From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org X-Spam-Level: X-Spam-Status: No, score=-9.0 required=3.0 tests=HEADER_FROM_DIFFERENT_DOMAINS, INCLUDES_PATCH,MAILING_LIST_MULTI,SIGNED_OFF_BY,SPF_PASS,USER_AGENT_GIT autolearn=ham autolearn_force=no version=3.4.0 Received: from mail.kernel.org (mail.kernel.org [198.145.29.99]) by smtp.lore.kernel.org (Postfix) with ESMTP id 2D052C10F14 for ; Sun, 21 Apr 2019 14:22:34 +0000 (UTC) Received: from vger.kernel.org (vger.kernel.org [209.132.180.67]) by mail.kernel.org (Postfix) with ESMTP id 06D0A20857 for ; Sun, 21 Apr 2019 14:22:34 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1727655AbfDUOWd (ORCPT ); Sun, 21 Apr 2019 10:22:33 -0400 Received: from ozlabs.org ([203.11.71.1]:45107 "EHLO ozlabs.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1727368AbfDUOWd (ORCPT ); Sun, 21 Apr 2019 10:22:33 -0400 Received: by ozlabs.org (Postfix, from userid 1034) id 44nBkG1nFPz9sML; Mon, 22 Apr 2019 00:22:29 +1000 (AEST) From: Michael Ellerman To: stable@vger.kernel.org, gregkh@linuxfoundation.org Cc: linuxppc-dev@ozlabs.org, diana.craciun@nxp.com, msuchanek@suse.de, npiggin@gmail.com, christophe.leroy@c-s.fr Subject: [PATCH stable v4.4 41/52] powerpc/64s: Add support for software count cache flush Date: Mon, 22 Apr 2019 00:20:26 +1000 Message-Id: <20190421142037.21881-42-mpe@ellerman.id.au> X-Mailer: git-send-email 2.20.1 In-Reply-To: <20190421142037.21881-1-mpe@ellerman.id.au> References: <20190421142037.21881-1-mpe@ellerman.id.au> MIME-Version: 1.0 Content-Transfer-Encoding: 8bit Sender: stable-owner@vger.kernel.org Precedence: bulk List-ID: X-Mailing-List: stable@vger.kernel.org commit ee13cb249fabdff8b90aaff61add347749280087 upstream. Some CPU revisions support a mode where the count cache needs to be flushed by software on context switch. Additionally some revisions may have a hardware accelerated flush, in which case the software flush sequence can be shortened. If we detect the appropriate flag from firmware we patch a branch into _switch() which takes us to a count cache flush sequence. That sequence in turn may be patched to return early if we detect that the CPU supports accelerating the flush sequence in hardware. Add debugfs support for reporting the state of the flush, as well as runtime disabling it. And modify the spectre_v2 sysfs file to report the state of the software flush. Signed-off-by: Michael Ellerman --- arch/powerpc/include/asm/asm-prototypes.h | 21 +++++ arch/powerpc/include/asm/security_features.h | 1 + arch/powerpc/kernel/entry_64.S | 54 +++++++++++ arch/powerpc/kernel/security.c | 98 +++++++++++++++++++- 4 files changed, 169 insertions(+), 5 deletions(-) create mode 100644 arch/powerpc/include/asm/asm-prototypes.h diff --git a/arch/powerpc/include/asm/asm-prototypes.h b/arch/powerpc/include/asm/asm-prototypes.h new file mode 100644 index 000000000000..8944c55591cf --- /dev/null +++ b/arch/powerpc/include/asm/asm-prototypes.h @@ -0,0 +1,21 @@ +#ifndef _ASM_POWERPC_ASM_PROTOTYPES_H +#define _ASM_POWERPC_ASM_PROTOTYPES_H +/* + * This file is for prototypes of C functions that are only called + * from asm, and any associated variables. + * + * Copyright 2016, Daniel Axtens, IBM Corporation. + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License + * as published by the Free Software Foundation; either version 2 + * of the License, or (at your option) any later version. + */ + +/* Patch sites */ +extern s32 patch__call_flush_count_cache; +extern s32 patch__flush_count_cache_return; + +extern long flush_count_cache; + +#endif /* _ASM_POWERPC_ASM_PROTOTYPES_H */ diff --git a/arch/powerpc/include/asm/security_features.h b/arch/powerpc/include/asm/security_features.h index a0d47bc18a5c..759597bf0fd8 100644 --- a/arch/powerpc/include/asm/security_features.h +++ b/arch/powerpc/include/asm/security_features.h @@ -22,6 +22,7 @@ enum stf_barrier_type { void setup_stf_barrier(void); void do_stf_barrier_fixups(enum stf_barrier_type types); +void setup_count_cache_flush(void); static inline void security_ftr_set(unsigned long feature) { diff --git a/arch/powerpc/kernel/entry_64.S b/arch/powerpc/kernel/entry_64.S index 9c2e58e0e55e..698bb51f5399 100644 --- a/arch/powerpc/kernel/entry_64.S +++ b/arch/powerpc/kernel/entry_64.S @@ -25,6 +25,7 @@ #include #include #include +#include #include #include #include @@ -450,6 +451,57 @@ _GLOBAL(ret_from_kernel_thread) li r3,0 b .Lsyscall_exit +#ifdef CONFIG_PPC_BOOK3S_64 + +#define FLUSH_COUNT_CACHE \ +1: nop; \ + patch_site 1b, patch__call_flush_count_cache + + +#define BCCTR_FLUSH .long 0x4c400420 + +.macro nops number + .rept \number + nop + .endr +.endm + +.balign 32 +.global flush_count_cache +flush_count_cache: + /* Save LR into r9 */ + mflr r9 + + .rept 64 + bl .+4 + .endr + b 1f + nops 6 + + .balign 32 + /* Restore LR */ +1: mtlr r9 + li r9,0x7fff + mtctr r9 + + BCCTR_FLUSH + +2: nop + patch_site 2b patch__flush_count_cache_return + + nops 3 + + .rept 278 + .balign 32 + BCCTR_FLUSH + nops 7 + .endr + + blr +#else +#define FLUSH_COUNT_CACHE +#endif /* CONFIG_PPC_BOOK3S_64 */ + /* * This routine switches between two different tasks. The process * state of one is saved on its kernel stack. Then the state @@ -513,6 +565,8 @@ BEGIN_FTR_SECTION END_FTR_SECTION_IFSET(CPU_FTR_ARCH_207S) #endif + FLUSH_COUNT_CACHE + #ifdef CONFIG_SMP /* We need a sync somewhere here to make sure that if the * previous task gets rescheduled on another CPU, it sees all diff --git a/arch/powerpc/kernel/security.c b/arch/powerpc/kernel/security.c index e04afd171b13..108d271a218d 100644 --- a/arch/powerpc/kernel/security.c +++ b/arch/powerpc/kernel/security.c @@ -10,12 +10,21 @@ #include #include +#include +#include #include #include unsigned long powerpc_security_features __read_mostly = SEC_FTR_DEFAULT; +enum count_cache_flush_type { + COUNT_CACHE_FLUSH_NONE = 0x1, + COUNT_CACHE_FLUSH_SW = 0x2, + COUNT_CACHE_FLUSH_HW = 0x4, +}; +static enum count_cache_flush_type count_cache_flush_type; + bool barrier_nospec_enabled; static bool no_nospec; @@ -160,17 +169,29 @@ ssize_t cpu_show_spectre_v2(struct device *dev, struct device_attribute *attr, c bcs = security_ftr_enabled(SEC_FTR_BCCTRL_SERIALISED); ccd = security_ftr_enabled(SEC_FTR_COUNT_CACHE_DISABLED); - if (bcs || ccd) { + if (bcs || ccd || count_cache_flush_type != COUNT_CACHE_FLUSH_NONE) { + bool comma = false; seq_buf_printf(&s, "Mitigation: "); - if (bcs) + if (bcs) { seq_buf_printf(&s, "Indirect branch serialisation (kernel only)"); + comma = true; + } + + if (ccd) { + if (comma) + seq_buf_printf(&s, ", "); + seq_buf_printf(&s, "Indirect branch cache disabled"); + comma = true; + } - if (bcs && ccd) + if (comma) seq_buf_printf(&s, ", "); - if (ccd) - seq_buf_printf(&s, "Indirect branch cache disabled"); + seq_buf_printf(&s, "Software count cache flush"); + + if (count_cache_flush_type == COUNT_CACHE_FLUSH_HW) + seq_buf_printf(&s, "(hardware accelerated)"); } else seq_buf_printf(&s, "Vulnerable"); @@ -325,4 +346,71 @@ static __init int stf_barrier_debugfs_init(void) } device_initcall(stf_barrier_debugfs_init); #endif /* CONFIG_DEBUG_FS */ + +static void toggle_count_cache_flush(bool enable) +{ + if (!enable || !security_ftr_enabled(SEC_FTR_FLUSH_COUNT_CACHE)) { + patch_instruction_site(&patch__call_flush_count_cache, PPC_INST_NOP); + count_cache_flush_type = COUNT_CACHE_FLUSH_NONE; + pr_info("count-cache-flush: software flush disabled.\n"); + return; + } + + patch_branch_site(&patch__call_flush_count_cache, + (u64)&flush_count_cache, BRANCH_SET_LINK); + + if (!security_ftr_enabled(SEC_FTR_BCCTR_FLUSH_ASSIST)) { + count_cache_flush_type = COUNT_CACHE_FLUSH_SW; + pr_info("count-cache-flush: full software flush sequence enabled.\n"); + return; + } + + patch_instruction_site(&patch__flush_count_cache_return, PPC_INST_BLR); + count_cache_flush_type = COUNT_CACHE_FLUSH_HW; + pr_info("count-cache-flush: hardware assisted flush sequence enabled\n"); +} + +void setup_count_cache_flush(void) +{ + toggle_count_cache_flush(true); +} + +#ifdef CONFIG_DEBUG_FS +static int count_cache_flush_set(void *data, u64 val) +{ + bool enable; + + if (val == 1) + enable = true; + else if (val == 0) + enable = false; + else + return -EINVAL; + + toggle_count_cache_flush(enable); + + return 0; +} + +static int count_cache_flush_get(void *data, u64 *val) +{ + if (count_cache_flush_type == COUNT_CACHE_FLUSH_NONE) + *val = 0; + else + *val = 1; + + return 0; +} + +DEFINE_SIMPLE_ATTRIBUTE(fops_count_cache_flush, count_cache_flush_get, + count_cache_flush_set, "%llu\n"); + +static __init int count_cache_flush_debugfs_init(void) +{ + debugfs_create_file("count_cache_flush", 0600, powerpc_debugfs_root, + NULL, &fops_count_cache_flush); + return 0; +} +device_initcall(count_cache_flush_debugfs_init); +#endif /* CONFIG_DEBUG_FS */ #endif /* CONFIG_PPC_BOOK3S_64 */ -- 2.20.1 From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org X-Spam-Level: X-Spam-Status: No, score=-9.0 required=3.0 tests=HEADER_FROM_DIFFERENT_DOMAINS, INCLUDES_PATCH,MAILING_LIST_MULTI,SIGNED_OFF_BY,SPF_PASS,USER_AGENT_GIT autolearn=unavailable autolearn_force=no version=3.4.0 Received: from mail.kernel.org (mail.kernel.org [198.145.29.99]) by smtp.lore.kernel.org (Postfix) with ESMTP id 55BB3C10F14 for ; Sun, 21 Apr 2019 16:11:53 +0000 (UTC) Received: from lists.ozlabs.org (lists.ozlabs.org [203.11.71.2]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by mail.kernel.org (Postfix) with ESMTPS id 9D2B420821 for ; Sun, 21 Apr 2019 16:11:52 +0000 (UTC) DMARC-Filter: OpenDMARC Filter v1.3.2 mail.kernel.org 9D2B420821 Authentication-Results: mail.kernel.org; dmarc=none (p=none dis=none) header.from=ellerman.id.au Authentication-Results: mail.kernel.org; spf=pass smtp.mailfrom=linuxppc-dev-bounces+linuxppc-dev=archiver.kernel.org@lists.ozlabs.org Received: from lists.ozlabs.org (lists.ozlabs.org [IPv6:2401:3900:2:1::3]) by lists.ozlabs.org (Postfix) with ESMTP id 44nF8Q3chzzDqW0 for ; Mon, 22 Apr 2019 02:11:50 +1000 (AEST) Received: from ozlabs.org (bilbo.ozlabs.org [IPv6:2401:3900:2:1::2]) (using TLSv1.3 with cipher TLS_AES_256_GCM_SHA384 (256/256 bits) key-exchange X25519 server-signature RSA-PSS (2048 bits)) (No client certificate requested) by lists.ozlabs.org (Postfix) with ESMTPS id 44nBkH3B7fzDqMH for ; Mon, 22 Apr 2019 00:22:31 +1000 (AEST) Authentication-Results: lists.ozlabs.org; dmarc=none (p=none dis=none) header.from=ellerman.id.au Received: by ozlabs.org (Postfix) id 44nBkG5Rsbz9sNL; Mon, 22 Apr 2019 00:22:30 +1000 (AEST) Received: by ozlabs.org (Postfix, from userid 1034) id 44nBkG1nFPz9sML; Mon, 22 Apr 2019 00:22:29 +1000 (AEST) From: Michael Ellerman To: stable@vger.kernel.org, gregkh@linuxfoundation.org Subject: [PATCH stable v4.4 41/52] powerpc/64s: Add support for software count cache flush Date: Mon, 22 Apr 2019 00:20:26 +1000 Message-Id: <20190421142037.21881-42-mpe@ellerman.id.au> X-Mailer: git-send-email 2.20.1 In-Reply-To: <20190421142037.21881-1-mpe@ellerman.id.au> References: <20190421142037.21881-1-mpe@ellerman.id.au> MIME-Version: 1.0 Content-Transfer-Encoding: 8bit X-BeenThere: linuxppc-dev@lists.ozlabs.org X-Mailman-Version: 2.1.29 Precedence: list List-Id: Linux on PowerPC Developers Mail List List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Cc: diana.craciun@nxp.com, linuxppc-dev@ozlabs.org, msuchanek@suse.de, npiggin@gmail.com Errors-To: linuxppc-dev-bounces+linuxppc-dev=archiver.kernel.org@lists.ozlabs.org Sender: "Linuxppc-dev" commit ee13cb249fabdff8b90aaff61add347749280087 upstream. Some CPU revisions support a mode where the count cache needs to be flushed by software on context switch. Additionally some revisions may have a hardware accelerated flush, in which case the software flush sequence can be shortened. If we detect the appropriate flag from firmware we patch a branch into _switch() which takes us to a count cache flush sequence. That sequence in turn may be patched to return early if we detect that the CPU supports accelerating the flush sequence in hardware. Add debugfs support for reporting the state of the flush, as well as runtime disabling it. And modify the spectre_v2 sysfs file to report the state of the software flush. Signed-off-by: Michael Ellerman --- arch/powerpc/include/asm/asm-prototypes.h | 21 +++++ arch/powerpc/include/asm/security_features.h | 1 + arch/powerpc/kernel/entry_64.S | 54 +++++++++++ arch/powerpc/kernel/security.c | 98 +++++++++++++++++++- 4 files changed, 169 insertions(+), 5 deletions(-) create mode 100644 arch/powerpc/include/asm/asm-prototypes.h diff --git a/arch/powerpc/include/asm/asm-prototypes.h b/arch/powerpc/include/asm/asm-prototypes.h new file mode 100644 index 000000000000..8944c55591cf --- /dev/null +++ b/arch/powerpc/include/asm/asm-prototypes.h @@ -0,0 +1,21 @@ +#ifndef _ASM_POWERPC_ASM_PROTOTYPES_H +#define _ASM_POWERPC_ASM_PROTOTYPES_H +/* + * This file is for prototypes of C functions that are only called + * from asm, and any associated variables. + * + * Copyright 2016, Daniel Axtens, IBM Corporation. + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License + * as published by the Free Software Foundation; either version 2 + * of the License, or (at your option) any later version. + */ + +/* Patch sites */ +extern s32 patch__call_flush_count_cache; +extern s32 patch__flush_count_cache_return; + +extern long flush_count_cache; + +#endif /* _ASM_POWERPC_ASM_PROTOTYPES_H */ diff --git a/arch/powerpc/include/asm/security_features.h b/arch/powerpc/include/asm/security_features.h index a0d47bc18a5c..759597bf0fd8 100644 --- a/arch/powerpc/include/asm/security_features.h +++ b/arch/powerpc/include/asm/security_features.h @@ -22,6 +22,7 @@ enum stf_barrier_type { void setup_stf_barrier(void); void do_stf_barrier_fixups(enum stf_barrier_type types); +void setup_count_cache_flush(void); static inline void security_ftr_set(unsigned long feature) { diff --git a/arch/powerpc/kernel/entry_64.S b/arch/powerpc/kernel/entry_64.S index 9c2e58e0e55e..698bb51f5399 100644 --- a/arch/powerpc/kernel/entry_64.S +++ b/arch/powerpc/kernel/entry_64.S @@ -25,6 +25,7 @@ #include #include #include +#include #include #include #include @@ -450,6 +451,57 @@ _GLOBAL(ret_from_kernel_thread) li r3,0 b .Lsyscall_exit +#ifdef CONFIG_PPC_BOOK3S_64 + +#define FLUSH_COUNT_CACHE \ +1: nop; \ + patch_site 1b, patch__call_flush_count_cache + + +#define BCCTR_FLUSH .long 0x4c400420 + +.macro nops number + .rept \number + nop + .endr +.endm + +.balign 32 +.global flush_count_cache +flush_count_cache: + /* Save LR into r9 */ + mflr r9 + + .rept 64 + bl .+4 + .endr + b 1f + nops 6 + + .balign 32 + /* Restore LR */ +1: mtlr r9 + li r9,0x7fff + mtctr r9 + + BCCTR_FLUSH + +2: nop + patch_site 2b patch__flush_count_cache_return + + nops 3 + + .rept 278 + .balign 32 + BCCTR_FLUSH + nops 7 + .endr + + blr +#else +#define FLUSH_COUNT_CACHE +#endif /* CONFIG_PPC_BOOK3S_64 */ + /* * This routine switches between two different tasks. The process * state of one is saved on its kernel stack. Then the state @@ -513,6 +565,8 @@ BEGIN_FTR_SECTION END_FTR_SECTION_IFSET(CPU_FTR_ARCH_207S) #endif + FLUSH_COUNT_CACHE + #ifdef CONFIG_SMP /* We need a sync somewhere here to make sure that if the * previous task gets rescheduled on another CPU, it sees all diff --git a/arch/powerpc/kernel/security.c b/arch/powerpc/kernel/security.c index e04afd171b13..108d271a218d 100644 --- a/arch/powerpc/kernel/security.c +++ b/arch/powerpc/kernel/security.c @@ -10,12 +10,21 @@ #include #include +#include +#include #include #include unsigned long powerpc_security_features __read_mostly = SEC_FTR_DEFAULT; +enum count_cache_flush_type { + COUNT_CACHE_FLUSH_NONE = 0x1, + COUNT_CACHE_FLUSH_SW = 0x2, + COUNT_CACHE_FLUSH_HW = 0x4, +}; +static enum count_cache_flush_type count_cache_flush_type; + bool barrier_nospec_enabled; static bool no_nospec; @@ -160,17 +169,29 @@ ssize_t cpu_show_spectre_v2(struct device *dev, struct device_attribute *attr, c bcs = security_ftr_enabled(SEC_FTR_BCCTRL_SERIALISED); ccd = security_ftr_enabled(SEC_FTR_COUNT_CACHE_DISABLED); - if (bcs || ccd) { + if (bcs || ccd || count_cache_flush_type != COUNT_CACHE_FLUSH_NONE) { + bool comma = false; seq_buf_printf(&s, "Mitigation: "); - if (bcs) + if (bcs) { seq_buf_printf(&s, "Indirect branch serialisation (kernel only)"); + comma = true; + } + + if (ccd) { + if (comma) + seq_buf_printf(&s, ", "); + seq_buf_printf(&s, "Indirect branch cache disabled"); + comma = true; + } - if (bcs && ccd) + if (comma) seq_buf_printf(&s, ", "); - if (ccd) - seq_buf_printf(&s, "Indirect branch cache disabled"); + seq_buf_printf(&s, "Software count cache flush"); + + if (count_cache_flush_type == COUNT_CACHE_FLUSH_HW) + seq_buf_printf(&s, "(hardware accelerated)"); } else seq_buf_printf(&s, "Vulnerable"); @@ -325,4 +346,71 @@ static __init int stf_barrier_debugfs_init(void) } device_initcall(stf_barrier_debugfs_init); #endif /* CONFIG_DEBUG_FS */ + +static void toggle_count_cache_flush(bool enable) +{ + if (!enable || !security_ftr_enabled(SEC_FTR_FLUSH_COUNT_CACHE)) { + patch_instruction_site(&patch__call_flush_count_cache, PPC_INST_NOP); + count_cache_flush_type = COUNT_CACHE_FLUSH_NONE; + pr_info("count-cache-flush: software flush disabled.\n"); + return; + } + + patch_branch_site(&patch__call_flush_count_cache, + (u64)&flush_count_cache, BRANCH_SET_LINK); + + if (!security_ftr_enabled(SEC_FTR_BCCTR_FLUSH_ASSIST)) { + count_cache_flush_type = COUNT_CACHE_FLUSH_SW; + pr_info("count-cache-flush: full software flush sequence enabled.\n"); + return; + } + + patch_instruction_site(&patch__flush_count_cache_return, PPC_INST_BLR); + count_cache_flush_type = COUNT_CACHE_FLUSH_HW; + pr_info("count-cache-flush: hardware assisted flush sequence enabled\n"); +} + +void setup_count_cache_flush(void) +{ + toggle_count_cache_flush(true); +} + +#ifdef CONFIG_DEBUG_FS +static int count_cache_flush_set(void *data, u64 val) +{ + bool enable; + + if (val == 1) + enable = true; + else if (val == 0) + enable = false; + else + return -EINVAL; + + toggle_count_cache_flush(enable); + + return 0; +} + +static int count_cache_flush_get(void *data, u64 *val) +{ + if (count_cache_flush_type == COUNT_CACHE_FLUSH_NONE) + *val = 0; + else + *val = 1; + + return 0; +} + +DEFINE_SIMPLE_ATTRIBUTE(fops_count_cache_flush, count_cache_flush_get, + count_cache_flush_set, "%llu\n"); + +static __init int count_cache_flush_debugfs_init(void) +{ + debugfs_create_file("count_cache_flush", 0600, powerpc_debugfs_root, + NULL, &fops_count_cache_flush); + return 0; +} +device_initcall(count_cache_flush_debugfs_init); +#endif /* CONFIG_DEBUG_FS */ #endif /* CONFIG_PPC_BOOK3S_64 */ -- 2.20.1