linux-kernel - Re: [PATCH 1/3] notifier: Fix broken error handling pattern

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Date:   Mon, 24 Jun 2019 17:21:07 -0500
From:   Josh Poimboeuf <jpoimboe@...hat.com>
To:     Peter Zijlstra <peterz@...radead.org>
Cc:     Jessica Yu <jeyu@...nel.org>, linux-kernel@...r.kernel.org,
        jikos@...nel.org, mbenes@...e.cz, pmladek@...e.com, ast@...nel.org,
        daniel@...earbox.net, akpm@...ux-foundation.org,
        "Rafael J. Wysocki" <rjw@...ysocki.net>,
        Pavel Machek <pavel@....cz>, Len Brown <len.brown@...el.com>,
        Sam Protsenko <semen.protsenko@...aro.org>,
        Thomas Gleixner <tglx@...utronix.de>,
        Greg Kroah-Hartman <gregkh@...uxfoundation.org>,
        Alexios Zavras <alexios.zavras@...el.com>,
        Allison Randal <allison@...utok.net>,
        Vasily Averin <vvs@...tuozzo.com>,
        Todd Brandt <todd.e.brandt@...ux.intel.com>,
        linux-pm@...r.kernel.org
Subject: Re: [PATCH 1/3] notifier: Fix broken error handling pattern

On Mon, Jun 24, 2019 at 11:18:44AM +0200, Peter Zijlstra wrote:
> The current notifiers have the following error handling pattern all
> over the place:
> 
> 	int nr;
> 
> 	ret = __foo_notifier_call_chain(&chain, val_up, v, -1, &nr);
> 	if (err & NOTIFIER_STOP_MASK)

s/err/ret/

> 		__foo_notifier_call_chain(&chain, val_down, v, nr-1, NULL)
> 
> And aside from the endless repetition thereof, it is broken. Consider
> blocking notifiers; both calls take and drop the rwsem, this means
> that the notifier list can change in between the two calls, making @nr
> meaningless.
> 
> Fix this by replacing all the __foo_notifier_call_chain() functions
> with foo_notifier_call_chain_error() that embeds the above patter, but
> ensures it is inside a single lock region.

The name "notifier_call_chain_error()" seems confusing, it almost sounds
like it's notifying an error code.  Then again, I can't really think of
a more reasonably succinct name.

> @@ -25,8 +25,23 @@ static int cpu_pm_notify(enum cpu_pm_eve
>  	 * RCU know this.
>  	 */
>  	rcu_irq_enter_irqson();
> -	ret = __atomic_notifier_call_chain(&cpu_pm_notifier_chain, event, NULL,
> -		nr_to_call, nr_calls);
> +	ret = atomic_notifier_call_chain(&cpu_pm_notifier_chain, event, NULL);
> +	rcu_irq_exit_irqson();
> +
> +	return notifier_to_errno(ret);
> +}
> +
> +static int cpu_pm_notify_error(enum cpu_pm_event event_up, enum cpu_pm_event event_down)
> +{
> +	int ret;
> +
> +	/*
> +	 * __atomic_notifier_call_chain has a RCU read critical section, which

__atomic_notifier_call_chain() no longer exists.

> +	 * could be disfunctional in cpu idle. Copy RCU_NONIDLE code to let

"dysfunctional"

> @@ -156,43 +169,30 @@ int atomic_notifier_chain_unregister(str
>  }
>  EXPORT_SYMBOL_GPL(atomic_notifier_chain_unregister);
>  
> -/**
> - *	__atomic_notifier_call_chain - Call functions in an atomic notifier chain
> - *	@nh: Pointer to head of the atomic notifier chain
> - *	@val: Value passed unmodified to notifier function
> - *	@v: Pointer passed unmodified to notifier function
> - *	@nr_to_call: See the comment for notifier_call_chain.
> - *	@nr_calls: See the comment for notifier_call_chain.
> - *
> - *	Calls each function in a notifier chain in turn.  The functions
> - *	run in an atomic context, so they must not block.
> - *	This routine uses RCU to synchronize with changes to the chain.
> - *
> - *	If the return value of the notifier can be and'ed
> - *	with %NOTIFY_STOP_MASK then atomic_notifier_call_chain()
> - *	will return immediately, with the return value of
> - *	the notifier function which halted execution.
> - *	Otherwise the return value is the return value
> - *	of the last notifier function called.
> - */

Why remove the useful comment?

Ditto for the blocking, raw, srcu, comments.

-- 
Josh