From:   Lukasz Stelmach <l.stelmach@samsung.com>
To:     Masahiro Yamada <masahiroy@kernel.org>
Cc:     Marek Szyprowski <m.szyprowski@samsung.com>,
        linux-kbuild@vger.kernel.org, linux-kernel@vger.kernel.org
Subject: Re: [PATCH v2] merge_config.sh: do not report some differencess
 between input and output
In-Reply-To: <CAK7LNASjsRYyRCAy_EujsX1o2ybR3KUQRA2HuPxeN=ppUpBa_A@mail.gmail.com>
        (Masahiro Yamada's message of "Thu, 16 Feb 2023 21:17:24 +0900")
User-Agent: Gnus/5.13 (Gnus v5.13) Emacs/27.1 (gnu/linux)
Date:   Fri, 24 Feb 2023 00:30:03 +0100
Message-ID: <dleftjzg946jus.fsf%l.stelmach@samsung.com>
MIME-Version: 1.0
Content-Type: multipart/signed; boundary="=-=-="; micalg="pgp-sha256";
        protocol="application/pgp-signature"
CMS-TYPE: 201P
References: <CAK7LNASjsRYyRCAy_EujsX1o2ybR3KUQRA2HuPxeN=ppUpBa_A@mail.gmail.com>
        <CGME20230223233013eucas1p116af5ebe8e229aa6fc8636728b738156@eucas1p1.samsung.com>
Precedence: bulk

--=-=-=
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: quoted-printable

It was <2023-02-16 czw 21:17>, when Masahiro Yamada wrote:
> On Wed, Feb 8, 2023 at 10:08 PM =C5=81ukasz Stelmach <l.stelmach@samsung.=
com> wrote:
>>
>> If an input config file contains CONFIG_FOO=3Dn the output one
>> will contain a line '# CONFIG_FOO is not set'. merge_config.sh
>> should not report it as difference because the end result of
>> CONFIG_FOO being disabled is achieved.
>>
>> Inexistence of CONFIG_FOO (because of unment dependencies) in case
>> CONFIG_FOO=3Dn is requested, should also be ignored.
>>
>> Change-Id: I129f3a0b4205a76d8c42020f8adb72b1889d75fb
>> Signed-off-by: =C5=81ukasz Stelmach <l.stelmach@samsung.com>
>> ---
>> Changes in v2:
>> - suppress reports reports if an option was "not set" in input files
>>   but is missing from the filnal .config due to unmet dependecies.
>> - apply the same logic to suppress some reports during the merging
>>   phase
>>
>> BTW. Do you think adding "| sort -u" after "grep -w" to avoid reports
>> about repeated entries may make sense or do you want such reports to
>> be printed.
>>
>>  scripts/kconfig/merge_config.sh | 30 ++++++++++++++++++++----------
>>  1 file changed, 20 insertions(+), 10 deletions(-)
>>
>> diff --git a/scripts/kconfig/merge_config.sh b/scripts/kconfig/merge_con=
fig.sh
>> index e5b46980c22a..1086bdc7abf2 100755
>> --- a/scripts/kconfig/merge_config.sh
>> +++ b/scripts/kconfig/merge_config.sh
>> @@ -144,12 +144,17 @@ for ORIG_MERGE_FILE in $MERGE_LIST ; do
>>                         echo
>>                         BUILTIN_FLAG=3Dtrue
>>                 elif [ "x$PREV_VAL" !=3D "x$NEW_VAL" ] ; then
>> -                       echo Value of $CFG is redefined by fragment $ORI=
G_MERGE_FILE:
>> -                       echo Previous  value: $PREV_VAL
>> -                       echo New value:       $NEW_VAL
>> -                       echo
>> -                       if [ "$STRICT" =3D "true" ]; then
>> -                               STRICT_MODE_VIOLATED=3Dtrue
>> +                       if [ \( "x$PREV_VAL" !=3D "x$CFG=3Dn" -a \
>> +                               "x$PREV_VAL" !=3D "x# $CFG is not set" \=
) -o \
>> +                            \( "x$NEW_VAL" !=3D "x"  -a \
>
>
>
> In which case does $NEW_VAL become empty?

You are right, there is now such case.

> I think it is opposite.
> $PREV_VAL might be empty, $NEW_VAL may specified as =3Dn.

Let me try to determine when we would like to see complaints about
changes being made to config options


| PV\NV | y | m | n | n-s |
|-------+---+---+---+-----|
| y     | o | x | x | x   |
| m     | x | o | x | x   |
| n     | x | x | o | /   |
| n-s   | x | x | / | o   |
| empty | o | o | o | o   |

o - OK, don't rport
x - switched, report
/ - "n" and "not-set" are synonyms, don't report

This gives us the following conditions three conditions under which to repo=
rt value changes.

(PREV_VAL !=3D "") &&=20
(NEW_VAL !=3D PREV_VAL) &&=20
((PREV_VAL !=3D "n" && PREV_VAL !=3D "not-set") ||
 (NEW_VAL  !=3D "n" && NEw_VA L !=3D "not-set"))

Does it make sense?

>
>
>
>
>> +                               "x$NEW_VAL" !=3D "x# $CFG is not set" \)=
 ]; then
>> +                               echo Value of $CFG is redefined by fragm=
ent $ORIG_MERGE_FILE:
>> +                               echo Previous  value: $PREV_VAL
>> +                               echo New value:       $NEW_VAL
>> +                               echo
>> +                               if [ "$STRICT" =3D "true" ]; then
>> +                                       STRICT_MODE_VIOLATED=3Dtrue
>> +                               fi
>>                         fi
>>                 elif [ "$WARNREDUN" =3D "true" ]; then
>>                         echo Value of $CFG is redundant by fragment $ORI=
G_MERGE_FILE:
>> @@ -196,9 +201,14 @@ for CFG in $(sed -n -e "$SED_CONFIG_EXP1" -e "$SED_=
CONFIG_EXP2" $TMP_FILE); do
>>         REQUESTED_VAL=3D$(grep -w -e "$CFG" $TMP_FILE)
>>         ACTUAL_VAL=3D$(grep -w -e "$CFG" "$KCONFIG_CONFIG" || true)
>>         if [ "x$REQUESTED_VAL" !=3D "x$ACTUAL_VAL" ] ; then
>> -               echo "Value requested for $CFG not in final .config"
>> -               echo "Requested value:  $REQUESTED_VAL"
>> -               echo "Actual value:     $ACTUAL_VAL"
>> -               echo ""
>> +               if [ \( "x$REQUESTED_VAL" !=3D "x$CFG=3Dn" -a \
>> +                       "x$REQUESTED_VAL" !=3D "x# $CFG is not set" \) -=
o \
>> +                    \( "x$ACTUAL_VAL" !=3D "x"  -a \
>> +                       "x$ACTUAL_VAL" !=3D "x# $CFG is not set" \) ]; t=
hen
>> +                       echo "Value requested for $CFG not in final .con=
fig"
>> +                       echo "Requested value:  $REQUESTED_VAL"
>> +                       echo "Actual value:     $ACTUAL_VAL"
>> +                       echo ""
>> +               fi
>>         fi
>>  done
>> --
>> 2.30.2
>>

=2D-=20
=C5=81ukasz Stelmach
Samsung R&D Institute Poland
Samsung Electronics

--=-=-=
Content-Type: application/pgp-signature; name="signature.asc"

-----BEGIN PGP SIGNATURE-----

iQEzBAEBCAAdFiEEXpuyqjq9kGEVr9UQsK4enJilgBAFAmP39vsACgkQsK4enJil
gBBsfgf9G/QGrR5+NFhRFZvXhR9O4ahwH3U0ZdLT081ElbmlfZbz9p5jyog7EOSH
2Og5kDoD3fvtFAE3jyMRY9UCvnU6BO7GmYawGf/iymMSmhnmbgpvZYOEHgXnQqhg
Dx6V9+VPWIlLw0PUF/SWaZ1M7yawaYayyakPvgedT3bobxKka1zRRw7hPtf31MqQ
wVEvj3mcj0h0g8D5qP/mJZoDxs3okFXkxj70fFPNsjBs9HTxSTa9/Eptl0qtyG/X
YKkyg+ene1IMgB65OjjeYFXYFwEV35jFR2Q6ThIP3iZa44bz4AtR+QF1UUGfHj8b
mFy4UV3hlj00Cz7ZoB+FuPH6klfQyw==
=As5Z
-----END PGP SIGNATURE-----
--=-=-=--