Re: [PATCH] vmcore_info: expose hardware error recovery statistics via sysfs

[Baoquan He <bhe@redhat.com>]

On 01/29/26 at 05:34am, Breno Leitao wrote:
> Add a sysfs file at /sys/kernel/vmcore_stats and expose hardware error
> recovery statistics that are already tracked by the kernel. This allows
> userspace monitoring tools to track recovered hardware errors without
> requiring kernel crashes.

I don't understand. If w/o requring kernel crashes, why do you call it
vmcore_stats? It's a normal showing of hardware error recovery
statistics tracked by kernel, can we name it /sys/kernel/hwerr_stats?
It's obviously having nothiing to do with vmcore, isn't it?

> 
> This is useful to track recoverable hardware errors in a time series,
> even if the host doesn't crash.
> 
> Create a generic vmcore_stats sysfs, and add a section for
> hwerr_recovery that shows the counts per subsystem and timestamps:
> 
>   - cpu: CPU-related errors (MCE, ARM processor errors)
>   - memory: Memory-related errors
>   - pci: PCI/PCIe AER non-fatal errors
>   - cxl: CXL errors
>   - other: Other hardware errors
> 
> Example output:
>   hwerr_recovery:
>     cpu: 0 (0)
>     memory: 2 (1738148257)
>     pci: 1 (1738147000)
>     cxl: 0 (0)
>     other: 0 (0)
> 
> The value in parentheses is the timestamp (seconds since epoch) of the
> last error of that type, or 0 if no errors have occurred.
> 
> These statistics provide visibility into the health of the system's
> hardware and can be used by system administrators to proactively detect
> failing components before they cause system crashes.
> 
> Signed-off-by: Breno Leitao <leitao@debian.org>
> ---
> To: akpm@linux-foundation.org
> Cc: kexec@lists.infradead.org
> To: bhe@redhat.com
> Cc: linux-kernel@vger.kernel.org
> Cc: dyoung@redhat.com
> Cc: tony.luck@intel.com
> Cc: xueshuai@linux.alibaba.com
> Cc: vgoyal@redhat.com
> Cc: zhiquan1.li@intel.com
> Cc: olja@meta.com
> ---
>  .../ABI/testing/sysfs-kernel-vmcore_stats          | 23 ++++++++++++++++
>  kernel/vmcore_info.c                               | 31 ++++++++++++++++++++++
>  2 files changed, 54 insertions(+)
> 
> diff --git a/Documentation/ABI/testing/sysfs-kernel-vmcore_stats b/Documentation/ABI/testing/sysfs-kernel-vmcore_stats
> new file mode 100644
> index 0000000000000..b42f18d24c00b
> --- /dev/null
> +++ b/Documentation/ABI/testing/sysfs-kernel-vmcore_stats
> @@ -0,0 +1,23 @@
> +What:		/sys/kernel/vmcore_stats
> +Date:		January 2026
> +KernelVersion:	6.20
> +Contact:	Breno Leitao <leitao@debian.org>
> +Description:
> +		Shows statistics related to vmcore functionality. Currently
> +		includes hardware error recovery statistics.
> +
> +		Format:
> +		  Recovered hardware errors:
> +		    metric: count (timestamp)
> +
> +		Statistics about recoverable hardware errors that the kernel
> +		has handled since boot. Each metric shows the count and
> +		timestamp (seconds since epoch) of the last error in
> +		parentheses (0 if no errors have occurred).
> +
> +		Metrics:
> +		    - cpu: CPU-related errors (MCE, ARM processor errors)
> +		    - memory: Memory-related errors
> +		    - pci: PCI/PCIe AER non-fatal errors
> +		    - cxl: CXL (Compute Express Link) errors
> +		    - other: Other hardware errors
> diff --git a/kernel/vmcore_info.c b/kernel/vmcore_info.c
> index fe9bf8db1922e..5974b4be08cbc 100644
> --- a/kernel/vmcore_info.c
> +++ b/kernel/vmcore_info.c
> @@ -6,6 +6,8 @@
>  
>  #include <linux/buildid.h>
>  #include <linux/init.h>
> +#include <linux/kobject.h>
> +#include <linux/sysfs.h>
>  #include <linux/utsname.h>
>  #include <linux/vmalloc.h>
>  #include <linux/sizes.h>
> @@ -135,6 +137,31 @@ void hwerr_log_error_type(enum hwerr_error_type src)
>  }
>  EXPORT_SYMBOL_GPL(hwerr_log_error_type);
>  
> +/* sysfs interface for hardware error recovery statistics */
> +static ssize_t vmcore_stats_show(struct kobject *kobj,
> +				 struct kobj_attribute *attr, char *buf)
> +{
> +	return sysfs_emit(buf,
> +			  "Recovered hardware errors:\n"
> +			  "  cpu: %d (%lld)\n"
> +			  "  memory: %d (%lld)\n"
> +			  "  pci: %d (%lld)\n"
> +			  "  cxl: %d (%lld)\n"
> +			  "  other: %d (%lld)\n",
> +			  atomic_read(&hwerr_data[HWERR_RECOV_CPU].count),
> +			  (long long)READ_ONCE(hwerr_data[HWERR_RECOV_CPU].timestamp),
> +			  atomic_read(&hwerr_data[HWERR_RECOV_MEMORY].count),
> +			  (long long)READ_ONCE(hwerr_data[HWERR_RECOV_MEMORY].timestamp),
> +			  atomic_read(&hwerr_data[HWERR_RECOV_PCI].count),
> +			  (long long)READ_ONCE(hwerr_data[HWERR_RECOV_PCI].timestamp),
> +			  atomic_read(&hwerr_data[HWERR_RECOV_CXL].count),
> +			  (long long)READ_ONCE(hwerr_data[HWERR_RECOV_CXL].timestamp),
> +			  atomic_read(&hwerr_data[HWERR_RECOV_OTHERS].count),
> +			  (long long)READ_ONCE(hwerr_data[HWERR_RECOV_OTHERS].timestamp));
> +}
> +
> +static struct kobj_attribute vmcore_stats_attr = __ATTR_RO(vmcore_stats);
> +
>  static int __init crash_save_vmcoreinfo_init(void)
>  {
>  	vmcoreinfo_data = (unsigned char *)get_zeroed_page(GFP_KERNEL);
> @@ -244,6 +271,10 @@ static int __init crash_save_vmcoreinfo_init(void)
>  	arch_crash_save_vmcoreinfo();
>  	update_vmcoreinfo_note();
>  
> +	/* Create /sys/kernel/vmcore_stats */
> +	if (sysfs_create_file(kernel_kobj, &vmcore_stats_attr.attr))
> +		pr_warn("Failed to create vmcore_stats sysfs file\n");
> +
>  	return 0;
>  }
>  
> 
> ---
> base-commit: 8dfce8991b95d8625d0a1d2896e42f93b9d7f68d
> change-id: 20260129-vmcoreinfo_sysfs-ff4687979cd5
> 
> Best regards,
> --  
> Breno Leitao <leitao@debian.org>
>