Chau Ly | a743e38 | 2024-10-26 11:12:22 +0000 | [diff] [blame] | 1 | #pragma once |
| 2 | |
| 3 | #include "libpldm/pldm.h" |
| 4 | |
| 5 | #include "common/instance_id.hpp" |
| 6 | #include "common/types.hpp" |
| 7 | #include "oem_event_manager.hpp" |
| 8 | #include "platform-mc/manager.hpp" |
| 9 | #include "requester/handler.hpp" |
| 10 | #include "requester/request.hpp" |
| 11 | |
| 12 | namespace pldm |
| 13 | { |
| 14 | namespace oem_ampere |
| 15 | { |
| 16 | using namespace pldm::pdr; |
| 17 | |
| 18 | using EventToMsgMap_t = std::unordered_map<uint8_t, std::string>; |
| 19 | |
| 20 | enum sensor_ids |
| 21 | { |
Chau Ly | cebf476 | 2024-10-03 09:02:54 +0000 | [diff] [blame^] | 22 | DDR_STATUS = 51, |
Chau Ly | 3de0d94 | 2024-10-03 08:57:11 +0000 | [diff] [blame] | 23 | PCIE_HOT_PLUG = 169, |
Chau Ly | a743e38 | 2024-10-26 11:12:22 +0000 | [diff] [blame] | 24 | BOOT_OVERALL = 175, |
| 25 | }; |
| 26 | |
| 27 | namespace boot |
| 28 | { |
| 29 | namespace status |
| 30 | { |
| 31 | enum boot_status |
| 32 | { |
| 33 | BOOT_STATUS_SUCCESS = 0x80, |
| 34 | BOOT_STATUS_FAILURE = 0x81, |
| 35 | }; |
| 36 | } // namespace status |
| 37 | namespace stage |
| 38 | { |
| 39 | enum boot_stage |
| 40 | { |
| 41 | UEFI_STATUS_CLASS_CODE_MIN = 0x00, |
| 42 | UEFI_STATUS_CLASS_CODE_MAX = 0x7f, |
| 43 | SECPRO = 0x90, |
| 44 | MPRO = 0x91, |
| 45 | ATF_BL1 = 0x92, |
| 46 | ATF_BL2 = 0x93, |
| 47 | DDR_INITIALIZATION = 0x94, |
| 48 | DDR_TRAINING = 0x95, |
| 49 | S0_DDR_TRAINING_FAILURE = 0x96, |
| 50 | ATF_BL31 = 0x97, |
| 51 | ATF_BL32 = 0x98, |
| 52 | S1_DDR_TRAINING_FAILURE = 0x99, |
| 53 | }; |
| 54 | } // namespace stage |
| 55 | } // namespace boot |
| 56 | |
| 57 | enum class log_level : int |
| 58 | { |
| 59 | OK, |
Chau Ly | 3de0d94 | 2024-10-03 08:57:11 +0000 | [diff] [blame] | 60 | WARNING, |
| 61 | CRITICAL, |
Chau Ly | a743e38 | 2024-10-26 11:12:22 +0000 | [diff] [blame] | 62 | BIOSFWPANIC, |
| 63 | }; |
| 64 | |
Chau Ly | 3de0d94 | 2024-10-03 08:57:11 +0000 | [diff] [blame] | 65 | /* |
| 66 | * PresentReading value format |
| 67 | * FIELD | COMMENT |
| 68 | * Bit 31 | Reserved |
| 69 | * Bit 30:24 | Media slot number (0 - 63) This field can be used by UEFI |
| 70 | * | to indicate the media slot number (such as NVMe/SSD slot) |
| 71 | * | (7 bits) |
| 72 | * Bit 23 | Operation status: 1 = operation failed |
| 73 | * | 0 = operation successful |
| 74 | * Bit 22 | Action: 0 - Insertion 1 - Removal |
| 75 | * Bit 21:18 | Function (4 bits) |
| 76 | * Bit 17:13 | Device (5 bits) |
| 77 | * Bit 12:5 | Bus (8 bits) |
| 78 | * Bit 4:0 | Segment (5 bits) |
| 79 | */ |
| 80 | typedef union |
| 81 | { |
| 82 | uint32_t value; |
| 83 | struct |
| 84 | { |
| 85 | uint32_t segment:5; |
| 86 | uint32_t bus:8; |
| 87 | uint32_t device:5; |
| 88 | uint32_t function:4; |
| 89 | uint32_t action:1; |
| 90 | uint32_t opStatus:1; |
| 91 | uint32_t mediaSlot:7; |
| 92 | uint32_t reserved:1; |
| 93 | } __attribute__((packed)) bits; |
| 94 | } PCIeHotPlugEventRecord_t; |
| 95 | |
Chau Ly | cebf476 | 2024-10-03 09:02:54 +0000 | [diff] [blame^] | 96 | typedef union |
| 97 | { |
| 98 | uint32_t value; |
| 99 | struct |
| 100 | { |
| 101 | uint32_t type:2; |
| 102 | uint32_t mcuRankIdx:3; |
| 103 | uint32_t reserved_1:3; // byte0 |
| 104 | uint32_t sliceNum:4; |
| 105 | uint32_t upperNibbStatErr:1; |
| 106 | uint32_t lowerNibbStatErr:1; |
| 107 | uint32_t reserved_2:2; // byte1 |
| 108 | uint32_t syndrome:4; |
| 109 | uint32_t reserved_3:4; // byte2 |
| 110 | uint32_t reserved_byte:8; |
| 111 | } __attribute__((packed)) bits; |
| 112 | } DIMMTrainingFailure_t; |
| 113 | |
| 114 | namespace ddr |
| 115 | { |
| 116 | namespace status |
| 117 | { |
| 118 | enum ddr_status |
| 119 | { |
| 120 | NO_SYSTEM_LEVEL_ERROR = 0x01, |
| 121 | ECC_INITIALIZATION_FAILURE = 0x04, |
| 122 | CONFIGURATION_FAILURE = 0x05, |
| 123 | TRAINING_FAILURE = 0x06, |
| 124 | OTHER_FAILURE = 0x07, |
| 125 | BOOT_FAILURE_NO_VALID_CONFIG = 0x08, |
| 126 | FAILSAFE_ACTIVATED_NEXT_BOOT_SUCCESS = 0x09, |
| 127 | }; |
| 128 | } |
| 129 | } // namespace ddr |
| 130 | |
| 131 | namespace dimm |
| 132 | { |
| 133 | namespace status |
| 134 | { |
| 135 | enum dimm_status |
| 136 | { |
| 137 | INSTALLED_NO_ERROR = 0x01, |
| 138 | NOT_INSTALLED = 0x02, |
| 139 | OTHER_FAILURE = 0x07, |
| 140 | INSTALLED_BUT_DISABLED = 0x10, |
| 141 | TRAINING_FAILURE = 0x12, |
| 142 | PMIC_HIGH_TEMP = 0x13, |
| 143 | TSx_HIGH_TEMP = 0x14, |
| 144 | SPD_HUB_HIGH_TEMP = 0x15, |
| 145 | PMIC_TEMP_ALERT = 0x16, |
| 146 | }; |
| 147 | } // namespace status |
| 148 | |
| 149 | namespace training_failure |
| 150 | { |
| 151 | enum dimm_training_failure_type |
| 152 | { |
| 153 | PHY_TRAINING_FAILURE_TYPE = 0x01, |
| 154 | DIMM_TRAINING_FAILURE_TYPE = 0x02, |
| 155 | }; |
| 156 | |
| 157 | namespace phy_syndrome |
| 158 | { |
| 159 | enum phy_training_failure_syndrome |
| 160 | { |
| 161 | NA = 0x00, |
| 162 | PHY_TRAINING_SETUP_FAILURE = 0x01, |
| 163 | CA_LEVELING = 0x02, |
| 164 | PHY_WRITE_LEVEL_FAILURE = 0x03, |
| 165 | PHY_READ_GATE_LEVELING_FAILURE = 0x04, |
| 166 | PHY_READ_LEVEL_FAILURE = 0x05, |
| 167 | WRITE_DQ_LEVELING = 0x06, |
| 168 | PHY_SW_TRAINING_FAILURE = 0x07, |
| 169 | }; |
| 170 | } // namespace phy_syndrome |
| 171 | |
| 172 | namespace dimm_syndrome |
| 173 | { |
| 174 | enum dimm_training_failure_syndrome |
| 175 | { |
| 176 | NA = 0x00, |
| 177 | DRAM_VREFDQ_TRAINING_FAILURE = 0x01, |
| 178 | LRDIMM_DB_TRAINING_FAILURE = 0x02, |
| 179 | LRDRIMM_DB_SW_TRAINING_FAILURE = 0x03, |
| 180 | }; |
| 181 | } // namespace dimm_syndrome |
| 182 | } // namespace training_failure |
| 183 | } // namespace dimm |
| 184 | |
Chau Ly | a743e38 | 2024-10-26 11:12:22 +0000 | [diff] [blame] | 185 | /** |
| 186 | * @brief OemEventManager |
| 187 | * |
| 188 | * |
| 189 | */ |
| 190 | class OemEventManager |
| 191 | { |
| 192 | public: |
| 193 | OemEventManager() = delete; |
| 194 | OemEventManager(const OemEventManager&) = delete; |
| 195 | OemEventManager(OemEventManager&&) = delete; |
| 196 | OemEventManager& operator=(const OemEventManager&) = delete; |
| 197 | OemEventManager& operator=(OemEventManager&&) = delete; |
| 198 | virtual ~OemEventManager() = default; |
| 199 | |
| 200 | explicit OemEventManager( |
| 201 | sdeventplus::Event& event, |
| 202 | requester::Handler<requester::Request>* /* handler */, |
| 203 | pldm::InstanceIdDb& /* instanceIdDb */) : event(event) {}; |
| 204 | |
| 205 | /** @brief Decode sensor event messages and handle correspondingly. |
| 206 | * |
| 207 | * @param[in] request - the request message of sensor event |
| 208 | * @param[in] payloadLength - the payload length of sensor event |
| 209 | * @param[in] formatVersion - the format version of sensor event |
| 210 | * @param[in] tid - TID |
| 211 | * @param[in] eventDataOffset - the event data offset of sensor event |
| 212 | * |
| 213 | * @return int - returned error code |
| 214 | */ |
| 215 | int handleSensorEvent(const pldm_msg* request, size_t payloadLength, |
| 216 | uint8_t /* formatVersion */, pldm_tid_t tid, |
| 217 | size_t eventDataOffset); |
| 218 | |
| 219 | protected: |
| 220 | /** @brief Create prefix string for logging message. |
| 221 | * |
| 222 | * @param[in] tid - TID |
| 223 | * @param[in] sensorId - Sensor ID |
| 224 | * |
| 225 | * @return std::string - the prefeix string |
| 226 | */ |
| 227 | std::string prefixMsgStrCreation(pldm_tid_t tid, uint16_t sensorId); |
| 228 | |
| 229 | /** @brief Log the message into Redfish SEL. |
| 230 | * |
| 231 | * @param[in] description - the logging message |
| 232 | * @param[in] logLevel - the logging level |
| 233 | */ |
| 234 | void sendJournalRedfish(const std::string& description, |
| 235 | log_level& logLevel); |
| 236 | |
| 237 | /** @brief Convert the one-hot DIMM index byte into a string of DIMM |
| 238 | * indexes. |
| 239 | * |
| 240 | * @param[in] dimmIdxs - the one-hot DIMM index byte |
| 241 | * |
| 242 | * @return std::string - the string of DIMM indexes |
| 243 | */ |
| 244 | std::string dimmIdxsToString(uint32_t dimmIdxs); |
| 245 | |
Chau Ly | cebf476 | 2024-10-03 09:02:54 +0000 | [diff] [blame^] | 246 | /** @brief Convert the DIMM training failure into logging string. |
| 247 | * |
| 248 | * @param[in] failureInfo - the one-hot DIMM index byte |
| 249 | * |
| 250 | * @return std::string - the returned logging string |
| 251 | */ |
| 252 | std::string dimmTrainingFailureToMsg(uint32_t failureInfo); |
| 253 | |
Chau Ly | 3de0d94 | 2024-10-03 08:57:11 +0000 | [diff] [blame] | 254 | /** @brief Handle numeric sensor event message from PCIe hot-plug sensor. |
| 255 | * |
| 256 | * @param[in] tid - TID |
| 257 | * @param[in] sensorId - Sensor ID |
| 258 | * @param[in] presentReading - the present reading of the sensor |
| 259 | */ |
| 260 | void handlePCIeHotPlugEvent(pldm_tid_t tid, uint16_t sensorId, |
| 261 | uint32_t presentReading); |
| 262 | |
Chau Ly | a743e38 | 2024-10-26 11:12:22 +0000 | [diff] [blame] | 263 | /** @brief Handle numeric sensor event message from boot overall sensor. |
| 264 | * |
| 265 | * @param[in] tid - TID |
| 266 | * @param[in] sensorId - Sensor ID |
| 267 | * @param[in] presentReading - the present reading of the sensor |
| 268 | */ |
| 269 | void handleBootOverallEvent(pldm_tid_t /*tid*/, uint16_t /*sensorId*/, |
| 270 | uint32_t presentReading); |
| 271 | |
Chau Ly | cebf476 | 2024-10-03 09:02:54 +0000 | [diff] [blame^] | 272 | /** @brief Handle numeric sensor event message from DIMM status sensor. |
| 273 | * |
| 274 | * @param[in] tid - TID |
| 275 | * @param[in] sensorId - Sensor ID |
| 276 | * @param[in] presentReading - the present reading of the sensor |
| 277 | */ |
| 278 | void handleDIMMStatusEvent(pldm_tid_t tid, uint16_t sensorId, |
| 279 | uint32_t presentReading); |
| 280 | |
| 281 | /** @brief Handle numeric sensor event message from DDR status sensor. |
| 282 | * |
| 283 | * @param[in] tid - TID |
| 284 | * @param[in] sensorId - Sensor ID |
| 285 | * @param[in] presentReading - the present reading of the sensor |
| 286 | */ |
| 287 | void handleDDRStatusEvent(pldm_tid_t tid, uint16_t sensorId, |
| 288 | uint32_t presentReading); |
| 289 | |
Chau Ly | a743e38 | 2024-10-26 11:12:22 +0000 | [diff] [blame] | 290 | /** @brief Handle numeric sensor event messages. |
| 291 | * |
| 292 | * @param[in] tid - TID |
| 293 | * @param[in] sensorId - Sensor ID |
| 294 | * @param[in] sensorData - the sensor data |
| 295 | * @param[in] sensorDataLength - the length of sensor data |
| 296 | * |
| 297 | * @return int - returned error code |
| 298 | */ |
| 299 | int processNumericSensorEvent(pldm_tid_t tid, uint16_t sensorId, |
| 300 | const uint8_t* sensorData, |
| 301 | size_t sensorDataLength); |
| 302 | |
| 303 | /** @brief Handle state sensor event messages. |
| 304 | * |
| 305 | * @param[in] tid - TID |
| 306 | * @param[in] sensorId - Sensor ID |
| 307 | * @param[in] sensorData - the sensor data |
| 308 | * @param[in] sensorDataLength - the length of sensor data |
| 309 | * |
| 310 | * @return int - returned error code |
| 311 | */ |
| 312 | int processStateSensorEvent(pldm_tid_t tid, uint16_t sensorId, |
| 313 | const uint8_t* sensorData, |
| 314 | size_t sensorDataLength); |
| 315 | |
| 316 | /** @brief Handle op state sensor event messages. |
| 317 | * |
| 318 | * @param[in] tid - TID |
| 319 | * @param[in] sensorId - Sensor ID |
| 320 | * @param[in] sensorData - the sensor data |
| 321 | * @param[in] sensorDataLength - the length of sensor data |
| 322 | * |
| 323 | * @return int - returned error code |
| 324 | */ |
| 325 | int processSensorOpStateEvent(pldm_tid_t tid, uint16_t sensorId, |
| 326 | const uint8_t* sensorData, |
| 327 | size_t sensorDataLength); |
| 328 | |
| 329 | /** @brief reference of main event loop of pldmd, primarily used to schedule |
| 330 | * work |
| 331 | */ |
| 332 | sdeventplus::Event& event; |
| 333 | }; |
| 334 | } // namespace oem_ampere |
| 335 | } // namespace pldm |