erofs: iomap support for non-tailpacking DIO

Add iomap support for non-tailpacking uncompressed data in order to
support DIO and DAX.

Direct I/O is useful in certain scenarios for uncompressed files.
For example, double pagecache can be avoid by direct I/O when
loop device is used for uncompressed files containing upper layer
compressed filesystem.

This adds iomap DIO support for non-tailpacking cases first and
tail-packing inline files are handled in the follow-up patch.

Link: https://lore.kernel.org/r/20210805003601.183063-2-hsiangkao@linux.alibaba.com
Cc: linux-fsdevel@vger.kernel.org
Reviewed-by: Chao Yu <chao@kernel.org>
Signed-off-by: Huang Jianan <huangjianan@oppo.com>
Signed-off-by: Gao Xiang <hsiangkao@linux.alibaba.com>

diff --git a/fs/erofs/Kconfig b/fs/erofs/Kconfig
index 906af0c..14b7470 100644 (file)
--- a/fs/erofs/Kconfig
+++ b/fs/erofs/Kconfig
@@ -3,6 +3,7 @@
 config EROFS_FS
        tristate "EROFS filesystem support"
        depends on BLOCK
+       select FS_IOMAP
        select LIBCRC32C
        help
          EROFS (Enhanced Read-Only File System) is a lightweight

diff --git a/fs/erofs/data.c b/fs/erofs/data.c
index 3787a5f..4158572 100644 (file)
--- a/fs/erofs/data.c
+++ b/fs/erofs/data.c
@@ -5,6 +5,7 @@
  */
 #include "internal.h"
 #include <linux/prefetch.h>
+#include <linux/iomap.h>
 
 #include <trace/events/erofs.h>
 
@@ -308,9 +309,102 @@ static sector_t erofs_bmap(struct address_space *mapping, sector_t block)
        return 0;
 }
 
+static int erofs_iomap_begin(struct inode *inode, loff_t offset, loff_t length,
+               unsigned int flags, struct iomap *iomap, struct iomap *srcmap)
+{
+       int ret;
+       struct erofs_map_blocks map;
+
+       map.m_la = offset;
+       map.m_llen = length;
+
+       ret = erofs_map_blocks_flatmode(inode, &map, EROFS_GET_BLOCKS_RAW);
+       if (ret < 0)
+               return ret;
+
+       iomap->bdev = inode->i_sb->s_bdev;
+       iomap->offset = map.m_la;
+       iomap->length = map.m_llen;
+       iomap->flags = 0;
+
+       if (!(map.m_flags & EROFS_MAP_MAPPED)) {
+               iomap->type = IOMAP_HOLE;
+               iomap->addr = IOMAP_NULL_ADDR;
+               if (!iomap->length)
+                       iomap->length = length;
+               return 0;
+       }
+
+       /* that shouldn't happen for now */
+       if (map.m_flags & EROFS_MAP_META) {
+               DBG_BUGON(1);
+               return -ENOTBLK;
+       }
+       iomap->type = IOMAP_MAPPED;
+       iomap->addr = map.m_pa;
+       return 0;
+}
+
+static const struct iomap_ops erofs_iomap_ops = {
+       .iomap_begin = erofs_iomap_begin,
+};
+
+static int erofs_prepare_dio(struct kiocb *iocb, struct iov_iter *to)
+{
+       struct inode *inode = file_inode(iocb->ki_filp);
+       loff_t align = iocb->ki_pos | iov_iter_count(to) |
+               iov_iter_alignment(to);
+       struct block_device *bdev = inode->i_sb->s_bdev;
+       unsigned int blksize_mask;
+
+       if (bdev)
+               blksize_mask = (1 << ilog2(bdev_logical_block_size(bdev))) - 1;
+       else
+               blksize_mask = (1 << inode->i_blkbits) - 1;
+
+       if (align & blksize_mask)
+               return -EINVAL;
+
+       /*
+        * Temporarily fall back tail-packing inline to buffered I/O instead
+        * since tail-packing inline support relies on an iomap core update.
+        */
+       if (EROFS_I(inode)->datalayout == EROFS_INODE_FLAT_INLINE &&
+           iocb->ki_pos + iov_iter_count(to) >
+                       rounddown(inode->i_size, EROFS_BLKSIZ))
+               return 1;
+       return 0;
+}
+
+static ssize_t erofs_file_read_iter(struct kiocb *iocb, struct iov_iter *to)
+{
+       /* no need taking (shared) inode lock since it's a ro filesystem */
+       if (!iov_iter_count(to))
+               return 0;
+
+       if (iocb->ki_flags & IOCB_DIRECT) {
+               int err = erofs_prepare_dio(iocb, to);
+
+               if (!err)
+                       return iomap_dio_rw(iocb, to, &erofs_iomap_ops,
+                                           NULL, 0);
+               if (err < 0)
+                       return err;
+       }
+       return filemap_read(iocb, to, 0);
+}
+
 /* for uncompressed (aligned) files and raw access for other files */
 const struct address_space_operations erofs_raw_access_aops = {
        .readpage = erofs_raw_access_readpage,
        .readahead = erofs_raw_access_readahead,
        .bmap = erofs_bmap,
+       .direct_IO = noop_direct_IO,
+};
+
+const struct file_operations erofs_file_fops = {
+       .llseek         = generic_file_llseek,
+       .read_iter      = erofs_file_read_iter,
+       .mmap           = generic_file_readonly_mmap,
+       .splice_read    = generic_file_splice_read,
 };

diff --git a/fs/erofs/inode.c b/fs/erofs/inode.c
index aa8a0d7..f296e1a 100644 (file)
--- a/fs/erofs/inode.c
+++ b/fs/erofs/inode.c
@@ -247,7 +247,10 @@ static int erofs_fill_inode(struct inode *inode, int isdir)
        switch (inode->i_mode & S_IFMT) {
        case S_IFREG:
                inode->i_op = &erofs_generic_iops;
-               inode->i_fop = &generic_ro_fops;
+               if (erofs_inode_is_data_compressed(vi->datalayout))
+                       inode->i_fop = &generic_ro_fops;
+               else
+                       inode->i_fop = &erofs_file_fops;
                break;
        case S_IFDIR:
                inode->i_op = &erofs_dir_iops;

diff --git a/fs/erofs/internal.h b/fs/erofs/internal.h
index 543c2ff..2669c78 100644 (file)
--- a/fs/erofs/internal.h
+++ b/fs/erofs/internal.h
@@ -371,6 +371,7 @@ static inline int z_erofs_map_blocks_iter(struct inode *inode,
 #endif /* !CONFIG_EROFS_FS_ZIP */
 
 /* data.c */
+extern const struct file_operations erofs_file_fops;
 struct page *erofs_get_meta_page(struct super_block *sb, erofs_blk_t blkaddr);
 
 /* inode.c */