1 /* 2 * UEFI Common Platform Error Record 3 * 4 * Copyright (C) 2010, Intel Corp. 5 * Author: Huang Ying <ying.huang@intel.com> 6 * 7 * This program is free software; you can redistribute it and/or 8 * modify it under the terms of the GNU General Public License version 9 * 2 as published by the Free Software Foundation. 10 * 11 * This program is distributed in the hope that it will be useful, 12 * but WITHOUT ANY WARRANTY; without even the implied warranty of 13 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the 14 * GNU General Public License for more details. 15 * 16 * You should have received a copy of the GNU General Public License 17 * along with this program; if not, write to the Free Software 18 * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA 19 */ 20 21 #ifndef LINUX_CPER_H 22 #define LINUX_CPER_H 23 24 #include <linux/uuid.h> 25 #include <linux/trace_seq.h> 26 27 /* CPER record signature and the size */ 28 #define CPER_SIG_RECORD "CPER" 29 #define CPER_SIG_SIZE 4 30 /* Used in signature_end field in struct cper_record_header */ 31 #define CPER_SIG_END 0xffffffff 32 33 /* 34 * CPER record header revision, used in revision field in struct 35 * cper_record_header 36 */ 37 #define CPER_RECORD_REV 0x0100 38 39 /* 40 * CPER record length contains the CPER fields which are relevant for further 41 * handling of a memory error in userspace (we don't carry all the fields 42 * defined in the UEFI spec because some of them don't make any sense.) 43 * Currently, a length of 256 should be more than enough. 44 */ 45 #define CPER_REC_LEN 256 46 /* 47 * Severity difinition for error_severity in struct cper_record_header 48 * and section_severity in struct cper_section_descriptor 49 */ 50 enum { 51 CPER_SEV_RECOVERABLE, 52 CPER_SEV_FATAL, 53 CPER_SEV_CORRECTED, 54 CPER_SEV_INFORMATIONAL, 55 }; 56 57 /* 58 * Validation bits difinition for validation_bits in struct 59 * cper_record_header. If set, corresponding fields in struct 60 * cper_record_header contain valid information. 61 * 62 * corresponds platform_id 63 */ 64 #define CPER_VALID_PLATFORM_ID 0x0001 65 /* corresponds timestamp */ 66 #define CPER_VALID_TIMESTAMP 0x0002 67 /* corresponds partition_id */ 68 #define CPER_VALID_PARTITION_ID 0x0004 69 70 /* 71 * Notification type used to generate error record, used in 72 * notification_type in struct cper_record_header 73 * 74 * Corrected Machine Check 75 */ 76 #define CPER_NOTIFY_CMC \ 77 GUID_INIT(0x2DCE8BB1, 0xBDD7, 0x450e, 0xB9, 0xAD, 0x9C, 0xF4, \ 78 0xEB, 0xD4, 0xF8, 0x90) 79 /* Corrected Platform Error */ 80 #define CPER_NOTIFY_CPE \ 81 GUID_INIT(0x4E292F96, 0xD843, 0x4a55, 0xA8, 0xC2, 0xD4, 0x81, \ 82 0xF2, 0x7E, 0xBE, 0xEE) 83 /* Machine Check Exception */ 84 #define CPER_NOTIFY_MCE \ 85 GUID_INIT(0xE8F56FFE, 0x919C, 0x4cc5, 0xBA, 0x88, 0x65, 0xAB, \ 86 0xE1, 0x49, 0x13, 0xBB) 87 /* PCI Express Error */ 88 #define CPER_NOTIFY_PCIE \ 89 GUID_INIT(0xCF93C01F, 0x1A16, 0x4dfc, 0xB8, 0xBC, 0x9C, 0x4D, \ 90 0xAF, 0x67, 0xC1, 0x04) 91 /* INIT Record (for IPF) */ 92 #define CPER_NOTIFY_INIT \ 93 GUID_INIT(0xCC5263E8, 0x9308, 0x454a, 0x89, 0xD0, 0x34, 0x0B, \ 94 0xD3, 0x9B, 0xC9, 0x8E) 95 /* Non-Maskable Interrupt */ 96 #define CPER_NOTIFY_NMI \ 97 GUID_INIT(0x5BAD89FF, 0xB7E6, 0x42c9, 0x81, 0x4A, 0xCF, 0x24, \ 98 0x85, 0xD6, 0xE9, 0x8A) 99 /* BOOT Error Record */ 100 #define CPER_NOTIFY_BOOT \ 101 GUID_INIT(0x3D61A466, 0xAB40, 0x409a, 0xA6, 0x98, 0xF3, 0x62, \ 102 0xD4, 0x64, 0xB3, 0x8F) 103 /* DMA Remapping Error */ 104 #define CPER_NOTIFY_DMAR \ 105 GUID_INIT(0x667DD791, 0xC6B3, 0x4c27, 0x8A, 0x6B, 0x0F, 0x8E, \ 106 0x72, 0x2D, 0xEB, 0x41) 107 108 /* 109 * Flags bits definitions for flags in struct cper_record_header 110 * If set, the error has been recovered 111 */ 112 #define CPER_HW_ERROR_FLAGS_RECOVERED 0x1 113 /* If set, the error is for previous boot */ 114 #define CPER_HW_ERROR_FLAGS_PREVERR 0x2 115 /* If set, the error is injected for testing */ 116 #define CPER_HW_ERROR_FLAGS_SIMULATED 0x4 117 118 /* 119 * CPER section header revision, used in revision field in struct 120 * cper_section_descriptor 121 */ 122 #define CPER_SEC_REV 0x0100 123 124 /* 125 * Validation bits difinition for validation_bits in struct 126 * cper_section_descriptor. If set, corresponding fields in struct 127 * cper_section_descriptor contain valid information. 128 * 129 * corresponds fru_id 130 */ 131 #define CPER_SEC_VALID_FRU_ID 0x1 132 /* corresponds fru_text */ 133 #define CPER_SEC_VALID_FRU_TEXT 0x2 134 135 /* 136 * Flags bits definitions for flags in struct cper_section_descriptor 137 * 138 * If set, the section is associated with the error condition 139 * directly, and should be focused on 140 */ 141 #define CPER_SEC_PRIMARY 0x0001 142 /* 143 * If set, the error was not contained within the processor or memory 144 * hierarchy and the error may have propagated to persistent storage 145 * or network 146 */ 147 #define CPER_SEC_CONTAINMENT_WARNING 0x0002 148 /* If set, the component must be re-initialized or re-enabled prior to use */ 149 #define CPER_SEC_RESET 0x0004 150 /* If set, Linux may choose to discontinue use of the resource */ 151 #define CPER_SEC_ERROR_THRESHOLD_EXCEEDED 0x0008 152 /* 153 * If set, resource could not be queried for error information due to 154 * conflicts with other system software or resources. Some fields of 155 * the section will be invalid 156 */ 157 #define CPER_SEC_RESOURCE_NOT_ACCESSIBLE 0x0010 158 /* 159 * If set, action has been taken to ensure error containment (such as 160 * poisoning data), but the error has not been fully corrected and the 161 * data has not been consumed. Linux may choose to take further 162 * corrective action before the data is consumed 163 */ 164 #define CPER_SEC_LATENT_ERROR 0x0020 165 166 /* 167 * Section type definitions, used in section_type field in struct 168 * cper_section_descriptor 169 * 170 * Processor Generic 171 */ 172 #define CPER_SEC_PROC_GENERIC \ 173 GUID_INIT(0x9876CCAD, 0x47B4, 0x4bdb, 0xB6, 0x5E, 0x16, 0xF1, \ 174 0x93, 0xC4, 0xF3, 0xDB) 175 /* Processor Specific: X86/X86_64 */ 176 #define CPER_SEC_PROC_IA \ 177 GUID_INIT(0xDC3EA0B0, 0xA144, 0x4797, 0xB9, 0x5B, 0x53, 0xFA, \ 178 0x24, 0x2B, 0x6E, 0x1D) 179 /* Processor Specific: IA64 */ 180 #define CPER_SEC_PROC_IPF \ 181 GUID_INIT(0xE429FAF1, 0x3CB7, 0x11D4, 0x0B, 0xCA, 0x07, 0x00, \ 182 0x80, 0xC7, 0x3C, 0x88, 0x81) 183 /* Processor Specific: ARM */ 184 #define CPER_SEC_PROC_ARM \ 185 GUID_INIT(0xE19E3D16, 0xBC11, 0x11E4, 0x9C, 0xAA, 0xC2, 0x05, \ 186 0x1D, 0x5D, 0x46, 0xB0) 187 /* Platform Memory */ 188 #define CPER_SEC_PLATFORM_MEM \ 189 GUID_INIT(0xA5BC1114, 0x6F64, 0x4EDE, 0xB8, 0x63, 0x3E, 0x83, \ 190 0xED, 0x7C, 0x83, 0xB1) 191 #define CPER_SEC_PCIE \ 192 GUID_INIT(0xD995E954, 0xBBC1, 0x430F, 0xAD, 0x91, 0xB4, 0x4D, \ 193 0xCB, 0x3C, 0x6F, 0x35) 194 /* Firmware Error Record Reference */ 195 #define CPER_SEC_FW_ERR_REC_REF \ 196 GUID_INIT(0x81212A96, 0x09ED, 0x4996, 0x94, 0x71, 0x8D, 0x72, \ 197 0x9C, 0x8E, 0x69, 0xED) 198 /* PCI/PCI-X Bus */ 199 #define CPER_SEC_PCI_X_BUS \ 200 GUID_INIT(0xC5753963, 0x3B84, 0x4095, 0xBF, 0x78, 0xED, 0xDA, \ 201 0xD3, 0xF9, 0xC9, 0xDD) 202 /* PCI Component/Device */ 203 #define CPER_SEC_PCI_DEV \ 204 GUID_INIT(0xEB5E4685, 0xCA66, 0x4769, 0xB6, 0xA2, 0x26, 0x06, \ 205 0x8B, 0x00, 0x13, 0x26) 206 #define CPER_SEC_DMAR_GENERIC \ 207 GUID_INIT(0x5B51FEF7, 0xC79D, 0x4434, 0x8F, 0x1B, 0xAA, 0x62, \ 208 0xDE, 0x3E, 0x2C, 0x64) 209 /* Intel VT for Directed I/O specific DMAr */ 210 #define CPER_SEC_DMAR_VT \ 211 GUID_INIT(0x71761D37, 0x32B2, 0x45cd, 0xA7, 0xD0, 0xB0, 0xFE, \ 212 0xDD, 0x93, 0xE8, 0xCF) 213 /* IOMMU specific DMAr */ 214 #define CPER_SEC_DMAR_IOMMU \ 215 GUID_INIT(0x036F84E1, 0x7F37, 0x428c, 0xA7, 0x9E, 0x57, 0x5F, \ 216 0xDF, 0xAA, 0x84, 0xEC) 217 218 #define CPER_PROC_VALID_TYPE 0x0001 219 #define CPER_PROC_VALID_ISA 0x0002 220 #define CPER_PROC_VALID_ERROR_TYPE 0x0004 221 #define CPER_PROC_VALID_OPERATION 0x0008 222 #define CPER_PROC_VALID_FLAGS 0x0010 223 #define CPER_PROC_VALID_LEVEL 0x0020 224 #define CPER_PROC_VALID_VERSION 0x0040 225 #define CPER_PROC_VALID_BRAND_INFO 0x0080 226 #define CPER_PROC_VALID_ID 0x0100 227 #define CPER_PROC_VALID_TARGET_ADDRESS 0x0200 228 #define CPER_PROC_VALID_REQUESTOR_ID 0x0400 229 #define CPER_PROC_VALID_RESPONDER_ID 0x0800 230 #define CPER_PROC_VALID_IP 0x1000 231 232 #define CPER_MEM_VALID_ERROR_STATUS 0x0001 233 #define CPER_MEM_VALID_PA 0x0002 234 #define CPER_MEM_VALID_PA_MASK 0x0004 235 #define CPER_MEM_VALID_NODE 0x0008 236 #define CPER_MEM_VALID_CARD 0x0010 237 #define CPER_MEM_VALID_MODULE 0x0020 238 #define CPER_MEM_VALID_BANK 0x0040 239 #define CPER_MEM_VALID_DEVICE 0x0080 240 #define CPER_MEM_VALID_ROW 0x0100 241 #define CPER_MEM_VALID_COLUMN 0x0200 242 #define CPER_MEM_VALID_BIT_POSITION 0x0400 243 #define CPER_MEM_VALID_REQUESTOR_ID 0x0800 244 #define CPER_MEM_VALID_RESPONDER_ID 0x1000 245 #define CPER_MEM_VALID_TARGET_ID 0x2000 246 #define CPER_MEM_VALID_ERROR_TYPE 0x4000 247 #define CPER_MEM_VALID_RANK_NUMBER 0x8000 248 #define CPER_MEM_VALID_CARD_HANDLE 0x10000 249 #define CPER_MEM_VALID_MODULE_HANDLE 0x20000 250 251 #define CPER_PCIE_VALID_PORT_TYPE 0x0001 252 #define CPER_PCIE_VALID_VERSION 0x0002 253 #define CPER_PCIE_VALID_COMMAND_STATUS 0x0004 254 #define CPER_PCIE_VALID_DEVICE_ID 0x0008 255 #define CPER_PCIE_VALID_SERIAL_NUMBER 0x0010 256 #define CPER_PCIE_VALID_BRIDGE_CONTROL_STATUS 0x0020 257 #define CPER_PCIE_VALID_CAPABILITY 0x0040 258 #define CPER_PCIE_VALID_AER_INFO 0x0080 259 260 #define CPER_PCIE_SLOT_SHIFT 3 261 262 #define CPER_ARM_VALID_MPIDR BIT(0) 263 #define CPER_ARM_VALID_AFFINITY_LEVEL BIT(1) 264 #define CPER_ARM_VALID_RUNNING_STATE BIT(2) 265 #define CPER_ARM_VALID_VENDOR_INFO BIT(3) 266 267 #define CPER_ARM_INFO_VALID_MULTI_ERR BIT(0) 268 #define CPER_ARM_INFO_VALID_FLAGS BIT(1) 269 #define CPER_ARM_INFO_VALID_ERR_INFO BIT(2) 270 #define CPER_ARM_INFO_VALID_VIRT_ADDR BIT(3) 271 #define CPER_ARM_INFO_VALID_PHYSICAL_ADDR BIT(4) 272 273 #define CPER_ARM_INFO_FLAGS_FIRST BIT(0) 274 #define CPER_ARM_INFO_FLAGS_LAST BIT(1) 275 #define CPER_ARM_INFO_FLAGS_PROPAGATED BIT(2) 276 #define CPER_ARM_INFO_FLAGS_OVERFLOW BIT(3) 277 278 /* 279 * All tables and structs must be byte-packed to match CPER 280 * specification, since the tables are provided by the system BIOS 281 */ 282 #pragma pack(1) 283 284 struct cper_record_header { 285 char signature[CPER_SIG_SIZE]; /* must be CPER_SIG_RECORD */ 286 __u16 revision; /* must be CPER_RECORD_REV */ 287 __u32 signature_end; /* must be CPER_SIG_END */ 288 __u16 section_count; 289 __u32 error_severity; 290 __u32 validation_bits; 291 __u32 record_length; 292 __u64 timestamp; 293 guid_t platform_id; 294 guid_t partition_id; 295 guid_t creator_id; 296 guid_t notification_type; 297 __u64 record_id; 298 __u32 flags; 299 __u64 persistence_information; 300 __u8 reserved[12]; /* must be zero */ 301 }; 302 303 struct cper_section_descriptor { 304 __u32 section_offset; /* Offset in bytes of the 305 * section body from the base 306 * of the record header */ 307 __u32 section_length; 308 __u16 revision; /* must be CPER_RECORD_REV */ 309 __u8 validation_bits; 310 __u8 reserved; /* must be zero */ 311 __u32 flags; 312 guid_t section_type; 313 guid_t fru_id; 314 __u32 section_severity; 315 __u8 fru_text[20]; 316 }; 317 318 /* Generic Processor Error Section */ 319 struct cper_sec_proc_generic { 320 __u64 validation_bits; 321 __u8 proc_type; 322 __u8 proc_isa; 323 __u8 proc_error_type; 324 __u8 operation; 325 __u8 flags; 326 __u8 level; 327 __u16 reserved; 328 __u64 cpu_version; 329 char cpu_brand[128]; 330 __u64 proc_id; 331 __u64 target_addr; 332 __u64 requestor_id; 333 __u64 responder_id; 334 __u64 ip; 335 }; 336 337 /* IA32/X64 Processor Error Section */ 338 struct cper_sec_proc_ia { 339 __u64 validation_bits; 340 __u8 lapic_id; 341 __u8 cpuid[48]; 342 }; 343 344 /* IA32/X64 Processor Error Information Structure */ 345 struct cper_ia_err_info { 346 guid_t err_type; 347 __u64 validation_bits; 348 __u64 check_info; 349 __u64 target_id; 350 __u64 requestor_id; 351 __u64 responder_id; 352 __u64 ip; 353 }; 354 355 /* IA32/X64 Processor Context Information Structure */ 356 struct cper_ia_proc_ctx { 357 __u16 reg_ctx_type; 358 __u16 reg_arr_size; 359 __u32 msr_addr; 360 __u64 mm_reg_addr; 361 }; 362 363 /* ARM Processor Error Section */ 364 struct cper_sec_proc_arm { 365 __u32 validation_bits; 366 __u16 err_info_num; /* Number of Processor Error Info */ 367 __u16 context_info_num; /* Number of Processor Context Info Records*/ 368 __u32 section_length; 369 __u8 affinity_level; 370 __u8 reserved[3]; /* must be zero */ 371 __u64 mpidr; 372 __u64 midr; 373 __u32 running_state; /* Bit 0 set - Processor running. PSCI = 0 */ 374 __u32 psci_state; 375 }; 376 377 /* ARM Processor Error Information Structure */ 378 struct cper_arm_err_info { 379 __u8 version; 380 __u8 length; 381 __u16 validation_bits; 382 __u8 type; 383 __u16 multiple_error; 384 __u8 flags; 385 __u64 error_info; 386 __u64 virt_fault_addr; 387 __u64 physical_fault_addr; 388 }; 389 390 /* ARM Processor Context Information Structure */ 391 struct cper_arm_ctx_info { 392 __u16 version; 393 __u16 type; 394 __u32 size; 395 }; 396 397 /* Old Memory Error Section UEFI 2.1, 2.2 */ 398 struct cper_sec_mem_err_old { 399 __u64 validation_bits; 400 __u64 error_status; 401 __u64 physical_addr; 402 __u64 physical_addr_mask; 403 __u16 node; 404 __u16 card; 405 __u16 module; 406 __u16 bank; 407 __u16 device; 408 __u16 row; 409 __u16 column; 410 __u16 bit_pos; 411 __u64 requestor_id; 412 __u64 responder_id; 413 __u64 target_id; 414 __u8 error_type; 415 }; 416 417 /* Memory Error Section UEFI >= 2.3 */ 418 struct cper_sec_mem_err { 419 __u64 validation_bits; 420 __u64 error_status; 421 __u64 physical_addr; 422 __u64 physical_addr_mask; 423 __u16 node; 424 __u16 card; 425 __u16 module; 426 __u16 bank; 427 __u16 device; 428 __u16 row; 429 __u16 column; 430 __u16 bit_pos; 431 __u64 requestor_id; 432 __u64 responder_id; 433 __u64 target_id; 434 __u8 error_type; 435 __u8 reserved; 436 __u16 rank; 437 __u16 mem_array_handle; /* card handle in UEFI 2.4 */ 438 __u16 mem_dev_handle; /* module handle in UEFI 2.4 */ 439 }; 440 441 struct cper_mem_err_compact { 442 __u64 validation_bits; 443 __u16 node; 444 __u16 card; 445 __u16 module; 446 __u16 bank; 447 __u16 device; 448 __u16 row; 449 __u16 column; 450 __u16 bit_pos; 451 __u64 requestor_id; 452 __u64 responder_id; 453 __u64 target_id; 454 __u16 rank; 455 __u16 mem_array_handle; 456 __u16 mem_dev_handle; 457 }; 458 459 struct cper_sec_pcie { 460 __u64 validation_bits; 461 __u32 port_type; 462 struct { 463 __u8 minor; 464 __u8 major; 465 __u8 reserved[2]; 466 } version; 467 __u16 command; 468 __u16 status; 469 __u32 reserved; 470 struct { 471 __u16 vendor_id; 472 __u16 device_id; 473 __u8 class_code[3]; 474 __u8 function; 475 __u8 device; 476 __u16 segment; 477 __u8 bus; 478 __u8 secondary_bus; 479 __u16 slot; 480 __u8 reserved; 481 } device_id; 482 struct { 483 __u32 lower; 484 __u32 upper; 485 } serial_number; 486 struct { 487 __u16 secondary_status; 488 __u16 control; 489 } bridge; 490 __u8 capability[60]; 491 __u8 aer_info[96]; 492 }; 493 494 /* Reset to default packing */ 495 #pragma pack() 496 497 u64 cper_next_record_id(void); 498 const char *cper_severity_str(unsigned int); 499 const char *cper_mem_err_type_str(unsigned int); 500 void cper_print_bits(const char *prefix, unsigned int bits, 501 const char * const strs[], unsigned int strs_size); 502 void cper_mem_err_pack(const struct cper_sec_mem_err *, 503 struct cper_mem_err_compact *); 504 const char *cper_mem_err_unpack(struct trace_seq *, 505 struct cper_mem_err_compact *); 506 507 #endif 508