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:   Thomas Gleixner <tglx@linutronix.de>
To:     linux-kernel@vger.kernel.org,
        Andrew Morton <akpm@linux-foundation.org>
Cc:     mm-commits@vger.kernel.org, mcgrof@kernel.org,
        keescook@chromium.org, chenhuacai@loongson.cn
Subject: Re: + kthread-unify-kernel_thread-and-user_mode_thread.patch added
 to mm-nonmm-unstable branch
In-Reply-To: <87352x22jc.fsf@email.froward.int.ebiederm.org>
References: <20230605231056.16BD1C433D2@smtp.kernel.org>
 <87352x22jc.fsf@email.froward.int.ebiederm.org>
Date:   Mon, 12 Jun 2023 09:21:17 +0200
Message-ID: <87wn096t8y.ffs@tglx>
MIME-Version: 1.0
Content-Type: text/plain
Precedence: bulk

On Sun, Jun 11 2023 at 14:59, Eric W. Biederman wrote:
> Andrew Morton <akpm@linux-foundation.org> writes:
>
>> The patch titled
>>      Subject: kthread: Unify kernel_thread() and user_mode_thread()
>> has been added to the -mm mm-nonmm-unstable branch.  Its filename is
>>      kthread-unify-kernel_thread-and-user_mode_thread.patch
>
> Andrew.
>
> My fuzzy memory thinks Linus asked for the current split.

Correct. It was in a discussion about a nasty security hole due to a
race in the original code which did _not_ have the distinction.

> Plus this change just obfuscates the code making the most important
> detail the argument to a boolean parameter.  Meaning you have to have
> an interface that has only 3 callers memorized to even begin to make
> sense of it.

Right. Losing the clear distinction of the function names is a horrible
idea.

If at all this should at least keep user_mode_thread() and
kernel_thread() as inline wrappers around a common function.

Just blindly unifying code is a patently bad idea.

Thanks,

        tglx