linux-fsdevel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Amir Goldstein <amir73il@gmail.com>
To: Christian Brauner <brauner@kernel.org>
Cc: Al Viro <viro@zeniv.linux.org.uk>, Jan Kara <jack@suse.cz>,
	Jens Axboe <axboe@kernel.dk>, Miklos Szeredi <miklos@szeredi.hu>,
	David Howells <dhowells@redhat.com>,
	linux-fsdevel@vger.kernel.org,
	Chuck Lever <chuck.lever@oracle.com>,
	Jeff Layton <jlayton@kernel.org>
Subject: [PATCH 03/15] splice: move permission hook out of splice_direct_to_actor()
Date: Tue, 14 Nov 2023 17:33:09 +0200	[thread overview]
Message-ID: <20231114153321.1716028-4-amir73il@gmail.com> (raw)
In-Reply-To: <20231114153321.1716028-1-amir73il@gmail.com>

vfs_splice_read() has a permission hook inside rw_verify_area() and
it is called from do_splice_direct() -> splice_direct_to_actor().

The callers of do_splice_direct() (e.g. vfs_copy_file_range()) already
call rw_verify_area() for the entire range, but the other caller of
splice_direct_to_actor() (nfsd) does not.

Add the rw_verify_area() checks in nfsd_splice_read() and use a
variant of vfs_splice_read() without rw_verify_area() check in
splice_direct_to_actor() to avoid the redundant rw_verify_area() checks.

This is needed for fanotify "pre content" events.

Signed-off-by: Amir Goldstein <amir73il@gmail.com>
---
 fs/nfsd/vfs.c |  5 ++++-
 fs/splice.c   | 58 +++++++++++++++++++++++++++++++--------------------
 2 files changed, 39 insertions(+), 24 deletions(-)

diff --git a/fs/nfsd/vfs.c b/fs/nfsd/vfs.c
index fbbea7498f02..5d704461e3b4 100644
--- a/fs/nfsd/vfs.c
+++ b/fs/nfsd/vfs.c
@@ -1046,7 +1046,10 @@ __be32 nfsd_splice_read(struct svc_rqst *rqstp, struct svc_fh *fhp,
 	ssize_t host_err;
 
 	trace_nfsd_read_splice(rqstp, fhp, offset, *count);
-	host_err = splice_direct_to_actor(file, &sd, nfsd_direct_splice_actor);
+	host_err = rw_verify_area(READ, file, &offset, *count);
+	if (!host_err)
+		host_err = splice_direct_to_actor(file, &sd,
+						  nfsd_direct_splice_actor);
 	return nfsd_finish_read(rqstp, fhp, file, offset, count, eof, host_err);
 }
 
diff --git a/fs/splice.c b/fs/splice.c
index 6e917db6f49a..6fc2c27e9520 100644
--- a/fs/splice.c
+++ b/fs/splice.c
@@ -944,27 +944,15 @@ static void do_splice_eof(struct splice_desc *sd)
 		sd->splice_eof(sd);
 }
 
-/**
- * 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.
+/*
+ * Callers already called rw_verify_area() on the entire range.
+ * No need to call it for sub ranges.
  */
-long vfs_splice_read(struct file *in, loff_t *ppos,
-		     struct pipe_inode_info *pipe, size_t len,
-		     unsigned int flags)
+static long do_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;
 
 	if (unlikely(!(in->f_mode & FMODE_READ)))
 		return -EBADF;
@@ -975,10 +963,6 @@ long vfs_splice_read(struct file *in, loff_t *ppos,
 	p_space = pipe->max_usage - pipe_occupancy(pipe->head, pipe->tail);
 	len = min_t(size_t, len, p_space << PAGE_SHIFT);
 
-	ret = rw_verify_area(READ, in, ppos, len);
-	if (unlikely(ret < 0))
-		return ret;
-
 	if (unlikely(len > MAX_RW_COUNT))
 		len = MAX_RW_COUNT;
 
@@ -992,6 +976,34 @@ long vfs_splice_read(struct file *in, loff_t *ppos,
 		return copy_splice_read(in, ppos, pipe, len, flags);
 	return in->f_op->splice_read(in, ppos, pipe, len, flags);
 }
+
+/**
+ * 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.
+ */
+long vfs_splice_read(struct file *in, loff_t *ppos,
+		     struct pipe_inode_info *pipe, size_t len,
+		     unsigned int flags)
+{
+	int ret;
+
+	ret = rw_verify_area(READ, in, ppos, len);
+	if (unlikely(ret < 0))
+		return ret;
+
+	return do_splice_read(in, ppos, pipe, len, flags);
+}
 EXPORT_SYMBOL_GPL(vfs_splice_read);
 
 /**
@@ -1066,7 +1078,7 @@ 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 = vfs_splice_read(in, &pos, pipe, len, flags);
+		ret = do_splice_read(in, &pos, pipe, len, flags);
 		if (unlikely(ret <= 0))
 			goto read_failure;
 
-- 
2.34.1


  parent reply	other threads:[~2023-11-14 15:33 UTC|newest]

Thread overview: 25+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2023-11-14 15:33 [PATCH 00/15] Tidy up file permission hooks Amir Goldstein
2023-11-14 15:33 ` [PATCH 01/15] ovl: add permission hooks outside of do_splice_direct() Amir Goldstein
2023-11-14 15:33 ` [PATCH 02/15] splice: remove permission hook from do_splice_direct() Amir Goldstein
2023-11-14 15:33 ` Amir Goldstein [this message]
2023-11-14 15:33 ` [PATCH 04/15] splice: move permission hook out of splice_file_to_pipe() Amir Goldstein
2023-11-14 15:33 ` [PATCH 05/15] splice: remove permission hook from iter_file_splice_write() Amir Goldstein
2023-11-21 14:56   ` Christian Brauner
2023-11-21 15:18     ` Amir Goldstein
2023-11-14 15:33 ` [PATCH 06/15] remap_range: move permission hooks out of do_clone_file_range() Amir Goldstein
2023-11-14 15:33 ` [PATCH 07/15] remap_range: move file_start_write() to after permission hook Amir Goldstein
2023-11-21 15:10   ` Christian Brauner
2023-11-21 15:47     ` Christian Brauner
2023-11-21 18:39     ` Amir Goldstein
2023-11-14 15:33 ` [PATCH 08/15] btrfs: " Amir Goldstein
2023-11-14 15:33 ` [PATCH 09/15] fs: move file_start_write() into vfs_iter_write() Amir Goldstein
2023-11-14 15:33 ` [PATCH 10/15] fs: move permission hook out of do_iter_write() Amir Goldstein
2023-11-21 15:34   ` Christian Brauner
2023-11-14 15:33 ` [PATCH 11/15] fs: move permission hook out of do_iter_read() Amir Goldstein
2023-11-21 15:28   ` Christian Brauner
2023-11-21 17:46     ` Amir Goldstein
2023-11-21 15:35   ` Christian Brauner
2023-11-14 15:33 ` [PATCH 12/15] fs: move kiocb_start_write() into vfs_iocb_iter_write() Amir Goldstein
2023-11-14 15:33 ` [PATCH 13/15] fs: create __sb_write_started() helper Amir Goldstein
2023-11-14 15:33 ` [PATCH 14/15] fs: create file_write_started() helper Amir Goldstein
2023-11-14 15:33 ` [PATCH 15/15] fs: create {sb,file}_write_not_started() helpers Amir Goldstein

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 \
    --in-reply-to=20231114153321.1716028-4-amir73il@gmail.com \
    --to=amir73il@gmail.com \
    --cc=axboe@kernel.dk \
    --cc=brauner@kernel.org \
    --cc=chuck.lever@oracle.com \
    --cc=dhowells@redhat.com \
    --cc=jack@suse.cz \
    --cc=jlayton@kernel.org \
    --cc=linux-fsdevel@vger.kernel.org \
    --cc=miklos@szeredi.hu \
    --cc=viro@zeniv.linux.org.uk \
    /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;
as well as URLs for NNTP newsgroup(s).