Received-SPF: pass (google.com: domain of linux-kernel+bounces-40126-linux.lists.archive=gmail.com@vger.kernel.org designates 147.75.48.161 as permitted sender) client-ip=147.75.48.161;
Precedence: bulk
MIME-Version: 1.0
References: <20231125163559.824210-1-masahiroy@kernel.org> <CAMuHMdWm6u1wX7efZQf=2XUAHascps76YQac6rdnQGhc8nop_Q@mail.gmail.com>
 <d21298d9-fed6-4e08-9780-dbcb388b9ccc@smile.fr> <CAK7LNASaG4DpHTb3YHMd8d8DJ5H3z0aiUcSqX+=7CZb99kRU8A@mail.gmail.com>
 <b65a68eb-6b96-41ff-bbb9-38cb2dee940e@smile.fr> <CAK7LNARVbjVkP=v7uQDB=Z+Ntcy9MiFa6WowTX9mA47YjS3zTg@mail.gmail.com>
 <928267c8-2c79-4f76-aa6e-26dc63c77e43@smile.fr>
In-Reply-To: <928267c8-2c79-4f76-aa6e-26dc63c77e43@smile.fr>
From: Masahiro Yamada <masahiroy@kernel.org>
Date: Fri, 26 Jan 2024 22:20:38 +0900
Message-ID: <CAK7LNAQu7WmgbGkMv7WUAv0WGW+6DTPAtcW+TA5XAF9cq0hYOg@mail.gmail.com>
Subject: Re: [PATCH 1/2] kconfig: remove unneeded symbol_empty variable
To: Yoann Congal <yoann.congal@smile.fr>
Cc: Geert Uytterhoeven <geert@linux-m68k.org>, linux-kbuild@vger.kernel.org, 
	linux-kernel@vger.kernel.org, 
	Linux-Renesas <linux-renesas-soc@vger.kernel.org>, 
	Vegard Nossum <vegard.nossum@oracle.com>
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

On Thu, Jan 25, 2024 at 11:42=E2=80=AFPM Yoann Congal <yoann.congal@smile.f=
r> wrote:
>
> Hi,
>
> Le 24/01/2024 =C3=A0 21:12, Masahiro Yamada a =C3=A9crit :
> > On Wed, Jan 24, 2024 at 5:56=E2=80=AFPM Yoann Congal <yoann.congal@smil=
e.fr> wrote:
> >> Le 24/01/2024 =C3=A0 09:09, Masahiro Yamada a =C3=A9crit :
> >>> On Wed, Jan 24, 2024 at 12:11=E2=80=AFAM Yoann Congal <yoann.congal@s=
mile.fr> wrote:
> >>>> For what it is worth, CONFIG_BASE_SMALL is defined as an int but is =
only used as a bool :
> >>>>    $ git grep BASE_SMALL
> >>>>   arch/x86/include/asm/mpspec.h:#if CONFIG_BASE_SMALL =3D=3D 0
> >>>>   drivers/tty/vt/vc_screen.c:#define CON_BUF_SIZE (CONFIG_BASE_SMALL=
 ? 256 : PAGE_SIZE)
> >>>>   include/linux/threads.h:#define PID_MAX_DEFAULT (CONFIG_BASE_SMALL=
 ? 0x1000 : 0x8000)
> >>>>   include/linux/threads.h:#define PID_MAX_LIMIT (CONFIG_BASE_SMALL ?=
 PAGE_SIZE * 8 : \
> >>>>   include/linux/udp.h:#define UDP_HTABLE_SIZE_MIN         (CONFIG_BA=
SE_SMALL ? 128 : 256)
> >>>>   include/linux/xarray.h:#define XA_CHUNK_SHIFT           (CONFIG_BA=
SE_SMALL ? 4 : 6)
> >>>>   init/Kconfig:   default 12 if !BASE_SMALL
> >>>>   init/Kconfig:   default 0 if BASE_SMALL
> >>>>   init/Kconfig:config BASE_SMALL
> >>>>   kernel/futex/core.c:#if CONFIG_BASE_SMALL
> >>>>   kernel/user.c:#define UIDHASH_BITS      (CONFIG_BASE_SMALL ? 3 : 7=
)
> >>>>
> >>>> Maybe we should change CONFIG_BASE_SMALL to the bool type?
> >>
> >> My first test shows that switching CONFIG_BASE_SMALL to bool type does=
 fix the LOG_CPU_MAX_BUF_SHIFT default value.
> >>
> >>>> I'll poke around to see if I can understand why a int=3D"0" is true =
for kconfig.
> >>
> >> Here's what I understood:
> >> To get the default value of LOG_CPU_MAX_BUF_SHIFT, kconfig calls sym_g=
et_default_prop(LOG_CPU_MAX_BUF_SHIFT)
> >> -> expr_calc_value("BASE_SMALL" as an expr)
> >> -> sym_calc_value(BASE_SMALL as a symbol) and returns sym->curr.tri
> >>
> >> But, if I understood correctly, sym_calc_value() does not set sym->cur=
r.tri in case of a int type config.
> >
> > Right.
>
> Thanks :)
>
> > The following will restore the original behavior.
> >
> >
> > --- a/scripts/kconfig/symbol.c
> > +++ b/scripts/kconfig/symbol.c
> > @@ -349,12 +349,15 @@ void sym_calc_value(struct symbol *sym)
> >         switch (sym->type) {
> >         case S_INT:
> >                 newval.val =3D "0";
> > +               newval.tri =3D no;
> >                 break;
> >         case S_HEX:
> >                 newval.val =3D "0x0";
> > +               newval.tri =3D no;
> >                 break;
> >         case S_STRING:
> >                 newval.val =3D "";
> > +               newval.tri =3D no;
> >                 break;
> >         case S_BOOLEAN:
> >         case S_TRISTATE:
> > >
> > But, I do not think that is the right thing to do.
> >
> > Presumably, turning CONFIG_BASE_SMALL is correct.
>
> I'm working on a patch to do that.
>

OK, please go ahead.


I will restore the Kconfig original behavior for now
and send a pull-req.


--=20
Best Regards
Masahiro Yamada