From nobody Sun Feb 8 16:31:13 2026 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 558FCC7EE25 for ; Fri, 9 Jun 2023 18:39:17 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S230016AbjFISis (ORCPT ); Fri, 9 Jun 2023 14:38:48 -0400 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:50746 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S231938AbjFIShz (ORCPT ); Fri, 9 Jun 2023 14:37:55 -0400 Received: from mga17.intel.com (mga17.intel.com [192.55.52.151]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id A4E583AB2 for ; Fri, 9 Jun 2023 11:37:31 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=intel.com; i=@intel.com; q=dns/txt; s=Intel; t=1686335851; x=1717871851; h=from:to:cc:subject:date:message-id:in-reply-to: references:mime-version:content-transfer-encoding; bh=cSHO1BH7WnTgrrvVGbc7GV85jY3NXc9iqn2mc1vIguw=; b=RRK7GH/x94E81wUK/rhlA+HweDkoCEaS8Eo5/mTas+aKSqcnuat1YC8D fqXRGsWKyNdrRbMJec4yMpeib+rtWRlr3GgJ1VU944dZlqgyhZt1S0dPm Egn/XhyUp/B+zuiXlFeyzIYdVtgUx/kNWtPKb6pvFkf7FZlsxDOjFrKab 4Oe7B1MFQxuaF1L9m2MxQuj1jkbcmkl96lI7xUI5RK5o8+el2o0VY8fT5 hfh+1+3jWskUiJXm4UoTGOMub6e95SZDPdPfxcZH7p/cyirRPZ7isDZFs wL+Ph8BUIAPujjkqKJOwvFKSP/1IMmVEWUl0e/q5guTginc5TVVfX6MTs g==; X-IronPort-AV: E=McAfee;i="6600,9927,10736"; a="338022230" X-IronPort-AV: E=Sophos;i="6.00,230,1681196400"; d="scan'208";a="338022230" Received: from orsmga002.jf.intel.com ([10.7.209.21]) by fmsmga107.fm.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 09 Jun 2023 11:37:25 -0700 X-ExtLoop1: 1 X-IronPort-AV: E=McAfee;i="6600,9927,10736"; a="710444135" X-IronPort-AV: E=Sophos;i="6.00,230,1681196400"; d="scan'208";a="710444135" Received: from black.fi.intel.com (HELO black.fi.intel.com.) ([10.237.72.28]) by orsmga002.jf.intel.com with ESMTP; 09 Jun 2023 11:37:22 -0700 From: Alexander Shishkin To: linux-kernel@vger.kernel.org, x86@kernel.org, Andy Lutomirski , Dave Hansen , Ravi Shankar , Tony Luck , Sohil Mehta Cc: Alexander Shishkin Subject: [RFC v3 12/12] x86/efi: Disable LASS enforcement when switching to EFI MM Date: Fri, 9 Jun 2023 21:36:32 +0300 Message-Id: <20230609183632.48706-13-alexander.shishkin@linux.intel.com> X-Mailer: git-send-email 2.39.2 In-Reply-To: <20230609183632.48706-1-alexander.shishkin@linux.intel.com> References: <20230609183632.48706-1-alexander.shishkin@linux.intel.com> MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable Precedence: bulk List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Type: text/plain; charset="utf-8" From: Sohil Mehta [Code is experimental and not yet ready to be merged upstream] PeterZ suggested that EFI memory can be mapped in user virtual address space which would trigger LASS violation upon access. It isn't exactly clear how and when these user address mapping happen. It may be possible this is related to EFI mixed mode. Link:https://lore.kernel.org/lkml/Y73S56t%2FwDIGEPlK@hirez.programming.kick= s-ass.net/ stac()/clac() calls in the EFI MM enter and exit functions trigger objtool warnings due to switch_mm() not being classified as func_uaccess_safe. Refer Objtool warnings section #9 in the document tools/objtool/Documentation/objtool.txt. This would need to be resolved before even considering merging. Signed-off-by: Sohil Mehta Signed-off-by: Alexander Shishkin --- arch/x86/platform/efi/efi_64.c | 6 ++++++ 1 file changed, 6 insertions(+) diff --git a/arch/x86/platform/efi/efi_64.c b/arch/x86/platform/efi/efi_64.c index 232acf418cfb..20966efcd87a 100644 --- a/arch/x86/platform/efi/efi_64.c +++ b/arch/x86/platform/efi/efi_64.c @@ -473,9 +473,14 @@ void __init efi_dump_pagetable(void) * while the EFI-mm is borrowed. mmgrab()/mmdrop() is not used because the= mm * can not change under us. * It should be ensured that there are no concurrent calls to this functio= n. + * + * Disable LASS enforcement temporarily when switching to EFI MM since it = could + * be mapped into the low 64-bit virtual address space with address bit 63= set + * to 0. */ void efi_enter_mm(void) { + stac(); efi_prev_mm =3D current->active_mm; current->active_mm =3D &efi_mm; switch_mm(efi_prev_mm, &efi_mm, NULL); @@ -485,6 +490,7 @@ void efi_leave_mm(void) { current->active_mm =3D efi_prev_mm; switch_mm(&efi_mm, efi_prev_mm, NULL); + clac(); } =20 static DEFINE_SPINLOCK(efi_runtime_lock); --=20 2.39.2