linux-fsdevel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Bernd Schubert <bernd.schubert@fastmail.fm>
To: Hao Xu <hao.xu@linux.dev>, fuse-devel@lists.sourceforge.net
Cc: miklos@szeredi.hu, linux-fsdevel@vger.kernel.org,
	Wanpeng Li <wanpengli@tencent.com>,
	cgxu519@mykernel.net, Dharmendra Singh <dsingh@ddn.com>
Subject: [RFC] [PATCH] fuse: DIO writes always use the same code path
Date: Wed, 5 Jul 2023 12:23:40 +0200	[thread overview]
Message-ID: <a77b34fe-dbe7-388f-c559-932b1cd44583@fastmail.fm> (raw)
In-Reply-To: <20230630094602.230573-1-hao.xu@linux.dev>

From: Bernd Schubert <bschubert@ddn.com>

In commit 153524053bbb0d27bb2e0be36d1b46862e9ce74c DIO
writes can be handled in parallel, as long as the file
is not extended. So far this only works when daemon/server
side set FOPEN_DIRECT_IO and FOPEN_PARALLEL_DIRECT_WRITES,
but O_DIRECT (iocb->ki_flags & IOCB_DIRECT) went another
code path that doesn't have the parallel DIO write
optimization.
Given that fuse_direct_write_iter has to handle page writes
and invalidation anyway (for mmap), the DIO handler in
fuse_cache_write_iter() is removed and DIO writes are now
only handled by fuse_direct_write_iter().

Note: Correctness of this patch depends on a non-merged
series from Hao Xu <hao.xu@linux.dev>
( fuse: add a new fuse init flag to relax restrictions in no cache mode)
---
  fs/fuse/file.c |   38 +++++---------------------------------
  1 file changed, 5 insertions(+), 33 deletions(-)

diff --git a/fs/fuse/file.c b/fs/fuse/file.c
index 89d97f6188e0..1490329b536c 100644
--- a/fs/fuse/file.c
+++ b/fs/fuse/file.c
@@ -1337,11 +1337,9 @@ static ssize_t fuse_cache_write_iter(struct kiocb 
*iocb, struct iov_iter *from)
  	struct file *file = iocb->ki_filp;
  	struct address_space *mapping = file->f_mapping;
  	ssize_t written = 0;
-	ssize_t written_buffered = 0;
  	struct inode *inode = mapping->host;
  	ssize_t err;
  	struct fuse_conn *fc = get_fuse_conn(inode);
-	loff_t endbyte = 0;

  	if (fc->writeback_cache) {
  		/* Update size (EOF optimization) and mode (SUID clearing) */
@@ -1377,37 +1375,10 @@ static ssize_t fuse_cache_write_iter(struct 
kiocb *iocb, struct iov_iter *from)
  	if (err)
  		goto out;

-	if (iocb->ki_flags & IOCB_DIRECT) {
-		loff_t pos = iocb->ki_pos;
-		written = generic_file_direct_write(iocb, from);
-		if (written < 0 || !iov_iter_count(from))
-			goto out;
-
-		pos += written;
-
-		written_buffered = fuse_perform_write(iocb, mapping, from, pos);
-		if (written_buffered < 0) {
-			err = written_buffered;
-			goto out;
-		}
-		endbyte = pos + written_buffered - 1;
-
-		err = filemap_write_and_wait_range(file->f_mapping, pos,
-						   endbyte);
-		if (err)
-			goto out;
-
-		invalidate_mapping_pages(file->f_mapping,
-					 pos >> PAGE_SHIFT,
-					 endbyte >> PAGE_SHIFT);
+	written = fuse_perform_write(iocb, mapping, from, iocb->ki_pos);
+	if (written >= 0)
+		iocb->ki_pos += written;

-		written += written_buffered;
-		iocb->ki_pos = pos + written_buffered;
-	} else {
-		written = fuse_perform_write(iocb, mapping, from, iocb->ki_pos);
-		if (written >= 0)
-			iocb->ki_pos += written;
-	}
  out:
  	current->backing_dev_info = NULL;
  	inode_unlock(inode);
@@ -1691,7 +1662,8 @@ static ssize_t fuse_file_write_iter(struct kiocb 
*iocb, struct iov_iter *from)
  	if (FUSE_IS_DAX(inode))
  		return fuse_dax_write_iter(iocb, from);

-	if (!(ff->open_flags & FOPEN_DIRECT_IO))
+	if (!(ff->open_flags & FOPEN_DIRECT_IO) &&
+	    !(iocb->ki_flags & IOCB_DIRECT))
  		return fuse_cache_write_iter(iocb, from);
  	else
  		return fuse_direct_write_iter(iocb, from);


  parent reply	other threads:[~2023-07-05 10:23 UTC|newest]

Thread overview: 28+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2023-06-30  9:45 [PATCH v3 0/3] fuse: add a new fuse init flag to relax restrictions in no cache mode Hao Xu
2023-06-30  9:46 ` [PATCH 1/3] fuse: invalidate page cache pages before direct write Hao Xu
2023-06-30 10:32   ` Bernd Schubert
2023-07-21  3:34   ` [fuse-devel] " Jiachen Zhang
2023-06-30  9:46 ` [PATCH 2/3] fuse: add a new fuse init flag to relax restrictions in no cache mode Hao Xu
2023-06-30 10:35   ` Bernd Schubert
2023-06-30  9:46 ` [PATCH 3/3] fuse: write back dirty pages before direct write in direct_io_relax mode Hao Xu
2023-06-30 10:40   ` Bernd Schubert
2023-07-21  6:35   ` [External] [fuse-devel] " Jiachen Zhang
2023-07-21 11:27     ` Hao Xu
2023-07-21 11:56       ` Bernd Schubert
2023-07-25 10:11         ` Hao Xu
2023-07-25 13:00           ` Bernd Schubert
2023-07-25 16:57             ` Hao Xu
2023-07-25 17:59               ` Bernd Schubert
2023-07-27  9:42                 ` Hao Xu
2023-07-26 11:07               ` Jiachen Zhang
2023-07-26 13:15                 ` Bernd Schubert
2023-07-27  2:24                   ` Jiachen Zhang
2023-07-27 10:31                 ` Hao Xu
2023-07-28  2:57                   ` Jiachen Zhang
2023-07-27 10:48                 ` Hao Xu
2023-07-05 10:23 ` Bernd Schubert [this message]
2023-07-06 14:43   ` [RFC] [PATCH] fuse: DIO writes always use the same code path Christoph Hellwig
2023-07-07 13:36     ` Bernd Schubert
2023-07-17  8:03   ` Hao Xu
2023-07-17 21:17     ` Bernd Schubert
2023-07-20  7:32 ` [PATCH v3 0/3] fuse: add a new fuse init flag to relax restrictions in no cache mode Hao Xu

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=a77b34fe-dbe7-388f-c559-932b1cd44583@fastmail.fm \
    --to=bernd.schubert@fastmail.fm \
    --cc=cgxu519@mykernel.net \
    --cc=dsingh@ddn.com \
    --cc=fuse-devel@lists.sourceforge.net \
    --cc=hao.xu@linux.dev \
    --cc=linux-fsdevel@vger.kernel.org \
    --cc=miklos@szeredi.hu \
    --cc=wanpengli@tencent.com \
    /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).