[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <dcd8eb4c-f151-e1f7-cec9-8b66e860f748@linux-m68k.org>
Date: Thu, 30 Sep 2021 22:31:06 +1000
From: Greg Ungerer <gerg@...ux-m68k.org>
To: Geert Uytterhoeven <geert@...ux-m68k.org>,
Randy Dunlap <rdunlap@...radead.org>
Cc: LKML <linux-kernel@...r.kernel.org>,
linux-m68k <linux-m68k@...ts.linux-m68k.org>
Subject: Re: m68k: default value for MEMORY_RESERVE?
Hi Randy, Geert,
On 28/9/21 5:20 pm, Geert Uytterhoeven wrote:
> Hi Randy,
>
> On Mon, Sep 27, 2021 at 11:26 PM Randy Dunlap <rdunlap@...radead.org> wrote:
>> Would it be safe (or unsafe) to add a default value for MEMORY_RESERVE?
>
> No idea ;-)
Me neither :-)
Looking at the code I am pretty sure a default value would be ok.
I am not sure what the need for MEMORY_RESERVE was for.
It is only used on the uCsimm and uCdimm platforms, which are DragonBall
(68328) based devices. Not obvious to me why it was needed.
>> As it is, kconfig can generate a randconfig with
>> "CONFIG_MEMORY_RESERVE=" (no value) since it has no default.
>>
>> Then a following 'make all' finds an invalid symbol value and restarts
>> the config:
>>
>> .config:214:warning: symbol value '' invalid for MEMORY_RESERVE
>> * Restart config...
>> Memory reservation (MiB) (MEMORY_RESERVE) [] (NEW)
>>
>> That's OK until I try to script/automate randconfig builds.
>>
>> Can we do anything about this or should I just add it to my
>> ignore-list?
>
> Looking at its sole user, I guess "default 0" would be fine.
> Greg?
Yep, I think that would be fine. Looks to be harmless if set to 0.
Regards
Greg
Powered by blists - more mailing lists