Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.18 as permitted sender) client-ip=23.128.96.18;
Date:   Tue, 6 Apr 2021 15:40:22 +0100
From:   Matthew Wilcox <willy@infradead.org>
To:     Christoph Hellwig <hch@infradead.org>
Cc:     "Kirill A. Shutemov" <kirill@shutemov.name>, linux-mm@kvack.org,
        linux-kernel@vger.kernel.org, linux-fsdevel@vger.kernel.org,
        linux-cachefs@redhat.com, linux-afs@lists.infradead.org
Subject: Re: [PATCH v6 01/27] mm: Introduce struct folio
Message-ID: <20210406144022.GR2531743@casper.infradead.org>
References: <20210331184728.1188084-1-willy@infradead.org>
 <20210331184728.1188084-2-willy@infradead.org>
 <20210406122918.h5dsnbjhmwpfasf4@box.shutemov.name>
 <20210406124807.GO2531743@casper.infradead.org>
 <20210406143150.GA3082513@infradead.org>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <20210406143150.GA3082513@infradead.org>
Precedence: bulk

On Tue, Apr 06, 2021 at 03:31:50PM +0100, Christoph Hellwig wrote:
> > > As Christoph, I'm not a fan of this :/
> > 
> > What would you prefer?
> 
> Looking at your full folio series on git.infradead.org, there are a
> total of 12 references to non-page members of struct folio, assuming
> my crude grep that expects a folio to be named folio did not miss any.

Hmm ... I count more in the filesystems:

fs/afs/dir.c:   struct afs_vnode *dvnode = AFS_FS_I(folio->page.mapping->host);
fs/afs/dir.c:   _enter("{%lu},%zu,%zu", folio->page.index, offset, length);
fs/afs/file.c:  _enter("{%lu},%zu,%zu", folio->page.index, offset, length);
fs/afs/write.c:         folio->page.index);
fs/befs/linuxvfs.c:     struct inode *inode = folio->page.mapping->host;
fs/btrfs/disk-io.c:     tree = &BTRFS_I(folio->page.mapping->host)->io_tree;
fs/btrfs/disk-io.c:             btrfs_warn(BTRFS_I(folio->page.mapping->host)->root->fs_info,
fs/btrfs/extent_io.c:   struct btrfs_inode *inode = BTRFS_I(folios[0]->page.mapping->host);
fs/btrfs/file.c:                if (folio->page.mapping != inode->i_mapping) {
fs/btrfs/free-space-cache.c:                    if (folio->page.mapping != inode->i_mapping) {
fs/btrfs/inode.c:               if (folio->page.mapping != mapping) {
fs/btrfs/inode.c:       struct btrfs_inode *inode = BTRFS_I(folio->page.mapping->host);
fs/buffer.c:    spin_lock(&folio->page.mapping->private_lock);
fs/buffer.c:    spin_unlock(&folio->page.mapping->private_lock);
fs/buffer.c:    block_in_file = (sector_t)folio->page.index <<
fs/ceph/addr.c:              mapping->host, folio, folio->page.index);
fs/ceph/addr.c:      mapping->host, folio, folio->page.index,
fs/ceph/addr.c: folio->page.private = (unsigned long)snapc;
fs/ceph/addr.c: inode = folio->page.mapping->host;
fs/ceph/addr.c:              inode, folio, folio->page.index, offset, length);
fs/ceph/addr.c:      inode, folio, folio->page.index);
fs/cifs/file.c: struct cifsInodeInfo *cifsi = CIFS_I(folio->page.mapping->host);
fs/ext4/inode.c:        struct inode *inode = folio->page.mapping->host;
fs/f2fs/data.c: struct inode *inode = folio->page.mapping->host;
fs/fuse/dir.c:  int err = fuse_readlink_page(folio->page.mapping->host, &folio->page);
fs/gfs2/aops.c: struct gfs2_sbd *sdp = GFS2_SB(folio->page.mapping->host);
fs/iomap/buffered-io.c: unsigned int nr_blocks = i_blocks_per_folio(folio->page.mapping->host,
fs/iomap/buffered-io.c: struct inode *inode = folio->page.mapping->host;
fs/iomap/buffered-io.c: BUG_ON(folio->page.index);
fs/iomap/buffered-io.c:         gfp_t gfp = mapping_gfp_constraint(folio->page.mapping,
fs/iomap/buffered-io.c: struct inode *inode = folio->page.mapping->host;
fs/iomap/buffered-io.c: struct inode *inode = folio->page.mapping->host;
fs/iomap/buffered-io.c: trace_iomap_releasepage(folio->page.mapping->host, folio_offset(folio),
fs/iomap/buffered-io.c: trace_iomap_invalidatepage(folio->page.mapping->host, offset, len);
fs/jffs2/file.c:        struct inode *inode = folio->page.mapping->host;
fs/mpage.c:     struct inode *inode = folio->page.mapping->host;
fs/mpage.c:             gfp = readahead_gfp_mask(folio->page.mapping);
fs/mpage.c:             gfp = mapping_gfp_constraint(folio->page.mapping, GFP_KERNEL);
fs/mpage.c:     block_in_file = (sector_t)folio->page.index << (PAGE_SHIFT - blkbits);
fs/mpage.c:             prefetchw(&folio->page.flags);
fs/nfs/file.c:  nfs_fscache_invalidate_page(&folio->page, folio->page.mapping->host);
fs/nfs/fscache.c:                nfs_i_fscache(inode), folio, folio->page.index,
fs/nfs/fscache.c:                folio->page.flags, inode);
fs/reiserfs/inode.c:    struct inode *inode = folio->page.mapping->host;
fs/remap_range.c:       if (folio1->page.index > folio2->page.index)
fs/ubifs/file.c:        struct inode *inode = folio->page.mapping->host;
fs/xfs/xfs_aops.c:      struct inode            *inode = folio->page.mapping->host;

(I didn't go through my whole series and do the conversion from
folio->page.x to folio->x yet)