diff options
Diffstat (limited to 'drivers/misc/habanalabs/include/common')
| -rw-r--r-- | drivers/misc/habanalabs/include/common/cpucp_if.h | 1320 | ||||
| -rw-r--r-- | drivers/misc/habanalabs/include/common/hl_boot_if.h | 702 | ||||
| -rw-r--r-- | drivers/misc/habanalabs/include/common/qman_if.h | 68 |
3 files changed, 2090 insertions, 0 deletions
diff --git a/drivers/misc/habanalabs/include/common/cpucp_if.h b/drivers/misc/habanalabs/include/common/cpucp_if.h new file mode 100644 index 000000000..baa5aa43b --- /dev/null +++ b/drivers/misc/habanalabs/include/common/cpucp_if.h @@ -0,0 +1,1320 @@ +/* SPDX-License-Identifier: GPL-2.0 + * + * Copyright 2020-2022 HabanaLabs, Ltd. + * All Rights Reserved. + * + */ + +#ifndef CPUCP_IF_H +#define CPUCP_IF_H + +#include <linux/types.h> +#include <linux/if_ether.h> + +#include "hl_boot_if.h" + +#define NUM_HBM_PSEUDO_CH 2 +#define NUM_HBM_CH_PER_DEV 8 +#define CPUCP_PKT_HBM_ECC_INFO_WR_PAR_SHIFT 0 +#define CPUCP_PKT_HBM_ECC_INFO_WR_PAR_MASK 0x00000001 +#define CPUCP_PKT_HBM_ECC_INFO_RD_PAR_SHIFT 1 +#define CPUCP_PKT_HBM_ECC_INFO_RD_PAR_MASK 0x00000002 +#define CPUCP_PKT_HBM_ECC_INFO_CA_PAR_SHIFT 2 +#define CPUCP_PKT_HBM_ECC_INFO_CA_PAR_MASK 0x00000004 +#define CPUCP_PKT_HBM_ECC_INFO_DERR_SHIFT 3 +#define CPUCP_PKT_HBM_ECC_INFO_DERR_MASK 0x00000008 +#define CPUCP_PKT_HBM_ECC_INFO_SERR_SHIFT 4 +#define CPUCP_PKT_HBM_ECC_INFO_SERR_MASK 0x00000010 +#define CPUCP_PKT_HBM_ECC_INFO_TYPE_SHIFT 5 +#define CPUCP_PKT_HBM_ECC_INFO_TYPE_MASK 0x00000020 +#define CPUCP_PKT_HBM_ECC_INFO_HBM_CH_SHIFT 6 +#define CPUCP_PKT_HBM_ECC_INFO_HBM_CH_MASK 0x000007C0 + +#define PLL_MAP_MAX_BITS 128 +#define PLL_MAP_LEN (PLL_MAP_MAX_BITS / 8) + +/* + * info of the pkt queue pointers in the first async occurrence + */ +struct cpucp_pkt_sync_err { + __le32 pi; + __le32 ci; +}; + +struct hl_eq_hbm_ecc_data { + /* SERR counter */ + __le32 sec_cnt; + /* DERR counter */ + __le32 dec_cnt; + /* Supplemental Information according to the mask bits */ + __le32 hbm_ecc_info; + /* Address in hbm where the ecc happened */ + __le32 first_addr; + /* SERR continuous address counter */ + __le32 sec_cont_cnt; + __le32 pad; +}; + +/* + * EVENT QUEUE + */ + +struct hl_eq_header { + __le32 reserved; + __le32 ctl; +}; + +struct hl_eq_ecc_data { + __le64 ecc_address; + __le64 ecc_syndrom; + __u8 memory_wrapper_idx; + __u8 is_critical; + __u8 pad[6]; +}; + +enum hl_sm_sei_cause { + SM_SEI_SO_OVERFLOW, + SM_SEI_LBW_4B_UNALIGNED, + SM_SEI_AXI_RESPONSE_ERR +}; + +struct hl_eq_sm_sei_data { + __le32 sei_log; + /* enum hl_sm_sei_cause */ + __u8 sei_cause; + __u8 pad[3]; +}; + +enum hl_fw_alive_severity { + FW_ALIVE_SEVERITY_MINOR, + FW_ALIVE_SEVERITY_CRITICAL +}; + +struct hl_eq_fw_alive { + __le64 uptime_seconds; + __le32 process_id; + __le32 thread_id; + /* enum hl_fw_alive_severity */ + __u8 severity; + __u8 pad[7]; +}; + +struct hl_eq_intr_cause { + __le64 intr_cause_data; +}; + +struct hl_eq_pcie_drain_ind_data { + struct hl_eq_intr_cause intr_cause; + __le64 drain_wr_addr_lbw; + __le64 drain_rd_addr_lbw; + __le64 drain_wr_addr_hbw; + __le64 drain_rd_addr_hbw; +}; + +struct hl_eq_razwi_lbw_info_regs { + __le32 rr_aw_razwi_reg; + __le32 rr_aw_razwi_id_reg; + __le32 rr_ar_razwi_reg; + __le32 rr_ar_razwi_id_reg; +}; + +struct hl_eq_razwi_hbw_info_regs { + __le32 rr_aw_razwi_hi_reg; + __le32 rr_aw_razwi_lo_reg; + __le32 rr_aw_razwi_id_reg; + __le32 rr_ar_razwi_hi_reg; + __le32 rr_ar_razwi_lo_reg; + __le32 rr_ar_razwi_id_reg; +}; + +/* razwi_happened masks */ +#define RAZWI_HAPPENED_HBW 0x1 +#define RAZWI_HAPPENED_LBW 0x2 +#define RAZWI_HAPPENED_AW 0x4 +#define RAZWI_HAPPENED_AR 0x8 + +struct hl_eq_razwi_info { + __le32 razwi_happened_mask; + union { + struct hl_eq_razwi_lbw_info_regs lbw; + struct hl_eq_razwi_hbw_info_regs hbw; + }; + __le32 pad; +}; + +struct hl_eq_razwi_with_intr_cause { + struct hl_eq_razwi_info razwi_info; + struct hl_eq_intr_cause intr_cause; +}; + +#define HBM_CA_ERR_CMD_LIFO_LEN 8 +#define HBM_RD_ERR_DATA_LIFO_LEN 8 +#define HBM_WR_PAR_CMD_LIFO_LEN 11 + +enum hl_hbm_sei_cause { + /* Command/address parity error event is split into 2 events due to + * size limitation: ODD suffix for odd HBM CK_t cycles and EVEN suffix + * for even HBM CK_t cycles + */ + HBM_SEI_CMD_PARITY_EVEN, + HBM_SEI_CMD_PARITY_ODD, + /* Read errors can be reflected as a combination of SERR/DERR/parity + * errors. Therefore, we define one event for all read error types. + * LKD will perform further proccessing. + */ + HBM_SEI_READ_ERR, + HBM_SEI_WRITE_DATA_PARITY_ERR, + HBM_SEI_CATTRIP, + HBM_SEI_MEM_BIST_FAIL, + HBM_SEI_DFI, + HBM_SEI_INV_TEMP_READ_OUT, + HBM_SEI_BIST_FAIL, +}; + +/* Masks for parsing hl_hbm_sei_headr fields */ +#define HBM_ECC_SERR_CNTR_MASK 0xFF +#define HBM_ECC_DERR_CNTR_MASK 0xFF00 +#define HBM_RD_PARITY_CNTR_MASK 0xFF0000 + +/* HBM index and MC index are known by the event_id */ +struct hl_hbm_sei_header { + union { + /* relevant only in case of HBM read error */ + struct { + __u8 ecc_serr_cnt; + __u8 ecc_derr_cnt; + __u8 read_par_cnt; + __u8 reserved; + }; + /* All other cases */ + __le32 cnt; + }; + __u8 sei_cause; /* enum hl_hbm_sei_cause */ + __u8 mc_channel; /* range: 0-3 */ + __u8 mc_pseudo_channel; /* range: 0-7 */ + __u8 is_critical; +}; + +#define HBM_RD_ADDR_SID_SHIFT 0 +#define HBM_RD_ADDR_SID_MASK 0x1 +#define HBM_RD_ADDR_BG_SHIFT 1 +#define HBM_RD_ADDR_BG_MASK 0x6 +#define HBM_RD_ADDR_BA_SHIFT 3 +#define HBM_RD_ADDR_BA_MASK 0x18 +#define HBM_RD_ADDR_COL_SHIFT 5 +#define HBM_RD_ADDR_COL_MASK 0x7E0 +#define HBM_RD_ADDR_ROW_SHIFT 11 +#define HBM_RD_ADDR_ROW_MASK 0x3FFF800 + +struct hbm_rd_addr { + union { + /* bit fields are only for FW use */ + struct { + u32 dbg_rd_err_addr_sid:1; + u32 dbg_rd_err_addr_bg:2; + u32 dbg_rd_err_addr_ba:2; + u32 dbg_rd_err_addr_col:6; + u32 dbg_rd_err_addr_row:15; + u32 reserved:6; + }; + __le32 rd_addr_val; + }; +}; + +#define HBM_RD_ERR_BEAT_SHIFT 2 +/* dbg_rd_err_misc fields: */ +/* Read parity is calculated per DW on every beat */ +#define HBM_RD_ERR_PAR_ERR_BEAT0_SHIFT 0 +#define HBM_RD_ERR_PAR_ERR_BEAT0_MASK 0x3 +#define HBM_RD_ERR_PAR_DATA_BEAT0_SHIFT 8 +#define HBM_RD_ERR_PAR_DATA_BEAT0_MASK 0x300 +/* ECC is calculated per PC on every beat */ +#define HBM_RD_ERR_SERR_BEAT0_SHIFT 16 +#define HBM_RD_ERR_SERR_BEAT0_MASK 0x10000 +#define HBM_RD_ERR_DERR_BEAT0_SHIFT 24 +#define HBM_RD_ERR_DERR_BEAT0_MASK 0x100000 + +struct hl_eq_hbm_sei_read_err_intr_info { + /* DFI_RD_ERR_REP_ADDR */ + struct hbm_rd_addr dbg_rd_err_addr; + /* DFI_RD_ERR_REP_ERR */ + union { + struct { + /* bit fields are only for FW use */ + u32 dbg_rd_err_par:8; + u32 dbg_rd_err_par_data:8; + u32 dbg_rd_err_serr:4; + u32 dbg_rd_err_derr:4; + u32 reserved:8; + }; + __le32 dbg_rd_err_misc; + }; + /* DFI_RD_ERR_REP_DM */ + __le32 dbg_rd_err_dm; + /* DFI_RD_ERR_REP_SYNDROME */ + __le32 dbg_rd_err_syndrome; + /* DFI_RD_ERR_REP_DATA */ + __le32 dbg_rd_err_data[HBM_RD_ERR_DATA_LIFO_LEN]; +}; + +struct hl_eq_hbm_sei_ca_par_intr_info { + /* 14 LSBs */ + __le16 dbg_row[HBM_CA_ERR_CMD_LIFO_LEN]; + /* 18 LSBs */ + __le32 dbg_col[HBM_CA_ERR_CMD_LIFO_LEN]; +}; + +#define WR_PAR_LAST_CMD_COL_SHIFT 0 +#define WR_PAR_LAST_CMD_COL_MASK 0x3F +#define WR_PAR_LAST_CMD_BG_SHIFT 6 +#define WR_PAR_LAST_CMD_BG_MASK 0xC0 +#define WR_PAR_LAST_CMD_BA_SHIFT 8 +#define WR_PAR_LAST_CMD_BA_MASK 0x300 +#define WR_PAR_LAST_CMD_SID_SHIFT 10 +#define WR_PAR_LAST_CMD_SID_MASK 0x400 + +/* Row address isn't latched */ +struct hbm_sei_wr_cmd_address { + /* DFI_DERR_LAST_CMD */ + union { + struct { + /* bit fields are only for FW use */ + u32 col:6; + u32 bg:2; + u32 ba:2; + u32 sid:1; + u32 reserved:21; + }; + __le32 dbg_wr_cmd_addr; + }; +}; + +struct hl_eq_hbm_sei_wr_par_intr_info { + /* entry 0: WR command address from the 1st cycle prior to the error + * entry 1: WR command address from the 2nd cycle prior to the error + * and so on... + */ + struct hbm_sei_wr_cmd_address dbg_last_wr_cmds[HBM_WR_PAR_CMD_LIFO_LEN]; + /* derr[0:1] - 1st HBM cycle DERR output + * derr[2:3] - 2nd HBM cycle DERR output + */ + __u8 dbg_derr; + /* extend to reach 8B */ + __u8 pad[3]; +}; + +/* + * this struct represents the following sei causes: + * command parity, ECC double error, ECC single error, dfi error, cattrip, + * temperature read-out, read parity error and write parity error. + * some only use the header while some have extra data. + */ +struct hl_eq_hbm_sei_data { + struct hl_hbm_sei_header hdr; + union { + struct hl_eq_hbm_sei_ca_par_intr_info ca_parity_even_info; + struct hl_eq_hbm_sei_ca_par_intr_info ca_parity_odd_info; + struct hl_eq_hbm_sei_read_err_intr_info read_err_info; + struct hl_eq_hbm_sei_wr_par_intr_info wr_parity_info; + }; +}; + +/* Engine/farm arc interrupt type */ +enum hl_engine_arc_interrupt_type { + /* Qman/farm ARC DCCM QUEUE FULL interrupt type */ + ENGINE_ARC_DCCM_QUEUE_FULL_IRQ = 1 +}; + +/* Data structure specifies details of payload of DCCM QUEUE FULL interrupt */ +struct hl_engine_arc_dccm_queue_full_irq { + /* Queue index value which caused DCCM QUEUE FULL */ + __le32 queue_index; + __le32 pad; +}; + +/* Data structure specifies details of QM/FARM ARC interrupt */ +struct hl_eq_engine_arc_intr_data { + /* ARC engine id e.g. DCORE0_TPC0_QM_ARC, DCORE0_TCP1_QM_ARC */ + __le32 engine_id; + __le32 intr_type; /* enum hl_engine_arc_interrupt_type */ + /* More info related to the interrupt e.g. queue index + * incase of DCCM_QUEUE_FULL interrupt. + */ + __le64 payload; + __le64 pad[5]; +}; + +struct hl_eq_entry { + struct hl_eq_header hdr; + union { + struct hl_eq_ecc_data ecc_data; + struct hl_eq_hbm_ecc_data hbm_ecc_data; /* Gaudi1 HBM */ + struct hl_eq_sm_sei_data sm_sei_data; + struct cpucp_pkt_sync_err pkt_sync_err; + struct hl_eq_fw_alive fw_alive; + struct hl_eq_intr_cause intr_cause; + struct hl_eq_pcie_drain_ind_data pcie_drain_ind_data; + struct hl_eq_razwi_info razwi_info; + struct hl_eq_razwi_with_intr_cause razwi_with_intr_cause; + struct hl_eq_hbm_sei_data sei_data; /* Gaudi2 HBM */ + struct hl_eq_engine_arc_intr_data arc_data; + __le64 data[7]; + }; +}; + +#define HL_EQ_ENTRY_SIZE sizeof(struct hl_eq_entry) + +#define EQ_CTL_READY_SHIFT 31 +#define EQ_CTL_READY_MASK 0x80000000 + +#define EQ_CTL_EVENT_TYPE_SHIFT 16 +#define EQ_CTL_EVENT_TYPE_MASK 0x0FFF0000 + +#define EQ_CTL_INDEX_SHIFT 0 +#define EQ_CTL_INDEX_MASK 0x0000FFFF + +enum pq_init_status { + PQ_INIT_STATUS_NA = 0, + PQ_INIT_STATUS_READY_FOR_CP, + PQ_INIT_STATUS_READY_FOR_HOST, + PQ_INIT_STATUS_READY_FOR_CP_SINGLE_MSI, + PQ_INIT_STATUS_LEN_NOT_POWER_OF_TWO_ERR, + PQ_INIT_STATUS_ILLEGAL_Q_ADDR_ERR +}; + +/* + * CpuCP Primary Queue Packets + * + * During normal operation, the host's kernel driver needs to send various + * messages to CpuCP, usually either to SET some value into a H/W periphery or + * to GET the current value of some H/W periphery. For example, SET the + * frequency of MME/TPC and GET the value of the thermal sensor. + * + * These messages can be initiated either by the User application or by the + * host's driver itself, e.g. power management code. In either case, the + * communication from the host's driver to CpuCP will *always* be in + * synchronous mode, meaning that the host will send a single message and poll + * until the message was acknowledged and the results are ready (if results are + * needed). + * + * This means that only a single message can be sent at a time and the host's + * driver must wait for its result before sending the next message. Having said + * that, because these are control messages which are sent in a relatively low + * frequency, this limitation seems acceptable. It's important to note that + * in case of multiple devices, messages to different devices *can* be sent + * at the same time. + * + * The message, inputs/outputs (if relevant) and fence object will be located + * on the device DDR at an address that will be determined by the host's driver. + * During device initialization phase, the host will pass to CpuCP that address. + * Most of the message types will contain inputs/outputs inside the message + * itself. The common part of each message will contain the opcode of the + * message (its type) and a field representing a fence object. + * + * When the host's driver wishes to send a message to CPU CP, it will write the + * message contents to the device DDR, clear the fence object and then write to + * the PSOC_ARC1_AUX_SW_INTR, to issue interrupt 121 to ARC Management CPU. + * + * Upon receiving the interrupt (#121), CpuCP will read the message from the + * DDR. In case the message is a SET operation, CpuCP will first perform the + * operation and then write to the fence object on the device DDR. In case the + * message is a GET operation, CpuCP will first fill the results section on the + * device DDR and then write to the fence object. If an error occurred, CpuCP + * will fill the rc field with the right error code. + * + * In the meantime, the host's driver will poll on the fence object. Once the + * host sees that the fence object is signaled, it will read the results from + * the device DDR (if relevant) and resume the code execution in the host's + * driver. + * + * To use QMAN packets, the opcode must be the QMAN opcode, shifted by 8 + * so the value being put by the host's driver matches the value read by CpuCP + * + * Non-QMAN packets should be limited to values 1 through (2^8 - 1) + * + * Detailed description: + * + * CPUCP_PACKET_DISABLE_PCI_ACCESS - + * After receiving this packet the embedded CPU must NOT issue PCI + * transactions (read/write) towards the Host CPU. This also include + * sending MSI-X interrupts. + * This packet is usually sent before the device is moved to D3Hot state. + * + * CPUCP_PACKET_ENABLE_PCI_ACCESS - + * After receiving this packet the embedded CPU is allowed to issue PCI + * transactions towards the Host CPU, including sending MSI-X interrupts. + * This packet is usually send after the device is moved to D0 state. + * + * CPUCP_PACKET_TEMPERATURE_GET - + * Fetch the current temperature / Max / Max Hyst / Critical / + * Critical Hyst of a specified thermal sensor. The packet's + * arguments specify the desired sensor and the field to get. + * + * CPUCP_PACKET_VOLTAGE_GET - + * Fetch the voltage / Max / Min of a specified sensor. The packet's + * arguments specify the sensor and type. + * + * CPUCP_PACKET_CURRENT_GET - + * Fetch the current / Max / Min of a specified sensor. The packet's + * arguments specify the sensor and type. + * + * CPUCP_PACKET_FAN_SPEED_GET - + * Fetch the speed / Max / Min of a specified fan. The packet's + * arguments specify the sensor and type. + * + * CPUCP_PACKET_PWM_GET - + * Fetch the pwm value / mode of a specified pwm. The packet's + * arguments specify the sensor and type. + * + * CPUCP_PACKET_PWM_SET - + * Set the pwm value / mode of a specified pwm. The packet's + * arguments specify the sensor, type and value. + * + * CPUCP_PACKET_FREQUENCY_SET - + * Set the frequency of a specified PLL. The packet's arguments specify + * the PLL and the desired frequency. The actual frequency in the device + * might differ from the requested frequency. + * + * CPUCP_PACKET_FREQUENCY_GET - + * Fetch the frequency of a specified PLL. The packet's arguments specify + * the PLL. + * + * CPUCP_PACKET_LED_SET - + * Set the state of a specified led. The packet's arguments + * specify the led and the desired state. + * + * CPUCP_PACKET_I2C_WR - + * Write 32-bit value to I2C device. The packet's arguments specify the + * I2C bus, address and value. + * + * CPUCP_PACKET_I2C_RD - + * Read 32-bit value from I2C device. The packet's arguments specify the + * I2C bus and address. + * + * CPUCP_PACKET_INFO_GET - + * Fetch information from the device as specified in the packet's + * structure. The host's driver passes the max size it allows the CpuCP to + * write to the structure, to prevent data corruption in case of + * mismatched driver/FW versions. + * + * CPUCP_PACKET_FLASH_PROGRAM_REMOVED - this packet was removed + * + * CPUCP_PACKET_UNMASK_RAZWI_IRQ - + * Unmask the given IRQ. The IRQ number is specified in the value field. + * The packet is sent after receiving an interrupt and printing its + * relevant information. + * + * CPUCP_PACKET_UNMASK_RAZWI_IRQ_ARRAY - + * Unmask the given IRQs. The IRQs numbers are specified in an array right + * after the cpucp_packet structure, where its first element is the array + * length. The packet is sent after a soft reset was done in order to + * handle any interrupts that were sent during the reset process. + * + * CPUCP_PACKET_TEST - + * Test packet for CpuCP connectivity. The CPU will put the fence value + * in the result field. + * + * CPUCP_PACKET_FREQUENCY_CURR_GET - + * Fetch the current frequency of a specified PLL. The packet's arguments + * specify the PLL. + * + * CPUCP_PACKET_MAX_POWER_GET - + * Fetch the maximal power of the device. + * + * CPUCP_PACKET_MAX_POWER_SET - + * Set the maximal power of the device. The packet's arguments specify + * the power. + * + * CPUCP_PACKET_EEPROM_DATA_GET - + * Get EEPROM data from the CpuCP kernel. The buffer is specified in the + * addr field. The CPU will put the returned data size in the result + * field. In addition, the host's driver passes the max size it allows the + * CpuCP to write to the structure, to prevent data corruption in case of + * mismatched driver/FW versions. + * + * CPUCP_PACKET_NIC_INFO_GET - + * Fetch information from the device regarding the NIC. the host's driver + * passes the max size it allows the CpuCP to write to the structure, to + * prevent data corruption in case of mismatched driver/FW versions. + * + * CPUCP_PACKET_TEMPERATURE_SET - + * Set the value of the offset property of a specified thermal sensor. + * The packet's arguments specify the desired sensor and the field to + * set. + * + * CPUCP_PACKET_VOLTAGE_SET - + * Trigger the reset_history property of a specified voltage sensor. + * The packet's arguments specify the desired sensor and the field to + * set. + * + * CPUCP_PACKET_CURRENT_SET - + * Trigger the reset_history property of a specified current sensor. + * The packet's arguments specify the desired sensor and the field to + * set. + * + * CPUCP_PACKET_PCIE_THROUGHPUT_GET - + * Get throughput of PCIe. + * The packet's arguments specify the transaction direction (TX/RX). + * The window measurement is 10[msec], and the return value is in KB/sec. + * + * CPUCP_PACKET_PCIE_REPLAY_CNT_GET + * Replay count measures number of "replay" events, which is basicly + * number of retries done by PCIe. + * + * CPUCP_PACKET_TOTAL_ENERGY_GET - + * Total Energy is measurement of energy from the time FW Linux + * is loaded. It is calculated by multiplying the average power + * by time (passed from armcp start). The units are in MilliJouls. + * + * CPUCP_PACKET_PLL_INFO_GET - + * Fetch frequencies of PLL from the required PLL IP. + * The packet's arguments specify the device PLL type + * Pll type is the PLL from device pll_index enum. + * The result is composed of 4 outputs, each is 16-bit + * frequency in MHz. + * + * CPUCP_PACKET_POWER_GET - + * Fetch the present power consumption of the device (Current * Voltage). + * + * CPUCP_PACKET_NIC_PFC_SET - + * Enable/Disable the NIC PFC feature. The packet's arguments specify the + * NIC port, relevant lanes to configure and one bit indication for + * enable/disable. + * + * CPUCP_PACKET_NIC_FAULT_GET - + * Fetch the current indication for local/remote faults from the NIC MAC. + * The result is 32-bit value of the relevant register. + * + * CPUCP_PACKET_NIC_LPBK_SET - + * Enable/Disable the MAC loopback feature. The packet's arguments specify + * the NIC port, relevant lanes to configure and one bit indication for + * enable/disable. + * + * CPUCP_PACKET_NIC_MAC_INIT - + * Configure the NIC MAC channels. The packet's arguments specify the + * NIC port and the speed. + * + * CPUCP_PACKET_MSI_INFO_SET - + * set the index number for each supported msi type going from + * host to device + * + * CPUCP_PACKET_NIC_XPCS91_REGS_GET - + * Fetch the un/correctable counters values from the NIC MAC. + * + * CPUCP_PACKET_NIC_STAT_REGS_GET - + * Fetch various NIC MAC counters from the NIC STAT. + * + * CPUCP_PACKET_NIC_STAT_REGS_CLR - + * Clear the various NIC MAC counters in the NIC STAT. + * + * CPUCP_PACKET_NIC_STAT_REGS_ALL_GET - + * Fetch all NIC MAC counters from the NIC STAT. + * + * CPUCP_PACKET_IS_IDLE_CHECK - + * Check if the device is IDLE in regard to the DMA/compute engines + * and QMANs. The f/w will return a bitmask where each bit represents + * a different engine or QMAN according to enum cpucp_idle_mask. + * The bit will be 1 if the engine is NOT idle. + * + * CPUCP_PACKET_HBM_REPLACED_ROWS_INFO_GET - + * Fetch all HBM replaced-rows and prending to be replaced rows data. + * + * CPUCP_PACKET_HBM_PENDING_ROWS_STATUS - + * Fetch status of HBM rows pending replacement and need a reboot to + * be replaced. + * + * CPUCP_PACKET_POWER_SET - + * Resets power history of device to 0 + * + * CPUCP_PACKET_ENGINE_CORE_ASID_SET - + * Packet to perform engine core ASID configuration + * + * CPUCP_PACKET_SEC_ATTEST_GET - + * Get the attestaion data that is collected during various stages of the + * boot sequence. the attestation data is also hashed with some unique + * number (nonce) provided by the host to prevent replay attacks. + * public key and certificate also provided as part of the FW response. + * + * CPUCP_PACKET_MONITOR_DUMP_GET - + * Get monitors registers dump from the CpuCP kernel. + * The CPU will put the registers dump in the a buffer allocated by the driver + * which address is passed via the CpuCp packet. In addition, the host's driver + * passes the max size it allows the CpuCP to write to the structure, to prevent + * data corruption in case of mismatched driver/FW versions. + * Relevant only to Gaudi. + * + * CPUCP_PACKET_ACTIVE_STATUS_SET - + * LKD sends FW indication whether device is free or in use, this indication is reported + * also to the BMC. + */ + +enum cpucp_packet_id { + CPUCP_PACKET_DISABLE_PCI_ACCESS = 1, /* internal */ + CPUCP_PACKET_ENABLE_PCI_ACCESS, /* internal */ + CPUCP_PACKET_TEMPERATURE_GET, /* sysfs */ + CPUCP_PACKET_VOLTAGE_GET, /* sysfs */ + CPUCP_PACKET_CURRENT_GET, /* sysfs */ + CPUCP_PACKET_FAN_SPEED_GET, /* sysfs */ + CPUCP_PACKET_PWM_GET, /* sysfs */ + CPUCP_PACKET_PWM_SET, /* sysfs */ + CPUCP_PACKET_FREQUENCY_SET, /* sysfs */ + CPUCP_PACKET_FREQUENCY_GET, /* sysfs */ + CPUCP_PACKET_LED_SET, /* debugfs */ + CPUCP_PACKET_I2C_WR, /* debugfs */ + CPUCP_PACKET_I2C_RD, /* debugfs */ + CPUCP_PACKET_INFO_GET, /* IOCTL */ + CPUCP_PACKET_FLASH_PROGRAM_REMOVED, + CPUCP_PACKET_UNMASK_RAZWI_IRQ, /* internal */ + CPUCP_PACKET_UNMASK_RAZWI_IRQ_ARRAY, /* internal */ + CPUCP_PACKET_TEST, /* internal */ + CPUCP_PACKET_FREQUENCY_CURR_GET, /* sysfs */ + CPUCP_PACKET_MAX_POWER_GET, /* sysfs */ + CPUCP_PACKET_MAX_POWER_SET, /* sysfs */ + CPUCP_PACKET_EEPROM_DATA_GET, /* sysfs */ + CPUCP_PACKET_NIC_INFO_GET, /* internal */ + CPUCP_PACKET_TEMPERATURE_SET, /* sysfs */ + CPUCP_PACKET_VOLTAGE_SET, /* sysfs */ + CPUCP_PACKET_CURRENT_SET, /* sysfs */ + CPUCP_PACKET_PCIE_THROUGHPUT_GET, /* internal */ + CPUCP_PACKET_PCIE_REPLAY_CNT_GET, /* internal */ + CPUCP_PACKET_TOTAL_ENERGY_GET, /* internal */ + CPUCP_PACKET_PLL_INFO_GET, /* internal */ + CPUCP_PACKET_NIC_STATUS, /* internal */ + CPUCP_PACKET_POWER_GET, /* internal */ + CPUCP_PACKET_NIC_PFC_SET, /* internal */ + CPUCP_PACKET_NIC_FAULT_GET, /* internal */ + CPUCP_PACKET_NIC_LPBK_SET, /* internal */ + CPUCP_PACKET_NIC_MAC_CFG, /* internal */ + CPUCP_PACKET_MSI_INFO_SET, /* internal */ + CPUCP_PACKET_NIC_XPCS91_REGS_GET, /* internal */ + CPUCP_PACKET_NIC_STAT_REGS_GET, /* internal */ + CPUCP_PACKET_NIC_STAT_REGS_CLR, /* internal */ + CPUCP_PACKET_NIC_STAT_REGS_ALL_GET, /* internal */ + CPUCP_PACKET_IS_IDLE_CHECK, /* internal */ + CPUCP_PACKET_HBM_REPLACED_ROWS_INFO_GET,/* internal */ + CPUCP_PACKET_HBM_PENDING_ROWS_STATUS, /* internal */ + CPUCP_PACKET_POWER_SET, /* internal */ + CPUCP_PACKET_RESERVED, /* not used */ + CPUCP_PACKET_ENGINE_CORE_ASID_SET, /* internal */ + CPUCP_PACKET_RESERVED2, /* not used */ + CPUCP_PACKET_SEC_ATTEST_GET, /* internal */ + CPUCP_PACKET_RESERVED3, /* not used */ + CPUCP_PACKET_RESERVED4, /* not used */ + CPUCP_PACKET_MONITOR_DUMP_GET, /* debugfs */ + CPUCP_PACKET_RESERVED5, /* not used */ + CPUCP_PACKET_RESERVED6, /* not used */ + CPUCP_PACKET_RESERVED7, /* not used */ + CPUCP_PACKET_RESERVED8, /* not used */ + CPUCP_PACKET_RESERVED9, /* not used */ + CPUCP_PACKET_ACTIVE_STATUS_SET, /* internal */ + CPUCP_PACKET_ID_MAX /* must be last */ +}; + +#define CPUCP_PACKET_FENCE_VAL 0xFE8CE7A5 + +#define CPUCP_PKT_CTL_RC_SHIFT 12 +#define CPUCP_PKT_CTL_RC_MASK 0x0000F000 + +#define CPUCP_PKT_CTL_OPCODE_SHIFT 16 +#define CPUCP_PKT_CTL_OPCODE_MASK 0x1FFF0000 + +#define CPUCP_PKT_RES_PLL_OUT0_SHIFT 0 +#define CPUCP_PKT_RES_PLL_OUT0_MASK 0x000000000000FFFFull +#define CPUCP_PKT_RES_PLL_OUT1_SHIFT 16 +#define CPUCP_PKT_RES_PLL_OUT1_MASK 0x00000000FFFF0000ull +#define CPUCP_PKT_RES_PLL_OUT2_SHIFT 32 +#define CPUCP_PKT_RES_PLL_OUT2_MASK 0x0000FFFF00000000ull +#define CPUCP_PKT_RES_PLL_OUT3_SHIFT 48 +#define CPUCP_PKT_RES_PLL_OUT3_MASK 0xFFFF000000000000ull + +#define CPUCP_PKT_VAL_PFC_IN1_SHIFT 0 +#define CPUCP_PKT_VAL_PFC_IN1_MASK 0x0000000000000001ull +#define CPUCP_PKT_VAL_PFC_IN2_SHIFT 1 +#define CPUCP_PKT_VAL_PFC_IN2_MASK 0x000000000000001Eull + +#define CPUCP_PKT_VAL_LPBK_IN1_SHIFT 0 +#define CPUCP_PKT_VAL_LPBK_IN1_MASK 0x0000000000000001ull +#define CPUCP_PKT_VAL_LPBK_IN2_SHIFT 1 +#define CPUCP_PKT_VAL_LPBK_IN2_MASK 0x000000000000001Eull + +#define CPUCP_PKT_VAL_MAC_CNT_IN1_SHIFT 0 +#define CPUCP_PKT_VAL_MAC_CNT_IN1_MASK 0x0000000000000001ull +#define CPUCP_PKT_VAL_MAC_CNT_IN2_SHIFT 1 +#define CPUCP_PKT_VAL_MAC_CNT_IN2_MASK 0x00000000FFFFFFFEull + +/* heartbeat status bits */ +#define CPUCP_PKT_HB_STATUS_EQ_FAULT_SHIFT 0 +#define CPUCP_PKT_HB_STATUS_EQ_FAULT_MASK 0x00000001 + +struct cpucp_packet { + union { + __le64 value; /* For SET packets */ + __le64 result; /* For GET packets */ + __le64 addr; /* For PQ */ + }; + + __le32 ctl; + + __le32 fence; /* Signal to host that message is completed */ + + union { + struct {/* For temperature/current/voltage/fan/pwm get/set */ + __le16 sensor_index; + __le16 type; + }; + + struct { /* For I2C read/write */ + __u8 i2c_bus; + __u8 i2c_addr; + __u8 i2c_reg; + /* + * In legacy implemetations, i2c_len was not present, + * was unused and just added as pad. + * So if i2c_len is 0, it is treated as legacy + * and r/w 1 Byte, else if i2c_len is specified, + * its treated as new multibyte r/w support. + */ + __u8 i2c_len; + }; + + struct {/* For PLL info fetch */ + __le16 pll_type; + /* TODO pll_reg is kept temporary before removal */ + __le16 pll_reg; + }; + + /* For any general request */ + __le32 index; + + /* For frequency get/set */ + __le32 pll_index; + + /* For led set */ + __le32 led_index; + + /* For get CpuCP info/EEPROM data/NIC info */ + __le32 data_max_size; + + /* + * For any general status bitmask. Shall be used whenever the + * result cannot be used to hold general purpose data. + */ + __le32 status_mask; + + /* random, used once number, for security packets */ + __le32 nonce; + }; + + /* For NIC requests */ + __le32 port_index; +}; + +struct cpucp_unmask_irq_arr_packet { + struct cpucp_packet cpucp_pkt; + __le32 length; + __le32 irqs[]; +}; + +struct cpucp_nic_status_packet { + struct cpucp_packet cpucp_pkt; + __le32 length; + __le32 data[]; +}; + +struct cpucp_array_data_packet { + struct cpucp_packet cpucp_pkt; + __le32 length; + __le32 data[]; +}; + +enum cpucp_led_index { + CPUCP_LED0_INDEX = 0, + CPUCP_LED1_INDEX, + CPUCP_LED2_INDEX +}; + +/* + * enum cpucp_packet_rc - Error return code + * @cpucp_packet_success -> in case of success. + * @cpucp_packet_invalid -> this is to support Goya and Gaudi platform. + * @cpucp_packet_fault -> in case of processing error like failing to + * get device binding or semaphore etc. + * @cpucp_packet_invalid_pkt -> when cpucp packet is un-supported. This is + * supported Greco onwards. + * @cpucp_packet_invalid_params -> when checking parameter like length of buffer + * or attribute value etc. Supported Greco onwards. + * @cpucp_packet_rc_max -> It indicates size of enum so should be at last. + */ +enum cpucp_packet_rc { + cpucp_packet_success, + cpucp_packet_invalid, + cpucp_packet_fault, + cpucp_packet_invalid_pkt, + cpucp_packet_invalid_params, + cpucp_packet_rc_max +}; + +/* + * cpucp_temp_type should adhere to hwmon_temp_attributes + * defined in Linux kernel hwmon.h file + */ +enum cpucp_temp_type { + cpucp_temp_input, + cpucp_temp_min = 4, + cpucp_temp_min_hyst, + cpucp_temp_max = 6, + cpucp_temp_max_hyst, + cpucp_temp_crit, + cpucp_temp_crit_hyst, + cpucp_temp_offset = 19, + cpucp_temp_lowest = 21, + cpucp_temp_highest = 22, + cpucp_temp_reset_history = 23, + cpucp_temp_warn = 24, + cpucp_temp_max_crit = 25, + cpucp_temp_max_warn = 26, +}; + +enum cpucp_in_attributes { + cpucp_in_input, + cpucp_in_min, + cpucp_in_max, + cpucp_in_lowest = 6, + cpucp_in_highest = 7, + cpucp_in_reset_history +}; + +enum cpucp_curr_attributes { + cpucp_curr_input, + cpucp_curr_min, + cpucp_curr_max, + cpucp_curr_lowest = 6, + cpucp_curr_highest = 7, + cpucp_curr_reset_history +}; + +enum cpucp_fan_attributes { + cpucp_fan_input, + cpucp_fan_min = 2, + cpucp_fan_max +}; + +enum cpucp_pwm_attributes { + cpucp_pwm_input, + cpucp_pwm_enable +}; + +enum cpucp_pcie_throughput_attributes { + cpucp_pcie_throughput_tx, + cpucp_pcie_throughput_rx +}; + +/* TODO temporary kept before removal */ +enum cpucp_pll_reg_attributes { + cpucp_pll_nr_reg, + cpucp_pll_nf_reg, + cpucp_pll_od_reg, + cpucp_pll_div_factor_reg, + cpucp_pll_div_sel_reg +}; + +/* TODO temporary kept before removal */ +enum cpucp_pll_type_attributes { + cpucp_pll_cpu, + cpucp_pll_pci, +}; + +/* + * cpucp_power_type aligns with hwmon_power_attributes + * defined in Linux kernel hwmon.h file + */ +enum cpucp_power_type { + CPUCP_POWER_INPUT = 8, + CPUCP_POWER_INPUT_HIGHEST = 9, + CPUCP_POWER_RESET_INPUT_HISTORY = 11 +}; + +/* + * MSI type enumeration table for all ASICs and future SW versions. + * For future ASIC-LKD compatibility, we can only add new enumerations. + * at the end of the table (before CPUCP_NUM_OF_MSI_TYPES). + * Changing the order of entries or removing entries is not allowed. + */ +enum cpucp_msi_type { + CPUCP_EVENT_QUEUE_MSI_TYPE, + CPUCP_NIC_PORT1_MSI_TYPE, + CPUCP_NIC_PORT3_MSI_TYPE, + CPUCP_NIC_PORT5_MSI_TYPE, + CPUCP_NIC_PORT7_MSI_TYPE, + CPUCP_NIC_PORT9_MSI_TYPE, + CPUCP_NUM_OF_MSI_TYPES +}; + +/* + * PLL enumeration table used for all ASICs and future SW versions. + * For future ASIC-LKD compatibility, we can only add new enumerations. + * at the end of the table. + * Changing the order of entries or removing entries is not allowed. + */ +enum pll_index { + CPU_PLL = 0, + PCI_PLL = 1, + NIC_PLL = 2, + DMA_PLL = 3, + MESH_PLL = 4, + MME_PLL = 5, + TPC_PLL = 6, + IF_PLL = 7, + SRAM_PLL = 8, + NS_PLL = 9, + HBM_PLL = 10, + MSS_PLL = 11, + DDR_PLL = 12, + VID_PLL = 13, + BANK_PLL = 14, + MMU_PLL = 15, + IC_PLL = 16, + MC_PLL = 17, + EMMC_PLL = 18, + PLL_MAX +}; + +enum rl_index { + TPC_RL = 0, + MME_RL, + EDMA_RL, +}; + +enum pvt_index { + PVT_SW, + PVT_SE, + PVT_NW, + PVT_NE +}; + +/* Event Queue Packets */ + +struct eq_generic_event { + __le64 data[7]; +}; + +/* + * CpuCP info + */ + +#define CARD_NAME_MAX_LEN 16 +#define CPUCP_MAX_SENSORS 128 +#define CPUCP_MAX_NICS 128 +#define CPUCP_LANES_PER_NIC 4 +#define CPUCP_NIC_QSFP_EEPROM_MAX_LEN 1024 +#define CPUCP_MAX_NIC_LANES (CPUCP_MAX_NICS * CPUCP_LANES_PER_NIC) +#define CPUCP_NIC_MASK_ARR_LEN ((CPUCP_MAX_NICS + 63) / 64) +#define CPUCP_NIC_POLARITY_ARR_LEN ((CPUCP_MAX_NIC_LANES + 63) / 64) +#define CPUCP_HBM_ROW_REPLACE_MAX 32 + +struct cpucp_sensor { + __le32 type; + __le32 flags; +}; + +/** + * struct cpucp_card_types - ASIC card type. + * @cpucp_card_type_pci: PCI card. + * @cpucp_card_type_pmc: PCI Mezzanine Card. + */ +enum cpucp_card_types { + cpucp_card_type_pci, + cpucp_card_type_pmc +}; + +#define CPUCP_SEC_CONF_ENABLED_SHIFT 0 +#define CPUCP_SEC_CONF_ENABLED_MASK 0x00000001 + +#define CPUCP_SEC_CONF_FLASH_WP_SHIFT 1 +#define CPUCP_SEC_CONF_FLASH_WP_MASK 0x00000002 + +#define CPUCP_SEC_CONF_EEPROM_WP_SHIFT 2 +#define CPUCP_SEC_CONF_EEPROM_WP_MASK 0x00000004 + +/** + * struct cpucp_security_info - Security information. + * @config: configuration bit field + * @keys_num: number of stored keys + * @revoked_keys: revoked keys bit field + * @min_svn: minimal security version + */ +struct cpucp_security_info { + __u8 config; + __u8 keys_num; + __u8 revoked_keys; + __u8 min_svn; +}; + +/** + * struct cpucp_info - Info from CpuCP that is necessary to the host's driver + * @sensors: available sensors description. + * @kernel_version: CpuCP linux kernel version. + * @reserved: reserved field. + * @card_type: card configuration type. + * @card_location: in a server, each card has different connections topology + * depending on its location (relevant for PMC card type) + * @cpld_version: CPLD programmed F/W version. + * @infineon_version: Infineon main DC-DC version. + * @fuse_version: silicon production FUSE information. + * @thermal_version: thermald S/W version. + * @cpucp_version: CpuCP S/W version. + * @infineon_second_stage_version: Infineon 2nd stage DC-DC version. + * @dram_size: available DRAM size. + * @card_name: card name that will be displayed in HWMON subsystem on the host + * @tpc_binning_mask: TPC binning mask, 1 bit per TPC instance + * (0 = functional, 1 = binned) + * @decoder_binning_mask: Decoder binning mask, 1 bit per decoder instance + * (0 = functional, 1 = binned), maximum 1 per dcore + * @sram_binning: Categorize SRAM functionality + * (0 = fully functional, 1 = lower-half is not functional, + * 2 = upper-half is not functional) + * @sec_info: security information + * @pll_map: Bit map of supported PLLs for current ASIC version. + * @mme_binning_mask: MME binning mask, + * bits [0:6] <==> dcore0 mme fma + * bits [7:13] <==> dcore1 mme fma + * bits [14:20] <==> dcore0 mme ima + * bits [21:27] <==> dcore1 mme ima + * For each group, if the 6th bit is set then first 5 bits + * represent the col's idx [0-31], otherwise these bits are + * ignored, and col idx 32 is binned. 7th bit is don't care. + * @dram_binning_mask: DRAM binning mask, 1 bit per dram instance + * (0 = functional 1 = binned) + * @memory_repair_flag: eFuse flag indicating memory repair + * @edma_binning_mask: EDMA binning mask, 1 bit per EDMA instance + * (0 = functional 1 = binned) + * @xbar_binning_mask: Xbar binning mask, 1 bit per Xbar instance + * (0 = functional 1 = binned) + * @interposer_version: Interposer version programmed in eFuse + * @substrate_version: Substrate version programmed in eFuse + * @fw_os_version: Firmware OS Version + */ +struct cpucp_info { + struct cpucp_sensor sensors[CPUCP_MAX_SENSORS]; + __u8 kernel_version[VERSION_MAX_LEN]; + __le32 reserved; + __le32 card_type; + __le32 card_location; + __le32 cpld_version; + __le32 infineon_version; + __u8 fuse_version[VERSION_MAX_LEN]; + __u8 thermal_version[VERSION_MAX_LEN]; + __u8 cpucp_version[VERSION_MAX_LEN]; + __le32 infineon_second_stage_version; + __le64 dram_size; + char card_name[CARD_NAME_MAX_LEN]; + __le64 tpc_binning_mask; + __le64 decoder_binning_mask; + __u8 sram_binning; + __u8 dram_binning_mask; + __u8 memory_repair_flag; + __u8 edma_binning_mask; + __u8 xbar_binning_mask; + __u8 interposer_version; + __u8 substrate_version; + __u8 reserved2; + struct cpucp_security_info sec_info; + __le32 reserved3; + __u8 pll_map[PLL_MAP_LEN]; + __le64 mme_binning_mask; + __u8 fw_os_version[VERSION_MAX_LEN]; +}; + +struct cpucp_mac_addr { + __u8 mac_addr[ETH_ALEN]; +}; + +enum cpucp_serdes_type { + TYPE_1_SERDES_TYPE, + TYPE_2_SERDES_TYPE, + HLS1_SERDES_TYPE, + HLS1H_SERDES_TYPE, + HLS2_SERDES_TYPE, + UNKNOWN_SERDES_TYPE, + MAX_NUM_SERDES_TYPE = UNKNOWN_SERDES_TYPE +}; + +struct cpucp_nic_info { + struct cpucp_mac_addr mac_addrs[CPUCP_MAX_NICS]; + __le64 link_mask[CPUCP_NIC_MASK_ARR_LEN]; + __le64 pol_tx_mask[CPUCP_NIC_POLARITY_ARR_LEN]; + __le64 pol_rx_mask[CPUCP_NIC_POLARITY_ARR_LEN]; + __le64 link_ext_mask[CPUCP_NIC_MASK_ARR_LEN]; + __u8 qsfp_eeprom[CPUCP_NIC_QSFP_EEPROM_MAX_LEN]; + __le64 auto_neg_mask[CPUCP_NIC_MASK_ARR_LEN]; + __le16 serdes_type; /* enum cpucp_serdes_type */ + __le16 tx_swap_map[CPUCP_MAX_NICS]; + __u8 reserved[6]; +}; + +#define PAGE_DISCARD_MAX 64 + +struct page_discard_info { + __u8 num_entries; + __u8 reserved[7]; + __le32 mmu_page_idx[PAGE_DISCARD_MAX]; +}; + +/* + * struct ser_val - the SER (symbol error rate) value is represented by "integer * 10 ^ -exp". + * @integer: the integer part of the SER value; + * @exp: the exponent part of the SER value. + */ +struct ser_val { + __le16 integer; + __le16 exp; +}; + +/* + * struct cpucp_nic_status - describes the status of a NIC port. + * @port: NIC port index. + * @bad_format_cnt: e.g. CRC. + * @responder_out_of_sequence_psn_cnt: e.g NAK. + * @high_ber_reinit_cnt: link reinit due to high BER. + * @correctable_err_cnt: e.g. bit-flip. + * @uncorrectable_err_cnt: e.g. MAC errors. + * @retraining_cnt: re-training counter. + * @up: is port up. + * @pcs_link: has PCS link. + * @phy_ready: is PHY ready. + * @auto_neg: is Autoneg enabled. + * @timeout_retransmission_cnt: timeout retransmission events + * @high_ber_cnt: high ber events + */ +struct cpucp_nic_status { + __le32 port; + __le32 bad_format_cnt; + __le32 responder_out_of_sequence_psn_cnt; + __le32 high_ber_reinit; + __le32 correctable_err_cnt; + __le32 uncorrectable_err_cnt; + __le32 retraining_cnt; + __u8 up; + __u8 pcs_link; + __u8 phy_ready; + __u8 auto_neg; + __le32 timeout_retransmission_cnt; + __le32 high_ber_cnt; +}; + +enum cpucp_hbm_row_replace_cause { + REPLACE_CAUSE_DOUBLE_ECC_ERR, + REPLACE_CAUSE_MULTI_SINGLE_ECC_ERR, +}; + +struct cpucp_hbm_row_info { + __u8 hbm_idx; + __u8 pc; + __u8 sid; + __u8 bank_idx; + __le16 row_addr; + __u8 replaced_row_cause; /* enum cpucp_hbm_row_replace_cause */ + __u8 pad; +}; + +struct cpucp_hbm_row_replaced_rows_info { + __le16 num_replaced_rows; + __u8 pad[6]; + struct cpucp_hbm_row_info replaced_rows[CPUCP_HBM_ROW_REPLACE_MAX]; +}; + +enum cpu_reset_status { + CPU_RST_STATUS_NA = 0, + CPU_RST_STATUS_SOFT_RST_DONE = 1, +}; + +#define SEC_PCR_DATA_BUF_SZ 256 +#define SEC_PCR_QUOTE_BUF_SZ 510 /* (512 - 2) 2 bytes used for size */ +#define SEC_SIGNATURE_BUF_SZ 255 /* (256 - 1) 1 byte used for size */ +#define SEC_PUB_DATA_BUF_SZ 510 /* (512 - 2) 2 bytes used for size */ +#define SEC_CERTIFICATE_BUF_SZ 2046 /* (2048 - 2) 2 bytes used for size */ + +/* + * struct cpucp_sec_attest_info - attestation report of the boot + * @pcr_data: raw values of the PCR registers + * @pcr_num_reg: number of PCR registers in the pcr_data array + * @pcr_reg_len: length of each PCR register in the pcr_data array (bytes) + * @nonce: number only used once. random number provided by host. this also + * passed to the quote command as a qualifying data. + * @pcr_quote_len: length of the attestation quote data (bytes) + * @pcr_quote: attestation report data structure + * @quote_sig_len: length of the attestation report signature (bytes) + * @quote_sig: signature structure of the attestation report + * @pub_data_len: length of the public data (bytes) + * @public_data: public key for the signed attestation + * (outPublic + name + qualifiedName) + * @certificate_len: length of the certificate (bytes) + * @certificate: certificate for the attestation signing key + */ +struct cpucp_sec_attest_info { + __u8 pcr_data[SEC_PCR_DATA_BUF_SZ]; + __u8 pcr_num_reg; + __u8 pcr_reg_len; + __le16 pad0; + __le32 nonce; + __le16 pcr_quote_len; + __u8 pcr_quote[SEC_PCR_QUOTE_BUF_SZ]; + __u8 quote_sig_len; + __u8 quote_sig[SEC_SIGNATURE_BUF_SZ]; + __le16 pub_data_len; + __u8 public_data[SEC_PUB_DATA_BUF_SZ]; + __le16 certificate_len; + __u8 certificate[SEC_CERTIFICATE_BUF_SZ]; +}; + +/* + * struct cpucp_dev_info_signed - device information signed by a secured device + * @info: device information structure as defined above + * @nonce: number only used once. random number provided by host. this number is + * hashed and signed along with the device information. + * @info_sig_len: length of the attestation signature (bytes) + * @info_sig: signature of the info + nonce data. + * @pub_data_len: length of the public data (bytes) + * @public_data: public key info signed info data + * (outPublic + name + qualifiedName) + * @certificate_len: length of the certificate (bytes) + * @certificate: certificate for the signing key + */ +struct cpucp_dev_info_signed { + struct cpucp_info info; /* assumed to be 64bit aligned */ + __le32 nonce; + __le32 pad0; + __u8 info_sig_len; + __u8 info_sig[SEC_SIGNATURE_BUF_SZ]; + __le16 pub_data_len; + __u8 public_data[SEC_PUB_DATA_BUF_SZ]; + __le16 certificate_len; + __u8 certificate[SEC_CERTIFICATE_BUF_SZ]; +}; + +/* + * struct dcore_monitor_regs_data - DCORE monitor regs data. + * the structure follows sync manager block layout. relevant only to Gaudi. + * @mon_pay_addrl: array of payload address low bits. + * @mon_pay_addrh: array of payload address high bits. + * @mon_pay_data: array of payload data. + * @mon_arm: array of monitor arm. + * @mon_status: array of monitor status. + */ +struct dcore_monitor_regs_data { + __le32 mon_pay_addrl[512]; + __le32 mon_pay_addrh[512]; + __le32 mon_pay_data[512]; + __le32 mon_arm[512]; + __le32 mon_status[512]; +}; + +/* contains SM data for each SYNC_MNGR (relevant only to Gaudi) */ +struct cpucp_monitor_dump { + struct dcore_monitor_regs_data sync_mngr_w_s; + struct dcore_monitor_regs_data sync_mngr_e_s; + struct dcore_monitor_regs_data sync_mngr_w_n; + struct dcore_monitor_regs_data sync_mngr_e_n; +}; + +#endif /* CPUCP_IF_H */ diff --git a/drivers/misc/habanalabs/include/common/hl_boot_if.h b/drivers/misc/habanalabs/include/common/hl_boot_if.h new file mode 100644 index 000000000..e0ea51cc7 --- /dev/null +++ b/drivers/misc/habanalabs/include/common/hl_boot_if.h @@ -0,0 +1,702 @@ +/* SPDX-License-Identifier: GPL-2.0 + * + * Copyright 2018-2020 HabanaLabs, Ltd. + * All Rights Reserved. + * + */ + +#ifndef HL_BOOT_IF_H +#define HL_BOOT_IF_H + +#define LKD_HARD_RESET_MAGIC 0xED7BD694 /* deprecated - do not use */ +#define HL_POWER9_HOST_MAGIC 0x1DA30009 + +#define BOOT_FIT_SRAM_OFFSET 0x200000 + +#define VERSION_MAX_LEN 128 + +enum cpu_boot_err { + CPU_BOOT_ERR_DRAM_INIT_FAIL = 0, + CPU_BOOT_ERR_FIT_CORRUPTED = 1, + CPU_BOOT_ERR_TS_INIT_FAIL = 2, + CPU_BOOT_ERR_DRAM_SKIPPED = 3, + CPU_BOOT_ERR_BMC_WAIT_SKIPPED = 4, + CPU_BOOT_ERR_NIC_DATA_NOT_RDY = 5, + CPU_BOOT_ERR_NIC_FW_FAIL = 6, + CPU_BOOT_ERR_SECURITY_NOT_RDY = 7, + CPU_BOOT_ERR_SECURITY_FAIL = 8, + CPU_BOOT_ERR_EFUSE_FAIL = 9, + CPU_BOOT_ERR_PRI_IMG_VER_FAIL = 10, + CPU_BOOT_ERR_SEC_IMG_VER_FAIL = 11, + CPU_BOOT_ERR_PLL_FAIL = 12, + CPU_BOOT_ERR_DEVICE_UNUSABLE_FAIL = 13, + CPU_BOOT_ERR_BOOT_FW_CRIT_ERR = 18, + CPU_BOOT_ERR_BINNING_FAIL = 19, + CPU_BOOT_ERR_TPM_FAIL = 20, + CPU_BOOT_ERR_TMP_THRESH_INIT_FAIL = 21, + CPU_BOOT_ERR_EEPROM_FAIL = 22, + CPU_BOOT_ERR_ENABLED = 31, + CPU_BOOT_ERR_SCND_EN = 63, + CPU_BOOT_ERR_LAST = 64 /* we have 2 registers of 32 bits */ +}; + +/* + * CPU error bits in BOOT_ERROR registers + * + * CPU_BOOT_ERR0_DRAM_INIT_FAIL DRAM initialization failed. + * DRAM is not reliable to use. + * + * CPU_BOOT_ERR0_FIT_CORRUPTED FIT data integrity verification of the + * image provided by the host has failed. + * + * CPU_BOOT_ERR0_TS_INIT_FAIL Thermal Sensor initialization failed. + * Boot continues as usual, but keep in + * mind this is a warning. + * + * CPU_BOOT_ERR0_DRAM_SKIPPED DRAM initialization has been skipped. + * Skipping DRAM initialization has been + * requested (e.g. strap, command, etc.) + * and FW skipped the DRAM initialization. + * Host can initialize the DRAM. + * + * CPU_BOOT_ERR0_BMC_WAIT_SKIPPED Waiting for BMC data will be skipped. + * Meaning the BMC data might not be + * available until reset. + * + * CPU_BOOT_ERR0_NIC_DATA_NOT_RDY NIC data from BMC is not ready. + * BMC has not provided the NIC data yet. + * Once provided this bit will be cleared. + * + * CPU_BOOT_ERR0_NIC_FW_FAIL NIC FW loading failed. + * The NIC FW loading and initialization + * failed. This means NICs are not usable. + * + * CPU_BOOT_ERR0_SECURITY_NOT_RDY Chip security initialization has been + * started, but is not ready yet - chip + * cannot be accessed. + * + * CPU_BOOT_ERR0_SECURITY_FAIL Security related tasks have failed. + * The tasks are security init (root of + * trust), boot authentication (chain of + * trust), data packets authentication. + * + * CPU_BOOT_ERR0_EFUSE_FAIL Reading from eFuse failed. + * The PCI device ID might be wrong. + * + * CPU_BOOT_ERR0_PRI_IMG_VER_FAIL Verification of primary image failed. + * It mean that ppboot checksum + * verification for the preboot primary + * image has failed to match expected + * checksum. Trying to program image again + * might solve this. + * + * CPU_BOOT_ERR0_SEC_IMG_VER_FAIL Verification of secondary image failed. + * It mean that ppboot checksum + * verification for the preboot secondary + * image has failed to match expected + * checksum. Trying to program image again + * might solve this. + * + * CPU_BOOT_ERR0_PLL_FAIL PLL settings failed, meaning that one + * of the PLLs remains in REF_CLK + * + * CPU_BOOT_ERR0_DEVICE_UNUSABLE_FAIL Device is unusable and customer support + * should be contacted. + * + * CPU_BOOT_ERR0_BOOT_FW_CRIT_ERR Critical error was detected during + * the execution of ppboot or preboot. + * for example: stack overflow. + * + * CPU_BOOT_ERR0_BINNING_FAIL Binning settings failed, meaning + * malfunctioning components might still be + * in use. + * + * CPU_BOOT_ERR0_TPM_FAIL TPM verification flow failed. + * + * CPU_BOOT_ERR0_TMP_THRESH_INIT_FAIL Failed to set threshold for tmperature + * sensor. + * + * CPU_BOOT_ERR_EEPROM_FAIL Failed reading EEPROM data. Defaults + * are used. + * + * CPU_BOOT_ERR0_ENABLED Error registers enabled. + * This is a main indication that the + * running FW populates the error + * registers. Meaning the error bits are + * not garbage, but actual error statuses. + */ +#define CPU_BOOT_ERR0_DRAM_INIT_FAIL (1 << CPU_BOOT_ERR_DRAM_INIT_FAIL) +#define CPU_BOOT_ERR0_FIT_CORRUPTED (1 << CPU_BOOT_ERR_FIT_CORRUPTED) +#define CPU_BOOT_ERR0_TS_INIT_FAIL (1 << CPU_BOOT_ERR_TS_INIT_FAIL) +#define CPU_BOOT_ERR0_DRAM_SKIPPED (1 << CPU_BOOT_ERR_DRAM_SKIPPED) +#define CPU_BOOT_ERR0_BMC_WAIT_SKIPPED (1 << CPU_BOOT_ERR_BMC_WAIT_SKIPPED) +#define CPU_BOOT_ERR0_NIC_DATA_NOT_RDY (1 << CPU_BOOT_ERR_NIC_DATA_NOT_RDY) +#define CPU_BOOT_ERR0_NIC_FW_FAIL (1 << CPU_BOOT_ERR_NIC_FW_FAIL) +#define CPU_BOOT_ERR0_SECURITY_NOT_RDY (1 << CPU_BOOT_ERR_SECURITY_NOT_RDY) +#define CPU_BOOT_ERR0_SECURITY_FAIL (1 << CPU_BOOT_ERR_SECURITY_FAIL) +#define CPU_BOOT_ERR0_EFUSE_FAIL (1 << CPU_BOOT_ERR_EFUSE_FAIL) +#define CPU_BOOT_ERR0_PRI_IMG_VER_FAIL (1 << CPU_BOOT_ERR_PRI_IMG_VER_FAIL) +#define CPU_BOOT_ERR0_SEC_IMG_VER_FAIL (1 << CPU_BOOT_ERR_SEC_IMG_VER_FAIL) +#define CPU_BOOT_ERR0_PLL_FAIL (1 << CPU_BOOT_ERR_PLL_FAIL) +#define CPU_BOOT_ERR0_DEVICE_UNUSABLE_FAIL (1 << CPU_BOOT_ERR_DEVICE_UNUSABLE_FAIL) +#define CPU_BOOT_ERR0_BOOT_FW_CRIT_ERR (1 << CPU_BOOT_ERR_BOOT_FW_CRIT_ERR) +#define CPU_BOOT_ERR0_BINNING_FAIL (1 << CPU_BOOT_ERR_BINNING_FAIL) +#define CPU_BOOT_ERR0_TPM_FAIL (1 << CPU_BOOT_ERR_TPM_FAIL) +#define CPU_BOOT_ERR0_TMP_THRESH_INIT_FAIL (1 << CPU_BOOT_ERR_TMP_THRESH_INIT_FAIL) +#define CPU_BOOT_ERR0_EEPROM_FAIL (1 << CPU_BOOT_ERR_EEPROM_FAIL) +#define CPU_BOOT_ERR0_ENABLED (1 << CPU_BOOT_ERR_ENABLED) +#define CPU_BOOT_ERR1_ENABLED (1 << CPU_BOOT_ERR_ENABLED) + +enum cpu_boot_dev_sts { + CPU_BOOT_DEV_STS_SECURITY_EN = 0, + CPU_BOOT_DEV_STS_DEBUG_EN = 1, + CPU_BOOT_DEV_STS_WATCHDOG_EN = 2, + CPU_BOOT_DEV_STS_DRAM_INIT_EN = 3, + CPU_BOOT_DEV_STS_BMC_WAIT_EN = 4, + CPU_BOOT_DEV_STS_E2E_CRED_EN = 5, + CPU_BOOT_DEV_STS_HBM_CRED_EN = 6, + CPU_BOOT_DEV_STS_RL_EN = 7, + CPU_BOOT_DEV_STS_SRAM_SCR_EN = 8, + CPU_BOOT_DEV_STS_DRAM_SCR_EN = 9, + CPU_BOOT_DEV_STS_FW_HARD_RST_EN = 10, + CPU_BOOT_DEV_STS_PLL_INFO_EN = 11, + CPU_BOOT_DEV_STS_SP_SRAM_EN = 12, + CPU_BOOT_DEV_STS_CLK_GATE_EN = 13, + CPU_BOOT_DEV_STS_HBM_ECC_EN = 14, + CPU_BOOT_DEV_STS_PKT_PI_ACK_EN = 15, + CPU_BOOT_DEV_STS_FW_LD_COM_EN = 16, + CPU_BOOT_DEV_STS_FW_IATU_CONF_EN = 17, + CPU_BOOT_DEV_STS_FW_NIC_MAC_EN = 18, + CPU_BOOT_DEV_STS_DYN_PLL_EN = 19, + CPU_BOOT_DEV_STS_GIC_PRIVILEGED_EN = 20, + CPU_BOOT_DEV_STS_EQ_INDEX_EN = 21, + CPU_BOOT_DEV_STS_MULTI_IRQ_POLL_EN = 22, + CPU_BOOT_DEV_STS_FW_NIC_STAT_XPCS91_EN = 23, + CPU_BOOT_DEV_STS_FW_NIC_STAT_EXT_EN = 24, + CPU_BOOT_DEV_STS_IS_IDLE_CHECK_EN = 25, + CPU_BOOT_DEV_STS_MAP_HWMON_EN = 26, + CPU_BOOT_DEV_STS_ENABLED = 31, + CPU_BOOT_DEV_STS_SCND_EN = 63, + CPU_BOOT_DEV_STS_LAST = 64 /* we have 2 registers of 32 bits */ +}; + +/* + * BOOT DEVICE STATUS bits in BOOT_DEVICE_STS registers + * + * CPU_BOOT_DEV_STS0_SECURITY_EN Security is Enabled. + * This is an indication for security + * enabled in FW, which means that + * all conditions for security are met: + * device is indicated as security enabled, + * registers are protected, and device + * uses keys for image verification. + * Initialized in: preboot + * + * CPU_BOOT_DEV_STS0_DEBUG_EN Debug is enabled. + * Enabled when JTAG or DEBUG is enabled + * in FW. + * Initialized in: preboot + * + * CPU_BOOT_DEV_STS0_WATCHDOG_EN Watchdog is enabled. + * Watchdog is enabled in FW. + * Initialized in: preboot + * + * CPU_BOOT_DEV_STS0_DRAM_INIT_EN DRAM initialization is enabled. + * DRAM initialization has been done in FW. + * Initialized in: u-boot + * + * CPU_BOOT_DEV_STS0_BMC_WAIT_EN Waiting for BMC data enabled. + * If set, it means that during boot, + * FW waited for BMC data. + * Initialized in: u-boot + * + * CPU_BOOT_DEV_STS0_E2E_CRED_EN E2E credits initialized. + * FW initialized E2E credits. + * Initialized in: u-boot + * + * CPU_BOOT_DEV_STS0_HBM_CRED_EN HBM credits initialized. + * FW initialized HBM credits. + * Initialized in: u-boot + * + * CPU_BOOT_DEV_STS0_RL_EN Rate limiter initialized. + * FW initialized rate limiter. + * Initialized in: u-boot + * + * CPU_BOOT_DEV_STS0_SRAM_SCR_EN SRAM scrambler enabled. + * FW initialized SRAM scrambler. + * Initialized in: linux + * + * CPU_BOOT_DEV_STS0_DRAM_SCR_EN DRAM scrambler enabled. + * FW initialized DRAM scrambler. + * Initialized in: u-boot + * + * CPU_BOOT_DEV_STS0_FW_HARD_RST_EN FW hard reset procedure is enabled. + * FW has the hard reset procedure + * implemented. This means that FW will + * perform hard reset procedure on + * receiving the halt-machine event. + * Initialized in: preboot, u-boot, linux + * + * CPU_BOOT_DEV_STS0_PLL_INFO_EN FW retrieval of PLL info is enabled. + * Initialized in: linux + * + * CPU_BOOT_DEV_STS0_SP_SRAM_EN SP SRAM is initialized and available + * for use. + * Initialized in: preboot + * + * CPU_BOOT_DEV_STS0_CLK_GATE_EN Clock Gating enabled. + * FW initialized Clock Gating. + * Initialized in: preboot + * + * CPU_BOOT_DEV_STS0_HBM_ECC_EN HBM ECC handling Enabled. + * FW handles HBM ECC indications. + * Initialized in: linux + * + * CPU_BOOT_DEV_STS0_PKT_PI_ACK_EN Packets ack value used in the armcpd + * is set to the PI counter. + * Initialized in: linux + * + * CPU_BOOT_DEV_STS0_FW_LD_COM_EN Flexible FW loading communication + * protocol is enabled. + * Initialized in: preboot + * + * CPU_BOOT_DEV_STS0_FW_IATU_CONF_EN FW iATU configuration is enabled. + * This bit if set, means the iATU has been + * configured and is ready for use. + * Initialized in: ppboot + * + * CPU_BOOT_DEV_STS0_FW_NIC_MAC_EN NIC MAC channels init is done by FW and + * any access to them is done via the FW. + * Initialized in: linux + * + * CPU_BOOT_DEV_STS0_DYN_PLL_EN Dynamic PLL configuration is enabled. + * FW sends to host a bitmap of supported + * PLLs. + * Initialized in: linux + * + * CPU_BOOT_DEV_STS0_GIC_PRIVILEGED_EN GIC access permission only from + * previleged entity. FW sets this status + * bit for host. If this bit is set then + * GIC can not be accessed from host. + * Initialized in: linux + * + * CPU_BOOT_DEV_STS0_EQ_INDEX_EN Event Queue (EQ) index is a running + * index for each new event sent to host. + * This is used as a method in host to + * identify that the waiting event in + * queue is actually a new event which + * was not served before. + * Initialized in: linux + * + * CPU_BOOT_DEV_STS0_MULTI_IRQ_POLL_EN Use multiple scratchpad interfaces to + * prevent IRQs overriding each other. + * Initialized in: linux + * + * CPU_BOOT_DEV_STS0_FW_NIC_STAT_XPCS91_EN + * NIC STAT and XPCS91 access is restricted + * and is done via FW only. + * Initialized in: linux + * + * CPU_BOOT_DEV_STS0_FW_NIC_STAT_EXT_EN + * NIC STAT get all is supported. + * Initialized in: linux + * + * CPU_BOOT_DEV_STS0_IS_IDLE_CHECK_EN + * F/W checks if the device is idle by reading defined set + * of registers. It returns a bitmask of all the engines, + * where a bit is set if the engine is not idle. + * Initialized in: linux + * + * CPU_BOOT_DEV_STS0_MAP_HWMON_EN + * If set, means f/w supports proprietary + * HWMON enum mapping to cpucp enums. + * Initialized in: linux + * + * CPU_BOOT_DEV_STS0_ENABLED Device status register enabled. + * This is a main indication that the + * running FW populates the device status + * register. Meaning the device status + * bits are not garbage, but actual + * statuses. + * Initialized in: preboot + * + */ +#define CPU_BOOT_DEV_STS0_SECURITY_EN (1 << CPU_BOOT_DEV_STS_SECURITY_EN) +#define CPU_BOOT_DEV_STS0_DEBUG_EN (1 << CPU_BOOT_DEV_STS_DEBUG_EN) +#define CPU_BOOT_DEV_STS0_WATCHDOG_EN (1 << CPU_BOOT_DEV_STS_WATCHDOG_EN) +#define CPU_BOOT_DEV_STS0_DRAM_INIT_EN (1 << CPU_BOOT_DEV_STS_DRAM_INIT_EN) +#define CPU_BOOT_DEV_STS0_BMC_WAIT_EN (1 << CPU_BOOT_DEV_STS_BMC_WAIT_EN) +#define CPU_BOOT_DEV_STS0_E2E_CRED_EN (1 << CPU_BOOT_DEV_STS_E2E_CRED_EN) +#define CPU_BOOT_DEV_STS0_HBM_CRED_EN (1 << CPU_BOOT_DEV_STS_HBM_CRED_EN) +#define CPU_BOOT_DEV_STS0_RL_EN (1 << CPU_BOOT_DEV_STS_RL_EN) +#define CPU_BOOT_DEV_STS0_SRAM_SCR_EN (1 << CPU_BOOT_DEV_STS_SRAM_SCR_EN) +#define CPU_BOOT_DEV_STS0_DRAM_SCR_EN (1 << CPU_BOOT_DEV_STS_DRAM_SCR_EN) +#define CPU_BOOT_DEV_STS0_FW_HARD_RST_EN (1 << CPU_BOOT_DEV_STS_FW_HARD_RST_EN) +#define CPU_BOOT_DEV_STS0_PLL_INFO_EN (1 << CPU_BOOT_DEV_STS_PLL_INFO_EN) +#define CPU_BOOT_DEV_STS0_SP_SRAM_EN (1 << CPU_BOOT_DEV_STS_SP_SRAM_EN) +#define CPU_BOOT_DEV_STS0_CLK_GATE_EN (1 << CPU_BOOT_DEV_STS_CLK_GATE_EN) +#define CPU_BOOT_DEV_STS0_HBM_ECC_EN (1 << CPU_BOOT_DEV_STS_HBM_ECC_EN) +#define CPU_BOOT_DEV_STS0_PKT_PI_ACK_EN (1 << CPU_BOOT_DEV_STS_PKT_PI_ACK_EN) +#define CPU_BOOT_DEV_STS0_FW_LD_COM_EN (1 << CPU_BOOT_DEV_STS_FW_LD_COM_EN) +#define CPU_BOOT_DEV_STS0_FW_IATU_CONF_EN (1 << CPU_BOOT_DEV_STS_FW_IATU_CONF_EN) +#define CPU_BOOT_DEV_STS0_FW_NIC_MAC_EN (1 << CPU_BOOT_DEV_STS_FW_NIC_MAC_EN) +#define CPU_BOOT_DEV_STS0_DYN_PLL_EN (1 << CPU_BOOT_DEV_STS_DYN_PLL_EN) +#define CPU_BOOT_DEV_STS0_GIC_PRIVILEGED_EN (1 << CPU_BOOT_DEV_STS_GIC_PRIVILEGED_EN) +#define CPU_BOOT_DEV_STS0_EQ_INDEX_EN (1 << CPU_BOOT_DEV_STS_EQ_INDEX_EN) +#define CPU_BOOT_DEV_STS0_MULTI_IRQ_POLL_EN (1 << CPU_BOOT_DEV_STS_MULTI_IRQ_POLL_EN) +#define CPU_BOOT_DEV_STS0_FW_NIC_STAT_XPCS91_EN (1 << CPU_BOOT_DEV_STS_FW_NIC_STAT_XPCS91_EN) +#define CPU_BOOT_DEV_STS0_FW_NIC_STAT_EXT_EN (1 << CPU_BOOT_DEV_STS_FW_NIC_STAT_EXT_EN) +#define CPU_BOOT_DEV_STS0_IS_IDLE_CHECK_EN (1 << CPU_BOOT_DEV_STS_IS_IDLE_CHECK_EN) +#define CPU_BOOT_DEV_STS0_MAP_HWMON_EN (1 << CPU_BOOT_DEV_STS_MAP_HWMON_EN) +#define CPU_BOOT_DEV_STS0_ENABLED (1 << CPU_BOOT_DEV_STS_ENABLED) +#define CPU_BOOT_DEV_STS1_ENABLED (1 << CPU_BOOT_DEV_STS_ENABLED) + +enum cpu_boot_status { + CPU_BOOT_STATUS_NA = 0, /* Default value after reset of chip */ + CPU_BOOT_STATUS_IN_WFE = 1, + CPU_BOOT_STATUS_DRAM_RDY = 2, + CPU_BOOT_STATUS_SRAM_AVAIL = 3, + CPU_BOOT_STATUS_IN_BTL = 4, /* BTL is H/W FSM */ + CPU_BOOT_STATUS_IN_PREBOOT = 5, + CPU_BOOT_STATUS_IN_SPL, /* deprecated - not reported */ + CPU_BOOT_STATUS_IN_UBOOT = 7, + CPU_BOOT_STATUS_DRAM_INIT_FAIL, /* deprecated - will be removed */ + CPU_BOOT_STATUS_FIT_CORRUPTED, /* deprecated - will be removed */ + /* U-Boot console prompt activated, commands are not processed */ + CPU_BOOT_STATUS_UBOOT_NOT_READY = 10, + /* Finished NICs init, reported after DRAM and NICs */ + CPU_BOOT_STATUS_NIC_FW_RDY = 11, + CPU_BOOT_STATUS_TS_INIT_FAIL, /* deprecated - will be removed */ + CPU_BOOT_STATUS_DRAM_SKIPPED, /* deprecated - will be removed */ + CPU_BOOT_STATUS_BMC_WAITING_SKIPPED, /* deprecated - will be removed */ + /* Last boot loader progress status, ready to receive commands */ + CPU_BOOT_STATUS_READY_TO_BOOT = 15, + /* Internal Boot finished, ready for boot-fit */ + CPU_BOOT_STATUS_WAITING_FOR_BOOT_FIT = 16, + /* Internal Security has been initialized, device can be accessed */ + CPU_BOOT_STATUS_SECURITY_READY = 17, +}; + +enum kmd_msg { + KMD_MSG_NA = 0, + KMD_MSG_GOTO_WFE, + KMD_MSG_FIT_RDY, + KMD_MSG_SKIP_BMC, + RESERVED, + KMD_MSG_RST_DEV, + KMD_MSG_LAST +}; + +enum cpu_msg_status { + CPU_MSG_CLR = 0, + CPU_MSG_OK, + CPU_MSG_ERR, +}; + +/* communication registers mapping - consider ABI when changing */ +struct cpu_dyn_regs { + __le32 cpu_pq_base_addr_low; + __le32 cpu_pq_base_addr_high; + __le32 cpu_pq_length; + __le32 cpu_pq_init_status; + __le32 cpu_eq_base_addr_low; + __le32 cpu_eq_base_addr_high; + __le32 cpu_eq_length; + __le32 cpu_eq_ci; + __le32 cpu_cq_base_addr_low; + __le32 cpu_cq_base_addr_high; + __le32 cpu_cq_length; + __le32 cpu_pf_pq_pi; + __le32 cpu_boot_dev_sts0; + __le32 cpu_boot_dev_sts1; + __le32 cpu_boot_err0; + __le32 cpu_boot_err1; + __le32 cpu_boot_status; + __le32 fw_upd_sts; + __le32 fw_upd_cmd; + __le32 fw_upd_pending_sts; + __le32 fuse_ver_offset; + __le32 preboot_ver_offset; + __le32 uboot_ver_offset; + __le32 hw_state; + __le32 kmd_msg_to_cpu; + __le32 cpu_cmd_status_to_host; + __le32 gic_host_pi_upd_irq; + __le32 gic_tpc_qm_irq_ctrl; + __le32 gic_mme_qm_irq_ctrl; + __le32 gic_dma_qm_irq_ctrl; + __le32 gic_nic_qm_irq_ctrl; + __le32 gic_dma_core_irq_ctrl; + __le32 gic_host_halt_irq; + __le32 gic_host_ints_irq; + __le32 gic_host_soft_rst_irq; + __le32 gic_rot_qm_irq_ctrl; + __le32 cpu_rst_status; + __le32 eng_arc_irq_ctrl; + __le32 reserved1[20]; /* reserve for future use */ +}; + +/* TODO: remove the desc magic after the code is updated to use message */ +/* HCDM - Habana Communications Descriptor Magic */ +#define HL_COMMS_DESC_MAGIC 0x4843444D +#define HL_COMMS_DESC_VER 1 + +/* HCMv - Habana Communications Message + header version */ +#define HL_COMMS_MSG_MAGIC_VALUE 0x48434D00 +#define HL_COMMS_MSG_MAGIC_MASK 0xFFFFFF00 +#define HL_COMMS_MSG_MAGIC_VER_MASK 0xFF + +#define HL_COMMS_MSG_MAGIC_VER(ver) (HL_COMMS_MSG_MAGIC_VALUE | \ + ((ver) & HL_COMMS_MSG_MAGIC_VER_MASK)) +#define HL_COMMS_MSG_MAGIC_V0 HL_COMMS_DESC_MAGIC +#define HL_COMMS_MSG_MAGIC_V1 HL_COMMS_MSG_MAGIC_VER(1) + +#define HL_COMMS_MSG_MAGIC HL_COMMS_MSG_MAGIC_V1 + +#define HL_COMMS_MSG_MAGIC_VALIDATE_MAGIC(magic) \ + (((magic) & HL_COMMS_MSG_MAGIC_MASK) == \ + HL_COMMS_MSG_MAGIC_VALUE) + +#define HL_COMMS_MSG_MAGIC_VALIDATE_VERSION(magic, ver) \ + (((magic) & HL_COMMS_MSG_MAGIC_VER_MASK) >= \ + ((ver) & HL_COMMS_MSG_MAGIC_VER_MASK)) + +#define HL_COMMS_MSG_MAGIC_VALIDATE(magic, ver) \ + (HL_COMMS_MSG_MAGIC_VALIDATE_MAGIC((magic)) && \ + HL_COMMS_MSG_MAGIC_VALIDATE_VERSION((magic), (ver))) + +enum comms_msg_type { + HL_COMMS_DESC_TYPE = 0, + HL_COMMS_RESET_CAUSE_TYPE = 1, + HL_COMMS_FW_CFG_SKIP_TYPE = 2, + HL_COMMS_BINNING_CONF_TYPE = 3, +}; + +/* + * Binning information shared between LKD and FW + * @tpc_mask - TPC binning information + * @dec_mask - Decoder binning information + * @hbm_mask - HBM binning information + * @edma_mask - EDMA binning information + * @mme_mask_l - MME binning information lower 32 + * @mme_mask_h - MME binning information upper 32 + * @reserved - reserved field for 64 bit alignment + */ +struct lkd_fw_binning_info { + __le64 tpc_mask; + __le32 dec_mask; + __le32 hbm_mask; + __le32 edma_mask; + __le32 mme_mask_l; + __le32 mme_mask_h; + __le32 reserved; +}; + +/* TODO: remove this struct after the code is updated to use message */ +/* this is the comms descriptor header - meta data */ +struct comms_desc_header { + __le32 magic; /* magic for validation */ + __le32 crc32; /* CRC32 of the descriptor w/o header */ + __le16 size; /* size of the descriptor w/o header */ + __u8 version; /* descriptor version */ + __u8 reserved[5]; /* pad to 64 bit */ +}; + +/* this is the comms message header - meta data */ +struct comms_msg_header { + __le32 magic; /* magic for validation */ + __le32 crc32; /* CRC32 of the message w/o header */ + __le16 size; /* size of the message w/o header */ + __u8 version; /* message payload version */ + __u8 type; /* message type */ + __u8 reserved[4]; /* pad to 64 bit */ +}; + +/* this is the main FW descriptor - consider ABI when changing */ +struct lkd_fw_comms_desc { + struct comms_desc_header header; + struct cpu_dyn_regs cpu_dyn_regs; + char fuse_ver[VERSION_MAX_LEN]; + char cur_fw_ver[VERSION_MAX_LEN]; + /* can be used for 1 more version w/o ABI change */ + char reserved0[VERSION_MAX_LEN]; + __le64 img_addr; /* address for next FW component load */ +}; + +enum comms_reset_cause { + HL_RESET_CAUSE_UNKNOWN = 0, + HL_RESET_CAUSE_HEARTBEAT = 1, + HL_RESET_CAUSE_TDR = 2, +}; + +/* TODO: remove define after struct name is aligned on all projects */ +#define lkd_msg_comms lkd_fw_comms_msg + +/* this is the comms message descriptor */ +struct lkd_fw_comms_msg { + struct comms_msg_header header; + /* union for future expantions of new messages */ + union { + struct { + struct cpu_dyn_regs cpu_dyn_regs; + char fuse_ver[VERSION_MAX_LEN]; + char cur_fw_ver[VERSION_MAX_LEN]; + /* can be used for 1 more version w/o ABI change */ + char reserved0[VERSION_MAX_LEN]; + /* address for next FW component load */ + __le64 img_addr; + }; + struct { + __u8 reset_cause; + }; + struct { + __u8 fw_cfg_skip; /* 1 - skip, 0 - don't skip */ + }; + struct lkd_fw_binning_info binning_info; + }; +}; + +/* + * LKD commands: + * + * COMMS_NOOP Used to clear the command register and no actual + * command is send. + * + * COMMS_CLR_STS Clear status command - FW should clear the + * status register. Used for synchronization + * between the commands as part of the race free + * protocol. + * + * COMMS_RST_STATE Reset the current communication state which is + * kept by FW for proper responses. + * Should be used in the beginning of the + * communication cycle to clean any leftovers from + * previous communication attempts. + * + * COMMS_PREP_DESC Prepare descriptor for setting up the + * communication and other dynamic data: + * struct lkd_fw_comms_desc. + * This command has a parameter stating the next FW + * component size, so the FW can actually prepare a + * space for it and in the status response provide + * the descriptor offset. The Offset of the next FW + * data component is a part of the descriptor + * structure. + * + * COMMS_DATA_RDY The FW data has been uploaded and is ready for + * validation. + * + * COMMS_EXEC Execute the next FW component. + * + * COMMS_RST_DEV Reset the device. + * + * COMMS_GOTO_WFE Execute WFE command. Allowed only on non-secure + * devices. + * + * COMMS_SKIP_BMC Perform actions required for BMC-less servers. + * Do not wait for BMC response. + * + * COMMS_PREP_DESC_ELBI Same as COMMS_PREP_DESC only that the memory + * space is allocated in a ELBI access only + * address range. + * + */ +enum comms_cmd { + COMMS_NOOP = 0, + COMMS_CLR_STS = 1, + COMMS_RST_STATE = 2, + COMMS_PREP_DESC = 3, + COMMS_DATA_RDY = 4, + COMMS_EXEC = 5, + COMMS_RST_DEV = 6, + COMMS_GOTO_WFE = 7, + COMMS_SKIP_BMC = 8, + COMMS_PREP_DESC_ELBI = 10, + COMMS_INVLD_LAST +}; + +#define COMMS_COMMAND_SIZE_SHIFT 0 +#define COMMS_COMMAND_SIZE_MASK 0x1FFFFFF +#define COMMS_COMMAND_CMD_SHIFT 27 +#define COMMS_COMMAND_CMD_MASK 0xF8000000 + +/* + * LKD command to FW register structure + * @size - FW component size + * @cmd - command from enum comms_cmd + */ +struct comms_command { + union { /* bit fields are only for FW use */ + struct { + u32 size :25; /* 32MB max. */ + u32 reserved :2; + enum comms_cmd cmd :5; /* 32 commands */ + }; + __le32 val; + }; +}; + +/* + * FW status + * + * COMMS_STS_NOOP Used to clear the status register and no actual + * status is provided. + * + * COMMS_STS_ACK Command has been received and recognized. + * + * COMMS_STS_OK Command execution has finished successfully. + * + * COMMS_STS_ERR Command execution was unsuccessful and resulted + * in error. + * + * COMMS_STS_VALID_ERR FW validation has failed. + * + * COMMS_STS_TIMEOUT_ERR Command execution has timed out. + */ +enum comms_sts { + COMMS_STS_NOOP = 0, + COMMS_STS_ACK = 1, + COMMS_STS_OK = 2, + COMMS_STS_ERR = 3, + COMMS_STS_VALID_ERR = 4, + COMMS_STS_TIMEOUT_ERR = 5, + COMMS_STS_INVLD_LAST +}; + +/* RAM types for FW components loading - defines the base address */ +enum comms_ram_types { + COMMS_SRAM = 0, + COMMS_DRAM = 1, +}; + +#define COMMS_STATUS_OFFSET_SHIFT 0 +#define COMMS_STATUS_OFFSET_MASK 0x03FFFFFF +#define COMMS_STATUS_OFFSET_ALIGN_SHIFT 2 +#define COMMS_STATUS_RAM_TYPE_SHIFT 26 +#define COMMS_STATUS_RAM_TYPE_MASK 0x0C000000 +#define COMMS_STATUS_STATUS_SHIFT 28 +#define COMMS_STATUS_STATUS_MASK 0xF0000000 + +/* + * FW status to LKD register structure + * @offset - an offset from the base of the ram_type shifted right by + * 2 bits (always aligned to 32 bits). + * Allows a maximum addressable offset of 256MB from RAM base. + * Example: for real offset in RAM of 0x800000 (8MB), the value + * in offset field is (0x800000 >> 2) = 0x200000. + * @ram_type - the RAM type that should be used for offset from + * enum comms_ram_types + * @status - status from enum comms_sts + */ +struct comms_status { + union { /* bit fields are only for FW use */ + struct { + u32 offset :26; + enum comms_ram_types ram_type :2; + enum comms_sts status :4; /* 16 statuses */ + }; + __le32 val; + }; +}; + +#endif /* HL_BOOT_IF_H */ diff --git a/drivers/misc/habanalabs/include/common/qman_if.h b/drivers/misc/habanalabs/include/common/qman_if.h new file mode 100644 index 000000000..7ed773957 --- /dev/null +++ b/drivers/misc/habanalabs/include/common/qman_if.h @@ -0,0 +1,68 @@ +/* SPDX-License-Identifier: GPL-2.0 + * + * Copyright 2016-2018 HabanaLabs, Ltd. + * All Rights Reserved. + * + */ + +#ifndef QMAN_IF_H +#define QMAN_IF_H + +#include <linux/types.h> + +/* + * PRIMARY QUEUE + */ + +struct hl_bd { + __le64 ptr; + __le32 len; + __le32 ctl; +}; + +#define HL_BD_SIZE sizeof(struct hl_bd) + +/* + * S/W CTL FIELDS. + * + * BD_CTL_REPEAT_VALID tells the CP whether the repeat field in the BD CTL is + * valid. 1 means the repeat field is valid, 0 means not-valid, + * i.e. repeat == 1 + */ +#define BD_CTL_REPEAT_VALID_SHIFT 24 +#define BD_CTL_REPEAT_VALID_MASK 0x01000000 + +#define BD_CTL_SHADOW_INDEX_SHIFT 0 +#define BD_CTL_SHADOW_INDEX_MASK 0x00000FFF + +/* + * H/W CTL FIELDS + */ + +#define BD_CTL_COMP_OFFSET_SHIFT 16 +#define BD_CTL_COMP_OFFSET_MASK 0x0FFF0000 + +#define BD_CTL_COMP_DATA_SHIFT 0 +#define BD_CTL_COMP_DATA_MASK 0x0000FFFF + +/* + * COMPLETION QUEUE + */ + +struct hl_cq_entry { + __le32 data; +}; + +#define HL_CQ_ENTRY_SIZE sizeof(struct hl_cq_entry) + +#define CQ_ENTRY_READY_SHIFT 31 +#define CQ_ENTRY_READY_MASK 0x80000000 + +#define CQ_ENTRY_SHADOW_INDEX_VALID_SHIFT 30 +#define CQ_ENTRY_SHADOW_INDEX_VALID_MASK 0x40000000 + +#define CQ_ENTRY_SHADOW_INDEX_SHIFT BD_CTL_SHADOW_INDEX_SHIFT +#define CQ_ENTRY_SHADOW_INDEX_MASK BD_CTL_SHADOW_INDEX_MASK + + +#endif /* QMAN_IF_H */ |