[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <CAOesGMiAKi+uqxUmcTXKvAZ_Eiup4aVvcmQ1jqMMazmDG+m-7Q@mail.gmail.com>
Date: Wed, 7 Dec 2016 14:21:54 -0800
From: Olof Johansson <olof@...om.net>
To: Laura Abbott <labbott@...hat.com>
Cc: Arnd Bergmann <arnd@...db.de>,
Bartlomiej Zolnierkiewicz <b.zolnierkie@...sung.com>,
"linux-arm-kernel@...ts.infradead.org"
<linux-arm-kernel@...ts.infradead.org>,
"linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>,
Russell King <linux@....linux.org.uk>,
Laura Abbott <laura@...bott.name>
Subject: Re: [RFC PATCH 00/23] arm: defconfigs: use kconfig fragments
On Wed, Dec 7, 2016 at 1:56 PM, Laura Abbott <labbott@...hat.com> wrote:
> On 12/07/2016 01:35 PM, Arnd Bergmann wrote:
>> On Wednesday, December 7, 2016 1:14:02 PM CET Olof Johansson wrote:
>>>>
>>>> - A "debug" fragment would be nice, to turn on common options that
>>>> add a lot of useful runtime checks at the expense of performance
>>>> or code size.
>>>
>>> Hmm, some of these might work but several useful debug options (in
>>> particular DEBUG_LL for early errors) are per-system/platform.
>>
>> I was thinking mostly of architecture-independent options, i.e.
>> the stuff that is in lib/Kconfig.debug but isn't too expensive
>> to be run in a regular test environment. Enabling those
>> for a build/boot automation environment would be particularly
>> useful as you'd catch more bugs that get introduced through
>> a random patch.
>>
>>>> - A "distro" fragment that turns on all loadable modules that are
>>>> enabled by common distributions (e.g. two or more of
>>>> debian/fedora/opensuse/gentoo), to let you build a drop-in
>>>> replacement kernel for a shipping distro. This would also allow
>>>> the distros to strip their own config files and just specify
>>>> whatever differs from the others.
>>>
>>> Keeping this in sync with the distro kernel could be a bit awkward
>>> (and possibly churny).
>>
>> It would certainly need buy-in from distro maintainers. I've discussed
>> this with Laura Abbott in the past, and she was interested in
>> principle.
>>
>> Arnd
>>
>
> Yes, I've gotten the request from multiple people now about having
> some Fedora config in mainline. I agree that churn and keeping in
> sync would be a concern. For a first pass, I was going to propose
> a minimal set of options that are unlikely to need to churn. Once
> those are agreed on, everything else could become a separate .config.
> My plan was to send a patch out around -rc2/-rc3 each cycle to sync
> up.
That sounds reasonable to me -- we could add it as
fedora_<something>_defconfig. Ideally just one per distro (for their
multi-platform target), if possible. Makes sense to get coverage of
this on builders, etc, as well.
-Olof
Powered by blists - more mailing lists