From: "Jason A. Donenfeld" <Jason@zx2c4.com>
To: linux-kernel@vger.kernel.org, viro@zeniv.linux.org.uk,
Jens Axboe <axboe@kernel.dk>,
linux-fsdevel@vger.kernel.org
Cc: "Jason A. Donenfeld" <Jason@zx2c4.com>
Subject: [PATCH 4/6] fs: check FMODE_LSEEK to control internal pipe splicing
Date: Fri, 24 Jun 2022 18:56:29 +0200 [thread overview]
Message-ID: <20220624165631.2124632-5-Jason@zx2c4.com> (raw)
In-Reply-To: <20220624165631.2124632-1-Jason@zx2c4.com>
The original direct splicing mechanism from Jens required the input to
be a regular file because it was avoiding the special socket case. It
also recognized blkdevs as being close enough to a regular file. But it
forgot about chardevs, which behave the same way and work fine here.
This is an okayish heuristic, but it doesn't totally work. For example,
a few chardevs should be spliceable here. And a few regular files
shouldn't. This patch fixes this by instead checking whether FMODE_LSEEK
is set, which represents decently enough what we need rewinding for when
splicing to internal pipes.
Fixes: b92ce5589374 ("[PATCH] splice: add direct fd <-> fd splicing support")
Cc: Jens Axboe <axboe@kernel.dk>
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
---
fs/splice.c | 10 ++++------
1 file changed, 4 insertions(+), 6 deletions(-)
diff --git a/fs/splice.c b/fs/splice.c
index 047b79db8eb5..93a2c9bf6249 100644
--- a/fs/splice.c
+++ b/fs/splice.c
@@ -814,17 +814,15 @@ ssize_t splice_direct_to_actor(struct file *in, struct splice_desc *sd,
{
struct pipe_inode_info *pipe;
long ret, bytes;
- umode_t i_mode;
size_t len;
int i, flags, more;
/*
- * We require the input being a regular file, as we don't want to
- * randomly drop data for eg socket -> socket splicing. Use the
- * piped splicing for that!
+ * We require the input to be seekable, as we don't want to randomly
+ * drop data for eg socket -> socket splicing. Use the piped splicing
+ * for that!
*/
- i_mode = file_inode(in)->i_mode;
- if (unlikely(!S_ISREG(i_mode) && !S_ISBLK(i_mode)))
+ if (unlikely(!(in->f_mode & FMODE_LSEEK)))
return -EINVAL;
/*
--
2.35.1
next prev parent reply other threads:[~2022-06-24 16:57 UTC|newest]
Thread overview: 18+ messages / expand[flat|nested] mbox.gz Atom feed top
2022-06-24 16:56 [PATCH 0/6] cleanup llseek and splice Jason A. Donenfeld
2022-06-24 16:56 ` [PATCH 1/6] ksmbd: use vfs_llseek instead of dereferencing NULL Jason A. Donenfeld
2022-06-24 21:20 ` Al Viro
2022-06-25 4:10 ` Namjae Jeon
2022-06-24 16:56 ` [PATCH 2/6] fs: do not set no_llseek in fops Jason A. Donenfeld
2022-06-24 21:48 ` Al Viro
2022-06-25 10:46 ` Jason A. Donenfeld
2022-06-24 16:56 ` [PATCH 3/6] fs: clear FMODE_LSEEK if no llseek function Jason A. Donenfeld
2022-06-24 17:04 ` Al Viro
2022-06-24 17:09 ` Jason A. Donenfeld
2022-06-24 17:13 ` Al Viro
2022-06-24 17:17 ` Jason A. Donenfeld
2022-06-24 16:56 ` Jason A. Donenfeld [this message]
2022-06-24 16:56 ` [PATCH 5/6] dma-buf: remove useless FMODE_LSEEK flag Jason A. Donenfeld
2022-06-24 21:51 ` Al Viro
2022-06-24 16:56 ` [PATCH 6/6] vfio: do not set " Jason A. Donenfeld
2022-06-24 21:53 ` Al Viro
2022-06-25 10:52 ` Jason A. Donenfeld
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=20220624165631.2124632-5-Jason@zx2c4.com \
--to=jason@zx2c4.com \
--cc=axboe@kernel.dk \
--cc=linux-fsdevel@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--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
all inboxes | Powered by JetHome®