I've been playing with NTFS-3G on an ARM system and have been looking at why performance is so low compared with other file systems (I get 2.1MB/s on write; FAT32, EXT3 and HFS+ all do about 17MB/s).
One of the things that struck me was that in ntfs_fuse_read and ntfs_fuse_write most of the time is spent opening and closing the internal inode.. Moving the ntfs_inode_open to ntfs_fuse_open and ntfs_inode_close to a newly created ntfs_fuse_release, my performance jumps to 6.1MB/s - a bit better. For the most part this seems to work (Iozone seems to have some issues that go away if I use debug mode or use this version of NTFS-3G on an Intel platform). I have noticed that ntfs_inode_open seems to create a new inode - might this cause problems if multiple open of the file were performed? Is there a reason for the approach that has been taken? -- Sven ------------------------------------------------------------------------- This SF.Net email is sponsored by the Moblin Your Move Developer's challenge Build the coolest Linux based applications with Moblin SDK & win great prizes Grand prize is a trip for two to an Open Source event anywhere in the world http://moblin-contest.org/redirect.php?banner_id=100&url=/ _______________________________________________ ntfs-3g-devel mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/ntfs-3g-devel
