From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org X-Spam-Level: X-Spam-Status: No, score=-10.8 required=3.0 tests=BAYES_00,DKIM_SIGNED, DKIM_VALID,DKIM_VALID_AU,HEADER_FROM_DIFFERENT_DOMAINS,INCLUDES_PATCH, MAILING_LIST_MULTI,SPF_HELO_NONE,SPF_PASS,URIBL_BLOCKED autolearn=ham autolearn_force=no version=3.4.0 Received: from mail.kernel.org (mail.kernel.org [198.145.29.99]) by smtp.lore.kernel.org (Postfix) with ESMTP id C5768C47082 for ; Mon, 31 May 2021 17:06:15 +0000 (UTC) Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by mail.kernel.org (Postfix) with ESMTP id A5E9C600EF for ; Mon, 31 May 2021 17:06:15 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S232043AbhEaRHy (ORCPT ); Mon, 31 May 2021 13:07:54 -0400 Received: from Galois.linutronix.de ([193.142.43.55]:55514 "EHLO galois.linutronix.de" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S233721AbhEaQc2 (ORCPT ); Mon, 31 May 2021 12:32:28 -0400 From: John Ogness DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=linutronix.de; s=2020; t=1622478644; h=from:from:reply-to:subject:subject:date:date:message-id:message-id: to:to:cc:cc:mime-version:mime-version:content-type:content-type: in-reply-to:in-reply-to:references:references; bh=cefR/DtEcUyRZ8qc11g2NcI+aJPhU+kuvC8iQhInVMA=; b=Mk/taBgxowuo4I3RBEs+YxImg4/sI6Zr4+03Zx+d/LpWbdEMBYUa2TQF5IPIEnWqD8K4sR rmwewO+kYXUdcfT0fDRR83TP+yPI33o8yq/3r2Ud6S3eHEtLpMHDPbrMA0OyGBZyfc28as ai983E+pkcci1wPmFcIWtdC1QM8XFOluLt1kFDjBMS1+726dCSdb/bHe3DKqVP8dOJJR3v YaSUqyeq036LwYOJSsQidq1Sw4m4KZnc0KByZSTRW81VU/dT05rRdQRwwPBeMcTAcVWkMb ztNn2yNHfxVnV8aCj1NgTKKcTn+3wHagtNRYBaYCwz1UNCBS3VGXXgdclIjQKA== DKIM-Signature: v=1; a=ed25519-sha256; c=relaxed/relaxed; d=linutronix.de; s=2020e; t=1622478644; h=from:from:reply-to:subject:subject:date:date:message-id:message-id: to:to:cc:cc:mime-version:mime-version:content-type:content-type: in-reply-to:in-reply-to:references:references; bh=cefR/DtEcUyRZ8qc11g2NcI+aJPhU+kuvC8iQhInVMA=; b=SZh3BUGPPABae9lerayRXuLT/DGm3svJi1sw8XUyfOFIRBxKnZdsCuGOlvpMse4wn7jSEP odIoz9/KJ5SpFrAA== To: Petr Mladek Cc: Sergey Senozhatsky , Sergey Senozhatsky , Steven Rostedt , Thomas Gleixner , linux-kernel@vger.kernel.org, Sergey Senozhatsky , Andrew Morton , Stephen Rothwell , Dmitry Safonov <0x7f454c46@gmail.com>, Valentin Schneider , Daniel Bristot de Oliveira , Peter Zijlstra , Stephen Boyd , Alexander Potapenko , "Paul E. McKenney" Subject: Re: [PATCH next v1 1/2] dump_stack: move cpu lock to printk.c In-Reply-To: <20210531162051.2325-2-john.ogness@linutronix.de> References: <20210531162051.2325-1-john.ogness@linutronix.de> <20210531162051.2325-2-john.ogness@linutronix.de> Date: Mon, 31 May 2021 18:30:43 +0200 Message-ID: <87mtsayifg.fsf@jogness.linutronix.de> MIME-Version: 1.0 Content-Type: text/plain Precedence: bulk List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On 2021-05-31, John Ogness wrote: > diff --git a/kernel/printk/printk.c b/kernel/printk/printk.c > index 114e9963f903..98feead621ff 100644 > --- a/kernel/printk/printk.c > +++ b/kernel/printk/printk.c > @@ -3531,4 +3531,96 @@ void kmsg_dump_rewind(struct kmsg_dump_iter *iter) > } > EXPORT_SYMBOL_GPL(kmsg_dump_rewind); > > +#ifdef CONFIG_SMP > +static atomic_t printk_cpulock_owner = ATOMIC_INIT(-1); > + > +/* > + * printk_cpu_lock: Acquire the printk cpu-reentrant spinning lock. > + * @cpu_store: A buffer to store lock state. > + * @flags: A buffer to store irq state. > + * > + * If no processor has the lock, the calling processor takes the lock and > + * becomes the owner. If the calling processor is already the owner of the > + * lock, this function succeeds immediately. If the lock is locked by another > + * processor, that function spins until the calling processor becomes the > + * owner. > + * > + * It is safe to call this function from any context and state. > + */ > +void printk_cpu_lock(unsigned int *cpu_store, unsigned long *flags) > +{ > + unsigned int cpu; > + > + for (;;) { > + cpu = get_cpu(); > + > + *cpu_store = atomic_read(&printk_cpulock_owner); > + > + if (*cpu_store == -1) { > + local_irq_save(*flags); > + > + /* > + * Guarantee loads an stores from the previous lock > + * owner are visible to this CPU once it is the lock > + * owner. This pairs with cpu_lock:A. The above paragraph is totally broken. It should read: + * Guarantee loads and stores from the previous lock + * owner are visible to this CPU once it is the lock + * owner. This pairs with cpu_unlock:B. John Ogness > + * > + * Memory barrier involvement: > + * > + * If cpu_lock:A reads from cpu_unlock:B, then > + * cpu_lock:B reads from cpu_unlock:A. > + * > + * Relies on: > + * > + * RELEASE from cpu_unlock:A to cpu_unlock:B > + * matching > + * ACQUIRE from cpu_lock:A to cpu_lock:B > + */ > + if (atomic_try_cmpxchg_acquire(&printk_cpulock_owner, > + cpu_store, cpu)) { /* LMM(cpu_lock:A) */ > + > + /* This CPU begins loading/storing data: LMM(cpu_lock:B) */ > + break; > + } > + > + local_irq_restore(*flags); > + > + } else if (*cpu_store == cpu) { > + break; > + } > + > + put_cpu(); > + cpu_relax(); > + } > +} > +EXPORT_SYMBOL(printk_cpu_lock); > + > +/* > + * printk_cpu_unlock: Release the printk cpu-reentrant spinning lock. > + * @cpu_store: The current lock state. > + * @flags: The current irq state. > + * > + * Release the lock. The calling processor must be the owner of the lock. > + * > + * It is safe to call this function from any context and state. > + */ > +void printk_cpu_unlock(unsigned int cpu_store, unsigned long flags) > +{ > + if (cpu_store == -1) { > + /* This CPU is finished loading/storing data: LMM(cpu_unlock:A) */ > + > + /* > + * Guarantee loads an stores from this CPU when it is the lock > + * owner are visible to the next lock owner. This pairs with > + * cpu_lock:A. > + */ > + atomic_set_release(&printk_cpulock_owner, cpu_store); /* LMM(cpu_unlock:B) */ > + > + local_irq_restore(flags); > + } > + > + put_cpu(); > +} > +EXPORT_SYMBOL(printk_cpu_unlock); > +#endif /* CONFIG_SMP */ > + > #endif