nerd2 / gexto Goto Github PK

I think the implemented extent traversal logic is incorrect.
This is not documented anywhere but ext4_extent_idx structs after the ext4_extent_header are sorted by the ei_block.

The current logic picks the ext4_extent_idx which satisfies ext4_extent_idx.ei_block < file_block_no but that will always choose the first ext4_extent_idx struct. This will lead to incorrect results for larger files.

The linux kernel implementation does a binary search: look at ext4_ext_binsearch_idx.

I'm using https://github.com/gokrazy/gokrazy to create bootable application images. It needs to add an (ext4) partition for user data to the image. Can I create that ext filesystem and write it to the image? Any hint would be appreciated.

i don't want to use it in production data, but only for newly created vm images. But do you have some expensive tests for all functions to guarantee data safety?

Hi,

Thanks for this great library!
How easy/hard would it be to start supporting 256 inode size instead of only 128?
That'd be awesome to have

Cheers!

Hey!
Great project! It is a really interesting implementation of ext2/3/4. I was planning on integrating your project into something I am working on. It would be great if you could a license so I could use it.

Usually the MIT or Apache 2.0 are good enough.

You should be using sb.s_desc_size if sb.s_rev_level == 1 and 32 if ifsb.s_rev_level == 0` over here because in ext4 the block group descriptor could be larger.