From patchwork Fri Jul 1 15:27:24 2022 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Ard Biesheuvel X-Patchwork-Id: 12903530 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org Received: from bombadil.infradead.org (bombadil.infradead.org [198.137.202.133]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by smtp.lore.kernel.org (Postfix) with ESMTPS id 1F6D7CCA479 for ; Fri, 1 Jul 2022 15:29:11 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; q=dns/txt; c=relaxed/relaxed; d=lists.infradead.org; s=bombadil.20210309; h=Sender: Content-Transfer-Encoding:Content-Type:List-Subscribe:List-Help:List-Post: List-Archive:List-Unsubscribe:List-Id:MIME-Version:References:In-Reply-To: Message-Id:Date:Subject:Cc:To:From:Reply-To:Content-ID:Content-Description: Resent-Date:Resent-From:Resent-Sender:Resent-To:Resent-Cc:Resent-Message-ID: List-Owner; bh=2QCKxT/p+bsDiKmaVEOs900yFQBjF94hAAEwRFbjZCk=; b=rA6GE6HzNb3VhQ 9sNxSVHzmppfpCWgKfhUL027JnZOJ7P26kM9TEUEWXnm4XPPjgCw+cTdCgDujbLZweNUl2DnK0pk8 CGYqU3hzQOctsOJGuJ430n354Xz1QRbKzn+EnccZcKReDggcbWasTS1IgCJImVXVUdIheEu07bTjT v4StL4HzCcB2WdMM7B/IBrLlUOont9D0Ehk2c375nwVmoDcMlqDRXjE5EMxi93AjO7P2GfHk9vpme ZhaWBCCmlnZ2LBFBhed7Z6RIm6NSAtYonnsHGbnnBOVwAjEV74dGazI1tvl6oSPfTnfCDZJ516I/X GfaQ7vJxUJxI9cocmXDA==; Received: from localhost ([::1] helo=bombadil.infradead.org) by bombadil.infradead.org with esmtp (Exim 4.94.2 #2 (Red Hat Linux)) id 1o7IYy-005fl0-2W; Fri, 01 Jul 2022 15:28:08 +0000 Received: from dfw.source.kernel.org ([139.178.84.217]) by bombadil.infradead.org with esmtps (Exim 4.94.2 #2 (Red Hat Linux)) id 1o7IYY-005fdu-Db for linux-arm-kernel@lists.infradead.org; Fri, 01 Jul 2022 15:27:45 +0000 Received: from smtp.kernel.org (relay.kernel.org [52.25.139.140]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by dfw.source.kernel.org (Postfix) with ESMTPS id E516762417; Fri, 1 Jul 2022 15:27:41 +0000 (UTC) Received: by smtp.kernel.org (Postfix) with ESMTPSA id EB45DC341CA; Fri, 1 Jul 2022 15:27:39 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=kernel.org; s=k20201202; t=1656689261; bh=W/Tmterm1NBpvEWAaoUhVS/5l8Pb2MvLmj0ZsTH/Qz0=; h=From:To:Cc:Subject:Date:In-Reply-To:References:From; b=tACFaIxKCUrSc3758FDEQFzVmCAWF3REetYFLAXkUJOndHDHNAFtKPQKcOtEZPdOv ghbaj0Nv+DTxUMkNH25lxEuN3Oxo7ow5r7v3IpXdj4nkLKZRaF5pW/xOdRtnc5GJEq 9VloXMmxpdPTimnxxnIdTML4bPN+DGOnQXm6FRrVq6GKTyqJI59UHMCfs5ItLexsAO 0KMkvvU7VE4zc3h8GuI/a3WpQDQ0AidFGObZDUHtQQh3kFPaqh6DW63xQZUvGQqKI4 Qy31mf6lJ/FpJEZc8cw38VmzOkj7CiA+7n/1YCDBfa1UMfkMG5FNjeR734TIwD1qXC piOlmkqhIsxrQ== From: Ard Biesheuvel To: linux-arm-kernel@lists.infradead.org Cc: catalin.marinas@arm.com, will@kernel.org, mark.rutland@arm.com, maz@kernel.org, Ard Biesheuvel Subject: [PATCH v4 3/3] arm64: implement dynamic shadow call stack for Clang Date: Fri, 1 Jul 2022 17:27:24 +0200 Message-Id: <20220701152724.3343599-4-ardb@kernel.org> X-Mailer: git-send-email 2.35.1 In-Reply-To: <20220701152724.3343599-1-ardb@kernel.org> References: <20220701152724.3343599-1-ardb@kernel.org> MIME-Version: 1.0 X-Developer-Signature: v=1; a=openpgp-sha256; l=15612; h=from:subject; bh=W/Tmterm1NBpvEWAaoUhVS/5l8Pb2MvLmj0ZsTH/Qz0=; b=owEB7QES/pANAwAKAcNPIjmS2Y8kAcsmYgBivxJbKZtfuBxt3Sechv874Uo2vtgOBNjcKqG+lJpq REw0bXiJAbMEAAEKAB0WIQT72WJ8QGnJQhU3VynDTyI5ktmPJAUCYr8SWwAKCRDDTyI5ktmPJBwdC/ 0WiPnzBb7iwOtj6nbsG0GI+FVUwQ6dVE9Hn687mOkrimlyLYE6BvPkCX7VfMyS20ARD0fFXUt/Zhax 6FM4ucc2XkJytHQ9MP7pjM0jNMlbmYJZ2O8Biz1dMmMVmddPZMrCRfl5U94Nr1iNuNn/yf114IDIWD eAoSPtRx1lr8TLFA1qpx/laXIaSKCF+D9JWjcdYzA7g9nnjwC6sDVen9M7Rn/fvYWNZivhfua5NDYS my5mvI3SMpXhu43Avl4qqGl4Jwtt1ptnMl00QjPbJcxElArPATZmqJDEq0vaIagu67V74gHCUq4snk UGhRWs8oBdhdyVpi+3kjNSMeqoOTnNuWWmgSS3r/W9XTPy0aL+xUFRAMW6y29OM+Qs3LeeSyHi2cOM 9pZ+dQqGfIjwGPJLQiilNyLc/vaxvxY8AOycxtgaizd33g6pk+h4EbW5vuewA7kQL/tl7mmjAzuLEe 67zGAxHuArBp7wVRzzXLJiQhJny1GJCzGIkU0YwX+ZzPA= X-Developer-Key: i=ardb@kernel.org; a=openpgp; fpr=F43D03328115A198C90016883D200E9CA6329909 X-CRM114-Version: 20100106-BlameMichelson ( TRE 0.8.0 (BSD) ) MR-646709E3 X-CRM114-CacheID: sfid-20220701_082742_672468_3270A6F9 X-CRM114-Status: GOOD ( 31.83 ) X-BeenThere: linux-arm-kernel@lists.infradead.org X-Mailman-Version: 2.1.34 Precedence: list List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Sender: "linux-arm-kernel" Errors-To: linux-arm-kernel-bounces+linux-arm-kernel=archiver.kernel.org@lists.infradead.org Implement dynamic shadow call stack support on Clang, by parsing the unwind tables at init time to locate all occurrences of PACIASP/AUTIASP instructions, and replacing them with the shadow call stack push and pop instructions, respectively. This is useful because the overhead of the shadow call stack is difficult to justify on hardware that implements pointer authentication (PAC), and given that the PAC instructions are executed as NOPs on hardware that doesn't, we can just replace them without breaking anything. As PACIASP/AUTIASP are guaranteed to be paired with respect to manipulations of the return address, replacing them 1:1 with shadow call stack pushes and pops is guaranteed to result in the desired behavior. Signed-off-by: Ard Biesheuvel --- arch/arm64/Kconfig | 9 + arch/arm64/Makefile | 10 +- arch/arm64/include/asm/scs.h | 47 ++++ arch/arm64/kernel/Makefile | 2 + arch/arm64/kernel/head.S | 3 + arch/arm64/kernel/irq.c | 2 +- arch/arm64/kernel/module.c | 8 + arch/arm64/kernel/patch-scs.c | 257 ++++++++++++++++++++ arch/arm64/kernel/sdei.c | 2 +- arch/arm64/kernel/setup.c | 4 + 10 files changed, 340 insertions(+), 4 deletions(-) diff --git a/arch/arm64/Kconfig b/arch/arm64/Kconfig index a9314b684f3c..a5156aca1b24 100644 --- a/arch/arm64/Kconfig +++ b/arch/arm64/Kconfig @@ -2082,6 +2082,15 @@ config ARCH_NR_GPIO If unsure, leave the default value. +config UNWIND_PATCH_PAC_INTO_SCS + bool "Enable shadow call stack dynamically using code patching" + # needs Clang with https://reviews.llvm.org/D111780 incorporated + depends on CC_IS_CLANG && CLANG_VERSION >= 150000 + depends on ARM64_PTR_AUTH_KERNEL && CC_HAS_BRANCH_PROT_PAC_RET + depends on SHADOW_CALL_STACK + select UNWIND_TABLES + select DYNAMIC_SCS + endmenu # "Kernel Features" menu "Boot options" diff --git a/arch/arm64/Makefile b/arch/arm64/Makefile index 4fbca56fa602..e439ebbd167d 100644 --- a/arch/arm64/Makefile +++ b/arch/arm64/Makefile @@ -77,10 +77,16 @@ branch-prot-flags-$(CONFIG_CC_HAS_SIGN_RETURN_ADDRESS) := -msign-return-address= # We enable additional protection for leaf functions as there is some # narrow potential for ROP protection benefits and no substantial # performance impact has been observed. +PACRET-y := pac-ret+leaf + +# Using a shadow call stack in leaf functions is too costly, so avoid PAC there +# as well when we may be patching PAC into SCS +PACRET-$(CONFIG_UNWIND_PATCH_PAC_INTO_SCS) := pac-ret + ifeq ($(CONFIG_ARM64_BTI_KERNEL),y) -branch-prot-flags-$(CONFIG_CC_HAS_BRANCH_PROT_PAC_RET_BTI) := -mbranch-protection=pac-ret+leaf+bti +branch-prot-flags-$(CONFIG_CC_HAS_BRANCH_PROT_PAC_RET_BTI) := -mbranch-protection=$(PACRET-y)+bti else -branch-prot-flags-$(CONFIG_CC_HAS_BRANCH_PROT_PAC_RET) := -mbranch-protection=pac-ret+leaf +branch-prot-flags-$(CONFIG_CC_HAS_BRANCH_PROT_PAC_RET) := -mbranch-protection=$(PACRET-y) endif # -march=armv8.3-a enables the non-nops instructions for PAC, to avoid the # compiler to generate them and consequently to break the single image contract diff --git a/arch/arm64/include/asm/scs.h b/arch/arm64/include/asm/scs.h index 8297bccf0784..e5726d737e37 100644 --- a/arch/arm64/include/asm/scs.h +++ b/arch/arm64/include/asm/scs.h @@ -24,6 +24,53 @@ .endm #endif /* CONFIG_SHADOW_CALL_STACK */ + +#else + +#include +#include + +#ifdef CONFIG_UNWIND_PATCH_PAC_INTO_SCS +static inline bool should_patch_pac_into_scs(void) +{ + /* + * We only enable the shadow call stack dynamically if we are running + * on a system that does not implement PAC or BTI. PAC and SCS provide + * roughly the same level of protection, and BTI relies on the PACIASP + * instructions serving as landing pads, preventing us from patching + * those instructions into something else. + */ + u64 reg = read_sysreg_s(SYS_ID_AA64ISAR1_EL1); + + if (reg & ((0xf << ID_AA64ISAR1_APA_SHIFT) | + (0xf << ID_AA64ISAR1_API_SHIFT))) + return false; + + reg = read_sysreg_s(SYS_ID_AA64ISAR2_EL1); + if (reg & (0xf << ID_AA64ISAR2_APA3_SHIFT)) + return false; + + if (IS_ENABLED(CONFIG_ARM64_BTI_KERNEL)) { + reg = read_sysreg_s(SYS_ID_AA64PFR1_EL1); + if (reg & (0xf << ID_AA64PFR1_BT_SHIFT)) + return false; + } + return true; +} + +static inline void dynamic_scs_init(void) +{ + if (should_patch_pac_into_scs()) { + pr_info("Enabling dynamic shadow call stack\n"); + static_branch_enable(&dynamic_scs_enabled); + } +} +#else +static inline void dynamic_scs_init(void) {} +#endif + +int scs_patch(const u8 eh_frame[], int size); + #endif /* __ASSEMBLY __ */ #endif /* _ASM_SCS_H */ diff --git a/arch/arm64/kernel/Makefile b/arch/arm64/kernel/Makefile index 1add7b01efa7..420b53b24f07 100644 --- a/arch/arm64/kernel/Makefile +++ b/arch/arm64/kernel/Makefile @@ -79,6 +79,8 @@ obj-$(CONFIG_ARM64_PTR_AUTH) += pointer_auth.o obj-$(CONFIG_ARM64_MTE) += mte.o obj-y += vdso-wrap.o obj-$(CONFIG_COMPAT_VDSO) += vdso32-wrap.o +obj-$(CONFIG_UNWIND_PATCH_PAC_INTO_SCS) += patch-scs.o +CFLAGS_patch-scs.o += -mbranch-protection=none # Force dependency (vdso*-wrap.S includes vdso.so through incbin) $(obj)/vdso-wrap.o: $(obj)/vdso/vdso.so diff --git a/arch/arm64/kernel/head.S b/arch/arm64/kernel/head.S index c300b43659dc..7187a979faba 100644 --- a/arch/arm64/kernel/head.S +++ b/arch/arm64/kernel/head.S @@ -443,6 +443,9 @@ SYM_FUNC_START_LOCAL(__primary_switched) #endif mov x0, x22 // pass FDT address in x0 bl init_feature_override // Parse cpu feature overrides +#ifdef CONFIG_UNWIND_PATCH_PAC_INTO_SCS + bl scs_patch_vmlinux +#endif mov x0, x20 bl switch_to_vhe // Prefer VHE if possible ldp x29, x30, [sp], #16 diff --git a/arch/arm64/kernel/irq.c b/arch/arm64/kernel/irq.c index bda49430c9ea..c284ec35c27c 100644 --- a/arch/arm64/kernel/irq.c +++ b/arch/arm64/kernel/irq.c @@ -39,7 +39,7 @@ static void init_irq_scs(void) { int cpu; - if (!IS_ENABLED(CONFIG_SHADOW_CALL_STACK)) + if (!scs_is_enabled()) return; for_each_possible_cpu(cpu) diff --git a/arch/arm64/kernel/module.c b/arch/arm64/kernel/module.c index f2d4bb14bfab..111dc6414e6d 100644 --- a/arch/arm64/kernel/module.c +++ b/arch/arm64/kernel/module.c @@ -15,9 +15,11 @@ #include #include #include +#include #include #include #include +#include #include void *module_alloc(unsigned long size) @@ -529,5 +531,11 @@ int module_finalize(const Elf_Ehdr *hdr, if (s) apply_alternatives_module((void *)s->sh_addr, s->sh_size); + if (scs_is_dynamic()) { + s = find_section(hdr, sechdrs, ".init.eh_frame"); + if (s) + scs_patch((void *)s->sh_addr, s->sh_size); + } + return module_init_ftrace_plt(hdr, sechdrs, me); } diff --git a/arch/arm64/kernel/patch-scs.c b/arch/arm64/kernel/patch-scs.c new file mode 100644 index 000000000000..1b3da02d5b74 --- /dev/null +++ b/arch/arm64/kernel/patch-scs.c @@ -0,0 +1,257 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* + * Copyright (C) 2022 - Google LLC + * Author: Ard Biesheuvel + */ + +#include +#include +#include +#include +#include +#include + +#include +#include + +// +// This minimal DWARF CFI parser is partially based on the code in +// arch/arc/kernel/unwind.c, and on the document below: +// https://refspecs.linuxbase.org/LSB_4.0.0/LSB-Core-generic/LSB-Core-generic/ehframechpt.html +// + +#define DW_CFA_nop 0x00 +#define DW_CFA_set_loc 0x01 +#define DW_CFA_advance_loc1 0x02 +#define DW_CFA_advance_loc2 0x03 +#define DW_CFA_advance_loc4 0x04 +#define DW_CFA_offset_extended 0x05 +#define DW_CFA_restore_extended 0x06 +#define DW_CFA_undefined 0x07 +#define DW_CFA_same_value 0x08 +#define DW_CFA_register 0x09 +#define DW_CFA_remember_state 0x0a +#define DW_CFA_restore_state 0x0b +#define DW_CFA_def_cfa 0x0c +#define DW_CFA_def_cfa_register 0x0d +#define DW_CFA_def_cfa_offset 0x0e +#define DW_CFA_def_cfa_expression 0x0f +#define DW_CFA_expression 0x10 +#define DW_CFA_offset_extended_sf 0x11 +#define DW_CFA_def_cfa_sf 0x12 +#define DW_CFA_def_cfa_offset_sf 0x13 +#define DW_CFA_val_offset 0x14 +#define DW_CFA_val_offset_sf 0x15 +#define DW_CFA_val_expression 0x16 +#define DW_CFA_lo_user 0x1c +#define DW_CFA_negate_ra_state 0x2d +#define DW_CFA_GNU_args_size 0x2e +#define DW_CFA_GNU_negative_offset_extended 0x2f +#define DW_CFA_hi_user 0x3f + +extern const u8 __eh_frame_start[], __eh_frame_end[]; + +enum { + PACIASP = 0xd503233f, + AUTIASP = 0xd50323bf, + SCS_PUSH = 0xf800865e, + SCS_POP = 0xf85f8e5e, +}; + +static void __always_inline scs_patch_loc(u64 loc) +{ + u32 insn = le32_to_cpup((void *)loc); + + switch (insn) { + case PACIASP: + *(u32 *)loc = cpu_to_le32(SCS_PUSH); + break; + case AUTIASP: + *(u32 *)loc = cpu_to_le32(SCS_POP); + break; + default: + /* + * While the DW_CFA_negate_ra_state directive is guaranteed to + * appear right after a PACIASP/AUTIASP instruction, it may + * also appear after a DW_CFA_restore_state directive that + * restores a state that is only partially accurate, and is + * followed by DW_CFA_negate_ra_state directive to toggle the + * PAC bit again. So we permit other instructions here, and ignore + * them. + */ + return; + } + dcache_clean_pou(loc, loc + sizeof(u32)); +} + +/* + * Skip one uleb128/sleb128 encoded quantity from the opcode stream. All bytes + * except the last one have bit #7 set. + */ +static int __always_inline skip_xleb128(const u8 **opcode, int size) +{ + u8 c; + + do { + c = *(*opcode)++; + size--; + } while (c & BIT(7)); + + return size; +} + +struct eh_frame { + /* + * The size of this frame if 0 < size < U32_MAX, 0 terminates the list. + */ + u32 size; + + /* + * The first frame is a Common Information Entry (CIE) frame, followed + * by one or more Frame Description Entry (FDE) frames. In the former + * case, this field is 0, otherwise it is the negated offset relative + * to the associated CIE frame. + */ + u32 cie_id_or_pointer; + + union { + struct { // CIE + u8 version; + u8 augmentation_string[]; + }; + + struct { // FDE + s32 initial_loc; + s32 range; + u8 opcodes[]; + }; + }; +}; + +static int noinstr scs_handle_fde_frame(const struct eh_frame *frame, + bool fde_has_augmentation_data, + int code_alignment_factor) +{ + int size = frame->size - offsetof(struct eh_frame, opcodes) + 4; + u64 loc = (u64)offset_to_ptr(&frame->initial_loc); + const u8 *opcode = frame->opcodes; + + if (fde_has_augmentation_data) { + int l; + + // assume single byte uleb128_t + if (WARN_ON(*opcode & BIT(7))) + return -ENOEXEC; + + l = *opcode++; + opcode += l; + size -= l + 1; + } + + /* + * Starting from 'loc', apply the CFA opcodes that advance the location + * pointer, and identify the locations of the PAC instructions. + */ + while (size-- > 0) { + switch (*opcode++) { + case DW_CFA_nop: + case DW_CFA_remember_state: + case DW_CFA_restore_state: + break; + + case DW_CFA_advance_loc1: + loc += *opcode++ * code_alignment_factor; + size--; + break; + + case DW_CFA_advance_loc2: + loc += *opcode++ * code_alignment_factor; + loc += (*opcode++ << 8) * code_alignment_factor; + size -= 2; + break; + + case DW_CFA_def_cfa: + case DW_CFA_offset_extended: + size = skip_xleb128(&opcode, size); + fallthrough; + case DW_CFA_def_cfa_offset: + case DW_CFA_def_cfa_offset_sf: + case DW_CFA_def_cfa_register: + case DW_CFA_same_value: + case DW_CFA_restore_extended: + case 0x80 ... 0xbf: + size = skip_xleb128(&opcode, size); + break; + + case DW_CFA_negate_ra_state: + scs_patch_loc(loc - 4); + break; + + case 0x40 ... 0x7f: + // advance loc + loc += (opcode[-1] & 0x3f) * code_alignment_factor; + break; + + case 0xc0 ... 0xff: + break; + + default: + pr_err("unhandled opcode: %02x in FDE frame %lx\n", opcode[-1], (uintptr_t)frame); + return -ENOEXEC; + } + } + return 0; +} + +int noinstr scs_patch(const u8 eh_frame[], int size) +{ + const u8 *p = eh_frame; + + while (size > 4) { + const struct eh_frame *frame = (const void *)p; + bool fde_has_augmentation_data = true; + int code_alignment_factor = 1; + int ret; + + if (frame->size == 0 || + frame->size == U32_MAX || + frame->size > size) + break; + + if (frame->cie_id_or_pointer == 0) { + const u8 *p = frame->augmentation_string; + + /* a 'z' in the augmentation string must come first */ + fde_has_augmentation_data = *p == 'z'; + + /* + * The code alignment factor is a uleb128 encoded field + * but given that the only sensible values are 1 or 4, + * there is no point in decoding the whole thing. + */ + p += strlen(p) + 1; + if (!WARN_ON(*p & BIT(7))) + code_alignment_factor = *p; + } else { + ret = scs_handle_fde_frame(frame, + fde_has_augmentation_data, + code_alignment_factor); + if (ret) + return ret; + } + + p += sizeof(frame->size) + frame->size; + size -= sizeof(frame->size) + frame->size; + } + return 0; +} + +asmlinkage void __init scs_patch_vmlinux(void) +{ + if (!should_patch_pac_into_scs()) + return; + + WARN_ON(scs_patch(__eh_frame_start, __eh_frame_end - __eh_frame_start)); + icache_inval_all_pou(); + isb(); +} diff --git a/arch/arm64/kernel/sdei.c b/arch/arm64/kernel/sdei.c index d20620a1c51a..30f3c7563694 100644 --- a/arch/arm64/kernel/sdei.c +++ b/arch/arm64/kernel/sdei.c @@ -144,7 +144,7 @@ static int init_sdei_scs(void) int cpu; int err = 0; - if (!IS_ENABLED(CONFIG_SHADOW_CALL_STACK)) + if (!scs_is_enabled()) return 0; for_each_possible_cpu(cpu) { diff --git a/arch/arm64/kernel/setup.c b/arch/arm64/kernel/setup.c index d0e6c7a291da..37e0ba95afc3 100644 --- a/arch/arm64/kernel/setup.c +++ b/arch/arm64/kernel/setup.c @@ -30,6 +30,7 @@ #include #include #include +#include #include #include @@ -42,6 +43,7 @@ #include #include #include +#include #include #include #include @@ -297,6 +299,8 @@ void __init __no_sanitize_address setup_arch(char **cmdline_p) jump_label_init(); parse_early_param(); + dynamic_scs_init(); + /* * Unmask asynchronous aborts and fiq after bringing up possible * earlycon. (Report possible System Errors once we can report this