1 /* SPDX-License-Identifier: GPL-2.0-only */
3 * UEFI Common Platform Error Record
5 * Copyright (C) 2010, Intel Corp.
6 * Author: Huang Ying <ying.huang@intel.com>
12 #include <linux/uuid.h>
13 #include <linux/trace_seq.h>
15 /* CPER record signature and the size */
16 #define CPER_SIG_RECORD "CPER"
17 #define CPER_SIG_SIZE 4
18 /* Used in signature_end field in struct cper_record_header */
19 #define CPER_SIG_END 0xffffffff
22 * CPER record header revision, used in revision field in struct
25 #define CPER_RECORD_REV 0x0100
28 * CPER record length contains the CPER fields which are relevant for further
29 * handling of a memory error in userspace (we don't carry all the fields
30 * defined in the UEFI spec because some of them don't make any sense.)
31 * Currently, a length of 256 should be more than enough.
33 #define CPER_REC_LEN 256
35 * Severity definition for error_severity in struct cper_record_header
36 * and section_severity in struct cper_section_descriptor
42 CPER_SEV_INFORMATIONAL,
46 * Validation bits definition for validation_bits in struct
47 * cper_record_header. If set, corresponding fields in struct
48 * cper_record_header contain valid information.
50 #define CPER_VALID_PLATFORM_ID 0x0001
51 #define CPER_VALID_TIMESTAMP 0x0002
52 #define CPER_VALID_PARTITION_ID 0x0004
55 * Notification type used to generate error record, used in
56 * notification_type in struct cper_record_header. These UUIDs are defined
57 * in the UEFI spec v2.7, sec N.2.1.
60 /* Corrected Machine Check */
61 #define CPER_NOTIFY_CMC \
62 GUID_INIT(0x2DCE8BB1, 0xBDD7, 0x450e, 0xB9, 0xAD, 0x9C, 0xF4, \
63 0xEB, 0xD4, 0xF8, 0x90)
64 /* Corrected Platform Error */
65 #define CPER_NOTIFY_CPE \
66 GUID_INIT(0x4E292F96, 0xD843, 0x4a55, 0xA8, 0xC2, 0xD4, 0x81, \
67 0xF2, 0x7E, 0xBE, 0xEE)
68 /* Machine Check Exception */
69 #define CPER_NOTIFY_MCE \
70 GUID_INIT(0xE8F56FFE, 0x919C, 0x4cc5, 0xBA, 0x88, 0x65, 0xAB, \
71 0xE1, 0x49, 0x13, 0xBB)
72 /* PCI Express Error */
73 #define CPER_NOTIFY_PCIE \
74 GUID_INIT(0xCF93C01F, 0x1A16, 0x4dfc, 0xB8, 0xBC, 0x9C, 0x4D, \
75 0xAF, 0x67, 0xC1, 0x04)
76 /* INIT Record (for IPF) */
77 #define CPER_NOTIFY_INIT \
78 GUID_INIT(0xCC5263E8, 0x9308, 0x454a, 0x89, 0xD0, 0x34, 0x0B, \
79 0xD3, 0x9B, 0xC9, 0x8E)
80 /* Non-Maskable Interrupt */
81 #define CPER_NOTIFY_NMI \
82 GUID_INIT(0x5BAD89FF, 0xB7E6, 0x42c9, 0x81, 0x4A, 0xCF, 0x24, \
83 0x85, 0xD6, 0xE9, 0x8A)
84 /* BOOT Error Record */
85 #define CPER_NOTIFY_BOOT \
86 GUID_INIT(0x3D61A466, 0xAB40, 0x409a, 0xA6, 0x98, 0xF3, 0x62, \
87 0xD4, 0x64, 0xB3, 0x8F)
88 /* DMA Remapping Error */
89 #define CPER_NOTIFY_DMAR \
90 GUID_INIT(0x667DD791, 0xC6B3, 0x4c27, 0x8A, 0x6B, 0x0F, 0x8E, \
91 0x72, 0x2D, 0xEB, 0x41)
93 /* CXL Event record UUIDs are formatted as GUIDs and reported in section type */
95 * General Media Event Record
96 * CXL rev 3.0 Section 8.2.9.2.1.1; Table 8-43
98 #define CPER_SEC_CXL_GEN_MEDIA_GUID \
99 GUID_INIT(0xfbcd0a77, 0xc260, 0x417f, \
100 0x85, 0xa9, 0x08, 0x8b, 0x16, 0x21, 0xeb, 0xa6)
103 * CXL rev 3.0 section 8.2.9.2.1.2; Table 8-44
105 #define CPER_SEC_CXL_DRAM_GUID \
106 GUID_INIT(0x601dcbb3, 0x9c06, 0x4eab, \
107 0xb8, 0xaf, 0x4e, 0x9b, 0xfb, 0x5c, 0x96, 0x24)
109 * Memory Module Event Record
110 * CXL rev 3.0 section 8.2.9.2.1.3; Table 8-45
112 #define CPER_SEC_CXL_MEM_MODULE_GUID \
113 GUID_INIT(0xfe927475, 0xdd59, 0x4339, \
114 0xa5, 0x86, 0x79, 0xba, 0xb1, 0x13, 0xb7, 0x74)
117 * Flags bits definitions for flags in struct cper_record_header
118 * If set, the error has been recovered
120 #define CPER_HW_ERROR_FLAGS_RECOVERED 0x1
121 /* If set, the error is for previous boot */
122 #define CPER_HW_ERROR_FLAGS_PREVERR 0x2
123 /* If set, the error is injected for testing */
124 #define CPER_HW_ERROR_FLAGS_SIMULATED 0x4
127 * CPER section header revision, used in revision field in struct
128 * cper_section_descriptor
130 #define CPER_SEC_REV 0x0100
133 * Validation bits definition for validation_bits in struct
134 * cper_section_descriptor. If set, corresponding fields in struct
135 * cper_section_descriptor contain valid information.
137 #define CPER_SEC_VALID_FRU_ID 0x1
138 #define CPER_SEC_VALID_FRU_TEXT 0x2
141 * Flags bits definitions for flags in struct cper_section_descriptor
143 * If set, the section is associated with the error condition
144 * directly, and should be focused on
146 #define CPER_SEC_PRIMARY 0x0001
148 * If set, the error was not contained within the processor or memory
149 * hierarchy and the error may have propagated to persistent storage
152 #define CPER_SEC_CONTAINMENT_WARNING 0x0002
153 /* If set, the component must be re-initialized or re-enabled prior to use */
154 #define CPER_SEC_RESET 0x0004
155 /* If set, Linux may choose to discontinue use of the resource */
156 #define CPER_SEC_ERROR_THRESHOLD_EXCEEDED 0x0008
158 * If set, resource could not be queried for error information due to
159 * conflicts with other system software or resources. Some fields of
160 * the section will be invalid
162 #define CPER_SEC_RESOURCE_NOT_ACCESSIBLE 0x0010
164 * If set, action has been taken to ensure error containment (such as
165 * poisoning data), but the error has not been fully corrected and the
166 * data has not been consumed. Linux may choose to take further
167 * corrective action before the data is consumed
169 #define CPER_SEC_LATENT_ERROR 0x0020
172 * Section type definitions, used in section_type field in struct
173 * cper_section_descriptor. These UUIDs are defined in the UEFI spec
177 /* Processor Generic */
178 #define CPER_SEC_PROC_GENERIC \
179 GUID_INIT(0x9876CCAD, 0x47B4, 0x4bdb, 0xB6, 0x5E, 0x16, 0xF1, \
180 0x93, 0xC4, 0xF3, 0xDB)
181 /* Processor Specific: X86/X86_64 */
182 #define CPER_SEC_PROC_IA \
183 GUID_INIT(0xDC3EA0B0, 0xA144, 0x4797, 0xB9, 0x5B, 0x53, 0xFA, \
184 0x24, 0x2B, 0x6E, 0x1D)
185 /* Processor Specific: IA64 */
186 #define CPER_SEC_PROC_IPF \
187 GUID_INIT(0xE429FAF1, 0x3CB7, 0x11D4, 0x0B, 0xCA, 0x07, 0x00, \
188 0x80, 0xC7, 0x3C, 0x88, 0x81)
189 /* Processor Specific: ARM */
190 #define CPER_SEC_PROC_ARM \
191 GUID_INIT(0xE19E3D16, 0xBC11, 0x11E4, 0x9C, 0xAA, 0xC2, 0x05, \
192 0x1D, 0x5D, 0x46, 0xB0)
193 /* Platform Memory */
194 #define CPER_SEC_PLATFORM_MEM \
195 GUID_INIT(0xA5BC1114, 0x6F64, 0x4EDE, 0xB8, 0x63, 0x3E, 0x83, \
196 0xED, 0x7C, 0x83, 0xB1)
197 #define CPER_SEC_PCIE \
198 GUID_INIT(0xD995E954, 0xBBC1, 0x430F, 0xAD, 0x91, 0xB4, 0x4D, \
199 0xCB, 0x3C, 0x6F, 0x35)
200 /* Firmware Error Record Reference */
201 #define CPER_SEC_FW_ERR_REC_REF \
202 GUID_INIT(0x81212A96, 0x09ED, 0x4996, 0x94, 0x71, 0x8D, 0x72, \
203 0x9C, 0x8E, 0x69, 0xED)
205 #define CPER_SEC_PCI_X_BUS \
206 GUID_INIT(0xC5753963, 0x3B84, 0x4095, 0xBF, 0x78, 0xED, 0xDA, \
207 0xD3, 0xF9, 0xC9, 0xDD)
208 /* PCI Component/Device */
209 #define CPER_SEC_PCI_DEV \
210 GUID_INIT(0xEB5E4685, 0xCA66, 0x4769, 0xB6, 0xA2, 0x26, 0x06, \
211 0x8B, 0x00, 0x13, 0x26)
212 #define CPER_SEC_DMAR_GENERIC \
213 GUID_INIT(0x5B51FEF7, 0xC79D, 0x4434, 0x8F, 0x1B, 0xAA, 0x62, \
214 0xDE, 0x3E, 0x2C, 0x64)
215 /* Intel VT for Directed I/O specific DMAr */
216 #define CPER_SEC_DMAR_VT \
217 GUID_INIT(0x71761D37, 0x32B2, 0x45cd, 0xA7, 0xD0, 0xB0, 0xFE, \
218 0xDD, 0x93, 0xE8, 0xCF)
219 /* IOMMU specific DMAr */
220 #define CPER_SEC_DMAR_IOMMU \
221 GUID_INIT(0x036F84E1, 0x7F37, 0x428c, 0xA7, 0x9E, 0x57, 0x5F, \
222 0xDF, 0xAA, 0x84, 0xEC)
224 #define CPER_PROC_VALID_TYPE 0x0001
225 #define CPER_PROC_VALID_ISA 0x0002
226 #define CPER_PROC_VALID_ERROR_TYPE 0x0004
227 #define CPER_PROC_VALID_OPERATION 0x0008
228 #define CPER_PROC_VALID_FLAGS 0x0010
229 #define CPER_PROC_VALID_LEVEL 0x0020
230 #define CPER_PROC_VALID_VERSION 0x0040
231 #define CPER_PROC_VALID_BRAND_INFO 0x0080
232 #define CPER_PROC_VALID_ID 0x0100
233 #define CPER_PROC_VALID_TARGET_ADDRESS 0x0200
234 #define CPER_PROC_VALID_REQUESTOR_ID 0x0400
235 #define CPER_PROC_VALID_RESPONDER_ID 0x0800
236 #define CPER_PROC_VALID_IP 0x1000
238 #define CPER_MEM_VALID_ERROR_STATUS 0x0001
239 #define CPER_MEM_VALID_PA 0x0002
240 #define CPER_MEM_VALID_PA_MASK 0x0004
241 #define CPER_MEM_VALID_NODE 0x0008
242 #define CPER_MEM_VALID_CARD 0x0010
243 #define CPER_MEM_VALID_MODULE 0x0020
244 #define CPER_MEM_VALID_BANK 0x0040
245 #define CPER_MEM_VALID_DEVICE 0x0080
246 #define CPER_MEM_VALID_ROW 0x0100
247 #define CPER_MEM_VALID_COLUMN 0x0200
248 #define CPER_MEM_VALID_BIT_POSITION 0x0400
249 #define CPER_MEM_VALID_REQUESTOR_ID 0x0800
250 #define CPER_MEM_VALID_RESPONDER_ID 0x1000
251 #define CPER_MEM_VALID_TARGET_ID 0x2000
252 #define CPER_MEM_VALID_ERROR_TYPE 0x4000
253 #define CPER_MEM_VALID_RANK_NUMBER 0x8000
254 #define CPER_MEM_VALID_CARD_HANDLE 0x10000
255 #define CPER_MEM_VALID_MODULE_HANDLE 0x20000
256 #define CPER_MEM_VALID_ROW_EXT 0x40000
257 #define CPER_MEM_VALID_BANK_GROUP 0x80000
258 #define CPER_MEM_VALID_BANK_ADDRESS 0x100000
259 #define CPER_MEM_VALID_CHIP_ID 0x200000
261 #define CPER_MEM_EXT_ROW_MASK 0x3
262 #define CPER_MEM_EXT_ROW_SHIFT 16
264 #define CPER_MEM_BANK_ADDRESS_MASK 0xff
265 #define CPER_MEM_BANK_GROUP_SHIFT 8
267 #define CPER_MEM_CHIP_ID_SHIFT 5
269 #define CPER_PCIE_VALID_PORT_TYPE 0x0001
270 #define CPER_PCIE_VALID_VERSION 0x0002
271 #define CPER_PCIE_VALID_COMMAND_STATUS 0x0004
272 #define CPER_PCIE_VALID_DEVICE_ID 0x0008
273 #define CPER_PCIE_VALID_SERIAL_NUMBER 0x0010
274 #define CPER_PCIE_VALID_BRIDGE_CONTROL_STATUS 0x0020
275 #define CPER_PCIE_VALID_CAPABILITY 0x0040
276 #define CPER_PCIE_VALID_AER_INFO 0x0080
278 #define CPER_PCIE_SLOT_SHIFT 3
280 #define CPER_ARM_VALID_MPIDR BIT(0)
281 #define CPER_ARM_VALID_AFFINITY_LEVEL BIT(1)
282 #define CPER_ARM_VALID_RUNNING_STATE BIT(2)
283 #define CPER_ARM_VALID_VENDOR_INFO BIT(3)
285 #define CPER_ARM_INFO_VALID_MULTI_ERR BIT(0)
286 #define CPER_ARM_INFO_VALID_FLAGS BIT(1)
287 #define CPER_ARM_INFO_VALID_ERR_INFO BIT(2)
288 #define CPER_ARM_INFO_VALID_VIRT_ADDR BIT(3)
289 #define CPER_ARM_INFO_VALID_PHYSICAL_ADDR BIT(4)
291 #define CPER_ARM_INFO_FLAGS_FIRST BIT(0)
292 #define CPER_ARM_INFO_FLAGS_LAST BIT(1)
293 #define CPER_ARM_INFO_FLAGS_PROPAGATED BIT(2)
294 #define CPER_ARM_INFO_FLAGS_OVERFLOW BIT(3)
296 #define CPER_ARM_CACHE_ERROR 0
297 #define CPER_ARM_TLB_ERROR 1
298 #define CPER_ARM_BUS_ERROR 2
299 #define CPER_ARM_VENDOR_ERROR 3
300 #define CPER_ARM_MAX_TYPE CPER_ARM_VENDOR_ERROR
302 #define CPER_ARM_ERR_VALID_TRANSACTION_TYPE BIT(0)
303 #define CPER_ARM_ERR_VALID_OPERATION_TYPE BIT(1)
304 #define CPER_ARM_ERR_VALID_LEVEL BIT(2)
305 #define CPER_ARM_ERR_VALID_PROC_CONTEXT_CORRUPT BIT(3)
306 #define CPER_ARM_ERR_VALID_CORRECTED BIT(4)
307 #define CPER_ARM_ERR_VALID_PRECISE_PC BIT(5)
308 #define CPER_ARM_ERR_VALID_RESTARTABLE_PC BIT(6)
309 #define CPER_ARM_ERR_VALID_PARTICIPATION_TYPE BIT(7)
310 #define CPER_ARM_ERR_VALID_TIME_OUT BIT(8)
311 #define CPER_ARM_ERR_VALID_ADDRESS_SPACE BIT(9)
312 #define CPER_ARM_ERR_VALID_MEM_ATTRIBUTES BIT(10)
313 #define CPER_ARM_ERR_VALID_ACCESS_MODE BIT(11)
315 #define CPER_ARM_ERR_TRANSACTION_SHIFT 16
316 #define CPER_ARM_ERR_TRANSACTION_MASK GENMASK(1,0)
317 #define CPER_ARM_ERR_OPERATION_SHIFT 18
318 #define CPER_ARM_ERR_OPERATION_MASK GENMASK(3,0)
319 #define CPER_ARM_ERR_LEVEL_SHIFT 22
320 #define CPER_ARM_ERR_LEVEL_MASK GENMASK(2,0)
321 #define CPER_ARM_ERR_PC_CORRUPT_SHIFT 25
322 #define CPER_ARM_ERR_PC_CORRUPT_MASK GENMASK(0,0)
323 #define CPER_ARM_ERR_CORRECTED_SHIFT 26
324 #define CPER_ARM_ERR_CORRECTED_MASK GENMASK(0,0)
325 #define CPER_ARM_ERR_PRECISE_PC_SHIFT 27
326 #define CPER_ARM_ERR_PRECISE_PC_MASK GENMASK(0,0)
327 #define CPER_ARM_ERR_RESTARTABLE_PC_SHIFT 28
328 #define CPER_ARM_ERR_RESTARTABLE_PC_MASK GENMASK(0,0)
329 #define CPER_ARM_ERR_PARTICIPATION_TYPE_SHIFT 29
330 #define CPER_ARM_ERR_PARTICIPATION_TYPE_MASK GENMASK(1,0)
331 #define CPER_ARM_ERR_TIME_OUT_SHIFT 31
332 #define CPER_ARM_ERR_TIME_OUT_MASK GENMASK(0,0)
333 #define CPER_ARM_ERR_ADDRESS_SPACE_SHIFT 32
334 #define CPER_ARM_ERR_ADDRESS_SPACE_MASK GENMASK(1,0)
335 #define CPER_ARM_ERR_MEM_ATTRIBUTES_SHIFT 34
336 #define CPER_ARM_ERR_MEM_ATTRIBUTES_MASK GENMASK(8,0)
337 #define CPER_ARM_ERR_ACCESS_MODE_SHIFT 43
338 #define CPER_ARM_ERR_ACCESS_MODE_MASK GENMASK(0,0)
341 * All tables and structs must be byte-packed to match CPER
342 * specification, since the tables are provided by the system BIOS
346 /* Record Header, UEFI v2.7 sec N.2.1 */
347 struct cper_record_header {
348 char signature[CPER_SIG_SIZE]; /* must be CPER_SIG_RECORD */
349 u16 revision; /* must be CPER_RECORD_REV */
350 u32 signature_end; /* must be CPER_SIG_END */
359 guid_t notification_type;
362 u64 persistence_information;
363 u8 reserved[12]; /* must be zero */
366 /* Section Descriptor, UEFI v2.7 sec N.2.2 */
367 struct cper_section_descriptor {
368 u32 section_offset; /* Offset in bytes of the
369 * section body from the base
370 * of the record header */
372 u16 revision; /* must be CPER_RECORD_REV */
374 u8 reserved; /* must be zero */
378 u32 section_severity;
382 /* Generic Processor Error Section, UEFI v2.7 sec N.2.4.1 */
383 struct cper_sec_proc_generic {
401 /* IA32/X64 Processor Error Section, UEFI v2.7 sec N.2.4.2 */
402 struct cper_sec_proc_ia {
408 /* IA32/X64 Processor Error Information Structure, UEFI v2.7 sec N.2.4.2.1 */
409 struct cper_ia_err_info {
419 /* IA32/X64 Processor Context Information Structure, UEFI v2.7 sec N.2.4.2.2 */
420 struct cper_ia_proc_ctx {
427 /* ARM Processor Error Section, UEFI v2.7 sec N.2.4.4 */
428 struct cper_sec_proc_arm {
430 u16 err_info_num; /* Number of Processor Error Info */
431 u16 context_info_num; /* Number of Processor Context Info Records*/
434 u8 reserved[3]; /* must be zero */
437 u32 running_state; /* Bit 0 set - Processor running. PSCI = 0 */
441 /* ARM Processor Error Information Structure, UEFI v2.7 sec N.2.4.4.1 */
442 struct cper_arm_err_info {
451 u64 physical_fault_addr;
454 /* ARM Processor Context Information Structure, UEFI v2.7 sec N.2.4.4.2 */
455 struct cper_arm_ctx_info {
461 /* Old Memory Error Section, UEFI v2.1, v2.2 */
462 struct cper_sec_mem_err_old {
466 u64 physical_addr_mask;
481 /* Memory Error Section (UEFI >= v2.3), UEFI v2.8 sec N.2.5 */
482 struct cper_sec_mem_err {
486 u64 physical_addr_mask;
501 u16 mem_array_handle; /* "card handle" in UEFI 2.4 */
502 u16 mem_dev_handle; /* "module handle" in UEFI 2.4 */
505 struct cper_mem_err_compact {
519 u16 mem_array_handle;
524 static inline u32 cper_get_mem_extension(u64 mem_valid, u8 mem_extended)
526 if (!(mem_valid & CPER_MEM_VALID_ROW_EXT))
528 return (mem_extended & CPER_MEM_EXT_ROW_MASK) << CPER_MEM_EXT_ROW_SHIFT;
531 /* PCI Express Error Section, UEFI v2.7 sec N.2.7 */
532 struct cper_sec_pcie {
560 u16 secondary_status;
567 /* Firmware Error Record Reference, UEFI v2.7 sec N.2.10 */
568 struct cper_sec_fw_err_rec_ref {
572 u64 record_identifier;
573 guid_t record_identifier_guid;
576 /* Reset to default packing */
579 extern const char *const cper_proc_error_type_strs[4];
581 u64 cper_next_record_id(void);
582 const char *cper_severity_str(unsigned int);
583 const char *cper_mem_err_type_str(unsigned int);
584 const char *cper_mem_err_status_str(u64 status);
585 void cper_print_bits(const char *prefix, unsigned int bits,
586 const char * const strs[], unsigned int strs_size);
587 void cper_mem_err_pack(const struct cper_sec_mem_err *,
588 struct cper_mem_err_compact *);
589 const char *cper_mem_err_unpack(struct trace_seq *,
590 struct cper_mem_err_compact *);
591 void cper_print_proc_arm(const char *pfx,
592 const struct cper_sec_proc_arm *proc);
593 void cper_print_proc_ia(const char *pfx,
594 const struct cper_sec_proc_ia *proc);
595 int cper_mem_err_location(struct cper_mem_err_compact *mem, char *msg);
596 int cper_dimm_err_location(struct cper_mem_err_compact *mem, char *msg);
598 struct acpi_hest_generic_status;
599 void cper_estatus_print(const char *pfx,
600 const struct acpi_hest_generic_status *estatus);
601 int cper_estatus_check_header(const struct acpi_hest_generic_status *estatus);
602 int cper_estatus_check(const struct acpi_hest_generic_status *estatus);