From patchwork Wed Apr 27 18:46:59 2022 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Kalesh Singh X-Patchwork-Id: 12829340 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 A34BDC433EF for ; Wed, 27 Apr 2022 18:51:18 +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:Cc:From:Subject:References:Mime-Version :Message-Id:In-Reply-To:Date:Reply-To:To:Content-ID:Content-Description: Resent-Date:Resent-From:Resent-Sender:Resent-To:Resent-Cc:Resent-Message-ID: List-Owner; bh=neCG5OY6RuaJq6309/DLuQGNIO7NxzVFKYTTwp7ixKQ=; b=paUEJ3OrcxNy1N JElEiRz+gSxeB3Uq1R7XGDqtyUL9TOWxa2m5BQkTzFteqGMPOC1llSRxl3WfdGZfrH5RgZDGOcxUG DFOBnKjYDJrIZ0VatxDFELNZbGXLuBDf9kk6xRGOTSJrEWYD85oENuHA7mFLfO3wkBybM/kHsE6Dd 1bITFqstkbtHeHajLfCuMXBqRR5w5oXzzi8c9ih8Pn1kj3K7QH6WpSfMjRtJjz+P2EPsv0dCs18IA Z4fOGDwWfdq1vkGd70PQsmpIDxrs99Ka1zqFAbxyq7vI2Aphk4ohIxpQ/YIgK+qxvlOI5gn+4ZDlK Etf9eFTdpGXi4aerPk+Q==; Received: from localhost ([::1] helo=bombadil.infradead.org) by bombadil.infradead.org with esmtp (Exim 4.94.2 #2 (Red Hat Linux)) id 1njmjs-0038LV-Lu; Wed, 27 Apr 2022 18:50:12 +0000 Received: from mail-yb1-xb49.google.com ([2607:f8b0:4864:20::b49]) by bombadil.infradead.org with esmtps (Exim 4.94.2 #2 (Red Hat Linux)) id 1njmje-0038Fi-DW for linux-arm-kernel@lists.infradead.org; Wed, 27 Apr 2022 18:50:00 +0000 Received: by mail-yb1-xb49.google.com with SMTP id o64-20020a257343000000b006483069a28aso2481084ybc.3 for ; Wed, 27 Apr 2022 11:49:57 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=20210112; h=date:in-reply-to:message-id:mime-version:references:subject:from:cc; bh=sH7KcOi4dXvStzx7CoNtlhTdwpi2Lae2JmM1YdLUdWE=; b=s9/HxvVMTsEgNz9+Lpu4CzAWwPaS8lKm6a14jcjfp9W5ZNjguoXROmEzBjK10UxFy9 tLEgGfH+WEKIS2zMNNyasptxhMVOoByB2BwCLxZXYIcnpyaSqMGJc4z2t4MVPCY4w9JL ZCKxdO7JZDeysjHv5IZmfjpMDcF5T4q09OxiwAHWwqtWyNANAY2LhSL/xuZCMBG/R0rF P6+3VRPc6JTUKrdTqz6Fagjbtr4+B+8XIbTIx4b/IWL52UZADMw4qnBRMJZwnZuSRQ0T iJgsxqBWuys5Dq5QelGIak2Kksgt0qMdgtaUGmLfBG979WMjpi55K1Nwa3phalPPNiSM GqcQ== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20210112; h=x-gm-message-state:date:in-reply-to:message-id:mime-version :references:subject:from:cc; bh=sH7KcOi4dXvStzx7CoNtlhTdwpi2Lae2JmM1YdLUdWE=; b=DgHD5jQkg82aQ4hnfdkloSSwQBd07JnuhuRu4ao/3NLCEkAkzjO07gDSTddUdap91/ 6ZaGKRpaZ5yDDBfIr3/3eZg40nQxeKJGtcMvvA2ia5n04ChfUgRU9JRakNzYoLRJ4lLb 4//IQ1xw2V48YctdUj7+LTWDR96Pw/Q4lDSSGpWaNGu9VJcwjTADvi0ejz8ZzUf+WoIC Lkqsm8E41B4+aWAB+/CUhhB39GE4TduB6FQ6mUD9VqERk6ACW+ont3z4Iaj7JsSdCc/F UaPQ3ygW85+Qy2xicBncf57lwLiKnTOteOZKHo5jY2Dq8YKaYt/hY1t4kkFHNSc6hJkH 7KYQ== X-Gm-Message-State: AOAM532FCEDW/0V5yAph6qct4RmMsBKDOPS2mJ+0TIjzdC1dBy0w/M1q 7KxdMagHwrh7RscTrfYYQmiGVuYg1ktO/uRjUA== X-Google-Smtp-Source: ABdhPJz/i5n/h2ljJRmeANt+JijtHbBq8jAiICNxa0BDMlFN2U3BSZscHogO0CRADOxNer7VUjlhyrYX3FQL4ZZXmA== X-Received: from kaleshsingh.mtv.corp.google.com ([2620:15c:211:200:16ec:6da0:8cc5:5f24]) (user=kaleshsingh job=sendgmr) by 2002:a25:3006:0:b0:646:ddbc:dba6 with SMTP id w6-20020a253006000000b00646ddbcdba6mr21665015ybw.113.1651085396777; Wed, 27 Apr 2022 11:49:56 -0700 (PDT) Date: Wed, 27 Apr 2022 11:46:59 -0700 In-Reply-To: <20220427184716.1949239-1-kaleshsingh@google.com> Message-Id: <20220427184716.1949239-5-kaleshsingh@google.com> Mime-Version: 1.0 References: <20220427184716.1949239-1-kaleshsingh@google.com> X-Mailer: git-send-email 2.36.0.rc2.479.g8af0fa9b8e-goog Subject: [PATCH 4/4] KVM: arm64: Unwind and dump nVHE hypervisor stacktrace From: Kalesh Singh Cc: mark.rutland@arm.com, will@kernel.org, maz@kernel.org, qperret@google.com, tabba@google.com, surenb@google.com, kernel-team@android.com, Kalesh Singh , James Morse , Alexandru Elisei , Suzuki K Poulose , Catalin Marinas , Mark Brown , Masami Hiramatsu , Peter Collingbourne , Alexei Starovoitov , "Madhavan T. Venkataraman" , Andrew Jones , Marco Elver , Kefeng Wang , Zenghui Yu , Keir Fraser , linux-arm-kernel@lists.infradead.org, kvmarm@lists.cs.columbia.edu, linux-kernel@vger.kernel.org X-CRM114-Version: 20100106-BlameMichelson ( TRE 0.8.0 (BSD) ) MR-646709E3 X-CRM114-CacheID: sfid-20220427_114958_509784_98A8999C X-CRM114-Status: GOOD ( 21.65 ) 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 On hyp_panic(), the hypervisor dumps the addresses for its stacktrace entries to a page shared with the host. The host then symbolizes and prints the hyp stacktrace before panicking itself. Example stacktrace: [ 122.051187] kvm [380]: Invalid host exception to nVHE hyp! [ 122.052467] kvm [380]: nVHE HYP call trace: [ 122.052814] kvm [380]: [] __kvm_nvhe___pkvm_vcpu_init_traps+0x1f0/0x1f0 [ 122.053865] kvm [380]: [] __kvm_nvhe_hyp_panic+0x130/0x1c0 [ 122.054367] kvm [380]: [] __kvm_nvhe___kvm_vcpu_run+0x10/0x10 [ 122.054878] kvm [380]: [] __kvm_nvhe_handle___kvm_vcpu_run+0x30/0x50 [ 122.055412] kvm [380]: [] __kvm_nvhe_handle_trap+0xbc/0x160 [ 122.055911] kvm [380]: [] __kvm_nvhe___host_exit+0x64/0x64 [ 122.056417] kvm [380]: ---- end of nVHE HYP call trace ---- Signed-off-by: Kalesh Singh Reviewed-by: Mark Brown --- arch/arm64/include/asm/stacktrace.h | 42 ++++++++++++++-- arch/arm64/kernel/stacktrace.c | 75 +++++++++++++++++++++++++++++ arch/arm64/kvm/handle_exit.c | 4 ++ arch/arm64/kvm/hyp/nvhe/switch.c | 4 ++ 4 files changed, 121 insertions(+), 4 deletions(-) diff --git a/arch/arm64/include/asm/stacktrace.h b/arch/arm64/include/asm/stacktrace.h index f5af9a94c5a6..3063912107b0 100644 --- a/arch/arm64/include/asm/stacktrace.h +++ b/arch/arm64/include/asm/stacktrace.h @@ -5,6 +5,7 @@ #ifndef __ASM_STACKTRACE_H #define __ASM_STACKTRACE_H +#include #include #include #include @@ -19,10 +20,12 @@ enum stack_type { #ifndef __KVM_NVHE_HYPERVISOR__ STACK_TYPE_TASK, STACK_TYPE_IRQ, - STACK_TYPE_OVERFLOW, STACK_TYPE_SDEI_NORMAL, STACK_TYPE_SDEI_CRITICAL, +#else /* __KVM_NVHE_HYPERVISOR__ */ + STACK_TYPE_HYP, #endif /* !__KVM_NVHE_HYPERVISOR__ */ + STACK_TYPE_OVERFLOW, STACK_TYPE_UNKNOWN, __NR_STACK_TYPES }; @@ -55,6 +58,9 @@ static inline bool on_stack(unsigned long sp, unsigned long size, extern void dump_backtrace(struct pt_regs *regs, struct task_struct *tsk, const char *loglvl); +extern void hyp_dump_backtrace(unsigned long hyp_offset); + +DECLARE_PER_CPU(unsigned long, kvm_arm_hyp_stacktrace_page); DECLARE_PER_CPU(unsigned long *, irq_stack_ptr); static inline bool on_irq_stack(unsigned long sp, unsigned long size, @@ -91,8 +97,32 @@ static inline bool on_overflow_stack(unsigned long sp, unsigned long size, static inline bool on_overflow_stack(unsigned long sp, unsigned long size, struct stack_info *info) { return false; } #endif -#endif /* !__KVM_NVHE_HYPERVISOR__ */ +#else /* __KVM_NVHE_HYPERVISOR__ */ + +extern void hyp_save_backtrace(void); + +DECLARE_PER_CPU(unsigned long [PAGE_SIZE/sizeof(long)], overflow_stack); +DECLARE_PER_CPU(struct kvm_nvhe_init_params, kvm_init_params); + +static inline bool on_overflow_stack(unsigned long sp, unsigned long size, + struct stack_info *info) +{ + unsigned long low = (unsigned long)this_cpu_ptr(overflow_stack); + unsigned long high = low + PAGE_SIZE; + + return on_stack(sp, size, low, high, STACK_TYPE_OVERFLOW, info); +} + +static inline bool on_hyp_stack(unsigned long sp, unsigned long size, + struct stack_info *info) +{ + struct kvm_nvhe_init_params *params = this_cpu_ptr(&kvm_init_params); + unsigned long high = params->stack_hyp_va; + unsigned long low = high - PAGE_SIZE; + return on_stack(sp, size, low, high, STACK_TYPE_HYP, info); +} +#endif /* !__KVM_NVHE_HYPERVISOR__ */ /* * We can only safely access per-cpu stacks from current in a non-preemptible @@ -105,6 +135,9 @@ static inline bool on_accessible_stack(const struct task_struct *tsk, if (info) info->type = STACK_TYPE_UNKNOWN; + if (on_overflow_stack(sp, size, info)) + return true; + #ifndef __KVM_NVHE_HYPERVISOR__ if (on_task_stack(tsk, sp, size, info)) return true; @@ -112,10 +145,11 @@ static inline bool on_accessible_stack(const struct task_struct *tsk, return false; if (on_irq_stack(sp, size, info)) return true; - if (on_overflow_stack(sp, size, info)) - return true; if (on_sdei_stack(sp, size, info)) return true; +#else /* __KVM_NVHE_HYPERVISOR__ */ + if (on_hyp_stack(sp, size, info)) + return true; #endif /* !__KVM_NVHE_HYPERVISOR__ */ return false; diff --git a/arch/arm64/kernel/stacktrace.c b/arch/arm64/kernel/stacktrace.c index f346b4c66f1c..c81dea9760ac 100644 --- a/arch/arm64/kernel/stacktrace.c +++ b/arch/arm64/kernel/stacktrace.c @@ -104,6 +104,7 @@ static int notrace __unwind_next(struct task_struct *tsk, * * TASK -> IRQ -> OVERFLOW -> SDEI_NORMAL * TASK -> SDEI_NORMAL -> SDEI_CRITICAL -> OVERFLOW + * HYP -> OVERFLOW * * ... but the nesting itself is strict. Once we transition from one * stack to another, it's never valid to unwind back to that first @@ -242,7 +243,81 @@ noinline notrace void arch_stack_walk(stack_trace_consume_fn consume_entry, unwind(task, &state, consume_entry, cookie); } + +/** + * Symbolizes and dumps the hypervisor backtrace from the shared + * stacktrace page. + */ +noinline notrace void hyp_dump_backtrace(unsigned long hyp_offset) +{ + unsigned long *stacktrace_pos = + (unsigned long *)*this_cpu_ptr(&kvm_arm_hyp_stacktrace_page); + unsigned long va_mask = GENMASK_ULL(vabits_actual - 1, 0); + unsigned long pc = *stacktrace_pos++; + + kvm_err("nVHE HYP call trace:\n"); + + while (pc) { + pc &= va_mask; /* Mask tags */ + pc += hyp_offset; /* Convert to kern addr */ + kvm_err("[<%016lx>] %pB\n", pc, (void *)pc); + pc = *stacktrace_pos++; + } + + kvm_err("---- end of nVHE HYP call trace ----\n"); +} #else /* __KVM_NVHE_HYPERVISOR__ */ DEFINE_PER_CPU(unsigned long [PAGE_SIZE/sizeof(long)], overflow_stack) __aligned(16); + +static int notrace unwind_next(struct task_struct *tsk, + struct unwind_state *state) +{ + struct stack_info info; + + return __unwind_next(tsk, state, &info); +} + +/** + * Saves a hypervisor stacktrace entry (address) to the shared stacktrace page. + */ +static bool hyp_save_backtrace_entry(void *arg, unsigned long where) +{ + struct kvm_nvhe_init_params *params = this_cpu_ptr(&kvm_init_params); + unsigned long **stacktrace_pos = (unsigned long **)arg; + unsigned long stacktrace_start, stacktrace_end; + + stacktrace_start = (unsigned long)params->stacktrace_hyp_va; + stacktrace_end = stacktrace_start + PAGE_SIZE - (2 * sizeof(long)); + + if ((unsigned long) *stacktrace_pos > stacktrace_end) + return false; + + /* Save the entry to the current pos in stacktrace page */ + **stacktrace_pos = where; + + /* A zero entry delimits the end of the stacktrace. */ + *(*stacktrace_pos + 1) = 0UL; + + /* Increment the current pos */ + ++*stacktrace_pos; + + return true; +} + +/** + * Saves hypervisor stacktrace to the shared stacktrace page. + */ +noinline notrace void hyp_save_backtrace(void) +{ + struct kvm_nvhe_init_params *params = this_cpu_ptr(&kvm_init_params); + void *stacktrace_start = (void *)params->stacktrace_hyp_va; + struct unwind_state state; + + unwind_init(&state, (unsigned long)__builtin_frame_address(0), + _THIS_IP_); + + unwind(NULL, &state, hyp_save_backtrace_entry, &stacktrace_start); +} + #endif /* !__KVM_NVHE_HYPERVISOR__ */ diff --git a/arch/arm64/kvm/handle_exit.c b/arch/arm64/kvm/handle_exit.c index a377b871bf58..ee5adc9bdb8c 100644 --- a/arch/arm64/kvm/handle_exit.c +++ b/arch/arm64/kvm/handle_exit.c @@ -17,6 +17,7 @@ #include #include #include +#include #include #include @@ -323,6 +324,9 @@ void __noreturn __cold nvhe_hyp_panic_handler(u64 esr, u64 spsr, (void *)panic_addr); } + /* Dump the hypervisor stacktrace */ + hyp_dump_backtrace(hyp_offset); + /* * Hyp has panicked and we're going to handle that by panicking the * kernel. The kernel offset will be revealed in the panic so we're diff --git a/arch/arm64/kvm/hyp/nvhe/switch.c b/arch/arm64/kvm/hyp/nvhe/switch.c index 978f1b94fb25..95d810e86c7d 100644 --- a/arch/arm64/kvm/hyp/nvhe/switch.c +++ b/arch/arm64/kvm/hyp/nvhe/switch.c @@ -25,6 +25,7 @@ #include #include #include +#include #include #include @@ -395,6 +396,9 @@ asmlinkage void __noreturn hyp_panic(void) __sysreg_restore_state_nvhe(host_ctxt); } + /* Save the hypervisor stacktrace */ + hyp_save_backtrace(); + __hyp_do_panic(host_ctxt, spsr, elr, par); unreachable(); }