linux-fsdevel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Joanne Koong <joannelkoong@gmail.com>
To: brauner@kernel.org, miklos@szeredi.hu
Cc: hch@infradead.org, djwong@kernel.org,
	hsiangkao@linux.alibaba.com, linux-block@vger.kernel.org,
	gfs2@lists.linux.dev, linux-fsdevel@vger.kernel.org,
	kernel-team@meta.com, linux-xfs@vger.kernel.org,
	linux-doc@vger.kernel.org
Subject: [PATCH v2 16/16] fuse: remove fc->blkbits workaround for partial writes
Date: Mon,  8 Sep 2025 11:51:22 -0700	[thread overview]
Message-ID: <20250908185122.3199171-17-joannelkoong@gmail.com> (raw)
In-Reply-To: <20250908185122.3199171-1-joannelkoong@gmail.com>

Now that fuse is integrated with iomap for read/readahead, we can remove
the workaround that was added in commit bd24d2108e9c ("fuse: fix fuseblk
i_blkbits for iomap partial writes"), which was previously needed to
avoid a race condition where an iomap partial write may be overwritten
by a read if blocksize < PAGE_SIZE. Now that fuse does iomap
read/readahead, this is protected against since there is granular
uptodate tracking of blocks, which means this workaround can be removed.

Signed-off-by: Joanne Koong <joannelkoong@gmail.com>
---
 fs/fuse/dir.c    |  2 +-
 fs/fuse/fuse_i.h |  8 --------
 fs/fuse/inode.c  | 13 +------------
 3 files changed, 2 insertions(+), 21 deletions(-)

diff --git a/fs/fuse/dir.c b/fs/fuse/dir.c
index 5c569c3cb53f..ebee7e0b1cd3 100644
--- a/fs/fuse/dir.c
+++ b/fs/fuse/dir.c
@@ -1199,7 +1199,7 @@ static void fuse_fillattr(struct mnt_idmap *idmap, struct inode *inode,
 	if (attr->blksize != 0)
 		blkbits = ilog2(attr->blksize);
 	else
-		blkbits = fc->blkbits;
+		blkbits = inode->i_sb->s_blocksize_bits;
 
 	stat->blksize = 1 << blkbits;
 }
diff --git a/fs/fuse/fuse_i.h b/fs/fuse/fuse_i.h
index cc428d04be3e..1647eb7ca6fa 100644
--- a/fs/fuse/fuse_i.h
+++ b/fs/fuse/fuse_i.h
@@ -975,14 +975,6 @@ struct fuse_conn {
 		/* Request timeout (in jiffies). 0 = no timeout */
 		unsigned int req_timeout;
 	} timeout;
-
-	/*
-	 * This is a workaround until fuse uses iomap for reads.
-	 * For fuseblk servers, this represents the blocksize passed in at
-	 * mount time and for regular fuse servers, this is equivalent to
-	 * inode->i_blkbits.
-	 */
-	u8 blkbits;
 };
 
 /*
diff --git a/fs/fuse/inode.c b/fs/fuse/inode.c
index 7ddfd2b3cc9c..3bfd83469d9f 100644
--- a/fs/fuse/inode.c
+++ b/fs/fuse/inode.c
@@ -292,7 +292,7 @@ void fuse_change_attributes_common(struct inode *inode, struct fuse_attr *attr,
 	if (attr->blksize)
 		fi->cached_i_blkbits = ilog2(attr->blksize);
 	else
-		fi->cached_i_blkbits = fc->blkbits;
+		fi->cached_i_blkbits = inode->i_sb->s_blocksize_bits;
 
 	/*
 	 * Don't set the sticky bit in i_mode, unless we want the VFS
@@ -1810,21 +1810,10 @@ int fuse_fill_super_common(struct super_block *sb, struct fuse_fs_context *ctx)
 		err = -EINVAL;
 		if (!sb_set_blocksize(sb, ctx->blksize))
 			goto err;
-		/*
-		 * This is a workaround until fuse hooks into iomap for reads.
-		 * Use PAGE_SIZE for the blocksize else if the writeback cache
-		 * is enabled, buffered writes go through iomap and a read may
-		 * overwrite partially written data if blocksize < PAGE_SIZE
-		 */
-		fc->blkbits = sb->s_blocksize_bits;
-		if (ctx->blksize != PAGE_SIZE &&
-		    !sb_set_blocksize(sb, PAGE_SIZE))
-			goto err;
 #endif
 	} else {
 		sb->s_blocksize = PAGE_SIZE;
 		sb->s_blocksize_bits = PAGE_SHIFT;
-		fc->blkbits = sb->s_blocksize_bits;
 	}
 
 	sb->s_subtype = ctx->subtype;
-- 
2.47.3


      parent reply	other threads:[~2025-09-08 18:52 UTC|newest]

Thread overview: 60+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2025-09-08 18:51 [PATCH v2 00/16] fuse: use iomap for buffered reads + readahead Joanne Koong
2025-09-08 18:51 ` [PATCH v2 01/16] iomap: move async bio read logic into helper function Joanne Koong
2025-09-11 11:09   ` Christoph Hellwig
2025-09-12 16:01     ` Joanne Koong
2025-09-08 18:51 ` [PATCH v2 02/16] iomap: move read/readahead bio submission " Joanne Koong
2025-09-11 11:09   ` Christoph Hellwig
2025-09-08 18:51 ` [PATCH v2 03/16] iomap: rename cur_folio_in_bio to folio_owned Joanne Koong
2025-09-11 11:10   ` Christoph Hellwig
2025-09-08 18:51 ` [PATCH v2 04/16] iomap: store read/readahead bio generically Joanne Koong
2025-09-11 11:11   ` Christoph Hellwig
2025-09-12 16:10     ` Joanne Koong
2025-09-08 18:51 ` [PATCH v2 05/16] iomap: propagate iomap_read_folio() error to caller Joanne Koong
2025-09-11 11:13   ` Christoph Hellwig
2025-09-12 16:28     ` Joanne Koong
2025-09-15 16:05       ` Christoph Hellwig
2025-09-08 18:51 ` [PATCH v2 06/16] iomap: iterate over entire folio in iomap_readpage_iter() Joanne Koong
2025-09-11 11:15   ` Christoph Hellwig
2025-09-08 18:51 ` [PATCH v2 07/16] iomap: rename iomap_readpage_iter() to iomap_read_folio_iter() Joanne Koong
2025-09-11 11:16   ` Christoph Hellwig
2025-09-08 18:51 ` [PATCH v2 08/16] iomap: rename iomap_readpage_ctx struct to iomap_read_folio_ctx Joanne Koong
2025-09-11 11:16   ` Christoph Hellwig
2025-09-08 18:51 ` [PATCH v2 09/16] iomap: add public start/finish folio read helpers Joanne Koong
2025-09-11 11:16   ` Christoph Hellwig
2025-09-08 18:51 ` [PATCH v2 10/16] iomap: make iomap_read_folio_ctx->folio_owned internal Joanne Koong
2025-09-11 11:17   ` Christoph Hellwig
2025-09-08 18:51 ` [PATCH v2 11/16] iomap: add caller-provided callbacks for read and readahead Joanne Koong
2025-09-09  0:14   ` Gao Xiang
2025-09-09  0:40     ` Gao Xiang
2025-09-09 15:24     ` Joanne Koong
2025-09-09 23:21       ` Gao Xiang
2025-09-10 17:41         ` Joanne Koong
2025-09-11 11:19           ` Christoph Hellwig
2025-09-11 11:26   ` Christoph Hellwig
2025-09-12 17:36     ` Joanne Koong
2025-09-08 18:51 ` [PATCH v2 12/16] iomap: add bias for async read requests Joanne Koong
2025-09-11 11:31   ` Christoph Hellwig
2025-09-12 17:30     ` Joanne Koong
2025-09-15 16:05       ` Christoph Hellwig
2025-09-16 19:14       ` Joanne Koong
2025-09-19 15:04         ` Christoph Hellwig
2025-09-19 17:58           ` Joanne Koong
2025-09-08 18:51 ` [PATCH v2 13/16] iomap: move read/readahead logic out of CONFIG_BLOCK guard Joanne Koong
2025-09-09  2:14   ` Gao Xiang
2025-09-09 15:33     ` Joanne Koong
2025-09-10  4:59       ` Gao Xiang
2025-09-11 11:37         ` Christoph Hellwig
2025-09-11 12:29           ` Gao Xiang
2025-09-11 19:45             ` Joanne Koong
2025-09-12  0:06               ` Gao Xiang
2025-09-12  1:09                 ` Gao Xiang
2025-09-12  1:10                   ` Gao Xiang
2025-09-12 19:56                     ` Joanne Koong
2025-09-12 20:09                       ` Joanne Koong
2025-09-12 23:35                         ` Gao Xiang
2025-09-12 23:20                       ` Gao Xiang
2025-09-11 11:44   ` Christoph Hellwig
2025-09-16 23:23     ` Joanne Koong
2025-09-08 18:51 ` [PATCH v2 14/16] fuse: use iomap for read_folio Joanne Koong
2025-09-08 18:51 ` [PATCH v2 15/16] fuse: use iomap for readahead Joanne Koong
2025-09-08 18:51 ` Joanne Koong [this message]

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=20250908185122.3199171-17-joannelkoong@gmail.com \
    --to=joannelkoong@gmail.com \
    --cc=brauner@kernel.org \
    --cc=djwong@kernel.org \
    --cc=gfs2@lists.linux.dev \
    --cc=hch@infradead.org \
    --cc=hsiangkao@linux.alibaba.com \
    --cc=kernel-team@meta.com \
    --cc=linux-block@vger.kernel.org \
    --cc=linux-doc@vger.kernel.org \
    --cc=linux-fsdevel@vger.kernel.org \
    --cc=linux-xfs@vger.kernel.org \
    --cc=miklos@szeredi.hu \
    /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).