[PATCH v4] trace: ras: add ARM processor error information trace event

Fri Jun 23 06:42:07 PDT 2017

On Fri, 23 Jun 2017 19:13:43 +0800
Xie XiuQi <xiexiuqi at huawei.com> wrote:

> Add a new trace event for ARM processor error information, so that
> the user will know what error occurred. With this information the
> user may take appropriate action.
> 
> These trace events are consistent with the ARM processor error
> information table which defined in UEFI 2.6 spec section N.2.4.4.1.
> 
> ---
> v4: use __print_flags instead of __print_symbolic, because ARM_PROC_ERR_FLAGS
>     might have more than on bit set.
>     setting up default values for __entry to avoid a lot of else branches.
>     set flags to 0 by default instead of ~0.
>     fix a typo
>     rename arm_proc_err to arm_proc_err_event
>     remove "ARM Processor Error: " prefix
>     rebase on Tyler's patchset v17 "Add UEFI 2.6 and ACPI 6.1 updates for RAS on ARM64"
> 
> v3: no change
> 
> v2: add trace enabled condition as Steven's suggestion.

Where's the trace enabled now?

>     fix a typo.
> 
> https://patchwork.kernel.org/patch/9653767/
> ---
> 
> Cc: Steven Rostedt <rostedt at goodmis.org>
> Cc: Tyler Baicar <tbaicar at codeaurora.org>
> Signed-off-by: Xie XiuQi <xiexiuqi at huawei.com>
> ---
>  drivers/ras/ras.c       |  8 +++++
>  include/linux/cper.h    |  5 ++++
>  include/ras/ras_event.h | 79 +++++++++++++++++++++++++++++++++++++++++++++++++
>  3 files changed, 92 insertions(+)
> 
> diff --git a/drivers/ras/ras.c b/drivers/ras/ras.c
> index 39701a5..785e25d 100644
> --- a/drivers/ras/ras.c
> +++ b/drivers/ras/ras.c
> @@ -22,7 +22,14 @@ void log_non_standard_event(const uuid_le *sec_type, const uuid_le *fru_id,
>  
>  void log_arm_hw_error(struct cper_sec_proc_arm *err)
>  {
> +	int i;
> +	struct cper_arm_err_info *err_info;
> +
>  	trace_arm_event(err);

	if (!trace_arm_err_info_event_enabled())
		return;

?

> +
> +	err_info = (struct cper_arm_err_info *)(err + 1);
> +	for (i = 0; i < err->err_info_num; i++, err_info++)
> +		trace_arm_err_info_event(err_info);
>  }
>  
>  static int __init ras_init(void)
> @@ -42,6 +49,7 @@ static int __init ras_init(void)
>  EXPORT_TRACEPOINT_SYMBOL_GPL(mc_event);
>  EXPORT_TRACEPOINT_SYMBOL_GPL(non_standard_event);
>  EXPORT_TRACEPOINT_SYMBOL_GPL(arm_event);
> +EXPORT_TRACEPOINT_SYMBOL_GPL(arm_err_info_event);
>  
>  int __init parse_ras_param(char *str)
>  {
> diff --git a/include/linux/cper.h b/include/linux/cper.h
> index 4c671fc..17546bf 100644
> --- a/include/linux/cper.h
> +++ b/include/linux/cper.h
> @@ -275,6 +275,11 @@ enum {
>  #define CPER_ARM_INFO_FLAGS_PROPAGATED		BIT(2)
>  #define CPER_ARM_INFO_FLAGS_OVERFLOW		BIT(3)
>  
> +#define CPER_ARM_INFO_TYPE_CACHE		0
> +#define CPER_ARM_INFO_TYPE_TLB			1
> +#define CPER_ARM_INFO_TYPE_BUS			2
> +#define CPER_ARM_INFO_TYPE_UARCH		3
> +
>  /*
>   * All tables and structs must be byte-packed to match CPER
>   * specification, since the tables are provided by the system BIOS
> diff --git a/include/ras/ras_event.h b/include/ras/ras_event.h
> index 429f46f..c38a367 100644
> --- a/include/ras/ras_event.h
> +++ b/include/ras/ras_event.h
> @@ -206,6 +206,85 @@
>  		  __entry->running_state, __entry->psci_state)
>  );
>  
> +#define ARM_PROC_ERR_TYPE	\
> +	EM ( CPER_ARM_INFO_TYPE_CACHE, "cache error" )	\
> +	EM ( CPER_ARM_INFO_TYPE_TLB,  "TLB error" )	\
> +	EM ( CPER_ARM_INFO_TYPE_BUS, "bus error" )	\
> +	EMe ( CPER_ARM_INFO_TYPE_UARCH, "micro-architectural error" )
> +
> +/*
> + * First define the enums in MM_ACTION_RESULT to be exported to userspace
> + * via TRACE_DEFINE_ENUM().
> + */
> +#undef EM
> +#undef EMe
> +#define EM(a, b) TRACE_DEFINE_ENUM(a);
> +#define EMe(a, b)	TRACE_DEFINE_ENUM(a);
> +
> +ARM_PROC_ERR_TYPE
> +
> +/*
> + * Now redefine the EM() and EMe() macros to map the enums to the strings
> + * that will be printed in the output.
> + */
> +#undef EM
> +#undef EMe
> +#define EM(a, b)		{ a, b },
> +#define EMe(a, b)	{ a, b }
> +
> +#define show_proc_err_flags(flags) __print_flags(flags, "|",			\
> +	{ CPER_ARM_INFO_FLAGS_FIRST,		"First error captured" },	\
> +	{ CPER_ARM_INFO_FLAGS_LAST,		"Last error captured" },	\
> +	{ CPER_ARM_INFO_FLAGS_PROPAGATED,	"Propagated" },			\
> +	{ CPER_ARM_INFO_FLAGS_OVERFLOW,		"Overflow" })
> +
> +TRACE_EVENT(arm_err_info_event,
> +
> +	TP_PROTO(const struct cper_arm_err_info *err),
> +
> +	TP_ARGS(err),
> +
> +	TP_STRUCT__entry(
> +		__field(u8, type)
> +		__field(u16, multiple_error)
> +		__field(u8, flags)

Wouldn't you want flags after type? To keep multiple_error on a 2 byte
boundary.

-- Steve

> +		__field(u64, error_info)
> +		__field(u64, virt_fault_addr)
> +		__field(u64, physical_fault_addr)
> +	),
> +
> +	TP_fast_assign(
> +		__entry->type = err->type;
> +		__entry->multiple_error = ~0;
> +		memset(&__entry->flags, 0,
> +			sizeof(*__entry) - offsetof(typeof(*__entry), flags));
> +
> +		if (err->validation_bits & CPER_ARM_INFO_VALID_MULTI_ERR)
> +			__entry->multiple_error = err->multiple_error;
> +
> +		if (err->validation_bits & CPER_ARM_INFO_VALID_FLAGS)
> +			__entry->flags = err->flags;
> +
> +		if (err->validation_bits & CPER_ARM_INFO_VALID_ERR_INFO)
> +			__entry->error_info = err->error_info;
> +
> +		if (err->validation_bits & CPER_ARM_INFO_VALID_VIRT_ADDR)
> +			__entry->virt_fault_addr = err->virt_fault_addr;
> +
> +		if (err->validation_bits & CPER_ARM_INFO_VALID_PHYSICAL_ADDR)
> +			__entry->physical_fault_addr = err->physical_fault_addr;
> +	),
> +
> +	TP_printk("type: %s; count: %u; flags: %s;"
> +		  " error info: %016llx; virtual address: %016llx;"
> +		  " physical address: %016llx",
> +		  __print_symbolic(__entry->type, ARM_PROC_ERR_TYPE),
> +		  __entry->multiple_error,
> +		  show_proc_err_flags(__entry->flags),
> +		  __entry->error_info, __entry->virt_fault_addr,
> +		  __entry->physical_fault_addr)
> +);
> +
>  /*
>   * Non-Standard Section Report
>   *