From nobody Sat Apr 11 06:41:05 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 41DB2C25B0E for ; Mon, 15 Aug 2022 04:15:09 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S240370AbiHOEPH (ORCPT ); Mon, 15 Aug 2022 00:15:07 -0400 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:43504 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S230387AbiHOEPG (ORCPT ); Mon, 15 Aug 2022 00:15:06 -0400 Received: from mga12.intel.com (mga12.intel.com [192.55.52.136]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id 2CDBB13E3D for ; Sun, 14 Aug 2022 21:15:05 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=intel.com; i=@intel.com; q=dns/txt; s=Intel; t=1660536905; x=1692072905; h=from:to:cc:subject:date:message-id:in-reply-to: references:mime-version:content-transfer-encoding; bh=4tAN31RYYpJnhaGsltGsp329SN2LQE0u8p1NIAAYm9M=; b=ZhV6y7t1IlR2Dwwsss1dodkx5I4MK46JKRskNqzHDNQchF4mtpefeQyV DGRDh/2zX7KpcgRWP2Sz4aNwErDD98H6w4uHEw/imgdibEv0IvCfqGws4 bg7hesf7mBGYQOZCtO2d2tczRT4C4++l0/16QITkHtfUw3CAz7e1qrhE3 OquDbEnxp84OcvibZz4ipyUm0px2wpziXSRIl1lh3SI1P9dzAiDt2dyOi eHjg5I3u1PpY9S+cZN7LkpmJLdO0UsU8G5EoN2h5Dc35KjDWwkoRUI+49 JxWFa6zWE/TKM1NXA7FBapt8UtAXcm89YNJvG7tGakWf0Q1Mej3fokuna A==; X-IronPort-AV: E=McAfee;i="6400,9594,10439"; a="271658890" X-IronPort-AV: E=Sophos;i="5.93,237,1654585200"; d="scan'208";a="271658890" Received: from fmsmga005.fm.intel.com ([10.253.24.32]) by fmsmga106.fm.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 14 Aug 2022 21:15:04 -0700 X-IronPort-AV: E=Sophos;i="5.93,237,1654585200"; d="scan'208";a="934354071" Received: from dmalka-mobl.ger.corp.intel.com (HELO box.shutemov.name) ([10.249.42.26]) by fmsmga005-auth.fm.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 14 Aug 2022 21:15:01 -0700 Received: by box.shutemov.name (Postfix, from userid 1000) id 385CE104A64; Mon, 15 Aug 2022 07:18:04 +0300 (+03) From: "Kirill A. Shutemov" To: Dave Hansen , Andy Lutomirski , Peter Zijlstra Cc: x86@kernel.org, Kostya Serebryany , Andrey Ryabinin , Andrey Konovalov , Alexander Potapenko , Taras Madan , Dmitry Vyukov , "H . J . Lu" , Andi Kleen , Rick Edgecombe , linux-mm@kvack.org, linux-kernel@vger.kernel.org, "Kirill A. Shutemov" Subject: [PATCHv6 01/11] x86/mm: Fix CR3_ADDR_MASK Date: Mon, 15 Aug 2022 07:17:53 +0300 Message-Id: <20220815041803.17954-2-kirill.shutemov@linux.intel.com> X-Mailer: git-send-email 2.35.1 In-Reply-To: <20220815041803.17954-1-kirill.shutemov@linux.intel.com> References: <20220815041803.17954-1-kirill.shutemov@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" The mask must not include bits above physical address mask. These bits are reserved and can be used for other things. Bits 61 and 62 are used for Linear Address Masking. Signed-off-by: Kirill A. Shutemov Reviewed-by: Rick Edgecombe Reviewed-by: Alexander Potapenko Tested-by: Alexander Potapenko Acked-by: Peter Zijlstra (Intel) --- arch/x86/include/asm/processor-flags.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/x86/include/asm/processor-flags.h b/arch/x86/include/asm/= processor-flags.h index 02c2cbda4a74..a7f3d9100adb 100644 --- a/arch/x86/include/asm/processor-flags.h +++ b/arch/x86/include/asm/processor-flags.h @@ -35,7 +35,7 @@ */ #ifdef CONFIG_X86_64 /* Mask off the address space ID and SME encryption bits. */ -#define CR3_ADDR_MASK __sme_clr(0x7FFFFFFFFFFFF000ull) +#define CR3_ADDR_MASK __sme_clr(PHYSICAL_PAGE_MASK) #define CR3_PCID_MASK 0xFFFull #define CR3_NOFLUSH BIT_ULL(63) =20 --=20 2.35.1 From nobody Sat Apr 11 06:41:05 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 D5F4DC25B0E for ; Mon, 15 Aug 2022 04:15:11 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S240415AbiHOEPJ (ORCPT ); Mon, 15 Aug 2022 00:15:09 -0400 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:43512 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S232332AbiHOEPG (ORCPT ); Mon, 15 Aug 2022 00:15:06 -0400 Received: from mga04.intel.com (mga04.intel.com [192.55.52.120]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id 9AEA01401B for ; Sun, 14 Aug 2022 21:15:05 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=intel.com; i=@intel.com; q=dns/txt; s=Intel; t=1660536905; x=1692072905; h=from:to:cc:subject:date:message-id:in-reply-to: references:mime-version:content-transfer-encoding; bh=+Rvd/xrgf0B0RTbzc+Lnz5LTd3QpoAfiklT1FBTqlT4=; b=NbM/OlfKOabkmUR3228TVR3y+/OFryVXBVfXs9M74xP8V6YfEGhCakf5 77CehFhJel2VDcDoWXBycCI1akVdafE9wZNkEqKZ8mPhhFyCpaCnmGRyu 1WEx85nc2xDbQBFReJoQld1rWua99awY96yoZW6quX5TSkA1A8j6ulSt9 grvqwLidls0KqUWSzhhC+yEEWZTELbTxB5MpikyUbRhZ8cITTcQC9H9Sn /LP4aHjdHP9aZ0nAgzt6zMir7pFhDSDuB9tf0SNOwvL/qNW5m3tR1d4s/ PBhlqZvxVtMWGLWxmHHHbW5Mhp2mZhzb+lkMR7kd+gLOmYac1i8R+yoVA g==; X-IronPort-AV: E=McAfee;i="6400,9594,10439"; a="290639847" X-IronPort-AV: E=Sophos;i="5.93,237,1654585200"; d="scan'208";a="290639847" Received: from orsmga002.jf.intel.com ([10.7.209.21]) by fmsmga104.fm.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 14 Aug 2022 21:15:05 -0700 X-IronPort-AV: E=Sophos;i="5.93,237,1654585200"; d="scan'208";a="606530103" Received: from dmalka-mobl.ger.corp.intel.com (HELO box.shutemov.name) ([10.249.42.26]) by orsmga002-auth.jf.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 14 Aug 2022 21:15:01 -0700 Received: by box.shutemov.name (Postfix, from userid 1000) id 413D1104A65; Mon, 15 Aug 2022 07:18:04 +0300 (+03) From: "Kirill A. Shutemov" To: Dave Hansen , Andy Lutomirski , Peter Zijlstra Cc: x86@kernel.org, Kostya Serebryany , Andrey Ryabinin , Andrey Konovalov , Alexander Potapenko , Taras Madan , Dmitry Vyukov , "H . J . Lu" , Andi Kleen , Rick Edgecombe , linux-mm@kvack.org, linux-kernel@vger.kernel.org, "Kirill A. Shutemov" Subject: [PATCHv6 02/11] x86: CPUID and CR3/CR4 flags for Linear Address Masking Date: Mon, 15 Aug 2022 07:17:54 +0300 Message-Id: <20220815041803.17954-3-kirill.shutemov@linux.intel.com> X-Mailer: git-send-email 2.35.1 In-Reply-To: <20220815041803.17954-1-kirill.shutemov@linux.intel.com> References: <20220815041803.17954-1-kirill.shutemov@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" Enumerate Linear Address Masking and provide defines for CR3 and CR4 flags. Signed-off-by: Kirill A. Shutemov Reviewed-by: Alexander Potapenko Tested-by: Alexander Potapenko Acked-by: Peter Zijlstra (Intel) --- arch/x86/include/asm/cpufeatures.h | 1 + arch/x86/include/asm/processor-flags.h | 2 ++ arch/x86/include/uapi/asm/processor-flags.h | 6 ++++++ 3 files changed, 9 insertions(+) diff --git a/arch/x86/include/asm/cpufeatures.h b/arch/x86/include/asm/cpuf= eatures.h index 235dc85c91c3..73c0cf5bd8a1 100644 --- a/arch/x86/include/asm/cpufeatures.h +++ b/arch/x86/include/asm/cpufeatures.h @@ -308,6 +308,7 @@ /* Intel-defined CPU features, CPUID level 0x00000007:1 (EAX), word 12 */ #define X86_FEATURE_AVX_VNNI (12*32+ 4) /* AVX VNNI instructions */ #define X86_FEATURE_AVX512_BF16 (12*32+ 5) /* AVX512 BFLOAT16 instruction= s */ +#define X86_FEATURE_LAM (12*32+26) /* Linear Address Masking */ =20 /* AMD-defined CPU features, CPUID level 0x80000008 (EBX), word 13 */ #define X86_FEATURE_CLZERO (13*32+ 0) /* CLZERO instruction */ diff --git a/arch/x86/include/asm/processor-flags.h b/arch/x86/include/asm/= processor-flags.h index a7f3d9100adb..d8cccadc83a6 100644 --- a/arch/x86/include/asm/processor-flags.h +++ b/arch/x86/include/asm/processor-flags.h @@ -28,6 +28,8 @@ * On systems with SME, one bit (in a variable position!) is stolen to ind= icate * that the top-level paging structure is encrypted. * + * On systemms with LAM, bits 61 and 62 are used to indicate LAM mode. + * * All of the remaining bits indicate the physical address of the top-level * paging structure. * diff --git a/arch/x86/include/uapi/asm/processor-flags.h b/arch/x86/include= /uapi/asm/processor-flags.h index c47cc7f2feeb..d898432947ff 100644 --- a/arch/x86/include/uapi/asm/processor-flags.h +++ b/arch/x86/include/uapi/asm/processor-flags.h @@ -82,6 +82,10 @@ #define X86_CR3_PCID_BITS 12 #define X86_CR3_PCID_MASK (_AC((1UL << X86_CR3_PCID_BITS) - 1, UL)) =20 +#define X86_CR3_LAM_U57_BIT 61 /* Activate LAM for userspace, 62:57 bits m= asked */ +#define X86_CR3_LAM_U57 _BITULL(X86_CR3_LAM_U57_BIT) +#define X86_CR3_LAM_U48_BIT 62 /* Activate LAM for userspace, 62:48 bits m= asked */ +#define X86_CR3_LAM_U48 _BITULL(X86_CR3_LAM_U48_BIT) #define X86_CR3_PCID_NOFLUSH_BIT 63 /* Preserve old PCID */ #define X86_CR3_PCID_NOFLUSH _BITULL(X86_CR3_PCID_NOFLUSH_BIT) =20 @@ -132,6 +136,8 @@ #define X86_CR4_PKE _BITUL(X86_CR4_PKE_BIT) #define X86_CR4_CET_BIT 23 /* enable Control-flow Enforcement Technology = */ #define X86_CR4_CET _BITUL(X86_CR4_CET_BIT) +#define X86_CR4_LAM_SUP_BIT 28 /* LAM for supervisor pointers */ +#define X86_CR4_LAM_SUP _BITUL(X86_CR4_LAM_SUP_BIT) =20 /* * x86-64 Task Priority Register, CR8 --=20 2.35.1 From nobody Sat Apr 11 06:41:05 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 27D34C25B06 for ; Mon, 15 Aug 2022 04:15:23 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S240598AbiHOEPV (ORCPT ); Mon, 15 Aug 2022 00:15:21 -0400 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:43530 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S240369AbiHOEPH (ORCPT ); Mon, 15 Aug 2022 00:15:07 -0400 Received: from mga06.intel.com (mga06b.intel.com [134.134.136.31]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id C3E4314022 for ; Sun, 14 Aug 2022 21:15:05 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=intel.com; i=@intel.com; q=dns/txt; s=Intel; t=1660536905; x=1692072905; h=from:to:cc:subject:date:message-id:in-reply-to: references:mime-version:content-transfer-encoding; bh=SM0U3bqgSaP76tIugfGQ0W5Ix7XtZZL6qmL0RNE/wKI=; b=KVZ/N7S8bW5MW2t+MwHNLzA6HbG3nIuYKxP7gceGMcqHdOQxUkmYGjRY bA5t7CcJLjgQzSO66HdJPk6JJV4bnTHa+r4izjmcanBOItcn40YbguR1N 45ZqBkpBYCm172yIThJgPHIAHxvlyUZJRlPv2aPqV09p5BZc0AX9muryV xnJLrSsISJLoV7/297YTB+uTRYFvkmd2QIJlgZwtodbc8cRO+sMMknsza 1FJoktBv6KYMXp8hzAKvEKy9m6y6+lvZYU60+YghcoclsxbGKoggtu5VQ lOElAgHoGaqnfeAqoidqhJdZq4BOdmqb7xgIi3Gsp9QkZA2qG2pYq8fI1 A==; X-IronPort-AV: E=McAfee;i="6400,9594,10439"; a="353628394" X-IronPort-AV: E=Sophos;i="5.93,237,1654585200"; d="scan'208";a="353628394" Received: from fmsmga001.fm.intel.com ([10.253.24.23]) by orsmga104.jf.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 14 Aug 2022 21:15:05 -0700 X-IronPort-AV: E=Sophos;i="5.93,237,1654585200"; d="scan'208";a="748800368" Received: from dmalka-mobl.ger.corp.intel.com (HELO box.shutemov.name) ([10.249.42.26]) by fmsmga001-auth.fm.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 14 Aug 2022 21:15:01 -0700 Received: by box.shutemov.name (Postfix, from userid 1000) id 4BC9E104A66; Mon, 15 Aug 2022 07:18:04 +0300 (+03) From: "Kirill A. Shutemov" To: Dave Hansen , Andy Lutomirski , Peter Zijlstra Cc: x86@kernel.org, Kostya Serebryany , Andrey Ryabinin , Andrey Konovalov , Alexander Potapenko , Taras Madan , Dmitry Vyukov , "H . J . Lu" , Andi Kleen , Rick Edgecombe , linux-mm@kvack.org, linux-kernel@vger.kernel.org, "Kirill A. Shutemov" Subject: [PATCHv6 03/11] mm: Pass down mm_struct to untagged_addr() Date: Mon, 15 Aug 2022 07:17:55 +0300 Message-Id: <20220815041803.17954-4-kirill.shutemov@linux.intel.com> X-Mailer: git-send-email 2.35.1 In-Reply-To: <20220815041803.17954-1-kirill.shutemov@linux.intel.com> References: <20220815041803.17954-1-kirill.shutemov@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" Intel Linear Address Masking (LAM) brings per-mm untagging rules. Pass down mm_struct to the untagging helper. It will help to apply untagging policy correctly. In most cases, current->mm is the one to use, but there are some exceptions, such as get_user_page_remote(). Move dummy implementation of untagged_addr() from to . can override the implementation. Moving the dummy header outside helps to avoid header hell if you need to defer mm_struct within the helper. Signed-off-by: Kirill A. Shutemov Reviewed-by: Rick Edgecombe Reviewed-by: Alexander Potapenko Tested-by: Alexander Potapenko Acked-by: Peter Zijlstra (Intel) --- arch/arm64/include/asm/memory.h | 4 ++-- arch/arm64/include/asm/signal.h | 2 +- arch/arm64/include/asm/uaccess.h | 4 ++-- arch/arm64/kernel/hw_breakpoint.c | 2 +- arch/arm64/kernel/traps.c | 4 ++-- arch/arm64/mm/fault.c | 10 +++++----- arch/sparc/include/asm/pgtable_64.h | 2 +- arch/sparc/include/asm/uaccess_64.h | 2 ++ drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c | 2 +- drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c | 2 +- drivers/gpu/drm/radeon/radeon_gem.c | 2 +- drivers/infiniband/hw/mlx4/mr.c | 2 +- drivers/media/common/videobuf2/frame_vector.c | 2 +- drivers/media/v4l2-core/videobuf-dma-contig.c | 2 +- drivers/staging/media/atomisp/pci/hmm/hmm_bo.c | 2 +- drivers/tee/tee_shm.c | 2 +- drivers/vfio/vfio_iommu_type1.c | 2 +- fs/proc/task_mmu.c | 2 +- include/linux/mm.h | 11 ----------- include/linux/uaccess.h | 15 +++++++++++++++ lib/strncpy_from_user.c | 2 +- lib/strnlen_user.c | 2 +- mm/gup.c | 6 +++--- mm/madvise.c | 2 +- mm/mempolicy.c | 6 +++--- mm/migrate.c | 2 +- mm/mincore.c | 2 +- mm/mlock.c | 4 ++-- mm/mmap.c | 2 +- mm/mprotect.c | 2 +- mm/mremap.c | 2 +- mm/msync.c | 2 +- virt/kvm/kvm_main.c | 2 +- 33 files changed, 59 insertions(+), 53 deletions(-) diff --git a/arch/arm64/include/asm/memory.h b/arch/arm64/include/asm/memor= y.h index 9dd08cd339c3..5b24ef93c6b9 100644 --- a/arch/arm64/include/asm/memory.h +++ b/arch/arm64/include/asm/memory.h @@ -227,8 +227,8 @@ static inline unsigned long kaslr_offset(void) #define __untagged_addr(addr) \ ((__force __typeof__(addr))sign_extend64((__force u64)(addr), 55)) =20 -#define untagged_addr(addr) ({ \ - u64 __addr =3D (__force u64)(addr); \ +#define untagged_addr(mm, addr) ({ \ + u64 __addr =3D (__force u64)(addr); \ __addr &=3D __untagged_addr(__addr); \ (__force __typeof__(addr))__addr; \ }) diff --git a/arch/arm64/include/asm/signal.h b/arch/arm64/include/asm/signa= l.h index ef449f5f4ba8..0899c355c398 100644 --- a/arch/arm64/include/asm/signal.h +++ b/arch/arm64/include/asm/signal.h @@ -18,7 +18,7 @@ static inline void __user *arch_untagged_si_addr(void __u= ser *addr, if (sig =3D=3D SIGTRAP && si_code =3D=3D TRAP_BRKPT) return addr; =20 - return untagged_addr(addr); + return untagged_addr(current->mm, addr); } #define arch_untagged_si_addr arch_untagged_si_addr =20 diff --git a/arch/arm64/include/asm/uaccess.h b/arch/arm64/include/asm/uacc= ess.h index 2fc9f0861769..0e17f44cf997 100644 --- a/arch/arm64/include/asm/uaccess.h +++ b/arch/arm64/include/asm/uaccess.h @@ -44,7 +44,7 @@ static inline int access_ok(const void __user *addr, unsi= gned long size) */ if (IS_ENABLED(CONFIG_ARM64_TAGGED_ADDR_ABI) && (current->flags & PF_KTHREAD || test_thread_flag(TIF_TAGGED_ADDR))) - addr =3D untagged_addr(addr); + addr =3D untagged_addr(current->mm, addr); =20 return likely(__access_ok(addr, size)); } @@ -217,7 +217,7 @@ static inline void __user *__uaccess_mask_ptr(const voi= d __user *ptr) " csel %0, %1, xzr, eq\n" : "=3D&r" (safe_ptr) : "r" (ptr), "r" (TASK_SIZE_MAX - 1), - "r" (untagged_addr(ptr)) + "r" (untagged_addr(current->mm, ptr)) : "cc"); =20 csdb(); diff --git a/arch/arm64/kernel/hw_breakpoint.c b/arch/arm64/kernel/hw_break= point.c index b29a311bb055..d637cee7b771 100644 --- a/arch/arm64/kernel/hw_breakpoint.c +++ b/arch/arm64/kernel/hw_breakpoint.c @@ -715,7 +715,7 @@ static u64 get_distance_from_watchpoint(unsigned long a= ddr, u64 val, u64 wp_low, wp_high; u32 lens, lene; =20 - addr =3D untagged_addr(addr); + addr =3D untagged_addr(current->mm, addr); =20 lens =3D __ffs(ctrl->len); lene =3D __fls(ctrl->len); diff --git a/arch/arm64/kernel/traps.c b/arch/arm64/kernel/traps.c index b7fed33981f7..9edef0e155b6 100644 --- a/arch/arm64/kernel/traps.c +++ b/arch/arm64/kernel/traps.c @@ -476,7 +476,7 @@ void arm64_notify_segfault(unsigned long addr) int code; =20 mmap_read_lock(current->mm); - if (find_vma(current->mm, untagged_addr(addr)) =3D=3D NULL) + if (find_vma(current->mm, untagged_addr(current->mm, addr)) =3D=3D NULL) code =3D SEGV_MAPERR; else code =3D SEGV_ACCERR; @@ -540,7 +540,7 @@ static void user_cache_maint_handler(unsigned long esr,= struct pt_regs *regs) int ret =3D 0; =20 tagged_address =3D pt_regs_read_reg(regs, rt); - address =3D untagged_addr(tagged_address); + address =3D untagged_addr(current->mm, tagged_address); =20 switch (crm) { case ESR_ELx_SYS64_ISS_CRM_DC_CVAU: /* DC CVAU, gets promoted */ diff --git a/arch/arm64/mm/fault.c b/arch/arm64/mm/fault.c index c33f1fad2745..1fa0f1166ac0 100644 --- a/arch/arm64/mm/fault.c +++ b/arch/arm64/mm/fault.c @@ -454,7 +454,7 @@ static void set_thread_esr(unsigned long address, unsig= ned long esr) static void do_bad_area(unsigned long far, unsigned long esr, struct pt_regs *regs) { - unsigned long addr =3D untagged_addr(far); + unsigned long addr =3D untagged_addr(current->mm, far); =20 /* * If we are in kernel mode at this point, we have no context to @@ -524,7 +524,7 @@ static int __kprobes do_page_fault(unsigned long far, u= nsigned long esr, vm_fault_t fault; unsigned long vm_flags; unsigned int mm_flags =3D FAULT_FLAG_DEFAULT; - unsigned long addr =3D untagged_addr(far); + unsigned long addr =3D untagged_addr(mm, far); =20 if (kprobe_page_fault(regs, esr)) return 0; @@ -679,7 +679,7 @@ static int __kprobes do_translation_fault(unsigned long= far, unsigned long esr, struct pt_regs *regs) { - unsigned long addr =3D untagged_addr(far); + unsigned long addr =3D untagged_addr(current->mm, far); =20 if (is_ttbr0_addr(addr)) return do_page_fault(far, esr, regs); @@ -723,7 +723,7 @@ static int do_sea(unsigned long far, unsigned long esr,= struct pt_regs *regs) * UNKNOWN for synchronous external aborts. Mask them out now * so that userspace doesn't see them. */ - siaddr =3D untagged_addr(far); + siaddr =3D untagged_addr(current->mm, far); } arm64_notify_die(inf->name, regs, inf->sig, inf->code, siaddr, esr); =20 @@ -813,7 +813,7 @@ static const struct fault_info fault_info[] =3D { void do_mem_abort(unsigned long far, unsigned long esr, struct pt_regs *re= gs) { const struct fault_info *inf =3D esr_to_fault_info(esr); - unsigned long addr =3D untagged_addr(far); + unsigned long addr =3D untagged_addr(current->mm, far); =20 if (!inf->fn(far, esr, regs)) return; diff --git a/arch/sparc/include/asm/pgtable_64.h b/arch/sparc/include/asm/p= gtable_64.h index a779418ceba9..aa996ffe5c8c 100644 --- a/arch/sparc/include/asm/pgtable_64.h +++ b/arch/sparc/include/asm/pgtable_64.h @@ -1052,7 +1052,7 @@ static inline unsigned long __untagged_addr(unsigned = long start) =20 return start; } -#define untagged_addr(addr) \ +#define untagged_addr(mm, addr) \ ((__typeof__(addr))(__untagged_addr((unsigned long)(addr)))) =20 static inline bool pte_access_permitted(pte_t pte, bool write) diff --git a/arch/sparc/include/asm/uaccess_64.h b/arch/sparc/include/asm/u= access_64.h index 94266a5c5b04..b825a5dd0210 100644 --- a/arch/sparc/include/asm/uaccess_64.h +++ b/arch/sparc/include/asm/uaccess_64.h @@ -8,8 +8,10 @@ =20 #include #include +#include #include #include +#include =20 #include #include diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c b/drivers/gpu= /drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c index a699134a1e8c..71babd0eb70a 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c @@ -1653,7 +1653,7 @@ int amdgpu_amdkfd_gpuvm_alloc_memory_of_gpu( if (flags & KFD_IOC_ALLOC_MEM_FLAGS_USERPTR) { if (!offset || !*offset) return -EINVAL; - user_addr =3D untagged_addr(*offset); + user_addr =3D untagged_addr(current->mm, *offset); } else if (flags & (KFD_IOC_ALLOC_MEM_FLAGS_DOORBELL | KFD_IOC_ALLOC_MEM_FLAGS_MMIO_REMAP)) { bo_type =3D ttm_bo_type_sg; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c b/drivers/gpu/drm/amd/= amdgpu/amdgpu_gem.c index 8ef31d687ef3..691dfb3f2c0e 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c @@ -382,7 +382,7 @@ int amdgpu_gem_userptr_ioctl(struct drm_device *dev, vo= id *data, uint32_t handle; int r; =20 - args->addr =3D untagged_addr(args->addr); + args->addr =3D untagged_addr(current->mm, args->addr); =20 if (offset_in_page(args->addr | args->size)) return -EINVAL; diff --git a/drivers/gpu/drm/radeon/radeon_gem.c b/drivers/gpu/drm/radeon/r= adeon_gem.c index 261fcbae88d7..cba2f4b19838 100644 --- a/drivers/gpu/drm/radeon/radeon_gem.c +++ b/drivers/gpu/drm/radeon/radeon_gem.c @@ -371,7 +371,7 @@ int radeon_gem_userptr_ioctl(struct drm_device *dev, vo= id *data, uint32_t handle; int r; =20 - args->addr =3D untagged_addr(args->addr); + args->addr =3D untagged_addr(current->mm, args->addr); =20 if (offset_in_page(args->addr | args->size)) return -EINVAL; diff --git a/drivers/infiniband/hw/mlx4/mr.c b/drivers/infiniband/hw/mlx4/m= r.c index 04a67b481608..b2860feeae3c 100644 --- a/drivers/infiniband/hw/mlx4/mr.c +++ b/drivers/infiniband/hw/mlx4/mr.c @@ -379,7 +379,7 @@ static struct ib_umem *mlx4_get_umem_mr(struct ib_devic= e *device, u64 start, * again */ if (!ib_access_writable(access_flags)) { - unsigned long untagged_start =3D untagged_addr(start); + unsigned long untagged_start =3D untagged_addr(current->mm, start); struct vm_area_struct *vma; =20 mmap_read_lock(current->mm); diff --git a/drivers/media/common/videobuf2/frame_vector.c b/drivers/media/= common/videobuf2/frame_vector.c index 542dde9d2609..7e62f7a2555d 100644 --- a/drivers/media/common/videobuf2/frame_vector.c +++ b/drivers/media/common/videobuf2/frame_vector.c @@ -47,7 +47,7 @@ int get_vaddr_frames(unsigned long start, unsigned int nr= _frames, if (WARN_ON_ONCE(nr_frames > vec->nr_allocated)) nr_frames =3D vec->nr_allocated; =20 - start =3D untagged_addr(start); + start =3D untagged_addr(mm, start); =20 ret =3D pin_user_pages_fast(start, nr_frames, FOLL_FORCE | FOLL_WRITE | FOLL_LONGTERM, diff --git a/drivers/media/v4l2-core/videobuf-dma-contig.c b/drivers/media/= v4l2-core/videobuf-dma-contig.c index 52312ce2ba05..a1444f8afa05 100644 --- a/drivers/media/v4l2-core/videobuf-dma-contig.c +++ b/drivers/media/v4l2-core/videobuf-dma-contig.c @@ -157,8 +157,8 @@ static void videobuf_dma_contig_user_put(struct videobu= f_dma_contig_memory *mem) static int videobuf_dma_contig_user_get(struct videobuf_dma_contig_memory = *mem, struct videobuf_buffer *vb) { - unsigned long untagged_baddr =3D untagged_addr(vb->baddr); struct mm_struct *mm =3D current->mm; + unsigned long untagged_baddr =3D untagged_addr(mm, vb->baddr); struct vm_area_struct *vma; unsigned long prev_pfn, this_pfn; unsigned long pages_done, user_address; diff --git a/drivers/staging/media/atomisp/pci/hmm/hmm_bo.c b/drivers/stagi= ng/media/atomisp/pci/hmm/hmm_bo.c index f50494123f03..a43c65950554 100644 --- a/drivers/staging/media/atomisp/pci/hmm/hmm_bo.c +++ b/drivers/staging/media/atomisp/pci/hmm/hmm_bo.c @@ -794,7 +794,7 @@ static int alloc_user_pages(struct hmm_buffer_object *b= o, * and map to user space */ =20 - userptr =3D untagged_addr(userptr); + userptr =3D untagged_addr(current->mm, userptr); =20 if (vma->vm_flags & (VM_IO | VM_PFNMAP)) { page_nr =3D pin_user_pages((unsigned long)userptr, bo->pgnr, diff --git a/drivers/tee/tee_shm.c b/drivers/tee/tee_shm.c index f2b1bcefcadd..386be09cb2cd 100644 --- a/drivers/tee/tee_shm.c +++ b/drivers/tee/tee_shm.c @@ -261,7 +261,7 @@ register_shm_helper(struct tee_context *ctx, unsigned l= ong addr, shm->flags =3D flags; shm->ctx =3D ctx; shm->id =3D id; - addr =3D untagged_addr(addr); + addr =3D untagged_addr(current->mm, addr); start =3D rounddown(addr, PAGE_SIZE); shm->offset =3D addr - start; shm->size =3D length; diff --git a/drivers/vfio/vfio_iommu_type1.c b/drivers/vfio/vfio_iommu_type= 1.c index db516c90a977..1ab5adba6482 100644 --- a/drivers/vfio/vfio_iommu_type1.c +++ b/drivers/vfio/vfio_iommu_type1.c @@ -562,7 +562,7 @@ static int vaddr_get_pfns(struct mm_struct *mm, unsigne= d long vaddr, goto done; } =20 - vaddr =3D untagged_addr(vaddr); + vaddr =3D untagged_addr(mm, vaddr); =20 retry: vma =3D vma_lookup(mm, vaddr); diff --git a/fs/proc/task_mmu.c b/fs/proc/task_mmu.c index a3398d0f1927..9255596b690f 100644 --- a/fs/proc/task_mmu.c +++ b/fs/proc/task_mmu.c @@ -1662,7 +1662,7 @@ static ssize_t pagemap_read(struct file *file, char _= _user *buf, /* watch out for wraparound */ start_vaddr =3D end_vaddr; if (svpfn <=3D (ULONG_MAX >> PAGE_SHIFT)) - start_vaddr =3D untagged_addr(svpfn << PAGE_SHIFT); + start_vaddr =3D untagged_addr(mm, svpfn << PAGE_SHIFT); =20 /* Ensure the address is inside the task */ if (start_vaddr > mm->task_size) diff --git a/include/linux/mm.h b/include/linux/mm.h index 3bedc449c14d..ae2f8c9fbc4d 100644 --- a/include/linux/mm.h +++ b/include/linux/mm.h @@ -95,17 +95,6 @@ extern int mmap_rnd_compat_bits __read_mostly; #include #include =20 -/* - * Architectures that support memory tagging (assigning tags to memory reg= ions, - * embedding these tags into addresses that point to these memory regions,= and - * checking that the memory and the pointer tags match on memory accesses) - * redefine this macro to strip tags from pointers. - * It's defined as noop for architectures that don't support memory taggin= g. - */ -#ifndef untagged_addr -#define untagged_addr(addr) (addr) -#endif - #ifndef __pa_symbol #define __pa_symbol(x) __pa(RELOC_HIDE((unsigned long)(x), 0)) #endif diff --git a/include/linux/uaccess.h b/include/linux/uaccess.h index 47e5d374c7eb..aed9555aed67 100644 --- a/include/linux/uaccess.h +++ b/include/linux/uaccess.h @@ -10,6 +10,21 @@ =20 #include =20 +/* + * Architectures that support memory tagging (assigning tags to memory reg= ions, + * embedding these tags into addresses that point to these memory regions,= and + * checking that the memory and the pointer tags match on memory accesses) + * redefine this macro to strip tags from pointers. + * + * Passing down mm_struct allows to define untagging rules on per-process + * basis. + * + * It's defined as noop for architectures that don't support memory taggin= g. + */ +#ifndef untagged_addr +#define untagged_addr(mm, addr) (addr) +#endif + /* * Architectures should provide two primitives (raw_copy_{to,from}_user()) * and get rid of their private instances of copy_{to,from}_user() and diff --git a/lib/strncpy_from_user.c b/lib/strncpy_from_user.c index 6432b8c3e431..6e1e2aa0c994 100644 --- a/lib/strncpy_from_user.c +++ b/lib/strncpy_from_user.c @@ -121,7 +121,7 @@ long strncpy_from_user(char *dst, const char __user *sr= c, long count) return 0; =20 max_addr =3D TASK_SIZE_MAX; - src_addr =3D (unsigned long)untagged_addr(src); + src_addr =3D (unsigned long)untagged_addr(current->mm, src); if (likely(src_addr < max_addr)) { unsigned long max =3D max_addr - src_addr; long retval; diff --git a/lib/strnlen_user.c b/lib/strnlen_user.c index feeb935a2299..abc096a68f05 100644 --- a/lib/strnlen_user.c +++ b/lib/strnlen_user.c @@ -97,7 +97,7 @@ long strnlen_user(const char __user *str, long count) return 0; =20 max_addr =3D TASK_SIZE_MAX; - src_addr =3D (unsigned long)untagged_addr(str); + src_addr =3D (unsigned long)untagged_addr(current->mm, str); if (likely(src_addr < max_addr)) { unsigned long max =3D max_addr - src_addr; long retval; diff --git a/mm/gup.c b/mm/gup.c index 732825157430..9f2a14f7e77a 100644 --- a/mm/gup.c +++ b/mm/gup.c @@ -1125,7 +1125,7 @@ static long __get_user_pages(struct mm_struct *mm, if (!nr_pages) return 0; =20 - start =3D untagged_addr(start); + start =3D untagged_addr(mm, start); =20 VM_BUG_ON(!!pages !=3D !!(gup_flags & (FOLL_GET | FOLL_PIN))); =20 @@ -1307,7 +1307,7 @@ int fixup_user_fault(struct mm_struct *mm, struct vm_area_struct *vma; vm_fault_t ret; =20 - address =3D untagged_addr(address); + address =3D untagged_addr(mm, address); =20 if (unlocked) fault_flags |=3D FAULT_FLAG_ALLOW_RETRY | FAULT_FLAG_KILLABLE; @@ -2935,7 +2935,7 @@ static int internal_get_user_pages_fast(unsigned long= start, if (!(gup_flags & FOLL_FAST_ONLY)) might_lock_read(¤t->mm->mmap_lock); =20 - start =3D untagged_addr(start) & PAGE_MASK; + start =3D untagged_addr(current->mm, start) & PAGE_MASK; len =3D nr_pages << PAGE_SHIFT; if (check_add_overflow(start, len, &end)) return 0; diff --git a/mm/madvise.c b/mm/madvise.c index 5f0f0948a50e..fb78881329c4 100644 --- a/mm/madvise.c +++ b/mm/madvise.c @@ -1373,7 +1373,7 @@ int do_madvise(struct mm_struct *mm, unsigned long st= art, size_t len_in, int beh size_t len; struct blk_plug plug; =20 - start =3D untagged_addr(start); + start =3D untagged_addr(mm, start); =20 if (!madvise_behavior_valid(behavior)) return -EINVAL; diff --git a/mm/mempolicy.c b/mm/mempolicy.c index b73d3248d976..92819bccb082 100644 --- a/mm/mempolicy.c +++ b/mm/mempolicy.c @@ -1456,7 +1456,7 @@ static long kernel_mbind(unsigned long start, unsigne= d long len, int lmode =3D mode; int err; =20 - start =3D untagged_addr(start); + start =3D untagged_addr(current->mm, start); err =3D sanitize_mpol_flags(&lmode, &mode_flags); if (err) return err; @@ -1479,7 +1479,7 @@ SYSCALL_DEFINE4(set_mempolicy_home_node, unsigned lon= g, start, unsigned long, le unsigned long end; int err =3D -ENOENT; =20 - start =3D untagged_addr(start); + start =3D untagged_addr(mm, start); if (start & ~PAGE_MASK) return -EINVAL; /* @@ -1682,7 +1682,7 @@ static int kernel_get_mempolicy(int __user *policy, if (nmask !=3D NULL && maxnode < nr_node_ids) return -EINVAL; =20 - addr =3D untagged_addr(addr); + addr =3D untagged_addr(current->mm, addr); =20 err =3D do_get_mempolicy(&pval, &nodes, addr, flags); =20 diff --git a/mm/migrate.c b/mm/migrate.c index 6a1597c92261..e05e56605bce 100644 --- a/mm/migrate.c +++ b/mm/migrate.c @@ -1768,7 +1768,7 @@ static int do_pages_move(struct mm_struct *mm, nodema= sk_t task_nodes, goto out_flush; if (get_user(node, nodes + i)) goto out_flush; - addr =3D (unsigned long)untagged_addr(p); + addr =3D (unsigned long)untagged_addr(mm, p); =20 err =3D -ENODEV; if (node < 0 || node >=3D MAX_NUMNODES) diff --git a/mm/mincore.c b/mm/mincore.c index fa200c14185f..72c55bd9d184 100644 --- a/mm/mincore.c +++ b/mm/mincore.c @@ -236,7 +236,7 @@ SYSCALL_DEFINE3(mincore, unsigned long, start, size_t, = len, unsigned long pages; unsigned char *tmp; =20 - start =3D untagged_addr(start); + start =3D untagged_addr(current->mm, start); =20 /* Check the start address: needs to be page-aligned.. */ if (start & ~PAGE_MASK) diff --git a/mm/mlock.c b/mm/mlock.c index b14e929084cc..1ea100fe3db4 100644 --- a/mm/mlock.c +++ b/mm/mlock.c @@ -571,7 +571,7 @@ static __must_check int do_mlock(unsigned long start, s= ize_t len, vm_flags_t fla unsigned long lock_limit; int error =3D -ENOMEM; =20 - start =3D untagged_addr(start); + start =3D untagged_addr(current->mm, start); =20 if (!can_do_mlock()) return -EPERM; @@ -634,7 +634,7 @@ SYSCALL_DEFINE2(munlock, unsigned long, start, size_t, = len) { int ret; =20 - start =3D untagged_addr(start); + start =3D untagged_addr(current->mm, start); =20 len =3D PAGE_ALIGN(len + (offset_in_page(start))); start &=3D PAGE_MASK; diff --git a/mm/mmap.c b/mm/mmap.c index c035020d0c89..90047dc5098a 100644 --- a/mm/mmap.c +++ b/mm/mmap.c @@ -2877,7 +2877,7 @@ EXPORT_SYMBOL(vm_munmap); =20 SYSCALL_DEFINE2(munmap, unsigned long, addr, size_t, len) { - addr =3D untagged_addr(addr); + addr =3D untagged_addr(current->mm, addr); return __vm_munmap(addr, len, true); } =20 diff --git a/mm/mprotect.c b/mm/mprotect.c index 3a23dde73723..78adf9635194 100644 --- a/mm/mprotect.c +++ b/mm/mprotect.c @@ -669,7 +669,7 @@ static int do_mprotect_pkey(unsigned long start, size_t= len, (prot & PROT_READ); struct mmu_gather tlb; =20 - start =3D untagged_addr(start); + start =3D untagged_addr(current->mm, start); =20 prot &=3D ~(PROT_GROWSDOWN|PROT_GROWSUP); if (grows =3D=3D (PROT_GROWSDOWN|PROT_GROWSUP)) /* can't be both */ diff --git a/mm/mremap.c b/mm/mremap.c index b522cd0259a0..f76648bc4f67 100644 --- a/mm/mremap.c +++ b/mm/mremap.c @@ -906,7 +906,7 @@ SYSCALL_DEFINE5(mremap, unsigned long, addr, unsigned l= ong, old_len, * * See Documentation/arm64/tagged-address-abi.rst for more information. */ - addr =3D untagged_addr(addr); + addr =3D untagged_addr(mm, addr); =20 if (flags & ~(MREMAP_FIXED | MREMAP_MAYMOVE | MREMAP_DONTUNMAP)) return ret; diff --git a/mm/msync.c b/mm/msync.c index 137d1c104f3e..5fe989bd3c4b 100644 --- a/mm/msync.c +++ b/mm/msync.c @@ -37,7 +37,7 @@ SYSCALL_DEFINE3(msync, unsigned long, start, size_t, len,= int, flags) int unmapped_error =3D 0; int error =3D -EINVAL; =20 - start =3D untagged_addr(start); + start =3D untagged_addr(mm, start); =20 if (flags & ~(MS_ASYNC | MS_INVALIDATE | MS_SYNC)) goto out; diff --git a/virt/kvm/kvm_main.c b/virt/kvm/kvm_main.c index 515dfe9d3bcf..d2239aa85cf5 100644 --- a/virt/kvm/kvm_main.c +++ b/virt/kvm/kvm_main.c @@ -1943,7 +1943,7 @@ int __kvm_set_memory_region(struct kvm *kvm, return -EINVAL; /* We can read the guest memory with __xxx_user() later on. */ if ((mem->userspace_addr & (PAGE_SIZE - 1)) || - (mem->userspace_addr !=3D untagged_addr(mem->userspace_addr)) || + (mem->userspace_addr !=3D untagged_addr(kvm->mm, mem->userspace_addr)= ) || !access_ok((void __user *)(unsigned long)mem->userspace_addr, mem->memory_size)) return -EINVAL; --=20 2.35.1 From nobody Sat Apr 11 06:41:05 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 69A16C00140 for ; Mon, 15 Aug 2022 04:15:19 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S240531AbiHOEPR (ORCPT ); Mon, 15 Aug 2022 00:15:17 -0400 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:43518 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S240291AbiHOEPH (ORCPT ); Mon, 15 Aug 2022 00:15:07 -0400 Received: from mga12.intel.com (mga12.intel.com [192.55.52.136]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id 7ECF31402D for ; Sun, 14 Aug 2022 21:15:06 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=intel.com; i=@intel.com; q=dns/txt; s=Intel; t=1660536906; x=1692072906; h=from:to:cc:subject:date:message-id:in-reply-to: references:mime-version:content-transfer-encoding; bh=bSNkJ9cWetXtIf8kGg4S5rqxqS7SRVBkhPDApVxReb8=; b=mCkU+YMbfdmePDquZhctuww4YXgkqTdiywpihIyH6qwA7vGitO27BTeR 2o3hcgO2BGP7QgTYv8oJA1Eyi3mxi5dV4dkUyaPAZXMCEfQchpNVHicay DfujVaoXJh7RbAVfDArD812UbInxRv3BrEDlOQUhh4ucpiRcYJi1VAZ31 icgKHSk1/leKta1aF/2n41mZa4/soWhq5E2c/RLfekWkCGrDp9jTZl9ND oYI2ZmBTE7vOICmCJKZyMCN9TJBSRpAp5N1eeDv/8SEFa4GFOBBoFaBB8 rruTswfo9hn3hNhePmxH2rMxcSmPvpXJmyEjOrrbcH7P/gMTyxLQarr2n A==; X-IronPort-AV: E=McAfee;i="6400,9594,10439"; a="271658897" X-IronPort-AV: E=Sophos;i="5.93,237,1654585200"; d="scan'208";a="271658897" Received: from orsmga004.jf.intel.com ([10.7.209.38]) by fmsmga106.fm.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 14 Aug 2022 21:15:05 -0700 X-IronPort-AV: E=Sophos;i="5.93,237,1654585200"; d="scan'208";a="732818242" Received: from dmalka-mobl.ger.corp.intel.com (HELO box.shutemov.name) ([10.249.42.26]) by orsmga004-auth.jf.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 14 Aug 2022 21:15:01 -0700 Received: by box.shutemov.name (Postfix, from userid 1000) id 54F00104A69; Mon, 15 Aug 2022 07:18:04 +0300 (+03) From: "Kirill A. Shutemov" To: Dave Hansen , Andy Lutomirski , Peter Zijlstra Cc: x86@kernel.org, Kostya Serebryany , Andrey Ryabinin , Andrey Konovalov , Alexander Potapenko , Taras Madan , Dmitry Vyukov , "H . J . Lu" , Andi Kleen , Rick Edgecombe , linux-mm@kvack.org, linux-kernel@vger.kernel.org, "Kirill A. Shutemov" Subject: [PATCHv6 04/11] x86/mm: Handle LAM on context switch Date: Mon, 15 Aug 2022 07:17:56 +0300 Message-Id: <20220815041803.17954-5-kirill.shutemov@linux.intel.com> X-Mailer: git-send-email 2.35.1 In-Reply-To: <20220815041803.17954-1-kirill.shutemov@linux.intel.com> References: <20220815041803.17954-1-kirill.shutemov@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" Linear Address Masking mode for userspace pointers encoded in CR3 bits. The mode is selected per-thread. Add new thread features indicate that the thread has Linear Address Masking enabled. switch_mm_irqs_off() now respects these flags and constructs CR3 accordingly. The active LAM mode gets recorded in the tlb_state. Signed-off-by: Kirill A. Shutemov Tested-by: Alexander Potapenko Acked-by: Peter Zijlstra (Intel) --- arch/x86/include/asm/mmu.h | 3 +++ arch/x86/include/asm/mmu_context.h | 24 +++++++++++++++++ arch/x86/include/asm/tlbflush.h | 35 ++++++++++++++++++++++++ arch/x86/mm/tlb.c | 43 +++++++++++++++++++----------- 4 files changed, 90 insertions(+), 15 deletions(-) diff --git a/arch/x86/include/asm/mmu.h b/arch/x86/include/asm/mmu.h index 5d7494631ea9..002889ca8978 100644 --- a/arch/x86/include/asm/mmu.h +++ b/arch/x86/include/asm/mmu.h @@ -40,6 +40,9 @@ typedef struct { =20 #ifdef CONFIG_X86_64 unsigned short flags; + + /* Active LAM mode: X86_CR3_LAM_U48 or X86_CR3_LAM_U57 or 0 (disabled) */ + unsigned long lam_cr3_mask; #endif =20 struct mutex lock; diff --git a/arch/x86/include/asm/mmu_context.h b/arch/x86/include/asm/mmu_= context.h index b8d40ddeab00..69c943b2ae90 100644 --- a/arch/x86/include/asm/mmu_context.h +++ b/arch/x86/include/asm/mmu_context.h @@ -91,6 +91,29 @@ static inline void switch_ldt(struct mm_struct *prev, st= ruct mm_struct *next) } #endif =20 +#ifdef CONFIG_X86_64 +static inline unsigned long mm_lam_cr3_mask(struct mm_struct *mm) +{ + return mm->context.lam_cr3_mask; +} + +static inline void dup_lam(struct mm_struct *oldmm, struct mm_struct *mm) +{ + mm->context.lam_cr3_mask =3D oldmm->context.lam_cr3_mask; +} + +#else + +static inline unsigned long mm_lam_cr3_mask(struct mm_struct *mm) +{ + return 0; +} + +static inline void dup_lam(struct mm_struct *oldmm, struct mm_struct *mm) +{ +} +#endif + #define enter_lazy_tlb enter_lazy_tlb extern void enter_lazy_tlb(struct mm_struct *mm, struct task_struct *tsk); =20 @@ -168,6 +191,7 @@ static inline int arch_dup_mmap(struct mm_struct *oldmm= , struct mm_struct *mm) { arch_dup_pkeys(oldmm, mm); paravirt_arch_dup_mmap(oldmm, mm); + dup_lam(oldmm, mm); return ldt_dup_context(oldmm, mm); } =20 diff --git a/arch/x86/include/asm/tlbflush.h b/arch/x86/include/asm/tlbflus= h.h index cda3118f3b27..1ad080163363 100644 --- a/arch/x86/include/asm/tlbflush.h +++ b/arch/x86/include/asm/tlbflush.h @@ -101,6 +101,16 @@ struct tlb_state { */ bool invalidate_other; =20 +#ifdef CONFIG_X86_64 + /* + * Active LAM mode. + * + * X86_CR3_LAM_U57/U48 shifted right by X86_CR3_LAM_U57_BIT or 0 if LAM + * disabled. + */ + u8 lam; +#endif + /* * Mask that contains TLB_NR_DYN_ASIDS+1 bits to indicate * the corresponding user PCID needs a flush next time we @@ -357,6 +367,30 @@ static inline bool huge_pmd_needs_flush(pmd_t oldpmd, = pmd_t newpmd) } #define huge_pmd_needs_flush huge_pmd_needs_flush =20 +#ifdef CONFIG_X86_64 +static inline unsigned long tlbstate_lam_cr3_mask(void) +{ + unsigned long lam =3D this_cpu_read(cpu_tlbstate.lam); + + return lam << X86_CR3_LAM_U57_BIT; +} + +static inline void set_tlbstate_cr3_lam_mask(unsigned long mask) +{ + this_cpu_write(cpu_tlbstate.lam, mask >> X86_CR3_LAM_U57_BIT); +} + +#else + +static inline unsigned long tlbstate_lam_cr3_mask(void) +{ + return 0; +} + +static inline void set_tlbstate_cr3_lam_mask(u64 mask) +{ +} +#endif #endif /* !MODULE */ =20 static inline void __native_tlb_flush_global(unsigned long cr4) @@ -364,4 +398,5 @@ static inline void __native_tlb_flush_global(unsigned l= ong cr4) native_write_cr4(cr4 ^ X86_CR4_PGE); native_write_cr4(cr4); } + #endif /* _ASM_X86_TLBFLUSH_H */ diff --git a/arch/x86/mm/tlb.c b/arch/x86/mm/tlb.c index c1e31e9a85d7..fdc0b69b5da7 100644 --- a/arch/x86/mm/tlb.c +++ b/arch/x86/mm/tlb.c @@ -154,17 +154,18 @@ static inline u16 user_pcid(u16 asid) return ret; } =20 -static inline unsigned long build_cr3(pgd_t *pgd, u16 asid) +static inline unsigned long build_cr3(pgd_t *pgd, u16 asid, unsigned long = lam) { if (static_cpu_has(X86_FEATURE_PCID)) { - return __sme_pa(pgd) | kern_pcid(asid); + return __sme_pa(pgd) | kern_pcid(asid) | lam; } else { VM_WARN_ON_ONCE(asid !=3D 0); - return __sme_pa(pgd); + return __sme_pa(pgd) | lam; } } =20 -static inline unsigned long build_cr3_noflush(pgd_t *pgd, u16 asid) +static inline unsigned long build_cr3_noflush(pgd_t *pgd, u16 asid, + unsigned long lam) { VM_WARN_ON_ONCE(asid > MAX_ASID_AVAILABLE); /* @@ -173,7 +174,7 @@ static inline unsigned long build_cr3_noflush(pgd_t *pg= d, u16 asid) * boot because all CPU's the have same capabilities: */ VM_WARN_ON_ONCE(!boot_cpu_has(X86_FEATURE_PCID)); - return __sme_pa(pgd) | kern_pcid(asid) | CR3_NOFLUSH; + return __sme_pa(pgd) | kern_pcid(asid) | lam | CR3_NOFLUSH; } =20 /* @@ -274,15 +275,16 @@ static inline void invalidate_user_asid(u16 asid) (unsigned long *)this_cpu_ptr(&cpu_tlbstate.user_pcid_flush_mask)); } =20 -static void load_new_mm_cr3(pgd_t *pgdir, u16 new_asid, bool need_flush) +static void load_new_mm_cr3(pgd_t *pgdir, u16 new_asid, unsigned long lam, + bool need_flush) { unsigned long new_mm_cr3; =20 if (need_flush) { invalidate_user_asid(new_asid); - new_mm_cr3 =3D build_cr3(pgdir, new_asid); + new_mm_cr3 =3D build_cr3(pgdir, new_asid, lam); } else { - new_mm_cr3 =3D build_cr3_noflush(pgdir, new_asid); + new_mm_cr3 =3D build_cr3_noflush(pgdir, new_asid, lam); } =20 /* @@ -491,6 +493,8 @@ void switch_mm_irqs_off(struct mm_struct *prev, struct = mm_struct *next, { struct mm_struct *real_prev =3D this_cpu_read(cpu_tlbstate.loaded_mm); u16 prev_asid =3D this_cpu_read(cpu_tlbstate.loaded_mm_asid); + unsigned long prev_lam =3D tlbstate_lam_cr3_mask(); + unsigned long new_lam =3D mm_lam_cr3_mask(next); bool was_lazy =3D this_cpu_read(cpu_tlbstate_shared.is_lazy); unsigned cpu =3D smp_processor_id(); u64 next_tlb_gen; @@ -520,7 +524,7 @@ void switch_mm_irqs_off(struct mm_struct *prev, struct = mm_struct *next, * isn't free. */ #ifdef CONFIG_DEBUG_VM - if (WARN_ON_ONCE(__read_cr3() !=3D build_cr3(real_prev->pgd, prev_asid)))= { + if (WARN_ON_ONCE(__read_cr3() !=3D build_cr3(real_prev->pgd, prev_asid, p= rev_lam))) { /* * If we were to BUG here, we'd be very likely to kill * the system so hard that we don't see the call trace. @@ -554,6 +558,7 @@ void switch_mm_irqs_off(struct mm_struct *prev, struct = mm_struct *next, if (real_prev =3D=3D next) { VM_WARN_ON(this_cpu_read(cpu_tlbstate.ctxs[prev_asid].ctx_id) !=3D next->context.ctx_id); + VM_WARN_ON(prev_lam !=3D new_lam); =20 /* * Even in lazy TLB mode, the CPU should stay set in the @@ -622,15 +627,16 @@ void switch_mm_irqs_off(struct mm_struct *prev, struc= t mm_struct *next, barrier(); } =20 + set_tlbstate_cr3_lam_mask(new_lam); if (need_flush) { this_cpu_write(cpu_tlbstate.ctxs[new_asid].ctx_id, next->context.ctx_id); this_cpu_write(cpu_tlbstate.ctxs[new_asid].tlb_gen, next_tlb_gen); - load_new_mm_cr3(next->pgd, new_asid, true); + load_new_mm_cr3(next->pgd, new_asid, new_lam, true); =20 trace_tlb_flush(TLB_FLUSH_ON_TASK_SWITCH, TLB_FLUSH_ALL); } else { /* The new ASID is already up to date. */ - load_new_mm_cr3(next->pgd, new_asid, false); + load_new_mm_cr3(next->pgd, new_asid, new_lam, false); =20 trace_tlb_flush(TLB_FLUSH_ON_TASK_SWITCH, 0); } @@ -691,6 +697,10 @@ void initialize_tlbstate_and_flush(void) /* Assert that CR3 already references the right mm. */ WARN_ON((cr3 & CR3_ADDR_MASK) !=3D __pa(mm->pgd)); =20 + /* LAM expected to be disabled in CR3 and init_mm */ + WARN_ON(cr3 & (X86_CR3_LAM_U48 | X86_CR3_LAM_U57)); + WARN_ON(mm_lam_cr3_mask(&init_mm)); + /* * Assert that CR4.PCIDE is set if needed. (CR4.PCIDE initialization * doesn't work like other CR4 bits because it can only be set from @@ -699,8 +709,8 @@ void initialize_tlbstate_and_flush(void) WARN_ON(boot_cpu_has(X86_FEATURE_PCID) && !(cr4_read_shadow() & X86_CR4_PCIDE)); =20 - /* Force ASID 0 and force a TLB flush. */ - write_cr3(build_cr3(mm->pgd, 0)); + /* Disable LAM, force ASID 0 and force a TLB flush. */ + write_cr3(build_cr3(mm->pgd, 0, 0)); =20 /* Reinitialize tlbstate. */ this_cpu_write(cpu_tlbstate.last_user_mm_spec, LAST_USER_MM_INIT); @@ -708,6 +718,7 @@ void initialize_tlbstate_and_flush(void) this_cpu_write(cpu_tlbstate.next_asid, 1); this_cpu_write(cpu_tlbstate.ctxs[0].ctx_id, mm->context.ctx_id); this_cpu_write(cpu_tlbstate.ctxs[0].tlb_gen, tlb_gen); + set_tlbstate_cr3_lam_mask(0); =20 for (i =3D 1; i < TLB_NR_DYN_ASIDS; i++) this_cpu_write(cpu_tlbstate.ctxs[i].ctx_id, 0); @@ -1071,8 +1082,10 @@ void flush_tlb_kernel_range(unsigned long start, uns= igned long end) */ unsigned long __get_current_cr3_fast(void) { - unsigned long cr3 =3D build_cr3(this_cpu_read(cpu_tlbstate.loaded_mm)->pg= d, - this_cpu_read(cpu_tlbstate.loaded_mm_asid)); + unsigned long cr3 =3D + build_cr3(this_cpu_read(cpu_tlbstate.loaded_mm)->pgd, + this_cpu_read(cpu_tlbstate.loaded_mm_asid), + tlbstate_lam_cr3_mask()); =20 /* For now, be very restrictive about when this can be called. */ VM_WARN_ON(in_nmi() || preemptible()); --=20 2.35.1 From nobody Sat Apr 11 06:41:05 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 66D60C00140 for ; Mon, 15 Aug 2022 04:15:28 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S240663AbiHOEP0 (ORCPT ); Mon, 15 Aug 2022 00:15:26 -0400 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:43604 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S240454AbiHOEPL (ORCPT ); Mon, 15 Aug 2022 00:15:11 -0400 Received: from mga12.intel.com (mga12.intel.com [192.55.52.136]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id BA05013E3D for ; Sun, 14 Aug 2022 21:15:10 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=intel.com; i=@intel.com; q=dns/txt; s=Intel; t=1660536910; x=1692072910; h=from:to:cc:subject:date:message-id:in-reply-to: references:mime-version:content-transfer-encoding; bh=KTcDC6oRvgqapGCb7AxZ9WYL7Fb8j5UPhWrve0lM//Y=; b=Re3XlVrhM+eQ5cYmulmbSvBKpNQoeXXkTjwLTQsAcWIESLqmsildA5or 8RscjUbq0wE3wABa3kUWfLWod45Z4bgoeefpb9K9eou7bzpKPII3Ukxya PmgYZoumOZOQnKi+HoEtVexhecjfGIKWs3MUxNZm2bHVWZ00SJfEEwaJh rGue+Be8URu6RRSWfcR7cNq+3tJlDSxLTdy7GfuaLoCYxC/c0Xq5pJuAk DkYgRz384QizS5YY5kBYpwJuoS3DdKbbq5aAfXJCiRMRtKylEzTCB/0dE 42CqJ/+J53DaxswofAcUzxLQHrtXHFh91A3IBETkVPbwWGceeFzjOM4Dc Q==; X-IronPort-AV: E=McAfee;i="6400,9594,10439"; a="271658915" X-IronPort-AV: E=Sophos;i="5.93,237,1654585200"; d="scan'208";a="271658915" Received: from fmsmga005.fm.intel.com ([10.253.24.32]) by fmsmga106.fm.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 14 Aug 2022 21:15:10 -0700 X-IronPort-AV: E=Sophos;i="5.93,237,1654585200"; d="scan'208";a="934354104" Received: from dmalka-mobl.ger.corp.intel.com (HELO box.shutemov.name) ([10.249.42.26]) by fmsmga005-auth.fm.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 14 Aug 2022 21:15:07 -0700 Received: by box.shutemov.name (Postfix, from userid 1000) id 5DE3B104A6A; Mon, 15 Aug 2022 07:18:04 +0300 (+03) From: "Kirill A. Shutemov" To: Dave Hansen , Andy Lutomirski , Peter Zijlstra Cc: x86@kernel.org, Kostya Serebryany , Andrey Ryabinin , Andrey Konovalov , Alexander Potapenko , Taras Madan , Dmitry Vyukov , "H . J . Lu" , Andi Kleen , Rick Edgecombe , linux-mm@kvack.org, linux-kernel@vger.kernel.org, "Kirill A. Shutemov" Subject: [PATCHv6 05/11] x86/uaccess: Provide untagged_addr() and remove tags before address check Date: Mon, 15 Aug 2022 07:17:57 +0300 Message-Id: <20220815041803.17954-6-kirill.shutemov@linux.intel.com> X-Mailer: git-send-email 2.35.1 In-Reply-To: <20220815041803.17954-1-kirill.shutemov@linux.intel.com> References: <20220815041803.17954-1-kirill.shutemov@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" untagged_addr() is a helper used by the core-mm to strip tag bits and get the address to the canonical shape. In only handles userspace addresses. The untagging mask is stored in mmu_context and will be set on enabling LAM for the process. The tags must not be included into check whether it's okay to access the userspace address. Strip tags in access_ok(). get_user() and put_user() don't use access_ok(), but check access against TASK_SIZE directly in assembly. Strip tags, before calling into the assembly helper. Signed-off-by: Kirill A. Shutemov Tested-by: Alexander Potapenko Acked-by: Peter Zijlstra (Intel) --- arch/x86/include/asm/mmu.h | 3 +++ arch/x86/include/asm/mmu_context.h | 11 ++++++++ arch/x86/include/asm/uaccess.h | 42 +++++++++++++++++++++++++++--- arch/x86/kernel/process.c | 3 +++ 4 files changed, 56 insertions(+), 3 deletions(-) diff --git a/arch/x86/include/asm/mmu.h b/arch/x86/include/asm/mmu.h index 002889ca8978..2fdb390040b5 100644 --- a/arch/x86/include/asm/mmu.h +++ b/arch/x86/include/asm/mmu.h @@ -43,6 +43,9 @@ typedef struct { =20 /* Active LAM mode: X86_CR3_LAM_U48 or X86_CR3_LAM_U57 or 0 (disabled) */ unsigned long lam_cr3_mask; + + /* Significant bits of the virtual address. Excludes tag bits. */ + u64 untag_mask; #endif =20 struct mutex lock; diff --git a/arch/x86/include/asm/mmu_context.h b/arch/x86/include/asm/mmu_= context.h index 69c943b2ae90..5bd3d46685dc 100644 --- a/arch/x86/include/asm/mmu_context.h +++ b/arch/x86/include/asm/mmu_context.h @@ -100,6 +100,12 @@ static inline unsigned long mm_lam_cr3_mask(struct mm_= struct *mm) static inline void dup_lam(struct mm_struct *oldmm, struct mm_struct *mm) { mm->context.lam_cr3_mask =3D oldmm->context.lam_cr3_mask; + mm->context.untag_mask =3D oldmm->context.untag_mask; +} + +static inline void mm_reset_untag_mask(struct mm_struct *mm) +{ + mm->context.untag_mask =3D -1UL; } =20 #else @@ -112,6 +118,10 @@ static inline unsigned long mm_lam_cr3_mask(struct mm_= struct *mm) static inline void dup_lam(struct mm_struct *oldmm, struct mm_struct *mm) { } + +static inline void mm_reset_untag_mask(struct mm_struct *mm) +{ +} #endif =20 #define enter_lazy_tlb enter_lazy_tlb @@ -138,6 +148,7 @@ static inline int init_new_context(struct task_struct *= tsk, mm->context.execute_only_pkey =3D -1; } #endif + mm_reset_untag_mask(mm); init_new_context_ldt(mm); return 0; } diff --git a/arch/x86/include/asm/uaccess.h b/arch/x86/include/asm/uaccess.h index 913e593a3b45..803241dfc473 100644 --- a/arch/x86/include/asm/uaccess.h +++ b/arch/x86/include/asm/uaccess.h @@ -6,6 +6,7 @@ */ #include #include +#include #include #include #include @@ -20,6 +21,30 @@ static inline bool pagefault_disabled(void); # define WARN_ON_IN_IRQ() #endif =20 +#ifdef CONFIG_X86_64 +/* + * Mask out tag bits from the address. + * + * Magic with the 'sign' allows to untag userspace pointer without any bra= nches + * while leaving kernel addresses intact. + */ +#define untagged_addr(mm, addr) ({ \ + u64 __addr =3D (__force u64)(addr); \ + s64 sign =3D (s64)__addr >> 63; \ + __addr &=3D (mm)->context.untag_mask | sign; \ + (__force __typeof__(addr))__addr; \ +}) + +#define untagged_ptr(mm, ptr) ({ \ + u64 __ptrval =3D (__force u64)(ptr); \ + __ptrval =3D untagged_addr(mm, __ptrval); \ + (__force __typeof__(*(ptr)) *)__ptrval; \ +}) +#else +#define untagged_addr(mm, addr) (addr) +#define untagged_ptr(mm, ptr) (ptr) +#endif + /** * access_ok - Checks if a user space pointer is valid * @addr: User space pointer to start of block to check @@ -40,7 +65,7 @@ static inline bool pagefault_disabled(void); #define access_ok(addr, size) \ ({ \ WARN_ON_IN_IRQ(); \ - likely(__access_ok(addr, size)); \ + likely(__access_ok(untagged_addr(current->mm, addr), size)); \ }) =20 #include @@ -125,7 +150,13 @@ extern int __get_user_bad(void); * Return: zero on success, or -EFAULT on error. * On error, the variable @x is set to zero. */ -#define get_user(x,ptr) ({ might_fault(); do_get_user_call(get_user,x,ptr)= ; }) +#define get_user(x,ptr) \ +({ \ + __typeof__(*(ptr)) __user *__ptr_clean; \ + __ptr_clean =3D untagged_ptr(current->mm, ptr); \ + might_fault(); \ + do_get_user_call(get_user,x,__ptr_clean); \ +}) =20 /** * __get_user - Get a simple variable from user space, with less checking. @@ -222,7 +253,12 @@ extern void __put_user_nocheck_8(void); * * Return: zero on success, or -EFAULT on error. */ -#define put_user(x, ptr) ({ might_fault(); do_put_user_call(put_user,x,ptr= ); }) +#define put_user(x, ptr) ({ \ + __typeof__(*(ptr)) __user *__ptr_clean; \ + __ptr_clean =3D untagged_ptr(current->mm, ptr); \ + might_fault(); \ + do_put_user_call(put_user,x,__ptr_clean); \ +}) =20 /** * __put_user - Write a simple value into user space, with less checking. diff --git a/arch/x86/kernel/process.c b/arch/x86/kernel/process.c index 58a6ea472db9..b0e86fb11ffa 100644 --- a/arch/x86/kernel/process.c +++ b/arch/x86/kernel/process.c @@ -47,6 +47,7 @@ #include #include #include +#include =20 #include "process.h" =20 @@ -367,6 +368,8 @@ void arch_setup_new_exec(void) task_clear_spec_ssb_noexec(current); speculation_ctrl_update(read_thread_flags()); } + + mm_reset_untag_mask(current->mm); } =20 #ifdef CONFIG_X86_IOPL_IOPERM --=20 2.35.1 From nobody Sat Apr 11 06:41:05 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 04861C00140 for ; Mon, 15 Aug 2022 04:15:32 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S240670AbiHOEP3 (ORCPT ); Mon, 15 Aug 2022 00:15:29 -0400 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:43614 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S240455AbiHOEPM (ORCPT ); Mon, 15 Aug 2022 00:15:12 -0400 Received: from mga05.intel.com (mga05.intel.com [192.55.52.43]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id 529781403E for ; Sun, 14 Aug 2022 21:15:11 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=intel.com; i=@intel.com; q=dns/txt; s=Intel; t=1660536911; x=1692072911; h=from:to:cc:subject:date:message-id:in-reply-to: references:mime-version:content-transfer-encoding; bh=rGmRjWLsZd0MykLtxfW7HGQQOCTU/72uM1KezIjGuQ4=; b=Tfk+yOKG+GXvspXq7RqwDYpGwLiTVgt1/tfZmEpGRw21mljPopbsDfLB HfhKfGHfBsR8Um+Je/FAPW2h8ybGLKuuff0KhhJEMRqiblWnEQWAPllC9 K7o2KouZPYX5T1PsKQeHIJZdY53BqQjB78gf4qKNOvnXV2FMtRa9Nd53M 7amkx/rLor0094FuU0fYz1BlIhxsHjFfcEiJ22Xens00FL0AOtTzFS/Bi hGrpxIlKChOtzfrwDRiCb+15Avb6wNlOHg8n5yWrqRDrAPVS6Uju5Lg0a yynIlvTqA/aaclG3G+5yn/MFP8kNohj2w7qN1jmxo4EQIqMnDQMj8JEqS g==; X-IronPort-AV: E=McAfee;i="6400,9594,10439"; a="378175188" X-IronPort-AV: E=Sophos;i="5.93,237,1654585200"; d="scan'208";a="378175188" Received: from orsmga006.jf.intel.com ([10.7.209.51]) by fmsmga105.fm.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 14 Aug 2022 21:15:10 -0700 X-IronPort-AV: E=Sophos;i="5.93,237,1654585200"; d="scan'208";a="582747731" Received: from dmalka-mobl.ger.corp.intel.com (HELO box.shutemov.name) ([10.249.42.26]) by orsmga006-auth.jf.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 14 Aug 2022 21:15:07 -0700 Received: by box.shutemov.name (Postfix, from userid 1000) id 67CDA104A6B; Mon, 15 Aug 2022 07:18:04 +0300 (+03) From: "Kirill A. Shutemov" To: Dave Hansen , Andy Lutomirski , Peter Zijlstra Cc: x86@kernel.org, Kostya Serebryany , Andrey Ryabinin , Andrey Konovalov , Alexander Potapenko , Taras Madan , Dmitry Vyukov , "H . J . Lu" , Andi Kleen , Rick Edgecombe , linux-mm@kvack.org, linux-kernel@vger.kernel.org, "Kirill A. Shutemov" Subject: [PATCHv6 06/11] x86/mm: Provide arch_prctl() interface for LAM Date: Mon, 15 Aug 2022 07:17:58 +0300 Message-Id: <20220815041803.17954-7-kirill.shutemov@linux.intel.com> X-Mailer: git-send-email 2.35.1 In-Reply-To: <20220815041803.17954-1-kirill.shutemov@linux.intel.com> References: <20220815041803.17954-1-kirill.shutemov@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" Add a couple of arch_prctl() handles: - ARCH_ENABLE_TAGGED_ADDR enabled LAM. The argument is required number of tag bits. It is rounded up to the nearest LAM mode that can provide it. For now only LAM_U57 is supported, with 6 tag bits. - ARCH_GET_UNTAG_MASK returns untag mask. It can indicates where tag bits located in the address. - ARCH_GET_MAX_TAG_BITS returns the maximum tag bits user can request. Zero if LAM is not supported. Signed-off-by: Kirill A. Shutemov Acked-by: Peter Zijlstra (Intel) Reviewed-by: Alexander Potapenko Tested-by: Alexander Potapenko --- arch/x86/include/uapi/asm/prctl.h | 4 ++ arch/x86/kernel/process_64.c | 65 ++++++++++++++++++++++++++++++- 2 files changed, 68 insertions(+), 1 deletion(-) diff --git a/arch/x86/include/uapi/asm/prctl.h b/arch/x86/include/uapi/asm/= prctl.h index 500b96e71f18..a31e27b95b19 100644 --- a/arch/x86/include/uapi/asm/prctl.h +++ b/arch/x86/include/uapi/asm/prctl.h @@ -20,4 +20,8 @@ #define ARCH_MAP_VDSO_32 0x2002 #define ARCH_MAP_VDSO_64 0x2003 =20 +#define ARCH_GET_UNTAG_MASK 0x4001 +#define ARCH_ENABLE_TAGGED_ADDR 0x4002 +#define ARCH_GET_MAX_TAG_BITS 0x4003 + #endif /* _ASM_X86_PRCTL_H */ diff --git a/arch/x86/kernel/process_64.c b/arch/x86/kernel/process_64.c index 1962008fe743..4f9f0f8ccd26 100644 --- a/arch/x86/kernel/process_64.c +++ b/arch/x86/kernel/process_64.c @@ -742,6 +742,60 @@ static long prctl_map_vdso(const struct vdso_image *im= age, unsigned long addr) } #endif =20 +static void enable_lam_func(void *mm) +{ + struct mm_struct *loaded_mm =3D this_cpu_read(cpu_tlbstate.loaded_mm); + unsigned long lam_mask; + unsigned long cr3; + + if (loaded_mm !=3D mm) + return; + + lam_mask =3D READ_ONCE(loaded_mm->context.lam_cr3_mask); + + /* Update CR3 to get LAM active on the CPU */ + cr3 =3D __read_cr3(); + cr3 &=3D ~(X86_CR3_LAM_U48 | X86_CR3_LAM_U57); + cr3 |=3D lam_mask; + write_cr3(cr3); + set_tlbstate_cr3_lam_mask(lam_mask); +} + +static int prctl_enable_tagged_addr(struct mm_struct *mm, unsigned long nr= _bits) +{ + int ret =3D 0; + + if (!cpu_feature_enabled(X86_FEATURE_LAM)) + return -ENODEV; + + mutex_lock(&mm->context.lock); + + /* Already enabled? */ + if (mm->context.lam_cr3_mask) { + ret =3D -EBUSY; + goto out; + } + + if (!nr_bits) { + ret =3D -EINVAL; + goto out; + } else if (nr_bits <=3D 6) { + mm->context.lam_cr3_mask =3D X86_CR3_LAM_U57; + mm->context.untag_mask =3D ~GENMASK(62, 57); + } else { + ret =3D -EINVAL; + goto out; + } + + /* Make lam_cr3_mask and untag_mask visible on other CPUs */ + smp_mb(); + + on_each_cpu_mask(mm_cpumask(mm), enable_lam_func, mm, true); +out: + mutex_unlock(&mm->context.lock); + return ret; +} + long do_arch_prctl_64(struct task_struct *task, int option, unsigned long = arg2) { int ret =3D 0; @@ -829,7 +883,16 @@ long do_arch_prctl_64(struct task_struct *task, int op= tion, unsigned long arg2) case ARCH_MAP_VDSO_64: return prctl_map_vdso(&vdso_image_64, arg2); #endif - + case ARCH_GET_UNTAG_MASK: + return put_user(task->mm->context.untag_mask, + (unsigned long __user *)arg2); + case ARCH_ENABLE_TAGGED_ADDR: + return prctl_enable_tagged_addr(task->mm, arg2); + case ARCH_GET_MAX_TAG_BITS: + if (!cpu_feature_enabled(X86_FEATURE_LAM)) + return put_user(0, (unsigned long __user *)arg2); + else + return put_user(6, (unsigned long __user *)arg2); default: ret =3D -EINVAL; break; --=20 2.35.1 From nobody Sat Apr 11 06:41:05 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 0FDE3C00140 for ; Mon, 15 Aug 2022 04:17:39 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S240851AbiHOEPl (ORCPT ); Mon, 15 Aug 2022 00:15:41 -0400 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:44136 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S240369AbiHOEPW (ORCPT ); Mon, 15 Aug 2022 00:15:22 -0400 Received: from mga11.intel.com (mga11.intel.com [192.55.52.93]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id 259AE14082 for ; Sun, 14 Aug 2022 21:15:12 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=intel.com; i=@intel.com; q=dns/txt; s=Intel; t=1660536912; x=1692072912; h=from:to:cc:subject:date:message-id:in-reply-to: references:mime-version:content-transfer-encoding; bh=ds9Ls73+GWfzKa4/92sftksfxawt2Atz4orB+uMy8Go=; b=fTyW3NyUKxRNDn80gH5c+6bBcs0MkJwUhE8vNsdjNO8cvnb+hImYrfLw DcSn88hkFJTWPYWevZkmPw2Ew6IEvQDuGXeRsS9KU9EbudsDH3zSCa/qq q60IwTT9PxtaG13GYOQ5HOvivFAPxPTaZvqLBAcB/U69N2F+/YDhMMiHP A7DduYuFJq2HPb5VuCn26J/44yFZZqstFqn5/y3CGaLJM7ERvcwQye2pw mZzVelQGhYZ/y3twtCIvj7YYsqVW77nGQ6ZYbIyMoWnZlhV4tOn7heM+v nwDXVMvIAN+E07bGTx+RJMcPMp//HlsmgAWFIzfDm9FmN7W0P5WsUrsKU Q==; X-IronPort-AV: E=McAfee;i="6400,9594,10439"; a="289455874" X-IronPort-AV: E=Sophos;i="5.93,237,1654585200"; d="scan'208";a="289455874" Received: from orsmga003.jf.intel.com ([10.7.209.27]) by fmsmga102.fm.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 14 Aug 2022 21:15:11 -0700 X-IronPort-AV: E=Sophos;i="5.93,237,1654585200"; d="scan'208";a="557160547" Received: from dmalka-mobl.ger.corp.intel.com (HELO box.shutemov.name) ([10.249.42.26]) by orsmga003-auth.jf.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 14 Aug 2022 21:15:07 -0700 Received: by box.shutemov.name (Postfix, from userid 1000) id 71CB2104A6C; Mon, 15 Aug 2022 07:18:04 +0300 (+03) From: "Kirill A. Shutemov" To: Dave Hansen , Andy Lutomirski , Peter Zijlstra Cc: x86@kernel.org, Kostya Serebryany , Andrey Ryabinin , Andrey Konovalov , Alexander Potapenko , Taras Madan , Dmitry Vyukov , "H . J . Lu" , Andi Kleen , Rick Edgecombe , linux-mm@kvack.org, linux-kernel@vger.kernel.org, "Kirill A. Shutemov" Subject: [PATCHv6 07/11] x86: Expose untagging mask in /proc/$PID/arch_status Date: Mon, 15 Aug 2022 07:17:59 +0300 Message-Id: <20220815041803.17954-8-kirill.shutemov@linux.intel.com> X-Mailer: git-send-email 2.35.1 In-Reply-To: <20220815041803.17954-1-kirill.shutemov@linux.intel.com> References: <20220815041803.17954-1-kirill.shutemov@linux.intel.com> MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable Precedence: bulk List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Type: text/plain; charset="utf-8" Add a line in /proc/$PID/arch_status to report untag_mask. It can be used to find out LAM status of the process from the outside. It is useful for debuggers. Signed-off-by: Kirill A. Shutemov Tested-by: Alexander Potapenko Acked-by: Peter Zijlstra (Intel) --- arch/x86/include/asm/mmu_context.h | 10 +++++ arch/x86/kernel/Makefile | 2 + arch/x86/kernel/fpu/xstate.c | 47 ----------------------- arch/x86/kernel/proc.c | 60 ++++++++++++++++++++++++++++++ 4 files changed, 72 insertions(+), 47 deletions(-) create mode 100644 arch/x86/kernel/proc.c diff --git a/arch/x86/include/asm/mmu_context.h b/arch/x86/include/asm/mmu_= context.h index 5bd3d46685dc..b0e9ea23758b 100644 --- a/arch/x86/include/asm/mmu_context.h +++ b/arch/x86/include/asm/mmu_context.h @@ -103,6 +103,11 @@ static inline void dup_lam(struct mm_struct *oldmm, st= ruct mm_struct *mm) mm->context.untag_mask =3D oldmm->context.untag_mask; } =20 +static inline unsigned long mm_untag_mask(struct mm_struct *mm) +{ + return mm->context.untag_mask; +} + static inline void mm_reset_untag_mask(struct mm_struct *mm) { mm->context.untag_mask =3D -1UL; @@ -119,6 +124,11 @@ static inline void dup_lam(struct mm_struct *oldmm, st= ruct mm_struct *mm) { } =20 +static inline unsigned long mm_untag_mask(struct mm_struct *mm) +{ + return -1UL; +} + static inline void mm_reset_untag_mask(struct mm_struct *mm) { } diff --git a/arch/x86/kernel/Makefile b/arch/x86/kernel/Makefile index a20a5ebfacd7..fada0e36031b 100644 --- a/arch/x86/kernel/Makefile +++ b/arch/x86/kernel/Makefile @@ -139,6 +139,8 @@ obj-$(CONFIG_UNWINDER_GUESS) +=3D unwind_guess.o =20 obj-$(CONFIG_AMD_MEM_ENCRYPT) +=3D sev.o =20 +obj-$(CONFIG_PROC_FS) +=3D proc.o + ### # 64 bit specific files ifeq ($(CONFIG_X86_64),y) diff --git a/arch/x86/kernel/fpu/xstate.c b/arch/x86/kernel/fpu/xstate.c index c8340156bfd2..838a6f0627fd 100644 --- a/arch/x86/kernel/fpu/xstate.c +++ b/arch/x86/kernel/fpu/xstate.c @@ -10,8 +10,6 @@ #include #include #include -#include -#include #include =20 #include @@ -1745,48 +1743,3 @@ long fpu_xstate_prctl(int option, unsigned long arg2) return -EINVAL; } } - -#ifdef CONFIG_PROC_PID_ARCH_STATUS -/* - * Report the amount of time elapsed in millisecond since last AVX512 - * use in the task. - */ -static void avx512_status(struct seq_file *m, struct task_struct *task) -{ - unsigned long timestamp =3D READ_ONCE(task->thread.fpu.avx512_timestamp); - long delta; - - if (!timestamp) { - /* - * Report -1 if no AVX512 usage - */ - delta =3D -1; - } else { - delta =3D (long)(jiffies - timestamp); - /* - * Cap to LONG_MAX if time difference > LONG_MAX - */ - if (delta < 0) - delta =3D LONG_MAX; - delta =3D jiffies_to_msecs(delta); - } - - seq_put_decimal_ll(m, "AVX512_elapsed_ms:\t", delta); - seq_putc(m, '\n'); -} - -/* - * Report architecture specific information - */ -int proc_pid_arch_status(struct seq_file *m, struct pid_namespace *ns, - struct pid *pid, struct task_struct *task) -{ - /* - * Report AVX512 state if the processor and build option supported. - */ - if (cpu_feature_enabled(X86_FEATURE_AVX512F)) - avx512_status(m, task); - - return 0; -} -#endif /* CONFIG_PROC_PID_ARCH_STATUS */ diff --git a/arch/x86/kernel/proc.c b/arch/x86/kernel/proc.c new file mode 100644 index 000000000000..9765b4d05ce4 --- /dev/null +++ b/arch/x86/kernel/proc.c @@ -0,0 +1,60 @@ +// SPDX-License-Identifier: GPL-2.0-only +#include +#include +#include +#include +#include + +/* + * Report the amount of time elapsed in millisecond since last AVX512 + * use in the task. + */ +static void avx512_status(struct seq_file *m, struct task_struct *task) +{ + unsigned long timestamp =3D READ_ONCE(task->thread.fpu.avx512_timestamp); + long delta; + + if (!timestamp) { + /* + * Report -1 if no AVX512 usage + */ + delta =3D -1; + } else { + delta =3D (long)(jiffies - timestamp); + /* + * Cap to LONG_MAX if time difference > LONG_MAX + */ + if (delta < 0) + delta =3D LONG_MAX; + delta =3D jiffies_to_msecs(delta); + } + + seq_put_decimal_ll(m, "AVX512_elapsed_ms:\t", delta); + seq_putc(m, '\n'); +} + +/* + * Report architecture specific information + */ +int proc_pid_arch_status(struct seq_file *m, struct pid_namespace *ns, + struct pid *pid, struct task_struct *task) +{ + struct mm_struct *mm; + unsigned long untag_mask =3D -1UL; + + /* + * Report AVX512 state if the processor and build option supported. + */ + if (cpu_feature_enabled(X86_FEATURE_AVX512F)) + avx512_status(m, task); + + mm =3D get_task_mm(task); + if (mm) { + untag_mask =3D mm_untag_mask(task->mm); + mmput(mm); + } + + seq_printf(m, "untag_mask:\t%#lx\n", untag_mask); + + return 0; +} --=20 2.35.1 From nobody Sat Apr 11 06:41:05 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 1BD3BC00140 for ; Mon, 15 Aug 2022 04:15:57 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S230479AbiHOEPg (ORCPT ); Mon, 15 Aug 2022 00:15:36 -0400 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:44138 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S240419AbiHOEPW (ORCPT ); Mon, 15 Aug 2022 00:15:22 -0400 Received: from mga07.intel.com (mga07.intel.com [134.134.136.100]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id 0978114081 for ; Sun, 14 Aug 2022 21:15:12 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=intel.com; i=@intel.com; q=dns/txt; s=Intel; t=1660536912; x=1692072912; h=from:to:cc:subject:date:message-id:in-reply-to: references:mime-version:content-transfer-encoding; bh=3EjyBFdwK4uZ3bVM+MhvlpEXA4APOBBafZji25n6VHM=; b=ibPi8DCBzoQNkYSTV/pbRR/bj/cvRVb39nkYRlh4T2GV1Fr/K93zGfhu u6at5lMlfuRmU+f2YO5wT+TOZ39eKQy9aohMzCO4f3vc0R1nmKyb4+7xd scnUWvxXQ9hwM1y9/AhYnyUokpQZkUBWtsY4nQPX66dZhej7yoBuMa+wd r/6hhfFSlhKsRtFrIAytP+xxyJOwITmuCKSvKjTQR2GoOJp3tcS1Viv6y ovqpmHAtyCaRTqUkHVNwiF7LCsLHmJQMNMA+VSIFwMDE1Qiuf2hlkorod pV/2ga90ob3W6FwfmjLU+yeqwC7TYsNmOtvFPDPQSbkbIacMoWBNKztYh Q==; X-IronPort-AV: E=McAfee;i="6400,9594,10439"; a="355892012" X-IronPort-AV: E=Sophos;i="5.93,237,1654585200"; d="scan'208";a="355892012" Received: from fmsmga004.fm.intel.com ([10.253.24.48]) by orsmga105.jf.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 14 Aug 2022 21:15:11 -0700 X-IronPort-AV: E=Sophos;i="5.93,237,1654585200"; d="scan'208";a="674717535" Received: from dmalka-mobl.ger.corp.intel.com (HELO box.shutemov.name) ([10.249.42.26]) by fmsmga004-auth.fm.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 14 Aug 2022 21:15:07 -0700 Received: by box.shutemov.name (Postfix, from userid 1000) id 7C3F5104A6D; Mon, 15 Aug 2022 07:18:04 +0300 (+03) From: "Kirill A. Shutemov" To: Dave Hansen , Andy Lutomirski , Peter Zijlstra Cc: x86@kernel.org, Kostya Serebryany , Andrey Ryabinin , Andrey Konovalov , Alexander Potapenko , Taras Madan , Dmitry Vyukov , "H . J . Lu" , Andi Kleen , Rick Edgecombe , linux-mm@kvack.org, linux-kernel@vger.kernel.org, Weihong Zhang , "Kirill A . Shutemov" Subject: [PATCHv6 08/11] selftests/x86/lam: Add malloc and tag-bits test cases for linear-address masking Date: Mon, 15 Aug 2022 07:18:00 +0300 Message-Id: <20220815041803.17954-9-kirill.shutemov@linux.intel.com> X-Mailer: git-send-email 2.35.1 In-Reply-To: <20220815041803.17954-1-kirill.shutemov@linux.intel.com> References: <20220815041803.17954-1-kirill.shutemov@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: Weihong Zhang LAM is supported only in 64-bit mode and applies only addresses used for da= ta accesses. In 64-bit mode, linear address have 64 bits. LAM is applied to 64= -bit linear address and allow software to use high bits for metadata. LAM supports configurations that differ regarding which pointer bits are ma= sked and can be used for metadata. LAM includes following mode: - LAM_U57, pointer bits in positions 62:57 are masked (LAM width 6), allows bits 62:57 of a user pointer to be used as metadata. There are two arch_prctls: ARCH_ENABLE_TAGGED_ADDR: enable LAM mode, mask high bits of a user pointer. ARCH_GET_UNTAG_MASK: get current untagged mask. ARCH_GET_MAX_TAG_BITS: the maximum tag bits user can request. zero if LAM is not supported. The LAM mode is for pre-process, a process has only one chance to set LAM m= ode. But there is no API to disable LAM mode. So all of test cases are run under child process. Functions of this test: MALLOC - LAM_U57 masks bits 57:62 of a user pointer. Process on user space can dereference such pointers. - Disable LAM, dereference a pointer with metadata above 48 bit or 57 bit lead to trigger SIGSEGV. TAG_BITS - Max tag bits of LAM_U57 is 6. Signed-off-by: Weihong Zhang Signed-off-by: Kirill A. Shutemov Acked-by: Peter Zijlstra (Intel) --- tools/testing/selftests/x86/Makefile | 2 +- tools/testing/selftests/x86/lam.c | 317 +++++++++++++++++++++++++++ 2 files changed, 318 insertions(+), 1 deletion(-) create mode 100644 tools/testing/selftests/x86/lam.c diff --git a/tools/testing/selftests/x86/Makefile b/tools/testing/selftests= /x86/Makefile index 0388c4d60af0..c1a16a9d4f2f 100644 --- a/tools/testing/selftests/x86/Makefile +++ b/tools/testing/selftests/x86/Makefile @@ -18,7 +18,7 @@ TARGETS_C_32BIT_ONLY :=3D entry_from_vm86 test_syscall_vd= so unwind_vdso \ test_FCMOV test_FCOMI test_FISTTP \ vdso_restorer TARGETS_C_64BIT_ONLY :=3D fsgsbase sysret_rip syscall_numbering \ - corrupt_xstate_header amx + corrupt_xstate_header amx lam # Some selftests require 32bit support enabled also on 64bit systems TARGETS_C_32BIT_NEEDED :=3D ldt_gdt ptrace_syscall =20 diff --git a/tools/testing/selftests/x86/lam.c b/tools/testing/selftests/x8= 6/lam.c new file mode 100644 index 000000000000..4c6c6dbf7db6 --- /dev/null +++ b/tools/testing/selftests/x86/lam.c @@ -0,0 +1,317 @@ +// SPDX-License-Identifier: GPL-2.0 +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#include "../kselftest.h" + +#ifndef __x86_64__ +# error This test is 64-bit only +#endif + +/* LAM modes, these definitions were copied from kernel code */ +#define LAM_NONE 0 +#define LAM_U57_BITS 6 +/* arch prctl for LAM */ +#define ARCH_GET_UNTAG_MASK 0x4001 +#define ARCH_ENABLE_TAGGED_ADDR 0x4002 +#define ARCH_GET_MAX_TAG_BITS 0x4003 + +/* Specified test function bits */ +#define FUNC_MALLOC 0x1 +#define FUNC_BITS 0x2 + +#define TEST_MASK 0x3 + +#define MALLOC_LEN 32 + +struct testcases { + unsigned int later; + int expected; /* 2: SIGSEGV Error; 1: other errors */ + unsigned long lam; + uint64_t addr; + int (*test_func)(struct testcases *test); + const char *msg; +}; + +int tests_cnt; +jmp_buf segv_env; + +static void segv_handler(int sig) +{ + ksft_print_msg("Get segmentation fault(%d).", sig); + siglongjmp(segv_env, 1); +} + +static inline int cpu_has_lam(void) +{ + unsigned int cpuinfo[4]; + + __cpuid_count(0x7, 1, cpuinfo[0], cpuinfo[1], cpuinfo[2], cpuinfo[3]); + + return (cpuinfo[0] & (1 << 26)); +} + +/* + * Set tagged address and read back untag mask. + * check if the untagged mask is expected. + */ +static int set_lam(unsigned long lam) +{ + int ret =3D 0; + uint64_t ptr =3D 0; + + if (lam !=3D LAM_U57_BITS && lam !=3D LAM_NONE) + return -1; + + /* Skip check return */ + syscall(SYS_arch_prctl, ARCH_ENABLE_TAGGED_ADDR, lam); + + /* Get untagged mask */ + syscall(SYS_arch_prctl, ARCH_GET_UNTAG_MASK, &ptr); + + /* Check mask returned is expected */ + if (lam =3D=3D LAM_U57_BITS) + ret =3D (ptr !=3D ~(0x3fULL << 57)); + else if (lam =3D=3D LAM_NONE) + ret =3D (ptr !=3D -1ULL); + + return ret; +} + +static unsigned long get_default_tag_bits(void) +{ + pid_t pid; + int lam =3D LAM_NONE; + int ret =3D 0; + + pid =3D fork(); + if (pid < 0) { + perror("Fork failed."); + ret =3D 1; + } else if (pid =3D=3D 0) { + /* Set LAM mode in parent process */ + if (set_lam(LAM_U57_BITS) =3D=3D 0) + lam =3D LAM_U57_BITS; + else + lam =3D LAM_NONE; + exit(lam); + } else { + wait(&ret); + lam =3D WEXITSTATUS(ret); + } + + return lam; +} + +/* According to LAM mode, set metadata in high bits */ +static uint64_t get_metadata(uint64_t src, unsigned long lam) +{ + uint64_t metadata; + + srand(time(NULL)); + /* Get a random value as metadata */ + metadata =3D rand(); + + switch (lam) { + case LAM_U57_BITS: /* Set metadata in bits 62:57 */ + metadata =3D (src & ~(0x3fULL << 57)) | ((metadata & 0x3f) << 57); + break; + default: + metadata =3D src; + break; + } + + return metadata; +} + +/* + * Set metadata in user pointer, compare new pointer with original pointer. + * both pointers should point to the same address. + */ +static int handle_lam_test(void *src, unsigned int lam) +{ + char *ptr; + + strcpy((char *)src, "USER POINTER"); + + ptr =3D (char *)get_metadata((uint64_t)src, lam); + if (src =3D=3D ptr) + return 0; + + /* Copy a string into the pointer with metadata */ + strcpy((char *)ptr, "METADATA POINTER"); + + return (!!strcmp((char *)src, (char *)ptr)); +} + + +int handle_max_bits(struct testcases *test) +{ + unsigned long exp_bits =3D get_default_tag_bits(); + unsigned long bits =3D 0; + + if (exp_bits !=3D LAM_NONE) + exp_bits =3D LAM_U57_BITS; + + /* Get LAM max tag bits */ + if (syscall(SYS_arch_prctl, ARCH_GET_MAX_TAG_BITS, &bits) =3D=3D -1) + return 1; + + return (exp_bits !=3D bits); +} + +/* + * Test lam feature through dereference pointer get from malloc. + * @return 0: Pass test. 1: Get failure during test 2: Get SIGSEGV + */ +static int handle_malloc(struct testcases *test) +{ + char *ptr =3D NULL; + int ret =3D 0; + + if (test->later =3D=3D 0 && test->lam !=3D 0) + if (set_lam(test->lam) =3D=3D -1) + return 1; + + ptr =3D (char *)malloc(MALLOC_LEN); + if (ptr =3D=3D NULL) { + perror("malloc() failure\n"); + return 1; + } + + /* Set signal handler */ + if (sigsetjmp(segv_env, 1) =3D=3D 0) { + signal(SIGSEGV, segv_handler); + ret =3D handle_lam_test(ptr, test->lam); + } else { + ret =3D 2; + } + + if (test->later !=3D 0 && test->lam !=3D 0) + if (set_lam(test->lam) =3D=3D -1 && ret =3D=3D 0) + ret =3D 1; + + free(ptr); + + return ret; +} + +static int fork_test(struct testcases *test) +{ + int ret, child_ret; + pid_t pid; + + pid =3D fork(); + if (pid < 0) { + perror("Fork failed."); + ret =3D 1; + } else if (pid =3D=3D 0) { + ret =3D test->test_func(test); + exit(ret); + } else { + wait(&child_ret); + ret =3D WEXITSTATUS(child_ret); + } + + return ret; +} + +static void run_test(struct testcases *test, int count) +{ + int i, ret =3D 0; + + for (i =3D 0; i < count; i++) { + struct testcases *t =3D test + i; + + /* fork a process to run test case */ + ret =3D fork_test(t); + if (ret !=3D 0) + ret =3D (t->expected =3D=3D ret); + else + ret =3D !(t->expected); + + tests_cnt++; + ksft_test_result(ret, t->msg); + } +} + +static struct testcases malloc_cases[] =3D { + { + .later =3D 0, + .lam =3D LAM_U57_BITS, + .test_func =3D handle_malloc, + .msg =3D "MALLOC: LAM_U57. Dereferencing pointer with metadata\n", + }, + { + .later =3D 1, + .expected =3D 2, + .lam =3D LAM_U57_BITS, + .test_func =3D handle_malloc, + .msg =3D "MALLOC:[Negtive] Disable LAM. Dereferencing pointer with metad= ata.\n", + }, +}; + + +static struct testcases bits_cases[] =3D { + { + .test_func =3D handle_max_bits, + .msg =3D "BITS: Check default tag bits\n", + }, +}; + +static void cmd_help(void) +{ + printf("usage: lam [-h] [-t test list]\n"); + printf("\t-t test list: run tests specified in the test list, default:0x%= x\n", TEST_MASK); + printf("\t\t0x1:malloc; 0x2:max_bits;\n"); + printf("\t-h: help\n"); +} + +int main(int argc, char **argv) +{ + int c =3D 0; + unsigned int tests =3D TEST_MASK; + + tests_cnt =3D 0; + + if (!cpu_has_lam()) { + ksft_print_msg("Unsupported LAM feature!\n"); + return -1; + } + + while ((c =3D getopt(argc, argv, "ht:")) !=3D -1) { + switch (c) { + case 't': + tests =3D strtoul(optarg, NULL, 16); + if (!(tests & TEST_MASK)) { + ksft_print_msg("Invalid argument!\n"); + return -1; + } + break; + case 'h': + cmd_help(); + return 0; + default: + ksft_print_msg("Invalid argument\n"); + return -1; + } + } + + if (tests & FUNC_MALLOC) + run_test(malloc_cases, ARRAY_SIZE(malloc_cases)); + + if (tests & FUNC_BITS) + run_test(bits_cases, ARRAY_SIZE(bits_cases)); + + ksft_set_plan(tests_cnt); + + return ksft_exit_pass(); +} --=20 2.35.1 From nobody Sat Apr 11 06:41:05 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 EC8E9C25B06 for ; Mon, 15 Aug 2022 04:15:57 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S240954AbiHOEPp (ORCPT ); Mon, 15 Aug 2022 00:15:45 -0400 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:44162 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S240434AbiHOEPW (ORCPT ); Mon, 15 Aug 2022 00:15:22 -0400 Received: from mga01.intel.com (mga01.intel.com [192.55.52.88]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id 6ABE31402F for ; Sun, 14 Aug 2022 21:15:12 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=intel.com; i=@intel.com; q=dns/txt; s=Intel; t=1660536912; x=1692072912; h=from:to:cc:subject:date:message-id:in-reply-to: references:mime-version:content-transfer-encoding; bh=pFEmrndyS4U/nV5J9Y9zEWLJbyzb9Y8r4SpnzcrYa1c=; b=kVOPo4z/TMt+gf3Ia4AJQPYwSFz9UHdxLks3tbbYVMRymLs+oo4F+W7H Lr6KUee/V6KLiHoRX3aO1LrwlfyN3A4BH4n0fDGZd7lcak8TCB/jwExD9 2FtRFa8wgWEaDD1dHmg9e7PgpqW2pAg2NkjCzWXw7UlFDssQTAOeQehiB jdwVXGa3Frb7zEuLHmNVziKJ/jvuNUz9kGZ9pEvqf/DLhoSZs6dyHY6lr 489/JMV5JAHnmsAORjmZBiO2yCfD3T/6l2/8zrDK//sAbNYKvu5Etkj1w d3QuInF6/6FT4BPG94XSJ/L5yB43gPy3gIUbs+Hsa723T5C9A2IQyfLdP w==; X-IronPort-AV: E=McAfee;i="6400,9594,10439"; a="317867216" X-IronPort-AV: E=Sophos;i="5.93,237,1654585200"; d="scan'208";a="317867216" Received: from orsmga004.jf.intel.com ([10.7.209.38]) by fmsmga101.fm.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 14 Aug 2022 21:15:11 -0700 X-IronPort-AV: E=Sophos;i="5.93,237,1654585200"; d="scan'208";a="732818295" Received: from dmalka-mobl.ger.corp.intel.com (HELO box.shutemov.name) ([10.249.42.26]) by orsmga004-auth.jf.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 14 Aug 2022 21:15:07 -0700 Received: by box.shutemov.name (Postfix, from userid 1000) id 866D2104A6E; Mon, 15 Aug 2022 07:18:04 +0300 (+03) From: "Kirill A. Shutemov" To: Dave Hansen , Andy Lutomirski , Peter Zijlstra Cc: x86@kernel.org, Kostya Serebryany , Andrey Ryabinin , Andrey Konovalov , Alexander Potapenko , Taras Madan , Dmitry Vyukov , "H . J . Lu" , Andi Kleen , Rick Edgecombe , linux-mm@kvack.org, linux-kernel@vger.kernel.org, Weihong Zhang , "Kirill A . Shutemov" Subject: [PATCHv6 09/11] selftests/x86/lam: Add mmap and SYSCALL test cases for linear-address masking Date: Mon, 15 Aug 2022 07:18:01 +0300 Message-Id: <20220815041803.17954-10-kirill.shutemov@linux.intel.com> X-Mailer: git-send-email 2.35.1 In-Reply-To: <20220815041803.17954-1-kirill.shutemov@linux.intel.com> References: <20220815041803.17954-1-kirill.shutemov@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: Weihong Zhang Add mmap and SYSCALL test cases. SYSCALL test cases: - LAM supports set metadata in high bits 62:57 (LAM_U57) of a user pointer= , pass the pointer to SYSCALL, SYSCALL can dereference the pointer and return c= orrect result. - Disable LAM, pass a pointer with metadata in high bits to SYSCALL, SYSCALL returns -1 (EFAULT). MMAP test cases: - Enable LAM_U57, MMAP with low address (below bits 47), set metadata in high bits of the address, dereference the address should be allowed. - Enable LAM_U57, MMAP with high address (above bits 47), set metadata in high bits of the address, dereference the address should be allowed. Signed-off-by: Weihong Zhang Signed-off-by: Kirill A. Shutemov Acked-by: Peter Zijlstra (Intel) --- tools/testing/selftests/x86/lam.c | 135 +++++++++++++++++++++++++++++- 1 file changed, 132 insertions(+), 3 deletions(-) diff --git a/tools/testing/selftests/x86/lam.c b/tools/testing/selftests/x8= 6/lam.c index 4c6c6dbf7db6..e9e92ab7fea8 100644 --- a/tools/testing/selftests/x86/lam.c +++ b/tools/testing/selftests/x86/lam.c @@ -7,6 +7,7 @@ #include #include #include +#include #include #include =20 @@ -27,11 +28,18 @@ /* Specified test function bits */ #define FUNC_MALLOC 0x1 #define FUNC_BITS 0x2 +#define FUNC_MMAP 0x4 +#define FUNC_SYSCALL 0x8 =20 -#define TEST_MASK 0x3 +#define TEST_MASK 0xf + +#define LOW_ADDR (0x1UL << 30) +#define HIGH_ADDR (0x3UL << 48) =20 #define MALLOC_LEN 32 =20 +#define PAGE_SIZE (4 << 10) + struct testcases { unsigned int later; int expected; /* 2: SIGSEGV Error; 1: other errors */ @@ -47,6 +55,7 @@ jmp_buf segv_env; static void segv_handler(int sig) { ksft_print_msg("Get segmentation fault(%d).", sig); + siglongjmp(segv_env, 1); } =20 @@ -59,6 +68,16 @@ static inline int cpu_has_lam(void) return (cpuinfo[0] & (1 << 26)); } =20 +/* Check 5-level page table feature in CPUID.(EAX=3D07H, ECX=3D00H):ECX.[b= it 16] */ +static inline int cpu_has_la57(void) +{ + unsigned int cpuinfo[4]; + + __cpuid_count(0x7, 0, cpuinfo[0], cpuinfo[1], cpuinfo[2], cpuinfo[3]); + + return (cpuinfo[2] & (1 << 16)); +} + /* * Set tagged address and read back untag mask. * check if the untagged mask is expected. @@ -204,6 +223,68 @@ static int handle_malloc(struct testcases *test) return ret; } =20 +static int handle_mmap(struct testcases *test) +{ + void *ptr; + unsigned int flags =3D MAP_PRIVATE | MAP_ANONYMOUS | MAP_FIXED; + int ret =3D 0; + + if (test->later =3D=3D 0 && test->lam !=3D 0) + if (set_lam(test->lam) !=3D 0) + return 1; + + ptr =3D mmap((void *)test->addr, PAGE_SIZE, PROT_READ | PROT_WRITE, + flags, -1, 0); + if (ptr =3D=3D MAP_FAILED) { + if (test->addr =3D=3D HIGH_ADDR) + if (!cpu_has_la57()) + return 3; /* unsupport LA57 */ + return 1; + } + + if (test->later !=3D 0 && test->lam !=3D 0) + if (set_lam(test->lam) !=3D 0) + ret =3D 1; + + if (ret =3D=3D 0) { + if (sigsetjmp(segv_env, 1) =3D=3D 0) { + signal(SIGSEGV, segv_handler); + ret =3D handle_lam_test(ptr, test->lam); + } else { + ret =3D 2; + } + } + + munmap(ptr, PAGE_SIZE); + return ret; +} + +static int handle_syscall(struct testcases *test) +{ + struct utsname unme, *pu; + int ret =3D 0; + + if (test->later =3D=3D 0 && test->lam !=3D 0) + if (set_lam(test->lam) !=3D 0) + return 1; + + if (sigsetjmp(segv_env, 1) =3D=3D 0) { + signal(SIGSEGV, segv_handler); + pu =3D (struct utsname *)get_metadata((uint64_t)&unme, test->lam); + ret =3D uname(pu); + if (ret < 0) + ret =3D 1; + } else { + ret =3D 2; + } + + if (test->later !=3D 0 && test->lam !=3D 0) + if (set_lam(test->lam) !=3D -1 && ret =3D=3D 0) + ret =3D 1; + + return ret; +} + static int fork_test(struct testcases *test) { int ret, child_ret; @@ -259,7 +340,6 @@ static struct testcases malloc_cases[] =3D { }, }; =20 - static struct testcases bits_cases[] =3D { { .test_func =3D handle_max_bits, @@ -267,11 +347,54 @@ static struct testcases bits_cases[] =3D { }, }; =20 +static struct testcases syscall_cases[] =3D { + { + .later =3D 0, + .lam =3D LAM_U57_BITS, + .test_func =3D handle_syscall, + .msg =3D "SYSCALL: LAM_U57. syscall with metadata\n", + }, + { + .later =3D 1, + .expected =3D 1, + .lam =3D LAM_U57_BITS, + .test_func =3D handle_syscall, + .msg =3D "SYSCALL:[Negtive] Disable LAM. Dereferencing pointer with meta= data.\n", + }, +}; + +static struct testcases mmap_cases[] =3D { + { + .later =3D 1, + .expected =3D 0, + .lam =3D LAM_U57_BITS, + .addr =3D HIGH_ADDR, + .test_func =3D handle_mmap, + .msg =3D "MMAP: First mmap high address, then set LAM_U57.\n", + }, + { + .later =3D 0, + .expected =3D 0, + .lam =3D LAM_U57_BITS, + .addr =3D HIGH_ADDR, + .test_func =3D handle_mmap, + .msg =3D "MMAP: First LAM_U57, then High address.\n", + }, + { + .later =3D 0, + .expected =3D 0, + .lam =3D LAM_U57_BITS, + .addr =3D LOW_ADDR, + .test_func =3D handle_mmap, + .msg =3D "MMAP: First LAM_U57, then Low address.\n", + }, +}; + static void cmd_help(void) { printf("usage: lam [-h] [-t test list]\n"); printf("\t-t test list: run tests specified in the test list, default:0x%= x\n", TEST_MASK); - printf("\t\t0x1:malloc; 0x2:max_bits;\n"); + printf("\t\t0x1:malloc; 0x2:max_bits; 0x4:mmap; 0x8:syscall.\n"); printf("\t-h: help\n"); } =20 @@ -311,6 +434,12 @@ int main(int argc, char **argv) if (tests & FUNC_BITS) run_test(bits_cases, ARRAY_SIZE(bits_cases)); =20 + if (tests & FUNC_MMAP) + run_test(mmap_cases, ARRAY_SIZE(mmap_cases)); + + if (tests & FUNC_SYSCALL) + run_test(syscall_cases, ARRAY_SIZE(syscall_cases)); + ksft_set_plan(tests_cnt); =20 return ksft_exit_pass(); --=20 2.35.1 From nobody Sat Apr 11 06:41:05 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 17B12C25B0E for ; Mon, 15 Aug 2022 04:15:58 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S241034AbiHOEPu (ORCPT ); Mon, 15 Aug 2022 00:15:50 -0400 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:44176 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S240605AbiHOEPW (ORCPT ); Mon, 15 Aug 2022 00:15:22 -0400 Received: from mga01.intel.com (mga01.intel.com [192.55.52.88]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id 41DFE14088 for ; Sun, 14 Aug 2022 21:15:13 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=intel.com; i=@intel.com; q=dns/txt; s=Intel; t=1660536913; x=1692072913; h=from:to:cc:subject:date:message-id:in-reply-to: references:mime-version:content-transfer-encoding; bh=P+SYU9cTeui4ntbsYjOdd9WQd3XioWrSFFVkN1SOniI=; b=cQBPOsdlmNayMNpdOztA9Zw/Z37fKqM56jruxqlPG8Otz0yAmmdO8YAu xDd786FbkuZXZ/CaTRELtgJ0HfisukrqUt8nWMt2kXVjR9KKpA8ZQqlHH 7rySO6Zu3w4c/ikafDdFt1jSWhmO3cQIFEo0sQzArsGCvj9pFgJy7jKMq oMvdTvaMnBkA+2kBzqpm4IjGJdcjSj+7SxAQNsl+bbSdyBhskFYdw5NIC OfNsnUoenYVMLNVAATan4tbp368X48kuaZHfmreHPKHbNzTMaZXQ9vI1+ +ELONO5cH8PteII4IdFDBxr3cF7a6k5ob2ZrROKMdq4Ut19bIgsGFiY7I A==; X-IronPort-AV: E=McAfee;i="6400,9594,10439"; a="317867219" X-IronPort-AV: E=Sophos;i="5.93,237,1654585200"; d="scan'208";a="317867219" Received: from orsmga004.jf.intel.com ([10.7.209.38]) by fmsmga101.fm.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 14 Aug 2022 21:15:12 -0700 X-IronPort-AV: E=Sophos;i="5.93,237,1654585200"; d="scan'208";a="732818298" Received: from dmalka-mobl.ger.corp.intel.com (HELO box.shutemov.name) ([10.249.42.26]) by orsmga004-auth.jf.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 14 Aug 2022 21:15:07 -0700 Received: by box.shutemov.name (Postfix, from userid 1000) id 90A89104A6F; Mon, 15 Aug 2022 07:18:04 +0300 (+03) From: "Kirill A. Shutemov" To: Dave Hansen , Andy Lutomirski , Peter Zijlstra Cc: x86@kernel.org, Kostya Serebryany , Andrey Ryabinin , Andrey Konovalov , Alexander Potapenko , Taras Madan , Dmitry Vyukov , "H . J . Lu" , Andi Kleen , Rick Edgecombe , linux-mm@kvack.org, linux-kernel@vger.kernel.org, Weihong Zhang , "Kirill A . Shutemov" Subject: [PATCHv6 10/11] selftests/x86/lam: Add io_uring test cases for linear-address masking Date: Mon, 15 Aug 2022 07:18:02 +0300 Message-Id: <20220815041803.17954-11-kirill.shutemov@linux.intel.com> X-Mailer: git-send-email 2.35.1 In-Reply-To: <20220815041803.17954-1-kirill.shutemov@linux.intel.com> References: <20220815041803.17954-1-kirill.shutemov@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: Weihong Zhang LAM should be supported in kernel thread, using io_uring to verify LAM feat= ure. The test cases implement read a file through io_uring, the test cases choos= e an iovec array as receiving buffer, which used to receive data, according to L= AM mode, set metadata in high bits of these buffer. io_uring can deal with these buffers that pointed to pointers with the meta= data in high bits. Signed-off-by: Weihong Zhang Signed-off-by: Kirill A. Shutemov Acked-by: Peter Zijlstra (Intel) --- tools/testing/selftests/x86/lam.c | 341 +++++++++++++++++++++++++++++- 1 file changed, 339 insertions(+), 2 deletions(-) diff --git a/tools/testing/selftests/x86/lam.c b/tools/testing/selftests/x8= 6/lam.c index e9e92ab7fea8..19208ba6047a 100644 --- a/tools/testing/selftests/x86/lam.c +++ b/tools/testing/selftests/x86/lam.c @@ -9,8 +9,12 @@ #include #include #include +#include +#include #include =20 +#include +#include #include "../kselftest.h" =20 #ifndef __x86_64__ @@ -30,8 +34,9 @@ #define FUNC_BITS 0x2 #define FUNC_MMAP 0x4 #define FUNC_SYSCALL 0x8 +#define FUNC_URING 0x10 =20 -#define TEST_MASK 0xf +#define TEST_MASK 0x1f =20 #define LOW_ADDR (0x1UL << 30) #define HIGH_ADDR (0x3UL << 48) @@ -40,6 +45,13 @@ =20 #define PAGE_SIZE (4 << 10) =20 +#define barrier() ({ \ + __asm__ __volatile__("" : : : "memory"); \ +}) + +#define URING_QUEUE_SZ 1 +#define URING_BLOCK_SZ 2048 + struct testcases { unsigned int later; int expected; /* 2: SIGSEGV Error; 1: other errors */ @@ -49,6 +61,33 @@ struct testcases { const char *msg; }; =20 +/* Used by CQ of uring, source file handler and file's size */ +struct file_io { + int file_fd; + off_t file_sz; + struct iovec iovecs[]; +}; + +struct io_uring_queue { + unsigned int *head; + unsigned int *tail; + unsigned int *ring_mask; + unsigned int *ring_entries; + unsigned int *flags; + unsigned int *array; + union { + struct io_uring_cqe *cqes; + struct io_uring_sqe *sqes; + } queue; + size_t ring_sz; +}; + +struct io_ring { + int ring_fd; + struct io_uring_queue sq_ring; + struct io_uring_queue cq_ring; +}; + int tests_cnt; jmp_buf segv_env; =20 @@ -285,6 +324,285 @@ static int handle_syscall(struct testcases *test) return ret; } =20 +int sys_uring_setup(unsigned int entries, struct io_uring_params *p) +{ + return (int)syscall(__NR_io_uring_setup, entries, p); +} + +int sys_uring_enter(int fd, unsigned int to, unsigned int min, unsigned in= t flags) +{ + return (int)syscall(__NR_io_uring_enter, fd, to, min, flags, NULL, 0); +} + +/* Init submission queue and completion queue */ +int mmap_io_uring(struct io_uring_params p, struct io_ring *s) +{ + struct io_uring_queue *sring =3D &s->sq_ring; + struct io_uring_queue *cring =3D &s->cq_ring; + + sring->ring_sz =3D p.sq_off.array + p.sq_entries * sizeof(unsigned int); + cring->ring_sz =3D p.cq_off.cqes + p.cq_entries * sizeof(struct io_uring_= cqe); + + if (p.features & IORING_FEAT_SINGLE_MMAP) { + if (cring->ring_sz > sring->ring_sz) + sring->ring_sz =3D cring->ring_sz; + + cring->ring_sz =3D sring->ring_sz; + } + + void *sq_ptr =3D mmap(0, sring->ring_sz, PROT_READ | PROT_WRITE, + MAP_SHARED | MAP_POPULATE, s->ring_fd, + IORING_OFF_SQ_RING); + + if (sq_ptr =3D=3D MAP_FAILED) { + perror("sub-queue!"); + return 1; + } + + void *cq_ptr =3D sq_ptr; + + if (!(p.features & IORING_FEAT_SINGLE_MMAP)) { + cq_ptr =3D mmap(0, cring->ring_sz, PROT_READ | PROT_WRITE, + MAP_SHARED | MAP_POPULATE, s->ring_fd, + IORING_OFF_CQ_RING); + if (cq_ptr =3D=3D MAP_FAILED) { + perror("cpl-queue!"); + munmap(sq_ptr, sring->ring_sz); + return 1; + } + } + + sring->head =3D sq_ptr + p.sq_off.head; + sring->tail =3D sq_ptr + p.sq_off.tail; + sring->ring_mask =3D sq_ptr + p.sq_off.ring_mask; + sring->ring_entries =3D sq_ptr + p.sq_off.ring_entries; + sring->flags =3D sq_ptr + p.sq_off.flags; + sring->array =3D sq_ptr + p.sq_off.array; + + /* Map a queue as mem map */ + s->sq_ring.queue.sqes =3D mmap(0, p.sq_entries * sizeof(struct io_uring_s= qe), + PROT_READ | PROT_WRITE, MAP_SHARED | MAP_POPULATE, + s->ring_fd, IORING_OFF_SQES); + if (s->sq_ring.queue.sqes =3D=3D MAP_FAILED) { + munmap(sq_ptr, sring->ring_sz); + if (sq_ptr !=3D cq_ptr) { + ksft_print_msg("failed to mmap uring queue!"); + munmap(cq_ptr, cring->ring_sz); + return 1; + } + } + + cring->head =3D cq_ptr + p.cq_off.head; + cring->tail =3D cq_ptr + p.cq_off.tail; + cring->ring_mask =3D cq_ptr + p.cq_off.ring_mask; + cring->ring_entries =3D cq_ptr + p.cq_off.ring_entries; + cring->queue.cqes =3D cq_ptr + p.cq_off.cqes; + + return 0; +} + +/* Init io_uring queues */ +int setup_io_uring(struct io_ring *s) +{ + struct io_uring_params para; + + memset(¶, 0, sizeof(para)); + s->ring_fd =3D sys_uring_setup(URING_QUEUE_SZ, ¶); + if (s->ring_fd < 0) + return 1; + + return mmap_io_uring(para, s); +} + +/* + * Get data from completion queue. the data buffer saved the file data + * return 0: success; others: error; + */ +int handle_uring_cq(struct io_ring *s) +{ + struct file_io *fi =3D NULL; + struct io_uring_queue *cring =3D &s->cq_ring; + struct io_uring_cqe *cqe; + unsigned int head; + off_t len =3D 0; + + head =3D *cring->head; + + do { + barrier(); + if (head =3D=3D *cring->tail) + break; + /* Get the entry */ + cqe =3D &cring->queue.cqes[head & *s->cq_ring.ring_mask]; + fi =3D (struct file_io *)cqe->user_data; + if (cqe->res < 0) + break; + + int blocks =3D (int)(fi->file_sz + URING_BLOCK_SZ - 1) / URING_BLOCK_SZ; + + for (int i =3D 0; i < blocks; i++) + len +=3D fi->iovecs[i].iov_len; + + head++; + } while (1); + + *cring->head =3D head; + barrier(); + + return (len !=3D fi->file_sz); +} + +/* + * Submit squeue. specify via IORING_OP_READV. + * the buffer need to be set metadata according to LAM mode + */ +int handle_uring_sq(struct io_ring *ring, struct file_io *fi, unsigned lon= g lam) +{ + int file_fd =3D fi->file_fd; + struct io_uring_queue *sring =3D &ring->sq_ring; + unsigned int index =3D 0, cur_block =3D 0, tail =3D 0, next_tail =3D 0; + struct io_uring_sqe *sqe; + + off_t remain =3D fi->file_sz; + int blocks =3D (int)(remain + URING_BLOCK_SZ - 1) / URING_BLOCK_SZ; + + while (remain) { + off_t bytes =3D remain; + void *buf; + + if (bytes > URING_BLOCK_SZ) + bytes =3D URING_BLOCK_SZ; + + fi->iovecs[cur_block].iov_len =3D bytes; + + if (posix_memalign(&buf, URING_BLOCK_SZ, URING_BLOCK_SZ)) + return 1; + + fi->iovecs[cur_block].iov_base =3D (void *)get_metadata((uint64_t)buf, l= am); + remain -=3D bytes; + cur_block++; + } + + next_tail =3D *sring->tail; + tail =3D next_tail; + next_tail++; + + barrier(); + + index =3D tail & *ring->sq_ring.ring_mask; + + sqe =3D &ring->sq_ring.queue.sqes[index]; + sqe->fd =3D file_fd; + sqe->flags =3D 0; + sqe->opcode =3D IORING_OP_READV; + sqe->addr =3D (unsigned long)fi->iovecs; + sqe->len =3D blocks; + sqe->off =3D 0; + sqe->user_data =3D (uint64_t)fi; + + sring->array[index] =3D index; + tail =3D next_tail; + + if (*sring->tail !=3D tail) { + *sring->tail =3D tail; + barrier(); + } + + if (sys_uring_enter(ring->ring_fd, 1, 1, IORING_ENTER_GETEVENTS) < 0) + return 1; + + return 0; +} + +/* + * Test LAM in async I/O and io_uring, read current binery through io_uring + * Set metadata in pointers to iovecs buffer. + */ +int do_uring(unsigned long lam) +{ + struct io_ring *ring; + struct file_io *fi; + struct stat st; + int ret =3D 1; + char path[PATH_MAX]; + + /* get current process path */ + if (readlink("/proc/self/exe", path, PATH_MAX) <=3D 0) + return 1; + + int file_fd =3D open(path, O_RDONLY); + + if (file_fd < 0) + return 1; + + if (fstat(file_fd, &st) < 0) + return 1; + + off_t file_sz =3D st.st_size; + + int blocks =3D (int)(file_sz + URING_BLOCK_SZ - 1) / URING_BLOCK_SZ; + + fi =3D malloc(sizeof(*fi) + sizeof(struct iovec) * blocks); + if (!fi) + return 1; + + fi->file_sz =3D file_sz; + fi->file_fd =3D file_fd; + + ring =3D malloc(sizeof(*ring)); + if (!ring) + return 1; + + memset(ring, 0, sizeof(struct io_ring)); + + if (setup_io_uring(ring)) + goto out; + + if (handle_uring_sq(ring, fi, lam)) + goto out; + + ret =3D handle_uring_cq(ring); + +out: + free(ring); + + for (int i =3D 0; i < blocks; i++) { + if (fi->iovecs[i].iov_base) { + uint64_t addr =3D ((uint64_t)fi->iovecs[i].iov_base); + + switch (lam) { + case LAM_U57_BITS: /* Clear bits 62:57 */ + addr =3D (addr & ~(0x3fULL << 57)); + break; + } + free((void *)addr); + fi->iovecs[i].iov_base =3D NULL; + } + } + + free(fi); + + return ret; +} + +int handle_uring(struct testcases *test) +{ + int ret =3D 0; + + if (test->later =3D=3D 0 && test->lam !=3D 0) + if (set_lam(test->lam) !=3D 0) + return 1; + + if (sigsetjmp(segv_env, 1) =3D=3D 0) { + signal(SIGSEGV, segv_handler); + ret =3D do_uring(test->lam); + } else { + ret =3D 2; + } + + return ret; +} + static int fork_test(struct testcases *test) { int ret, child_ret; @@ -324,6 +642,22 @@ static void run_test(struct testcases *test, int count) } } =20 +static struct testcases uring_cases[] =3D { + { + .later =3D 0, + .lam =3D LAM_U57_BITS, + .test_func =3D handle_uring, + .msg =3D "URING: LAM_U57. Dereferencing pointer with metadata\n", + }, + { + .later =3D 1, + .expected =3D 1, + .lam =3D LAM_U57_BITS, + .test_func =3D handle_uring, + .msg =3D "URING:[Negtive] Disable LAM. Dereferencing pointer with metada= ta.\n", + }, +}; + static struct testcases malloc_cases[] =3D { { .later =3D 0, @@ -394,7 +728,7 @@ static void cmd_help(void) { printf("usage: lam [-h] [-t test list]\n"); printf("\t-t test list: run tests specified in the test list, default:0x%= x\n", TEST_MASK); - printf("\t\t0x1:malloc; 0x2:max_bits; 0x4:mmap; 0x8:syscall.\n"); + printf("\t\t0x1:malloc; 0x2:max_bits; 0x4:mmap; 0x8:syscall; 0x10:io_urin= g.\n"); printf("\t-h: help\n"); } =20 @@ -440,6 +774,9 @@ int main(int argc, char **argv) if (tests & FUNC_SYSCALL) run_test(syscall_cases, ARRAY_SIZE(syscall_cases)); =20 + if (tests & FUNC_URING) + run_test(uring_cases, ARRAY_SIZE(uring_cases)); + ksft_set_plan(tests_cnt); =20 return ksft_exit_pass(); --=20 2.35.1 From nobody Sat Apr 11 06:41:05 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 264D5C25B06 for ; Mon, 15 Aug 2022 04:17:39 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S240996AbiHOEPr (ORCPT ); Mon, 15 Aug 2022 00:15:47 -0400 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:44164 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S240604AbiHOEPW (ORCPT ); Mon, 15 Aug 2022 00:15:22 -0400 Received: from mga12.intel.com (mga12.intel.com [192.55.52.136]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id 6CCAC14085 for ; Sun, 14 Aug 2022 21:15:12 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=intel.com; i=@intel.com; q=dns/txt; s=Intel; t=1660536912; x=1692072912; h=from:to:cc:subject:date:message-id:in-reply-to: references:mime-version:content-transfer-encoding; bh=IDZAMF3H0jOJUSFvgSQfeFD1dZ2psaQLK4sileStRHs=; b=GBIXTVBMC1mJPETz0E4WI9uAwJNGBy8dqLiM5IeaI7LZWDWlW97gRhcj jjgKxaWfPDiXG1w5Hq9wHoG7qPcA8DouvsXFHTXntvXYMfcsQQY1N/vLH ol3I70kB1AeMgBD4bdsESXLW20oBLl0QLKEri5v2IBHZ2e8ABpN8VxEeN bmLHRdYZYyUmHjwiWTAsOw7DNlJazrCGzGOd0PuYqiXJN0VLS88euRDoM lYhRiAPLhcLuqWl6UyN3d4y1N4jSpPkKXx+2eJdYwxSDEQARWF1Yqs/Ku oJ03yF8THjxwYpTNU8dTQH4QMTggqzgr/dO/+aJeGfNNAO97YM9i6FLKn A==; X-IronPort-AV: E=McAfee;i="6400,9594,10439"; a="271658919" X-IronPort-AV: E=Sophos;i="5.93,237,1654585200"; d="scan'208";a="271658919" Received: from orsmga008.jf.intel.com ([10.7.209.65]) by fmsmga106.fm.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 14 Aug 2022 21:15:12 -0700 X-IronPort-AV: E=Sophos;i="5.93,237,1654585200"; d="scan'208";a="635346906" Received: from dmalka-mobl.ger.corp.intel.com (HELO box.shutemov.name) ([10.249.42.26]) by orsmga008-auth.jf.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 14 Aug 2022 21:15:07 -0700 Received: by box.shutemov.name (Postfix, from userid 1000) id 9A162104A70; Mon, 15 Aug 2022 07:18:04 +0300 (+03) From: "Kirill A. Shutemov" To: Dave Hansen , Andy Lutomirski , Peter Zijlstra Cc: x86@kernel.org, Kostya Serebryany , Andrey Ryabinin , Andrey Konovalov , Alexander Potapenko , Taras Madan , Dmitry Vyukov , "H . J . Lu" , Andi Kleen , Rick Edgecombe , linux-mm@kvack.org, linux-kernel@vger.kernel.org, Weihong Zhang , "Kirill A . Shutemov" Subject: [PATCHv6 11/11] selftests/x86/lam: Add inherit test cases for linear-address masking Date: Mon, 15 Aug 2022 07:18:03 +0300 Message-Id: <20220815041803.17954-12-kirill.shutemov@linux.intel.com> X-Mailer: git-send-email 2.35.1 In-Reply-To: <20220815041803.17954-1-kirill.shutemov@linux.intel.com> References: <20220815041803.17954-1-kirill.shutemov@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: Weihong Zhang LAM is enabled per-thread and gets inherited on fork(2)/clone(2). exec() reverts LAM status to the default disabled state. There are two test scenarios: - Fork test cases: These cases were used to test the inheritance of LAM for per-thread, Child process generated by fork() should inherit LAM feature from parent process, Child process can get the LAM mode same as parent process. - Execve test cases: Processes generated by execve() are different from processes generated by fork(), these processes revert LAM status to disabled status. Signed-off-by: Weihong Zhang Signed-off-by: Kirill A. Shutemov Acked-by: Peter Zijlstra (Intel) --- tools/testing/selftests/x86/lam.c | 123 +++++++++++++++++++++++++++++- 1 file changed, 120 insertions(+), 3 deletions(-) diff --git a/tools/testing/selftests/x86/lam.c b/tools/testing/selftests/x8= 6/lam.c index 19208ba6047a..074fa0cd65ef 100644 --- a/tools/testing/selftests/x86/lam.c +++ b/tools/testing/selftests/x86/lam.c @@ -35,8 +35,9 @@ #define FUNC_MMAP 0x4 #define FUNC_SYSCALL 0x8 #define FUNC_URING 0x10 +#define FUNC_INHERITE 0x20 =20 -#define TEST_MASK 0x1f +#define TEST_MASK 0x3f =20 #define LOW_ADDR (0x1UL << 30) #define HIGH_ADDR (0x3UL << 48) @@ -169,6 +170,28 @@ static unsigned long get_default_tag_bits(void) return lam; } =20 +/* + * Set tagged address and read back untag mask. + * check if the untag mask is expected. + */ +static int get_lam(void) +{ + uint64_t ptr =3D 0; + int ret =3D -1; + /* Get untagged mask */ + if (syscall(SYS_arch_prctl, ARCH_GET_UNTAG_MASK, &ptr) =3D=3D -1) + return -1; + + /* Check mask returned is expected */ + if (ptr =3D=3D ~(0x3fULL << 57)) + ret =3D LAM_U57_BITS; + else if (ptr =3D=3D -1ULL) + ret =3D LAM_NONE; + + + return ret; +} + /* According to LAM mode, set metadata in high bits */ static uint64_t get_metadata(uint64_t src, unsigned long lam) { @@ -623,6 +646,72 @@ static int fork_test(struct testcases *test) return ret; } =20 +static int handle_execve(struct testcases *test) +{ + int ret, child_ret; + int lam =3D test->lam; + pid_t pid; + + pid =3D fork(); + if (pid < 0) { + perror("Fork failed."); + ret =3D 1; + } else if (pid =3D=3D 0) { + char path[PATH_MAX]; + + /* Set LAM mode in parent process */ + if (set_lam(lam) !=3D 0) + return 1; + + /* Get current binary's path and the binary was run by execve */ + if (readlink("/proc/self/exe", path, PATH_MAX) <=3D 0) + exit(-1); + + /* run binary to get LAM mode and return to parent process */ + if (execlp(path, path, "-t 0x0", NULL) < 0) { + perror("error on exec"); + exit(-1); + } + } else { + wait(&child_ret); + ret =3D WEXITSTATUS(child_ret); + if (ret !=3D LAM_NONE) + return 1; + } + + return 0; +} + +static int handle_inheritance(struct testcases *test) +{ + int ret, child_ret; + int lam =3D test->lam; + pid_t pid; + + /* Set LAM mode in parent process */ + if (set_lam(lam) !=3D 0) + return 1; + + pid =3D fork(); + if (pid < 0) { + perror("Fork failed."); + return 1; + } else if (pid =3D=3D 0) { + /* Set LAM mode in parent process */ + int child_lam =3D get_lam(); + + exit(child_lam); + } else { + wait(&child_ret); + ret =3D WEXITSTATUS(child_ret); + + if (lam !=3D ret) + return 1; + } + + return 0; +} + static void run_test(struct testcases *test, int count) { int i, ret =3D 0; @@ -724,11 +813,26 @@ static struct testcases mmap_cases[] =3D { }, }; =20 +static struct testcases inheritance_cases[] =3D { + { + .expected =3D 0, + .lam =3D LAM_U57_BITS, + .test_func =3D handle_inheritance, + .msg =3D "FORK: LAM_U57, child process should get LAM mode same as paren= t\n", + }, + { + .expected =3D 0, + .lam =3D LAM_U57_BITS, + .test_func =3D handle_execve, + .msg =3D "EXECVE: LAM_U57, child process should get disabled LAM mode\n", + }, +}; + static void cmd_help(void) { printf("usage: lam [-h] [-t test list]\n"); printf("\t-t test list: run tests specified in the test list, default:0x%= x\n", TEST_MASK); - printf("\t\t0x1:malloc; 0x2:max_bits; 0x4:mmap; 0x8:syscall; 0x10:io_urin= g.\n"); + printf("\t\t0x1:malloc; 0x2:max_bits; 0x4:mmap; 0x8:syscall; 0x10:io_urin= g; 0x20:inherit;\n"); printf("\t-h: help\n"); } =20 @@ -748,7 +852,7 @@ int main(int argc, char **argv) switch (c) { case 't': tests =3D strtoul(optarg, NULL, 16); - if (!(tests & TEST_MASK)) { + if (tests && !(tests & TEST_MASK)) { ksft_print_msg("Invalid argument!\n"); return -1; } @@ -762,6 +866,16 @@ int main(int argc, char **argv) } } =20 + /* + * When tests is 0, it is not a real test case; + * the option used by test case(execve) to check the lam mode in + * process generated by execve, the process read back lam mode and + * check with lam mode in parent process. + */ + if (!tests) + return (get_lam()); + + /* Run test cases */ if (tests & FUNC_MALLOC) run_test(malloc_cases, ARRAY_SIZE(malloc_cases)); =20 @@ -777,6 +891,9 @@ int main(int argc, char **argv) if (tests & FUNC_URING) run_test(uring_cases, ARRAY_SIZE(uring_cases)); =20 + if (tests & FUNC_INHERITE) + run_test(inheritance_cases, ARRAY_SIZE(inheritance_cases)); + ksft_set_plan(tests_cnt); =20 return ksft_exit_pass(); --=20 2.35.1