ntfs-3g speed

[daniel_I_l]

If I have a lot of source code saved on a ntfs (windows) partition and want to compile it, would it be slower to use ntfs-3g to mount the partition and compile directly from that partition than it would if I where compiling directly from the linux partition?
How much overhead does the ntfs-3g cause if any?
Thanks.

[Hlingler]

We've been through this discussion before. Please search the forum. Answer here: http://www.ntfs-3g.org/support.html#slow, see also: http://www.ntfs-3g.org/support.html#cpu100

V

[daniel_I_l]

Thanks. for the links.

[szaka]

I compile a lot on NTFS. Actually the compilation of the kernel and NTFS-3G (and soon gcc) is part of the NTFS-3G quality assurance process: http://ntfs-3g.org/quality.html

I didn't notice much difference in performance compared to other file systems during compilations. The typical bottleneck is disk seeks and none of the current stable file system is optimized well for this, yet. This will significantly change in the future (btrfs, ext4, ntfs-3g new releases, etc).

[forkbomb]

Out of curiosity, do you ever run into issues with permissions being applied properly?

Though I guess you wouldn't. Chmoding and Chowning doesn't take place until make, anyway.

[szaka]

Never but the enhanced NTFS-3G driver fully supports permissions and ownerships: http://pagesperso-orange.fr/b.andre/security.html

Actually the NTFS-3G project maintains the POSIX File System Test Suite, what our driver fully passes without any fault: http://ntfs-3g.org/pjd-fstest.html

[forkbomb]

Oh, yeahhhh...

I knew that. Just forgot about it.