From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org Received: from aws-us-west-2-korg-lkml-1.web.codeaurora.org (localhost.localdomain [127.0.0.1]) by smtp.lore.kernel.org (Postfix) with ESMTP id 03727C982C1 for ; Thu, 17 Sep 2026 05:44:38 +0000 (UTC) Subject: Re: [meta-oe][PATCH] smolkit: Add low-footprint drop-in replacement for polkit To: openembedded-devel@lists.openembedded.org From: "Severin Tillmann, DE-862" X-Originating-Location: Munich, Bavaria, DE (160.46.252.58) X-Originating-Platform: Linux Firefox 150 User-Agent: GROUPS.IO Web Poster MIME-Version: 1.0 Date: Wed, 16 Sep 2026 22:44:31 -0700 References: <20260914122927.87418-1-tillmann.severin@bmw.de> In-Reply-To: Message-ID: <546192.1789623871363509840@lists.openembedded.org> Content-Type: multipart/alternative; boundary="wJ4sYCYDJhNHHMY19JZx" List-Id: X-Webhook-Received: from 45-33-107-173.ip.linodeusercontent.com [45.33.107.173] by aws-us-west-2-korg-lkml-1.web.codeaurora.org with HTTPS for ; Thu, 17 Sep 2026 05:44:38 -0000 X-Groupsio-URL: https://lists.openembedded.org/g/openembedded-devel/message/130091 --wJ4sYCYDJhNHHMY19JZx Content-Type: text/plain; charset="utf-8" Content-Transfer-Encoding: quoted-printable Hi Ross, thanks a lot for your feedback! On Tue, Sep 15, 2026 at 06:30 AM, Ross Burton wrote: >=20 > On 14 Sep 2026, at 13:29, Severin Tillmann, DE-862 via > lists.openembedded.org > wrote: >=20 >> +CONFLICT_DISTRO_FEATURES =3D "polkit" >> +REQUIRED_DISTRO_FEATURES =3D "smolkit" >=20 > This makes the claim of =E2=80=9Cdrop-in replacement=E2=80=9D suspicious,= because it would > mean every recipe that currently depends on polkit and respects the polki= t > DISTRO_FEATURE needs to be modified. >=20 > Could smolkit not PROVIDE polkit, so the distro can just switch the > preferred provider and everything just works? You are of course right, smolkit is not a complete drop-in replacement (yet= ), it is only a drop-in replacement for a certain subset of polkit's interf= ace, I will try to rephrase that in the next version of the patch. smolkit = does not provide any helper tools like e.g. `pkexec` and also does not impl= ement the whole featureset that polkit has. This might lead to issues if ot= her recipes expect tools, libraries or other features polkit provides, but = smolkit does not provide. Do you think it nevertheless makes sense to switc= h to a PROVIDES based approach here? I have checked that the build succeeds= also with `PROVIDES` and `PREFERRED_PROVIDER_polkit =3D "smolkit"`, so it = should not make a big difference, I am just worried regarding expectations = from other recipes. I would send a patch v2 once it is decided which way to= go. >=20 >=20 >=20 >> +PV =3D "0.2+git" >> + >> +SRC_URI =3D "\ >> + git://github.com/bmwcarit/${BPN}.git;protocol=3Dhttps;branch=3Dmain \ >> +" >> +SRCREV =3D "46bbfd5c922a8d03a534398ee129ded822dfcafe" >=20 > I notice a lack of tags upstream=E2=80=A6 Tags are now present on the upstream repository. >=20 >=20 >> +S =3D "${UNPACKDIR}/${BB_GIT_DEFAULT_DESTSUFFIX}" >=20 > You can drop this line. >=20 > Finally, I absolutely love the name. >=20 > Ross I will drop the line in the next version and am happy that you like the nam= e ;-) Best regards, Tillmann --wJ4sYCYDJhNHHMY19JZx Content-Type: text/html; charset="utf-8" Content-Transfer-Encoding: quoted-printable
Hi Ross,
 
thanks a lot for your feedback!
 
On Tue, Sep 15, 2026 at 06:30 AM, Ross Burton wrote:
On 14 Sep 2026, at 13:29, Severin Tillmann, DE-862 via lists.op= enembedded.org <Tillmann.Severin=3Dbmw.de@lists.openembedded.org> wro= te:
+CONFLICT_DISTRO_FEATURES =3D "polkit"
+REQUIRED_DISTRO_FE= ATURES =3D "smolkit"
This makes the claim of “drop-in replacement” suspicious, becau= se it would mean every recipe that currently depends on polkit and respects= the polkit DISTRO_FEATURE needs to be modified.

Could smolkit n= ot PROVIDE polkit, so the distro can just switch the preferred provider and= everything just works?
You are of course right, smolkit is not a complete drop-in replacement (yet= ), it is only a drop-in replacement for a certain subset of polkit's interf= ace, I will try to rephrase that in the next version of the patch. smolkit = does not provide any helper tools like e.g. `pkexec` and also does not impl= ement the whole featureset that polkit has. This might lead to issues if ot= her recipes expect tools, libraries or other features polkit provides, but = smolkit does not provide. Do you think it nevertheless makes sense to switc= h to a PROVIDES based approach here? I have checked that the build succeeds= also with `PROVIDES` and `PREFERRED_PROVIDER_polkit =3D "smolkit"`, so it = should not make a big difference, I am just worried regarding expectations = from other recipes. I would send a patch v2 once it is decided which way to= go.
+PV =3D "0.2+git"
+
+SRC_URI =3D "\
+ git://gith= ub.com/bmwcarit/${BPN}.git;protocol=3Dhttps;branch=3Dmain \
+"
+S= RCREV =3D "46bbfd5c922a8d03a534398ee129ded822dfcafe"
I notice a lack of tags upstream…
Tags are now present on the upstream repository.
+S =3D "${UNPACKDIR}/${BB_GIT_DEFAULT_DESTSUFFIX}"
You can drop this line.

Finally, I absolutely love the name.

Ross
I will drop the line in the next version and am happy that you like the nam= e ;-)

Best regards,
Tillmann
--wJ4sYCYDJhNHHMY19JZx--