# Re: add file system helpers that take kernel pointers for the init code v4

- From: Christoph Hellwig \<hidden\>
- Date: 2020-07-31 06:33:23 
- List: linux-raid
- Page: https://ratatoskr.run/linux-raid/2020/07/185699
- Thread: https://ratatoskr.run/linux-raid/2020/07/185462/t.md

```text
On Fri, Jul 31, 2020 at 03:14:24AM +0100, Al Viro wrote:
> Christoph Hellwig (28):
> [snip]
>       initramfs: switch initramfs unpacking to struct file based APIs
>       initramfs: switch initramfs unpacking to struct file based APIs
> [snip]
> 
> It's not a bisect hazard, of course, but if you don't fold those
> together, you might at least want to give the second one a different
> commit summary...  I hadn't been able to find an analogue of #init_path on
> top of that either.
> 
> As it is, #init-user-pointers is fine (aside of that SNAFU with unfolded
> pair of commits), and so's the contents of #init_path part following what
> used to be #init-user-pointers, but it'll be an awful mess on merge in
> the current shape.
> 
> I can sort it out myself, if you don't mind that; again, I'm OK with
> the contents and I've no problem with doing reordering/folding.

I've fixed the folding issues in init-user-pointers and rebased init_path
on top of that.  Feel free to pull it.  I don't think it is worth
reposting.
```