[PATCH v7 4/8] crash: add generic infrastructure for crash hotplug support

[Baoquan He]

On 04/13/22 at 12:42pm, Eric DeVolder wrote:
> Upon CPU and memory changes, a generic crash_hotplug_handler()
> dispatches the hot plug/unplug event to the architecture specific
> arch_crash_hotplug_handler(). During the process, the kexec_mutex
> is held.
> 
> To support cpu hotplug, a callback is registered to capture the
> CPUHP_AP_ONLINE_DYN online and ofline events via
> cpuhp_setup_state_nocalls().
> 
> To support memory hotplug, a notifier is registered to capture the
> MEM_ONLINE and MEM_OFFLINE events via register_memory_notifier().
> 
> The cpu callback and memory notifier then call crash_hotplug_handler()
> to handle the hot plug/unplug event.
> 
> Signed-off-by: Eric DeVolder <eric.devolder at oracle.com>
> ---
>  include/linux/kexec.h |  16 +++++++
>  kernel/crash_core.c   | 101 ++++++++++++++++++++++++++++++++++++++++++
>  2 files changed, 117 insertions(+)
> 
> diff --git a/include/linux/kexec.h b/include/linux/kexec.h
> index f93f2591fc1e..02daff1f47dd 100644
> --- a/include/linux/kexec.h
> +++ b/include/linux/kexec.h
> @@ -306,6 +306,13 @@ struct kimage {
>  
>  	/* Information for loading purgatory */
>  	struct purgatory_info purgatory_info;
> +
> +#ifdef CONFIG_CRASH_HOTPLUG
> +	bool hotplug_event;
> +	unsigned int offlinecpu;
> +	bool elfcorehdr_index_valid;
> +	int elfcorehdr_index;
> +#endif
>  #endif
>  
>  #ifdef CONFIG_IMA_KEXEC
> @@ -322,6 +329,15 @@ struct kimage {
>  	unsigned long elf_load_addr;
>  };
>  
> +#ifdef CONFIG_CRASH_HOTPLUG
> +void arch_crash_hotplug_handler(struct kimage *image,
> +	unsigned int hp_action, unsigned int cpu);
> +#define KEXEC_CRASH_HP_REMOVE_CPU   0
> +#define KEXEC_CRASH_HP_ADD_CPU      1
> +#define KEXEC_CRASH_HP_REMOVE_MEMORY 2
> +#define KEXEC_CRASH_HP_ADD_MEMORY   3
> +#endif /* CONFIG_CRASH_HOTPLUG */
> +
>  /* kexec interface functions */
>  extern void machine_kexec(struct kimage *image);
>  extern int machine_kexec_prepare(struct kimage *image);
> diff --git a/kernel/crash_core.c b/kernel/crash_core.c
> index 256cf6db573c..ecf746243ab2 100644
> --- a/kernel/crash_core.c
> +++ b/kernel/crash_core.c
> @@ -9,12 +9,17 @@
>  #include <linux/init.h>
>  #include <linux/utsname.h>
>  #include <linux/vmalloc.h>
> +#include <linux/highmem.h>
> +#include <linux/memory.h>
> +#include <linux/cpuhotplug.h>
>  
>  #include <asm/page.h>
>  #include <asm/sections.h>
>  
>  #include <crypto/sha1.h>
>  
> +#include "kexec_internal.h"
> +
>  /* vmcoreinfo stuff */
>  unsigned char *vmcoreinfo_data;
>  size_t vmcoreinfo_size;
> @@ -491,3 +496,99 @@ static int __init crash_save_vmcoreinfo_init(void)
>  }
>  
>  subsys_initcall(crash_save_vmcoreinfo_init);
> +
> +#ifdef CONFIG_CRASH_HOTPLUG
> +void __weak arch_crash_hotplug_handler(struct kimage *image,
> +	unsigned int hp_action, unsigned int cpu)
> +{
> +	pr_warn("crash hp: %s not implemented", __func__);
> +}
> +
> +static void crash_hotplug_handler(unsigned int hp_action,
> +	unsigned int cpu)
> +{
> +	/* Obtain lock while changing crash information */
> +	if (!mutex_trylock(&kexec_mutex))
> +		return;
> +
> +	/* Check kdump is loaded */
> +	if (kexec_crash_image) {
> +		pr_debug("crash hp: hp_action %u, cpu %u", hp_action, cpu);
> +
> +		/* Needed in order for the segments to be updated */
> +		arch_kexec_unprotect_crashkres();
> +
> +		/* Flag to differentiate between normal load and hotplug */
> +		kexec_crash_image->hotplug_event = true;
> +
> +		/* Now invoke arch-specific update handler */
> +		arch_crash_hotplug_handler(kexec_crash_image, hp_action, cpu);
> +
> +		/* No longer handling a hotplug event */
> +		kexec_crash_image->hotplug_event = false;
> +
> +		/* Change back to read-only */
> +		arch_kexec_protect_crashkres();
> +	}
> +
> +	/* Release lock now that update complete */
> +	mutex_unlock(&kexec_mutex);
> +}
> +
> +#if defined(CONFIG_MEMORY_HOTPLUG)
> +static int crash_memhp_notifier(struct notifier_block *nb,
> +	unsigned long val, void *v)
> +{
> +	struct memory_notify *mhp = v;
> +
> +	switch (val) {
> +	case MEM_ONLINE:
> +		crash_hotplug_handler(KEXEC_CRASH_HP_ADD_MEMORY, -1U);
We don't differentiate the memory add/remove, cpu add, except of cpu
remove. Means the hp_action only differentiate cpu remove from the other
action. Maybe only making two types?

#define KEXEC_CRASH_HP_REMOVE_CPU   0
#define KEXEC_CRASH_HP_UPDATE_OTHER      1

And define a new macro to replace the magic number?

#define KEXEC_CRASH_HP_INVALID_CPU   -1U
> +		break;
> +
> +	case MEM_OFFLINE:
> +		crash_hotplug_handler(KEXEC_CRASH_HP_REMOVE_MEMORY, -1U);
> +		break;
> +	}
> +	return NOTIFY_OK;
> +}
> +
> +static struct notifier_block crash_memhp_nb = {
> +	.notifier_call = crash_memhp_notifier,
> +	.priority = 0
> +};
> +#endif
> +
> +#if defined(CONFIG_HOTPLUG_CPU)
> +static int crash_cpuhp_online(unsigned int cpu)
> +{
> +	crash_hotplug_handler(KEXEC_CRASH_HP_ADD_CPU, cpu);
 Should making the cpu as -1U?
 	crash_hotplug_handler(KEXEC_CRASH_HP_ADD_CPU, -1U);
> +	return 0;
> +}
> +
> +static int crash_cpuhp_offline(unsigned int cpu)
> +{
> +	crash_hotplug_handler(KEXEC_CRASH_HP_REMOVE_CPU, cpu);
> +	return 0;
> +}
> +#endif
> +
> +static int __init crash_hotplug_init(void)
> +{
> +	int result = 0;
> +
> +#if defined(CONFIG_MEMORY_HOTPLUG)
> +	register_memory_notifier(&crash_memhp_nb);
> +#endif
> +
> +#if defined(CONFIG_HOTPLUG_CPU)
> +	result = cpuhp_setup_state_nocalls(CPUHP_AP_ONLINE_DYN,
> +				"crash/cpuhp",
> +				crash_cpuhp_online, crash_cpuhp_offline);
> +#endif
> +
> +	return result;
> +}
> +
> +subsys_initcall(crash_hotplug_init);
> +#endif /* CONFIG_CRASH_HOTPLUG */
> -- 
> 2.27.0
>