Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::3:6 as permitted sender) client-ip=2620:137:e000::3:6;
Date:   Fri, 15 Sep 2023 23:18:15 +0200
From:   Peter Zijlstra <peterz@infradead.org>
To:     Linus Torvalds <torvalds@linux-foundation.org>
Cc:     Bartosz Golaszewski <bartosz.golaszewski@linaro.org>,
        Alexey Dobriyan <adobriyan@gmail.com>,
        linux-kernel@vger.kernel.org,
        Linus Walleij <linus.walleij@linaro.org>,
        akpm@linux-foundation.org
Subject: Re: Buggy __free(kfree) usage pattern already in tree
Message-ID: <20230915211815.GA34041@noisy.programming.kicks-ass.net>
References: <07c32bf1-6c1a-49d9-b97d-f0ae4a2b42ab@p183>
 <CAHk-=whO1+-4ALjFWSE0kzytz1kEbWPvy3xWvcUP1dJ4t-QqkA@mail.gmail.com>
 <CACMJSetxQi+t3SBXu6OvBbmxV8AbX2CfdSA9JvF1chLJSU9Ppw@mail.gmail.com>
 <CAHk-=wgRHiV5VSxtfXA4S6aLUmcQYEuB67u3BJPJPtuESs1JyA@mail.gmail.com>
 <CACMJSevZQgik7S-62fz9H7+Mib+W0CgYMV4GyWjYV7N_E6iHVQ@mail.gmail.com>
 <CACMJSevrJ5KSPAZVheXkNaYj8KQFD8ck55kU_E4vEj4vzR8wnQ@mail.gmail.com>
 <CAHk-=wicfvWPuRVDG5R1mZSxD8Xg=-0nLOiHay2T_UJ0yDX42g@mail.gmail.com>
 <20230915210851.GA23174@noisy.programming.kicks-ass.net>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <20230915210851.GA23174@noisy.programming.kicks-ass.net>
Precedence: bulk

On Fri, Sep 15, 2023 at 11:08:51PM +0200, Peter Zijlstra wrote:

> But a little later in that same function I then have:
> 
> 	do {
> 		struct rw_semaphore *exec_update_lock __free(up_read) = NULL;
> 		if (task) {
> 			err = down_read_interruptible(&task->signal->exec_update_lock);
> 			if (err)
> 				return err;
> 
> 			exec_update_lock = &task->signal->exec_update_lock;
> 
> 			if (!perf_check_permissions(&attr, task))
> 				return -EACCESS;
> 		}
> 
> 		... stuff serialized against exec *if* this is a task event ...
> 
> 	} while (0);
> 
> 
> And that might be a little harder to 'fix'.

Hmm, perhaps I can do a class for it and write the thing like:

	do {
		CLASS(cond_exec_update_lock, exec_lock_guard)(task); /* allow task == NULL */
		if (task && !exec_lock_guard)
			return -EINTR;

		if (task && !perf_check_permissions(&attr, task))
			return -EACCESS;

		... the rest ...

	} while (0);

that might be nicer..