Skip to contentSkip to conversation

erofs: dio/dax support for non-tailpacking cases

PATCH New Gao Xiang 1 participant 3 messages Jul 12, 2021

Subsystems (3)
lore
Share & tools
lore

Gao Xiang

author
#0 lore raw

v2 cover erofs: dio/dax support for non-tailpacking cases

Hi folks,

This patchset mainly adds preliminary EROFS iomap dio/dax support for non-tailpacking uncompressed cases.

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.

Also, DAX is quite useful for some VM use cases in order to save guest memory extremely by using the minimal lightweight EROFS.

Tail-packing inline iomap support will be handled later since currently iomap doesn't support such data pattern, which is independent to non-tailpacking cases.

Comments are welcome. Thanks for your time on reading this!

Thanks,
Gao Xiang

changes since v1:
- allow 'dax=always' and 'dax=never' to keep in sync with ext4/xfs

Gao Xiang (1):
erofs: dax support for non-tailpacking regular file

Huang Jianan (1):
erofs: iomap support for non-tailpacking DIO

 fs/erofs/Kconfig    |   1 +
 fs/erofs/data.c     | 142 +++++++++++++++++++++++++++++++++++++++++++-
 fs/erofs/inode.c    |   9 ++-
 fs/erofs/internal.h |   4 ++
 fs/erofs/super.c    |  60 ++++++++++++++++++-
 5 files changed, 212 insertions(+), 4 deletions(-)
Signature · 1 line
2.24.4
author
#1 lore raw

PATCH v2 1/2 erofs: iomap support for non-tailpacking DIO

From: Huang Jianan <[email protected]>

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 can be handled later.

---
 fs/erofs/Kconfig    |   1 +
 fs/erofs/data.c     | 102 ++++++++++++++++++++++++++++++++++++++++++++
 fs/erofs/inode.c    |   5 ++-
 fs/erofs/internal.h |   1 +
 4 files changed, 108 insertions(+), 1 deletion(-)
fs/erofs/Kconfig +1 -0
index 906af0c1998c..14b747026742 100644
@@ -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
fs/erofs/data.c +102 -0

117 lines not on this page · raw message

fs/erofs/inode.c +4 -1
index aa8a0d770ba3..00edb7562fea 100644
@@ -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 = &erofs_file_fops;
+		else
+			inode->i_fop = &generic_ro_fops;
 		break;
 	case S_IFDIR:
 		inode->i_op = &erofs_dir_iops;
fs/erofs/internal.h +1 -0
index 543c2ff97d30..2669c785d548 100644
@@ -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 */
Signature · 1 line
2.24.4
author
#2 lore raw

PATCH v2 2/2 erofs: dax support for non-tailpacking regular file

In reply to Gao Xiang

DAX is quite useful for some VM use cases in order to save guest memory extremely with minimal lightweight EROFS.

In order to prepare for such use cases, add preliminary dax support for non-tailpacking regular files for now.

Tested with the DRAM-emulated PMEM and the EROFS image generated by "mkfs.erofs -Enoinline_data enwik9.fsdax.img enwik9"

---
 fs/erofs/data.c     | 42 +++++++++++++++++++++++++++++--
 fs/erofs/inode.c    |  4 +++
 fs/erofs/internal.h |  3 +++
 fs/erofs/super.c    | 60 +++++++++++++++++++++++++++++++++++++++++++--
 4 files changed, 105 insertions(+), 4 deletions(-)
fs/erofs/data.c +40 -2
index 0f82b4cb474c..00493855319a 100644
@@ -6,7 +6,7 @@
 #include "internal.h"
 #include <linux/prefetch.h>
 #include <linux/iomap.h>
-
+#include <linux/dax.h>
 #include <trace/events/erofs.h>
 
 static void erofs_readendio(struct bio *bio)
@@ -323,6 +323,7 @@ static int erofs_iomap_begin(struct inode *inode, loff_t offset, loff_t length,
 		return ret;
 
 	iomap->bdev = inode->i_sb->s_bdev;
+	iomap->dax_dev = EROFS_I_SB(inode)->dax_dev;
 	iomap->offset = map.m_la;
 	iomap->length = map.m_llen;
 
@@ -382,6 +383,10 @@ static ssize_t erofs_file_read_iter(struct kiocb *iocb, struct iov_iter *to)
 	if (!iov_iter_count(to))
 		return 0;
 
+#ifdef CONFIG_FS_DAX
+	if (IS_DAX(iocb->ki_filp->f_mapping->host))
+		return dax_iomap_rw(iocb, to, &erofs_iomap_ops);
+#endif
 	if (iocb->ki_flags & IOCB_DIRECT) {
 		int err = erofs_prepare_dio(iocb, to);
 
@@ -410,9 +415,42 @@ const struct address_space_operations erofs_raw_access_aops = {
 	.direct_IO = noop_direct_IO,
 };
 
+#ifdef CONFIG_FS_DAX
+static vm_fault_t erofs_dax_huge_fault(struct vm_fault *vmf,
+		enum page_entry_size pe_size)
+{
+	return dax_iomap_fault(vmf, pe_size, NULL, NULL, &erofs_iomap_ops);
+}
+
+static vm_fault_t erofs_dax_fault(struct vm_fault *vmf)
+{
+	return erofs_dax_huge_fault(vmf, PE_SIZE_PTE);
+}
+
+static const struct vm_operations_struct erofs_dax_vm_ops = {
+	.fault		= erofs_dax_fault,
+	.huge_fault	= erofs_dax_huge_fault,
+};
+
+static int erofs_file_mmap(struct file *file, struct vm_area_struct *vma)
+{
+	if (!IS_DAX(file_inode(file)))
+		return generic_file_readonly_mmap(file, vma);
+
+	if ((vma->vm_flags & VM_SHARED) && (vma->vm_flags & VM_MAYWRITE))
+		return -EINVAL;
+
+	vma->vm_ops = &erofs_dax_vm_ops;
+	vma->vm_flags |= VM_HUGEPAGE;
+	return 0;
+}
+#else
+#define erofs_file_mmap	generic_file_readonly_mmap
+#endif
+
 const struct file_operations erofs_file_fops = {
 	.llseek		= generic_file_llseek,
 	.read_iter	= erofs_file_read_iter,
-	.mmap		= generic_file_readonly_mmap,
+	.mmap		= erofs_file_mmap,
 	.splice_read	= generic_file_splice_read,
 };
fs/erofs/inode.c +4 -0
index 00edb7562fea..e875fba18159 100644
@@ -174,6 +174,10 @@ static struct page *erofs_read_inode(struct inode *inode,
 	inode->i_mtime.tv_nsec = inode->i_ctime.tv_nsec;
 	inode->i_atime.tv_nsec = inode->i_ctime.tv_nsec;
 
+	inode->i_flags &= ~S_DAX;
+	if (test_opt(&sbi->ctx, DAX_ALWAYS) && S_ISREG(inode->i_mode) &&
+	    vi->datalayout == EROFS_INODE_FLAT_PLAIN)
+		inode->i_flags |= S_DAX;
 	if (!nblks)
 		/* measure inode.i_blocks as generic filesystems */
 		inode->i_blocks = roundup(inode->i_size, EROFS_BLKSIZ) >> 9;
fs/erofs/internal.h +3 -0
index 2669c785d548..7c9abfc93109 100644
@@ -83,6 +83,7 @@ struct erofs_sb_info {
 
 	struct erofs_sb_lz4_info lz4;
 #endif	/* CONFIG_EROFS_FS_ZIP */
+	struct dax_device *dax_dev;
 	u32 blocks;
 	u32 meta_blkaddr;
 #ifdef CONFIG_EROFS_FS_XATTR
@@ -115,6 +116,8 @@ struct erofs_sb_info {
 /* Mount flags set via mount options or defaults */
 #define EROFS_MOUNT_XATTR_USER		0x00000010
 #define EROFS_MOUNT_POSIX_ACL		0x00000020
+#define EROFS_MOUNT_DAX_ALWAYS		0x00000040
+#define EROFS_MOUNT_DAX_NEVER		0x00000080
 
 #define clear_opt(ctx, option)	((ctx)->mount_opt &= ~EROFS_MOUNT_##option)
 #define set_opt(ctx, option)	((ctx)->mount_opt |= EROFS_MOUNT_##option)
fs/erofs/super.c +58 -2

120 lines not on this page · raw message

Signature · 1 line
2.24.4