Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20;
From:   Michael Ellerman <mpe@ellerman.id.au>
To:     Christophe Leroy <christophe.leroy@csgroup.eu>,
        Nicholas Piggin <npiggin@gmail.com>
Cc:     "linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>,
        "linuxppc-dev@lists.ozlabs.org" <linuxppc-dev@lists.ozlabs.org>
Subject: Re: [PATCH v2 06/19] powerpc/cputable: Split cpu_specs[] out of
 cputable.h
In-Reply-To: <f75c2d9d-423f-f8c4-5bcd-cdea6c33e206@csgroup.eu>
References: <828f6a64eeb51ce9abfa1d4e84c521a02fecebb8.1663606875.git.christophe.leroy@csgroup.eu>
 <a44b865e0318286155273b10cdf524ab697928c1.1663606875.git.christophe.leroy@csgroup.eu>
 <CN13QQX0NEA2.M6SABG46NYYN@bobo>
 <f75c2d9d-423f-f8c4-5bcd-cdea6c33e206@csgroup.eu>
Date:   Fri, 07 Oct 2022 20:59:40 +1100
Message-ID: <87k05cnecz.fsf@mpe.ellerman.id.au>
MIME-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: quoted-printable
Precedence: bulk

Christophe Leroy <christophe.leroy@csgroup.eu> writes:
> Le 20/09/2022 =C3=A0 10:56, Nicholas Piggin a =C3=A9crit=C2=A0:
>> On Tue Sep 20, 2022 at 3:01 AM AEST, Christophe Leroy wrote:
>>=20
>> This series is a nice cleanup. No comments yet but kernel/ is getting
>> pretty crowded. Should we make some subdirectories for subarch things
>> like mm has?
>>=20
>> Can do that after your series. Probably requires another merge window
>> to do it.
>>=20
>
> By the way, I'm wondering how we decide whether some code goes in=20
> arch/powerpc/kernel/ or in arch/powerpc/lib/

On a case-by-case basis? :)

I think our lib is *mostly* routines that are library like, eg. string
copy, memcpy, memcmp etc. Though there's also single step code, code
patching etc.

I guess one thing they have in common is that they're (somewhat) self
contained routines that do a specific thing, and that they are called
from various other parts of the kernel.

On the other hand code in kernel *tends* to be more larger things, like
early boot sequence, interrupt/syscall entry, module loading etc.

But really kernel is where everything lives that doesn't have anywhere
else to go, so it's a bit of a dumping ground.


Talking specifically about all these CPU files, I think we could create
an arch/powerpc/cpu for them. On x86 they have arch/x86/kernel/cpu, but
the kernel seems redundant.

With all the CPU specs, CPU setups, and dt_cpu_ftrs.c we'd have ~4500
lines that could go in arch/powerpc/cpu. Which seems like enough to
justify a directory.

cheers