unicode: Add utf8-data module

utf8data.h contains a large database table which is an auto-generated
decodification trie for the unicode normalization functions.

Allow building it into a separate module.

Based on a patch from Shreeya Patel <shreeya.patel@collabora.com>.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Gabriel Krisman Bertazi <krisman@collabora.com>

unicode: cache the normalization tables in struct unicode_map

Instead of repeatedly looking up the version add pointers to the
NFD and NFD+CF tables to struct unicode_map, and pass a
unicode_map plus index to the functions using the normalization
tables.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Gabriel Krisman Bertazi <krisman@collabora.com>

unicode: pass a UNICODE_AGE() tripple to utf8_load

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>
Signed-off-by: Gabriel Krisman Bertazi <krisman@collabora.com>

unicode: mark the version field in struct unicode_map unsigned

unicode version tripplets are always unsigned.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Gabriel Krisman Bertazi <krisman@collabora.com>
Signed-off-by: Gabriel Krisman Bertazi <krisman@collabora.com>

unicode: remove the charset field from struct unicode_map

It is hardcoded and only used for a f2fs sysfs file where it can be
hardcoded just as easily.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Gabriel Krisman Bertazi <krisman@collabora.com>
Signed-off-by: Gabriel Krisman Bertazi <krisman@collabora.com>

unicode: Add utf8_casefold_hash

This adds a case insensitive hash function to allow taking the hash
without needing to allocate a casefolded copy of the string.

The existing d_hash implementations for casefolding allocate memory
within rcu-walk, by avoiding it we can be more efficient and avoid
worrying about a failed allocation.

Signed-off-by: Daniel Rosenberg <drosen@google.com>
Reviewed-by: Gabriel Krisman Bertazi <krisman@collabora.com>
Reviewed-by: Eric Biggers <ebiggers@google.com>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>

ext4: optimize case-insensitive lookups

Temporarily cache a casefolded version of the file name under lookup in
ext4_filename, to avoid repeatedly casefolding it. I got up to 30%
speedup on lookups of large directories (>100k entries), depending on
the length of the string under lookup.

Signed-off-by: Gabriel Krisman Bertazi <krisman@collabora.com>
Signed-off-by: Theodore Ts'o <tytso@mit.edu>

unicode: implement higher level API for string handling

This patch integrates the utf8n patches with some higher level API to
perform UTF-8 string comparison, normalization and casefolding
operations. Implemented is a variation of NFD, and casefold is
performed by doing full casefold on top of NFD. These algorithms are
based on the core implemented by Olaf Weber from SGI.

Signed-off-by: Gabriel Krisman Bertazi <krisman@collabora.co.uk>
Signed-off-by: Theodore Ts'o <tytso@mit.edu>