RE: [PATCH v3 3/6] x86/hyperv: Skip LP/VP creation on kexec

[Michael Kelley]

From: Jork Loeser <jloeser@xxxxxxxxxxxxxxxxxxx> Sent: Tuesday, April 7, 2026 6:37 PM
> 
> After a kexec the logical processors and virtual processors already
> exist in the hypervisor because they were created by the previous
> kernel. Attempting to add them again causes either a BUG_ON or
> corrupted VP state leading to MCEs in the new kernel.
> 
> Add hv_lp_exists() to probe whether an LP is already present by
> calling HVCALL_GET_LOGICAL_PROCESSOR_RUN_TIME. When it succeeds the
> LP exists and we skip the add-LP and create-VP loops entirely.
> 
> Also add hv_call_notify_all_processors_started() which informs the
> hypervisor that all processors are online. This is required after
> adding LPs (fresh boot) and is a no-op on kexec since we skip that
> path.

Adding hv_call_notify_all_processors_started() seems like it should be
a separate patch. And this paragraph in the commit message leaves me
with questions:  Is it really "required"?  If it is, how does the existing
upstream code ever work? Does the change need to be backported
to stable kernels? If it isn't *really* required, what are the implications
of not doing it?

> 
> Co-developed-by: Anirudh Rayabharam <anrayabh@xxxxxxxxxxxxxxxxxxx>
> Signed-off-by: Anirudh Rayabharam <anrayabh@xxxxxxxxxxxxxxxxxxx>
> Co-developed-by: Stanislav Kinsburskii <stanislav.kinsburskii@xxxxxxxxx>
> Signed-off-by: Stanislav Kinsburskii <stanislav.kinsburskii@xxxxxxxxx>
> Co-developed-by: Mukesh Rathor <mrathor@xxxxxxxxxxxxxxxxxxx>
> Signed-off-by: Mukesh Rathor <mrathor@xxxxxxxxxxxxxxxxxxx>
> Signed-off-by: Jork Loeser <jloeser@xxxxxxxxxxxxxxxxxxx>
> ---
>  arch/x86/kernel/cpu/mshyperv.c |  7 +++++
>  drivers/hv/hv_proc.c           | 47 ++++++++++++++++++++++++++++++++++
>  include/asm-generic/mshyperv.h | 10 ++++++++
>  include/hyperv/hvgdk_mini.h    |  1 +
>  include/hyperv/hvhdk_mini.h    | 12 +++++++++
>  5 files changed, 77 insertions(+)
> 
> diff --git a/arch/x86/kernel/cpu/mshyperv.c b/arch/x86/kernel/cpu/mshyperv.c
> index e498b6b2ef19..b5b6a58b67b0 100644
> --- a/arch/x86/kernel/cpu/mshyperv.c
> +++ b/arch/x86/kernel/cpu/mshyperv.c
> @@ -431,6 +431,10 @@ static void __init hv_smp_prepare_cpus(unsigned int max_cpus)
>  	}
> 
>  #ifdef CONFIG_X86_64
> +	/* If AP LPs exist, we are in a kexec'd kernel and VPs already exist */
> +	if (num_present_cpus() == 1 || hv_lp_exists(1))
> +		return;
> +
>  	for_each_present_cpu(i) {
>  		if (i == 0)
>  			continue;
> @@ -438,6 +442,9 @@ static void __init hv_smp_prepare_cpus(unsigned int max_cpus)
>  		BUG_ON(ret);
>  	}
> 
> +	ret = hv_call_notify_all_processors_started();
> +	WARN_ON(ret);
> +
>  	for_each_present_cpu(i) {
>  		if (i == 0)
>  			continue;

An observation:  hv_smp_prepare_cpus() is getting to be a bit of a mess.
It handles both the SNP case and the root case, which aren't really related.
I could envision having hv_smp_prepare_cpus_for_snp() and
hv_smp_prepare_cpus_for_root() in order to separate the two cases
cleanly.

Then hv_smp_prepare_cpus_for_root() calls four functions in hv_proc.c,
all of which require stubs for the case where MSHV root isn't being built.
Better would be to move the root version of prepare CPUs functionality
into a new function in hv_proc.c, and only have a stub for that single
function. Three of the other four called functions could then become static.
The #ifdef CONFIG_X86_64 could also go away since hv_proc.c is only
built for x64.

I'll probably submit a separate patch to implement these suggested
cleanups, unless someone else wants to do it first.

Michael

> diff --git a/drivers/hv/hv_proc.c b/drivers/hv/hv_proc.c
> index 3cb4b2a3035c..57b2c64197cb 100644
> --- a/drivers/hv/hv_proc.c
> +++ b/drivers/hv/hv_proc.c
> @@ -239,3 +239,50 @@ int hv_call_create_vp(int node, u64 partition_id, u32 vp_index, u32 flags)
>  	return ret;
>  }
>  EXPORT_SYMBOL_GPL(hv_call_create_vp);
> +
> +int hv_call_notify_all_processors_started(void)
> +{
> +	struct hv_input_notify_partition_event *input;
> +	u64 status;
> +	unsigned long irq_flags;
> +	int ret = 0;
> +
> +	local_irq_save(irq_flags);
> +	input = *this_cpu_ptr(hyperv_pcpu_input_arg);
> +	memset(input, 0, sizeof(*input));
> +	input->event = HV_PARTITION_ALL_LOGICAL_PROCESSORS_STARTED;
> +	status = hv_do_hypercall(HVCALL_NOTIFY_PARTITION_EVENT,
> +				 input, NULL);
> +	local_irq_restore(irq_flags);
> +
> +	if (!hv_result_success(status)) {
> +		hv_status_err(status, "\n");
> +		ret = hv_result_to_errno(status);
> +	}
> +	return ret;
> +}
> +
> +bool hv_lp_exists(u32 lp_index)
> +{
> +	struct hv_input_get_logical_processor_run_time *input;
> +	struct hv_output_get_logical_processor_run_time *output;
> +	unsigned long flags;
> +	u64 status;
> +
> +	local_irq_save(flags);
> +	input = *this_cpu_ptr(hyperv_pcpu_input_arg);
> +	output = *this_cpu_ptr(hyperv_pcpu_output_arg);
> +
> +	input->lp_index = lp_index;
> +	status = hv_do_hypercall(HVCALL_GET_LOGICAL_PROCESSOR_RUN_TIME,
> +				 input, output);
> +	local_irq_restore(flags);
> +
> +	if (!hv_result_success(status) &&
> +	    hv_result(status) != HV_STATUS_INVALID_LP_INDEX) {
> +		hv_status_err(status, "\n");
> +		BUG();
> +	}
> +
> +	return hv_result_success(status);
> +}
> diff --git a/include/asm-generic/mshyperv.h b/include/asm-generic/mshyperv.h
> index d37b68238c97..bf601d67cecb 100644
> --- a/include/asm-generic/mshyperv.h
> +++ b/include/asm-generic/mshyperv.h
> @@ -347,6 +347,8 @@ bool hv_result_needs_memory(u64 status);
>  int hv_deposit_memory_node(int node, u64 partition_id, u64 status);
>  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_notify_all_processors_started(void);
> +bool hv_lp_exists(u32 lp_index);
>  int hv_call_create_vp(int node, u64 partition_id, u32 vp_index, u32 flags);
> 
>  #else /* CONFIG_MSHV_ROOT */
> @@ -366,6 +368,14 @@ static inline int hv_call_add_logical_proc(int node, u32 lp_index, u32 acpi_id)
>  {
>  	return -EOPNOTSUPP;
>  }
> +static inline int hv_call_notify_all_processors_started(void)
> +{
> +	return -EOPNOTSUPP;
> +}
> +static inline bool hv_lp_exists(u32 lp_index)
> +{
> +	return false;
> +}
>  static inline int hv_call_create_vp(int node, u64 partition_id, u32 vp_index, u32 flags)
>  {
>  	return -EOPNOTSUPP;
> diff --git a/include/hyperv/hvgdk_mini.h b/include/hyperv/hvgdk_mini.h
> index f9600f87186a..6a4e8b9d570f 100644
> --- a/include/hyperv/hvgdk_mini.h
> +++ b/include/hyperv/hvgdk_mini.h
> @@ -435,6 +435,7 @@ union hv_vp_assist_msr_contents {	 /*
> HV_REGISTER_VP_ASSIST_PAGE */
>  /* HV_CALL_CODE */
>  #define HVCALL_FLUSH_VIRTUAL_ADDRESS_SPACE		0x0002
>  #define HVCALL_FLUSH_VIRTUAL_ADDRESS_LIST		0x0003
> +#define HVCALL_GET_LOGICAL_PROCESSOR_RUN_TIME		0x0004
>  #define HVCALL_NOTIFY_LONG_SPIN_WAIT			0x0008
>  #define HVCALL_SEND_IPI					0x000b
>  #define HVCALL_ENABLE_VP_VTL				0x000f
> diff --git a/include/hyperv/hvhdk_mini.h b/include/hyperv/hvhdk_mini.h
> index 091c03e26046..b4cb2fa26e9b 100644
> --- a/include/hyperv/hvhdk_mini.h
> +++ b/include/hyperv/hvhdk_mini.h
> @@ -362,6 +362,7 @@ union hv_partition_event_input {
> 
>  enum hv_partition_event {
>  	HV_PARTITION_EVENT_ROOT_CRASHDUMP = 2,
> +	HV_PARTITION_ALL_LOGICAL_PROCESSORS_STARTED = 4,
>  };
> 
>  struct hv_input_notify_partition_event {
> @@ -369,6 +370,17 @@ struct hv_input_notify_partition_event {
>  	union hv_partition_event_input input;
>  } __packed;
> 
> +struct hv_input_get_logical_processor_run_time {
> +	u32 lp_index;
> +} __packed;
> +
> +struct hv_output_get_logical_processor_run_time {
> +	u64 global_time;
> +	u64 local_run_time;
> +	u64 rsvdz0;
> +	u64 hypervisor_time;
> +} __packed;
> +
>  struct hv_lp_startup_status {
>  	u64 hv_status;
>  	u64 substatus1;
> --
> 2.43.0
>