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=-1.0 required=3.0 tests=HEADER_FROM_DIFFERENT_DOMAINS, MAILING_LIST_MULTI,SPF_PASS 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 DAFA5C43219 for ; Fri, 26 Apr 2019 15:06:07 +0000 (UTC) Received: from vger.kernel.org (vger.kernel.org [209.132.180.67]) by mail.kernel.org (Postfix) with ESMTP id B7036206E0 for ; Fri, 26 Apr 2019 15:06:07 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1726455AbfDZPGG (ORCPT ); Fri, 26 Apr 2019 11:06:06 -0400 Received: from usa-sjc-mx-foss1.foss.arm.com ([217.140.101.70]:43860 "EHLO foss.arm.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1725984AbfDZPGG (ORCPT ); Fri, 26 Apr 2019 11:06:06 -0400 Received: from usa-sjc-imap-foss1.foss.arm.com (unknown [10.72.51.249]) by usa-sjc-mx-foss1.foss.arm.com (Postfix) with ESMTP id 8F74980D; Fri, 26 Apr 2019 08:06:05 -0700 (PDT) Received: from [10.1.196.50] (e108454-lin.cambridge.arm.com [10.1.196.50]) by usa-sjc-imap-foss1.foss.arm.com (Postfix) with ESMTPSA id AD28C3F5C1; Fri, 26 Apr 2019 08:06:03 -0700 (PDT) Subject: Re: [PATCH v4 3/3] arm64/fpsimd: Don't disable softirq when touching FPSIMD/SVE state To: Dave Martin Cc: linux-kernel@vger.kernel.org, linux-arm-kernel@lists.infradead.org, ard.biesheuvel@linaro.org, julien.thierry@arm.com, marc.zyngier@arm.com, catalin.marinas@arm.com, suzuki.poulose@arm.com, will.deacon@arm.com, christoffer.dall@arm.com, james.morse@arm.com References: <20190426143740.31973-1-julien.grall@arm.com> <20190426143740.31973-4-julien.grall@arm.com> <20190426145232.GK3567@e103592.cambridge.arm.com> From: Julien Grall Message-ID: <322340c7-0c97-76f8-8ab8-875040b4459c@arm.com> Date: Fri, 26 Apr 2019 16:06:02 +0100 User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:60.0) Gecko/20100101 Thunderbird/60.6.1 MIME-Version: 1.0 In-Reply-To: <20190426145232.GK3567@e103592.cambridge.arm.com> Content-Type: text/plain; charset=utf-8; format=flowed Content-Language: en-US Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org Precedence: bulk List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Hi, On 26/04/2019 15:52, Dave Martin wrote: > On Fri, Apr 26, 2019 at 03:37:40PM +0100, Julien Grall wrote: >> When the kernel is compiled with CONFIG_KERNEL_MODE_NEON, some part of >> the kernel may be able to use FPSIMD/SVE. This is for instance the case >> for crypto code. >> >> Any use of FPSIMD/SVE in the kernel are clearly marked by using the >> function kernel_neon_{begin, end}. Furthermore, this can only be used >> when may_use_simd() returns true. >> >> The current implementation of may_use_simd() allows softirq to use >> FPSIMD/SVE unless it is currently in use (i.e kernel_neon_busy is true). >> When in use, softirqs usually fall back to a software method. >> >> At the moment, as a softirq may use FPSIMD/SVE, softirqs are disabled >> when touching the FPSIMD/SVE context. This has the drawback to disable >> all softirqs even if they are not using FPSIMD/SVE. >> >> Since a softirq is supposed to check may_use_simd() anyway before >> attempting to use FPSIMD/SVE, there is limited reason to keep softirq >> disabled when touching the FPSIMD/SVE context. Instead, we can simply >> disable preemption and mark the FPSIMD/SVE context as in use by setting >> CPU's kernel_neon_busy flag. > > fpsimd_context_busy? Yes. > >> Two new helpers {get, put}_cpu_fpsimd_context is introduced to mark the >> area using FPSIMD/SVE context and uses them in replacement of > > Paragraph mangled during edit? Possibly, I will update it. > > -> "are introduced ... and they are used to replace ..." > >> local_bh_{disable, enable}. The functions kernel_neon_{begin, end} are >> also re-implemented to use the new helpers. >> >> Additionally, double-underscored versions of the helpers are provided to >> be used in function called with interrupt masked. They are used for >> sanity and also help to mark place where the FPSIMD context can be >> manipulate freely. > > For the benefit of other readers, this should be more explicit. Also, > the distinction between the normal and __ helpers is that the latter > can be caller with preemption disabled. > > To clarify the impact, we can say something like > > "These are only relevant on paths where irqs are disabled anyway, so > they are not needed for correctness in the current code. Let's use them > anyway though: this marks the critical sections clearly and will help > to avoid mistakes during future maintenance." How about the following commit message? arm64/fpsimd: Don't disable softirq when touching FPSIMD/SVE state When the kernel is compiled with CONFIG_KERNEL_MODE_NEON, some part of the kernel may be able to use FPSIMD/SVE. This is for instance the case for crypto code. Any use of FPSIMD/SVE in the kernel are clearly marked by using the function kernel_neon_{begin, end}. Furthermore, this can only be used when may_use_simd() returns true. The current implementation of may_use_simd() allows softirq to use FPSIMD/SVE unless it is currently in use (i.e kernel_neon_busy is true). When in use, softirqs usually fall back to a software method. At the moment, as a softirq may use FPSIMD/SVE, softirqs are disabled when touching the FPSIMD/SVE context. This has the drawback to disable all softirqs even if they are not using FPSIMD/SVE. Since a softirq is supposed to check may_use_simd() anyway before attempting to use FPSIMD/SVE, there is limited reason to keep softirq disabled when touching the FPSIMD/SVE context. Instead, we can simply disable preemption and mark the FPSIMD/SVE context as in use by setting CPU's fpsimd_context_busy flag. Two new helpers {get, put}_cpu_fpsimd_context are introduced to mark the area using FPSIMD/SVE context and they are used to replace local_bh_{disable, enable}. The functions kernel_neon_{begin, end} are also re-implemented to use the new helpers. Additionally, double-underscored versions of the helpers are provided to called when preemption is already disabled. These are only relevant on paths where irqs are disabled anyway, so they are not needed for correctness in the current code. Let's use them anyway though: this marks critical sections clearly and will help to avoid mistakes during future maintenance. The change has been benchmarked on Linux 5.1-rc4 with defconfig. On Juno2: * hackbench 100 process 1000 (10 times) * .7% quicker On ThunderX 2: * hackbench 1000 process 1000 (20 times) * 3.4% quicker > > [...] > > (Sorry to nitpick) That's fine, I should have be more careful when rework the commit message. Cheers, -- Julien Grall