Received-SPF: pass (google.com: domain of linux-ext4+bounces-1589-linux.lists.archive=gmail.com@vger.kernel.org designates 139.178.88.99 as permitted sender) client-ip=139.178.88.99;
From: Luis Henriques <lhenriques@suse.de>
To: Jan Kara <jack@suse.cz>
Cc: Christian Brauner <brauner@kernel.org>,  Theodore Ts'o <tytso@mit.edu>,
  Andreas Dilger <adilger.kernel@dilger.ca>,  Alexander Viro
 <viro@zeniv.linux.org.uk>,  Miklos Szeredi <miklos@szeredi.hu>,  Amir
 Goldstein <amir73il@gmail.com>,  linux-ext4@vger.kernel.org,
  linux-fsdevel@vger.kernel.org,  linux-unionfs@vger.kernel.org,
  linux-kernel@vger.kernel.org
Subject: Re: [PATCH 1/3] fs_parser: handle parameters that can be empty and
 don't have a value
In-Reply-To: <20240308230911.r5a4xn6f5vp24hil@quack3> (Jan Kara's message of
	"Sat, 9 Mar 2024 00:09:11 +0100")
References: <20240229163011.16248-1-lhenriques@suse.de>
	<20240229163011.16248-2-lhenriques@suse.de>
	<20240301-gegossen-seestern-683681ea75d1@brauner>
	<87il269crs.fsf@suse.de> <20240307151356.ishrtxrsge2i5mjn@quack3>
	<20240308-fahrdienst-torten-eae8f3eed3b4@brauner>
	<87a5n9t4le.fsf@suse.de> <20240308230911.r5a4xn6f5vp24hil@quack3>
Date: Mon, 11 Mar 2024 10:26:05 +0000
Message-ID: <87r0gh6p4y.fsf@suse.de>
Precedence: bulk
MIME-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: quoted-printable

Jan Kara <jack@suse.cz> writes:

> On Fri 08-03-24 10:12:13, Luis Henriques wrote:
>> Christian Brauner <brauner@kernel.org> writes:
>>=20
>> > On Thu, Mar 07, 2024 at 04:13:56PM +0100, Jan Kara wrote:
>> >> On Fri 01-03-24 15:45:27, Luis Henriques wrote:
>> >> > Christian Brauner <brauner@kernel.org> writes:
>> >> >=20
>> >> > > On Thu, Feb 29, 2024 at 04:30:08PM +0000, Luis Henriques wrote:
>> >> > >> Currently, only parameters that have the fs_parameter_spec 'type=
' set to
>> >> > >> NULL are handled as 'flag' types.  However, parameters that have=
 the
>> >> > >> 'fs_param_can_be_empty' flag set and their value is NULL should =
also be
>> >> > >> handled as 'flag' type, as their type is set to 'fs_value_is_fla=
g'.
>> >> > >>=20
>> >> > >> Signed-off-by: Luis Henriques <lhenriques@suse.de>
>> >> > >> ---
>> >> > >>  fs/fs_parser.c | 3 ++-
>> >> > >>  1 file changed, 2 insertions(+), 1 deletion(-)
>> >> > >>=20
>> >> > >> diff --git a/fs/fs_parser.c b/fs/fs_parser.c
>> >> > >> index edb3712dcfa5..53f6cb98a3e0 100644
>> >> > >> --- a/fs/fs_parser.c
>> >> > >> +++ b/fs/fs_parser.c
>> >> > >> @@ -119,7 +119,8 @@ int __fs_parse(struct p_log *log,
>> >> > >>  	/* Try to turn the type we were given into the type desired by=
 the
>> >> > >>  	 * parameter and give an error if we can't.
>> >> > >>  	 */
>> >> > >> -	if (is_flag(p)) {
>> >> > >> +	if (is_flag(p) ||
>> >> > >> +	    (!param->string && (p->flags & fs_param_can_be_empty))) {
>> >> > >>  		if (param->type !=3D fs_value_is_flag)
>> >> > >>  			return inval_plog(log, "Unexpected value for '%s'",
>> >> > >>  				      param->key);
>> >> > >
>> >> > > If the parameter was derived from FSCONFIG_SET_STRING in fsconfig=
() then
>> >> > > param->string is guaranteed to not be NULL. So really this is only
>> >> > > about:
>> >> > >
>> >> > > FSCONFIG_SET_FD
>> >> > > FSCONFIG_SET_BINARY
>> >> > > FSCONFIG_SET_PATH
>> >> > > FSCONFIG_SET_PATH_EMPTY
>> >> > >
>> >> > > and those values being used without a value. What filesystem does=
 this?
>> >> > > I don't see any.
>> >> > >
>> >> > > The tempting thing to do here is to to just remove fs_param_can_b=
e_empty
>> >> > > from every helper that isn't fs_param_is_string() until we actual=
ly have
>> >> > > a filesystem that wants to use any of the above as flags. Will lo=
se a
>> >> > > lot of code that isn't currently used.
>> >> >=20
>> >> > Right, I find it quite confusing and I may be fixing the issue in t=
he
>> >> > wrong place.  What I'm seeing with ext4 when I mount a filesystem u=
sing
>> >> > the option '-o usrjquota' is that fs_parse() will get:
>> >> >=20
>> >> >  * p->type is set to fs_param_is_string
>> >> >    ('p' is a struct fs_parameter_spec, ->type is a function)
>> >> >  * param->type is set to fs_value_is_flag
>> >> >    ('param' is a struct fs_parameter, ->type is an enum)
>> >> >=20
>> >> > This is because ext4 will use the __fsparam macro to set define a
>> >> > fs_param_spec as a fs_param_is_string but will also set the
>> >> > fs_param_can_be_empty; and the fsconfig() syscall will get that par=
ameter
>> >> > as a flag.  That's why param->string will be NULL in this case.
>> >>=20
>> >> So I'm a bit confused here. Valid variants of these quota options are=
 like
>> >> "usrjquota=3D<filename>" (to set quota file name) or "usrjquota=3D" (=
to clear
>> >> quota file name). The variant "usrjquota" should ideally be rejected
>> >> because it doesn't make a good sense and only adds to confusion. Now =
as far
>> >> as I'm reading fs/ext4/super.c: parse_options() (and as far as my tes=
ting
>> >> shows) this is what is happening so what is exactly the problem you're
>> >> trying to fix?
>> >
>> > mount(8) has no way of easily knowing that for something like
>> > mount -o usrjquota /dev/sda1 /mnt that "usrjquota" is supposed to be
>> > set as an empty string via FSCONFIG_SET_STRING. For mount(8) it is
>> > indistinguishable from a flag because it's specified without an
>> > argument. So mount(8) passes FSCONFIG_SET_FLAG and it seems strange th=
at
>> > we should require mount(8) to know what mount options are strings or n=
o.
>> > I've ran into this issue before myself when using the mount api
>> > programatically.
>>=20
>> Right.  A simple usecase is to try to do:
>>=20
>>   mount -t ext4 -o usrjquota=3D /dev/sda1 /mnt/
>>=20
>> It will fail, and this has been broken for a while.
>
> I see. But you have to have new enough mount that is using fsconfig, don't
> you? Because for me in my test VM this works just fine...

Oh, interesting.  FTR I'm using mount from util-linux 2.39.3, but I
haven't tried this with older versions.

Cheers,
--=20
Lu=C3=ADs


>
> But anyway, I get the point. Thanks for educating me :)
>
> 								Honza
> --=20
> Jan Kara <jack@suse.com>
> SUSE Labs, CR