[PATCH v4 0/3] Casefolding in F2FS

[PATCH v4 0/3] Casefolding in F2FS

[Daniel Rosenberg]

These patches are largely based on the casefolding patches for ext4

v4: Added FS_CASEFOLD_FL flag, added documentation that escaped the last
    format-patch, moved setting dentry ops to f2fs_setup_casefold
v3: Addressed feedback, apart from F2FS_CASEFOLD_FL/FS_CASEFOLD_FL
    Added sysfs file "encoding" to see the encoding set on a filesystem
v2: Rebased patches again master, changed f2fs_msg to f2fs_info/f2fs_err

Daniel Rosenberg (3):
  fs: Reserve flag for casefolding
  f2fs: include charset encoding information in the superblock
  f2fs: Support case-insensitive file name lookups

 Documentation/ABI/testing/sysfs-fs-f2fs |   7 ++
 Documentation/filesystems/f2fs.txt      |   3 +
 fs/f2fs/dir.c                           | 126 ++++++++++++++++++++++--
 fs/f2fs/f2fs.h                          |  21 +++-
 fs/f2fs/file.c                          |  16 ++-
 fs/f2fs/hash.c                          |  35 ++++++-
 fs/f2fs/inline.c                        |   4 +-
 fs/f2fs/inode.c                         |   4 +-
 fs/f2fs/namei.c                         |  21 ++++
 fs/f2fs/super.c                         |  96 ++++++++++++++++++
 fs/f2fs/sysfs.c                         |  23 +++++
 include/linux/f2fs_fs.h                 |   9 +-
 include/uapi/linux/fs.h                 |   1 +
 tools/include/uapi/linux/fs.h           |   1 +
 14 files changed, 347 insertions(+), 20 deletions(-)

-- 
2.22.0.657.g960e92d24f-goog

[PATCH v4 1/3] fs: Reserve flag for casefolding

[Daniel Rosenberg]

In preparation for including the casefold feature within f2fs, elevate
the EXT4_CASEFOLD_FL flag to FS_CASEFOLD_FL.

Signed-off-by: Daniel Rosenberg <drosen@google.com>
---
 include/uapi/linux/fs.h       | 1 +
 tools/include/uapi/linux/fs.h | 1 +
 2 files changed, 2 insertions(+)

diff --git a/include/uapi/linux/fs.h b/include/uapi/linux/fs.h
index 59c71fa8c553a..2a616aa3f6862 100644
--- a/include/uapi/linux/fs.h
+++ b/include/uapi/linux/fs.h
@@ -311,6 +311,7 @@ struct fscrypt_key {
 #define FS_NOCOW_FL			0x00800000 /* Do not cow file */
 #define FS_INLINE_DATA_FL		0x10000000 /* Reserved for ext4 */
 #define FS_PROJINHERIT_FL		0x20000000 /* Create with parents projid */
+#define FS_CASEFOLD_FL			0x40000000 /* Folder is case insensitive */
 #define FS_RESERVED_FL			0x80000000 /* reserved for ext2 lib */
 
 #define FS_FL_USER_VISIBLE		0x0003DFFF /* User visible flags */
diff --git a/tools/include/uapi/linux/fs.h b/tools/include/uapi/linux/fs.h
index 59c71fa8c553a..2a616aa3f6862 100644
--- a/tools/include/uapi/linux/fs.h
+++ b/tools/include/uapi/linux/fs.h
@@ -311,6 +311,7 @@ struct fscrypt_key {
 #define FS_NOCOW_FL			0x00800000 /* Do not cow file */
 #define FS_INLINE_DATA_FL		0x10000000 /* Reserved for ext4 */
 #define FS_PROJINHERIT_FL		0x20000000 /* Create with parents projid */
+#define FS_CASEFOLD_FL			0x40000000 /* Folder is case insensitive */
 #define FS_RESERVED_FL			0x80000000 /* reserved for ext2 lib */
 
 #define FS_FL_USER_VISIBLE		0x0003DFFF /* User visible flags */
-- 
2.22.0.657.g960e92d24f-goog

[PATCH v4 2/3] f2fs: include charset encoding information in the superblock

[Daniel Rosenberg]

Add charset encoding to f2fs to support casefolding. It is modeled after
the same feature introduced in commit c83ad55eaa91 ("ext4: include charset
encoding information in the superblock")

Currently this is not compatible with encryption, similar to the current
ext4 imlpementation. This will change in the future.

From the ext4 patch:
"""
The s_encoding field stores a magic number indicating the encoding
format and version used globally by file and directory names in the
filesystem.  The s_encoding_flags defines policies for using the charset
encoding, like how to handle invalid sequences.  The magic number is
mapped to the exact charset table, but the mapping is specific to ext4.
Since we don't have any commitment to support old encodings, the only
encoding I am supporting right now is utf8-12.1.0.

The current implementation prevents the user from enabling encoding and
per-directory encryption on the same filesystem at the same time.  The
incompatibility between these features lies in how we do efficient
directory searches when we cannot be sure the encryption of the user
provided fname will match the actual hash stored in the disk without
decrypting every directory entry, because of normalization cases.  My
quickest solution is to simply block the concurrent use of these
features for now, and enable it later, once we have a better solution.
"""

Signed-off-by: Daniel Rosenberg <drosen@google.com>
---
 Documentation/ABI/testing/sysfs-fs-f2fs |  7 ++
 Documentation/filesystems/f2fs.txt      |  3 +
 fs/f2fs/f2fs.h                          |  6 ++
 fs/f2fs/super.c                         | 95 +++++++++++++++++++++++++
 fs/f2fs/sysfs.c                         | 23 ++++++
 include/linux/f2fs_fs.h                 |  9 ++-
 6 files changed, 142 insertions(+), 1 deletion(-)

diff --git a/Documentation/ABI/testing/sysfs-fs-f2fs b/Documentation/ABI/testing/sysfs-fs-f2fs
index dca326e0ee3e1..7ab2b1b5e255f 100644
--- a/Documentation/ABI/testing/sysfs-fs-f2fs
+++ b/Documentation/ABI/testing/sysfs-fs-f2fs
@@ -251,3 +251,10 @@ Description:
 		If checkpoint=disable, it displays the number of blocks that are unusable.
                 If checkpoint=enable it displays the enumber of blocks that would be unusable
                 if checkpoint=disable were to be set.
+
+What:		/sys/fs/f2fs/<disk>/encoding
+Date		July 2019
+Contact:	"Daniel Rosenberg" <drosen@google.com>
+Description:
+		Displays name and version of the encoding set for the filesystem.
+                If no encoding is set, displays (none)
diff --git a/Documentation/filesystems/f2fs.txt b/Documentation/filesystems/f2fs.txt
index 496fa28b24925..5fa38ab373cab 100644
--- a/Documentation/filesystems/f2fs.txt
+++ b/Documentation/filesystems/f2fs.txt
@@ -413,6 +413,9 @@ Files in /sys/fs/f2fs/<devname>
                               that would be unusable if checkpoint=disable were
                               to be set.
 
+encoding 	              This shows the encoding used for casefolding.
+                              If casefolding is not enabled, returns (none)
+
 ================================================================================
 USAGE
 ================================================================================
diff --git a/fs/f2fs/f2fs.h b/fs/f2fs/f2fs.h
index 17382da7f0bd9..c6c7904572d0d 100644
--- a/fs/f2fs/f2fs.h
+++ b/fs/f2fs/f2fs.h
@@ -153,6 +153,7 @@ struct f2fs_mount_info {
 #define F2FS_FEATURE_LOST_FOUND		0x0200
 #define F2FS_FEATURE_VERITY		0x0400	/* reserved */
 #define F2FS_FEATURE_SB_CHKSUM		0x0800
+#define F2FS_FEATURE_CASEFOLD		0x1000
 
 #define __F2FS_HAS_FEATURE(raw_super, mask)				\
 	((raw_super->feature & cpu_to_le32(mask)) != 0)
@@ -1169,6 +1170,10 @@ struct f2fs_sb_info {
 	int valid_super_block;			/* valid super block no */
 	unsigned long s_flag;				/* flags for sbi */
 	struct mutex writepages;		/* mutex for writepages() */
+#ifdef CONFIG_UNICODE
+	struct unicode_map *s_encoding;
+	__u16 s_encoding_flags;
+#endif
 
 #ifdef CONFIG_BLK_DEV_ZONED
 	unsigned int blocks_per_blkz;		/* F2FS blocks per zone */
@@ -3562,6 +3567,7 @@ F2FS_FEATURE_FUNCS(quota_ino, QUOTA_INO);
 F2FS_FEATURE_FUNCS(inode_crtime, INODE_CRTIME);
 F2FS_FEATURE_FUNCS(lost_found, LOST_FOUND);
 F2FS_FEATURE_FUNCS(sb_chksum, SB_CHKSUM);
+F2FS_FEATURE_FUNCS(casefold, CASEFOLD);
 
 #ifdef CONFIG_BLK_DEV_ZONED
 static inline bool f2fs_blkz_is_seq(struct f2fs_sb_info *sbi, int devi,
diff --git a/fs/f2fs/super.c b/fs/f2fs/super.c
index 6de6cda440315..068db78a1e03e 100644
--- a/fs/f2fs/super.c
+++ b/fs/f2fs/super.c
@@ -23,6 +23,7 @@
 #include <linux/f2fs_fs.h>
 #include <linux/sysfs.h>
 #include <linux/quota.h>
+#include <linux/unicode.h>
 
 #include "f2fs.h"
 #include "node.h"
@@ -222,6 +223,36 @@ void f2fs_printk(struct f2fs_sb_info *sbi, const char *fmt, ...)
 	va_end(args);
 }
 
+#ifdef CONFIG_UNICODE
+static const struct f2fs_sb_encodings {
+	__u16 magic;
+	char *name;
+	char *version;
+} f2fs_sb_encoding_map[] = {
+	{F2FS_ENC_UTF8_12_1, "utf8", "12.1.0"},
+};
+
+static int f2fs_sb_read_encoding(const struct f2fs_super_block *sb,
+				 const struct f2fs_sb_encodings **encoding,
+				 __u16 *flags)
+{
+	__u16 magic = le16_to_cpu(sb->s_encoding);
+	int i;
+
+	for (i = 0; i < ARRAY_SIZE(f2fs_sb_encoding_map); i++)
+		if (magic == f2fs_sb_encoding_map[i].magic)
+			break;
+
+	if (i >= ARRAY_SIZE(f2fs_sb_encoding_map))
+		return -EINVAL;
+
+	*encoding = &f2fs_sb_encoding_map[i];
+	*flags = le16_to_cpu(sb->s_encoding_flags);
+
+	return 0;
+}
+#endif
+
 static inline void limit_reserve_root(struct f2fs_sb_info *sbi)
 {
 	block_t limit = min((sbi->user_block_count << 1) / 1000,
@@ -798,6 +829,13 @@ static int parse_options(struct super_block *sb, char *options)
 		return -EINVAL;
 	}
 #endif
+#ifndef CONFIG_UNICODE
+	if (f2fs_sb_has_casefold(sbi)) {
+		f2fs_err(sbi,
+			"Filesystem with casefold feature cannot be mounted without CONFIG_UNICODE");
+		return -EINVAL;
+	}
+#endif
 
 	if (F2FS_IO_SIZE_BITS(sbi) && !test_opt(sbi, LFS)) {
 		f2fs_err(sbi, "Should set mode=lfs with %uKB-sized IO",
@@ -1089,6 +1127,9 @@ static void f2fs_put_super(struct super_block *sb)
 	destroy_percpu_info(sbi);
 	for (i = 0; i < NR_PAGE_TYPE; i++)
 		kvfree(sbi->write_io[i]);
+#ifdef CONFIG_UNICODE
+	utf8_unload(sbi->s_encoding);
+#endif
 	kvfree(sbi);
 }
 
@@ -3031,6 +3072,52 @@ static int f2fs_scan_devices(struct f2fs_sb_info *sbi)
 	return 0;
 }
 
+static int f2fs_setup_casefold(struct f2fs_sb_info *sbi)
+{
+#ifdef CONFIG_UNICODE
+	if (f2fs_sb_has_casefold(sbi) && !sbi->s_encoding) {
+		const struct f2fs_sb_encodings *encoding_info;
+		struct unicode_map *encoding;
+		__u16 encoding_flags;
+
+		if (f2fs_sb_has_encrypt(sbi)) {
+			f2fs_err(sbi,
+				"Can't mount with encoding and encryption");
+			return -EINVAL;
+		}
+
+		if (f2fs_sb_read_encoding(sbi->raw_super, &encoding_info,
+					  &encoding_flags)) {
+			f2fs_err(sbi,
+				 "Encoding requested by superblock is unknown");
+			return -EINVAL;
+		}
+
+		encoding = utf8_load(encoding_info->version);
+		if (IS_ERR(encoding)) {
+			f2fs_err(sbi,
+				 "can't mount with superblock charset: %s-%s "
+				 "not supported by the kernel. flags: 0x%x.",
+				 encoding_info->name, encoding_info->version,
+				 encoding_flags);
+			return PTR_ERR(encoding);
+		}
+		f2fs_info(sbi, "Using encoding defined by superblock: "
+			 "%s-%s with flags 0x%hx", encoding_info->name,
+			 encoding_info->version?:"\b", encoding_flags);
+
+		sbi->s_encoding = encoding;
+		sbi->s_encoding_flags = encoding_flags;
+	}
+#else
+	if (f2fs_sb_has_casefold(sbi)) {
+		f2fs_err(sbi, "Filesystem with casefold feature cannot be mounted without CONFIG_UNICODE");
+		return -EINVAL;
+	}
+#endif
+	return 0;
+}
+
 static void f2fs_tuning_parameters(struct f2fs_sb_info *sbi)
 {
 	struct f2fs_sm_info *sm_i = SM_I(sbi);
@@ -3127,6 +3214,10 @@ static int f2fs_fill_super(struct super_block *sb, void *data, int silent)
 				le32_to_cpu(raw_super->log_blocksize);
 	sb->s_max_links = F2FS_LINK_MAX;
 
+	err = f2fs_setup_casefold(sbi);
+	if (err)
+		goto free_options;
+
 #ifdef CONFIG_QUOTA
 	sb->dq_op = &f2fs_quota_operations;
 	sb->s_qcop = &f2fs_quotactl_ops;
@@ -3477,6 +3568,10 @@ static int f2fs_fill_super(struct super_block *sb, void *data, int silent)
 free_bio_info:
 	for (i = 0; i < NR_PAGE_TYPE; i++)
 		kvfree(sbi->write_io[i]);
+
+#ifdef CONFIG_UNICODE
+	utf8_unload(sbi->s_encoding);
+#endif
 free_options:
 #ifdef CONFIG_QUOTA
 	for (i = 0; i < MAXQUOTAS; i++)
diff --git a/fs/f2fs/sysfs.c b/fs/f2fs/sysfs.c
index 3aeacd0aacfd2..f9fcca695db9f 100644
--- a/fs/f2fs/sysfs.c
+++ b/fs/f2fs/sysfs.c
@@ -10,6 +10,7 @@
 #include <linux/proc_fs.h>
 #include <linux/f2fs_fs.h>
 #include <linux/seq_file.h>
+#include <linux/unicode.h>
 
 #include "f2fs.h"
 #include "segment.h"
@@ -81,6 +82,19 @@ static ssize_t unusable_show(struct f2fs_attr *a,
 		(unsigned long long)unusable);
 }
 
+static ssize_t encoding_show(struct f2fs_attr *a,
+		struct f2fs_sb_info *sbi, char *buf)
+{
+#ifdef CONFIG_UNICODE
+	if (f2fs_sb_has_casefold(sbi))
+		return snprintf(buf, PAGE_SIZE, "%s (%d.%d.%d)\n",
+			sbi->s_encoding->charset,
+			(sbi->s_encoding->version >> 16) & 0xff,
+			(sbi->s_encoding->version >> 8) & 0xff,
+			sbi->s_encoding->version & 0xff);
+#endif
+	return snprintf(buf, PAGE_SIZE, "(none)");
+}
 
 static ssize_t lifetime_write_kbytes_show(struct f2fs_attr *a,
 		struct f2fs_sb_info *sbi, char *buf)
@@ -134,6 +148,9 @@ static ssize_t features_show(struct f2fs_attr *a,
 	if (f2fs_sb_has_sb_chksum(sbi))
 		len += snprintf(buf + len, PAGE_SIZE - len, "%s%s",
 				len ? ", " : "", "sb_checksum");
+	if (f2fs_sb_has_casefold(sbi))
+		len += snprintf(buf + len, PAGE_SIZE - len, "%s%s",
+				len ? ", " : "", "casefold");
 	len += snprintf(buf + len, PAGE_SIZE - len, "\n");
 	return len;
 }
@@ -365,6 +382,7 @@ enum feat_id {
 	FEAT_INODE_CRTIME,
 	FEAT_LOST_FOUND,
 	FEAT_SB_CHECKSUM,
+	FEAT_CASEFOLD,
 };
 
 static ssize_t f2fs_feature_show(struct f2fs_attr *a,
@@ -382,6 +400,7 @@ static ssize_t f2fs_feature_show(struct f2fs_attr *a,
 	case FEAT_INODE_CRTIME:
 	case FEAT_LOST_FOUND:
 	case FEAT_SB_CHECKSUM:
+	case FEAT_CASEFOLD:
 		return snprintf(buf, PAGE_SIZE, "supported\n");
 	}
 	return 0;
@@ -455,6 +474,7 @@ F2FS_GENERAL_RO_ATTR(lifetime_write_kbytes);
 F2FS_GENERAL_RO_ATTR(features);
 F2FS_GENERAL_RO_ATTR(current_reserved_blocks);
 F2FS_GENERAL_RO_ATTR(unusable);
+F2FS_GENERAL_RO_ATTR(encoding);
 
 #ifdef CONFIG_FS_ENCRYPTION
 F2FS_FEATURE_RO_ATTR(encryption, FEAT_CRYPTO);
@@ -471,6 +491,7 @@ F2FS_FEATURE_RO_ATTR(quota_ino, FEAT_QUOTA_INO);
 F2FS_FEATURE_RO_ATTR(inode_crtime, FEAT_INODE_CRTIME);
 F2FS_FEATURE_RO_ATTR(lost_found, FEAT_LOST_FOUND);
 F2FS_FEATURE_RO_ATTR(sb_checksum, FEAT_SB_CHECKSUM);
+F2FS_FEATURE_RO_ATTR(casefold, FEAT_CASEFOLD);
 
 #define ATTR_LIST(name) (&f2fs_attr_##name.attr)
 static struct attribute *f2fs_attrs[] = {
@@ -515,6 +536,7 @@ static struct attribute *f2fs_attrs[] = {
 	ATTR_LIST(features),
 	ATTR_LIST(reserved_blocks),
 	ATTR_LIST(current_reserved_blocks),
+	ATTR_LIST(encoding),
 	NULL,
 };
 ATTRIBUTE_GROUPS(f2fs);
@@ -535,6 +557,7 @@ static struct attribute *f2fs_feat_attrs[] = {
 	ATTR_LIST(inode_crtime),
 	ATTR_LIST(lost_found),
 	ATTR_LIST(sb_checksum),
+	ATTR_LIST(casefold),
 	NULL,
 };
 ATTRIBUTE_GROUPS(f2fs_feat);
diff --git a/include/linux/f2fs_fs.h b/include/linux/f2fs_fs.h
index 65559900d4d76..b7c9c7f721339 100644
--- a/include/linux/f2fs_fs.h
+++ b/include/linux/f2fs_fs.h
@@ -36,6 +36,11 @@
 
 #define F2FS_MAX_QUOTAS		3
 
+#define F2FS_ENC_UTF8_12_1	1
+#define F2FS_ENC_STRICT_MODE_FL	(1 << 0)
+#define f2fs_has_strict_mode(sbi) \
+	(sbi->s_encoding_flags & F2FS_ENC_STRICT_MODE_FL)
+
 #define F2FS_IO_SIZE(sbi)	(1 << F2FS_OPTION(sbi).write_io_size_bits) /* Blocks */
 #define F2FS_IO_SIZE_KB(sbi)	(1 << (F2FS_OPTION(sbi).write_io_size_bits + 2)) /* KB */
 #define F2FS_IO_SIZE_BYTES(sbi)	(1 << (F2FS_OPTION(sbi).write_io_size_bits + 12)) /* B */
@@ -109,7 +114,9 @@ struct f2fs_super_block {
 	struct f2fs_device devs[MAX_DEVICES];	/* device list */
 	__le32 qf_ino[F2FS_MAX_QUOTAS];	/* quota inode numbers */
 	__u8 hot_ext_count;		/* # of hot file extension */
-	__u8 reserved[310];		/* valid reserved region */
+	__le16  s_encoding;		/* Filename charset encoding */
+	__le16  s_encoding_flags;	/* Filename charset encoding flags */
+	__u8 reserved[306];		/* valid reserved region */
 	__le32 crc;			/* checksum of superblock */
 } __packed;
 
-- 
2.22.0.657.g960e92d24f-goog

[PATCH v4 3/3] f2fs: Support case-insensitive file name lookups

[Daniel Rosenberg]

Modeled after commit b886ee3e778e ("ext4: Support case-insensitive file
name lookups")

"""
This patch implements the actual support for case-insensitive file name
lookups in f2fs, based on the feature bit and the encoding stored in the
superblock.

A filesystem that has the casefold feature set is able to configure
directories with the +F (F2FS_CASEFOLD_FL) attribute, enabling lookups
to succeed in that directory in a case-insensitive fashion, i.e: match
a directory entry even if the name used by userspace is not a byte per
byte match with the disk name, but is an equivalent case-insensitive
version of the Unicode string.  This operation is called a
case-insensitive file name lookup.

The feature is configured as an inode attribute applied to directories
and inherited by its children.  This attribute can only be enabled on
empty directories for filesystems that support the encoding feature,
thus preventing collision of file names that only differ by case.

* dcache handling:

For a +F directory, F2Fs only stores the first equivalent name dentry
used in the dcache. This is done to prevent unintentional duplication of
dentries in the dcache, while also allowing the VFS code to quickly find
the right entry in the cache despite which equivalent string was used in
a previous lookup, without having to resort to ->lookup().

d_hash() of casefolded directories is implemented as the hash of the
casefolded string, such that we always have a well-known bucket for all
the equivalencies of the same string. d_compare() uses the
utf8_strncasecmp() infrastructure, which handles the comparison of
equivalent, same case, names as well.

For now, negative lookups are not inserted in the dcache, since they
would need to be invalidated anyway, because we can't trust missing file
dentries.  This is bad for performance but requires some leveraging of
the vfs layer to fix.  We can live without that for now, and so does
everyone else.

* on-disk data:

Despite using a specific version of the name as the internal
representation within the dcache, the name stored and fetched from the
disk is a byte-per-byte match with what the user requested, making this
implementation 'name-preserving'. i.e. no actual information is lost
when writing to storage.

DX is supported by modifying the hashes used in +F directories to make
them case/encoding-aware.  The new disk hashes are calculated as the
hash of the full casefolded string, instead of the string directly.
This allows us to efficiently search for file names in the htree without
requiring the user to provide an exact name.

* Dealing with invalid sequences:

By default, when a invalid UTF-8 sequence is identified, ext4 will treat
it as an opaque byte sequence, ignoring the encoding and reverting to
the old behavior for that unique file.  This means that case-insensitive
file name lookup will not work only for that file.  An optional bit can
be set in the superblock telling the filesystem code and userspace tools
to enforce the encoding.  When that optional bit is set, any attempt to
create a file name using an invalid UTF-8 sequence will fail and return
an error to userspace.

* Normalization algorithm:

The UTF-8 algorithms used to compare strings in f2fs is implemented
in fs/unicode, and is based on a previous version developed by
SGI.  It implements the Canonical decomposition (NFD) algorithm
described by the Unicode specification 12.1, or higher, combined with
the elimination of ignorable code points (NFDi) and full
case-folding (CF) as documented in fs/unicode/utf8_norm.c.

NFD seems to be the best normalization method for F2FS because:

  - It has a lower cost than NFC/NFKC (which requires
    decomposing to NFD as an intermediary step)
  - It doesn't eliminate important semantic meaning like
    compatibility decompositions.

Although:

- This implementation is not completely linguistic accurate, because
different languages have conflicting rules, which would require the
specialization of the filesystem to a given locale, which brings all
sorts of problems for removable media and for users who use more than
one language.
"""

Signed-off-by: Daniel Rosenberg <drosen@google.com>
---
 fs/f2fs/dir.c    | 126 +++++++++++++++++++++++++++++++++++++++++++----
 fs/f2fs/f2fs.h   |  15 ++++--
 fs/f2fs/file.c   |  16 +++++-
 fs/f2fs/hash.c   |  35 ++++++++++++-
 fs/f2fs/inline.c |   4 +-
 fs/f2fs/inode.c  |   4 +-
 fs/f2fs/namei.c  |  21 ++++++++
 fs/f2fs/super.c  |   1 +
 8 files changed, 203 insertions(+), 19 deletions(-)

diff --git a/fs/f2fs/dir.c b/fs/f2fs/dir.c
index 85a1528f319f2..2913483473f30 100644
--- a/fs/f2fs/dir.c
+++ b/fs/f2fs/dir.c
@@ -8,6 +8,7 @@
 #include <linux/fs.h>
 #include <linux/f2fs_fs.h>
 #include <linux/sched/signal.h>
+#include <linux/unicode.h>
 #include "f2fs.h"
 #include "node.h"
 #include "acl.h"
@@ -81,7 +82,8 @@ static unsigned long dir_block_index(unsigned int level,
 	return bidx;
 }
 
-static struct f2fs_dir_entry *find_in_block(struct page *dentry_page,
+static struct f2fs_dir_entry *find_in_block(struct inode *dir,
+				struct page *dentry_page,
 				struct fscrypt_name *fname,
 				f2fs_hash_t namehash,
 				int *max_slots,
@@ -93,7 +95,7 @@ static struct f2fs_dir_entry *find_in_block(struct page *dentry_page,
 
 	dentry_blk = (struct f2fs_dentry_block *)page_address(dentry_page);
 
-	make_dentry_ptr_block(NULL, &d, dentry_blk);
+	make_dentry_ptr_block(dir, &d, dentry_blk);
 	de = f2fs_find_target_dentry(fname, namehash, max_slots, &d);
 	if (de)
 		*res_page = dentry_page;
@@ -101,6 +103,39 @@ static struct f2fs_dir_entry *find_in_block(struct page *dentry_page,
 	return de;
 }
 
+#ifdef CONFIG_UNICODE
+/*
+ * Test whether a case-insensitive directory entry matches the filename
+ * being searched for.
+ *
+ * Returns: 0 if the directory entry matches, more than 0 if it
+ * doesn't match or less than zero on error.
+ */
+int f2fs_ci_compare(const struct inode *parent, const struct qstr *name,
+		    const struct qstr *entry)
+{
+	const struct f2fs_sb_info *sbi = F2FS_SB(parent->i_sb);
+	const struct unicode_map *um = sbi->s_encoding;
+	int ret;
+
+	ret = utf8_strncasecmp(um, name, entry);
+	if (ret < 0) {
+		/* Handle invalid character sequence as either an error
+		 * or as an opaque byte sequence.
+		 */
+		if (f2fs_has_strict_mode(sbi))
+			return -EINVAL;
+
+		if (name->len != entry->len)
+			return 1;
+
+		return !!memcmp(name->name, entry->name, name->len);
+	}
+
+	return ret;
+}
+#endif
+
 struct f2fs_dir_entry *f2fs_find_target_dentry(struct fscrypt_name *fname,
 			f2fs_hash_t namehash, int *max_slots,
 			struct f2fs_dentry_ptr *d)
@@ -108,6 +143,9 @@ struct f2fs_dir_entry *f2fs_find_target_dentry(struct fscrypt_name *fname,
 	struct f2fs_dir_entry *de;
 	unsigned long bit_pos = 0;
 	int max_len = 0;
+#ifdef CONFIG_UNICODE
+	struct qstr entry;
+#endif
 
 	if (max_slots)
 		*max_slots = 0;
@@ -119,16 +157,28 @@ struct f2fs_dir_entry *f2fs_find_target_dentry(struct fscrypt_name *fname,
 		}
 
 		de = &d->dentry[bit_pos];
+#ifdef CONFIG_UNICODE
+		entry.name = d->filename[bit_pos];
+		entry.len = de->name_len;
+#endif
 
 		if (unlikely(!de->name_len)) {
 			bit_pos++;
 			continue;
 		}
+		if (de->hash_code == namehash) {
+#ifdef CONFIG_UNICODE
+			if (F2FS_SB(d->inode->i_sb)->s_encoding &&
+					IS_CASEFOLDED(d->inode) &&
+					!f2fs_ci_compare(d->inode,
+						fname->usr_fname, &entry))
+				goto found;
 
-		if (de->hash_code == namehash &&
-		    fscrypt_match_name(fname, d->filename[bit_pos],
-				       le16_to_cpu(de->name_len)))
-			goto found;
+#endif
+			if (fscrypt_match_name(fname, d->filename[bit_pos],
+						le16_to_cpu(de->name_len)))
+				goto found;
+		}
 
 		if (max_slots && max_len > *max_slots)
 			*max_slots = max_len;
@@ -157,7 +207,7 @@ static struct f2fs_dir_entry *find_in_level(struct inode *dir,
 	struct f2fs_dir_entry *de = NULL;
 	bool room = false;
 	int max_slots;
-	f2fs_hash_t namehash = f2fs_dentry_hash(&name, fname);
+	f2fs_hash_t namehash = f2fs_dentry_hash(dir, &name, fname);
 
 	nbucket = dir_buckets(level, F2FS_I(dir)->i_dir_level);
 	nblock = bucket_blocks(level);
@@ -179,8 +229,8 @@ static struct f2fs_dir_entry *find_in_level(struct inode *dir,
 			}
 		}
 
-		de = find_in_block(dentry_page, fname, namehash, &max_slots,
-								res_page);
+		de = find_in_block(dir, dentry_page, fname, namehash,
+							&max_slots, res_page);
 		if (de)
 			break;
 
@@ -250,6 +300,14 @@ struct f2fs_dir_entry *f2fs_find_entry(struct inode *dir,
 	struct fscrypt_name fname;
 	int err;
 
+#ifdef CONFIG_UNICODE
+	if (f2fs_has_strict_mode(F2FS_I_SB(dir)) && IS_CASEFOLDED(dir) &&
+			utf8_validate(F2FS_I_SB(dir)->s_encoding, child)) {
+		*res_page = ERR_PTR(-EINVAL);
+		return NULL;
+	}
+#endif
+
 	err = fscrypt_setup_filename(dir, child, 1, &fname);
 	if (err) {
 		if (err == -ENOENT)
@@ -504,7 +562,7 @@ int f2fs_add_regular_entry(struct inode *dir, const struct qstr *new_name,
 
 	level = 0;
 	slots = GET_DENTRY_SLOTS(new_name->len);
-	dentry_hash = f2fs_dentry_hash(new_name, NULL);
+	dentry_hash = f2fs_dentry_hash(dir, new_name, NULL);
 
 	current_depth = F2FS_I(dir)->i_current_depth;
 	if (F2FS_I(dir)->chash == dentry_hash) {
@@ -943,3 +1001,51 @@ const struct file_operations f2fs_dir_operations = {
 	.compat_ioctl   = f2fs_compat_ioctl,
 #endif
 };
+
+#ifdef CONFIG_UNICODE
+static int f2fs_d_compare(const struct dentry *dentry, unsigned int len,
+			  const char *str, const struct qstr *name)
+{
+	struct qstr qstr = {.name = str, .len = len };
+
+	if (!IS_CASEFOLDED(dentry->d_parent->d_inode)) {
+		if (len != name->len)
+			return -1;
+		return memcmp(str, name, len);
+	}
+
+	return f2fs_ci_compare(dentry->d_parent->d_inode, name, &qstr);
+}
+
+static int f2fs_d_hash(const struct dentry *dentry, struct qstr *str)
+{
+	struct f2fs_sb_info *sbi = F2FS_SB(dentry->d_sb);
+	const struct unicode_map *um = sbi->s_encoding;
+	unsigned char *norm;
+	int len, ret = 0;
+
+	if (!IS_CASEFOLDED(dentry->d_inode))
+		return 0;
+
+	norm = f2fs_kmalloc(sbi, PATH_MAX, GFP_ATOMIC);
+	if (!norm)
+		return -ENOMEM;
+
+	len = utf8_casefold(um, str, norm, PATH_MAX);
+	if (len < 0) {
+		if (f2fs_has_strict_mode(sbi))
+			ret = -EINVAL;
+		goto out;
+	}
+	str->hash = full_name_hash(dentry, norm, len);
+out:
+	kvfree(norm);
+	return ret;
+}
+
+const struct dentry_operations f2fs_dentry_ops = {
+	.d_hash = f2fs_d_hash,
+	.d_compare = f2fs_d_compare,
+};
+#endif
+
diff --git a/fs/f2fs/f2fs.h b/fs/f2fs/f2fs.h
index c6c7904572d0d..31fd2a268ba14 100644
--- a/fs/f2fs/f2fs.h
+++ b/fs/f2fs/f2fs.h
@@ -2364,10 +2364,12 @@ static inline void f2fs_change_bit(unsigned int nr, char *addr)
 #define F2FS_INDEX_FL			0x00001000 /* hash-indexed directory */
 #define F2FS_DIRSYNC_FL			0x00010000 /* dirsync behaviour (directories only) */
 #define F2FS_PROJINHERIT_FL		0x20000000 /* Create with parents projid */
+#define F2FS_CASEFOLD_FL		0x40000000 /* Casefolded file */
 
 /* Flags that should be inherited by new inodes from their parent. */
 #define F2FS_FL_INHERITED (F2FS_SYNC_FL | F2FS_NODUMP_FL | F2FS_NOATIME_FL | \
-			   F2FS_DIRSYNC_FL | F2FS_PROJINHERIT_FL)
+			   F2FS_DIRSYNC_FL | F2FS_PROJINHERIT_FL | \
+			   F2FS_CASEFOLD_FL)
 
 /* Flags that are appropriate for regular files (all but dir-specific ones). */
 #define F2FS_REG_FLMASK		(~(F2FS_DIRSYNC_FL | F2FS_PROJINHERIT_FL))
@@ -2930,6 +2932,10 @@ int f2fs_update_extension_list(struct f2fs_sb_info *sbi, const char *name,
 							bool hot, bool set);
 struct dentry *f2fs_get_parent(struct dentry *child);
 
+extern int f2fs_ci_compare(const struct inode *parent,
+			   const struct qstr *name,
+			   const struct qstr *entry);
+
 /*
  * dir.c
  */
@@ -2993,8 +2999,8 @@ int f2fs_sanity_check_ckpt(struct f2fs_sb_info *sbi);
 /*
  * hash.c
  */
-f2fs_hash_t f2fs_dentry_hash(const struct qstr *name_info,
-				struct fscrypt_name *fname);
+f2fs_hash_t f2fs_dentry_hash(const struct inode *dir,
+		const struct qstr *name_info, struct fscrypt_name *fname);
 
 /*
  * node.c
@@ -3437,6 +3443,9 @@ static inline void f2fs_destroy_root_stats(void) { }
 #endif
 
 extern const struct file_operations f2fs_dir_operations;
+#ifdef CONFIG_UNICODE
+extern const struct dentry_operations f2fs_dentry_ops;
+#endif
 extern const struct file_operations f2fs_file_operations;
 extern const struct inode_operations f2fs_file_inode_operations;
 extern const struct address_space_operations f2fs_dblock_aops;
diff --git a/fs/f2fs/file.c b/fs/f2fs/file.c
index f8d46df8fa9ee..7ab2e6fea5b82 100644
--- a/fs/f2fs/file.c
+++ b/fs/f2fs/file.c
@@ -1660,7 +1660,16 @@ static int f2fs_setflags_common(struct inode *inode, u32 iflags, u32 mask)
 		return -EPERM;
 
 	oldflags = fi->i_flags;
+	if ((iflags ^ oldflags) & F2FS_CASEFOLD_FL) {
+		if (!f2fs_sb_has_casefold(F2FS_I_SB(inode)))
+			return -EOPNOTSUPP;
+
+		if (!S_ISDIR(inode->i_mode))
+			return -ENOTDIR;
 
+		if (!f2fs_empty_dir(inode))
+			return -ENOTEMPTY;
+	}
 	if ((iflags ^ oldflags) & (F2FS_APPEND_FL | F2FS_IMMUTABLE_FL))
 		if (!capable(CAP_LINUX_IMMUTABLE))
 			return -EPERM;
@@ -1699,6 +1708,7 @@ static const struct {
 	{ F2FS_INDEX_FL,	FS_INDEX_FL },
 	{ F2FS_DIRSYNC_FL,	FS_DIRSYNC_FL },
 	{ F2FS_PROJINHERIT_FL,	FS_PROJINHERIT_FL },
+	{ F2FS_CASEFOLD_FL,	FS_CASEFOLD_FL },
 };
 
 #define F2FS_GETTABLE_FS_FL (		\
@@ -1712,7 +1722,8 @@ static const struct {
 		FS_PROJINHERIT_FL |	\
 		FS_ENCRYPT_FL |		\
 		FS_INLINE_DATA_FL |	\
-		FS_NOCOW_FL)
+		FS_NOCOW_FL |		\
+		FS_CASEFOLD_FL)
 
 #define F2FS_SETTABLE_FS_FL (		\
 		FS_SYNC_FL |		\
@@ -1721,7 +1732,8 @@ static const struct {
 		FS_NODUMP_FL |		\
 		FS_NOATIME_FL |		\
 		FS_DIRSYNC_FL |		\
-		FS_PROJINHERIT_FL)
+		FS_PROJINHERIT_FL |	\
+		FS_CASEFOLD_FL)
 
 /* Convert f2fs on-disk i_flags to FS_IOC_{GET,SET}FLAGS flags */
 static inline u32 f2fs_iflags_to_fsflags(u32 iflags)
diff --git a/fs/f2fs/hash.c b/fs/f2fs/hash.c
index cc82f142f811f..99e79934f5088 100644
--- a/fs/f2fs/hash.c
+++ b/fs/f2fs/hash.c
@@ -14,6 +14,7 @@
 #include <linux/f2fs_fs.h>
 #include <linux/cryptohash.h>
 #include <linux/pagemap.h>
+#include <linux/unicode.h>
 
 #include "f2fs.h"
 
@@ -67,7 +68,7 @@ static void str2hashbuf(const unsigned char *msg, size_t len,
 		*buf++ = pad;
 }
 
-f2fs_hash_t f2fs_dentry_hash(const struct qstr *name_info,
+static f2fs_hash_t __f2fs_dentry_hash(const struct qstr *name_info,
 				struct fscrypt_name *fname)
 {
 	__u32 hash;
@@ -103,3 +104,35 @@ f2fs_hash_t f2fs_dentry_hash(const struct qstr *name_info,
 	f2fs_hash = cpu_to_le32(hash & ~F2FS_HASH_COL_BIT);
 	return f2fs_hash;
 }
+
+f2fs_hash_t f2fs_dentry_hash(const struct inode *dir,
+		const struct qstr *name_info, struct fscrypt_name *fname)
+{
+#ifdef CONFIG_UNICODE
+	struct f2fs_sb_info *sbi = F2FS_SB(dir->i_sb);
+	const struct unicode_map *um = sbi->s_encoding;
+	int r, dlen;
+	unsigned char *buff;
+	struct qstr *folded;
+
+	if (name_info->len && IS_CASEFOLDED(dir)) {
+		buff = f2fs_kzalloc(sbi, sizeof(char) * PATH_MAX, GFP_KERNEL);
+		if (!buff)
+			return -ENOMEM;
+
+		dlen = utf8_casefold(um, name_info, buff, PATH_MAX);
+		if (dlen < 0) {
+			kvfree(buff);
+			goto opaque_seq;
+		}
+		folded->name = buff;
+		folded->len = dlen;
+		r = __f2fs_dentry_hash(folded, fname);
+
+		kvfree(buff);
+		return r;
+	}
+opaque_seq:
+#endif
+	return __f2fs_dentry_hash(name_info, fname);
+}
diff --git a/fs/f2fs/inline.c b/fs/f2fs/inline.c
index 3613efca8c00c..354f71cf9e6ba 100644
--- a/fs/f2fs/inline.c
+++ b/fs/f2fs/inline.c
@@ -320,7 +320,7 @@ struct f2fs_dir_entry *f2fs_find_in_inline_dir(struct inode *dir,
 		return NULL;
 	}
 
-	namehash = f2fs_dentry_hash(&name, fname);
+	namehash = f2fs_dentry_hash(dir, &name, fname);
 
 	inline_dentry = inline_data_addr(dir, ipage);
 
@@ -580,7 +580,7 @@ int f2fs_add_inline_entry(struct inode *dir, const struct qstr *new_name,
 
 	f2fs_wait_on_page_writeback(ipage, NODE, true, true);
 
-	name_hash = f2fs_dentry_hash(new_name, NULL);
+	name_hash = f2fs_dentry_hash(dir, new_name, NULL);
 	f2fs_update_dentry(ino, mode, &d, new_name, name_hash, bit_pos);
 
 	set_page_dirty(ipage);
diff --git a/fs/f2fs/inode.c b/fs/f2fs/inode.c
index a33d7a849b2df..9a1f0d6616577 100644
--- a/fs/f2fs/inode.c
+++ b/fs/f2fs/inode.c
@@ -46,9 +46,11 @@ void f2fs_set_inode_flags(struct inode *inode)
 		new_fl |= S_DIRSYNC;
 	if (file_is_encrypt(inode))
 		new_fl |= S_ENCRYPTED;
+	if (flags & F2FS_CASEFOLD_FL)
+		new_fl |= S_CASEFOLD;
 	inode_set_flags(inode, new_fl,
 			S_SYNC|S_APPEND|S_IMMUTABLE|S_NOATIME|S_DIRSYNC|
-			S_ENCRYPTED);
+			S_ENCRYPTED|S_CASEFOLD);
 }
 
 static void __get_inode_rdev(struct inode *inode, struct f2fs_inode *ri)
diff --git a/fs/f2fs/namei.c b/fs/f2fs/namei.c
index c5b99042e6f2b..727de2f8620f2 100644
--- a/fs/f2fs/namei.c
+++ b/fs/f2fs/namei.c
@@ -489,6 +489,17 @@ static struct dentry *f2fs_lookup(struct inode *dir, struct dentry *dentry,
 		goto out_iput;
 	}
 out_splice:
+#ifdef CONFIG_UNICODE
+	if (!inode && IS_CASEFOLDED(dir)) {
+		/* Eventually we want to call d_add_ci(dentry, NULL)
+		 * for negative dentries in the encoding case as
+		 * well.  For now, prevent the negative dentry
+		 * from being cached.
+		 */
+		trace_f2fs_lookup_end(dir, dentry, ino, err);
+		return NULL;
+	}
+#endif
 	new = d_splice_alias(inode, dentry);
 	err = PTR_ERR_OR_ZERO(new);
 	trace_f2fs_lookup_end(dir, dentry, ino, err);
@@ -537,6 +548,16 @@ static int f2fs_unlink(struct inode *dir, struct dentry *dentry)
 		goto fail;
 	}
 	f2fs_delete_entry(de, page, dir, inode);
+#ifdef CONFIG_UNICODE
+	/* VFS negative dentries are incompatible with Encoding and
+	 * Case-insensitiveness. Eventually we'll want avoid
+	 * invalidating the dentries here, alongside with returning the
+	 * negative dentries at f2fs_lookup(), when it is  better
+	 * supported by the VFS for the CI case.
+	 */
+	if (IS_CASEFOLDED(dir))
+		d_invalidate(dentry);
+#endif
 	f2fs_unlock_op(sbi);
 
 	if (IS_DIRSYNC(dir))
diff --git a/fs/f2fs/super.c b/fs/f2fs/super.c
index 068db78a1e03e..20bf5fc3c5333 100644
--- a/fs/f2fs/super.c
+++ b/fs/f2fs/super.c
@@ -3108,6 +3108,7 @@ static int f2fs_setup_casefold(struct f2fs_sb_info *sbi)
 
 		sbi->s_encoding = encoding;
 		sbi->s_encoding_flags = encoding_flags;
+		sbi->sb->s_d_op = &f2fs_dentry_ops;
 	}
 #else
 	if (f2fs_sb_has_casefold(sbi)) {
-- 
2.22.0.657.g960e92d24f-goog

Re: [f2fs-dev] [PATCH v4 1/3] fs: Reserve flag for casefolding

[Chao Yu]

On 2019-7-24 7:05, Daniel Rosenberg via Linux-f2fs-devel wrote:
> In preparation for including the casefold feature within f2fs, elevate
> the EXT4_CASEFOLD_FL flag to FS_CASEFOLD_FL.
> 
> Signed-off-by: Daniel Rosenberg <drosen@google.com>

Reviewed-by: Chao Yu <yuchao0@huawei.com>

Thanks,

Re: [f2fs-dev] [PATCH v4 2/3] f2fs: include charset encoding information in the superblock

[Chao Yu]

On 2019-7-24 7:05, Daniel Rosenberg via Linux-f2fs-devel wrote:
> Add charset encoding to f2fs to support casefolding. It is modeled after
> the same feature introduced in commit c83ad55eaa91 ("ext4: include charset
> encoding information in the superblock")
> 
> Currently this is not compatible with encryption, similar to the current
> ext4 imlpementation. This will change in the future.
> 
> From the ext4 patch:
> """
> The s_encoding field stores a magic number indicating the encoding
> format and version used globally by file and directory names in the
> filesystem.  The s_encoding_flags defines policies for using the charset
> encoding, like how to handle invalid sequences.  The magic number is
> mapped to the exact charset table, but the mapping is specific to ext4.
> Since we don't have any commitment to support old encodings, the only
> encoding I am supporting right now is utf8-12.1.0.
> 
> The current implementation prevents the user from enabling encoding and
> per-directory encryption on the same filesystem at the same time.  The
> incompatibility between these features lies in how we do efficient
> directory searches when we cannot be sure the encryption of the user
> provided fname will match the actual hash stored in the disk without
> decrypting every directory entry, because of normalization cases.  My
> quickest solution is to simply block the concurrent use of these
> features for now, and enable it later, once we have a better solution.
> """
> 
> Signed-off-by: Daniel Rosenberg <drosen@google.com>

Reviewed-by: Chao Yu <yuchao0@huawei.com>

Thanks,

Re: [f2fs-dev] [PATCH v4 3/3] f2fs: Support case-insensitive file name lookups

[Chao Yu]

On 2019-7-24 7:05, Daniel Rosenberg via Linux-f2fs-devel wrote:
>  /* Flags that are appropriate for regular files (all but dir-specific ones). */
>  #define F2FS_REG_FLMASK		(~(F2FS_DIRSYNC_FL | F2FS_PROJINHERIT_FL))

We missed to add F2FS_CASEFOLD_FL here to exclude it in F2FS_REG_FLMASK.

> @@ -1660,7 +1660,16 @@ static int f2fs_setflags_common(struct inode *inode, u32 iflags, u32 mask)
>  		return -EPERM;
>  
>  	oldflags = fi->i_flags;
> +	if ((iflags ^ oldflags) & F2FS_CASEFOLD_FL) {
> +		if (!f2fs_sb_has_casefold(F2FS_I_SB(inode)))
> +			return -EOPNOTSUPP;
> +
> +		if (!S_ISDIR(inode->i_mode))
> +			return -ENOTDIR;
>  
> +		if (!f2fs_empty_dir(inode))
> +			return -ENOTEMPTY;
> +	}

I applied the patches based on last Jaegeuk's dev branch, it seems we needs to
adjust above code a bit. Otherwise it looks good to me.

BTW, it looks the patchset works fine with generic/556 testcase.

Thanks,

Re: [f2fs-dev] [PATCH v4 3/3] f2fs: Support case-insensitive file name lookups

[Jaegeuk Kim]

On 07/28, Chao Yu wrote:
> On 2019-7-24 7:05, Daniel Rosenberg via Linux-f2fs-devel wrote:
> >  /* Flags that are appropriate for regular files (all but dir-specific ones). */
> >  #define F2FS_REG_FLMASK		(~(F2FS_DIRSYNC_FL | F2FS_PROJINHERIT_FL))
> 
> We missed to add F2FS_CASEFOLD_FL here to exclude it in F2FS_REG_FLMASK.

Applied.

> 
> > @@ -1660,7 +1660,16 @@ static int f2fs_setflags_common(struct inode *inode, u32 iflags, u32 mask)
> >  		return -EPERM;
> >  
> >  	oldflags = fi->i_flags;
> > +	if ((iflags ^ oldflags) & F2FS_CASEFOLD_FL) {
> > +		if (!f2fs_sb_has_casefold(F2FS_I_SB(inode)))
> > +			return -EOPNOTSUPP;
> > +
> > +		if (!S_ISDIR(inode->i_mode))
> > +			return -ENOTDIR;
> >  
> > +		if (!f2fs_empty_dir(inode))
> > +			return -ENOTEMPTY;
> > +	}

Modified like this:
@@ -1665,6 +1665,13 @@ static int f2fs_setflags_common(struct inode *inode, u32 iflags, u32 mask)
        if (IS_NOQUOTA(inode))
                return -EPERM;

+       if ((iflags ^ fi->i_flags) & F2FS_CASEFOLD_FL) {
+               if (!f2fs_sb_has_casefold(F2FS_I_SB(inode)))
+                       return -EOPNOTSUPP;
+               if (!f2fs_empty_dir(inode))
+                       return -ENOTEMPTY;
+       }
+

Note that, directory is checked by above change.

I've uploaded in f2fs.git, so could you check it out and test a bit?

Thanks,

> 
> I applied the patches based on last Jaegeuk's dev branch, it seems we needs to
> adjust above code a bit. Otherwise it looks good to me.
> 
> BTW, it looks the patchset works fine with generic/556 testcase.
> 
> Thanks,

Re: [f2fs-dev] [PATCH v4 3/3] f2fs: Support case-insensitive file name lookups

[Chao Yu]

On 2019/7/29 14:27, Jaegeuk Kim wrote:
> On 07/28, Chao Yu wrote:
>> On 2019-7-24 7:05, Daniel Rosenberg via Linux-f2fs-devel wrote:
>>>  /* Flags that are appropriate for regular files (all but dir-specific ones). */
>>>  #define F2FS_REG_FLMASK		(~(F2FS_DIRSYNC_FL | F2FS_PROJINHERIT_FL))
>>
>> We missed to add F2FS_CASEFOLD_FL here to exclude it in F2FS_REG_FLMASK.
> 
> Applied.
> 
>>
>>> @@ -1660,7 +1660,16 @@ static int f2fs_setflags_common(struct inode *inode, u32 iflags, u32 mask)
>>>  		return -EPERM;
>>>  
>>>  	oldflags = fi->i_flags;
>>> +	if ((iflags ^ oldflags) & F2FS_CASEFOLD_FL) {
>>> +		if (!f2fs_sb_has_casefold(F2FS_I_SB(inode)))
>>> +			return -EOPNOTSUPP;
>>> +
>>> +		if (!S_ISDIR(inode->i_mode))
>>> +			return -ENOTDIR;
>>>  
>>> +		if (!f2fs_empty_dir(inode))
>>> +			return -ENOTEMPTY;
>>> +	}
> 
> Modified like this:
> @@ -1665,6 +1665,13 @@ static int f2fs_setflags_common(struct inode *inode, u32 iflags, u32 mask)
>         if (IS_NOQUOTA(inode))
>                 return -EPERM;
> 
> +       if ((iflags ^ fi->i_flags) & F2FS_CASEFOLD_FL) {
> +               if (!f2fs_sb_has_casefold(F2FS_I_SB(inode)))
> +                       return -EOPNOTSUPP;
> +               if (!f2fs_empty_dir(inode))
> +                       return -ENOTEMPTY;
> +       }
> +
> 
> Note that, directory is checked by above change.
> 
> I've uploaded in f2fs.git, so could you check it out and test a bit?

I've checked it out, it looks good to me now, and later I will test this new
version.

Reviewed-by: Chao Yu <yuchao0@huawei.com>

Thanks,

> 
> Thanks,
> 
>>
>> I applied the patches based on last Jaegeuk's dev branch, it seems we needs to
>> adjust above code a bit. Otherwise it looks good to me.
>>
>> BTW, it looks the patchset works fine with generic/556 testcase.
>>
>> Thanks,
> .
> 

Re: [f2fs-dev] [PATCH v4 3/3] f2fs: Support case-insensitive file name lookups

[Chao Yu]

On 2019-7-29 15:22, Chao Yu wrote:
> On 2019/7/29 14:27, Jaegeuk Kim wrote:
>> On 07/28, Chao Yu wrote:
>>> On 2019-7-24 7:05, Daniel Rosenberg via Linux-f2fs-devel wrote:
>>>>  /* Flags that are appropriate for regular files (all but dir-specific ones). */
>>>>  #define F2FS_REG_FLMASK		(~(F2FS_DIRSYNC_FL | F2FS_PROJINHERIT_FL))
>>>
>>> We missed to add F2FS_CASEFOLD_FL here to exclude it in F2FS_REG_FLMASK.
>>
>> Applied.
>>
>>>
>>>> @@ -1660,7 +1660,16 @@ static int f2fs_setflags_common(struct inode *inode, u32 iflags, u32 mask)
>>>>  		return -EPERM;
>>>>  
>>>>  	oldflags = fi->i_flags;
>>>> +	if ((iflags ^ oldflags) & F2FS_CASEFOLD_FL) {
>>>> +		if (!f2fs_sb_has_casefold(F2FS_I_SB(inode)))
>>>> +			return -EOPNOTSUPP;
>>>> +
>>>> +		if (!S_ISDIR(inode->i_mode))
>>>> +			return -ENOTDIR;
>>>>  
>>>> +		if (!f2fs_empty_dir(inode))
>>>> +			return -ENOTEMPTY;
>>>> +	}
>>
>> Modified like this:
>> @@ -1665,6 +1665,13 @@ static int f2fs_setflags_common(struct inode *inode, u32 iflags, u32 mask)
>>         if (IS_NOQUOTA(inode))
>>                 return -EPERM;
>>
>> +       if ((iflags ^ fi->i_flags) & F2FS_CASEFOLD_FL) {
>> +               if (!f2fs_sb_has_casefold(F2FS_I_SB(inode)))
>> +                       return -EOPNOTSUPP;
>> +               if (!f2fs_empty_dir(inode))
>> +                       return -ENOTEMPTY;
>> +       }
>> +
>>
>> Note that, directory is checked by above change.
>>
>> I've uploaded in f2fs.git, so could you check it out and test a bit?
> 
> I've checked it out, it looks good to me now, and later I will test this new
> version.
> 
> Reviewed-by: Chao Yu <yuchao0@huawei.com>

It can pass generic/556 as well.

Thanks,

> 
> Thanks,
> 
>>
>> Thanks,
>>
>>>
>>> I applied the patches based on last Jaegeuk's dev branch, it seems we needs to
>>> adjust above code a bit. Otherwise it looks good to me.
>>>
>>> BTW, it looks the patchset works fine with generic/556 testcase.
>>>
>>> Thanks,
>> .
>>

Re: [PATCH v4 3/3] f2fs: Support case-insensitive file name lookups

[Nathan Chancellor]

Hi all,

<snip>

> diff --git a/fs/f2fs/hash.c b/fs/f2fs/hash.c
> index cc82f142f811f..99e79934f5088 100644
> --- a/fs/f2fs/hash.c
> +++ b/fs/f2fs/hash.c
> @@ -14,6 +14,7 @@
>  #include <linux/f2fs_fs.h>
>  #include <linux/cryptohash.h>
>  #include <linux/pagemap.h>
> +#include <linux/unicode.h>
>  
>  #include "f2fs.h"
>  
> @@ -67,7 +68,7 @@ static void str2hashbuf(const unsigned char *msg, size_t len,
>  		*buf++ = pad;
>  }
>  
> -f2fs_hash_t f2fs_dentry_hash(const struct qstr *name_info,
> +static f2fs_hash_t __f2fs_dentry_hash(const struct qstr *name_info,
>  				struct fscrypt_name *fname)
>  {
>  	__u32 hash;
> @@ -103,3 +104,35 @@ f2fs_hash_t f2fs_dentry_hash(const struct qstr *name_info,
>  	f2fs_hash = cpu_to_le32(hash & ~F2FS_HASH_COL_BIT);
>  	return f2fs_hash;
>  }
> +
> +f2fs_hash_t f2fs_dentry_hash(const struct inode *dir,
> +		const struct qstr *name_info, struct fscrypt_name *fname)
> +{
> +#ifdef CONFIG_UNICODE
> +	struct f2fs_sb_info *sbi = F2FS_SB(dir->i_sb);
> +	const struct unicode_map *um = sbi->s_encoding;
> +	int r, dlen;
> +	unsigned char *buff;
> +	struct qstr *folded;
> +
> +	if (name_info->len && IS_CASEFOLDED(dir)) {
> +		buff = f2fs_kzalloc(sbi, sizeof(char) * PATH_MAX, GFP_KERNEL);
> +		if (!buff)
> +			return -ENOMEM;
> +
> +		dlen = utf8_casefold(um, name_info, buff, PATH_MAX);
> +		if (dlen < 0) {
> +			kvfree(buff);
> +			goto opaque_seq;
> +		}
> +		folded->name = buff;
> +		folded->len = dlen;
> +		r = __f2fs_dentry_hash(folded, fname);
> +
> +		kvfree(buff);
> +		return r;
> +	}
> +opaque_seq:
> +#endif
> +	return __f2fs_dentry_hash(name_info, fname);
> +}

Clang now warns:

fs/f2fs/hash.c:128:3: warning: variable 'folded' is uninitialized when used here [-Wuninitialized]
                folded->name = buff;
                ^~~~~~
fs/f2fs/hash.c:116:21: note: initialize the variable 'folded' to silence this warning
        struct qstr *folded;
                           ^
                            = NULL
1 warning generated.

I assume that it wants to be initialized with f2fs_kzalloc as well but
I am not familiar with this code and what it expects to do.

Please look into this when you get a chance!
Nathan

Re: [PATCH v4 3/3] f2fs: Support case-insensitive file name lookups

[Chao Yu]

Hi Nathan,

Thanks for the report! :)

On 2019/8/1 1:57, Nathan Chancellor wrote:
> Hi all,
> 
> <snip>
> 
>> diff --git a/fs/f2fs/hash.c b/fs/f2fs/hash.c
>> index cc82f142f811f..99e79934f5088 100644
>> --- a/fs/f2fs/hash.c
>> +++ b/fs/f2fs/hash.c
>> @@ -14,6 +14,7 @@
>>  #include <linux/f2fs_fs.h>
>>  #include <linux/cryptohash.h>
>>  #include <linux/pagemap.h>
>> +#include <linux/unicode.h>
>>  
>>  #include "f2fs.h"
>>  
>> @@ -67,7 +68,7 @@ static void str2hashbuf(const unsigned char *msg, size_t len,
>>  		*buf++ = pad;
>>  }
>>  
>> -f2fs_hash_t f2fs_dentry_hash(const struct qstr *name_info,
>> +static f2fs_hash_t __f2fs_dentry_hash(const struct qstr *name_info,
>>  				struct fscrypt_name *fname)
>>  {
>>  	__u32 hash;
>> @@ -103,3 +104,35 @@ f2fs_hash_t f2fs_dentry_hash(const struct qstr *name_info,
>>  	f2fs_hash = cpu_to_le32(hash & ~F2FS_HASH_COL_BIT);
>>  	return f2fs_hash;
>>  }
>> +
>> +f2fs_hash_t f2fs_dentry_hash(const struct inode *dir,
>> +		const struct qstr *name_info, struct fscrypt_name *fname)
>> +{
>> +#ifdef CONFIG_UNICODE
>> +	struct f2fs_sb_info *sbi = F2FS_SB(dir->i_sb);
>> +	const struct unicode_map *um = sbi->s_encoding;
>> +	int r, dlen;
>> +	unsigned char *buff;
>> +	struct qstr *folded;
>> +
>> +	if (name_info->len && IS_CASEFOLDED(dir)) {
>> +		buff = f2fs_kzalloc(sbi, sizeof(char) * PATH_MAX, GFP_KERNEL);
>> +		if (!buff)
>> +			return -ENOMEM;
>> +
>> +		dlen = utf8_casefold(um, name_info, buff, PATH_MAX);
>> +		if (dlen < 0) {
>> +			kvfree(buff);
>> +			goto opaque_seq;
>> +		}
>> +		folded->name = buff;
>> +		folded->len = dlen;
>> +		r = __f2fs_dentry_hash(folded, fname);
>> +
>> +		kvfree(buff);
>> +		return r;
>> +	}
>> +opaque_seq:
>> +#endif
>> +	return __f2fs_dentry_hash(name_info, fname);
>> +}
> 
> Clang now warns:
> 
> fs/f2fs/hash.c:128:3: warning: variable 'folded' is uninitialized when used here [-Wuninitialized]
>                 folded->name = buff;
>                 ^~~~~~
> fs/f2fs/hash.c:116:21: note: initialize the variable 'folded' to silence this warning
>         struct qstr *folded;
>                            ^
>                             = NULL
> 1 warning generated.
> 
> I assume that it wants to be initialized with f2fs_kzalloc as well but
> I am not familiar with this code and what it expects to do.
> 
> Please look into this when you get a chance!

That should be a bug, it needs to define a struct qstr type variable rather than
a pointer there.

Jaegeuk, could you fix this in you branch?

Thanks,

> Nathan
> .
> 

Re: [PATCH v4 3/3] f2fs: Support case-insensitive file name lookups

[Jaegeuk Kim]

On 08/01, Chao Yu wrote:
> Hi Nathan,
> 
> Thanks for the report! :)
> 
> On 2019/8/1 1:57, Nathan Chancellor wrote:
> > Hi all,
> > 
> > <snip>
> > 
> >> diff --git a/fs/f2fs/hash.c b/fs/f2fs/hash.c
> >> index cc82f142f811f..99e79934f5088 100644
> >> --- a/fs/f2fs/hash.c
> >> +++ b/fs/f2fs/hash.c
> >> @@ -14,6 +14,7 @@
> >>  #include <linux/f2fs_fs.h>
> >>  #include <linux/cryptohash.h>
> >>  #include <linux/pagemap.h>
> >> +#include <linux/unicode.h>
> >>  
> >>  #include "f2fs.h"
> >>  
> >> @@ -67,7 +68,7 @@ static void str2hashbuf(const unsigned char *msg, size_t len,
> >>  		*buf++ = pad;
> >>  }
> >>  
> >> -f2fs_hash_t f2fs_dentry_hash(const struct qstr *name_info,
> >> +static f2fs_hash_t __f2fs_dentry_hash(const struct qstr *name_info,
> >>  				struct fscrypt_name *fname)
> >>  {
> >>  	__u32 hash;
> >> @@ -103,3 +104,35 @@ f2fs_hash_t f2fs_dentry_hash(const struct qstr *name_info,
> >>  	f2fs_hash = cpu_to_le32(hash & ~F2FS_HASH_COL_BIT);
> >>  	return f2fs_hash;
> >>  }
> >> +
> >> +f2fs_hash_t f2fs_dentry_hash(const struct inode *dir,
> >> +		const struct qstr *name_info, struct fscrypt_name *fname)
> >> +{
> >> +#ifdef CONFIG_UNICODE
> >> +	struct f2fs_sb_info *sbi = F2FS_SB(dir->i_sb);
> >> +	const struct unicode_map *um = sbi->s_encoding;
> >> +	int r, dlen;
> >> +	unsigned char *buff;
> >> +	struct qstr *folded;
> >> +
> >> +	if (name_info->len && IS_CASEFOLDED(dir)) {
> >> +		buff = f2fs_kzalloc(sbi, sizeof(char) * PATH_MAX, GFP_KERNEL);
> >> +		if (!buff)
> >> +			return -ENOMEM;
> >> +
> >> +		dlen = utf8_casefold(um, name_info, buff, PATH_MAX);
> >> +		if (dlen < 0) {
> >> +			kvfree(buff);
> >> +			goto opaque_seq;
> >> +		}
> >> +		folded->name = buff;
> >> +		folded->len = dlen;
> >> +		r = __f2fs_dentry_hash(folded, fname);
> >> +
> >> +		kvfree(buff);
> >> +		return r;
> >> +	}
> >> +opaque_seq:
> >> +#endif
> >> +	return __f2fs_dentry_hash(name_info, fname);
> >> +}
> > 
> > Clang now warns:
> > 
> > fs/f2fs/hash.c:128:3: warning: variable 'folded' is uninitialized when used here [-Wuninitialized]
> >                 folded->name = buff;
> >                 ^~~~~~
> > fs/f2fs/hash.c:116:21: note: initialize the variable 'folded' to silence this warning
> >         struct qstr *folded;
> >                            ^
> >                             = NULL
> > 1 warning generated.
> > 
> > I assume that it wants to be initialized with f2fs_kzalloc as well but
> > I am not familiar with this code and what it expects to do.
> > 
> > Please look into this when you get a chance!
> 
> That should be a bug, it needs to define a struct qstr type variable rather than
> a pointer there.
> 
> Jaegeuk, could you fix this in you branch?

Yeah, let me apply this.

--- a/fs/f2fs/hash.c
+++ b/fs/f2fs/hash.c
@@ -113,25 +113,27 @@ f2fs_hash_t f2fs_dentry_hash(const struct inode *dir,
        const struct unicode_map *um = sbi->s_encoding;
        int r, dlen;
        unsigned char *buff;
-       struct qstr *folded;
+       struct qstr folded;

-       if (name_info->len && IS_CASEFOLDED(dir)) {
-               buff = f2fs_kzalloc(sbi, sizeof(char) * PATH_MAX, GFP_KERNEL);
-               if (!buff)
-                       return -ENOMEM;
+       if (!name_info->len || !IS_CASEFOLDED(dir))
+               goto opaque_seq;

-               dlen = utf8_casefold(um, name_info, buff, PATH_MAX);
-               if (dlen < 0) {
-                       kvfree(buff);
-                       goto opaque_seq;
-               }
-               folded->name = buff;
-               folded->len = dlen;
-               r = __f2fs_dentry_hash(folded, fname);
+       buff = f2fs_kzalloc(sbi, sizeof(char) * PATH_MAX, GFP_KERNEL);
+       if (!buff)
+               return -ENOMEM;

+       dlen = utf8_casefold(um, name_info, buff, PATH_MAX);
+       if (dlen < 0) {
                kvfree(buff);
-               return r;
+               goto opaque_seq;
        }
+       folded.name = buff;
+       folded.len = dlen;
+       r = __f2fs_dentry_hash(&folded, fname);
+
+       kvfree(buff);
+       return r;
+
 opaque_seq:
 #endif
        return __f2fs_dentry_hash(name_info, fname);

Re: [f2fs-dev] [PATCH v4 3/3] f2fs: Support case-insensitive file name lookups

[Chao Yu]

On 2019-7-24 7:05, Daniel Rosenberg via Linux-f2fs-devel wrote:
> +static int f2fs_d_compare(const struct dentry *dentry, unsigned int len,
> + const char *str, const struct qstr *name)
> +{
> +	struct qstr qstr = {.name = str, .len = len };
> +
> +	if (!IS_CASEFOLDED(dentry->d_parent->d_inode)) {
> +		if (len != name->len)
> +			return -1;
> +		return memcmp(str, name, len);

66883da1eee8 ("ext4: fix dcache lookup of !casefolded directories")

memcmp(str, name->name, len);