linux-kernel - Re: [PATCH 4/5] locking/percpu-rwsem: Extract __percpu_down_read

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, 18 Nov 2019 17:28:38 +0100
From:   Oleg Nesterov <oleg@...hat.com>
To:     Peter Zijlstra <peterz@...radead.org>
Cc:     mingo@...nel.org, will@...nel.org, tglx@...utronix.de,
        linux-kernel@...r.kernel.org, bigeasy@...utronix.de,
        juri.lelli@...hat.com, williams@...hat.com, bristot@...hat.com,
        longman@...hat.com, dave@...olabs.net, jack@...e.com
Subject: Re: [PATCH 4/5] locking/percpu-rwsem: Extract
 __percpu_down_read_trylock()

Hi Peter, sorry for delay.

I'll re-read this series tomorrow, but everything looks correct at first
glance...

Except one very minor problem in this patch, see below.

On 11/13, Peter Zijlstra wrote:
>
> -bool __percpu_down_read(struct percpu_rw_semaphore *sem, bool try)
> +static bool __percpu_down_read_trylock(struct percpu_rw_semaphore *sem)
>  {
>  	__this_cpu_inc(*sem->read_count);
>  
> @@ -70,14 +70,21 @@ bool __percpu_down_read(struct percpu_rw
>  	 * If !readers_block the critical section starts here, matched by the
>  	 * release in percpu_up_write().
>  	 */
> -	if (likely(!smp_load_acquire(&sem->readers_block)))
> +	if (likely(!atomic_read_acquire(&sem->readers_block)))

I don't think this can be compiled ;) ->readers_block is "int" until the next
patch makes it atomic_t and renames to ->block.

And. I think __percpu_down_read_trylock() should do

	if (atomic_read(&sem->block))
		return false;

at the start, before __this_cpu_inc(read_count).

Suppose that the pending writer sleeps in rcuwait_wait_event(readers_active_check).
If the new reader comes, it is better to not wake up that writer.

Oleg.