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 bombadil.infradead.org (bombadil.infradead.org [198.137.202.133]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by smtp.lore.kernel.org (Postfix) with ESMTPS id 3FFD1ECAAD3 for ; Mon, 5 Sep 2022 14:21:18 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; q=dns/txt; c=relaxed/relaxed; d=lists.infradead.org; s=bombadil.20210309; h=Sender:Content-Type: Content-Transfer-Encoding:List-Subscribe:List-Help:List-Post:List-Archive: List-Unsubscribe:List-Id:In-Reply-To:From:References:Cc:To:Subject: MIME-Version:Date:Message-ID:Reply-To:Content-ID:Content-Description: Resent-Date:Resent-From:Resent-Sender:Resent-To:Resent-Cc:Resent-Message-ID: List-Owner; bh=IWfN+CSeO+2ujN+iZL0ZSSzsjyGs+Y0XkXRwoOR37eo=; b=TqS6RfhFuOnLUQ TpmQQt9IhaXeOF9gKR+S0dZ8iNua4W816rXoS9m7pKjpcSA8CgiJUl8yU9yNrIU7LtitFcvxYvHuP bxB8i768jrbyLhDY3ZcdB0fzjWQnKrNawIWYwsLVIS3Qh3nbNFD7cY6xwcVW6NgZWJBJAGG5Bf7p+ sls+eG2SJ6arVEEUESFBWyxSHJ0bbdFyZJoUP4+ZHxBX9uzIeznKd8OuoDcHfDdrFuQkspZ1QwK90 FJOGRrR3KpWXvTs5BmZ2foJnPohp8CpaWNhh0sWDIzTbnR+LYQ0soeaVP25Uustxyw97fxN2Zh4dN khJoig+ouJXparuuXa7Q==; Received: from localhost ([::1] helo=bombadil.infradead.org) by bombadil.infradead.org with esmtp (Exim 4.94.2 #2 (Red Hat Linux)) id 1oVCx1-004K2R-Iv; Mon, 05 Sep 2022 14:19:47 +0000 Received: from desiato.infradead.org ([2001:8b0:10b:1:d65d:64ff:fe57:4e05]) by bombadil.infradead.org with esmtps (Exim 4.94.2 #2 (Red Hat Linux)) id 1oVA12-000vM7-4u for linux-arm-kernel@bombadil.infradead.org; Mon, 05 Sep 2022 11:11:44 +0000 DKIM-Signature: v=1; a=rsa-sha256; q=dns/txt; c=relaxed/relaxed; d=infradead.org; s=desiato.20200630; h=Content-Transfer-Encoding:Content-Type :In-Reply-To:From:References:Cc:To:Subject:MIME-Version:Date:Message-ID: Sender:Reply-To:Content-ID:Content-Description; bh=7eWV7gk4NPntLOR/lmgycaFf4ZZOkcJFD32xV930G0o=; b=na2N4odBu4cxckwMxnfDOBEXC+ b1c9X2poSfLCyWk96Fi0Eme0VG+J6jDwykvIDTt5Gmz+SGTrZrFx5+pma0gXvAVzTXTqDRiv7/eZp HR/C4o6UDJHJuyYs8JCf/5zgR2YpcSQEKTIEPw0a4QEcatcGzk9UsXYHVGrfusFipUokofiv11QzL n+664UD1V++K4jMixPWEQz2C4kaRcDxOicfPumh3qlmnWLYHL3m5SoDTOGFaWfHzhLWxAmdbkO9M6 j1BtgI8m3IIRd2xgTOMinBW6y5q7bkXs2N9NrrvHoXeX8dmwLQOH2Rk4llSd9rTblBWG7h0iV6dQZ oyF/nvwg==; Received: from foss.arm.com ([217.140.110.172]) by desiato.infradead.org with esmtp (Exim 4.94.2 #2 (Red Hat Linux)) id 1oV9d9-009X6X-S3 for linux-arm-kernel@lists.infradead.org; Mon, 05 Sep 2022 10:47:06 +0000 Received: from usa-sjc-imap-foss1.foss.arm.com (unknown [10.121.207.14]) by usa-sjc-mx-foss1.foss.arm.com (Postfix) with ESMTP id A86F7D6E; Mon, 5 Sep 2022 03:47:05 -0700 (PDT) Received: from [10.57.15.199] (unknown [10.57.15.199]) by usa-sjc-imap-foss1.foss.arm.com (Postfix) with ESMTPSA id 9E80E3F7B4; Mon, 5 Sep 2022 03:46:57 -0700 (PDT) Message-ID: <696a1a53-87b5-7517-3c05-82217c8eb190@arm.com> Date: Mon, 5 Sep 2022 11:46:50 +0100 MIME-Version: 1.0 User-Agent: Mozilla/5.0 (Windows NT 10.0; rv:102.0) Gecko/20100101 Thunderbird/102.2.1 Subject: Re: [PATCH] arm64: Remove d-cache clean operation at preserve_boot_args(). Content-Language: en-GB To: Jeungwoo Yoo , Catalin Marinas , Will Deacon Cc: Hyeonggon Yoo <42.hyeyoo@gmail.com>, Gwan-gyeong Mun , Sangyun Kim , linux-arm-kernel@lists.infradead.org, linux-kernel@vger.kernel.org References: <20220904193020.65556-1-casionwoo@gmail.com> From: Robin Murphy In-Reply-To: <20220904193020.65556-1-casionwoo@gmail.com> X-CRM114-Version: 20100106-BlameMichelson ( TRE 0.8.0 (BSD) ) MR-646709E3 X-CRM114-CacheID: sfid-20220905_114704_359565_09323158 X-CRM114-Status: GOOD ( 24.88 ) X-BeenThere: linux-arm-kernel@lists.infradead.org X-Mailman-Version: 2.1.34 Precedence: list List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Transfer-Encoding: 7bit Content-Type: text/plain; charset="us-ascii"; Format="flowed" Sender: "linux-arm-kernel" Errors-To: linux-arm-kernel-bounces+linux-arm-kernel=archiver.kernel.org@lists.infradead.org On 2022-09-04 20:30, Jeungwoo Yoo wrote: > Kernel expects only the clean operation as a booting requirement in > arm64 architecture [1], therefore, the kernel has to invalidate any > cache entries after accessing a memory in the booting time (before > enabling D-cache and MMU) not to overwrite the memory with the stale > cache entry. > > Same applied in preserve_boot_args(), kernel saves boot arguments into > 'boot_args' and invalidates the corresponding cache entry. However, > according to the 'dcache_inval_poc()' implementation, the cache entry > will be not only invalidated but also cleaned. That means if there is a > stale cache entry corresponding to the address of the 'boot_args', the > saved boot arguments in 'boot_args' will be overwritten by the stale > cache entry. Therefore, it uses 'dv ivac' instruction directly instead > of calling 'dcache_inval_poc()'. You've already said in the first paragraph that we expect these locations to be clean. Clean lines are not written back, so your reasoning here is spurious. If boot_args has somehow become dirtied such that the clean operation *would* write back to memory, that can only mean one of two things: either the kernel image was not cleaned per the boot protocol, in which case there's every chance that other things will also go wrong elsewhere and there's not much we can do, or the prior stores hit in the cache (either unexpectedly or because the MMU was left on), in which case we almost certainly *would* want writeback here anyway. > The address of the 'boot_args' is aligned to the cache line size and the > size of 'boot_args' is 32 byte (8 byte * 4), therefore, a single > invalidate operation is enough to invalidate the cache line belonging to > the 'boot_args'. The architecture allows the CWG to be as small as 2 words, so this is clearly untrue. Thanks, Robin. > Sometimes clean operation is required not to lose any contents in the > cache entry but not the target of the invalidation. However, in this > case, there is no valid cache entries at a very early booting stage and > preserve_boot_args() is not called by any other (non-primary) CPUs. > Therefore, this invalidation operation will not introduce any problems. > > [1] in Documentation/arm64/booting.rst: > The address range corresponding to the loaded kernel image must be > cleaned to the PoC. > > Co-developed-by: Gwan-gyeong Mun > Signed-off-by: Gwan-gyeong Mun > > Co-developed-by: Sangyun Kim > Signed-off-by: Sangyun Kim > > Signed-off-by: Jeungwoo Yoo > --- > arch/arm64/kernel/head.S | 4 +--- > 1 file changed, 1 insertion(+), 3 deletions(-) > > diff --git a/arch/arm64/kernel/head.S b/arch/arm64/kernel/head.S > index cefe6a73ee54..916227666b07 100644 > --- a/arch/arm64/kernel/head.S > +++ b/arch/arm64/kernel/head.S > @@ -121,9 +121,7 @@ SYM_CODE_START_LOCAL(preserve_boot_args) > > dmb sy // needed before dc ivac with > // MMU off > - > - add x1, x0, #0x20 // 4 x 8 bytes > - b dcache_inval_poc // tail call > + dc ivac, x0 // Invalidate potentially stale cache line > SYM_CODE_END(preserve_boot_args) > > SYM_FUNC_START_LOCAL(clear_page_tables) _______________________________________________ linux-arm-kernel mailing list linux-arm-kernel@lists.infradead.org http://lists.infradead.org/mailman/listinfo/linux-arm-kernel