.c b/block/fops.c
index d2e6be4e3d1c..6c9aa028af6e 100644
--- a/block/fops.c
+++ b/block/fops.c
@@ -691,7 +691,7 @@ const struct file_operations def_blk_fops = {
#ifdef CONFIG_COMPAT
.compat_ioctl = compat_blkdev_ioctl,
#endif
- .splice_read = generic_file_splice_read,
+ .splice_read = filemap_splice_read,
.splice_write = iter_file_splice_write,
.fallocate = blkdev_fallocate,
};
diff --git a/fs/adfs/file.c b/fs/adfs/file.c
index 754afb14a6ff..ee80718aaeec 100644
--- a/fs/adfs/file.c
+++ b/fs/adfs/file.c
@@ -28,7 +28,7 @@ const struct file_operations adfs_file_operations = {
.mmap = generic_file_mmap,
.fsync = generic_file_fsync,
.write_iter = generic_file_write_iter,
- .splice_read = generic_file_splice_read,
+ .splice_read = filemap_splice_read,
};

const struct inode_operations adfs_file_inode_operations = {
diff --git a/fs/affs/file.c b/fs/affs/file.c
index 8daeed31e1af..e43f2f007ac1 100644
--- a/fs/affs/file.c
+++ b/fs/affs/file.c
@@ -1001,7 +1001,7 @@ const struct file_operations affs_file_operations = {
.open = affs_file_open,
.release = affs_file_release,
.fsync = affs_file_fsync,
- .splice_read = generic_file_splice_read,
+ .splice_read = filemap_splice_read,
};

const struct inode_operations affs_file_inode_operations = {
diff --git a/fs/afs/file.c b/fs/afs/file.c
index d8a6b09dadf7..d37dd201752b 100644
--- a/fs/afs/file.c
+++ b/fs/afs/file.c
@@ -603,5 +603,5 @@ static ssize_t afs_file_splice_read(struct file *in, loff_t *ppos,
if (ret < 0)
return ret;

- return generic_file_splice_read(in, ppos, pipe, len, flags);
+ return filemap_splice_read(in, ppos, pipe, len, flags);
}
diff --git a/fs/bfs/file.c b/fs/bfs/file.c
index 57ae5ee6deec..adc2230079c6 100644
--- a/fs/bfs/file.c
+++ b/fs/bfs/file.c
@@ -27,7 +27,7 @@ const struct file_operations bfs_file_operations = {
.read_iter = generic_file_read_iter,
.write_iter = generic_file_write_iter,
.mmap = generic_file_mmap,
- .splice_read = generic_file_splice_read,
+ .splice_read = filemap_splice_read,
};

static int bfs_move_block(unsigned long from, unsigned long to,
diff --git a/fs/btrfs/file.c b/fs/btrfs/file.c
index f649647392e0..71426c6408fa 100644
--- a/fs/btrfs/file.c
+++ b/fs/btrfs/file.c
@@ -3825,7 +3825,7 @@ static ssize_t btrfs_file_read_iter(struct kiocb *iocb, struct iov_iter *to)
const struct file_operations btrfs_file_operations = {
.llseek = btrfs_file_llseek,
.read_iter = btrfs_file_read_iter,
- .splice_read = generic_file_splice_read,
+ .splice_read = filemap_splice_read,
.write_iter = btrfs_file_write_iter,
.splice_write = iter_file_splice_write,
.mmap = btrfs_file_mmap,
diff --git a/fs/cramfs/inode.c b/fs/cramfs/inode.c
index 006ef68d7ff6..27c6597aa1be 100644
--- a/fs/cramfs/inode.c
+++ b/fs/cramfs/inode.c
@@ -473,7 +473,7 @@ static unsigned int cramfs_physmem_mmap_capabilities(struct file *file)
static const struct file_operations cramfs_physmem_fops = {
.llseek = generic_file_llseek,
.read_iter = generic_file_read_iter,
- .splice_read = generic_file_splice_read,
+ .splice_read = filemap_splice_read,
.mmap = cramfs_physmem_mmap,
#ifndef CONFIG_MMU
.get_unmapped_area = cramfs_physmem_get_unmapped_area,
diff --git a/fs/ecryptfs/file.c b/fs/ecryptfs/file.c
index 284395587be0..ce0a3c5ed0ca 100644
--- a/fs/ecryptfs/file.c
+++ b/fs/ecryptfs/file.c
@@ -47,7 +47,7 @@ static ssize_t ecryptfs_read_update_atime(struct kiocb *iocb,
/*
* ecryptfs_splice_read_update_atime
*
- * generic_file_splice_read updates the atime of upper layer inode. But, it
+ * filemap_splice_read updates the atime of upper layer inode. But, it
* doesn't give us a chance to update the atime of the lower layer inode. This
* function is a wrapper to generic_file_read. It updates the atime of the
* lower level inode if generic_file_read returns without any errors. This is
@@ -61,7 +61,7 @@ static ssize_t ecryptfs_splice_read_update_atime(struct file *in, loff_t *ppos,
ssize_t rc;
const struct path *path;

- rc = generic_file_splice_read(in, ppos, pipe, len, flags);
+ rc = filemap_splice_read(in, ppos, pipe, len, flags);
if (rc >= 0) {
path = ecryptfs_dentry_to_lower_path(in->f_path.dentry);
touch_atime(path);
diff --git a/fs/erofs/data.c b/fs/erofs/data.c
index 6fe9a779fa91..db5e4b7636ec 100644
--- a/fs/erofs/data.c
+++ b/fs/erofs/data.c
@@ -448,5 +448,5 @@ const struct file_operations erofs_file_fops = {
.llseek = generic_file_llseek,
.read_iter = erofs_file_read_iter,
.mmap = erofs_file_mmap,
- .splice_read = generic_file_splice_read,
+ .splice_read = filemap_splice_read,
};
diff --git a/fs/exfat/file.c b/fs/exfat/file.c
index e99183a74611..3cbd270e0cba 100644
--- a/fs/exfat/file.c
+++ b/fs/exfat/file.c
@@ -389,7 +389,7 @@ const struct file_operations exfat_file_operations = {
#endif
.mmap = generic_file_mmap,
.fsync = exfat_file_fsync,
- .splice_read = generic_file_splice_read,
+ .splice_read = filemap_splice_read,
.splice_write = iter_file_splice_write,
};

diff --git a/fs/ext2/file.c b/fs/ext2/file.c
index 6b4bebe982ca..d1ae0f0a3726 100644
--- a/fs/ext2/file.c
+++ b/fs/ext2/file.c
@@ -192,7 +192,7 @@ const struct file_operations ext2_file_operations = {
.release = ext2_release_file,
.fsync = ext2_fsync,
.get_unmapped_area = thp_get_unmapped_area,
- .splice_read = generic_file_splice_read,
+ .splice_read = filemap_splice_read,
.splice_write = iter_file_splice_write,
};

diff --git a/fs/ext4/file.c b/fs/ext4/file.c
index 9f8bbd9d131c..e8261900f4f3 100644
--- a/fs/ext4/file.c
+++ b/fs/ext4/file.c
@@ -155,7 +155,7 @@ static ssize_t ext4_file_splice_read(struct file *in, loff_t *ppos,

if (unlikely(ext4_forced_shutdown(EXT4_SB(inode->i_sb))))
return -EIO;
- return generic_file_splice_read(in, ppos, pipe, len, flags);
+ return filemap_splice_read(in, ppos, pipe, len, flags);
}

/*
diff --git a/fs/fat/file.c b/fs/fat/file.c
index 795a4fad5c40..456477946dd9 100644
--- a/fs/fat/file.c
+++ b/fs/fat/file.c
@@ -209,7 +209,7 @@ const struct file_operations fat_file_operations = {
.unlocked_ioctl = fat_generic_ioctl,
.compat_ioctl = compat_ptr_ioctl,
.fsync = fat_file_fsync,
- .splice_read = generic_file_splice_read,
+ .splice_read = filemap_splice_read,
.splice_write = iter_file_splice_write,
.fallocate = fat_fallocate,
};
diff --git a/fs/fuse/file.c b/fs/fuse/file.c
index 89d97f6188e0..4553124f5406 100644
--- a/fs/fuse/file.c
+++ b/fs/fuse/file.c
@@ -3252,7 +3252,7 @@ static const struct file_operations fuse_file_operations = {
.lock = fuse_file_lock,
.get_unmapped_area = thp_get_unmapped_area,
.flock = fuse_file_flock,
- .splice_read = generic_file_splice_read,
+ .splice_read = filemap_splice_read,
.splice_write = iter_file_splice_write,
.unlocked_ioctl = fuse_file_ioctl,
.compat_ioctl = fuse_file_compat_ioctl,
diff --git a/fs/gfs2/file.c b/fs/gfs2/file.c
index 300844f50dcd..0f5ad5165361 100644
--- a/fs/gfs2/file.c
+++ b/fs/gfs2/file.c
@@ -1568,7 +1568,7 @@ const struct file_operations gfs2_file_fops = {
.fsync = gfs2_fsync,
.lock = gfs2_lock,
.flock = gfs2_flock,
- .splice_read = generic_file_splice_read,
+ .splice_read = filemap_splice_read,
.splice_write = gfs2_file_splice_write,
.setlease = simple_nosetlease,
.fallocate = gfs2_fallocate,
@@ -1599,7 +1599,7 @@ const struct file_operations gfs2_file_fops_nolock = {
.open = gfs2_open,
.release = gfs2_release,
.fsync = gfs2_fsync,
- .splice_read = generic_file_splice_read,
+ .splice_read = filemap_splice_read,
.splice_write = gfs2_file_splice_write,
.setlease = generic_setlease,
.fallocate = gfs2_fallocate,
diff --git a/fs/hfs/inode.c b/fs/hfs/inode.c
index 1f7bd068acf0..441d7fc952e3 100644
--- a/fs/hfs/inode.c
+++ b/fs/hfs/inode.c
@@ -694,7 +694,7 @@ static const struct file_operations hfs_file_operations = {
.read_iter = generic_file_read_iter,
.write_iter = generic_file_write_iter,
.mmap = generic_file_mmap,
- .splice_read = generic_file_splice_read,
+ .splice_read = filemap_splice_read,
.fsync = hfs_file_fsync,
.open = hfs_file_open,
.release = hfs_file_release,
diff --git a/fs/hfsplus/inode.c b/fs/hfsplus/inode.c
index b21660475ac1..7d1a675e037d 100644
--- a/fs/hfsplus/inode.c
+++ b/fs/hfsplus/inode.c
@@ -372,7 +372,7 @@ static const struct file_operations hfsplus_file_operations = {
.read_iter = generic_file_read_iter,
.write_iter = generic_file_write_iter,
.mmap = generic_file_mmap,
- .splice_read = generic_file_splice_read,
+ .splice_read = filemap_splice_read,
.fsync = hfsplus_file_fsync,
.open = hfsplus_file_open,
.release = hfsplus_file_release,
diff --git a/fs/hostfs/hostfs_kern.c b/fs/hostfs/hostfs_kern.c
index 28b4f15c19eb..87998df499f4 100644
--- a/fs/hostfs/hostfs_kern.c
+++ b/fs/hostfs/hostfs_kern.c
@@ -381,7 +381,7 @@ static int hostfs_fsync(struct file *file, loff_t start, loff_t end,

static const struct file_operations hostfs_file_fops = {
.llseek = generic_file_llseek,
- .splice_read = generic_file_splice_read,
+ .splice_read = filemap_splice_read,
.splice_write = iter_file_splice_write,
.read_iter = generic_file_read_iter,
.write_iter = generic_file_write_iter,
diff --git a/fs/hpfs/file.c b/fs/hpfs/file.c
index 88952d4a631e..1bb8d97cd9ae 100644
--- a/fs/hpfs/file.c
+++ b/fs/hpfs/file.c
@@ -259,7 +259,7 @@ const struct file_operations hpfs_file_ops =
.mmap = generic_file_mmap,
.release = hpfs_file_release,
.fsync = hpfs_file_fsync,
- .splice_read = generic_file_splice_read,
+ .splice_read = filemap_splice_read,
.unlocked_ioctl = hpfs_ioctl,
.compat_ioctl = compat_ptr_ioctl,
};
diff --git a/fs/jffs2/file.c b/fs/jffs2/file.c
index 96b0275ce957..2345ca3f09ee 100644
--- a/fs/jffs2/file.c
+++ b/fs/jffs2/file.c
@@ -56,7 +56,7 @@ const struct file_operations jffs2_file_operations =
.unlocked_ioctl=jffs2_ioctl,
.mmap = generic_file_readonly_mmap,
.fsync = jffs2_fsync,
- .splice_read = generic_file_splice_read,
+ .splice_read = filemap_splice_read,
.splice_write = iter_file_splice_write,
};

diff --git a/fs/jfs/file.c b/fs/jfs/file.c
index 2ee35be49de1..01b6912e60f8 100644
--- a/fs/jfs/file.c
+++ b/fs/jfs/file.c
@@ -144,7 +144,7 @@ const struct file_operations jfs_file_operations = {
.read_iter = generic_file_read_iter,
.write_iter = generic_file_write_iter,
.mmap = generic_file_mmap,
- .splice_read = generic_file_splice_read,
+ .splice_read = filemap_splice_read,
.splice_write = iter_file_splice_write,
.fsync = jfs_fsync,
.release = jfs_release,
diff --git a/fs/minix/file.c b/fs/minix/file.c
index 0dd05d47724a..906d192ab7f3 100644
--- a/fs/minix/file.c
+++ b/fs/minix/file.c
@@ -19,7 +19,7 @@ const struct file_operations minix_file_operations = {
.write_iter = generic_file_write_iter,
.mmap = generic_file_mmap,
.fsync = generic_file_fsync,
- .splice_read = generic_file_splice_read,
+ .splice_read = filemap_splice_read,
};

static int minix_setattr(struct mnt_idmap *idmap,
diff --git a/fs/nilfs2/file.c b/fs/nilfs2/file.c
index a265d391ffe9..a9eb3487efb2 100644
--- a/fs/nilfs2/file.c
+++ b/fs/nilfs2/file.c
@@ -140,7 +140,7 @@ const struct file_operations nilfs_file_operations = {
.open = generic_file_open,
/* .release = nilfs_release_file, */
.fsync = nilfs_sync_file,
- .splice_read = generic_file_splice_read,
+ .splice_read = filemap_splice_read,
.splice_write = iter_file_splice_write,
};

diff --git a/fs/ntfs/file.c b/fs/ntfs/file.c
index c481b14e4fd9..e5e0ed58670b 100644
--- a/fs/ntfs/file.c
+++ b/fs/ntfs/file.c
@@ -1992,7 +1992,7 @@ const struct file_operations ntfs_file_ops = {
#endif /* NTFS_RW */
.mmap = generic_file_mmap,
.open = ntfs_file_open,
- .splice_read = generic_file_splice_read,
+ .splice_read = filemap_splice_read,
};

const struct inode_operations ntfs_file_inode_ops = {
diff --git a/fs/ntfs3/file.c b/fs/ntfs3/file.c
index 667c9dc68b58..036efd85f60c 100644
--- a/fs/ntfs3/file.c
+++ b/fs/ntfs3/file.c
@@ -770,7 +770,7 @@ static ssize_t ntfs_file_splice_read(struct file *in, loff_t *ppos,
return -EOPNOTSUPP;
}

- return generic_file_splice_read(in, ppos, pipe, len, flags);
+ return filemap_splice_read(in, ppos, pipe, len, flags);
}

/*
diff --git a/fs/ocfs2/file.c b/fs/ocfs2/file.c
index f7e00b5689d5..ff673c882816 100644
--- a/fs/ocfs2/file.c
+++ b/fs/ocfs2/file.c
@@ -2601,7 +2601,7 @@ static ssize_t ocfs2_file_splice_read(struct file *in, loff_t *ppos,
* rw_lock during read.
*
* Take and drop the meta data lock to update inode fields like i_size.
- * This allows the checks down below generic_file_splice_read() a
+ * This allows the checks down below filemap_splice_read() a
* chance of actually working.
*/
ret = ocfs2_inode_lock_atime(inode, in->f_path.mnt, &lock_level, true);
@@ -2827,7 +2827,7 @@ const struct file_operations ocfs2_fops_no_plocks = {
.compat_ioctl = ocfs2_compat_ioctl,
#endif
.flock = ocfs2_flock,
- .splice_read = generic_file_splice_read,
+ .splice_read = filemap_splice_read,
.splice_write = iter_file_splice_write,
.fallocate = ocfs2_fallocate,
.remap_file_range = ocfs2_remap_file_range,
diff --git a/fs/omfs/file.c b/fs/omfs/file.c
index 0101f1f87b56..de8f57ee39ec 100644
--- a/fs/omfs/file.c
+++ b/fs/omfs/file.c
@@ -334,7 +334,7 @@ const struct file_operations omfs_file_operations = {
.write_iter = generic_file_write_iter,
.mmap = generic_file_mmap,
.fsync = generic_file_fsync,
- .splice_read = generic_file_splice_read,
+ .splice_read = filemap_splice_read,
};

static int omfs_setattr(struct mnt_idmap *idmap,
diff --git a/fs/ramfs/file-mmu.c b/fs/ramfs/file-mmu.c
index 12af0490322f..c7a1aa3c882b 100644
--- a/fs/ramfs/file-mmu.c
+++ b/fs/ramfs/file-mmu.c
@@ -43,7 +43,7 @@ const struct file_operations ramfs_file_operations = {
.write_iter = generic_file_write_iter,
.mmap = generic_file_mmap,
.fsync = noop_fsync,
- .splice_read = generic_file_splice_read,
+ .splice_read = filemap_splice_read,
.splice_write = iter_file_splice_write,
.llseek = generic_file_llseek,
.get_unmapped_area = ramfs_mmu_get_unmapped_area,
diff --git a/fs/ramfs/file-nommu.c b/fs/ramfs/file-nommu.c
index 9fbb9b5256f7..efb1b4c1a0a4 100644
--- a/fs/ramfs/file-nommu.c
+++ b/fs/ramfs/file-nommu.c
@@ -43,7 +43,7 @@ const struct file_operations ramfs_file_operations = {
.read_iter = generic_file_read_iter,
.write_iter = generic_file_write_iter,
.fsync = noop_fsync,
- .splice_read = generic_file_splice_read,
+ .splice_read = filemap_splice_read,
.splice_write = iter_file_splice_write,
.llseek = generic_file_llseek,
};
diff --git a/fs/read_write.c b/fs/read_write.c
index a21ba3be7dbe..b07de77ef126 100644
--- a/fs/read_write.c
+++ b/fs/read_write.c
@@ -29,7 +29,7 @@ const struct file_operations generic_ro_fops = {
.llseek = generic_file_llseek,
.read_iter = generic_file_read_iter,
.mmap = generic_file_readonly_mmap,
- .splice_read = generic_file_splice_read,
+ .splice_read = filemap_splice_read,
};

EXPORT_SYMBOL(generic_ro_fops);
diff --git a/fs/reiserfs/file.c b/fs/reiserfs/file.c
index b54cc7048f02..8eb3ad3e8ae9 100644
--- a/fs/reiserfs/file.c
+++ b/fs/reiserfs/file.c
@@ -247,7 +247,7 @@ const struct file_operations reiserfs_file_operations = {
.fsync = reiserfs_sync_file,
.read_iter = generic_file_read_iter,
.write_iter = generic_file_write_iter,
- .splice_read = generic_file_splice_read,
+ .splice_read = filemap_splice_read,
.splice_write = iter_file_splice_write,
.llseek = generic_file_llseek,
};
diff --git a/fs/romfs/mmap-nommu.c b/fs/romfs/mmap-nommu.c
index 4578dc45e50a..4520ca413867 100644
--- a/fs/romfs/mmap-nommu.c
+++ b/fs/romfs/mmap-nommu.c
@@ -78,7 +78,7 @@ static unsigned romfs_mmap_capabilities(struct file *file)
const struct file_operations romfs_ro_fops = {
.llseek = generic_file_llseek,
.read_iter = generic_file_read_iter,
- .splice_read = generic_file_splice_read,
+ .splice_read = filemap_splice_read,
.mmap = romfs_mmap,
.get_unmapped_area = romfs_get_unmapped_area,
.mmap_capabilities = romfs_mmap_capabilities,
diff --git a/fs/sysv/file.c b/fs/sysv/file.c
index 50eb92557a0f..c645f60bdb7f 100644
--- a/fs/sysv/file.c
+++ b/fs/sysv/file.c
@@ -26,7 +26,7 @@ const struct file_operations sysv_file_operations = {
.write_iter = generic_file_write_iter,
.mmap = generic_file_mmap,
.fsync = generic_file_fsync,
- .splice_read = generic_file_splice_read,
+ .splice_read = filemap_splice_read,
};

static int sysv_setattr(struct mnt_idmap *idmap,
diff --git a/fs/ubifs/file.c b/fs/ubifs/file.c
index 979ab1d9d0c3..6738fe43040b 100644
--- a/fs/ubifs/file.c
+++ b/fs/ubifs/file.c
@@ -1669,7 +1669,7 @@ const struct file_operations ubifs_file_operations = {
.mmap = ubifs_file_mmap,
.fsync = ubifs_fsync,
.unlocked_ioctl = ubifs_ioctl,
- .splice_read = generic_file_splice_read,
+ .splice_read = filemap_splice_read,
.splice_write = iter_file_splice_write,
.open = fscrypt_file_open,
#ifdef CONFIG_COMPAT
diff --git a/fs/udf/file.c b/fs/udf/file.c
index 8238f742377b..29daf5d5cb67 100644
--- a/fs/udf/file.c
+++ b/fs/udf/file.c
@@ -209,7 +209,7 @@ const struct file_operations udf_file_operations = {
.write_iter = udf_file_write_iter,
.release = udf_release_file,
.fsync = generic_file_fsync,
- .splice_read = generic_file_splice_read,
+ .splice_read = filemap_splice_read,
.splice_write = iter_file_splice_write,
.llseek = generic_file_llseek,
};
diff --git a/fs/ufs/file.c b/fs/ufs/file.c
index 7e087581be7e..6558882a89ef 100644
--- a/fs/ufs/file.c
+++ b/fs/ufs/file.c
@@ -41,5 +41,5 @@ const struct file_operations ufs_file_operations = {
.mmap = generic_file_mmap,
.open = generic_file_open,
.fsync = generic_file_fsync,
- .splice_read = generic_file_splice_read,
+ .splice_read = filemap_splice_read,
};
diff --git a/fs/vboxsf/file.c b/fs/vboxsf/file.c
index 572aa1c43b37..2307f8037efc 100644
--- a/fs/vboxsf/file.c
+++ b/fs/vboxsf/file.c
@@ -217,7 +217,7 @@ const struct file_operations vboxsf_reg_fops = {
.open = vboxsf_file_open,
.release = vboxsf_file_release,
.fsync = noop_fsync,
- .splice_read = generic_file_splice_read,
+ .splice_read = filemap_splice_read,
};

const struct inode_operations vboxsf_reg_iops = {


2023-05-20 01:23:52

by David Howells

[permalink] [raw]
Subject: [PATCH v21 27/30] cifs: Use filemap_splice_read()

Make cifs use filemap_splice_read() rather than doing its own version of
generic_file_splice_read().

Signed-off-by: David Howells <[email protected]>
Reviewed-by: Christoph Hellwig <[email protected]>
Reviewed-by: Paulo Alcantara (SUSE) <[email protected]>
cc: Jens Axboe <[email protected]>
cc: Steve French <[email protected]>
cc: Al Viro <[email protected]>
cc: David Hildenbrand <[email protected]>
cc: John Hubbard <[email protected]>
cc: [email protected]
cc: [email protected]
cc: [email protected]
cc: [email protected]
---

Notes:
ver #21)
- Switch to filemap_splice_read() rather than generic_file_splice_read().

ver #20)
- Don't remove the export of filemap_splice_read().

ver #18)
- Split out from change to generic_file_splice_read().

fs/cifs/cifsfs.c | 8 ++++----
fs/cifs/cifsfs.h | 3 ---
fs/cifs/file.c | 16 ----------------
3 files changed, 4 insertions(+), 23 deletions(-)

diff --git a/fs/cifs/cifsfs.c b/fs/cifs/cifsfs.c
index fa2477bbcc86..4f4492eb975f 100644
--- a/fs/cifs/cifsfs.c
+++ b/fs/cifs/cifsfs.c
@@ -1376,7 +1376,7 @@ const struct file_operations cifs_file_ops = {
.fsync = cifs_fsync,
.flush = cifs_flush,
.mmap = cifs_file_mmap,
- .splice_read = cifs_splice_read,
+ .splice_read = filemap_splice_read,
.splice_write = iter_file_splice_write,
.llseek = cifs_llseek,
.unlocked_ioctl = cifs_ioctl,
@@ -1396,7 +1396,7 @@ const struct file_operations cifs_file_strict_ops = {
.fsync = cifs_strict_fsync,
.flush = cifs_flush,
.mmap = cifs_file_strict_mmap,
- .splice_read = cifs_splice_read,
+ .splice_read = filemap_splice_read,
.splice_write = iter_file_splice_write,
.llseek = cifs_llseek,
.unlocked_ioctl = cifs_ioctl,
@@ -1434,7 +1434,7 @@ const struct file_operations cifs_file_nobrl_ops = {
.fsync = cifs_fsync,
.flush = cifs_flush,
.mmap = cifs_file_mmap,
- .splice_read = cifs_splice_read,
+ .splice_read = filemap_splice_read,
.splice_write = iter_file_splice_write,
.llseek = cifs_llseek,
.unlocked_ioctl = cifs_ioctl,
@@ -1452,7 +1452,7 @@ const struct file_operations cifs_file_strict_nobrl_ops = {
.fsync = cifs_strict_fsync,
.flush = cifs_flush,
.mmap = cifs_file_strict_mmap,
- .splice_read = cifs_splice_read,
+ .splice_read = filemap_splice_read,
.splice_write = iter_file_splice_write,
.llseek = cifs_llseek,
.unlocked_ioctl = cifs_ioctl,
diff --git a/fs/cifs/cifsfs.h b/fs/cifs/cifsfs.h
index 74cd6fafb33e..d7274eefc666 100644
--- a/fs/cifs/cifsfs.h
+++ b/fs/cifs/cifsfs.h
@@ -100,9 +100,6 @@ extern ssize_t cifs_strict_readv(struct kiocb *iocb, struct iov_iter *to);
extern ssize_t cifs_user_writev(struct kiocb *iocb, struct iov_iter *from);
extern ssize_t cifs_direct_writev(struct kiocb *iocb, struct iov_iter *from);
extern ssize_t cifs_strict_writev(struct kiocb *iocb, struct iov_iter *from);
-extern ssize_t cifs_splice_read(struct file *in, loff_t *ppos,
- struct pipe_inode_info *pipe, size_t len,
- unsigned int flags);
extern int cifs_flock(struct file *pfile, int cmd, struct file_lock *plock);
extern int cifs_lock(struct file *, int, struct file_lock *);
extern int cifs_fsync(struct file *, loff_t, loff_t, int);
diff --git a/fs/cifs/file.c b/fs/cifs/file.c
index 023496207c18..375a8037a3f3 100644
--- a/fs/cifs/file.c
+++ b/fs/cifs/file.c
@@ -5078,19 +5078,3 @@ const struct address_space_operations cifs_addr_ops_smallbuf = {
.launder_folio = cifs_launder_folio,
.migrate_folio = filemap_migrate_folio,
};
-
-/*
- * Splice data from a file into a pipe.
- */
-ssize_t cifs_splice_read(struct file *in, loff_t *ppos,
- struct pipe_inode_info *pipe, size_t len,
- unsigned int flags)
-{
- if (unlikely(*ppos >= file_inode(in)->i_sb->s_maxbytes))
- return 0;
- if (unlikely(!len))
- return 0;
- if (in->f_flags & O_DIRECT)
- return copy_splice_read(in, ppos, pipe, len, flags);
- return filemap_splice_read(in, ppos, pipe, len, flags);
-}


2023-05-20 01:26:33

by David Howells

[permalink] [raw]
Subject: [PATCH v21 11/30] coda: Implement splice-read

Implement splice-read for coda by passing the request down a layer rather
than going through generic_file_splice_read() which is going to be changed
to assume that ->read_folio() is present on buffered files.

Signed-off-by: David Howells <[email protected]>
Acked-by: Jan Harkes <[email protected]>
cc: Christoph Hellwig <[email protected]>
cc: Jens Axboe <[email protected]>
cc: Al Viro <[email protected]>
cc: John Hubbard <[email protected]>
cc: David Hildenbrand <[email protected]>
cc: Matthew Wilcox <[email protected]>
cc: [email protected]
cc: [email protected]
cc: [email protected]
cc: [email protected]
cc: [email protected]
cc: [email protected]
---

Notes:
ver #17)
- Use vfs_splice_read() helper rather than open-coding checks.

fs/coda/file.c | 29 ++++++++++++++++++++++++++++-
1 file changed, 28 insertions(+), 1 deletion(-)

diff --git a/fs/coda/file.c b/fs/coda/file.c
index 3f3c81e6b1ab..12b26bd13564 100644
--- a/fs/coda/file.c
+++ b/fs/coda/file.c
@@ -23,6 +23,7 @@
#include <linux/slab.h>
#include <linux/uaccess.h>
#include <linux/uio.h>
+#include <linux/splice.h>

#include <linux/coda.h>
#include "coda_psdev.h"
@@ -94,6 +95,32 @@ coda_file_write_iter(struct kiocb *iocb, struct iov_iter *to)
return ret;
}

+static ssize_t
+coda_file_splice_read(struct file *coda_file, loff_t *ppos,
+ struct pipe_inode_info *pipe,
+ size_t len, unsigned int flags)
+{
+ struct inode *coda_inode = file_inode(coda_file);
+ struct coda_file_info *cfi = coda_ftoc(coda_file);
+ struct file *in = cfi->cfi_container;
+ loff_t ki_pos = *ppos;
+ ssize_t ret;
+
+ ret = venus_access_intent(coda_inode->i_sb, coda_i2f(coda_inode),
+ &cfi->cfi_access_intent,
+ len, ki_pos, CODA_ACCESS_TYPE_READ);
+ if (ret)
+ goto finish_read;
+
+ ret = vfs_splice_read(in, ppos, pipe, len, flags);
+
+finish_read:
+ venus_access_intent(coda_inode->i_sb, coda_i2f(coda_inode),
+ &cfi->cfi_access_intent,
+ len, ki_pos, CODA_ACCESS_TYPE_READ_FINISH);
+ return ret;
+}
+
static void
coda_vm_open(struct vm_area_struct *vma)
{
@@ -302,5 +329,5 @@ const struct file_operations coda_file_operations = {
.open = coda_open,
.release = coda_release,
.fsync = coda_fsync,
- .splice_read = generic_file_splice_read,
+ .splice_read = coda_file_splice_read,
};


2023-05-20 01:28:42

by David Howells

[permalink] [raw]
Subject: [PATCH v21 01/30] splice: Fix filemap of a blockdev

Fix filemap_splice_read() to use file->f_mapping->host, not file->f_inode,
as the source of the file size because in the case of a block device,
file->f_inode points to the block-special file (which is typically 0
length) and not the backing store.

Fixes: 07073eb01c5f ("splice: Add a func to do a splice from a buffered file without ITER_PIPE")
Signed-off-by: David Howells <[email protected]>
Reviewed-by: Christoph Hellwig <[email protected]>
cc: Steve French <[email protected]>
cc: Jens Axboe <[email protected]>
cc: Al Viro <[email protected]>
cc: David Hildenbrand <[email protected]>
cc: John Hubbard <[email protected]>
cc: [email protected]
cc: [email protected]
cc: [email protected]
---
mm/filemap.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/mm/filemap.c b/mm/filemap.c
index b4c9bd368b7e..a2006936a6ae 100644
--- a/mm/filemap.c
+++ b/mm/filemap.c
@@ -2900,7 +2900,7 @@ ssize_t filemap_splice_read(struct file *in, loff_t *ppos,
do {
cond_resched();

- if (*ppos >= i_size_read(file_inode(in)))
+ if (*ppos >= i_size_read(in->f_mapping->host))
break;

iocb.ki_pos = *ppos;
@@ -2916,7 +2916,7 @@ ssize_t filemap_splice_read(struct file *in, loff_t *ppos,
* part of the page is not copied back to userspace (unless
* another truncate extends the file - this is desired though).
*/
- isize = i_size_read(file_inode(in));
+ isize = i_size_read(in->f_mapping->host);
if (unlikely(*ppos >= isize))
break;
end_offset = min_t(loff_t, isize, *ppos + len);


2023-05-20 01:28:53

by David Howells

[permalink] [raw]
Subject: [PATCH v21 14/30] 9p: Add splice_read stub

Add a splice_read stub for 9p. We should use copy_splice_read() if
9PL_DIRECT is set and filemap_splice_read() otherwise. Note that this
doesn't seem to be particularly related to O_DIRECT.

Signed-off-by: David Howells <[email protected]>
cc: Christoph Hellwig <[email protected]>
cc: Al Viro <[email protected]>
cc: Jens Axboe <[email protected]>
cc: Dominique Martinet <[email protected]>
cc: Eric Van Hensbergen <[email protected]>
cc: Latchesar Ionkov <[email protected]>
cc: Christian Schoenebeck <[email protected]>
cc: [email protected]
cc: [email protected]
cc: [email protected]
cc: [email protected]
---
fs/9p/vfs_file.c | 26 ++++++++++++++++++++++++--
1 file changed, 24 insertions(+), 2 deletions(-)

diff --git a/fs/9p/vfs_file.c b/fs/9p/vfs_file.c
index 6c31b8c8112d..2996fb00387f 100644
--- a/fs/9p/vfs_file.c
+++ b/fs/9p/vfs_file.c
@@ -374,6 +374,28 @@ v9fs_file_read_iter(struct kiocb *iocb, struct iov_iter *to)
return ret;
}

+/*
+ * v9fs_file_splice_read - splice-read from a file
+ * @in: The 9p file to read from
+ * @ppos: Where to find/update the file position
+ * @pipe: The pipe to splice into
+ * @len: The maximum amount of data to splice
+ * @flags: SPLICE_F_* flags
+ */
+static ssize_t v9fs_file_splice_read(struct file *in, loff_t *ppos,
+ struct pipe_inode_info *pipe,
+ size_t len, unsigned int flags)
+{
+ struct p9_fid *fid = in->private_data;
+
+ p9_debug(P9_DEBUG_VFS, "fid %d count %zu offset %lld\n",
+ fid->fid, len, *ppos);
+
+ if (fid->mode & P9L_DIRECT)
+ return copy_splice_read(in, ppos, pipe, len, flags);
+ return filemap_splice_read(in, ppos, pipe, len, flags);
+}
+
/**
* v9fs_file_write_iter - write to a file
* @iocb: The operation parameters
@@ -569,7 +591,7 @@ const struct file_operations v9fs_file_operations = {
.release = v9fs_dir_release,
.lock = v9fs_file_lock,
.mmap = generic_file_readonly_mmap,
- .splice_read = generic_file_splice_read,
+ .splice_read = v9fs_file_splice_read,
.splice_write = iter_file_splice_write,
.fsync = v9fs_file_fsync,
};
@@ -583,7 +605,7 @@ const struct file_operations v9fs_file_operations_dotl = {
.lock = v9fs_file_lock_dotl,
.flock = v9fs_file_flock_dotl,
.mmap = v9fs_file_mmap,
- .splice_read = generic_file_splice_read,
+ .splice_read = v9fs_file_splice_read,
.splice_write = iter_file_splice_write,
.fsync = v9fs_file_fsync_dotl,
};


2023-05-20 01:34:32

by David Howells

[permalink] [raw]
Subject: [PATCH v21 19/30] f2fs: Provide a splice-read stub

Provide a splice_read stub for f2fs. This does some checks and tracing
before calling filemap_splice_read() and will update the iostats
afterwards. Direct I/O is handled by the caller.

Signed-off-by: David Howells <[email protected]>
cc: Christoph Hellwig <[email protected]>
cc: Al Viro <[email protected]>
cc: Jens Axboe <[email protected]>
cc: Jaegeuk Kim <[email protected]>
cc: Chao Yu <[email protected]>
cc: [email protected]
cc: [email protected]
cc: [email protected]
cc: [email protected]
---
fs/f2fs/file.c | 43 +++++++++++++++++++++++++++++++++++--------
1 file changed, 35 insertions(+), 8 deletions(-)

diff --git a/fs/f2fs/file.c b/fs/f2fs/file.c
index 5ac53d2627d2..3fce122997ca 100644
--- a/fs/f2fs/file.c
+++ b/fs/f2fs/file.c
@@ -4367,22 +4367,23 @@ static ssize_t f2fs_dio_read_iter(struct kiocb *iocb, struct iov_iter *to)
return ret;
}

-static void f2fs_trace_rw_file_path(struct kiocb *iocb, size_t count, int rw)
+static void f2fs_trace_rw_file_path(struct file *file, loff_t pos, size_t count,
+ int rw)
{
- struct inode *inode = file_inode(iocb->ki_filp);
+ struct inode *inode = file_inode(file);
char *buf, *path;

buf = f2fs_getname(F2FS_I_SB(inode));
if (!buf)
return;
- path = dentry_path_raw(file_dentry(iocb->ki_filp), buf, PATH_MAX);
+ path = dentry_path_raw(file_dentry(file), buf, PATH_MAX);
if (IS_ERR(path))
goto free_buf;
if (rw == WRITE)
- trace_f2fs_datawrite_start(inode, iocb->ki_pos, count,
+ trace_f2fs_datawrite_start(inode, pos, count,
current->pid, path, current->comm);
else
- trace_f2fs_dataread_start(inode, iocb->ki_pos, count,
+ trace_f2fs_dataread_start(inode, pos, count,
current->pid, path, current->comm);
free_buf:
f2fs_putname(buf);
@@ -4398,7 +4399,8 @@ static ssize_t f2fs_file_read_iter(struct kiocb *iocb, struct iov_iter *to)
return -EOPNOTSUPP;

if (trace_f2fs_dataread_start_enabled())
- f2fs_trace_rw_file_path(iocb, iov_iter_count(to), READ);
+ f2fs_trace_rw_file_path(iocb->ki_filp, iocb->ki_pos,
+ iov_iter_count(to), READ);

if (f2fs_should_use_dio(inode, iocb, to)) {
ret = f2fs_dio_read_iter(iocb, to);
@@ -4413,6 +4415,30 @@ static ssize_t f2fs_file_read_iter(struct kiocb *iocb, struct iov_iter *to)
return ret;
}

+static ssize_t f2fs_file_splice_read(struct file *in, loff_t *ppos,
+ struct pipe_inode_info *pipe,
+ size_t len, unsigned int flags)
+{
+ struct inode *inode = file_inode(in);
+ const loff_t pos = *ppos;
+ ssize_t ret;
+
+ if (!f2fs_is_compress_backend_ready(inode))
+ return -EOPNOTSUPP;
+
+ if (trace_f2fs_dataread_start_enabled())
+ f2fs_trace_rw_file_path(in, pos, len, READ);
+
+ ret = filemap_splice_read(in, ppos, pipe, len, flags);
+ if (ret > 0)
+ f2fs_update_iostat(F2FS_I_SB(inode), inode,
+ APP_BUFFERED_READ_IO, ret);
+
+ if (trace_f2fs_dataread_end_enabled())
+ trace_f2fs_dataread_end(inode, pos, ret);
+ return ret;
+}
+
static ssize_t f2fs_write_checks(struct kiocb *iocb, struct iov_iter *from)
{
struct file *file = iocb->ki_filp;
@@ -4714,7 +4740,8 @@ static ssize_t f2fs_file_write_iter(struct kiocb *iocb, struct iov_iter *from)
ret = preallocated;
} else {
if (trace_f2fs_datawrite_start_enabled())
- f2fs_trace_rw_file_path(iocb, orig_count, WRITE);
+ f2fs_trace_rw_file_path(iocb->ki_filp, iocb->ki_pos,
+ orig_count, WRITE);

/* Do the actual write. */
ret = dio ?
@@ -4919,7 +4946,7 @@ const struct file_operations f2fs_file_operations = {
#ifdef CONFIG_COMPAT
.compat_ioctl = f2fs_compat_ioctl,
#endif
- .splice_read = generic_file_splice_read,
+ .splice_read = f2fs_file_splice_read,
.splice_write = iter_file_splice_write,
.fadvise = f2fs_file_fadvise,
};


2023-05-20 01:35:56

by David Howells

[permalink] [raw]
Subject: [PATCH v21 04/30] splice: Clean up copy_splice_read() a bit

Do a couple of cleanups to copy_splice_read():

(1) Cast to struct page **, not void *.

(2) Simplify the calculation of the number of pages to keep/reclaim in
copy_splice_read().

Suggested-by: Christoph Hellwig <[email protected]>
Signed-off-by: David Howells <[email protected]>
Reviewed-by: Christoph Hellwig <[email protected]>
cc: Jens Axboe <[email protected]>
cc: Al Viro <[email protected]>
cc: David Hildenbrand <[email protected]>
cc: John Hubbard <[email protected]>
cc: [email protected]
cc: [email protected]
cc: [email protected]
---

Notes:
ver #21)
- direct_splice_read() got renamed to copy_splice_read().

fs/splice.c | 19 +++++++------------
1 file changed, 7 insertions(+), 12 deletions(-)

diff --git a/fs/splice.c b/fs/splice.c
index 2478e065bc53..f9a9be797b0c 100644
--- a/fs/splice.c
+++ b/fs/splice.c
@@ -311,7 +311,7 @@ ssize_t copy_splice_read(struct file *in, loff_t *ppos,
struct kiocb kiocb;
struct page **pages;
ssize_t ret;
- size_t used, npages, chunk, remain, reclaim;
+ size_t used, npages, chunk, remain, keep = 0;
int i;

/* Work out how much data we can actually add into the pipe */
@@ -325,7 +325,7 @@ ssize_t copy_splice_read(struct file *in, loff_t *ppos,
if (!bv)
return -ENOMEM;

- pages = (void *)(bv + npages);
+ pages = (struct page **)(bv + npages);
npages = alloc_pages_bulk_array(GFP_USER, npages, pages);
if (!npages) {
kfree(bv);
@@ -348,11 +348,8 @@ ssize_t copy_splice_read(struct file *in, loff_t *ppos,
kiocb.ki_pos = *ppos;
ret = call_read_iter(in, &kiocb, &to);

- reclaim = npages * PAGE_SIZE;
- remain = 0;
if (ret > 0) {
- reclaim -= ret;
- remain = ret;
+ keep = DIV_ROUND_UP(ret, PAGE_SIZE);
*ppos = kiocb.ki_pos;
file_accessed(in);
} else if (ret < 0) {
@@ -365,14 +362,12 @@ ssize_t copy_splice_read(struct file *in, loff_t *ppos,
}

/* Free any pages that didn't get touched at all. */
- reclaim /= PAGE_SIZE;
- if (reclaim) {
- npages -= reclaim;
- release_pages(pages + npages, reclaim);
- }
+ if (keep < npages)
+ release_pages(pages + keep, npages - keep);

/* Push the remaining pages into the pipe. */
- for (i = 0; i < npages; i++) {
+ remain = ret;
+ for (i = 0; i < keep; i++) {
struct pipe_buffer *buf = pipe_head_buf(pipe);

chunk = min_t(size_t, remain, PAGE_SIZE);


2023-05-20 04:41:22

by Christoph Hellwig

[permalink] [raw]

2023-05-20 04:43:47

by Christoph Hellwig

[permalink] [raw]
Subject: Re: [PATCH v21 01/30] splice: Fix filemap of a blockdev

Just noticed this now, but filemap of a blockdev sounds weird.

Maybe something like:

splice: use the correct inode in filemap_splice_read

might be better.


2023-05-20 09:17:13

by Christian Brauner

[permalink] [raw]
Subject: Re: [PATCH v21 01/30] splice: Fix filemap of a blockdev

On Sat, May 20, 2023 at 01:00:20AM +0100, David Howells wrote:
> Fix filemap_splice_read() to use file->f_mapping->host, not file->f_inode,
> as the source of the file size because in the case of a block device,
> file->f_inode points to the block-special file (which is typically 0
> length) and not the backing store.
>
> Fixes: 07073eb01c5f ("splice: Add a func to do a splice from a buffered file without ITER_PIPE")
> Signed-off-by: David Howells <[email protected]>
> Reviewed-by: Christoph Hellwig <[email protected]>
> cc: Steve French <[email protected]>
> cc: Jens Axboe <[email protected]>
> cc: Al Viro <[email protected]>
> cc: David Hildenbrand <[email protected]>
> cc: John Hubbard <[email protected]>
> cc: [email protected]
> cc: [email protected]
> cc: [email protected]
> ---

Reviewed-by: Christian Brauner <[email protected]>

2023-05-20 09:53:04

by Christian Brauner

[permalink] [raw]
Subject: Re: [PATCH v21 04/30] splice: Clean up copy_splice_read() a bit

On Sat, May 20, 2023 at 01:00:23AM +0100, David Howells wrote:
> Do a couple of cleanups to copy_splice_read():
>
> (1) Cast to struct page **, not void *.
>
> (2) Simplify the calculation of the number of pages to keep/reclaim in
> copy_splice_read().
>
> Suggested-by: Christoph Hellwig <[email protected]>
> Signed-off-by: David Howells <[email protected]>
> Reviewed-by: Christoph Hellwig <[email protected]>
> cc: Jens Axboe <[email protected]>
> cc: Al Viro <[email protected]>
> cc: David Hildenbrand <[email protected]>
> cc: John Hubbard <[email protected]>
> cc: [email protected]
> cc: [email protected]
> cc: [email protected]
> ---

Reviewed-by: Christian Brauner <[email protected]>

2023-05-20 09:54:39

by Christian Brauner

[permalink] [raw]
Subject: Re: [PATCH v21 05/30] splice: Make do_splice_to() generic and export it

On Sat, May 20, 2023 at 01:00:24AM +0100, David Howells wrote:
> Rename do_splice_to() to vfs_splice_read() and export it so that it can be
> used as a helper when calling down to a lower layer filesystem as it
> performs all the necessary checks[1].
>
> Signed-off-by: David Howells <[email protected]>
> Reviewed-by: Christoph Hellwig <[email protected]>
> cc: Miklos Szeredi <[email protected]>
> cc: Jens Axboe <[email protected]>
> cc: Al Viro <[email protected]>
> cc: John Hubbard <[email protected]>
> cc: David Hildenbrand <[email protected]>
> cc: Matthew Wilcox <[email protected]>
> cc: [email protected]
> cc: [email protected]
> cc: [email protected]
> cc: [email protected]
> Link: https://lore.kernel.org/r/CAJfpeguGksS3sCigmRi9hJdUec8qtM9f+_9jC1rJhsXT+dV01w@mail.gmail.com/ [1]
> ---

Reviewed-by: Christian Brauner <[email protected]>

2023-05-20 10:14:27

by Christian Brauner

[permalink] [raw]
Subject: Re: [PATCH v21 06/30] splice: Check for zero count in vfs_splice_read()

On Sat, May 20, 2023 at 01:00:25AM +0100, David Howells wrote:
> Make vfs_splice_read() return immediately if the length is 0.
>
> Signed-off-by: David Howells <[email protected]>
> cc: Christoph Hellwig <[email protected]>
> cc: Jens Axboe <[email protected]>
> cc: Al Viro <[email protected]>
> cc: [email protected]
> cc: [email protected]
> cc: [email protected]
> ---

Reviewed-by: Christian Brauner <[email protected]>

2023-05-20 10:51:56

by Christian Brauner

[permalink] [raw]
Subject: Re: [PATCH v21 29/30] splice: Remove generic_file_splice_read()

On Sat, May 20, 2023 at 01:00:48AM +0100, David Howells wrote:
> Remove generic_file_splice_read() as it has been replaced with calls to
> filemap_splice_read() and copy_splice_read().
>
> With this, ITER_PIPE is no longer used.
>
> Signed-off-by: David Howells <[email protected]>
> cc: Christoph Hellwig <[email protected]>
> cc: Jens Axboe <[email protected]>
> cc: Steve French <[email protected]>
> cc: Al Viro <[email protected]>
> cc: David Hildenbrand <[email protected]>
> cc: John Hubbard <[email protected]>
> cc: [email protected]
> cc: [email protected]
> cc: [email protected]
> cc: [email protected]
> ---

Reviewed-by: Christian Brauner <[email protected]>

2023-05-20 10:52:23

by Christian Brauner

[permalink] [raw]
Subject: Re: [PATCH v21 28/30] splice: Use filemap_splice_read() instead of generic_file_splice_read()

On Sat, May 20, 2023 at 01:00:47AM +0100, David Howells wrote:
> Replace pointers to generic_file_splice_read() with calls to
> filemap_splice_read().
>
> Signed-off-by: David Howells <[email protected]>
> cc: Christoph Hellwig <[email protected]>
> cc: Jens Axboe <[email protected]>
> cc: Al Viro <[email protected]>
> cc: David Hildenbrand <[email protected]>
> cc: John Hubbard <[email protected]>
> cc: [email protected]
> cc: [email protected]
> cc: [email protected]
> ---

Reviewed-by: Christian Brauner <[email protected]>

2023-05-21 12:36:06

by Masami Hiramatsu

[permalink] [raw]
Subject: Re: [PATCH v21 26/30] splice: Convert trace/seq to use copy_splice_read()

Hi David,

On Sat, 20 May 2023 01:00:45 +0100
David Howells <[email protected]> wrote:

> For the splice from the trace seq buffer, just use copy_splice_read().

So this is because you will remove generic_file_splice_read() (since
it's buggy), right?

>
> In the future, something better can probably be done by gifting pages from
> seq->buf into the pipe, but that would require changing seq->buf into a
> vmap over an array of pages.

So what we need is to introduce a vmap? We introduced splice support for
avoiding copy ringbuffer pages, but this drops it. Thus this will drop
performance of splice on ring buffer (trace file). If it is correct,
can you also add a note about that?

Thank you,

>
> Signed-off-by: David Howells <[email protected]>
> cc: Christoph Hellwig <[email protected]>
> cc: Al Viro <[email protected]>
> cc: Jens Axboe <[email protected]>
> cc: Steven Rostedt <[email protected]>
> cc: Masami Hiramatsu <[email protected]>
> cc: [email protected]
> cc: [email protected]
> cc: [email protected]
> cc: [email protected]
> cc: [email protected]
> ---
> kernel/trace/trace.c | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/kernel/trace/trace.c b/kernel/trace/trace.c
> index ebc59781456a..c210d02fac97 100644
> --- a/kernel/trace/trace.c
> +++ b/kernel/trace/trace.c
> @@ -5171,7 +5171,7 @@ static const struct file_operations tracing_fops = {
> .open = tracing_open,
> .read = seq_read,
> .read_iter = seq_read_iter,
> - .splice_read = generic_file_splice_read,
> + .splice_read = copy_splice_read,
> .write = tracing_write_stub,
> .llseek = tracing_lseek,
> .release = tracing_release,
>


--
Masami Hiramatsu (Google) <[email protected]>

2023-05-21 13:33:36

by David Howells

[permalink] [raw]
Subject: Re: [PATCH v21 26/30] splice: Convert trace/seq to use copy_splice_read()

Masami Hiramatsu (Google) <[email protected]> wrote:

> David Howells <[email protected]> wrote:
>
> > For the splice from the trace seq buffer, just use copy_splice_read().
>
> So this is because you will remove generic_file_splice_read() (since
> it's buggy), right?

An ITER_PIPE iterator has a problem if it gets reverted with other changes I
want to make. The problem is that it may not be valid to control the lifetime
of the data in the buffer with get_page(). The pages may need a pin taking
(FOLL_PIN) or the lifetime might be controlled with kfree() or rmmod.

> > In the future, something better can probably be done by gifting pages from
> > seq->buf into the pipe, but that would require changing seq->buf into a
> > vmap over an array of pages.
>
> ... We introduced splice support for avoiding copy ringbuffer pages, but
> this drops it. Thus this will drop performance of splice on ring buffer
> (trace file). If it is correct, can you also add a note about that?

Actually, no. There is no special splice support for tracing_fops. You
currently use generic_file_splice_read(), which wends its way down into
seq_read_iter. However, the seqfile stuff uses kvmalloc() to allocate the
buffer, so you are not allowed to splice page refs from kmalloc'd or vmalloc'd
memory into a pipe, so it doesn't. It calls copy_to_iter() which will cause
ITER_PIPE to allocate bufferage on an as-needed basis.

copy_splice_read() instead creates an ITER_BVEC and populates it up front
using the bulk allocator, so if you're splicing a lot of data, this ought to
be marginally faster.

> So what we need is to introduce a vmap?

We could implement seq_splice_read(). What we would need to do is to change
how the buffer is allocated: bulk allocate a bunch of arbitrary pages which we
then vmap(). When we need to splice, we read into the buffer, do a vunmap()
and then splice the pages holding the data we used into the pipe.

If we don't manage to splice all the data, we can continue splicing from the
pages we have left next time. If a read() comes along to view partially
spliced data, we would need to copy from the individual pages.

When we use up all the data, we discard all the pages we might have spliced
from and shuffle down the other pages, call the bulk allocator to replenish
the buffer and then vmap() it again.

Any pages we've spliced from must be discarded and replaced and not rewritten.

If a read() comes without the buffer having been spliced from, it can do as it
does now.

David


2023-05-23 14:41:49

by Steven Rostedt

[permalink] [raw]
Subject: Re: [PATCH v21 26/30] splice: Convert trace/seq to use copy_splice_read()

On Sat, 20 May 2023 01:00:45 +0100
David Howells <[email protected]> wrote:

> For the splice from the trace seq buffer, just use copy_splice_read().
>
> In the future, something better can probably be done by gifting pages from
> seq->buf into the pipe, but that would require changing seq->buf into a
> vmap over an array of pages.
>
> Signed-off-by: David Howells <[email protected]>
> cc: Christoph Hellwig <[email protected]>
> cc: Al Viro <[email protected]>
> cc: Jens Axboe <[email protected]>
> cc: Steven Rostedt <[email protected]>
> cc: Masami Hiramatsu <[email protected]>
> cc: [email protected]
> cc: [email protected]
> cc: [email protected]
> cc: [email protected]
> cc: [email protected]
> ---
> kernel/trace/trace.c | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/kernel/trace/trace.c b/kernel/trace/trace.c
> index ebc59781456a..c210d02fac97 100644
> --- a/kernel/trace/trace.c
> +++ b/kernel/trace/trace.c
> @@ -5171,7 +5171,7 @@ static const struct file_operations tracing_fops = {
> .open = tracing_open,
> .read = seq_read,
> .read_iter = seq_read_iter,
> - .splice_read = generic_file_splice_read,
> + .splice_read = copy_splice_read,

Anyway, for this change:

Acked-by: Steven Rostedt (Google) <[email protected]>

-- Steve

> .write = tracing_write_stub,
> .llseek = tracing_lseek,
> .release = tracing_release,


2023-07-06 00:25:14

by patchwork-bot+f2fs

[permalink] [raw]
Subject: Re: [f2fs-dev] [PATCH v21 19/30] f2fs: Provide a splice-read stub

Hello:

This patch was applied to jaegeuk/f2fs.git (dev)
by Jens Axboe <[email protected]>:

On Sat, 20 May 2023 01:00:38 +0100 you wrote:
> Provide a splice_read stub for f2fs. This does some checks and tracing
> before calling filemap_splice_read() and will update the iostats
> afterwards. Direct I/O is handled by the caller.
>
> Signed-off-by: David Howells <[email protected]>
> cc: Christoph Hellwig <[email protected]>
> cc: Al Viro <[email protected]>
> cc: Jens Axboe <[email protected]>
> cc: Jaegeuk Kim <[email protected]>
> cc: Chao Yu <[email protected]>
> cc: [email protected]
> cc: [email protected]
> cc: [email protected]
> cc: [email protected]
>
> [...]

Here is the summary with links:
- [f2fs-dev,v21,19/30] f2fs: Provide a splice-read stub
https://git.kernel.org/jaegeuk/f2fs/c/ceb11d0e2da2

You are awesome, thank you!
--
Deet-doot-dot, I am a bot.
https://korg.docs.kernel.org/patchwork/pwbot.html