Received: by 2002:a05:6a10:9848:0:0:0:0 with SMTP id x8csp241241pxf; Wed, 10 Mar 2021 05:27:46 -0800 (PST) X-Google-Smtp-Source: ABdhPJySsXdGtwggmOfpGtCrZ86ywq5tbC5TjWNoMBW1PMfGIUAH/dP/vbvUurPfCGCzZmZYa0AU X-Received: by 2002:a17:907:3e9e:: with SMTP id hs30mr3752126ejc.66.1615382866801; Wed, 10 Mar 2021 05:27:46 -0800 (PST) ARC-Seal: i=1; a=rsa-sha256; t=1615382866; cv=none; d=google.com; s=arc-20160816; b=nnzmaO+7Ck3ydCzh0TPti+8mZ6sC4tbwhsgWRPdg16h1lqb7U+jPUw2gu/RLIqFgWj W+Tf6NpQ2hTtyNw7YCNtep5JVcMjmhAxgiestENTDsm6+s/2zv8exwehU3F1o5UHvzvP vfbp903d1mivbeV4uZewqomQqhcKH6uchMyobjdJLfqEl8BPZiNtgEkzKalrrbMEm94H Kp57Ekr1aIC7j0YIl4H7guV0g88vwuPt4sLj1RAfBJZUhubeFYxbHkteDS1vqMERwQ5G xJFvMhgsXllNkejVQpX+zjGPwK3z2ZXZExsYV/NrqrV/rOIGqiGb5U1LpR3sM0MUT090 Wowg== ARC-Message-Signature: i=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=arc-20160816; h=list-id:precedence:content-transfer-encoding:mime-version :user-agent:references:in-reply-to:message-id:date:subject:cc:to :from:dkim-signature; bh=9DP66zrLkEkZjt7qrZvtCMw2UqvVHKiaGLPTzqRi2Ho=; b=JNh1pyBI0cdqZMLaZ+9yM0FCs0G7gxaQZ00AvMNk9wkLKvdbD7V4C/M+2qT3UMzdxu HH8aVdrEiGCcvTnmewn5EQHnjF3faFLnAFVvbFD2slvbtH47eifkh+S8XEvQ4GUhszFk 385LkflixfZJv8z8PRGt0NbFXA974Kz9XCp2PModQ+rDrGd0eNpDbyA+QpSXqXXAl+XD LJHg/BTRfjVfBuJKtP2dQbYmhGaeZP0CekWvu4C7zj6OhFi3TLRDq6JBEXGV53Tso/ra hNYGF8kFuoLsnU2qdjPXGZvLnN638Ibv5HUenOs2JL/4w+MErPK7cATIrlEHdh3cW9Sq roHQ== ARC-Authentication-Results: i=1; mx.google.com; dkim=pass header.i=@linuxfoundation.org header.s=korg header.b=WCm+DRYV; spf=pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.18 as permitted sender) smtp.mailfrom=linux-kernel-owner@vger.kernel.org; dmarc=pass (p=NONE sp=NONE dis=NONE) header.from=linuxfoundation.org Return-Path: Received: from vger.kernel.org (vger.kernel.org. [23.128.96.18]) by mx.google.com with ESMTP id cy27si11635822edb.529.2021.03.10.05.27.21; Wed, 10 Mar 2021 05:27:46 -0800 (PST) Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.18 as permitted sender) client-ip=23.128.96.18; Authentication-Results: mx.google.com; dkim=pass header.i=@linuxfoundation.org header.s=korg header.b=WCm+DRYV; spf=pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.18 as permitted sender) smtp.mailfrom=linux-kernel-owner@vger.kernel.org; dmarc=pass (p=NONE sp=NONE dis=NONE) header.from=linuxfoundation.org Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S233008AbhCJN0i (ORCPT + 99 others); Wed, 10 Mar 2021 08:26:38 -0500 Received: from mail.kernel.org ([198.145.29.99]:46278 "EHLO mail.kernel.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S233070AbhCJNYv (ORCPT ); Wed, 10 Mar 2021 08:24:51 -0500 Received: by mail.kernel.org (Postfix) with ESMTPSA id ACD7E64FFC; Wed, 10 Mar 2021 13:24:49 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=linuxfoundation.org; s=korg; t=1615382690; bh=5NjXXHsTx+pFKNRxYqkiO8VGzzFU8zfb09edpkYm5+A=; h=From:To:Cc:Subject:Date:In-Reply-To:References:From; b=WCm+DRYVyXnJ66q0ZrAzjF3Sr+F8lSqd/g/8kNBYb3W6aI5gf1FHj6x8lOT3J8WAA jqyDxIDhgablgC7RHkJD+HWx5xcryFOLop7bPxUWMOvaTEwsYe60Tvt1rR2XFFXdRr pqAuL2BYRFI60jrUl9wQ9NLdQRTc57CzE9SmFggA= From: gregkh@linuxfoundation.org To: linux-kernel@vger.kernel.org Cc: Greg Kroah-Hartman , stable@vger.kernel.org, Andrey Ryabinin , Will Deacon , Joerg Roedel Subject: [PATCH 5.4 06/24] iommu/amd: Fix sleeping in atomic in increase_address_space() Date: Wed, 10 Mar 2021 14:24:18 +0100 Message-Id: <20210310132320.741293325@linuxfoundation.org> X-Mailer: git-send-email 2.30.2 In-Reply-To: <20210310132320.550932445@linuxfoundation.org> References: <20210310132320.550932445@linuxfoundation.org> User-Agent: quilt/0.66 MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Precedence: bulk List-ID: X-Mailing-List: linux-kernel@vger.kernel.org From: Greg Kroah-Hartman From: Andrey Ryabinin commit 140456f994195b568ecd7fc2287a34eadffef3ca upstream. increase_address_space() calls get_zeroed_page(gfp) under spin_lock with disabled interrupts. gfp flags passed to increase_address_space() may allow sleeping, so it comes to this: BUG: sleeping function called from invalid context at mm/page_alloc.c:4342 in_atomic(): 1, irqs_disabled(): 1, pid: 21555, name: epdcbbf1qnhbsd8 Call Trace: dump_stack+0x66/0x8b ___might_sleep+0xec/0x110 __alloc_pages_nodemask+0x104/0x300 get_zeroed_page+0x15/0x40 iommu_map_page+0xdd/0x3e0 amd_iommu_map+0x50/0x70 iommu_map+0x106/0x220 vfio_iommu_type1_ioctl+0x76e/0x950 [vfio_iommu_type1] do_vfs_ioctl+0xa3/0x6f0 ksys_ioctl+0x66/0x70 __x64_sys_ioctl+0x16/0x20 do_syscall_64+0x4e/0x100 entry_SYSCALL_64_after_hwframe+0x44/0xa9 Fix this by moving get_zeroed_page() out of spin_lock/unlock section. Fixes: 754265bcab ("iommu/amd: Fix race in increase_address_space()") Signed-off-by: Andrey Ryabinin Acked-by: Will Deacon Cc: Link: https://lore.kernel.org/r/20210217143004.19165-1-arbn@yandex-team.com Signed-off-by: Joerg Roedel Signed-off-by: Andrey Ryabinin Signed-off-by: Greg Kroah-Hartman --- drivers/iommu/amd_iommu.c | 10 ++++++---- 1 file changed, 6 insertions(+), 4 deletions(-) --- a/drivers/iommu/amd_iommu.c +++ b/drivers/iommu/amd_iommu.c @@ -1469,25 +1469,27 @@ static bool increase_address_space(struc bool ret = false; u64 *pte; + pte = (void *)get_zeroed_page(gfp); + if (!pte) + return false; + spin_lock_irqsave(&domain->lock, flags); if (address <= PM_LEVEL_SIZE(domain->mode) || WARN_ON_ONCE(domain->mode == PAGE_MODE_6_LEVEL)) goto out; - pte = (void *)get_zeroed_page(gfp); - if (!pte) - goto out; - *pte = PM_LEVEL_PDE(domain->mode, iommu_virt_to_phys(domain->pt_root)); domain->pt_root = pte; domain->mode += 1; + pte = NULL; ret = true; out: spin_unlock_irqrestore(&domain->lock, flags); + free_page((unsigned long)pte); return ret; }