From: Andreas Gruenbacher <agruenba@redhat.com>
To: Linus Torvalds <torvalds@linux-foundation.org>
Cc: cluster-devel@redhat.com, Jan Kara <jack@suse.cz>,
Andreas Gruenbacher <agruenba@redhat.com>,
linux-kernel@vger.kernel.org,
Christoph Hellwig <hch@infradead.org>,
Alexander Viro <viro@zeniv.linux.org.uk>,
linux-fsdevel@vger.kernel.org, ocfs2-devel@oss.oracle.com
Subject: [Ocfs2-devel] [PATCH v2 6/6] gfs2: Fix mmap + page fault deadlocks for direct I/O
Date: Mon, 19 Jul 2021 00:39:32 +0200 [thread overview]
Message-ID: <20210718223932.2703330-7-agruenba@redhat.com> (raw)
In-Reply-To: <20210718223932.2703330-1-agruenba@redhat.com>
Direct I/O differs from buffered I/O in that it uses bio_iov_iter_get_pages for
grabbing page references and for manually faulting in pages instead of
triggering actual page faults. For disabling these manual page faults, it's
not enough to call pagefault_disable(); instead, we use the new
ITER_FLAG_FAST_ONLY flag for telling iomap_dio_rw to stop faulting pages in for
us.
Signed-off-by: Andreas Gruenbacher <agruenba@redhat.com>
---
fs/gfs2/file.c | 35 ++++++++++++++++++++++++++++++++++-
1 file changed, 34 insertions(+), 1 deletion(-)
diff --git a/fs/gfs2/file.c b/fs/gfs2/file.c
index 99df7934b4d8..6feb857a8a1c 100644
--- a/fs/gfs2/file.c
+++ b/fs/gfs2/file.c
@@ -763,21 +763,42 @@ static ssize_t gfs2_file_direct_read(struct kiocb *iocb, struct iov_iter *to,
struct file *file = iocb->ki_filp;
struct gfs2_inode *ip = GFS2_I(file->f_mapping->host);
size_t count = iov_iter_count(to);
+ size_t written = 0;
ssize_t ret;
+ /*
+ * In this function, we disable page faults when whe're holding the
+ * inode glock while doing I/O. If a page fault occurs, we drop the
+ * inode glock, fault in the pages manually, and then we retry. Other
+ * than in gfs2_file_read_iter, iomap_dio_rw can trigger implicit as
+ * well as manual page faults, and we need to disable both kinds
+ * separately.
+ */
+
if (!count)
return 0; /* skip atime */
gfs2_holder_init(ip->i_gl, LM_ST_DEFERRED, 0, gh);
+retry:
ret = gfs2_glock_nq(gh);
if (ret)
goto out_uninit;
+ pagefault_disable();
+ to->type |= ITER_FLAG_FAST_ONLY;
ret = iomap_dio_rw(iocb, to, &gfs2_iomap_ops, NULL, 0);
+ to->type &= ~ITER_FLAG_FAST_ONLY;
+ pagefault_enable();
+
gfs2_glock_dq(gh);
+ if (ret > 0)
+ written += ret;
+ if (unlikely(iov_iter_count(to) && (ret > 0 || ret == -EFAULT)) &&
+ fault_in_iov_iter(to))
+ goto retry;
out_uninit:
gfs2_holder_uninit(gh);
- return ret;
+ return written ? written : ret;
}
static ssize_t gfs2_file_direct_write(struct kiocb *iocb, struct iov_iter *from,
@@ -790,6 +811,12 @@ static ssize_t gfs2_file_direct_write(struct kiocb *iocb, struct iov_iter *from,
loff_t offset = iocb->ki_pos;
ssize_t ret;
+ /*
+ * In this function, we disable page faults when whe're holding the
+ * inode glock while doing I/O. If a page fault occurs, we drop the
+ * inode glock, fault in the pages manually, and then we retry.
+ */
+
/*
* Deferred lock, even if its a write, since we do no allocation on
* this path. All we need to change is the atime, and this lock mode
@@ -799,6 +826,7 @@ static ssize_t gfs2_file_direct_write(struct kiocb *iocb, struct iov_iter *from,
* VFS does.
*/
gfs2_holder_init(ip->i_gl, LM_ST_DEFERRED, 0, gh);
+retry:
ret = gfs2_glock_nq(gh);
if (ret)
goto out_uninit;
@@ -807,11 +835,16 @@ static ssize_t gfs2_file_direct_write(struct kiocb *iocb, struct iov_iter *from,
if (offset + len > i_size_read(&ip->i_inode))
goto out;
+ from->type |= ITER_FLAG_FAST_ONLY;
ret = iomap_dio_rw(iocb, from, &gfs2_iomap_ops, NULL, 0);
+ from->type &= ~ITER_FLAG_FAST_ONLY;
+
if (ret == -ENOTBLK)
ret = 0;
out:
gfs2_glock_dq(gh);
+ if (unlikely(ret == -EFAULT) && fault_in_iov_iter(from))
+ goto retry;
out_uninit:
gfs2_holder_uninit(gh);
return ret;
--
2.26.3
_______________________________________________
Ocfs2-devel mailing list
Ocfs2-devel@oss.oracle.com
https://oss.oracle.com/mailman/listinfo/ocfs2-devel
next prev parent reply other threads:[~2021-07-18 22:43 UTC|newest]
Thread overview: 12+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-07-18 22:39 [Ocfs2-devel] [PATCH v2 0/6] gfs2: Fix mmap + page fault deadlocks Andreas Gruenbacher
2021-07-18 22:39 ` [Ocfs2-devel] [PATCH v2 1/6] iov_iter: Introduce fault_in_iov_iter helper Andreas Gruenbacher
2021-07-19 19:26 ` Linus Torvalds
2021-07-18 22:39 ` [Ocfs2-devel] [PATCH v2 2/6] iomap: Fix iomap_dio_rw return value for page faults Andreas Gruenbacher
2021-07-18 22:39 ` [Ocfs2-devel] [PATCH v2 3/6] gfs2: Add wrapper for iomap_file_buffered_write Andreas Gruenbacher
2021-07-18 22:39 ` [Ocfs2-devel] [PATCH v2 4/6] gfs2: Fix mmap + page fault deadlocks for buffered I/O Andreas Gruenbacher
2021-07-18 22:39 ` [Ocfs2-devel] [PATCH v2 5/6] iov_iter: Introduce ITER_FLAG_FAST_ONLY flag Andreas Gruenbacher
2021-07-19 19:29 ` Linus Torvalds
2021-07-19 20:17 ` Al Viro
2021-07-18 22:39 ` Andreas Gruenbacher [this message]
2021-07-19 19:31 ` [Ocfs2-devel] [PATCH v2 0/6] gfs2: Fix mmap + page fault deadlocks Linus Torvalds
2021-07-19 20:14 ` Al Viro
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=20210718223932.2703330-7-agruenba@redhat.com \
--to=agruenba@redhat.com \
--cc=cluster-devel@redhat.com \
--cc=hch@infradead.org \
--cc=jack@suse.cz \
--cc=linux-fsdevel@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=ocfs2-devel@oss.oracle.com \
--cc=torvalds@linux-foundation.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;
as well as URLs for NNTP newsgroup(s).