diff mbox series

[v5,5/6] efi / ras: CCIX Link error reporting

Message ID 20191114133919.32290-6-Jonathan.Cameron@huawei.com (mailing list archive)
State New, archived
Headers show
Series CCIX Protocol error reporting. | expand

Commit Message

Jonathan Cameron Nov. 14, 2019, 1:39 p.m. UTC
The CCIX 1.0 Base Specification defines a protocol layer
CCIX link and related error reporting mechanism.
The UEFI 2.8 specification includes a CCIX CPER record for
firmware first handling to report these errors to the
operating system.

This patch is very similar to the support previously added
for CCIX Memory Errors and provides both logging and RAS
tracepoint for this error class.

Note that we currently don't print the raw message in the
trace point print functions.

Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
---
Changes since v4
* Fixed length array parameters
* Take advantage of being first print to string and simplify code.
* Fix buffer overflow risk in snprintf calls.
* Clarify comments on not printing the raw message for tracepoints.

drivers/acpi/apei/ghes.c         |   4 +
 drivers/firmware/efi/cper-ccix.c | 141 +++++++++++++++++++++++++++++++
 include/linux/cper.h             |  48 +++++++++++
 include/ras/ras_event.h          |  65 ++++++++++++++
 4 files changed, 258 insertions(+)
diff mbox series

Patch

diff --git a/drivers/acpi/apei/ghes.c b/drivers/acpi/apei/ghes.c
index 7455db97319c..22df8c14ec13 100644
--- a/drivers/acpi/apei/ghes.c
+++ b/drivers/acpi/apei/ghes.c
@@ -529,6 +529,10 @@  static void ghes_handle_ccix_per(struct acpi_hest_generic_data *gdata, int sev)
 		trace_ccix_port_error_event(payload, err_seq, sev,
 					    ccix_port_err_ven_len_get(payload));
 		break;
+	case CCIX_LINK_ERROR:
+		trace_ccix_link_error_event(payload, err_seq, sev,
+					    ccix_link_err_ven_len_get(payload));
+		break;
 	default:
 		/* Unknown error type */
 		pr_info("CCIX error of unknown or vendor defined type\n");
diff --git a/drivers/firmware/efi/cper-ccix.c b/drivers/firmware/efi/cper-ccix.c
index 59f8871b701f..ea17fb9140f9 100644
--- a/drivers/firmware/efi/cper-ccix.c
+++ b/drivers/firmware/efi/cper-ccix.c
@@ -508,6 +508,85 @@  static int cper_ccix_port_err_details(const char *pfx,
 	return 0;
 }
 
+static const char * const ccix_link_err_type_strs[] = {
+	"Generic Error",
+	"Credit Underflow",
+	"Credit Overflow",
+	"Unusable Credit Received",
+	"Link Credit Timeout",
+};
+
+static const char *cper_ccix_link_err_type_str(__u8 op)
+{
+	return op < ARRAY_SIZE(ccix_link_err_type_strs) ?
+		ccix_link_err_type_strs[op] : "Reserved";
+}
+
+static const char * const ccix_link_credit_type_strs[] = {
+	"Memory",
+	"Snoop",
+	"Data",
+	"Misc",
+};
+
+static const char *cper_ccix_link_credit_type_str(__u8 op)
+{
+	return op < ARRAY_SIZE(ccix_link_credit_type_strs) ?
+		ccix_link_credit_type_strs[op] : "Reserved";
+}
+
+static int cper_ccix_link_err_details(const char *pfx,
+				     struct acpi_hest_generic_data *gdata)
+{
+	struct cper_ccix_link_error *full_link_err;
+	struct cper_sec_ccix_link_error *link_err;
+	u16 vendor_data_len;
+	int i;
+
+	if (gdata->error_data_length < sizeof(*full_link_err))
+		return -ENOSPC;
+
+	full_link_err = acpi_hest_get_payload(gdata);
+
+	link_err = &full_link_err->link_record;
+
+	if (link_err->validation_bits & CCIX_LINK_ERR_TYPE_VALID)
+		printk("%s""Error Type: %s\n", pfx,
+		       cper_ccix_link_err_type_str(link_err->err_type));
+
+	if (link_err->validation_bits & CCIX_LINK_ERR_OP_VALID)
+		printk("%s""Operation: %s\n", pfx,
+		       cper_ccix_port_err_op_str(link_err->op_type));
+
+	if (link_err->validation_bits & CCIX_LINK_ERR_LINK_ID_VALID)
+		printk("%s""Link ID: %d\n", pfx, link_err->link_id);
+
+	if (link_err->validation_bits & CCIX_LINK_ERR_CREDIT_TYPE_VALID)
+		printk("%s""Credit Type: %s\n", pfx,
+		       cper_ccix_link_credit_type_str(link_err->credit_type));
+
+	if (link_err->validation_bits & CCIX_LINK_ERR_MESSAGE_VALID) {
+		for (i = 0; i < ARRAY_SIZE(link_err->message); i++)
+			printk("%s""Message%d: 0x%08x\n", pfx, i, link_err->message[i]);
+	}
+
+	if (link_err->validation_bits & CCIX_LINK_ERR_VENDOR_DATA_VALID) {
+		if (gdata->error_data_length < sizeof(*full_link_err) + 4)
+			return -ENOSPC;
+
+		vendor_data_len = link_err->vendor_data[0] & GENMASK(15, 0);
+		if (gdata->error_data_length < sizeof(*full_link_err) + vendor_data_len)
+			return -ENOSPC;
+
+		for (i = 0; i < vendor_data_len/4 - 1; i++)
+			printk("%s""Vendor%d: 0x%08x\n", pfx, i,
+			       link_err->vendor_data[i + 1]);
+
+	}
+
+	return 0;
+}
+
 int cper_print_ccix_per(const char *pfx, struct acpi_hest_generic_data *gdata)
 {
 	struct cper_sec_ccix_header *header = acpi_hest_get_payload(gdata);
@@ -575,6 +654,8 @@  int cper_print_ccix_per(const char *pfx, struct acpi_hest_generic_data *gdata)
 		return cper_ccix_atc_err_details(pfx, gdata);
 	case CCIX_PORT_ERROR:
 		return cper_ccix_port_err_details(pfx, gdata);
+	case CCIX_LINK_ERROR:
+		return cper_ccix_link_err_details(pfx, gdata);
 	default:
 		/* Vendor defined so no formatting be done */
 		break;
@@ -737,3 +818,63 @@  const char *cper_ccix_port_err_unpack(struct trace_seq *p,
 
 	return ret;
 }
+
+void cper_ccix_link_err_pack(const struct cper_sec_ccix_link_error *link_record,
+			     struct cper_ccix_link_err_compact *clink_err,
+			     const u16 vendor_data_len,
+			     u8 *vendor_data)
+{
+	clink_err->validation_bits = link_record->validation_bits;
+	clink_err->err_type = link_record->err_type;
+	clink_err->op_type = link_record->op_type;
+	clink_err->link_id = link_record->link_id;
+	clink_err->credit_type = link_record->credit_type;
+	memcpy(clink_err->message, link_record->message,
+	       sizeof(clink_err->message));
+
+	memcpy(vendor_data, &link_record->vendor_data[1], vendor_data_len);
+}
+
+static int cper_ccix_err_link_location(struct cper_ccix_link_err_compact *clink_err,
+				       char msg[CPER_REC_LEN])
+{
+	u32 len = CPER_REC_LEN - 1;
+	u32 n = 0;
+
+	if (!msg)
+		return 0;
+
+	if (clink_err->validation_bits & CCIX_LINK_ERR_TYPE_VALID)
+		n = snprintf(msg, len, "Error Type: %s ",
+			     cper_ccix_link_err_type_str(clink_err->err_type));
+
+
+	if (clink_err->validation_bits & CCIX_LINK_ERR_OP_VALID)
+		n += snprintf(msg + n, len - n, "Op: %s ",
+			      cper_ccix_port_err_op_str(clink_err->op_type));
+
+	if (clink_err->validation_bits & CCIX_LINK_ERR_LINK_ID_VALID)
+		n += snprintf(msg + n, len - n, "Link ID: %d ",
+			      clink_err->link_id);
+
+	if (clink_err->validation_bits & CCIX_LINK_ERR_CREDIT_TYPE_VALID)
+		n += snprintf(msg + n, len - n, "Credit Type: %s ",
+			      cper_ccix_link_credit_type_str(clink_err->credit_type));
+
+	/* Raw message not currently output */
+
+	return n;
+}
+
+const char *cper_ccix_link_err_unpack(struct trace_seq *p,
+				      struct cper_ccix_link_err_compact *clink_err)
+{
+	const char *ret = trace_seq_buffer_ptr(p);
+
+	if (cper_ccix_err_link_location(clink_err, rcd_decode_str))
+		trace_seq_printf(p, "%s", rcd_decode_str);
+
+	trace_seq_putc(p, '\0');
+
+	return ret;
+}
diff --git a/include/linux/cper.h b/include/linux/cper.h
index 5e315afc210e..d35be55351e3 100644
--- a/include/linux/cper.h
+++ b/include/linux/cper.h
@@ -742,6 +742,47 @@  struct cper_ccix_port_err_compact {
 	__u8	op_type;
 };
 
+struct cper_sec_ccix_link_error {
+	__u32	validation_bits;
+#define CCIX_LINK_ERR_OP_VALID			BIT(0)
+#define CCIX_LINK_ERR_TYPE_VALID		BIT(1)
+#define CCIX_LINK_ERR_LINK_ID_VALID		BIT(2)
+#define CCIX_LINK_ERR_CREDIT_TYPE_VALID		BIT(3)
+#define CCIX_LINK_ERR_MESSAGE_VALID		BIT(4)
+#define CCIX_LINK_ERR_VENDOR_DATA_VALID		BIT(5)
+	__u16	length; /* Includes vendor specific log info */
+	__u8	op_type;
+	__u8	err_type;
+	__u8	link_id;
+	__u8	credit_type;
+	__u16	reserved;
+	__u32	message[8];
+	__u32	vendor_data[];
+};
+
+struct cper_ccix_link_error {
+	struct cper_sec_ccix_header header;
+	__u32 ccix_header[CCIX_PER_LOG_HEADER_DWS];
+	struct cper_sec_ccix_link_error link_record;
+};
+
+static inline u16 ccix_link_err_ven_len_get(struct cper_ccix_link_error *link_err)
+{
+	if (link_err->link_record.validation_bits & CCIX_LINK_ERR_VENDOR_DATA_VALID)
+		return link_err->link_record.vendor_data[0] & 0xFFFF;
+	else
+		return 0;
+}
+
+struct cper_ccix_link_err_compact {
+	__u32	validation_bits;
+	__u32	message[8];
+	__u8	err_type;
+	__u8	op_type;
+	__u8	link_id;
+	__u8	credit_type;
+};
+
 /* Reset to default packing */
 #pragma pack()
 
@@ -787,6 +828,13 @@  void cper_ccix_port_err_pack(const struct cper_sec_ccix_port_error *port_record,
 const char *cper_ccix_port_err_unpack(struct trace_seq *p,
 				      struct cper_ccix_port_err_compact *cport_err);
 
+void cper_ccix_link_err_pack(const struct cper_sec_ccix_link_error *link_record,
+			     struct cper_ccix_link_err_compact *clink_err,
+			     const u16 vendor_data_len,
+			     u8 *vendor_data);
+const char *cper_ccix_link_err_unpack(struct trace_seq *p,
+				      struct cper_ccix_link_err_compact *clink_err);
+
 struct acpi_hest_generic_data;
 int cper_print_ccix_per(const char *pfx,
 			struct acpi_hest_generic_data *gdata);
diff --git a/include/ras/ras_event.h b/include/ras/ras_event.h
index b0c416382786..7cecfadb0b15 100644
--- a/include/ras/ras_event.h
+++ b/include/ras/ras_event.h
@@ -614,6 +614,71 @@  TRACE_EVENT(ccix_port_error_event,
 		__print_hex(__get_dynamic_array(vendor_data), __entry->vendor_data_length)
 	)
 );
+
+TRACE_EVENT(ccix_link_error_event,
+	TP_PROTO(struct cper_ccix_link_error *err,
+		 u32 err_seq,
+		 u8 sev, u16 ven_len),
+
+	TP_ARGS(err, err_seq, sev, ven_len),
+
+	TP_STRUCT__entry(
+		__field(u32, err_seq)
+		__field(u8, sev)
+		__field(u8, sevdetail)
+		__field(u8, source)
+		__field(u8, component)
+		__field(u64, pa)
+		__field(u8, pa_mask_lsb)
+		__field_struct(struct cper_ccix_link_err_compact, data)
+		__field(u16, vendor_data_length)
+		__dynamic_array(u8, vendor_data, ven_len)
+	),
+
+	TP_fast_assign(
+		__entry->err_seq = err_seq;
+
+		__entry->sev = sev;
+		__entry->sevdetail = FIELD_GET(CCIX_PER_LOG_DW1_SEV_UE_M |
+			CCIX_PER_LOG_DW1_SEV_NO_COMM_M |
+			CCIX_PER_LOG_DW1_SEV_DEGRADED_M |
+			CCIX_PER_LOG_DW1_SEV_DEFFERABLE_M,
+			err->ccix_header[1]);
+		if (err->header.validation_bits & 0x1)
+			__entry->source = err->header.source_id;
+		else
+			__entry->source = ~0;
+		__entry->component = FIELD_GET(CCIX_PER_LOG_DW1_COMP_TYPE_M,
+						   err->ccix_header[1]);
+		if (err->ccix_header[1] & CCIX_PER_LOG_DW1_ADDR_VAL_M) {
+			__entry->pa = (u64)err->ccix_header[2] << 32 |
+				(err->ccix_header[3] & 0xfffffffc);
+			__entry->pa_mask_lsb = err->ccix_header[4] & 0xff;
+		} else {
+			__entry->pa = ~0ull;
+			__entry->pa_mask_lsb = ~0;
+		}
+		/* Do not store the vendor data header length */
+		__entry->vendor_data_length = ven_len ? ven_len - 4 : 0;
+		cper_ccix_link_err_pack(&err->link_record, &__entry->data,
+					__entry->vendor_data_length,
+					__get_dynamic_array(vendor_data));
+	),
+
+	TP_printk("{%d} %s CCIX PER Link Error in %s SevUE:%d SevNoComm:%d SevDegraded:%d SevDeferred:%d physical addr: %016llx (mask: %x) %s vendor:%s",
+		__entry->err_seq,
+		cper_severity_str(__entry->sev),
+		cper_ccix_comp_type_str(__entry->component),
+		__entry->sevdetail & BIT(0) ? 1 : 0,
+		__entry->sevdetail & BIT(1) ? 1 : 0,
+		__entry->sevdetail & BIT(2) ? 1 : 0,
+		__entry->sevdetail & BIT(3) ? 1 : 0,
+		__entry->pa,
+		__entry->pa_mask_lsb,
+		cper_ccix_link_err_unpack(p, &__entry->data),
+		__print_hex(__get_dynamic_array(vendor_data), __entry->vendor_data_length)
+	)
+);
 #endif
 
 /*