Received-SPF: pass (google.com: domain of linux-kernel+bounces-107957-linux.lists.archive=gmail.com@vger.kernel.org designates 139.178.88.99 as permitted sender) client-ip=139.178.88.99;
Precedence: bulk
MIME-Version: 1.0
References: <20240319163656.2100766-1-glider@google.com> <20240319163656.2100766-3-glider@google.com>
In-Reply-To: <20240319163656.2100766-3-glider@google.com>
From: Linus Torvalds <torvalds@linux-foundation.org>
Date: Tue, 19 Mar 2024 10:58:03 -0700
Message-ID: <CAHk-=wiUf3Eqqz3PttTCBLyDKqwW2sdpeqjL+PuKtip15vDauA@mail.gmail.com>
Subject: Re: [PATCH v1 3/3] x86: call instrumentation hooks from copy_mc.c
To: Alexander Potapenko <glider@google.com>
Cc: akpm@linux-foundation.org, linux-kernel@vger.kernel.org, 
	linux-mm@kvack.org, kasan-dev@googlegroups.com, tglx@linutronix.de, 
	x86@kernel.org, Dmitry Vyukov <dvyukov@google.com>, Marco Elver <elver@google.com>, 
	Tetsuo Handa <penguin-kernel@i-love.sakura.ne.jp>
Content-Type: text/plain; charset="UTF-8"

On Tue, 19 Mar 2024 at 09:37, Alexander Potapenko <glider@google.com> wrote:
>
>         if (copy_mc_fragile_enabled) {
>                 __uaccess_begin();
> +               instrument_copy_to_user(dst, src, len);
>                 ret = copy_mc_fragile((__force void *)dst, src, len);
>                 __uaccess_end();

I'd actually prefer that instrument_copy_to_user() to be *outside* the
__uaccess_begin.

In fact, I'm a bit surprised that objtool didn't complain about it in that form.

__uaccess_begin() causes the CPU to accept kernel accesses to user
mode, and I don't think instrument_copy_to_user() has any business
actually touching user mode memory.

In fact it might be better to rename the function and change the prototype to

   instrument_src(src, len);

because you really can't sanely instrument the destination of a user
copy, but "instrument_src()" might be useful in other situations than
just user copies.

Hmm?

               Linus