Paragon Submits Third Version Of New NTFS Kernel Driver For Linux

erek

[H]F Junkie
Joined
Dec 19, 2005
Messages
10,875
"This is bollocks. First and foremost, d_splice_alias() *does* iput() on
failure, so you've got double-put there. What's more
* d_splice_alias(ERR_PTR(err), dentry) return err
* d_splice_alias(NULL, dentry) is equivalent to d_add(dentry, NULL) and returns NULL

IOW, all that boilerplate could be replaced with one line:

return d_splice_alias(dir_search(dir, &dentry->d_name, fnd), dentry);"


"Two weeks ago file-system driver vendor Paragon Software posted patches for their previously commercial NTFS Linux driver with hopes of getting the code mainlined. That initial patch drew some criticism for how it was handled but a week later a new version was published that split up the patches nicely and had other feedback. Prior to calling it a weekend, Paragon sent out a third version of the "NTFS3" Linux kernel driver patches..."

https://www.phoronix.com/forums/for...d-version-of-new-ntfs-kernel-driver-for-linux
 
I’m happy to see somebody doing the work, and I am happy they are being held to a high standard. While not “the best” file system format out there NTFS does have a lot of strengths and having write capabilities caked into the system would make my personal life easier. But it’s not something I would want half asses that’s for sure.
 
Yeah, it's a somewhat niche use-case, but there isn't any widespread availability for *nix FSes on Windows systems; and NTFS is a much more capable option than exFAT for a shared non-network drive (dual boot or USB).

It's giant single vendor code drop, so code review the crap out of it; but it get it approved once it's ready.
 
Yeah, it's a somewhat niche use-case, but there isn't any widespread availability for *nix FSes on Windows systems; and NTFS is a much more capable option than exFAT for a shared non-network drive (dual boot or USB).

It's giant single vendor code drop, so code review the crap out of it; but it get it approved once it's ready.
Given the nature of the update and the Linux community ad a whole the lack of vulgarity and adjectives in their comments I take as a sign the code is almost in a state where it can be approved.

I was going to say I was using the Seagate drivers currently and that they did a decent job and haven’t had any issues with them but while writing this post I went to verify that I was actually using them and not mis-remembering and upon inspection they are from Seagate but they are actually developed by Paragon so..... yay?
 
I've just been using the repo-available NTFS support when needed; if we're going to get real support, that would be nice for larger external volumes.

[of course, I'd rather just have transparent EXT4 / XFS / ZFS / BTRFS support in Windows, but baby steps where can be]
 
Arrgggghhhh ntfs, let it die.

If I wanted horrific small file performance I’d connect my hard disk with a piece of string

(I do love a good rant in a code / commit comments though)
 
Back
Top