Re: [PATCH] erofs: move erofs out of staging

From: Gao Xiang
Date: Sun Aug 18 2019 - 05:29:18 EST


On Sun, Aug 18, 2019 at 11:03:53AM +0200, Richard Weinberger wrote:
> ----- Urspr??ngliche Mail -----
> > I agree with you, but what can we do now is trying our best to fuzz
> > all the fields.
> >
> > So, what is your opinion about EROFS?
>
> All I'm saying is that you should not blindly trust the disk.

I completely agree with you, and I'm teaching EROFS to
make the little naughty boy more strong... (we already
have many error handling code, but I think I will teach him
more, yes.)

>
> Another thing that raises my attention is in superblock_read():
> memcpy(sbi->volume_name, layout->volume_name,
> sizeof(layout->volume_name));
>
> Where do you check whether ->volume_name has a NUL terminator?
> Currently this field has no user, maybe will add a check upon usage.
> But this kind of things makes me wonder.

Yes, I think this is a good point. :)
Since volume_name is not used currently, I will fix it when
it use late.

I will make a note here (or more straightforward, I will fix it
to avoid potential bug now.)

Thanks,
Gao Xiang

>
> Thanks,
> //richard