* [PATCH v0 1/6] x86/hyperv: Rename guest shutdown functions
2025-09-04 2:10 [PATCH v0 0/6] Hyper-V: Implement hypervisor core collection Mukesh Rathor
@ 2025-09-04 2:10 ` Mukesh Rathor
2025-09-04 17:22 ` Anirudh Rayabharam
2025-09-04 2:10 ` [PATCH v0 2/6] Hyper-V: Add two new hypercall numbers to guest ABI public header Mukesh Rathor
` (4 subsequent siblings)
5 siblings, 1 reply; 14+ messages in thread
From: Mukesh Rathor @ 2025-09-04 2:10 UTC (permalink / raw)
To: linux-hyperv, linux-kernel, linux-arch
Cc: kys, haiyangz, wei.liu, decui, tglx, mingo, bp, dave.hansen, x86,
hpa, arnd
This commit renames hv_machine_crash_shutdown to more appropriate
hv_guest_crash_shutdown and makes it applicable to guests only. This
in preparation for the subsequent hypervisor root/dom0 crash support
patches.
Signed-off-by: Mukesh Rathor <mrathor@linux.microsoft.com>
---
arch/x86/kernel/cpu/mshyperv.c | 5 +++--
1 file changed, 3 insertions(+), 2 deletions(-)
diff --git a/arch/x86/kernel/cpu/mshyperv.c b/arch/x86/kernel/cpu/mshyperv.c
index b8caf78c4336..f5d76d854d39 100644
--- a/arch/x86/kernel/cpu/mshyperv.c
+++ b/arch/x86/kernel/cpu/mshyperv.c
@@ -237,7 +237,7 @@ static void hv_machine_shutdown(void)
#endif /* CONFIG_KEXEC_CORE */
#ifdef CONFIG_CRASH_DUMP
-static void hv_machine_crash_shutdown(struct pt_regs *regs)
+static void hv_guest_crash_shutdown(struct pt_regs *regs)
{
if (hv_crash_handler)
hv_crash_handler(regs);
@@ -824,7 +824,8 @@ static void __init ms_hyperv_init_platform(void)
machine_ops.shutdown = hv_machine_shutdown;
#endif
#if defined(CONFIG_CRASH_DUMP)
- machine_ops.crash_shutdown = hv_machine_crash_shutdown;
+ if (!hv_root_partition)
+ machine_ops.crash_shutdown = hv_guest_crash_shutdown;
#endif
#endif
/*
--
2.36.1.vfs.0.0
^ permalink raw reply related [flat|nested] 14+ messages in thread* Re: [PATCH v0 1/6] x86/hyperv: Rename guest shutdown functions
2025-09-04 2:10 ` [PATCH v0 1/6] x86/hyperv: Rename guest shutdown functions Mukesh Rathor
@ 2025-09-04 17:22 ` Anirudh Rayabharam
0 siblings, 0 replies; 14+ messages in thread
From: Anirudh Rayabharam @ 2025-09-04 17:22 UTC (permalink / raw)
To: Mukesh Rathor
Cc: linux-hyperv, linux-kernel, linux-arch, kys, haiyangz, wei.liu,
decui, tglx, mingo, bp, dave.hansen, x86, hpa, arnd
On Wed, Sep 03, 2025 at 07:10:12PM -0700, Mukesh Rathor wrote:
> This commit renames hv_machine_crash_shutdown to more appropriate
https://www.kernel.org/doc/html/latest/process/submitting-patches.html
Please describe your changes in imperative mood. For e.g.:
“make xyzzy do frotz” instead of “[This patch] makes xyzzy do frotz”
Please fix this for all the patches in this series.
Thanks,
Anirudh.
> hv_guest_crash_shutdown and makes it applicable to guests only. This
> in preparation for the subsequent hypervisor root/dom0 crash support
> patches.
>
> Signed-off-by: Mukesh Rathor <mrathor@linux.microsoft.com>
> ---
> arch/x86/kernel/cpu/mshyperv.c | 5 +++--
> 1 file changed, 3 insertions(+), 2 deletions(-)
>
> diff --git a/arch/x86/kernel/cpu/mshyperv.c b/arch/x86/kernel/cpu/mshyperv.c
> index b8caf78c4336..f5d76d854d39 100644
> --- a/arch/x86/kernel/cpu/mshyperv.c
> +++ b/arch/x86/kernel/cpu/mshyperv.c
> @@ -237,7 +237,7 @@ static void hv_machine_shutdown(void)
> #endif /* CONFIG_KEXEC_CORE */
>
> #ifdef CONFIG_CRASH_DUMP
> -static void hv_machine_crash_shutdown(struct pt_regs *regs)
> +static void hv_guest_crash_shutdown(struct pt_regs *regs)
> {
> if (hv_crash_handler)
> hv_crash_handler(regs);
> @@ -824,7 +824,8 @@ static void __init ms_hyperv_init_platform(void)
> machine_ops.shutdown = hv_machine_shutdown;
> #endif
> #if defined(CONFIG_CRASH_DUMP)
> - machine_ops.crash_shutdown = hv_machine_crash_shutdown;
> + if (!hv_root_partition)
> + machine_ops.crash_shutdown = hv_guest_crash_shutdown;
> #endif
> #endif
> /*
> --
> 2.36.1.vfs.0.0
>
^ permalink raw reply [flat|nested] 14+ messages in thread
* [PATCH v0 2/6] Hyper-V: Add two new hypercall numbers to guest ABI public header
2025-09-04 2:10 [PATCH v0 0/6] Hyper-V: Implement hypervisor core collection Mukesh Rathor
2025-09-04 2:10 ` [PATCH v0 1/6] x86/hyperv: Rename guest shutdown functions Mukesh Rathor
@ 2025-09-04 2:10 ` Mukesh Rathor
2025-09-04 2:10 ` [PATCH v0 3/6] Hyper-V: Add definitions for hypervisor crash dump support Mukesh Rathor
` (3 subsequent siblings)
5 siblings, 0 replies; 14+ messages in thread
From: Mukesh Rathor @ 2025-09-04 2:10 UTC (permalink / raw)
To: linux-hyperv, linux-kernel, linux-arch
Cc: kys, haiyangz, wei.liu, decui, tglx, mingo, bp, dave.hansen, x86,
hpa, arnd
In preparation for upcoming crashdump patches, this commit copies two new
hypercall numbers to the guest ABI header published by Hyper-V. One to
notify hypervisor of an event that occurs in the root partition, other
to ask hypervisor to disable the hypervisor.
Signed-off-by: Mukesh Rathor <mrathor@linux.microsoft.com>
---
include/hyperv/hvgdk_mini.h | 2 ++
1 file changed, 2 insertions(+)
diff --git a/include/hyperv/hvgdk_mini.h b/include/hyperv/hvgdk_mini.h
index 79b7324e4ef5..b288906c79d0 100644
--- a/include/hyperv/hvgdk_mini.h
+++ b/include/hyperv/hvgdk_mini.h
@@ -469,6 +469,7 @@ union hv_vp_assist_msr_contents { /* HV_REGISTER_VP_ASSIST_PAGE */
#define HVCALL_MAP_DEVICE_INTERRUPT 0x007c
#define HVCALL_UNMAP_DEVICE_INTERRUPT 0x007d
#define HVCALL_RETARGET_INTERRUPT 0x007e
+#define HVCALL_NOTIFY_PARTITION_EVENT 0x0087
#define HVCALL_NOTIFY_PORT_RING_EMPTY 0x008b
#define HVCALL_REGISTER_INTERCEPT_RESULT 0x0091
#define HVCALL_ASSERT_VIRTUAL_INTERRUPT 0x0094
@@ -492,6 +493,7 @@ union hv_vp_assist_msr_contents { /* HV_REGISTER_VP_ASSIST_PAGE */
#define HVCALL_GET_VP_CPUID_VALUES 0x00f4
#define HVCALL_MMIO_READ 0x0106
#define HVCALL_MMIO_WRITE 0x0107
+#define HVCALL_DISABLE_HYP_EX 0x010f
/* HV_HYPERCALL_INPUT */
#define HV_HYPERCALL_RESULT_MASK GENMASK_ULL(15, 0)
--
2.36.1.vfs.0.0
^ permalink raw reply related [flat|nested] 14+ messages in thread* [PATCH v0 3/6] Hyper-V: Add definitions for hypervisor crash dump support
2025-09-04 2:10 [PATCH v0 0/6] Hyper-V: Implement hypervisor core collection Mukesh Rathor
2025-09-04 2:10 ` [PATCH v0 1/6] x86/hyperv: Rename guest shutdown functions Mukesh Rathor
2025-09-04 2:10 ` [PATCH v0 2/6] Hyper-V: Add two new hypercall numbers to guest ABI public header Mukesh Rathor
@ 2025-09-04 2:10 ` Mukesh Rathor
2025-09-04 2:10 ` [PATCH v0 4/6] x86/hyperv: Add trampoline asm code to transition from hypervisor Mukesh Rathor
` (2 subsequent siblings)
5 siblings, 0 replies; 14+ messages in thread
From: Mukesh Rathor @ 2025-09-04 2:10 UTC (permalink / raw)
To: linux-hyperv, linux-kernel, linux-arch
Cc: kys, haiyangz, wei.liu, decui, tglx, mingo, bp, dave.hansen, x86,
hpa, arnd
This commit adds data structures for hypervisor crash dump support
to the hypervisor host ABI public header file. Details of their usages
can be found in subsequent commits.
Signed-off-by: Mukesh Rathor <mrathor@linux.microsoft.com>
---
include/hyperv/hvhdk_mini.h | 55 +++++++++++++++++++++++++++++++++++++
1 file changed, 55 insertions(+)
diff --git a/include/hyperv/hvhdk_mini.h b/include/hyperv/hvhdk_mini.h
index 858f6a3925b3..ad9a8048fb4e 100644
--- a/include/hyperv/hvhdk_mini.h
+++ b/include/hyperv/hvhdk_mini.h
@@ -116,6 +116,17 @@ enum hv_system_property {
/* Add more values when needed */
HV_SYSTEM_PROPERTY_SCHEDULER_TYPE = 15,
HV_DYNAMIC_PROCESSOR_FEATURE_PROPERTY = 21,
+ HV_SYSTEM_PROPERTY_CRASHDUMPAREA = 47,
+};
+
+#define HV_PFN_RANGE_PGBITS 24 /* HV_SPA_PAGE_RANGE_ADDITIONAL_PAGES_BITS */
+union hv_pfn_range { /* HV_SPA_PAGE_RANGE */
+ u64 as_uint64;
+ struct {
+ /* 39:0: base pfn. 63:40: additional pages */
+ u64 base_pfn : 64 - HV_PFN_RANGE_PGBITS;
+ u64 add_pfns : HV_PFN_RANGE_PGBITS;
+ } __packed;
};
enum hv_dynamic_processor_feature_property {
@@ -142,6 +153,8 @@ struct hv_output_get_system_property {
#if IS_ENABLED(CONFIG_X86)
u64 hv_processor_feature_value;
#endif
+ union hv_pfn_range hv_cda_info; /* CrashdumpAreaAddress */
+ u64 hv_tramp_pa; /* CrashdumpTrampolineAddress */
};
} __packed;
@@ -234,6 +247,48 @@ union hv_gpa_page_access_state {
u8 as_uint8;
} __packed;
+enum hv_crashdump_action {
+ HV_CRASHDUMP_NONE = 0,
+ HV_CRASHDUMP_SUSPEND_ALL_VPS,
+ HV_CRASHDUMP_PREPARE_FOR_STATE_SAVE,
+ HV_CRASHDUMP_STATE_SAVED,
+ HV_CRASHDUMP_ENTRY,
+};
+
+struct hv_partition_event_root_crashdump_input {
+ u32 crashdump_action; /* enum hv_crashdump_action */
+} __packed;
+
+struct hv_input_disable_hyp_ex { /* HV_X64_INPUT_DISABLE_HYPERVISOR_EX */
+ u64 rip;
+ u64 arg;
+} __packed;
+
+struct hv_crashdump_area { /* HV_CRASHDUMP_AREA */
+ u32 version;
+ union {
+ u32 flags_as_uint32;
+ struct {
+ u32 cda_valid : 1;
+ u32 cda_unused : 31;
+ } __packed;
+ };
+ /* more unused fields */
+} __packed;
+
+union hv_partition_event_input {
+ struct hv_partition_event_root_crashdump_input crashdump_input;
+};
+
+enum hv_partition_event {
+ HV_PARTITION_EVENT_ROOT_CRASHDUMP = 2,
+};
+
+struct hv_input_notify_partition_event {
+ u32 event; /* enum hv_partition_event */
+ union hv_partition_event_input input;
+} __packed;
+
struct hv_lp_startup_status {
u64 hv_status;
u64 substatus1;
--
2.36.1.vfs.0.0
^ permalink raw reply related [flat|nested] 14+ messages in thread* [PATCH v0 4/6] x86/hyperv: Add trampoline asm code to transition from hypervisor
2025-09-04 2:10 [PATCH v0 0/6] Hyper-V: Implement hypervisor core collection Mukesh Rathor
` (2 preceding siblings ...)
2025-09-04 2:10 ` [PATCH v0 3/6] Hyper-V: Add definitions for hypervisor crash dump support Mukesh Rathor
@ 2025-09-04 2:10 ` Mukesh Rathor
2025-09-04 2:10 ` [PATCH v0 5/6] x86/hyperv: Implement hypervisor ram collection into vmcore Mukesh Rathor
2025-09-04 2:10 ` [PATCH v0 6/6] Hyper-V: Enable build of hypervisor crash collection files Mukesh Rathor
5 siblings, 0 replies; 14+ messages in thread
From: Mukesh Rathor @ 2025-09-04 2:10 UTC (permalink / raw)
To: linux-hyperv, linux-kernel, linux-arch
Cc: kys, haiyangz, wei.liu, decui, tglx, mingo, bp, dave.hansen, x86,
hpa, arnd
This commit introduces a small asm stub to transition from the hypervisor
to linux upon devirtualization. At a high level, during panic of either
the hypervisor or the dom0 (aka root), the nmi handler asks hypervisor
to devirtualize. As part of that, the arguments include the entry point
to return to linux. This asm stub implements the entry point.
The stub is entered in protected mode, uses temporary gdt and page table
to enable long mode and get to kernel entry point to restore full kernel
context to resume execution to kexec.
Signed-off-by: Mukesh Rathor <mrathor@linux.microsoft.com>
---
arch/x86/hyperv/hv_trampoline.S | 99 +++++++++++++++++++++++++++++++++
1 file changed, 99 insertions(+)
create mode 100644 arch/x86/hyperv/hv_trampoline.S
diff --git a/arch/x86/hyperv/hv_trampoline.S b/arch/x86/hyperv/hv_trampoline.S
new file mode 100644
index 000000000000..307663c8891e
--- /dev/null
+++ b/arch/x86/hyperv/hv_trampoline.S
@@ -0,0 +1,99 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+/*
+ * X86 specific Hyper-V kdump/crash related code.
+ *
+ * Copyright (C) 2025, Microsoft, Inc.
+ *
+ */
+#include <linux/linkage.h>
+#include <asm/alternative.h>
+#include <asm/msr.h>
+#include <asm/processor-flags.h>
+#include <asm/nospec-branch.h>
+
+/*
+ * void noreturn hv_crash_asm32(arg1)
+ * arg1 == edi == 32bit PA of struct hv_crash_trdata
+ *
+ * The hypervisor jumps here upon devirtualization in protected mode. This
+ * code gets copied to a page in the low 4G ie, 32bit space so it can run
+ * in the protected mode. Hence we cannot use any compile/link time offsets or
+ * addresses. It restores long mode via temporary gdt and page tables and
+ * eventually jumps to kernel code entry at HV_CRASHDATA_OFFS_C_entry.
+ *
+ * PreCondition (ie, Hypervisor call back ABI):
+ * o CR0 is set to 0x0021: PE(prot mode) and NE are set, paging is disabled
+ * o CR4 is set to 0x0
+ * o IA32_EFER is set to 0x901 (SCE and NXE are set)
+ * o EDI is set to the Arg passed to HVCALL_DISABLE_HYP_EX.
+ * o CS, DS, ES, FS, GS are all initialized with a base of 0 and limit 0xFFFF
+ * o IDTR, TR and GDTR are initialized with a base of 0 and limit of 0xFFFF
+ * o LDTR is initialized as invalid (limit of 0)
+ * o MSR PAT is power on default.
+ * o Other state/registers are cleared. All TLBs flushed.
+ *
+ * See Intel SDM 10.8.5
+ */
+
+#define HV_CRASHDATA_OFFS_TRAMPCR3 0x0 /* 0 */
+#define HV_CRASHDATA_OFFS_KERNCR3 0x8 /* 8 */
+#define HV_CRASHDATA_OFFS_GDTRLIMIT 0x12 /* 18 */
+#define HV_CRASHDATA_OFFS_CS_JMPTGT 0x28 /* 40 */
+#define HV_CRASHDATA_OFFS_C_entry 0x30 /* 48 */
+#define HV_CRASHDATA_TRAMPOLINE_CS 0x8
+
+ .text
+ .code32
+
+SYM_CODE_START(hv_crash_asm32)
+ movl $X86_CR4_PAE, %ecx
+ movl %ecx, %cr4
+
+ movl %edi, %ebx
+ add $HV_CRASHDATA_OFFS_TRAMPCR3, %ebx
+ movl %cs:(%ebx), %eax
+ movl %eax, %cr3
+
+ # Setup EFER for long mode now.
+ movl $MSR_EFER, %ecx
+ rdmsr
+ btsl $_EFER_LME, %eax
+ wrmsr
+
+ # Turn paging on using the temp 32bit trampoline page table.
+ movl %cr0, %eax
+ orl $(X86_CR0_PG), %eax
+ movl %eax, %cr0
+
+ /* since kernel cr3 could be above 4G, we need to be in the long mode
+ * before we can load 64bits of the kernel cr3. We use a temp gdt for
+ * that with CS.L=1 and CS.D=0 */
+ mov %edi, %eax
+ add $HV_CRASHDATA_OFFS_GDTRLIMIT, %eax
+ lgdtl %cs:(%eax)
+
+ /* not done yet, restore CS now to switch to CS.L=1 */
+ mov %edi, %eax
+ add $HV_CRASHDATA_OFFS_CS_JMPTGT, %eax
+ ljmp %cs:*(%eax)
+
+ /* we now run in full 64bit IA32-e long mode, CS.L=1 and CS.D=0 */
+ .code64
+ .balign 8
+SYM_INNER_LABEL(hv_crash_asm64_lbl, SYM_L_GLOBAL)
+ /* restore kernel page tables so we can jump to kernel code */
+ mov %edi, %eax
+ add $HV_CRASHDATA_OFFS_KERNCR3, %eax
+ movq %cs:(%eax), %rbx
+ movq %rbx, %cr3
+
+ mov %edi, %eax
+ add $HV_CRASHDATA_OFFS_C_entry, %eax
+ movq %cs:(%eax), %rbx
+ ANNOTATE_RETPOLINE_SAFE
+ jmp *%rbx
+
+ int $3
+
+SYM_INNER_LABEL(hv_crash_asm32_end, SYM_L_GLOBAL)
+SYM_CODE_END(hv_crash_asm32)
--
2.36.1.vfs.0.0
^ permalink raw reply related [flat|nested] 14+ messages in thread* [PATCH v0 5/6] x86/hyperv: Implement hypervisor ram collection into vmcore
2025-09-04 2:10 [PATCH v0 0/6] Hyper-V: Implement hypervisor core collection Mukesh Rathor
` (3 preceding siblings ...)
2025-09-04 2:10 ` [PATCH v0 4/6] x86/hyperv: Add trampoline asm code to transition from hypervisor Mukesh Rathor
@ 2025-09-04 2:10 ` Mukesh Rathor
2025-09-04 22:37 ` Stanislav Kinsburskii
2025-09-08 16:21 ` Stanislav Kinsburskii
2025-09-04 2:10 ` [PATCH v0 6/6] Hyper-V: Enable build of hypervisor crash collection files Mukesh Rathor
5 siblings, 2 replies; 14+ messages in thread
From: Mukesh Rathor @ 2025-09-04 2:10 UTC (permalink / raw)
To: linux-hyperv, linux-kernel, linux-arch
Cc: kys, haiyangz, wei.liu, decui, tglx, mingo, bp, dave.hansen, x86,
hpa, arnd
This commit introduces a new file to enable collection of hypervisor ram
into the vmcore collected by linux. By default, the hypervisor ram is locked,
ie, protected via hw page table. Hyper-V implements a disable hypercall which
essentially devirtualizes the system on the fly. This mechanism makes the
hypervisor ram accessible to linux without any extra work because it is
already mapped into linux address space. Details of the implementation
are available in the file prologue.
Signed-off-by: Mukesh Rathor <mrathor@linux.microsoft.com>
---
arch/x86/hyperv/hv_crash.c | 618 +++++++++++++++++++++++++++++++++++++
1 file changed, 618 insertions(+)
create mode 100644 arch/x86/hyperv/hv_crash.c
diff --git a/arch/x86/hyperv/hv_crash.c b/arch/x86/hyperv/hv_crash.c
new file mode 100644
index 000000000000..50c54d39f0e2
--- /dev/null
+++ b/arch/x86/hyperv/hv_crash.c
@@ -0,0 +1,618 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * X86 specific Hyper-V kdump/crash support module
+ *
+ * Copyright (C) 2025, Microsoft, Inc.
+ *
+ * This module implements hypervisor ram collection into vmcore for both
+ * cases of the hypervisor crash and linux dom0/root crash. Hyper-V implements
+ * a devirtualization hypercall with a 32bit protected mode ABI callback. This
+ * mechanism must be used to unlock hypervisor ram. Since the hypervisor ram
+ * is already mapped in linux, it is automatically collected into linux vmcore,
+ * and can be examined by the crash command (raw ram dump) or windbg.
+ *
+ * At a high level:
+ *
+ * Hypervisor Crash:
+ * Upon crash, hypervisor goes into minimal dispatch loop, a restrictive
+ * mode with very limited hypercall and msr support. Each cpu then injects
+ * NMIs into dom0/root vcpus. A shared page is used to check by linux if
+ * hypervisor has crashed in the nmi handler. This shared page is setup
+ * in hv_root_crash_init during boot.
+ *
+ * Linux Crash:
+ * In case of linux crash, the callback hv_crash_stop_other_cpus will send
+ * NMIs to all cpus, then proceed to the crash_nmi_callback where it waits
+ * for all cpus to be in NMI.
+ *
+ * NMI Handler (upon quorum):
+ * Hyper-V requires the disable hypervisor must be done from the bsp. So
+ * the bsp nmi handler saves current context, does some fixups and makes
+ * the hypercall to disable the hypervisor, ie, devirtualize. Hypervisor
+ * at that point will suspend all vcpus (except the bsp), unlock all its
+ * ram, and return to linux at the 32bit mode entry RIP.
+ *
+ * Linux 32bit entry trampoline will then restore long mode and call C
+ * function here to restore context and continue execution to crash kexec.
+ */
+
+#include <linux/delay.h>
+#include <linux/kexec.h>
+#include <linux/crash_dump.h>
+#include <linux/panic.h>
+#include <asm/apic.h>
+#include <asm/desc.h>
+#include <asm/page.h>
+#include <asm/pgalloc.h>
+#include <asm/mshyperv.h>
+#include <asm/nmi.h>
+#include <asm/idtentry.h>
+#include <asm/reboot.h>
+#include <asm/intel_pt.h>
+
+int hv_crash_enabled;
+EXPORT_SYMBOL_GPL(hv_crash_enabled);
+
+struct hv_crash_ctxt {
+ ulong rsp;
+ ulong cr0;
+ ulong cr2;
+ ulong cr4;
+ ulong cr8;
+
+ u16 cs;
+ u16 ss;
+ u16 ds;
+ u16 es;
+ u16 fs;
+ u16 gs;
+
+ u16 gdt_fill;
+ struct desc_ptr gdtr;
+ char idt_fill[6];
+ struct desc_ptr idtr;
+
+ u64 gsbase;
+ u64 efer;
+ u64 pat;
+};
+static struct hv_crash_ctxt hv_crash_ctxt;
+
+/* Shared hypervisor page that contains crash dump area we peek into.
+ * NB: windbg looks for "hv_cda" symbol so don't change it.
+ */
+static struct hv_crashdump_area *hv_cda;
+
+static u32 trampoline_pa, devirt_cr3arg;
+static atomic_t crash_cpus_wait;
+static void *hv_crash_ptpgs[4];
+static int hv_has_crashed, lx_has_crashed;
+
+/* This cannot be inlined as it needs stack */
+static noinline __noclone void hv_crash_restore_tss(void)
+{
+ load_TR_desc();
+}
+
+/* This cannot be inlined as it needs stack */
+static noinline void hv_crash_clear_kernpt(void)
+{
+ pgd_t *pgd;
+ p4d_t *p4d;
+
+ /* Clear entry so it's not confusing to someone looking at the core */
+ pgd = pgd_offset_k(trampoline_pa);
+ p4d = p4d_offset(pgd, trampoline_pa);
+ native_p4d_clear(p4d);
+}
+
+/*
+ * This is the C entry point from the asm glue code after the devirt hypercall.
+ * We enter here in IA32-e long mode, ie, full 64bit mode running on kernel
+ * page tables with our below 4G page identity mapped, but using a temporary
+ * GDT. ds/fs/gs/es are null. ss is not usable. bp is null. stack is not
+ * available. We restore kernel GDT, and rest of the context, and continue
+ * to kexec.
+ */
+static asmlinkage void __noreturn hv_crash_c_entry(void)
+{
+ struct hv_crash_ctxt *ctxt = &hv_crash_ctxt;
+
+ /* first thing, restore kernel gdt */
+ native_load_gdt(&ctxt->gdtr);
+
+ asm volatile("movw %%ax, %%ss" : : "a"(ctxt->ss));
+ asm volatile("movq %0, %%rsp" : : "m"(ctxt->rsp));
+
+ asm volatile("movw %%ax, %%ds" : : "a"(ctxt->ds));
+ asm volatile("movw %%ax, %%es" : : "a"(ctxt->es));
+ asm volatile("movw %%ax, %%fs" : : "a"(ctxt->fs));
+ asm volatile("movw %%ax, %%gs" : : "a"(ctxt->gs));
+
+ native_wrmsrq(MSR_IA32_CR_PAT, ctxt->pat);
+ asm volatile("movq %0, %%cr0" : : "r"(ctxt->cr0));
+
+ asm volatile("movq %0, %%cr8" : : "r"(ctxt->cr8));
+ asm volatile("movq %0, %%cr4" : : "r"(ctxt->cr4));
+ asm volatile("movq %0, %%cr2" : : "r"(ctxt->cr4));
+
+ native_load_idt(&ctxt->idtr);
+ native_wrmsrq(MSR_GS_BASE, ctxt->gsbase);
+ native_wrmsrq(MSR_EFER, ctxt->efer);
+
+ /* restore the original kernel CS now via far return */
+ asm volatile("movzwq %0, %%rax\n\t"
+ "pushq %%rax\n\t"
+ "pushq $1f\n\t"
+ "lretq\n\t"
+ "1:nop\n\t" : : "m"(ctxt->cs) : "rax");
+
+ /* We are in asmlinkage without stack frame, hence make a C function
+ * call which will buy stack frame to restore the tss or clear PT entry.
+ */
+ hv_crash_restore_tss();
+ hv_crash_clear_kernpt();
+
+ /* we are now fully in devirtualized normal kernel mode */
+ __crash_kexec(NULL);
+
+ BUG();
+}
+/* Tell gcc we are using lretq long jump in the above function intentionally */
+STACK_FRAME_NON_STANDARD(hv_crash_c_entry);
+
+static void hv_mark_tss_not_busy(void)
+{
+ struct desc_struct *desc = get_current_gdt_rw();
+ tss_desc tss;
+
+ memcpy(&tss, &desc[GDT_ENTRY_TSS], sizeof(tss_desc));
+ tss.type = 0x9; /* available 64-bit TSS. 0xB is busy TSS */
+ write_gdt_entry(desc, GDT_ENTRY_TSS, &tss, DESC_TSS);
+}
+
+/* Save essential context */
+static void hv_hvcrash_ctxt_save(void)
+{
+ struct hv_crash_ctxt *ctxt = &hv_crash_ctxt;
+
+ asm volatile("movq %%rsp,%0" : "=m"(ctxt->rsp));
+
+ ctxt->cr0 = native_read_cr0();
+ ctxt->cr4 = native_read_cr4();
+
+ asm volatile("movq %%cr2, %0" : "=a"(ctxt->cr2));
+ asm volatile("movq %%cr8, %0" : "=a"(ctxt->cr8));
+
+ asm volatile("movl %%cs, %%eax" : "=a"(ctxt->cs));
+ asm volatile("movl %%ss, %%eax" : "=a"(ctxt->ss));
+ asm volatile("movl %%ds, %%eax" : "=a"(ctxt->ds));
+ asm volatile("movl %%es, %%eax" : "=a"(ctxt->es));
+ asm volatile("movl %%fs, %%eax" : "=a"(ctxt->fs));
+ asm volatile("movl %%gs, %%eax" : "=a"(ctxt->gs));
+
+ native_store_gdt(&ctxt->gdtr);
+ store_idt(&ctxt->idtr);
+
+ ctxt->gsbase = __rdmsr(MSR_GS_BASE);
+ ctxt->efer = __rdmsr(MSR_EFER);
+ ctxt->pat = __rdmsr(MSR_IA32_CR_PAT);
+}
+
+/* Add trampoline page to the kernel pagetable for transition to kernel PT */
+static void hv_crash_fixup_kernpt(void)
+{
+ pgd_t *pgd;
+ p4d_t *p4d;
+
+ pgd = pgd_offset_k(trampoline_pa);
+ p4d = p4d_offset(pgd, trampoline_pa);
+
+ /* trampoline_pa is below 4G, so no pre-existing entry to clobber */
+ p4d_populate(&init_mm, p4d, (pud_t *)hv_crash_ptpgs[1]);
+ p4d->p4d = p4d->p4d & ~(_PAGE_NX); /* enable execute */
+}
+
+/*
+ * Now that all cpus are in nmi and spinning, we notify the hyp that dom0 has
+ * crashed and will collect core. This will cause the hyp to quiesce and
+ * suspend all VPs except the bsp.
+ */
+static void hv_notify_prepare_hyp(void)
+{
+ u64 status;
+ struct hv_input_notify_partition_event *input;
+ struct hv_partition_event_root_crashdump_input *cda;
+
+ input = *this_cpu_ptr(hyperv_pcpu_input_arg);
+ cda = &input->input.crashdump_input;
+ memset(input, 0, sizeof(*input));
+ input->event = HV_PARTITION_EVENT_ROOT_CRASHDUMP;
+
+ cda->crashdump_action = HV_CRASHDUMP_ENTRY;
+ status = hv_do_hypercall(HVCALL_NOTIFY_PARTITION_EVENT, input, NULL);
+ if (!hv_result_success(status))
+ return;
+
+ cda->crashdump_action = HV_CRASHDUMP_SUSPEND_ALL_VPS;
+ hv_do_hypercall(HVCALL_NOTIFY_PARTITION_EVENT, input, NULL);
+}
+
+/*
+ * Common function for all cpus before devirtualization.
+ *
+ * Hypervisor crash: all cpus get here in nmi context.
+ * Linux crash: the panicing cpu gets here at base level, all others in nmi
+ * context. Note, panicing cpu may not be the bsp.
+ *
+ * The function is not inlined so it will show on the stack. It is named so
+ * because the crash cmd looks for certain well known function names on the
+ * stack before looking into the cpu saved note in the elf section, and
+ * that work is currently incomplete.
+ *
+ * Notes:
+ * Hypervisor crash:
+ * - the hypervisor is in a very restrictive mode at this point and any
+ * vmexit it cannot handle would result in reboot. For example, console
+ * output from here would result in synic ipi hcall, which would result
+ * in reboot. So, no mumbo jumbo, just get to kexec as quickly as possible.
+ *
+ * Devirtualization is supported from the bsp only.
+ */
+static noinline __noclone void crash_nmi_callback(struct pt_regs *regs)
+{
+ struct hv_input_disable_hyp_ex *input;
+ u64 status;
+ int msecs = 1000, ccpu = smp_processor_id();
+
+ if (ccpu == 0) {
+ /* crash_save_cpu() will be done in the kexec path */
+ cpu_emergency_stop_pt(); /* disable performance trace */
+ atomic_inc(&crash_cpus_wait);
+ } else {
+ crash_save_cpu(regs, ccpu);
+ cpu_emergency_stop_pt(); /* disable performance trace */
+ atomic_inc(&crash_cpus_wait);
+ for (;;); /* cause no vmexits */
+ }
+
+ while (atomic_read(&crash_cpus_wait) < num_online_cpus() && msecs--)
+ mdelay(1);
+
+ stop_nmi();
+ if (!hv_has_crashed)
+ hv_notify_prepare_hyp();
+
+ if (crashing_cpu == -1)
+ crashing_cpu = ccpu; /* crash cmd uses this */
+
+ hv_hvcrash_ctxt_save();
+ hv_mark_tss_not_busy();
+ hv_crash_fixup_kernpt();
+
+ input = *this_cpu_ptr(hyperv_pcpu_input_arg);
+ memset(input, 0, sizeof(*input));
+ input->rip = trampoline_pa; /* PA of hv_crash_asm32 */
+ input->arg = devirt_cr3arg; /* PA of trampoline page table L4 */
+
+ status = hv_do_hypercall(HVCALL_DISABLE_HYP_EX, input, NULL);
+ if (!hv_result_success(status)) {
+ pr_emerg("%s: %s\n", __func__, hv_result_to_string(status));
+ pr_emerg("Hyper-V: disable hyp failed. kexec not possible\n");
+ }
+
+ native_wrmsrq(HV_X64_MSR_RESET, 1); /* get hv to reboot */
+}
+
+/*
+ * generic nmi callback handler: could be called without any crash also.
+ * hv crash: hypervisor injects nmi's into all cpus
+ * lx crash: panicing cpu sends nmi to all but self via crash_stop_other_cpus
+ */
+static int hv_crash_nmi_local(unsigned int cmd, struct pt_regs *regs)
+{
+ int ccpu = smp_processor_id();
+
+ if (!hv_has_crashed && hv_cda && hv_cda->cda_valid)
+ hv_has_crashed = 1;
+
+ if (!hv_has_crashed && !lx_has_crashed)
+ return NMI_DONE; /* ignore the nmi */
+
+ if (hv_has_crashed && !hv_crash_enabled) {
+ if (ccpu == 0) {
+ pr_emerg("Hyper-V: core collect not setup. Reboot\n");
+ native_wrmsrq(HV_X64_MSR_RESET, 1); /* reboot */
+ } else
+ for (;;)
+ cpu_relax();
+ }
+
+ crash_nmi_callback(regs);
+ return NMI_DONE;
+}
+
+/*
+ * hv_crash_stop_other_cpus() == smp_ops.crash_stop_other_cpus
+ *
+ * On normal linux panic, this is called twice: first from panic and then again
+ * from native_machine_crash_shutdown.
+ *
+ * In case of mshv, 3 ways to get here:
+ * 1. hv crash (only bsp will get here):
+ * BSP : nmi callback -> DisableHv -> hv_crash_asm32 -> hv_crash_c_entry
+ * -> __crash_kexec -> native_machine_crash_shutdown
+ * -> crash_smp_send_stop -> smp_ops.crash_stop_other_cpus
+ * linux panic:
+ * 2. panic cpu x: panic() -> crash_smp_send_stop
+ * -> smp_ops.crash_stop_other_cpus
+ * 3. bsp: native_machine_crash_shutdown -> crash_smp_send_stop
+ *
+ * NB: noclone and non standard stack because of call to crash_setup_regs().
+ */
+static void __noclone hv_crash_stop_other_cpus(void)
+{
+ static int crash_stop_done;
+ struct pt_regs lregs;
+ int ccpu = smp_processor_id();
+
+ if (hv_has_crashed)
+ return; /* all cpus already in nmi handler path */
+
+ if (crash_stop_done)
+ return;
+ crash_stop_done = 1;
+
+ /* linux has crashed: hv is healthy, we can ipi safely */
+ lx_has_crashed = 1;
+ wmb(); /* nmi handlers look at lx_has_crashed */
+
+ apic->send_IPI_allbutself(NMI_VECTOR);
+
+ if (crashing_cpu == -1)
+ crashing_cpu = ccpu; /* crash cmd uses this */
+
+ /* crash_setup_regs() happens in kexec also, but for the kexec cpu which
+ * is the bsp. We could be here on non-bsp cpu, collect regs if so.
+ */
+ if (ccpu)
+ crash_setup_regs(&lregs, NULL);
+
+ crash_nmi_callback(&lregs);
+}
+STACK_FRAME_NON_STANDARD(hv_crash_stop_other_cpus);
+
+/* This GDT is accessed in IA32-e compat mode which uses 32bits addresses */
+struct hv_gdtreg_32 {
+ u16 fill;
+ u16 limit;
+ u32 address;
+} __packed;
+
+/* We need a CS with L bit to goto IA32-e long mode from 32bit compat mode */
+struct hv_crash_tramp_gdt {
+ u64 null; /* index 0, selector 0, null selector */
+ u64 cs64; /* index 1, selector 8, cs64 selector */
+} __packed;
+
+/* No stack, so jump via far ptr in memory to load the 64bit CS */
+struct hv_cs_jmptgt {
+ u32 address;
+ u16 csval;
+ u16 fill;
+} __packed;
+
+/* This trampoline data is copied onto the trampoline page after the asm code */
+struct hv_crash_tramp_data {
+ u64 tramp32_cr3;
+ u64 kernel_cr3;
+ struct hv_gdtreg_32 gdtr32;
+ struct hv_crash_tramp_gdt tramp_gdt;
+ struct hv_cs_jmptgt cs_jmptgt;
+ u64 c_entry_addr;
+} __packed;
+
+/*
+ * Setup a temporary gdt to allow the asm code to switch to the long mode.
+ * Since the asm code is relocated/copied to a below 4G page, it cannot use rip
+ * relative addressing, hence, we must use trampoline_pa here. Also save
+ * other jmp and C entry targets for same reasons.
+ *
+ * Returns: 0 on success, -1 on error
+ */
+static int hv_crash_setup_trampdata(u64 trampoline_va)
+{
+ int size, offs;
+ void *dest;
+ struct hv_crash_tramp_data *tramp;
+
+ /* These must match exactly the ones in the corresponding asm file */
+ BUILD_BUG_ON(offsetof(struct hv_crash_tramp_data, tramp32_cr3) != 0);
+ BUILD_BUG_ON(offsetof(struct hv_crash_tramp_data, kernel_cr3) != 8);
+ BUILD_BUG_ON(offsetof(struct hv_crash_tramp_data, gdtr32.limit) != 18);
+ BUILD_BUG_ON(offsetof(struct hv_crash_tramp_data,
+ cs_jmptgt.address) != 40);
+
+ /* end is beyond last byte by 1 */
+ size = &hv_crash_asm32_end - &hv_crash_asm32;
+ if (size + sizeof(struct hv_crash_tramp_data) > PAGE_SIZE) {
+ pr_err("%s: trampoline page overflow\n", __func__);
+ return -1;
+ }
+
+ dest = (void *)trampoline_va;
+ memcpy(dest, &hv_crash_asm32, size);
+
+ dest += size;
+ dest = (void *)round_up((ulong)dest, 16);
+ tramp = (struct hv_crash_tramp_data *)dest;
+
+ /* see MAX_ASID_AVAILABLE in tlb.c: "PCID 0 is reserved for use by
+ * non-PCID-aware users". Build cr3 with pcid 0
+ */
+ tramp->tramp32_cr3 = __sme_pa(hv_crash_ptpgs[0]);
+
+ /* Note, when restoring X86_CR4_PCIDE, cr3[11:0] must be zero */
+ tramp->kernel_cr3 = __sme_pa(init_mm.pgd);
+
+ tramp->gdtr32.limit = sizeof(struct hv_crash_tramp_gdt);
+ tramp->gdtr32.address = trampoline_pa +
+ (ulong)&tramp->tramp_gdt - trampoline_va;
+
+ /* base:0 limit:0xfffff type:b dpl:0 P:1 L:1 D:0 avl:0 G:1 */
+ tramp->tramp_gdt.cs64 = 0x00af9a000000ffff;
+
+ tramp->cs_jmptgt.csval = 0x8;
+ offs = (ulong)&hv_crash_asm64_lbl - (ulong)&hv_crash_asm32;
+ tramp->cs_jmptgt.address = trampoline_pa + offs;
+
+ tramp->c_entry_addr = (u64)&hv_crash_c_entry;
+
+ memcpy(dest, tramp, sizeof(struct hv_crash_tramp_data));
+
+ devirt_cr3arg = trampoline_pa + (ulong)dest - trampoline_va;
+
+ return 0;
+}
+
+/*
+ * Build 32bit trampoline page table for transition from protected mode
+ * non-paging to long-mode paging. This transition needs pagetables below 4G.
+ */
+static void hv_crash_build_tramp_pt(void)
+{
+ pgd_t *pgd;
+ p4d_t *p4d;
+ pud_t *pud;
+ pmd_t *pmd;
+ pte_t *pte;
+ u64 pa, addr = trampoline_pa;
+
+ p4d = hv_crash_ptpgs[0] + pgd_index(addr) * sizeof(p4d);
+ pa = virt_to_phys(hv_crash_ptpgs[1]);
+ set_p4d(p4d, __p4d(_PAGE_TABLE | pa));
+ p4d->p4d &= ~(_PAGE_NX); /* disable no execute */
+
+ pud = hv_crash_ptpgs[1] + pud_index(addr) * sizeof(pud);
+ pa = virt_to_phys(hv_crash_ptpgs[2]);
+ set_pud(pud, __pud(_PAGE_TABLE | pa));
+
+ pmd = hv_crash_ptpgs[2] + pmd_index(addr) * sizeof(pmd);
+ pa = virt_to_phys(hv_crash_ptpgs[3]);
+ set_pmd(pmd, __pmd(_PAGE_TABLE | pa));
+
+ pte = hv_crash_ptpgs[3] + pte_index(addr) * sizeof(pte);
+ set_pte(pte, pfn_pte(addr >> PAGE_SHIFT, PAGE_KERNEL_EXEC));
+}
+
+/*
+ * Setup trampoline for devirtualization:
+ * - a page below 4G, ie 32bit addr containing asm glue code that mshv jmps to
+ * in protected mode.
+ * - 4 pages for a temporary page table that asm code uses to turn paging on
+ * - a temporary gdt to use in the compat mode.
+ *
+ * Returns: 0 on success
+ */
+static int hv_crash_trampoline_setup(void)
+{
+ int i, rc, order;
+ struct page *page;
+ u64 trampoline_va;
+ gfp_t flags32 = GFP_KERNEL | GFP_DMA32 | __GFP_ZERO;
+
+ /* page for 32bit trampoline assembly code + hv_crash_tramp_data */
+ page = alloc_page(flags32);
+ if (page == NULL) {
+ pr_err("%s: failed to alloc asm stub page\n", __func__);
+ return -1;
+ }
+
+ trampoline_va = (u64)page_to_virt(page);
+ trampoline_pa = (u32)page_to_phys(page);
+
+ order = 2; /* alloc 2^2 pages */
+ page = alloc_pages(flags32, order);
+ if (page == NULL) {
+ pr_err("%s: failed to alloc pt pages\n", __func__);
+ free_page(trampoline_va);
+ return -1;
+ }
+
+ for (i = 0; i < 4; i++, page++)
+ hv_crash_ptpgs[i] = page_to_virt(page);
+
+ hv_crash_build_tramp_pt();
+
+ rc = hv_crash_setup_trampdata(trampoline_va);
+ if (rc)
+ goto errout;
+
+ return 0;
+
+errout:
+ free_page(trampoline_va);
+ free_pages((ulong)hv_crash_ptpgs[0], order);
+
+ return rc;
+}
+
+/* Setup for kdump kexec to collect hypervisor ram when running as mshv root */
+void hv_root_crash_init(void)
+{
+ int rc;
+ struct hv_input_get_system_property *input;
+ struct hv_output_get_system_property *output;
+ unsigned long flags;
+ u64 status;
+ union hv_pfn_range cda_info;
+
+ if (pgtable_l5_enabled()) {
+ pr_err("Hyper-V: crash dump not yet supported on 5level PTs\n");
+ goto err_out;
+ }
+
+ local_irq_save(flags);
+ input = *this_cpu_ptr(hyperv_pcpu_input_arg);
+ output = *this_cpu_ptr(hyperv_pcpu_output_arg);
+
+ memset(input, 0, sizeof(*input));
+ memset(output, 0, sizeof(*output));
+ input->property_id = HV_SYSTEM_PROPERTY_CRASHDUMPAREA;
+
+ status = hv_do_hypercall(HVCALL_GET_SYSTEM_PROPERTY, input, output);
+ local_irq_restore(flags);
+ if (!hv_result_success(status))
+ goto prop_err_out;
+
+ cda_info.as_uint64 = output->hv_cda_info.as_uint64;
+
+ if (cda_info.base_pfn == 0) {
+ pr_err("Hyper-V: hypervisor crash dump area pfn is 0\n");
+ goto err_out;
+ }
+
+ hv_cda = phys_to_virt(cda_info.base_pfn << PAGE_SHIFT);
+
+ rc = hv_crash_trampoline_setup();
+ if (rc)
+ goto err_out;
+
+ register_nmi_handler(NMI_LOCAL, hv_crash_nmi_local, NMI_FLAG_FIRST,
+ "hv_crash_nmi");
+
+ smp_ops.crash_stop_other_cpus = hv_crash_stop_other_cpus;
+
+ crash_kexec_post_notifiers = true;
+ hv_crash_enabled = 1;
+ pr_info("Hyper-V: linux and hv kdump support enabled\n");
+
+ return;
+
+prop_err_out:
+ pr_err("Hyper-V: %s: property:%d %s\n", __func__, input->property_id,
+ hv_result_to_string(status));
+err_out:
+ pr_err("Hyper-V: only linux (but not hv) kdump support enabled\n");
+}
+
--
2.36.1.vfs.0.0
^ permalink raw reply related [flat|nested] 14+ messages in thread* Re: [PATCH v0 5/6] x86/hyperv: Implement hypervisor ram collection into vmcore
2025-09-04 2:10 ` [PATCH v0 5/6] x86/hyperv: Implement hypervisor ram collection into vmcore Mukesh Rathor
@ 2025-09-04 22:37 ` Stanislav Kinsburskii
2025-09-05 2:38 ` Mukesh R
2025-09-08 16:21 ` Stanislav Kinsburskii
1 sibling, 1 reply; 14+ messages in thread
From: Stanislav Kinsburskii @ 2025-09-04 22:37 UTC (permalink / raw)
To: Mukesh Rathor
Cc: linux-hyperv, linux-kernel, linux-arch, kys, haiyangz, wei.liu,
decui, tglx, mingo, bp, dave.hansen, x86, hpa, arnd
On Wed, Sep 03, 2025 at 07:10:16PM -0700, Mukesh Rathor wrote:
> This commit introduces a new file to enable collection of hypervisor ram
> into the vmcore collected by linux. By default, the hypervisor ram is locked,
> ie, protected via hw page table. Hyper-V implements a disable hypercall which
> essentially devirtualizes the system on the fly. This mechanism makes the
> hypervisor ram accessible to linux without any extra work because it is
> already mapped into linux address space. Details of the implementation
> are available in the file prologue.
>
> Signed-off-by: Mukesh Rathor <mrathor@linux.microsoft.com>
> ---
> arch/x86/hyperv/hv_crash.c | 618 +++++++++++++++++++++++++++++++++++++
> 1 file changed, 618 insertions(+)
> create mode 100644 arch/x86/hyperv/hv_crash.c
>
<snip>
> +
> +/*
> + * Common function for all cpus before devirtualization.
> + *
> + * Hypervisor crash: all cpus get here in nmi context.
> + * Linux crash: the panicing cpu gets here at base level, all others in nmi
> + * context. Note, panicing cpu may not be the bsp.
> + *
> + * The function is not inlined so it will show on the stack. It is named so
> + * because the crash cmd looks for certain well known function names on the
> + * stack before looking into the cpu saved note in the elf section, and
> + * that work is currently incomplete.
> + *
> + * Notes:
> + * Hypervisor crash:
> + * - the hypervisor is in a very restrictive mode at this point and any
> + * vmexit it cannot handle would result in reboot. For example, console
> + * output from here would result in synic ipi hcall, which would result
> + * in reboot. So, no mumbo jumbo, just get to kexec as quickly as possible.
> + *
> + * Devirtualization is supported from the bsp only.
> + */
> +static noinline __noclone void crash_nmi_callback(struct pt_regs *regs)
> +{
> + struct hv_input_disable_hyp_ex *input;
> + u64 status;
> + int msecs = 1000, ccpu = smp_processor_id();
> +
> + if (ccpu == 0) {
> + /* crash_save_cpu() will be done in the kexec path */
> + cpu_emergency_stop_pt(); /* disable performance trace */
> + atomic_inc(&crash_cpus_wait);
> + } else {
> + crash_save_cpu(regs, ccpu);
> + cpu_emergency_stop_pt(); /* disable performance trace */
> + atomic_inc(&crash_cpus_wait);
> + for (;;); /* cause no vmexits */
> + }
> +
> + while (atomic_read(&crash_cpus_wait) < num_online_cpus() && msecs--)
> + mdelay(1);
> +
> + stop_nmi();
> + if (!hv_has_crashed)
> + hv_notify_prepare_hyp();
> +
> + if (crashing_cpu == -1)
> + crashing_cpu = ccpu; /* crash cmd uses this */
> +
> + hv_hvcrash_ctxt_save();
> + hv_mark_tss_not_busy();
> + hv_crash_fixup_kernpt();
> +
> + input = *this_cpu_ptr(hyperv_pcpu_input_arg);
> + memset(input, 0, sizeof(*input));
> + input->rip = trampoline_pa; /* PA of hv_crash_asm32 */
> + input->arg = devirt_cr3arg; /* PA of trampoline page table L4 */
> +
> + status = hv_do_hypercall(HVCALL_DISABLE_HYP_EX, input, NULL);
> + if (!hv_result_success(status)) {
> + pr_emerg("%s: %s\n", __func__, hv_result_to_string(status));
> + pr_emerg("Hyper-V: disable hyp failed. kexec not possible\n");
These prints won't ever be printed to any console as prints in NMI
handler are deffered.
Also, how are they aligned with the notice in the comment on top of
the function stating that console output would lead to synic ipi call?
> + }
> +
> + native_wrmsrq(HV_X64_MSR_RESET, 1); /* get hv to reboot */
Resetting the machine from an NMI handler is sloppy.
There could be another NMI, which triggers the panic, leading to this handler.
NMI handlers servicing is batched meanining that not only this handler
won't output anything, but also any other prints from any other handlers
executed before the same lock won't be written out to consoles.
This introduces silent machine resets for the root partition. Can the
intrusive logic me moved to a tasklet?
> +}
> +
> +/*
> + * generic nmi callback handler: could be called without any crash also.
> + * hv crash: hypervisor injects nmi's into all cpus
> + * lx crash: panicing cpu sends nmi to all but self via crash_stop_other_cpus
> + */
> +static int hv_crash_nmi_local(unsigned int cmd, struct pt_regs *regs)
> +{
> + int ccpu = smp_processor_id();
> +
> + if (!hv_has_crashed && hv_cda && hv_cda->cda_valid)
> + hv_has_crashed = 1;
> +
> + if (!hv_has_crashed && !lx_has_crashed)
> + return NMI_DONE; /* ignore the nmi */
> +
> + if (hv_has_crashed && !hv_crash_enabled) {
> + if (ccpu == 0) {
> + pr_emerg("Hyper-V: core collect not setup. Reboot\n");
Same here: this print won't reach console.
> + native_wrmsrq(HV_X64_MSR_RESET, 1); /* reboot */
> + } else
> + for (;;)
> + cpu_relax();
> + }
> +
> + crash_nmi_callback(regs);
> + return NMI_DONE;
> +}
> +
<snip>
> +/* Setup for kdump kexec to collect hypervisor ram when running as mshv root */
> +void hv_root_crash_init(void)
> +{
> + int rc;
> + struct hv_input_get_system_property *input;
> + struct hv_output_get_system_property *output;
> + unsigned long flags;
> + u64 status;
> + union hv_pfn_range cda_info;
> +
> + if (pgtable_l5_enabled()) {
> + pr_err("Hyper-V: crash dump not yet supported on 5level PTs\n");
> + goto err_out;
> + }
> +
> + local_irq_save(flags);
> + input = *this_cpu_ptr(hyperv_pcpu_input_arg);
> + output = *this_cpu_ptr(hyperv_pcpu_output_arg);
> +
> + memset(input, 0, sizeof(*input));
> + memset(output, 0, sizeof(*output));
> + input->property_id = HV_SYSTEM_PROPERTY_CRASHDUMPAREA;
> +
> + status = hv_do_hypercall(HVCALL_GET_SYSTEM_PROPERTY, input, output);
> + local_irq_restore(flags);
> + if (!hv_result_success(status))
> + goto prop_err_out;
> +
> + cda_info.as_uint64 = output->hv_cda_info.as_uint64;
> +
> + if (cda_info.base_pfn == 0) {
> + pr_err("Hyper-V: hypervisor crash dump area pfn is 0\n");
> + goto err_out;
> + }
> +
> + hv_cda = phys_to_virt(cda_info.base_pfn << PAGE_SHIFT);
> +
> + rc = hv_crash_trampoline_setup();
> + if (rc)
> + goto err_out;
> +
> + register_nmi_handler(NMI_LOCAL, hv_crash_nmi_local, NMI_FLAG_FIRST,
> + "hv_crash_nmi");
> +
Rgistering NMI handler can fail and this should be handled.
Stas
> + smp_ops.crash_stop_other_cpus = hv_crash_stop_other_cpus;
> +
> + crash_kexec_post_notifiers = true;
> + hv_crash_enabled = 1;
> + pr_info("Hyper-V: linux and hv kdump support enabled\n");
> +
> + return;
> +
> +prop_err_out:
> + pr_err("Hyper-V: %s: property:%d %s\n", __func__, input->property_id,
> + hv_result_to_string(status));
> +err_out:
> + pr_err("Hyper-V: only linux (but not hv) kdump support enabled\n");
> +}
> +
> --
> 2.36.1.vfs.0.0
>
^ permalink raw reply [flat|nested] 14+ messages in thread* Re: [PATCH v0 5/6] x86/hyperv: Implement hypervisor ram collection into vmcore
2025-09-04 22:37 ` Stanislav Kinsburskii
@ 2025-09-05 2:38 ` Mukesh R
2025-09-09 17:29 ` Stanislav Kinsburskii
0 siblings, 1 reply; 14+ messages in thread
From: Mukesh R @ 2025-09-05 2:38 UTC (permalink / raw)
To: Stanislav Kinsburskii
Cc: linux-hyperv, linux-kernel, linux-arch, kys, haiyangz, wei.liu,
decui, tglx, mingo, bp, dave.hansen, x86, hpa, arnd
On 9/4/25 15:37, Stanislav Kinsburskii wrote:
> On Wed, Sep 03, 2025 at 07:10:16PM -0700, Mukesh Rathor wrote:
>> This commit introduces a new file to enable collection of hypervisor ram
>> into the vmcore collected by linux. By default, the hypervisor ram is locked,
>> ie, protected via hw page table. Hyper-V implements a disable hypercall which
>> essentially devirtualizes the system on the fly. This mechanism makes the
>> hypervisor ram accessible to linux without any extra work because it is
>> already mapped into linux address space. Details of the implementation
>> are available in the file prologue.
>>
>> Signed-off-by: Mukesh Rathor <mrathor@linux.microsoft.com>
>> ---
>> arch/x86/hyperv/hv_crash.c | 618 +++++++++++++++++++++++++++++++++++++
>> 1 file changed, 618 insertions(+)
>> create mode 100644 arch/x86/hyperv/hv_crash.c
>>
>
> <snip>
>
>> +
>> +/*
>> + * Common function for all cpus before devirtualization.
>> + *
>> + * Hypervisor crash: all cpus get here in nmi context.
>> + * Linux crash: the panicing cpu gets here at base level, all others in nmi
>> + * context. Note, panicing cpu may not be the bsp.
>> + *
>> + * The function is not inlined so it will show on the stack. It is named so
>> + * because the crash cmd looks for certain well known function names on the
>> + * stack before looking into the cpu saved note in the elf section, and
>> + * that work is currently incomplete.
>> + *
>> + * Notes:
>> + * Hypervisor crash:
>> + * - the hypervisor is in a very restrictive mode at this point and any
>> + * vmexit it cannot handle would result in reboot. For example, console
>> + * output from here would result in synic ipi hcall, which would result
>> + * in reboot. So, no mumbo jumbo, just get to kexec as quickly as possible.
>> + *
>> + * Devirtualization is supported from the bsp only.
>> + */
>> +static noinline __noclone void crash_nmi_callback(struct pt_regs *regs)
>> +{
>> + struct hv_input_disable_hyp_ex *input;
>> + u64 status;
>> + int msecs = 1000, ccpu = smp_processor_id();
>> +
>> + if (ccpu == 0) {
>> + /* crash_save_cpu() will be done in the kexec path */
>> + cpu_emergency_stop_pt(); /* disable performance trace */
>> + atomic_inc(&crash_cpus_wait);
>> + } else {
>> + crash_save_cpu(regs, ccpu);
>> + cpu_emergency_stop_pt(); /* disable performance trace */
>> + atomic_inc(&crash_cpus_wait);
>> + for (;;); /* cause no vmexits */
>> + }
>> +
>> + while (atomic_read(&crash_cpus_wait) < num_online_cpus() && msecs--)
>> + mdelay(1);
>> +
>> + stop_nmi();
>> + if (!hv_has_crashed)
>> + hv_notify_prepare_hyp();
>> +
>> + if (crashing_cpu == -1)
>> + crashing_cpu = ccpu; /* crash cmd uses this */
>> +
>> + hv_hvcrash_ctxt_save();
>> + hv_mark_tss_not_busy();
>> + hv_crash_fixup_kernpt();
>> +
>> + input = *this_cpu_ptr(hyperv_pcpu_input_arg);
>> + memset(input, 0, sizeof(*input));
>> + input->rip = trampoline_pa; /* PA of hv_crash_asm32 */
>> + input->arg = devirt_cr3arg; /* PA of trampoline page table L4 */
>> +
>> + status = hv_do_hypercall(HVCALL_DISABLE_HYP_EX, input, NULL);
>> + if (!hv_result_success(status)) {
>> + pr_emerg("%s: %s\n", __func__, hv_result_to_string(status));
>> + pr_emerg("Hyper-V: disable hyp failed. kexec not possible\n");
>
> These prints won't ever be printed to any console as prints in NMI
> handler are deffered.
It's mostly for debug. There are different config options allowing one
to build kernel easily dumping to either uart, led, speaker etc... There
are no easy ways to debug. kernel debuggers could trap EMERGENCY printks
also...
Are you 100% sure printk is async even if KERN_EMERG? If yes, I'd like to
propose someday to make it bypass all that for pr_emerg.
> Also, how are they aligned with the notice in the comment on top of
> the function stating that console output would lead to synic ipi call?
Comment says "Hypervisor Crash". Please reread the whole block.
>> + }
>> +
>> + native_wrmsrq(HV_X64_MSR_RESET, 1); /* get hv to reboot */
>
> Resetting the machine from an NMI handler is sloppy.
> There could be another NMI, which triggers the panic, leading to this handler.
> NMI handlers servicing is batched meanining that not only this handler
> won't output anything, but also any other prints from any other handlers
> executed before the same lock won't be written out to consoles.
>
> This introduces silent machine resets for the root partition. Can the
> intrusive logic me moved to a tasklet?
I really don't think you understand what is going on here. I've tried
telling you at least once in the past year, there is no return from the nmi
handler in case of hyp crash, and that this is panic mode, something
really bad has happened! It could be memory corruption, it could be
hw failure... The hyp goes in emergency mode that just mostly loops,
handling tiny number of hypercalls and msrs for support of dom0/root
like windows that implements custom core collection in raw mode.
Lastly, if disable hyp fails, things are in dire straits and we do last
ditch effort to print in case there is special debug printk or kernel
debugger over serial attached intercepting pr_emerg, or print to LED
for EMERGENCY etc ... before rebooting without hanging.
>> +}
>> +
>> +/*
>> + * generic nmi callback handler: could be called without any crash also.
>> + * hv crash: hypervisor injects nmi's into all cpus
>> + * lx crash: panicing cpu sends nmi to all but self via crash_stop_other_cpus
>> + */
>> +static int hv_crash_nmi_local(unsigned int cmd, struct pt_regs *regs)
>> +{
>> + int ccpu = smp_processor_id();
>> +
>> + if (!hv_has_crashed && hv_cda && hv_cda->cda_valid)
>> + hv_has_crashed = 1;
>> +
>> + if (!hv_has_crashed && !lx_has_crashed)
>> + return NMI_DONE; /* ignore the nmi */
>> +
>> + if (hv_has_crashed && !hv_crash_enabled) {
>> + if (ccpu == 0) {
>> + pr_emerg("Hyper-V: core collect not setup. Reboot\n");
>
> Same here: this print won't reach console.
>
>> + native_wrmsrq(HV_X64_MSR_RESET, 1); /* reboot */
>> + } else
>> + for (;;)
>> + cpu_relax();
>> + }
>> +
>> + crash_nmi_callback(regs);
>> + return NMI_DONE;
>> +}
>> +
>
> <snip>
>
>> +/* Setup for kdump kexec to collect hypervisor ram when running as mshv root */
>> +void hv_root_crash_init(void)
>> +{
>> + int rc;
>> + struct hv_input_get_system_property *input;
>> + struct hv_output_get_system_property *output;
>> + unsigned long flags;
>> + u64 status;
>> + union hv_pfn_range cda_info;
>> +
>> + if (pgtable_l5_enabled()) {
>> + pr_err("Hyper-V: crash dump not yet supported on 5level PTs\n");
>> + goto err_out;
>> + }
>> +
>> + local_irq_save(flags);
>> + input = *this_cpu_ptr(hyperv_pcpu_input_arg);
>> + output = *this_cpu_ptr(hyperv_pcpu_output_arg);
>> +
>> + memset(input, 0, sizeof(*input));
>> + memset(output, 0, sizeof(*output));
>> + input->property_id = HV_SYSTEM_PROPERTY_CRASHDUMPAREA;
>> +
>> + status = hv_do_hypercall(HVCALL_GET_SYSTEM_PROPERTY, input, output);
>> + local_irq_restore(flags);
>> + if (!hv_result_success(status))
>> + goto prop_err_out;
>> +
>> + cda_info.as_uint64 = output->hv_cda_info.as_uint64;
>> +
>> + if (cda_info.base_pfn == 0) {
>> + pr_err("Hyper-V: hypervisor crash dump area pfn is 0\n");
>> + goto err_out;
>> + }
>> +
>> + hv_cda = phys_to_virt(cda_info.base_pfn << PAGE_SHIFT);
>> +
>> + rc = hv_crash_trampoline_setup();
>> + if (rc)
>> + goto err_out;
>> +
>> + register_nmi_handler(NMI_LOCAL, hv_crash_nmi_local, NMI_FLAG_FIRST,
>> + "hv_crash_nmi");
>> +
>
> Rgistering NMI handler can fail and this should be handled.
Not sure if it "should be". There are other instances where this is
not done because the call is guaranteed to be not recursive.
> Stas
>
>> + smp_ops.crash_stop_other_cpus = hv_crash_stop_other_cpus;
>> +
>> + crash_kexec_post_notifiers = true;
>> + hv_crash_enabled = 1;
>> + pr_info("Hyper-V: linux and hv kdump support enabled\n");
>> +
>> + return;
>> +
>> +prop_err_out:
>> + pr_err("Hyper-V: %s: property:%d %s\n", __func__, input->property_id,
>> + hv_result_to_string(status));
>> +err_out:
>> + pr_err("Hyper-V: only linux (but not hv) kdump support enabled\n");
>> +}
>> +
>> --
>> 2.36.1.vfs.0.0
>>
^ permalink raw reply [flat|nested] 14+ messages in thread* Re: [PATCH v0 5/6] x86/hyperv: Implement hypervisor ram collection into vmcore
2025-09-05 2:38 ` Mukesh R
@ 2025-09-09 17:29 ` Stanislav Kinsburskii
2025-09-09 18:25 ` Mukesh R
0 siblings, 1 reply; 14+ messages in thread
From: Stanislav Kinsburskii @ 2025-09-09 17:29 UTC (permalink / raw)
To: Mukesh R
Cc: linux-hyperv, linux-kernel, linux-arch, kys, haiyangz, wei.liu,
decui, tglx, mingo, bp, dave.hansen, x86, hpa, arnd
On Thu, Sep 04, 2025 at 07:38:53PM -0700, Mukesh R wrote:
> On 9/4/25 15:37, Stanislav Kinsburskii wrote:
> > On Wed, Sep 03, 2025 at 07:10:16PM -0700, Mukesh Rathor wrote:
> >> +
> >> +/*
> >> + * Common function for all cpus before devirtualization.
> >> + *
> >> + * Hypervisor crash: all cpus get here in nmi context.
> >> + * Linux crash: the panicing cpu gets here at base level, all others in nmi
> >> + * context. Note, panicing cpu may not be the bsp.
> >> + *
> >> + * The function is not inlined so it will show on the stack. It is named so
> >> + * because the crash cmd looks for certain well known function names on the
> >> + * stack before looking into the cpu saved note in the elf section, and
> >> + * that work is currently incomplete.
> >> + *
> >> + * Notes:
> >> + * Hypervisor crash:
> >> + * - the hypervisor is in a very restrictive mode at this point and any
> >> + * vmexit it cannot handle would result in reboot. For example, console
> >> + * output from here would result in synic ipi hcall, which would result
> >> + * in reboot. So, no mumbo jumbo, just get to kexec as quickly as possible.
> >> + *
> >> + * Devirtualization is supported from the bsp only.
> >> + */
> >> +static noinline __noclone void crash_nmi_callback(struct pt_regs *regs)
> >> +{
> >> + struct hv_input_disable_hyp_ex *input;
> >> + u64 status;
> >> + int msecs = 1000, ccpu = smp_processor_id();
> >> +
> >> + if (ccpu == 0) {
> >> + /* crash_save_cpu() will be done in the kexec path */
> >> + cpu_emergency_stop_pt(); /* disable performance trace */
> >> + atomic_inc(&crash_cpus_wait);
> >> + } else {
> >> + crash_save_cpu(regs, ccpu);
> >> + cpu_emergency_stop_pt(); /* disable performance trace */
> >> + atomic_inc(&crash_cpus_wait);
> >> + for (;;); /* cause no vmexits */
> >> + }
> >> +
> >> + while (atomic_read(&crash_cpus_wait) < num_online_cpus() && msecs--)
> >> + mdelay(1);
> >> +
> >> + stop_nmi();
> >> + if (!hv_has_crashed)
> >> + hv_notify_prepare_hyp();
> >> +
> >> + if (crashing_cpu == -1)
> >> + crashing_cpu = ccpu; /* crash cmd uses this */
> >> +
> >> + hv_hvcrash_ctxt_save();
> >> + hv_mark_tss_not_busy();
> >> + hv_crash_fixup_kernpt();
> >> +
> >> + input = *this_cpu_ptr(hyperv_pcpu_input_arg);
> >> + memset(input, 0, sizeof(*input));
> >> + input->rip = trampoline_pa; /* PA of hv_crash_asm32 */
> >> + input->arg = devirt_cr3arg; /* PA of trampoline page table L4 */
> >> +
> >> + status = hv_do_hypercall(HVCALL_DISABLE_HYP_EX, input, NULL);
> >> + if (!hv_result_success(status)) {
> >> + pr_emerg("%s: %s\n", __func__, hv_result_to_string(status));
> >> + pr_emerg("Hyper-V: disable hyp failed. kexec not possible\n");
> >
> > These prints won't ever be printed to any console as prints in NMI
> > handler are deffered.
>
> It's mostly for debug. There are different config options allowing one
> to build kernel easily dumping to either uart, led, speaker etc... There
> are no easy ways to debug. kernel debuggers could trap EMERGENCY printks
> also...
>
> Are you 100% sure printk is async even if KERN_EMERG? If yes, I'd like to
> propose someday to make it bypass all that for pr_emerg.
>
Yes, I'm quite sure. Right now this looks like is dead code.
>
> > Also, how are they aligned with the notice in the comment on top of
> > the function stating that console output would lead to synic ipi call?
>
> Comment says "Hypervisor Crash". Please reread the whole block.
>
The comment states that in case of hypervisor crash "console
output from here would result in synic ipi hcall, which would result in
reboot".
So, why printing anything if it will simply lead to reboot?
> >
> > Resetting the machine from an NMI handler is sloppy.
> > There could be another NMI, which triggers the panic, leading to this handler.
> > NMI handlers servicing is batched meanining that not only this handler
> > won't output anything, but also any other prints from any other handlers
> > executed before the same lock won't be written out to consoles.
> >
> > This introduces silent machine resets for the root partition. Can the
> > intrusive logic me moved to a tasklet?
>
> I really don't think you understand what is going on here. I've tried
> telling you at least once in the past year, there is no return from the nmi
> handler in case of hyp crash, and that this is panic mode, something
> really bad has happened! It could be memory corruption, it could be
> hw failure... The hyp goes in emergency mode that just mostly loops,
> handling tiny number of hypercalls and msrs for support of dom0/root
> like windows that implements custom core collection in raw mode.
>
I wasn't clear.
I wasn't talking about a hypervisor crash. If it is so intrusive, that an
attempt to print things to console may lead to reboot, then there should
be no prints for this case.
But this same logic is also used for Linux crashes, when prints can and
should be printed to console.
Moreover, whe same logic is used for a case when there is no crash
kernel loaded, which as I said already leads to silent reboot if panic
has happened in NMI handler.
I believe this needs to be fixed.
Stas
^ permalink raw reply [flat|nested] 14+ messages in thread* Re: [PATCH v0 5/6] x86/hyperv: Implement hypervisor ram collection into vmcore
2025-09-09 17:29 ` Stanislav Kinsburskii
@ 2025-09-09 18:25 ` Mukesh R
0 siblings, 0 replies; 14+ messages in thread
From: Mukesh R @ 2025-09-09 18:25 UTC (permalink / raw)
To: Stanislav Kinsburskii
Cc: linux-hyperv, linux-kernel, linux-arch, kys, haiyangz, wei.liu,
decui, tglx, mingo, bp, dave.hansen, x86, hpa, arnd
On 9/9/25 10:29, Stanislav Kinsburskii wrote:
> On Thu, Sep 04, 2025 at 07:38:53PM -0700, Mukesh R wrote:
>> On 9/4/25 15:37, Stanislav Kinsburskii wrote:
>>> On Wed, Sep 03, 2025 at 07:10:16PM -0700, Mukesh Rathor wrote:
>>>> +
>>>> +/*
>>>> + * Common function for all cpus before devirtualization.
>>>> + *
>>>> + * Hypervisor crash: all cpus get here in nmi context.
>>>> + * Linux crash: the panicing cpu gets here at base level, all others in nmi
>>>> + * context. Note, panicing cpu may not be the bsp.
>>>> + *
>>>> + * The function is not inlined so it will show on the stack. It is named so
>>>> + * because the crash cmd looks for certain well known function names on the
>>>> + * stack before looking into the cpu saved note in the elf section, and
>>>> + * that work is currently incomplete.
>>>> + *
>>>> + * Notes:
>>>> + * Hypervisor crash:
>>>> + * - the hypervisor is in a very restrictive mode at this point and any
>>>> + * vmexit it cannot handle would result in reboot. For example, console
>>>> + * output from here would result in synic ipi hcall, which would result
>>>> + * in reboot. So, no mumbo jumbo, just get to kexec as quickly as possible.
>>>> + *
>>>> + * Devirtualization is supported from the bsp only.
>>>> + */
>>>> +static noinline __noclone void crash_nmi_callback(struct pt_regs *regs)
>>>> +{
>>>> + struct hv_input_disable_hyp_ex *input;
>>>> + u64 status;
>>>> + int msecs = 1000, ccpu = smp_processor_id();
>>>> +
>>>> + if (ccpu == 0) {
>>>> + /* crash_save_cpu() will be done in the kexec path */
>>>> + cpu_emergency_stop_pt(); /* disable performance trace */
>>>> + atomic_inc(&crash_cpus_wait);
>>>> + } else {
>>>> + crash_save_cpu(regs, ccpu);
>>>> + cpu_emergency_stop_pt(); /* disable performance trace */
>>>> + atomic_inc(&crash_cpus_wait);
>>>> + for (;;); /* cause no vmexits */
>>>> + }
>>>> +
>>>> + while (atomic_read(&crash_cpus_wait) < num_online_cpus() && msecs--)
>>>> + mdelay(1);
>>>> +
>>>> + stop_nmi();
>>>> + if (!hv_has_crashed)
>>>> + hv_notify_prepare_hyp();
>>>> +
>>>> + if (crashing_cpu == -1)
>>>> + crashing_cpu = ccpu; /* crash cmd uses this */
>>>> +
>>>> + hv_hvcrash_ctxt_save();
>>>> + hv_mark_tss_not_busy();
>>>> + hv_crash_fixup_kernpt();
>>>> +
>>>> + input = *this_cpu_ptr(hyperv_pcpu_input_arg);
>>>> + memset(input, 0, sizeof(*input));
>>>> + input->rip = trampoline_pa; /* PA of hv_crash_asm32 */
>>>> + input->arg = devirt_cr3arg; /* PA of trampoline page table L4 */
>>>> +
>>>> + status = hv_do_hypercall(HVCALL_DISABLE_HYP_EX, input, NULL);
>>>> + if (!hv_result_success(status)) {
>>>> + pr_emerg("%s: %s\n", __func__, hv_result_to_string(status));
>>>> + pr_emerg("Hyper-V: disable hyp failed. kexec not possible\n");
>>>
>>> These prints won't ever be printed to any console as prints in NMI
>>> handler are deffered.
>>
>> It's mostly for debug. There are different config options allowing one
>> to build kernel easily dumping to either uart, led, speaker etc... There
>> are no easy ways to debug. kernel debuggers could trap EMERGENCY printks
>> also...
>>
>> Are you 100% sure printk is async even if KERN_EMERG? If yes, I'd like to
>> propose someday to make it bypass all that for pr_emerg.
>>
>
> Yes, I'm quite sure. Right now this looks like is dead code.
>
>>
>>> Also, how are they aligned with the notice in the comment on top of
>>> the function stating that console output would lead to synic ipi call?
>>
>> Comment says "Hypervisor Crash". Please reread the whole block.
>>
>
> The comment states that in case of hypervisor crash "console
> output from here would result in synic ipi hcall, which would result in
> reboot".
> So, why printing anything if it will simply lead to reboot?
>
>>>
>>> Resetting the machine from an NMI handler is sloppy.
>>> There could be another NMI, which triggers the panic, leading to this handler.
>>> NMI handlers servicing is batched meanining that not only this handler
>>> won't output anything, but also any other prints from any other handlers
>>> executed before the same lock won't be written out to consoles.
>>>
>>> This introduces silent machine resets for the root partition. Can the
>>> intrusive logic me moved to a tasklet?
>>
>> I really don't think you understand what is going on here. I've tried
>> telling you at least once in the past year, there is no return from the nmi
>> handler in case of hyp crash, and that this is panic mode, something
>> really bad has happened! It could be memory corruption, it could be
>> hw failure... The hyp goes in emergency mode that just mostly loops,
>> handling tiny number of hypercalls and msrs for support of dom0/root
>> like windows that implements custom core collection in raw mode.
>>
>
> I wasn't clear.
> I wasn't talking about a hypervisor crash. If it is so intrusive, that an
> attempt to print things to console may lead to reboot, then there should
> be no prints for this case.
The line after the print is reboot!!
Ah, forget it! heck with the prints...
> But this same logic is also used for Linux crashes, when prints can and
> should be printed to console.
check the panic function to figure when/where it prints, then check
where the nmi is called from. that will help.
> Moreover, whe same logic is used for a case when there is no crash
> kernel loaded, which as I said already leads to silent reboot if panic
> has happened in NMI handler.
>
> I believe this needs to be fixed.
>
> Stas
>
^ permalink raw reply [flat|nested] 14+ messages in thread
* Re: [PATCH v0 5/6] x86/hyperv: Implement hypervisor ram collection into vmcore
2025-09-04 2:10 ` [PATCH v0 5/6] x86/hyperv: Implement hypervisor ram collection into vmcore Mukesh Rathor
2025-09-04 22:37 ` Stanislav Kinsburskii
@ 2025-09-08 16:21 ` Stanislav Kinsburskii
2025-09-08 17:38 ` Mukesh R
1 sibling, 1 reply; 14+ messages in thread
From: Stanislav Kinsburskii @ 2025-09-08 16:21 UTC (permalink / raw)
To: Mukesh Rathor
Cc: linux-hyperv, linux-kernel, linux-arch, kys, haiyangz, wei.liu,
decui, tglx, mingo, bp, dave.hansen, x86, hpa, arnd
On Wed, Sep 03, 2025 at 07:10:16PM -0700, Mukesh Rathor wrote:
> This commit introduces a new file to enable collection of hypervisor ram
> into the vmcore collected by linux. By default, the hypervisor ram is locked,
> ie, protected via hw page table. Hyper-V implements a disable hypercall which
> essentially devirtualizes the system on the fly. This mechanism makes the
> hypervisor ram accessible to linux without any extra work because it is
> already mapped into linux address space. Details of the implementation
> are available in the file prologue.
>
> Signed-off-by: Mukesh Rathor <mrathor@linux.microsoft.com>
> ---
> arch/x86/hyperv/hv_crash.c | 618 +++++++++++++++++++++++++++++++++++++
> 1 file changed, 618 insertions(+)
> create mode 100644 arch/x86/hyperv/hv_crash.c
>
> diff --git a/arch/x86/hyperv/hv_crash.c b/arch/x86/hyperv/hv_crash.c
> new file mode 100644
> index 000000000000..50c54d39f0e2
> --- /dev/null
> +++ b/arch/x86/hyperv/hv_crash.c
> +
<snip>
> +/*
> + * generic nmi callback handler: could be called without any crash also.
> + * hv crash: hypervisor injects nmi's into all cpus
> + * lx crash: panicing cpu sends nmi to all but self via crash_stop_other_cpus
> + */
> +static int hv_crash_nmi_local(unsigned int cmd, struct pt_regs *regs)
> +{
> + int ccpu = smp_processor_id();
> +
> + if (!hv_has_crashed && hv_cda && hv_cda->cda_valid)
> + hv_has_crashed = 1;
> +
> + if (!hv_has_crashed && !lx_has_crashed)
> + return NMI_DONE; /* ignore the nmi */
> +
> + if (hv_has_crashed && !hv_crash_enabled) {
> + if (ccpu == 0) {
> + pr_emerg("Hyper-V: core collect not setup. Reboot\n");
> + native_wrmsrq(HV_X64_MSR_RESET, 1); /* reboot */
> + } else
> + for (;;)
> + cpu_relax();
> + }
> +
> + crash_nmi_callback(regs);
> + return NMI_DONE;
> +}
One more thing.
It looks like the function above goes through the new logic even when
hypervisor is intact and there is no crash kernel loaded.
This is redundant and it should rather return back to the existent
generic kernel panic logic.
Stanislav
^ permalink raw reply [flat|nested] 14+ messages in thread* Re: [PATCH v0 5/6] x86/hyperv: Implement hypervisor ram collection into vmcore
2025-09-08 16:21 ` Stanislav Kinsburskii
@ 2025-09-08 17:38 ` Mukesh R
0 siblings, 0 replies; 14+ messages in thread
From: Mukesh R @ 2025-09-08 17:38 UTC (permalink / raw)
To: Stanislav Kinsburskii
Cc: linux-hyperv, linux-kernel, linux-arch, kys, haiyangz, wei.liu,
decui, tglx, mingo, bp, dave.hansen, x86, hpa, arnd
On 9/8/25 09:21, Stanislav Kinsburskii wrote:
> On Wed, Sep 03, 2025 at 07:10:16PM -0700, Mukesh Rathor wrote:
>> This commit introduces a new file to enable collection of hypervisor ram
>> into the vmcore collected by linux. By default, the hypervisor ram is locked,
>> ie, protected via hw page table. Hyper-V implements a disable hypercall which
>> essentially devirtualizes the system on the fly. This mechanism makes the
>> hypervisor ram accessible to linux without any extra work because it is
>> already mapped into linux address space. Details of the implementation
>> are available in the file prologue.
>>
>> Signed-off-by: Mukesh Rathor <mrathor@linux.microsoft.com>
>> ---
>> arch/x86/hyperv/hv_crash.c | 618 +++++++++++++++++++++++++++++++++++++
>> 1 file changed, 618 insertions(+)
>> create mode 100644 arch/x86/hyperv/hv_crash.c
>>
>> diff --git a/arch/x86/hyperv/hv_crash.c b/arch/x86/hyperv/hv_crash.c
>> new file mode 100644
>> index 000000000000..50c54d39f0e2
>> --- /dev/null
>> +++ b/arch/x86/hyperv/hv_crash.c
>> +
>
> <snip>
>
>> +/*
>> + * generic nmi callback handler: could be called without any crash also.
>> + * hv crash: hypervisor injects nmi's into all cpus
>> + * lx crash: panicing cpu sends nmi to all but self via crash_stop_other_cpus
>> + */
>> +static int hv_crash_nmi_local(unsigned int cmd, struct pt_regs *regs)
>> +{
>> + int ccpu = smp_processor_id();
>> +
>> + if (!hv_has_crashed && hv_cda && hv_cda->cda_valid)
>> + hv_has_crashed = 1;
>> +
>> + if (!hv_has_crashed && !lx_has_crashed)
>> + return NMI_DONE; /* ignore the nmi */
>> +
>> + if (hv_has_crashed && !hv_crash_enabled) {
>> + if (ccpu == 0) {
>> + pr_emerg("Hyper-V: core collect not setup. Reboot\n");
>> + native_wrmsrq(HV_X64_MSR_RESET, 1); /* reboot */
>> + } else
>> + for (;;)
>> + cpu_relax();
>> + }
>> +
>> + crash_nmi_callback(regs);
>> + return NMI_DONE;
>> +}
>
> One more thing.
> It looks like the function above goes through the new logic even when
> hypervisor is intact and there is no crash kernel loaded.
> This is redundant and it should rather return back to the existent
> generic kernel panic logic.
Yeah, that is already addressed in V1 coming up.
^ permalink raw reply [flat|nested] 14+ messages in thread
* [PATCH v0 6/6] Hyper-V: Enable build of hypervisor crash collection files
2025-09-04 2:10 [PATCH v0 0/6] Hyper-V: Implement hypervisor core collection Mukesh Rathor
` (4 preceding siblings ...)
2025-09-04 2:10 ` [PATCH v0 5/6] x86/hyperv: Implement hypervisor ram collection into vmcore Mukesh Rathor
@ 2025-09-04 2:10 ` Mukesh Rathor
5 siblings, 0 replies; 14+ messages in thread
From: Mukesh Rathor @ 2025-09-04 2:10 UTC (permalink / raw)
To: linux-hyperv, linux-kernel, linux-arch
Cc: kys, haiyangz, wei.liu, decui, tglx, mingo, bp, dave.hansen, x86,
hpa, arnd
This final commit enables build of the new files introduced in the
earlier commits and adds call to do the setup during boot.
Signed-off-by: Mukesh Rathor <mrathor@linux.microsoft.com>
---
arch/x86/hyperv/Makefile | 6 ++++++
arch/x86/hyperv/hv_init.c | 2 ++
include/asm-generic/mshyperv.h | 9 +++++++++
3 files changed, 17 insertions(+)
diff --git a/arch/x86/hyperv/Makefile b/arch/x86/hyperv/Makefile
index d55f494f471d..6f5d97cddd80 100644
--- a/arch/x86/hyperv/Makefile
+++ b/arch/x86/hyperv/Makefile
@@ -5,4 +5,10 @@ obj-$(CONFIG_HYPERV_VTL_MODE) += hv_vtl.o
ifdef CONFIG_X86_64
obj-$(CONFIG_PARAVIRT_SPINLOCKS) += hv_spinlock.o
+
+ ifdef CONFIG_MSHV_ROOT
+ CFLAGS_REMOVE_hv_trampoline.o += -pg
+ CFLAGS_hv_trampoline.o += -fno-stack-protector
+ obj-$(CONFIG_CRASH_DUMP) += hv_crash.o hv_trampoline.o
+ endif
endif
diff --git a/arch/x86/hyperv/hv_init.c b/arch/x86/hyperv/hv_init.c
index 6919f1af2e51..e3a4650b9ec7 100644
--- a/arch/x86/hyperv/hv_init.c
+++ b/arch/x86/hyperv/hv_init.c
@@ -513,6 +513,8 @@ void __init hyperv_init(void)
/* mark ram reserved for hypervisor as owned by hypervisor */
hv_mark_resources();
+
+ hv_root_crash_init();
} else {
hypercall_msr.guest_physical_address = vmalloc_to_pfn(hv_hypercall_pg);
wrmsrq(HV_X64_MSR_HYPERCALL, hypercall_msr.as_uint64);
diff --git a/include/asm-generic/mshyperv.h b/include/asm-generic/mshyperv.h
index 1d2ad1304ad4..57cc3e0cac90 100644
--- a/include/asm-generic/mshyperv.h
+++ b/include/asm-generic/mshyperv.h
@@ -369,6 +369,15 @@ int hv_call_deposit_pages(int node, u64 partition_id, u32 num_pages);
int hv_call_add_logical_proc(int node, u32 lp_index, u32 acpi_id);
int hv_call_create_vp(int node, u64 partition_id, u32 vp_index, u32 flags);
+#if CONFIG_CRASH_DUMP
+void hv_root_crash_init(void);
+void hv_crash_asm32(void);
+void hv_crash_asm64_lbl(void);
+void hv_crash_asm32_end(void);
+#else /* CONFIG_CRASH_DUMP */
+static inline void hv_root_crash_init(void) {}
+#endif /* CONFIG_CRASH_DUMP */
+
#else /* CONFIG_MSHV_ROOT */
static inline bool hv_root_partition(void) { return false; }
static inline bool hv_l1vh_partition(void) { return false; }
--
2.36.1.vfs.0.0
^ permalink raw reply related [flat|nested] 14+ messages in thread