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 Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by smtp.lore.kernel.org (Postfix) with ESMTP id 9736AC4167B for ; Fri, 30 Dec 2022 08:40:16 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S234790AbiL3IkO (ORCPT ); Fri, 30 Dec 2022 03:40:14 -0500 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:49478 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S230248AbiL3IkL (ORCPT ); Fri, 30 Dec 2022 03:40:11 -0500 Received: from formenos.hmeau.com (helcar.hmeau.com [216.24.177.18]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id C50A9F7C; Fri, 30 Dec 2022 00:40:08 -0800 (PST) Received: from loth.rohan.me.apana.org.au ([192.168.167.2]) by formenos.hmeau.com with smtp (Exim 4.94.2 #2 (Debian)) id 1pBAvk-00CHTe-4w; Fri, 30 Dec 2022 16:39:57 +0800 Received: by loth.rohan.me.apana.org.au (sSMTP sendmail emulation); Fri, 30 Dec 2022 16:39:56 +0800 Date: Fri, 30 Dec 2022 16:39:56 +0800 From: Herbert Xu To: Robert Elliott Cc: davem@davemloft.net, paulmck@kernel.org, frederic@kernel.org, quic_neeraju@quicinc.com, josh@joshtriplett.org, linux-crypto@vger.kernel.org, rcu@vger.kernel.org, linux-kernel@vger.kernel.org Subject: Re: [PATCH 3/3] crypto: tcrypt - suppress RCU stall warnings during speed tests Message-ID: References: <20221219202910.3063036-1-elliott@hpe.com> <20221219202910.3063036-4-elliott@hpe.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20221219202910.3063036-4-elliott@hpe.com> Precedence: bulk List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Mon, Dec 19, 2022 at 02:29:10PM -0600, Robert Elliott wrote: > Suppress RCU CPU stall warnings while running speed tests. > > The tcrypt module is intended only for developer usage, so > RCU stalls induced by those tests are not necessarily representative > of real problems. > > Speed tests need to disable interrupts or preemption to get results > that are not distorted by such interruptions. This triggers more > RCU stalls than normal invocations of the crypto functions. Where do we disable interrupts? That would seem to break the use of jiffies since it won't get updated anymore. Which particular test is still triggering the RCU warnings after fixing the hash problems that you've already identified? Cheers, -- Email: Herbert Xu Home Page: http://gondor.apana.org.au/~herbert/ PGP Key: http://gondor.apana.org.au/~herbert/pubkey.txt