Re: [PATCH 05/11] unicode: pass a UNICODE_AGE() tripple to utf8_load

Gabriel Krisman Bertazi [off-list ref] writes:

Christoph Hellwig [off-list ref] writes:

quoted

Don't bother with pointless string parsing when the caller can just pass
the version in the format that the core expects.  Also remove the
fallback to the latest version that none of the callers actually uses.

Signed-off-by: Christoph Hellwig <hch@lst.de>
---
 fs/ext4/super.c            | 10 ++++----
 fs/f2fs/super.c            | 10 ++++----
 fs/unicode/utf8-core.c     | 50 ++++----------------------------------
 fs/unicode/utf8-norm.c     | 11 ++-------
 fs/unicode/utf8-selftest.c | 15 ++++++------
 fs/unicode/utf8n.h         | 14 ++---------
 include/linux/unicode.h    | 11 ++++++++-
 7 files changed, 37 insertions(+), 84 deletions(-)

diff --git a/fs/ext4/super.c b/fs/ext4/super.c
index a68be582bba5..be418a30b52e 100644
--- a/fs/ext4/super.c
+++ b/fs/ext4/super.c

@@ -2016,9 +2016,9 @@ static const struct mount_opts {
 static const struct ext4_sb_encodings {
 	__u16 magic;
 	char *name;
-	char *version;
+	unsigned int version;
 } ext4_sb_encoding_map[] = {
-	{EXT4_ENC_UTF8_12_1, "utf8", "12.1.0"},
+	{EXT4_ENC_UTF8_12_1, "utf8", UNICODE_AGE(12, 1, 0)},
 };
 
 static const struct ext4_sb_encodings *

@@ -4308,15 +4308,15 @@ static int ext4_fill_super(struct super_block *sb, void *data, int silent)
 		encoding = utf8_load(encoding_info->version);
 		if (IS_ERR(encoding)) {
 			ext4_msg(sb, KERN_ERR,
-				 "can't mount with superblock charset: %s-%s "
+				 "can't mount with superblock charset: %s-0x%x "
 				 "not supported by the kernel. flags: 0x%x.",
 				 encoding_info->name, encoding_info->version,
 				 encoding_flags);
 			goto failed_mount;

"Using encoding defined by superblock: utf8-0xc0100 with flags 0x0"

This is much less readable than what we previously had:

"Using encoding defined by superblock: utf8-12.1.0 with flags 0x0"

It is minor, but can we do instead:

ext4_msg("... %u.%u.%u\n", (encoding_info->version>>12) & 0xff,
	 (encoding_info->version>>8) & 0xff), encoding_info->version & 0xff))

The rest of the series looks good and I can pick it up for 5.15, unless
someone has anything else to say?  It has lived on the list for a while
now.

Ugh, pressed reply too quickly.  Sorry for the multiple email reply.

In the summary line: tripple -> triple.

-- 
Gabriel Krisman Bertazi