From nobody Wed Nov 27 15:48:28 2024 Received: from smtp.kernel.org (aws-us-west-2-korg-mail-1.web.codeaurora.org [10.30.226.201]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by smtp.subspace.kernel.org (Postfix) with ESMTPS id 572101E47BB; Wed, 9 Oct 2024 18:10:35 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=10.30.226.201 ARC-Seal: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1728497436; cv=none; b=MxjD1Ae/+WAb8l4oPCRY8ZuYpRDSKRcWghIpyB1K//Fek544w+rh28ssAwDEh0+jyGh66UybMLVLlyp+WtWVHL9X282JSpWARkXpqflLzolvC6h8V1EZcOiuveX6/wxIAR9uxAdxI2G8Ce/GYCU6GDm/mHGLFRg7bm2W+/FqEsQ= ARC-Message-Signature: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1728497436; c=relaxed/simple; bh=Ax4VXUUK487+iOr8SxFFqUBHtKPyMu4ZOopc26VJUvE=; h=From:To:Cc:Subject:Date:Message-ID:In-Reply-To:References: MIME-Version; b=mtMGTusJIHEBTs/TqTBb6moZT/om0X4nNdjznVZ+QvjENWK7z7F9auCWNIivhMaSEbuvOkRiSh8xII0YRW9XyweT636Qe0XYO9UNrIXfeMhJ8KkxA3fGMLgikGt6ykn9yIv777MQh/Kc+wi1I2CKa3KWTl+SfAM0Jz7GYFjEIZs= ARC-Authentication-Results: i=1; smtp.subspace.kernel.org; dkim=pass (2048-bit key) header.d=kernel.org header.i=@kernel.org header.b=eL7QxUW+; arc=none smtp.client-ip=10.30.226.201 Authentication-Results: smtp.subspace.kernel.org; dkim=pass (2048-bit key) header.d=kernel.org header.i=@kernel.org header.b="eL7QxUW+" Received: by smtp.kernel.org (Postfix) with ESMTPSA id 8958EC4CED0; Wed, 9 Oct 2024 18:10:17 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=kernel.org; s=k20201202; t=1728497434; bh=Ax4VXUUK487+iOr8SxFFqUBHtKPyMu4ZOopc26VJUvE=; h=From:To:Cc:Subject:Date:In-Reply-To:References:From; b=eL7QxUW+vZYygNeCRUOO2URi13c/tSPIJ/AcToCEuebakImdibg401XM8nkONNJFH PlM7y2jGeGSA7OD//2rES0aSYta2Tgwfq78xgFB8+5tHg9fJxuNebM25et79NZE61i bTLAA4lMbFc0LJt7fvoMZmqR4vFfn5KTbUbTRbBR4WNpw6uWADts3KYL9/kIXi/ytM TqZ8nGvNZEO7o/D25gU0hM4JQSVnRP0orrd7Djoc1nLQ2vHHVEYY7Zu1gTL3VJZpGG mgGKLLXeTEhqhzCXe4mT+ctQSg7ER33P5FIlXyRrlSFmNlnuSfMQ+8wUv9T1uD7kSE OLt0EIR71kG4w== From: Mike Rapoport To: Andrew Morton Cc: Andreas Larsson , Andy Lutomirski , Ard Biesheuvel , Arnd Bergmann , Borislav Petkov , Brian Cain , Catalin Marinas , Christoph Hellwig , Christophe Leroy , Dave Hansen , Dinh Nguyen , Geert Uytterhoeven , Guo Ren , Helge Deller , Huacai Chen , Ingo Molnar , Johannes Berg , John Paul Adrian Glaubitz , Kent Overstreet , "Liam R. Howlett" , Luis Chamberlain , Mark Rutland , Masami Hiramatsu , Matt Turner , Max Filippov , Michael Ellerman , Michal Simek , Mike Rapoport , Oleg Nesterov , Palmer Dabbelt , Peter Zijlstra , Richard Weinberger , Russell King , Song Liu , Stafford Horne , Steven Rostedt , Thomas Bogendoerfer , Thomas Gleixner , Uladzislau Rezki , Vineet Gupta , Will Deacon , bpf@vger.kernel.org, linux-alpha@vger.kernel.org, linux-arch@vger.kernel.org, linux-arm-kernel@lists.infradead.org, linux-csky@vger.kernel.org, linux-hexagon@vger.kernel.org, linux-kernel@vger.kernel.org, linux-m68k@lists.linux-m68k.org, linux-mips@vger.kernel.org, linux-mm@kvack.org, linux-modules@vger.kernel.org, linux-openrisc@vger.kernel.org, linux-parisc@vger.kernel.org, linux-riscv@lists.infradead.org, linux-sh@vger.kernel.org, linux-snps-arc@lists.infradead.org, linux-trace-kernel@vger.kernel.org, linux-um@lists.infradead.org, linuxppc-dev@lists.ozlabs.org, loongarch@lists.linux.dev, sparclinux@vger.kernel.org, x86@kernel.org Subject: [PATCH v5 6/8] x86/module: perpare module loading for ROX allocations of text Date: Wed, 9 Oct 2024 21:08:14 +0300 Message-ID: <20241009180816.83591-7-rppt@kernel.org> X-Mailer: git-send-email 2.43.0 In-Reply-To: <20241009180816.83591-1-rppt@kernel.org> References: <20241009180816.83591-1-rppt@kernel.org> Precedence: bulk X-Mailing-List: linux-kernel@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable Content-Type: text/plain; charset="utf-8" From: "Mike Rapoport (Microsoft)" When module text memory will be allocated with ROX permissions, the memory at the actual address where the module will live will contain invalid instructions and there will be a writable copy that contains the actual module code. Update relocations and alternatives patching to deal with it. Signed-off-by: Mike Rapoport (Microsoft) Reviewed-by: Christoph Hellwig --- arch/um/kernel/um_arch.c | 11 +- arch/x86/entry/vdso/vma.c | 3 +- arch/x86/include/asm/alternative.h | 14 +-- arch/x86/kernel/alternative.c | 160 +++++++++++++++++------------ arch/x86/kernel/ftrace.c | 30 +++--- arch/x86/kernel/module.c | 45 +++++--- include/linux/module.h | 16 ++- 7 files changed, 161 insertions(+), 118 deletions(-) diff --git a/arch/um/kernel/um_arch.c b/arch/um/kernel/um_arch.c index f8de31a0c5d1..e8e8b54b3037 100644 --- a/arch/um/kernel/um_arch.c +++ b/arch/um/kernel/um_arch.c @@ -435,24 +435,25 @@ void __init arch_cpu_finalize_init(void) os_check_bugs(); } =20 -void apply_seal_endbr(s32 *start, s32 *end) +void apply_seal_endbr(s32 *start, s32 *end, struct module *mod) { } =20 -void apply_retpolines(s32 *start, s32 *end) +void apply_retpolines(s32 *start, s32 *end, struct module *mod) { } =20 -void apply_returns(s32 *start, s32 *end) +void apply_returns(s32 *start, s32 *end, struct module *mod) { } =20 void apply_fineibt(s32 *start_retpoline, s32 *end_retpoline, - s32 *start_cfi, s32 *end_cfi) + s32 *start_cfi, s32 *end_cfi, struct module *mod) { } =20 -void apply_alternatives(struct alt_instr *start, struct alt_instr *end) +void apply_alternatives(struct alt_instr *start, struct alt_instr *end, + struct module *mod) { } =20 diff --git a/arch/x86/entry/vdso/vma.c b/arch/x86/entry/vdso/vma.c index b8fed8b8b9cc..ed21151923c3 100644 --- a/arch/x86/entry/vdso/vma.c +++ b/arch/x86/entry/vdso/vma.c @@ -54,7 +54,8 @@ int __init init_vdso_image(const struct vdso_image *image) =20 apply_alternatives((struct alt_instr *)(image->data + image->alt), (struct alt_instr *)(image->data + image->alt + - image->alt_len)); + image->alt_len), + NULL); =20 return 0; } diff --git a/arch/x86/include/asm/alternative.h b/arch/x86/include/asm/alte= rnative.h index ca9ae606aab9..dc03a647776d 100644 --- a/arch/x86/include/asm/alternative.h +++ b/arch/x86/include/asm/alternative.h @@ -96,16 +96,16 @@ extern struct alt_instr __alt_instructions[], __alt_ins= tructions_end[]; * instructions were patched in already: */ extern int alternatives_patched; +struct module; =20 extern void alternative_instructions(void); -extern void apply_alternatives(struct alt_instr *start, struct alt_instr *= end); -extern void apply_retpolines(s32 *start, s32 *end); -extern void apply_returns(s32 *start, s32 *end); -extern void apply_seal_endbr(s32 *start, s32 *end); +extern void apply_alternatives(struct alt_instr *start, struct alt_instr *= end, + struct module *mod); +extern void apply_retpolines(s32 *start, s32 *end, struct module *mod); +extern void apply_returns(s32 *start, s32 *end, struct module *mod); +extern void apply_seal_endbr(s32 *start, s32 *end, struct module *mod); extern void apply_fineibt(s32 *start_retpoline, s32 *end_retpoine, - s32 *start_cfi, s32 *end_cfi); - -struct module; + s32 *start_cfi, s32 *end_cfi, struct module *mod); =20 struct callthunk_sites { s32 *call_start, *call_end; diff --git a/arch/x86/kernel/alternative.c b/arch/x86/kernel/alternative.c index d17518ca19b8..3b3fa93af3b1 100644 --- a/arch/x86/kernel/alternative.c +++ b/arch/x86/kernel/alternative.c @@ -392,8 +392,10 @@ EXPORT_SYMBOL(BUG_func); * Rewrite the "call BUG_func" replacement to point to the target of the * indirect pv_ops call "call *disp(%ip)". */ -static int alt_replace_call(u8 *instr, u8 *insn_buff, struct alt_instr *a) +static int alt_replace_call(u8 *instr, u8 *insn_buff, struct alt_instr *a, + struct module *mod) { + u8 *wr_instr =3D module_writable_address(mod, instr); void *target, *bug =3D &BUG_func; s32 disp; =20 @@ -403,14 +405,14 @@ static int alt_replace_call(u8 *instr, u8 *insn_buff,= struct alt_instr *a) } =20 if (a->instrlen !=3D 6 || - instr[0] !=3D CALL_RIP_REL_OPCODE || - instr[1] !=3D CALL_RIP_REL_MODRM) { + wr_instr[0] !=3D CALL_RIP_REL_OPCODE || + wr_instr[1] !=3D CALL_RIP_REL_MODRM) { pr_err("ALT_FLAG_DIRECT_CALL set for unrecognized indirect call\n"); BUG(); } =20 /* Skip CALL_RIP_REL_OPCODE and CALL_RIP_REL_MODRM */ - disp =3D *(s32 *)(instr + 2); + disp =3D *(s32 *)(wr_instr + 2); #ifdef CONFIG_X86_64 /* ff 15 00 00 00 00 call *0x0(%rip) */ /* target address is stored at "next instruction + disp". */ @@ -448,7 +450,8 @@ static inline u8 * instr_va(struct alt_instr *i) * to refetch changed I$ lines. */ void __init_or_module noinline apply_alternatives(struct alt_instr *start, - struct alt_instr *end) + struct alt_instr *end, + struct module *mod) { u8 insn_buff[MAX_PATCH_LEN]; u8 *instr, *replacement; @@ -477,6 +480,7 @@ void __init_or_module noinline apply_alternatives(struc= t alt_instr *start, */ for (a =3D start; a < end; a++) { int insn_buff_sz =3D 0; + u8 *wr_instr, *wr_replacement; =20 /* * In case of nested ALTERNATIVE()s the outer alternative might @@ -490,7 +494,11 @@ void __init_or_module noinline apply_alternatives(stru= ct alt_instr *start, } =20 instr =3D instr_va(a); + wr_instr =3D module_writable_address(mod, instr); + replacement =3D (u8 *)&a->repl_offset + a->repl_offset; + wr_replacement =3D module_writable_address(mod, replacement); + BUG_ON(a->instrlen > sizeof(insn_buff)); BUG_ON(a->cpuid >=3D (NCAPINTS + NBUGINTS) * 32); =20 @@ -501,9 +509,9 @@ void __init_or_module noinline apply_alternatives(struc= t alt_instr *start, * patch if feature is *NOT* present. */ if (!boot_cpu_has(a->cpuid) =3D=3D !(a->flags & ALT_FLAG_NOT)) { - memcpy(insn_buff, instr, a->instrlen); + memcpy(insn_buff, wr_instr, a->instrlen); optimize_nops(instr, insn_buff, a->instrlen); - text_poke_early(instr, insn_buff, a->instrlen); + text_poke_early(wr_instr, insn_buff, a->instrlen); continue; } =20 @@ -513,11 +521,12 @@ void __init_or_module noinline apply_alternatives(str= uct alt_instr *start, instr, instr, a->instrlen, replacement, a->replacementlen, a->flags); =20 - memcpy(insn_buff, replacement, a->replacementlen); + memcpy(insn_buff, wr_replacement, a->replacementlen); insn_buff_sz =3D a->replacementlen; =20 if (a->flags & ALT_FLAG_DIRECT_CALL) { - insn_buff_sz =3D alt_replace_call(instr, insn_buff, a); + insn_buff_sz =3D alt_replace_call(instr, insn_buff, a, + mod); if (insn_buff_sz < 0) continue; } @@ -527,11 +536,11 @@ void __init_or_module noinline apply_alternatives(str= uct alt_instr *start, =20 apply_relocation(insn_buff, instr, a->instrlen, replacement, a->replacem= entlen); =20 - DUMP_BYTES(ALT, instr, a->instrlen, "%px: old_insn: ", instr); + DUMP_BYTES(ALT, wr_instr, a->instrlen, "%px: old_insn: ", instr); DUMP_BYTES(ALT, replacement, a->replacementlen, "%px: rpl_insn: ", rep= lacement); DUMP_BYTES(ALT, insn_buff, insn_buff_sz, "%px: final_insn: ", instr); =20 - text_poke_early(instr, insn_buff, insn_buff_sz); + text_poke_early(wr_instr, insn_buff, insn_buff_sz); } =20 kasan_enable_current(); @@ -722,18 +731,20 @@ static int patch_retpoline(void *addr, struct insn *i= nsn, u8 *bytes) /* * Generated by 'objtool --retpoline'. */ -void __init_or_module noinline apply_retpolines(s32 *start, s32 *end) +void __init_or_module noinline apply_retpolines(s32 *start, s32 *end, + struct module *mod) { s32 *s; =20 for (s =3D start; s < end; s++) { void *addr =3D (void *)s + *s; + void *wr_addr =3D module_writable_address(mod, addr); struct insn insn; int len, ret; u8 bytes[16]; u8 op1, op2; =20 - ret =3D insn_decode_kernel(&insn, addr); + ret =3D insn_decode_kernel(&insn, wr_addr); if (WARN_ON_ONCE(ret < 0)) continue; =20 @@ -761,9 +772,9 @@ void __init_or_module noinline apply_retpolines(s32 *st= art, s32 *end) len =3D patch_retpoline(addr, &insn, bytes); if (len =3D=3D insn.length) { optimize_nops(addr, bytes, len); - DUMP_BYTES(RETPOLINE, ((u8*)addr), len, "%px: orig: ", addr); + DUMP_BYTES(RETPOLINE, ((u8*)wr_addr), len, "%px: orig: ", addr); DUMP_BYTES(RETPOLINE, ((u8*)bytes), len, "%px: repl: ", addr); - text_poke_early(addr, bytes, len); + text_poke_early(wr_addr, bytes, len); } } } @@ -799,7 +810,8 @@ static int patch_return(void *addr, struct insn *insn, = u8 *bytes) return i; } =20 -void __init_or_module noinline apply_returns(s32 *start, s32 *end) +void __init_or_module noinline apply_returns(s32 *start, s32 *end, + struct module *mod) { s32 *s; =20 @@ -808,12 +820,13 @@ void __init_or_module noinline apply_returns(s32 *sta= rt, s32 *end) =20 for (s =3D start; s < end; s++) { void *dest =3D NULL, *addr =3D (void *)s + *s; + void *wr_addr =3D module_writable_address(mod, addr); struct insn insn; int len, ret; u8 bytes[16]; u8 op; =20 - ret =3D insn_decode_kernel(&insn, addr); + ret =3D insn_decode_kernel(&insn, wr_addr); if (WARN_ON_ONCE(ret < 0)) continue; =20 @@ -833,32 +846,35 @@ void __init_or_module noinline apply_returns(s32 *sta= rt, s32 *end) =20 len =3D patch_return(addr, &insn, bytes); if (len =3D=3D insn.length) { - DUMP_BYTES(RET, ((u8*)addr), len, "%px: orig: ", addr); + DUMP_BYTES(RET, ((u8*)wr_addr), len, "%px: orig: ", addr); DUMP_BYTES(RET, ((u8*)bytes), len, "%px: repl: ", addr); - text_poke_early(addr, bytes, len); + text_poke_early(wr_addr, bytes, len); } } } #else -void __init_or_module noinline apply_returns(s32 *start, s32 *end) { } +void __init_or_module noinline apply_returns(s32 *start, s32 *end, + struct module *mod) { } #endif /* CONFIG_MITIGATION_RETHUNK */ =20 #else /* !CONFIG_MITIGATION_RETPOLINE || !CONFIG_OBJTOOL */ =20 -void __init_or_module noinline apply_retpolines(s32 *start, s32 *end) { } -void __init_or_module noinline apply_returns(s32 *start, s32 *end) { } +void __init_or_module noinline apply_retpolines(s32 *start, s32 *end, + struct module *mod) { } +void __init_or_module noinline apply_returns(s32 *start, s32 *end, + struct module *mod) { } =20 #endif /* CONFIG_MITIGATION_RETPOLINE && CONFIG_OBJTOOL */ =20 #ifdef CONFIG_X86_KERNEL_IBT =20 -static void poison_cfi(void *addr); +static void poison_cfi(void *addr, void *wr_addr); =20 -static void __init_or_module poison_endbr(void *addr, bool warn) +static void __init_or_module poison_endbr(void *addr, void *wr_addr, bool = warn) { u32 endbr, poison =3D gen_endbr_poison(); =20 - if (WARN_ON_ONCE(get_kernel_nofault(endbr, addr))) + if (WARN_ON_ONCE(get_kernel_nofault(endbr, wr_addr))) return; =20 if (!is_endbr(endbr)) { @@ -873,7 +889,7 @@ static void __init_or_module poison_endbr(void *addr, b= ool warn) */ DUMP_BYTES(ENDBR, ((u8*)addr), 4, "%px: orig: ", addr); DUMP_BYTES(ENDBR, ((u8*)&poison), 4, "%px: repl: ", addr); - text_poke_early(addr, &poison, 4); + text_poke_early(wr_addr, &poison, 4); } =20 /* @@ -882,22 +898,23 @@ static void __init_or_module poison_endbr(void *addr,= bool warn) * Seal the functions for indirect calls by clobbering the ENDBR instructi= ons * and the kCFI hash value. */ -void __init_or_module noinline apply_seal_endbr(s32 *start, s32 *end) +void __init_or_module noinline apply_seal_endbr(s32 *start, s32 *end, stru= ct module *mod) { s32 *s; =20 for (s =3D start; s < end; s++) { void *addr =3D (void *)s + *s; + void *wr_addr =3D module_writable_address(mod, addr); =20 - poison_endbr(addr, true); + poison_endbr(addr, wr_addr, true); if (IS_ENABLED(CONFIG_FINEIBT)) - poison_cfi(addr - 16); + poison_cfi(addr - 16, wr_addr - 16); } } =20 #else =20 -void __init_or_module apply_seal_endbr(s32 *start, s32 *end) { } +void __init_or_module apply_seal_endbr(s32 *start, s32 *end, struct module= *mod) { } =20 #endif /* CONFIG_X86_KERNEL_IBT */ =20 @@ -1119,7 +1136,7 @@ static u32 decode_caller_hash(void *addr) } =20 /* .retpoline_sites */ -static int cfi_disable_callers(s32 *start, s32 *end) +static int cfi_disable_callers(s32 *start, s32 *end, struct module *mod) { /* * Disable kCFI by patching in a JMP.d8, this leaves the hash immediate @@ -1131,6 +1148,7 @@ static int cfi_disable_callers(s32 *start, s32 *end) =20 for (s =3D start; s < end; s++) { void *addr =3D (void *)s + *s; + void *wr_addr =3D module_writable_address(mod, addr); u32 hash; =20 addr -=3D fineibt_caller_size; @@ -1138,13 +1156,13 @@ static int cfi_disable_callers(s32 *start, s32 *end) if (!hash) /* nocfi callers */ continue; =20 - text_poke_early(addr, jmp, 2); + text_poke_early(wr_addr, jmp, 2); } =20 return 0; } =20 -static int cfi_enable_callers(s32 *start, s32 *end) +static int cfi_enable_callers(s32 *start, s32 *end, struct module *mod) { /* * Re-enable kCFI, undo what cfi_disable_callers() did. @@ -1154,6 +1172,7 @@ static int cfi_enable_callers(s32 *start, s32 *end) =20 for (s =3D start; s < end; s++) { void *addr =3D (void *)s + *s; + void *wr_addr =3D module_writable_address(mod, addr); u32 hash; =20 addr -=3D fineibt_caller_size; @@ -1161,19 +1180,20 @@ static int cfi_enable_callers(s32 *start, s32 *end) if (!hash) /* nocfi callers */ continue; =20 - text_poke_early(addr, mov, 2); + text_poke_early(wr_addr, mov, 2); } =20 return 0; } =20 /* .cfi_sites */ -static int cfi_rand_preamble(s32 *start, s32 *end) +static int cfi_rand_preamble(s32 *start, s32 *end, struct module *mod) { s32 *s; =20 for (s =3D start; s < end; s++) { void *addr =3D (void *)s + *s; + void *wr_addr =3D module_writable_address(mod, addr); u32 hash; =20 hash =3D decode_preamble_hash(addr); @@ -1182,18 +1202,19 @@ static int cfi_rand_preamble(s32 *start, s32 *end) return -EINVAL; =20 hash =3D cfi_rehash(hash); - text_poke_early(addr + 1, &hash, 4); + text_poke_early(wr_addr + 1, &hash, 4); } =20 return 0; } =20 -static int cfi_rewrite_preamble(s32 *start, s32 *end) +static int cfi_rewrite_preamble(s32 *start, s32 *end, struct module *mod) { s32 *s; =20 for (s =3D start; s < end; s++) { void *addr =3D (void *)s + *s; + void *wr_addr =3D module_writable_address(mod, addr); u32 hash; =20 hash =3D decode_preamble_hash(addr); @@ -1201,59 +1222,62 @@ static int cfi_rewrite_preamble(s32 *start, s32 *en= d) addr, addr, 5, addr)) return -EINVAL; =20 - text_poke_early(addr, fineibt_preamble_start, fineibt_preamble_size); + text_poke_early(wr_addr, fineibt_preamble_start, fineibt_preamble_size); WARN_ON(*(u32 *)(addr + fineibt_preamble_hash) !=3D 0x12345678); - text_poke_early(addr + fineibt_preamble_hash, &hash, 4); + text_poke_early(wr_addr + fineibt_preamble_hash, &hash, 4); } =20 return 0; } =20 -static void cfi_rewrite_endbr(s32 *start, s32 *end) +static void cfi_rewrite_endbr(s32 *start, s32 *end, struct module *mod) { s32 *s; =20 for (s =3D start; s < end; s++) { void *addr =3D (void *)s + *s; + void *wr_addr =3D module_writable_address(mod, addr); =20 - poison_endbr(addr+16, false); + poison_endbr(addr+16, wr_addr, false); } } =20 /* .retpoline_sites */ -static int cfi_rand_callers(s32 *start, s32 *end) +static int cfi_rand_callers(s32 *start, s32 *end, struct module *mod) { s32 *s; =20 for (s =3D start; s < end; s++) { void *addr =3D (void *)s + *s; + void *wr_addr =3D module_writable_address(mod, addr); u32 hash; =20 addr -=3D fineibt_caller_size; hash =3D decode_caller_hash(addr); if (hash) { hash =3D -cfi_rehash(hash); - text_poke_early(addr + 2, &hash, 4); + text_poke_early(wr_addr + 2, &hash, 4); } } =20 return 0; } =20 -static int cfi_rewrite_callers(s32 *start, s32 *end) +static int cfi_rewrite_callers(s32 *start, s32 *end, struct module *mod) { s32 *s; =20 for (s =3D start; s < end; s++) { void *addr =3D (void *)s + *s; + void *wr_addr =3D module_writable_address(mod, addr); u32 hash; =20 addr -=3D fineibt_caller_size; hash =3D decode_caller_hash(addr); if (hash) { - text_poke_early(addr, fineibt_caller_start, fineibt_caller_size); + text_poke_early(wr_addr, fineibt_caller_start, fineibt_caller_size); WARN_ON(*(u32 *)(addr + fineibt_caller_hash) !=3D 0x12345678); - text_poke_early(addr + fineibt_caller_hash, &hash, 4); + text_poke_early(wr_addr + fineibt_caller_hash, &hash, 4); } /* rely on apply_retpolines() */ } @@ -1262,8 +1286,9 @@ static int cfi_rewrite_callers(s32 *start, s32 *end) } =20 static void __apply_fineibt(s32 *start_retpoline, s32 *end_retpoline, - s32 *start_cfi, s32 *end_cfi, bool builtin) + s32 *start_cfi, s32 *end_cfi, struct module *mod) { + bool builtin =3D mod ? false : true; int ret; =20 if (WARN_ONCE(fineibt_preamble_size !=3D 16, @@ -1281,7 +1306,7 @@ static void __apply_fineibt(s32 *start_retpoline, s32= *end_retpoline, * rewrite them. This disables all CFI. If this succeeds but any of the * later stages fails, we're without CFI. */ - ret =3D cfi_disable_callers(start_retpoline, end_retpoline); + ret =3D cfi_disable_callers(start_retpoline, end_retpoline, mod); if (ret) goto err; =20 @@ -1292,11 +1317,11 @@ static void __apply_fineibt(s32 *start_retpoline, s= 32 *end_retpoline, cfi_bpf_subprog_hash =3D cfi_rehash(cfi_bpf_subprog_hash); } =20 - ret =3D cfi_rand_preamble(start_cfi, end_cfi); + ret =3D cfi_rand_preamble(start_cfi, end_cfi, mod); if (ret) goto err; =20 - ret =3D cfi_rand_callers(start_retpoline, end_retpoline); + ret =3D cfi_rand_callers(start_retpoline, end_retpoline, mod); if (ret) goto err; } @@ -1308,7 +1333,7 @@ static void __apply_fineibt(s32 *start_retpoline, s32= *end_retpoline, return; =20 case CFI_KCFI: - ret =3D cfi_enable_callers(start_retpoline, end_retpoline); + ret =3D cfi_enable_callers(start_retpoline, end_retpoline, mod); if (ret) goto err; =20 @@ -1318,17 +1343,17 @@ static void __apply_fineibt(s32 *start_retpoline, s= 32 *end_retpoline, =20 case CFI_FINEIBT: /* place the FineIBT preamble at func()-16 */ - ret =3D cfi_rewrite_preamble(start_cfi, end_cfi); + ret =3D cfi_rewrite_preamble(start_cfi, end_cfi, mod); if (ret) goto err; =20 /* rewrite the callers to target func()-16 */ - ret =3D cfi_rewrite_callers(start_retpoline, end_retpoline); + ret =3D cfi_rewrite_callers(start_retpoline, end_retpoline, mod); if (ret) goto err; =20 /* now that nobody targets func()+0, remove ENDBR there */ - cfi_rewrite_endbr(start_cfi, end_cfi); + cfi_rewrite_endbr(start_cfi, end_cfi, mod); =20 if (builtin) pr_info("Using FineIBT CFI\n"); @@ -1347,7 +1372,7 @@ static inline void poison_hash(void *addr) *(u32 *)addr =3D 0; } =20 -static void poison_cfi(void *addr) +static void poison_cfi(void *addr, void *wr_addr) { switch (cfi_mode) { case CFI_FINEIBT: @@ -1359,8 +1384,8 @@ static void poison_cfi(void *addr) * ud2 * 1: nop */ - poison_endbr(addr, false); - poison_hash(addr + fineibt_preamble_hash); + poison_endbr(addr, wr_addr, false); + poison_hash(wr_addr + fineibt_preamble_hash); break; =20 case CFI_KCFI: @@ -1369,7 +1394,7 @@ static void poison_cfi(void *addr) * movl $0, %eax * .skip 11, 0x90 */ - poison_hash(addr + 1); + poison_hash(wr_addr + 1); break; =20 default: @@ -1380,22 +1405,21 @@ static void poison_cfi(void *addr) #else =20 static void __apply_fineibt(s32 *start_retpoline, s32 *end_retpoline, - s32 *start_cfi, s32 *end_cfi, bool builtin) + s32 *start_cfi, s32 *end_cfi, struct module *mod) { } =20 #ifdef CONFIG_X86_KERNEL_IBT -static void poison_cfi(void *addr) { } +static void poison_cfi(void *addr, void *wr_addr) { } #endif =20 #endif =20 void apply_fineibt(s32 *start_retpoline, s32 *end_retpoline, - s32 *start_cfi, s32 *end_cfi) + s32 *start_cfi, s32 *end_cfi, struct module *mod) { return __apply_fineibt(start_retpoline, end_retpoline, - start_cfi, end_cfi, - /* .builtin =3D */ false); + start_cfi, end_cfi, mod); } =20 #ifdef CONFIG_SMP @@ -1692,16 +1716,16 @@ void __init alternative_instructions(void) paravirt_set_cap(); =20 __apply_fineibt(__retpoline_sites, __retpoline_sites_end, - __cfi_sites, __cfi_sites_end, true); + __cfi_sites, __cfi_sites_end, NULL); =20 /* * Rewrite the retpolines, must be done before alternatives since * those can rewrite the retpoline thunks. */ - apply_retpolines(__retpoline_sites, __retpoline_sites_end); - apply_returns(__return_sites, __return_sites_end); + apply_retpolines(__retpoline_sites, __retpoline_sites_end, NULL); + apply_returns(__return_sites, __return_sites_end, NULL); =20 - apply_alternatives(__alt_instructions, __alt_instructions_end); + apply_alternatives(__alt_instructions, __alt_instructions_end, NULL); =20 /* * Now all calls are established. Apply the call thunks if @@ -1712,7 +1736,7 @@ void __init alternative_instructions(void) /* * Seal all functions that do not have their address taken. */ - apply_seal_endbr(__ibt_endbr_seal, __ibt_endbr_seal_end); + apply_seal_endbr(__ibt_endbr_seal, __ibt_endbr_seal_end, NULL); =20 #ifdef CONFIG_SMP /* Patch to UP if other cpus not imminent. */ diff --git a/arch/x86/kernel/ftrace.c b/arch/x86/kernel/ftrace.c index 8da0e66ca22d..b498897b213c 100644 --- a/arch/x86/kernel/ftrace.c +++ b/arch/x86/kernel/ftrace.c @@ -118,10 +118,13 @@ ftrace_modify_code_direct(unsigned long ip, const cha= r *old_code, return ret; =20 /* replace the text with the new text */ - if (ftrace_poke_late) + if (ftrace_poke_late) { text_poke_queue((void *)ip, new_code, MCOUNT_INSN_SIZE, NULL); - else - text_poke_early((void *)ip, new_code, MCOUNT_INSN_SIZE); + } else { + mutex_lock(&text_mutex); + text_poke((void *)ip, new_code, MCOUNT_INSN_SIZE); + mutex_unlock(&text_mutex); + } return 0; } =20 @@ -318,7 +321,7 @@ create_trampoline(struct ftrace_ops *ops, unsigned int = *tramp_size) unsigned const char op_ref[] =3D { 0x48, 0x8b, 0x15 }; unsigned const char retq[] =3D { RET_INSN_OPCODE, INT3_INSN_OPCODE }; union ftrace_op_code_union op_ptr; - int ret; + void *ret; =20 if (ops->flags & FTRACE_OPS_FL_SAVE_REGS) { start_offset =3D (unsigned long)ftrace_regs_caller; @@ -349,15 +352,15 @@ create_trampoline(struct ftrace_ops *ops, unsigned in= t *tramp_size) npages =3D DIV_ROUND_UP(*tramp_size, PAGE_SIZE); =20 /* Copy ftrace_caller onto the trampoline memory */ - ret =3D copy_from_kernel_nofault(trampoline, (void *)start_offset, size); - if (WARN_ON(ret < 0)) + ret =3D text_poke_copy(trampoline, (void *)start_offset, size); + if (WARN_ON(!ret)) goto fail; =20 ip =3D trampoline + size; if (cpu_feature_enabled(X86_FEATURE_RETHUNK)) __text_gen_insn(ip, JMP32_INSN_OPCODE, ip, x86_return_thunk, JMP32_INSN_= SIZE); else - memcpy(ip, retq, sizeof(retq)); + text_poke_copy(ip, retq, sizeof(retq)); =20 /* No need to test direct calls on created trampolines */ if (ops->flags & FTRACE_OPS_FL_SAVE_REGS) { @@ -365,8 +368,7 @@ create_trampoline(struct ftrace_ops *ops, unsigned int = *tramp_size) ip =3D trampoline + (jmp_offset - start_offset); if (WARN_ON(*(char *)ip !=3D 0x75)) goto fail; - ret =3D copy_from_kernel_nofault(ip, x86_nops[2], 2); - if (ret < 0) + if (!text_poke_copy(ip, x86_nops[2], 2)) goto fail; } =20 @@ -379,7 +381,7 @@ create_trampoline(struct ftrace_ops *ops, unsigned int = *tramp_size) */ =20 ptr =3D (unsigned long *)(trampoline + size + RET_SIZE); - *ptr =3D (unsigned long)ops; + text_poke_copy(ptr, &ops, sizeof(unsigned long)); =20 op_offset -=3D start_offset; memcpy(&op_ptr, trampoline + op_offset, OP_REF_SIZE); @@ -395,7 +397,7 @@ create_trampoline(struct ftrace_ops *ops, unsigned int = *tramp_size) op_ptr.offset =3D offset; =20 /* put in the new offset to the ftrace_ops */ - memcpy(trampoline + op_offset, &op_ptr, OP_REF_SIZE); + text_poke_copy(trampoline + op_offset, &op_ptr, OP_REF_SIZE); =20 /* put in the call to the function */ mutex_lock(&text_mutex); @@ -405,9 +407,9 @@ create_trampoline(struct ftrace_ops *ops, unsigned int = *tramp_size) * the depth accounting before the call already. */ dest =3D ftrace_ops_get_func(ops); - memcpy(trampoline + call_offset, - text_gen_insn(CALL_INSN_OPCODE, trampoline + call_offset, dest), - CALL_INSN_SIZE); + text_poke_copy_locked(trampoline + call_offset, + text_gen_insn(CALL_INSN_OPCODE, trampoline + call_offset, dest), + CALL_INSN_SIZE, false); mutex_unlock(&text_mutex); =20 /* ALLOC_TRAMP flags lets us know we created it */ diff --git a/arch/x86/kernel/module.c b/arch/x86/kernel/module.c index 837450b6e882..8984abd91c00 100644 --- a/arch/x86/kernel/module.c +++ b/arch/x86/kernel/module.c @@ -146,18 +146,21 @@ static int __write_relocate_add(Elf64_Shdr *sechdrs, } =20 if (apply) { - if (memcmp(loc, &zero, size)) { + void *wr_loc =3D module_writable_address(me, loc); + + if (memcmp(wr_loc, &zero, size)) { pr_err("x86/modules: Invalid relocation target, existing value is nonz= ero for type %d, loc %p, val %Lx\n", (int)ELF64_R_TYPE(rel[i].r_info), loc, val); return -ENOEXEC; } - write(loc, &val, size); + write(wr_loc, &val, size); } else { if (memcmp(loc, &val, size)) { pr_warn("x86/modules: Invalid relocation target, existing value does n= ot match expected value for type %d, loc %p, val %Lx\n", (int)ELF64_R_TYPE(rel[i].r_info), loc, val); return -ENOEXEC; } + /* FIXME: needs care for ROX module allocations */ write(loc, &zero, size); } } @@ -224,7 +227,7 @@ int module_finalize(const Elf_Ehdr *hdr, const Elf_Shdr *sechdrs, struct module *me) { - const Elf_Shdr *s, *alt =3D NULL, *locks =3D NULL, + const Elf_Shdr *s, *alt =3D NULL, *orc =3D NULL, *orc_ip =3D NULL, *retpolines =3D NULL, *returns =3D NULL, *ibt_endbr =3D NULL, *calls =3D NULL, *cfi =3D NULL; @@ -233,8 +236,6 @@ int module_finalize(const Elf_Ehdr *hdr, for (s =3D sechdrs; s < sechdrs + hdr->e_shnum; s++) { if (!strcmp(".altinstructions", secstrings + s->sh_name)) alt =3D s; - if (!strcmp(".smp_locks", secstrings + s->sh_name)) - locks =3D s; if (!strcmp(".orc_unwind", secstrings + s->sh_name)) orc =3D s; if (!strcmp(".orc_unwind_ip", secstrings + s->sh_name)) @@ -265,20 +266,20 @@ int module_finalize(const Elf_Ehdr *hdr, csize =3D cfi->sh_size; } =20 - apply_fineibt(rseg, rseg + rsize, cseg, cseg + csize); + apply_fineibt(rseg, rseg + rsize, cseg, cseg + csize, me); } if (retpolines) { void *rseg =3D (void *)retpolines->sh_addr; - apply_retpolines(rseg, rseg + retpolines->sh_size); + apply_retpolines(rseg, rseg + retpolines->sh_size, me); } if (returns) { void *rseg =3D (void *)returns->sh_addr; - apply_returns(rseg, rseg + returns->sh_size); + apply_returns(rseg, rseg + returns->sh_size, me); } if (alt) { /* patch .altinstructions */ void *aseg =3D (void *)alt->sh_addr; - apply_alternatives(aseg, aseg + alt->sh_size); + apply_alternatives(aseg, aseg + alt->sh_size, me); } if (calls || alt) { struct callthunk_sites cs =3D {}; @@ -297,8 +298,28 @@ int module_finalize(const Elf_Ehdr *hdr, } if (ibt_endbr) { void *iseg =3D (void *)ibt_endbr->sh_addr; - apply_seal_endbr(iseg, iseg + ibt_endbr->sh_size); + apply_seal_endbr(iseg, iseg + ibt_endbr->sh_size, me); } + + if (orc && orc_ip) + unwind_module_init(me, (void *)orc_ip->sh_addr, orc_ip->sh_size, + (void *)orc->sh_addr, orc->sh_size); + + return 0; +} + +int module_post_finalize(const Elf_Ehdr *hdr, + const Elf_Shdr *sechdrs, + struct module *me) +{ + const Elf_Shdr *s, *locks =3D NULL; + char *secstrings =3D (void *)hdr + sechdrs[hdr->e_shstrndx].sh_offset; + + for (s =3D sechdrs; s < sechdrs + hdr->e_shnum; s++) { + if (!strcmp(".smp_locks", secstrings + s->sh_name)) + locks =3D s; + } + if (locks) { void *lseg =3D (void *)locks->sh_addr; void *text =3D me->mem[MOD_TEXT].base; @@ -308,10 +329,6 @@ int module_finalize(const Elf_Ehdr *hdr, text, text_end); } =20 - if (orc && orc_ip) - unwind_module_init(me, (void *)orc_ip->sh_addr, orc_ip->sh_size, - (void *)orc->sh_addr, orc->sh_size); - return 0; } =20 diff --git a/include/linux/module.h b/include/linux/module.h index 7039f609c6ef..2c1a24ba99a3 100644 --- a/include/linux/module.h +++ b/include/linux/module.h @@ -376,15 +376,6 @@ struct module_memory { #endif }; =20 -#ifdef CONFIG_MODULES -void *module_writable_address(struct module *mod, void *loc); -#else -static inline void *module_writable_address(struct module *mod, void *loc) -{ - return loc; -} -#endif - #ifdef CONFIG_MODULES_TREE_LOOKUP /* Only touch one cacheline for common rbtree-for-core-layout case. */ #define __module_memory_align ____cacheline_aligned @@ -778,6 +769,8 @@ static inline bool is_livepatch_module(struct module *m= od) =20 void set_module_sig_enforced(void); =20 +void *module_writable_address(struct module *mod, void *loc); + #else /* !CONFIG_MODULES... */ =20 static inline struct module *__module_address(unsigned long addr) @@ -885,6 +878,11 @@ static inline bool module_is_coming(struct module *mod) { return false; } + +static inline void *module_writable_address(struct module *mod, void *loc) +{ + return loc; +} #endif /* CONFIG_MODULES */ =20 #ifdef CONFIG_SYSFS --=20 2.43.0