Re: [PATCH] erofs: add error log in erofs_fc_parse_param

[Gao Xiang]

On 2025/1/18 09:25, Ian Kent wrote:
> On 18/1/25 08:45, Ian Kent wrote:
> > On 17/1/25 16:52, Chen Linxuan wrote:
> > > While reading erofs code, I notice that `erofs_fc_parse_param` will
> > > return -ENOPARAM, which means that erofs do not support this option,
> > > without report anything when `fs_parse` return an unknown `opt`.
> > >
> > > But if an option is unknown to erofs, I mean that option not in
> > > `erofs_fs_parameters` at all, `fs_parse` will return -ENOPARAM,
> > > which means that `erofs_fs_parameters` should has returned earlier.
> >
> > I'm pretty sure than the vfs deals with reporting unknown options
> >
> > and returns -EINVAL already.
> >
> >
> > I think the caller oferofs_fc_parse_param() is vfs_parse_fs_param()
> >
> > and for an -ENOPARAM return will ultimately do this:
> >
> > return invalf(fc, "%s: Unknown parameter '%s'", fc->fs_type->name, param->key);
> >
> > which does this.
> >
> > The thing about this is the mount API macro deals with (or it should,
> >
> > although I'm not sure that's completely sorted out yet) logging the
> >
> > message to the console log as well as possibly making it available to
> >
> > mount api system calls. I'm pretty sure this change will prevent the
> >
> > error message being available for mount api system calls to retrieve.
>
> Actually no, removing the default switch branch does look like the right thing
>
> to do, my mistake. The call to fs_parse() will return -NOPARAM so the
>
> switch doesn't need to handle that case.
>
>
> Oops!

Yeah, actually dead code..

Thanks,
Gao Xiang

> > Ian
> >
> > >