Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.18 as permitted sender) client-ip=23.128.96.18;
From:   ebiederm@xmission.com (Eric W. Biederman)
To:     Alexey Gladkov <gladkov.alexey@gmail.com>
Cc:     LKML <linux-kernel@vger.kernel.org>,
        Kernel Hardening <kernel-hardening@lists.openwall.com>,
        Linux Containers <containers@lists.linux-foundation.org>,
        linux-mm@kvack.org, Andrew Morton <akpm@linux-foundation.org>,
        Christian Brauner <christian.brauner@ubuntu.com>,
        Jann Horn <jannh@google.com>, Jens Axboe <axboe@kernel.dk>,
        Kees Cook <keescook@chromium.org>,
        Linus Torvalds <torvalds@linux-foundation.org>,
        Oleg Nesterov <oleg@redhat.com>
References: <cover.1616533074.git.gladkov.alexey@gmail.com>
        <8f0c2888b4e92d51239e154b82d75972e7e39833.1616533074.git.gladkov.alexey@gmail.com>
        <m1y2dwllfg.fsf@fess.ebiederm.org>
        <20210406154444.icpvezlq3izzxf5t@example.org>
Date:   Wed, 07 Apr 2021 11:56:36 -0500
In-Reply-To: <20210406154444.icpvezlq3izzxf5t@example.org> (Alexey Gladkov's
        message of "Tue, 6 Apr 2021 17:44:44 +0200")
Message-ID: <m1eefmjanv.fsf@fess.ebiederm.org>
User-Agent: Gnus/5.13 (Gnus v5.13) Emacs/26.1 (gnu/linux)
MIME-Version: 1.0
Content-Type: text/plain
Subject: Re: [PATCH v9 4/8] Reimplement RLIMIT_NPROC on top of ucounts
Precedence: bulk

Alexey Gladkov <gladkov.alexey@gmail.com> writes:

> On Mon, Apr 05, 2021 at 11:56:35AM -0500, Eric W. Biederman wrote:
>>
>> Also when setting ns->ucount_max[] in create_user_ns because one value
>> is signed and the other is unsigned.  Care should be taken so that
>> rlimit_infinity is translated into the largest positive value the
>> type can hold.
>
> You mean like that ?
>
> ns->ucount_max[UCOUNT_RLIMIT_NPROC] = rlimit(RLIMIT_NPROC) <= LONG_MAX ?
> 	rlimit(RLIMIT_NPROC) : LONG_MAX;
> ns->ucount_max[UCOUNT_RLIMIT_MSGQUEUE] = rlimit(RLIMIT_MSGQUEUE) <= LONG_MAX ?
> 	rlimit(RLIMIT_MSGQUEUE) : LONG_MAX;
> ns->ucount_max[UCOUNT_RLIMIT_SIGPENDING] = rlimit(RLIMIT_SIGPENDING) <= LONG_MAX ?
> 	rlimit(RLIMIT_SIGPENDING) : LONG_MAX;
> ns->ucount_max[UCOUNT_RLIMIT_MEMLOCK] = rlimit(RLIMIT_MEMLOCK) <= LONG_MAX ?
> 	rlimit(RLIMIT_MEMLOCK) : LONG_MAX;

Yes.

I only got as far as:
if (rlimit(RLIMI_NNN) == RLIM_INFINITY) {
	ns->ucount_max[UCOUNT_LIMIT_NNN] = LONG_MAX;
} else {
	ns->ucount_max[UCOUNT_LMIT_NNN] = rlmit(RLIMIT_NNN);
}
But forcing everything about LONG_MAX to LONG_MAX actually looks better
in practice.  Especially as that is effectively RLIMIT_INFINITY anyway.

Eric