Timeline or roadmap to split lowmem mode and original mode check?

Timeline or roadmap to split lowmem mode and original mode check?

[Qu Wenruo]

[-- Attachment #1.1: Type: text/plain, Size: 762 bytes --]

Hi David,

There is the long planned work to split the original mode and lowmem
mode check into their own .[ch] files.

I found it harder and harder to locate repair functions for original and
lowmem mode when writing fixes for them.

I wonder if it's a good time to start the split work, and if you're OK
with it, my plan is to:

1) Start the split work based on 4.14.1

2) Split check code into:
   check/common.[ch]	<<< Not sure how many functions will be there
   check/lowmem.[ch]
   check/origin.[ch]
   check/main.c

Currently most fsck fixes and enhancement are from Fujitsu guys who I'm
quite familiar with, so rebasing their code won't be a big problem for me.

If this works for you, I can start the work asap.

Thanks,
Qu


[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 520 bytes --]

Re: Timeline or roadmap to split lowmem mode and original mode check?

[Nikolay Borisov]

On 17.01.2018 07:32, Qu Wenruo wrote:
> Hi David,
> 
> There is the long planned work to split the original mode and lowmem
> mode check into their own .[ch] files.
> 
> I found it harder and harder to locate repair functions for original and
> lowmem mode when writing fixes for them.
> 
> I wonder if it's a good time to start the split work, and if you're OK
> with it, my plan is to:
> 
> 1) Start the split work based on 4.14.1
> 
> 2) Split check code into:
>    check/common.[ch]	<<< Not sure how many functions will be there
>    check/lowmem.[ch]
>    check/origin.[ch]
>    check/main.c
> 
> Currently most fsck fixes and enhancement are from Fujitsu guys who I'm
> quite familiar with, so rebasing their code won't be a big problem for me.
> 
> If this works for you, I can start the work asap.

When you start working on that you might consider also documenting most
of the repair functions :)

> 
> Thanks,
> Qu
>