// SPDX-License-Identifier: GPL-2.0 /* * Copyright 2016-2022 HabanaLabs, Ltd. * All Rights Reserved. */ #include "habanalabs.h" #include "../include/common/hl_boot_if.h" #include #include #include #include #define FW_FILE_MAX_SIZE 0x1400000 /* maximum size of 20MB */ static char *extract_fw_ver_from_str(const char *fw_str) { char *str, *fw_ver, *whitespace; fw_ver = kmalloc(16, GFP_KERNEL); if (!fw_ver) return NULL; str = strnstr(fw_str, "fw-", VERSION_MAX_LEN); if (!str) goto free_fw_ver; /* Skip the fw- part */ str += 3; /* Copy until the next whitespace */ whitespace = strnstr(str, " ", 15); if (!whitespace) goto free_fw_ver; strscpy(fw_ver, str, whitespace - str + 1); return fw_ver; free_fw_ver: kfree(fw_ver); return NULL; } static int hl_request_fw(struct hl_device *hdev, const struct firmware **firmware_p, const char *fw_name) { size_t fw_size; int rc; rc = request_firmware(firmware_p, fw_name, hdev->dev); if (rc) { dev_err(hdev->dev, "Firmware file %s is not found! (error %d)\n", fw_name, rc); goto out; } fw_size = (*firmware_p)->size; if ((fw_size % 4) != 0) { dev_err(hdev->dev, "Illegal %s firmware size %zu\n", fw_name, fw_size); rc = -EINVAL; goto release_fw; } dev_dbg(hdev->dev, "%s firmware size == %zu\n", fw_name, fw_size); if (fw_size > FW_FILE_MAX_SIZE) { dev_err(hdev->dev, "FW file size %zu exceeds maximum of %u bytes\n", fw_size, FW_FILE_MAX_SIZE); rc = -EINVAL; goto release_fw; } return 0; release_fw: release_firmware(*firmware_p); out: return rc; } /** * hl_release_firmware() - release FW * * @fw: fw descriptor * * note: this inline function added to serve as a comprehensive mirror for the * hl_request_fw function. */ static inline void hl_release_firmware(const struct firmware *fw) { release_firmware(fw); } /** * hl_fw_copy_fw_to_device() - copy FW to device * * @hdev: pointer to hl_device structure. * @fw: fw descriptor * @dst: IO memory mapped address space to copy firmware to * @src_offset: offset in src FW to copy from * @size: amount of bytes to copy (0 to copy the whole binary) * * actual copy of FW binary data to device, shared by static and dynamic loaders */ static int hl_fw_copy_fw_to_device(struct hl_device *hdev, const struct firmware *fw, void __iomem *dst, u32 src_offset, u32 size) { const void *fw_data; /* size 0 indicates to copy the whole file */ if (!size) size = fw->size; if (src_offset + size > fw->size) { dev_err(hdev->dev, "size to copy(%u) and offset(%u) are invalid\n", size, src_offset); return -EINVAL; } fw_data = (const void *) fw->data; memcpy_toio(dst, fw_data + src_offset, size); return 0; } /** * hl_fw_copy_msg_to_device() - copy message to device * * @hdev: pointer to hl_device structure. * @msg: message * @dst: IO memory mapped address space to copy firmware to * @src_offset: offset in src message to copy from * @size: amount of bytes to copy (0 to copy the whole binary) * * actual copy of message data to device. */ static int hl_fw_copy_msg_to_device(struct hl_device *hdev, struct lkd_msg_comms *msg, void __iomem *dst, u32 src_offset, u32 size) { void *msg_data; /* size 0 indicates to copy the whole file */ if (!size) size = sizeof(struct lkd_msg_comms); if (src_offset + size > sizeof(struct lkd_msg_comms)) { dev_err(hdev->dev, "size to copy(%u) and offset(%u) are invalid\n", size, src_offset); return -EINVAL; } msg_data = (void *) msg; memcpy_toio(dst, msg_data + src_offset, size); return 0; } /** * hl_fw_load_fw_to_device() - Load F/W code to device's memory. * * @hdev: pointer to hl_device structure. * @fw_name: the firmware image name * @dst: IO memory mapped address space to copy firmware to * @src_offset: offset in src FW to copy from * @size: amount of bytes to copy (0 to copy the whole binary) * * Copy fw code from firmware file to device memory. * * Return: 0 on success, non-zero for failure. */ int hl_fw_load_fw_to_device(struct hl_device *hdev, const char *fw_name, void __iomem *dst, u32 src_offset, u32 size) { const struct firmware *fw; int rc; rc = hl_request_fw(hdev, &fw, fw_name); if (rc) return rc; rc = hl_fw_copy_fw_to_device(hdev, fw, dst, src_offset, size); hl_release_firmware(fw); return rc; } int hl_fw_send_pci_access_msg(struct hl_device *hdev, u32 opcode) { struct cpucp_packet pkt = {}; pkt.ctl = cpu_to_le32(opcode << CPUCP_PKT_CTL_OPCODE_SHIFT); return hdev->asic_funcs->send_cpu_message(hdev, (u32 *) &pkt, sizeof(pkt), 0, NULL); } int hl_fw_send_cpu_message(struct hl_device *hdev, u32 hw_queue_id, u32 *msg, u16 len, u32 timeout, u64 *result) { struct hl_hw_queue *queue = &hdev->kernel_queues[hw_queue_id]; struct asic_fixed_properties *prop = &hdev->asic_prop; struct cpucp_packet *pkt; dma_addr_t pkt_dma_addr; struct hl_bd *sent_bd; u32 tmp, expected_ack_val, pi; int rc; pkt = hdev->asic_funcs->cpu_accessible_dma_pool_alloc(hdev, len, &pkt_dma_addr); if (!pkt) { dev_err(hdev->dev, "Failed to allocate DMA memory for packet to CPU\n"); return -ENOMEM; } memcpy(pkt, msg, len); mutex_lock(&hdev->send_cpu_message_lock); /* CPU-CP messages can be sent during soft-reset */ if (hdev->disabled && !hdev->reset_info.is_in_soft_reset) { rc = 0; goto out; } if (hdev->device_cpu_disabled) { rc = -EIO; goto out; } /* set fence to a non valid value */ pkt->fence = cpu_to_le32(UINT_MAX); pi = queue->pi; /* * The CPU queue is a synchronous queue with an effective depth of * a single entry (although it is allocated with room for multiple * entries). We lock on it using 'send_cpu_message_lock' which * serializes accesses to the CPU queue. * Which means that we don't need to lock the access to the entire H/W * queues module when submitting a JOB to the CPU queue. */ hl_hw_queue_submit_bd(hdev, queue, hl_queue_inc_ptr(queue->pi), len, pkt_dma_addr); if (prop->fw_app_cpu_boot_dev_sts0 & CPU_BOOT_DEV_STS0_PKT_PI_ACK_EN) expected_ack_val = queue->pi; else expected_ack_val = CPUCP_PACKET_FENCE_VAL; rc = hl_poll_timeout_memory(hdev, &pkt->fence, tmp, (tmp == expected_ack_val), 1000, timeout, true); hl_hw_queue_inc_ci_kernel(hdev, hw_queue_id); if (rc == -ETIMEDOUT) { dev_err(hdev->dev, "Device CPU packet timeout (0x%x)\n", tmp); hdev->device_cpu_disabled = true; goto out; } tmp = le32_to_cpu(pkt->ctl); rc = (tmp & CPUCP_PKT_CTL_RC_MASK) >> CPUCP_PKT_CTL_RC_SHIFT; if (rc) { dev_err(hdev->dev, "F/W ERROR %d for CPU packet %d\n", rc, (tmp & CPUCP_PKT_CTL_OPCODE_MASK) >> CPUCP_PKT_CTL_OPCODE_SHIFT); rc = -EIO; } else if (result) { *result = le64_to_cpu(pkt->result); } /* Scrub previous buffer descriptor 'ctl' field which contains the * previous PI value written during packet submission. * We must do this or else F/W can read an old value upon queue wraparound. */ sent_bd = queue->kernel_address; sent_bd += hl_pi_2_offset(pi); sent_bd->ctl = cpu_to_le32(UINT_MAX); out: mutex_unlock(&hdev->send_cpu_message_lock); hdev->asic_funcs->cpu_accessible_dma_pool_free(hdev, len, pkt); return rc; } int hl_fw_unmask_irq(struct hl_device *hdev, u16 event_type) { struct cpucp_packet pkt; u64 result; int rc; memset(&pkt, 0, sizeof(pkt)); pkt.ctl = cpu_to_le32(CPUCP_PACKET_UNMASK_RAZWI_IRQ << CPUCP_PKT_CTL_OPCODE_SHIFT); pkt.value = cpu_to_le64(event_type); rc = hdev->asic_funcs->send_cpu_message(hdev, (u32 *) &pkt, sizeof(pkt), 0, &result); if (rc) dev_err(hdev->dev, "failed to unmask RAZWI IRQ %d", event_type); return rc; } int hl_fw_unmask_irq_arr(struct hl_device *hdev, const u32 *irq_arr, size_t irq_arr_size) { struct cpucp_unmask_irq_arr_packet *pkt; size_t total_pkt_size; u64 result; int rc; total_pkt_size = sizeof(struct cpucp_unmask_irq_arr_packet) + irq_arr_size; /* data should be aligned to 8 bytes in order to CPU-CP to copy it */ total_pkt_size = (total_pkt_size + 0x7) & ~0x7; /* total_pkt_size is casted to u16 later on */ if (total_pkt_size > USHRT_MAX) { dev_err(hdev->dev, "too many elements in IRQ array\n"); return -EINVAL; } pkt = kzalloc(total_pkt_size, GFP_KERNEL); if (!pkt) return -ENOMEM; pkt->length = cpu_to_le32(irq_arr_size / sizeof(irq_arr[0])); memcpy(&pkt->irqs, irq_arr, irq_arr_size); pkt->cpucp_pkt.ctl = cpu_to_le32(CPUCP_PACKET_UNMASK_RAZWI_IRQ_ARRAY << CPUCP_PKT_CTL_OPCODE_SHIFT); rc = hdev->asic_funcs->send_cpu_message(hdev, (u32 *) pkt, total_pkt_size, 0, &result); if (rc) dev_err(hdev->dev, "failed to unmask IRQ array\n"); kfree(pkt); return rc; } int hl_fw_test_cpu_queue(struct hl_device *hdev) { struct cpucp_packet test_pkt = {}; u64 result; int rc; test_pkt.ctl = cpu_to_le32(CPUCP_PACKET_TEST << CPUCP_PKT_CTL_OPCODE_SHIFT); test_pkt.value = cpu_to_le64(CPUCP_PACKET_FENCE_VAL); rc = hdev->asic_funcs->send_cpu_message(hdev, (u32 *) &test_pkt, sizeof(test_pkt), 0, &result); if (!rc) { if (result != CPUCP_PACKET_FENCE_VAL) dev_err(hdev->dev, "CPU queue test failed (%#08llx)\n", result); } else { dev_err(hdev->dev, "CPU queue test failed, error %d\n", rc); } return rc; } void *hl_fw_cpu_accessible_dma_pool_alloc(struct hl_device *hdev, size_t size, dma_addr_t *dma_handle) { u64 kernel_addr; kernel_addr = gen_pool_alloc(hdev->cpu_accessible_dma_pool, size); *dma_handle = hdev->cpu_accessible_dma_address + (kernel_addr - (u64) (uintptr_t) hdev->cpu_accessible_dma_mem); return (void *) (uintptr_t) kernel_addr; } void hl_fw_cpu_accessible_dma_pool_free(struct hl_device *hdev, size_t size, void *vaddr) { gen_pool_free(hdev->cpu_accessible_dma_pool, (u64) (uintptr_t) vaddr, size); } int hl_fw_send_heartbeat(struct hl_device *hdev) { struct cpucp_packet hb_pkt; u64 result; int rc; memset(&hb_pkt, 0, sizeof(hb_pkt)); hb_pkt.ctl = cpu_to_le32(CPUCP_PACKET_TEST << CPUCP_PKT_CTL_OPCODE_SHIFT); hb_pkt.value = cpu_to_le64(CPUCP_PACKET_FENCE_VAL); rc = hdev->asic_funcs->send_cpu_message(hdev, (u32 *) &hb_pkt, sizeof(hb_pkt), 0, &result); if ((rc) || (result != CPUCP_PACKET_FENCE_VAL)) return -EIO; if (le32_to_cpu(hb_pkt.status_mask) & CPUCP_PKT_HB_STATUS_EQ_FAULT_MASK) { dev_warn(hdev->dev, "FW reported EQ fault during heartbeat\n"); rc = -EIO; } return rc; } static bool fw_report_boot_dev0(struct hl_device *hdev, u32 err_val, u32 sts_val) { bool err_exists = false; if (!(err_val & CPU_BOOT_ERR0_ENABLED)) return false; if (err_val & CPU_BOOT_ERR0_DRAM_INIT_FAIL) { dev_err(hdev->dev, "Device boot error - DRAM initialization failed\n"); err_exists = true; } if (err_val & CPU_BOOT_ERR0_FIT_CORRUPTED) { dev_err(hdev->dev, "Device boot error - FIT image corrupted\n"); err_exists = true; } if (err_val & CPU_BOOT_ERR0_TS_INIT_FAIL) { dev_err(hdev->dev, "Device boot error - Thermal Sensor initialization failed\n"); err_exists = true; } if (err_val & CPU_BOOT_ERR0_BMC_WAIT_SKIPPED) { if (hdev->bmc_enable) { dev_err(hdev->dev, "Device boot error - Skipped waiting for BMC\n"); err_exists = true; } else { dev_info(hdev->dev, "Device boot message - Skipped waiting for BMC\n"); /* This is an info so we don't want it to disable the * device */ err_val &= ~CPU_BOOT_ERR0_BMC_WAIT_SKIPPED; } } if (err_val & CPU_BOOT_ERR0_NIC_DATA_NOT_RDY) { dev_err(hdev->dev, "Device boot error - Serdes data from BMC not available\n"); err_exists = true; } if (err_val & CPU_BOOT_ERR0_NIC_FW_FAIL) { dev_err(hdev->dev, "Device boot error - NIC F/W initialization failed\n"); err_exists = true; } if (err_val & CPU_BOOT_ERR0_SECURITY_NOT_RDY) { dev_err(hdev->dev, "Device boot warning - security not ready\n"); err_exists = true; } if (err_val & CPU_BOOT_ERR0_SECURITY_FAIL) { dev_err(hdev->dev, "Device boot error - security failure\n"); err_exists = true; } if (err_val & CPU_BOOT_ERR0_EFUSE_FAIL) { dev_err(hdev->dev, "Device boot error - eFuse failure\n"); err_exists = true; } if (err_val & CPU_BOOT_ERR0_SEC_IMG_VER_FAIL) { dev_err(hdev->dev, "Device boot error - Failed to load preboot secondary image\n"); err_exists = true; } if (err_val & CPU_BOOT_ERR0_PLL_FAIL) { dev_err(hdev->dev, "Device boot error - PLL failure\n"); err_exists = true; } if (err_val & CPU_BOOT_ERR0_DEVICE_UNUSABLE_FAIL) { /* Ignore this bit, don't prevent driver loading */ dev_dbg(hdev->dev, "device unusable status is set\n"); err_val &= ~CPU_BOOT_ERR0_DEVICE_UNUSABLE_FAIL; } if (sts_val & CPU_BOOT_DEV_STS0_ENABLED) dev_dbg(hdev->dev, "Device status0 %#x\n", sts_val); /* All warnings should go here in order not to reach the unknown error validation */ if (err_val & CPU_BOOT_ERR0_DRAM_SKIPPED) { dev_warn(hdev->dev, "Device boot warning - Skipped DRAM initialization\n"); /* This is a warning so we don't want it to disable the * device */ err_val &= ~CPU_BOOT_ERR0_DRAM_SKIPPED; } if (err_val & CPU_BOOT_ERR0_PRI_IMG_VER_FAIL) { dev_warn(hdev->dev, "Device boot warning - Failed to load preboot primary image\n"); /* This is a warning so we don't want it to disable the * device as we have a secondary preboot image */ err_val &= ~CPU_BOOT_ERR0_PRI_IMG_VER_FAIL; } if (err_val & CPU_BOOT_ERR0_TPM_FAIL) { dev_warn(hdev->dev, "Device boot warning - TPM failure\n"); /* This is a warning so we don't want it to disable the * device */ err_val &= ~CPU_BOOT_ERR0_TPM_FAIL; } if (!err_exists && (err_val & ~CPU_BOOT_ERR0_ENABLED)) { dev_err(hdev->dev, "Device boot error - unknown ERR0 error 0x%08x\n", err_val); err_exists = true; } /* return error only if it's in the predefined mask */ if (err_exists && ((err_val & ~CPU_BOOT_ERR0_ENABLED) & lower_32_bits(hdev->boot_error_status_mask))) return true; return false; } /* placeholder for ERR1 as no errors defined there yet */ static bool fw_report_boot_dev1(struct hl_device *hdev, u32 err_val, u32 sts_val) { /* * keep this variable to preserve the logic of the function. * this way it would require less modifications when error will be * added to DEV_ERR1 */ bool err_exists = false; if (!(err_val & CPU_BOOT_ERR1_ENABLED)) return false; if (sts_val & CPU_BOOT_DEV_STS1_ENABLED) dev_dbg(hdev->dev, "Device status1 %#x\n", sts_val); if (!err_exists && (err_val & ~CPU_BOOT_ERR1_ENABLED)) { dev_err(hdev->dev, "Device boot error - unknown ERR1 error 0x%08x\n", err_val); err_exists = true; } /* return error only if it's in the predefined mask */ if (err_exists && ((err_val & ~CPU_BOOT_ERR1_ENABLED) & upper_32_bits(hdev->boot_error_status_mask))) return true; return false; } static int fw_read_errors(struct hl_device *hdev, u32 boot_err0_reg, u32 boot_err1_reg, u32 cpu_boot_dev_status0_reg, u32 cpu_boot_dev_status1_reg) { u32 err_val, status_val; bool err_exists = false; /* Some of the firmware status codes are deprecated in newer f/w * versions. In those versions, the errors are reported * in different registers. Therefore, we need to check those * registers and print the exact errors. Moreover, there * may be multiple errors, so we need to report on each error * separately. Some of the error codes might indicate a state * that is not an error per-se, but it is an error in production * environment */ err_val = RREG32(boot_err0_reg); status_val = RREG32(cpu_boot_dev_status0_reg); err_exists = fw_report_boot_dev0(hdev, err_val, status_val); err_val = RREG32(boot_err1_reg); status_val = RREG32(cpu_boot_dev_status1_reg); err_exists |= fw_report_boot_dev1(hdev, err_val, status_val); if (err_exists) return -EIO; return 0; } int hl_fw_cpucp_info_get(struct hl_device *hdev, u32 sts_boot_dev_sts0_reg, u32 sts_boot_dev_sts1_reg, u32 boot_err0_reg, u32 boot_err1_reg) { struct asic_fixed_properties *prop = &hdev->asic_prop; struct cpucp_packet pkt = {}; dma_addr_t cpucp_info_dma_addr; void *cpucp_info_cpu_addr; char *kernel_ver; u64 result; int rc; cpucp_info_cpu_addr = hdev->asic_funcs->cpu_accessible_dma_pool_alloc(hdev, sizeof(struct cpucp_info), &cpucp_info_dma_addr); if (!cpucp_info_cpu_addr) { dev_err(hdev->dev, "Failed to allocate DMA memory for CPU-CP info packet\n"); return -ENOMEM; } memset(cpucp_info_cpu_addr, 0, sizeof(struct cpucp_info)); pkt.ctl = cpu_to_le32(CPUCP_PACKET_INFO_GET << CPUCP_PKT_CTL_OPCODE_SHIFT); pkt.addr = cpu_to_le64(cpucp_info_dma_addr); pkt.data_max_size = cpu_to_le32(sizeof(struct cpucp_info)); rc = hdev->asic_funcs->send_cpu_message(hdev, (u32 *) &pkt, sizeof(pkt), HL_CPUCP_INFO_TIMEOUT_USEC, &result); if (rc) { dev_err(hdev->dev, "Failed to handle CPU-CP info pkt, error %d\n", rc); goto out; } rc = fw_read_errors(hdev, boot_err0_reg, boot_err1_reg, sts_boot_dev_sts0_reg, sts_boot_dev_sts1_reg); if (rc) { dev_err(hdev->dev, "Errors in device boot\n"); goto out; } memcpy(&prop->cpucp_info, cpucp_info_cpu_addr, sizeof(prop->cpucp_info)); rc = hl_build_hwmon_channel_info(hdev, prop->cpucp_info.sensors); if (rc) { dev_err(hdev->dev, "Failed to build hwmon channel info, error %d\n", rc); rc = -EFAULT; goto out; } kernel_ver = extract_fw_ver_from_str(prop->cpucp_info.kernel_version); if (kernel_ver) { dev_info(hdev->dev, "Linux version %s", kernel_ver); kfree(kernel_ver); } /* assume EQ code doesn't need to check eqe index */ hdev->event_queue.check_eqe_index = false; /* Read FW application security bits again */ if (prop->fw_cpu_boot_dev_sts0_valid) { prop->fw_app_cpu_boot_dev_sts0 = RREG32(sts_boot_dev_sts0_reg); if (prop->fw_app_cpu_boot_dev_sts0 & CPU_BOOT_DEV_STS0_EQ_INDEX_EN) hdev->event_queue.check_eqe_index = true; } if (prop->fw_cpu_boot_dev_sts1_valid) prop->fw_app_cpu_boot_dev_sts1 = RREG32(sts_boot_dev_sts1_reg); out: hdev->asic_funcs->cpu_accessible_dma_pool_free(hdev, sizeof(struct cpucp_info), cpucp_info_cpu_addr); return rc; } static int hl_fw_send_msi_info_msg(struct hl_device *hdev) { struct cpucp_array_data_packet *pkt; size_t total_pkt_size, data_size; u64 result; int rc; /* skip sending this info for unsupported ASICs */ if (!hdev->asic_funcs->get_msi_info) return 0; data_size = CPUCP_NUM_OF_MSI_TYPES * sizeof(u32); total_pkt_size = sizeof(struct cpucp_array_data_packet) + data_size; /* data should be aligned to 8 bytes in order to CPU-CP to copy it */ total_pkt_size = (total_pkt_size + 0x7) & ~0x7; /* total_pkt_size is casted to u16 later on */ if (total_pkt_size > USHRT_MAX) { dev_err(hdev->dev, "CPUCP array data is too big\n"); return -EINVAL; } pkt = kzalloc(total_pkt_size, GFP_KERNEL); if (!pkt) return -ENOMEM; pkt->length = cpu_to_le32(CPUCP_NUM_OF_MSI_TYPES); memset((void *) &pkt->data, 0xFF, data_size); hdev->asic_funcs->get_msi_info(pkt->data); pkt->cpucp_pkt.ctl = cpu_to_le32(CPUCP_PACKET_MSI_INFO_SET << CPUCP_PKT_CTL_OPCODE_SHIFT); rc = hdev->asic_funcs->send_cpu_message(hdev, (u32 *)pkt, total_pkt_size, 0, &result); /* * in case packet result is invalid it means that FW does not support * this feature and will use default/hard coded MSI values. no reason * to stop the boot */ if (rc && result == cpucp_packet_invalid) rc = 0; if (rc) dev_err(hdev->dev, "failed to send CPUCP array data\n"); kfree(pkt); return rc; } int hl_fw_cpucp_handshake(struct hl_device *hdev, u32 sts_boot_dev_sts0_reg, u32 sts_boot_dev_sts1_reg, u32 boot_err0_reg, u32 boot_err1_reg) { int rc; rc = hl_fw_cpucp_info_get(hdev, sts_boot_dev_sts0_reg, sts_boot_dev_sts1_reg, boot_err0_reg, boot_err1_reg); if (rc) return rc; return hl_fw_send_msi_info_msg(hdev); } int hl_fw_get_eeprom_data(struct hl_device *hdev, void *data, size_t max_size) { struct cpucp_packet pkt = {}; void *eeprom_info_cpu_addr; dma_addr_t eeprom_info_dma_addr; u64 result; int rc; eeprom_info_cpu_addr = hdev->asic_funcs->cpu_accessible_dma_pool_alloc(hdev, max_size, &eeprom_info_dma_addr); if (!eeprom_info_cpu_addr) { dev_err(hdev->dev, "Failed to allocate DMA memory for CPU-CP EEPROM packet\n"); return -ENOMEM; } memset(eeprom_info_cpu_addr, 0, max_size); pkt.ctl = cpu_to_le32(CPUCP_PACKET_EEPROM_DATA_GET << CPUCP_PKT_CTL_OPCODE_SHIFT); pkt.addr = cpu_to_le64(eeprom_info_dma_addr); pkt.data_max_size = cpu_to_le32(max_size); rc = hdev->asic_funcs->send_cpu_message(hdev, (u32 *) &pkt, sizeof(pkt), HL_CPUCP_EEPROM_TIMEOUT_USEC, &result); if (rc) { dev_err(hdev->dev, "Failed to handle CPU-CP EEPROM packet, error %d\n", rc); goto out; } /* result contains the actual size */ memcpy(data, eeprom_info_cpu_addr, min((size_t)result, max_size)); out: hdev->asic_funcs->cpu_accessible_dma_pool_free(hdev, max_size, eeprom_info_cpu_addr); return rc; } int hl_fw_cpucp_pci_counters_get(struct hl_device *hdev, struct hl_info_pci_counters *counters) { struct cpucp_packet pkt = {}; u64 result; int rc; pkt.ctl = cpu_to_le32(CPUCP_PACKET_PCIE_THROUGHPUT_GET << CPUCP_PKT_CTL_OPCODE_SHIFT); /* Fetch PCI rx counter */ pkt.index = cpu_to_le32(cpucp_pcie_throughput_rx); rc = hdev->asic_funcs->send_cpu_message(hdev, (u32 *) &pkt, sizeof(pkt), HL_CPUCP_INFO_TIMEOUT_USEC, &result); if (rc) { dev_err(hdev->dev, "Failed to handle CPU-CP PCI info pkt, error %d\n", rc); return rc; } counters->rx_throughput = result; memset(&pkt, 0, sizeof(pkt)); pkt.ctl = cpu_to_le32(CPUCP_PACKET_PCIE_THROUGHPUT_GET << CPUCP_PKT_CTL_OPCODE_SHIFT); /* Fetch PCI tx counter */ pkt.index = cpu_to_le32(cpucp_pcie_throughput_tx); rc = hdev->asic_funcs->send_cpu_message(hdev, (u32 *) &pkt, sizeof(pkt), HL_CPUCP_INFO_TIMEOUT_USEC, &result); if (rc) { dev_err(hdev->dev, "Failed to handle CPU-CP PCI info pkt, error %d\n", rc); return rc; } counters->tx_throughput = result; /* Fetch PCI replay counter */ memset(&pkt, 0, sizeof(pkt)); pkt.ctl = cpu_to_le32(CPUCP_PACKET_PCIE_REPLAY_CNT_GET << CPUCP_PKT_CTL_OPCODE_SHIFT); rc = hdev->asic_funcs->send_cpu_message(hdev, (u32 *) &pkt, sizeof(pkt), HL_CPUCP_INFO_TIMEOUT_USEC, &result); if (rc) { dev_err(hdev->dev, "Failed to handle CPU-CP PCI info pkt, error %d\n", rc); return rc; } counters->replay_cnt = (u32) result; return rc; } int hl_fw_cpucp_total_energy_get(struct hl_device *hdev, u64 *total_energy) { struct cpucp_packet pkt = {}; u64 result; int rc; pkt.ctl = cpu_to_le32(CPUCP_PACKET_TOTAL_ENERGY_GET << CPUCP_PKT_CTL_OPCODE_SHIFT); rc = hdev->asic_funcs->send_cpu_message(hdev, (u32 *) &pkt, sizeof(pkt), HL_CPUCP_INFO_TIMEOUT_USEC, &result); if (rc) { dev_err(hdev->dev, "Failed to handle CpuCP total energy pkt, error %d\n", rc); return rc; } *total_energy = result; return rc; } int get_used_pll_index(struct hl_device *hdev, u32 input_pll_index, enum pll_index *pll_index) { struct asic_fixed_properties *prop = &hdev->asic_prop; u8 pll_byte, pll_bit_off; bool dynamic_pll; int fw_pll_idx; dynamic_pll = !!(prop->fw_app_cpu_boot_dev_sts0 & CPU_BOOT_DEV_STS0_DYN_PLL_EN); if (!dynamic_pll) { /* * in case we are working with legacy FW (each asic has unique * PLL numbering) use the driver based index as they are * aligned with fw legacy numbering */ *pll_index = input_pll_index; return 0; } /* retrieve a FW compatible PLL index based on * ASIC specific user request */ fw_pll_idx = hdev->asic_funcs->map_pll_idx_to_fw_idx(input_pll_index); if (fw_pll_idx < 0) { dev_err(hdev->dev, "Invalid PLL index (%u) error %d\n", input_pll_index, fw_pll_idx); return -EINVAL; } /* PLL map is a u8 array */ pll_byte = prop->cpucp_info.pll_map[fw_pll_idx >> 3]; pll_bit_off = fw_pll_idx & 0x7; if (!(pll_byte & BIT(pll_bit_off))) { dev_err(hdev->dev, "PLL index %d is not supported\n", fw_pll_idx); return -EINVAL; } *pll_index = fw_pll_idx; return 0; } int hl_fw_cpucp_pll_info_get(struct hl_device *hdev, u32 pll_index, u16 *pll_freq_arr) { struct cpucp_packet pkt; enum pll_index used_pll_idx; u64 result; int rc; rc = get_used_pll_index(hdev, pll_index, &used_pll_idx); if (rc) return rc; memset(&pkt, 0, sizeof(pkt)); pkt.ctl = cpu_to_le32(CPUCP_PACKET_PLL_INFO_GET << CPUCP_PKT_CTL_OPCODE_SHIFT); pkt.pll_type = __cpu_to_le16((u16)used_pll_idx); rc = hdev->asic_funcs->send_cpu_message(hdev, (u32 *) &pkt, sizeof(pkt), HL_CPUCP_INFO_TIMEOUT_USEC, &result); if (rc) { dev_err(hdev->dev, "Failed to read PLL info, error %d\n", rc); return rc; } pll_freq_arr[0] = FIELD_GET(CPUCP_PKT_RES_PLL_OUT0_MASK, result); pll_freq_arr[1] = FIELD_GET(CPUCP_PKT_RES_PLL_OUT1_MASK, result); pll_freq_arr[2] = FIELD_GET(CPUCP_PKT_RES_PLL_OUT2_MASK, result); pll_freq_arr[3] = FIELD_GET(CPUCP_PKT_RES_PLL_OUT3_MASK, result); return 0; } int hl_fw_cpucp_power_get(struct hl_device *hdev, u64 *power) { struct cpucp_packet pkt; u64 result; int rc; memset(&pkt, 0, sizeof(pkt)); pkt.ctl = cpu_to_le32(CPUCP_PACKET_POWER_GET << CPUCP_PKT_CTL_OPCODE_SHIFT); pkt.type = cpu_to_le16(CPUCP_POWER_INPUT); rc = hdev->asic_funcs->send_cpu_message(hdev, (u32 *) &pkt, sizeof(pkt), HL_CPUCP_INFO_TIMEOUT_USEC, &result); if (rc) { dev_err(hdev->dev, "Failed to read power, error %d\n", rc); return rc; } *power = result; return rc; } int hl_fw_dram_replaced_row_get(struct hl_device *hdev, struct cpucp_hbm_row_info *info) { struct cpucp_hbm_row_info *cpucp_repl_rows_info_cpu_addr; dma_addr_t cpucp_repl_rows_info_dma_addr; struct cpucp_packet pkt = {}; u64 result; int rc; cpucp_repl_rows_info_cpu_addr = hdev->asic_funcs->cpu_accessible_dma_pool_alloc(hdev, sizeof(struct cpucp_hbm_row_info), &cpucp_repl_rows_info_dma_addr); if (!cpucp_repl_rows_info_cpu_addr) { dev_err(hdev->dev, "Failed to allocate DMA memory for CPU-CP replaced rows info packet\n"); return -ENOMEM; } memset(cpucp_repl_rows_info_cpu_addr, 0, sizeof(struct cpucp_hbm_row_info)); pkt.ctl = cpu_to_le32(CPUCP_PACKET_HBM_REPLACED_ROWS_INFO_GET << CPUCP_PKT_CTL_OPCODE_SHIFT); pkt.addr = cpu_to_le64(cpucp_repl_rows_info_dma_addr); pkt.data_max_size = cpu_to_le32(sizeof(struct cpucp_hbm_row_info)); rc = hdev->asic_funcs->send_cpu_message(hdev, (u32 *) &pkt, sizeof(pkt), HL_CPUCP_INFO_TIMEOUT_USEC, &result); if (rc) { dev_err(hdev->dev, "Failed to handle CPU-CP replaced rows info pkt, error %d\n", rc); goto out; } memcpy(info, cpucp_repl_rows_info_cpu_addr, sizeof(*info)); out: hdev->asic_funcs->cpu_accessible_dma_pool_free(hdev, sizeof(struct cpucp_hbm_row_info), cpucp_repl_rows_info_cpu_addr); return rc; } int hl_fw_dram_pending_row_get(struct hl_device *hdev, u32 *pend_rows_num) { struct cpucp_packet pkt; u64 result; int rc; memset(&pkt, 0, sizeof(pkt)); pkt.ctl = cpu_to_le32(CPUCP_PACKET_HBM_PENDING_ROWS_STATUS << CPUCP_PKT_CTL_OPCODE_SHIFT); rc = hdev->asic_funcs->send_cpu_message(hdev, (u32 *) &pkt, sizeof(pkt), 0, &result); if (rc) { dev_err(hdev->dev, "Failed to handle CPU-CP pending rows info pkt, error %d\n", rc); goto out; } *pend_rows_num = (u32) result; out: return rc; } int hl_fw_cpucp_engine_core_asid_set(struct hl_device *hdev, u32 asid) { struct cpucp_packet pkt; int rc; memset(&pkt, 0, sizeof(pkt)); pkt.ctl = cpu_to_le32(CPUCP_PACKET_ENGINE_CORE_ASID_SET << CPUCP_PKT_CTL_OPCODE_SHIFT); pkt.value = cpu_to_le64(asid); rc = hdev->asic_funcs->send_cpu_message(hdev, (u32 *) &pkt, sizeof(pkt), HL_CPUCP_INFO_TIMEOUT_USEC, NULL); if (rc) dev_err(hdev->dev, "Failed on ASID configuration request for engine core, error %d\n", rc); return rc; } void hl_fw_ask_hard_reset_without_linux(struct hl_device *hdev) { struct static_fw_load_mgr *static_loader = &hdev->fw_loader.static_loader; int rc; if (hdev->asic_prop.dynamic_fw_load) { rc = hl_fw_dynamic_send_protocol_cmd(hdev, &hdev->fw_loader, COMMS_RST_DEV, 0, false, hdev->fw_loader.cpu_timeout); if (rc) dev_warn(hdev->dev, "Failed sending COMMS_RST_DEV\n"); } else { WREG32(static_loader->kmd_msg_to_cpu_reg, KMD_MSG_RST_DEV); } } void hl_fw_ask_halt_machine_without_linux(struct hl_device *hdev) { struct static_fw_load_mgr *static_loader = &hdev->fw_loader.static_loader; int rc; if (hdev->device_cpu_is_halted) return; /* Stop device CPU to make sure nothing bad happens */ if (hdev->asic_prop.dynamic_fw_load) { rc = hl_fw_dynamic_send_protocol_cmd(hdev, &hdev->fw_loader, COMMS_GOTO_WFE, 0, true, hdev->fw_loader.cpu_timeout); if (rc) dev_warn(hdev->dev, "Failed sending COMMS_GOTO_WFE\n"); } else { WREG32(static_loader->kmd_msg_to_cpu_reg, KMD_MSG_GOTO_WFE); msleep(static_loader->cpu_reset_wait_msec); /* Must clear this register in order to prevent preboot * from reading WFE after reboot */ WREG32(static_loader->kmd_msg_to_cpu_reg, KMD_MSG_NA); } hdev->device_cpu_is_halted = true; } static void detect_cpu_boot_status(struct hl_device *hdev, u32 status) { /* Some of the status codes below are deprecated in newer f/w * versions but we keep them here for backward compatibility */ switch (status) { case CPU_BOOT_STATUS_NA: dev_err(hdev->dev, "Device boot progress - BTL/ROM did NOT run\n"); break; case CPU_BOOT_STATUS_IN_WFE: dev_err(hdev->dev, "Device boot progress - Stuck inside WFE loop\n"); break; case CPU_BOOT_STATUS_IN_BTL: dev_err(hdev->dev, "Device boot progress - Stuck in BTL\n"); break; case CPU_BOOT_STATUS_IN_PREBOOT: dev_err(hdev->dev, "Device boot progress - Stuck in Preboot\n"); break; case CPU_BOOT_STATUS_IN_SPL: dev_err(hdev->dev, "Device boot progress - Stuck in SPL\n"); break; case CPU_BOOT_STATUS_IN_UBOOT: dev_err(hdev->dev, "Device boot progress - Stuck in u-boot\n"); break; case CPU_BOOT_STATUS_DRAM_INIT_FAIL: dev_err(hdev->dev, "Device boot progress - DRAM initialization failed\n"); break; case CPU_BOOT_STATUS_UBOOT_NOT_READY: dev_err(hdev->dev, "Device boot progress - Cannot boot\n"); break; case CPU_BOOT_STATUS_TS_INIT_FAIL: dev_err(hdev->dev, "Device boot progress - Thermal Sensor initialization failed\n"); break; case CPU_BOOT_STATUS_SECURITY_READY: dev_err(hdev->dev, "Device boot progress - Stuck in preboot after security initialization\n"); break; default: dev_err(hdev->dev, "Device boot progress - Invalid status code %d\n", status); break; } } static int hl_fw_read_preboot_caps(struct hl_device *hdev, u32 cpu_boot_status_reg, u32 sts_boot_dev_sts0_reg, u32 sts_boot_dev_sts1_reg, u32 boot_err0_reg, u32 boot_err1_reg, u32 timeout) { struct asic_fixed_properties *prop = &hdev->asic_prop; u32 status, reg_val; int rc; /* Need to check two possible scenarios: * * CPU_BOOT_STATUS_WAITING_FOR_BOOT_FIT - for newer firmwares where * the preboot is waiting for the boot fit * * All other status values - for older firmwares where the uboot was * loaded from the FLASH */ rc = hl_poll_timeout( hdev, cpu_boot_status_reg, status, (status == CPU_BOOT_STATUS_NIC_FW_RDY) || (status == CPU_BOOT_STATUS_READY_TO_BOOT) || (status == CPU_BOOT_STATUS_WAITING_FOR_BOOT_FIT), hdev->fw_poll_interval_usec, timeout); if (rc) { dev_err(hdev->dev, "CPU boot ready status timeout\n"); detect_cpu_boot_status(hdev, status); /* If we read all FF, then something is totally wrong, no point * of reading specific errors */ if (status != -1) fw_read_errors(hdev, boot_err0_reg, boot_err1_reg, sts_boot_dev_sts0_reg, sts_boot_dev_sts1_reg); return -EIO; } /* * the registers DEV_STS* contain FW capabilities/features. * We can rely on this registers only if bit CPU_BOOT_DEV_STS*_ENABLED * is set. * In the first read of this register we store the value of this * register ONLY if the register is enabled (which will be propagated * to next stages) and also mark the register as valid. * In case it is not enabled the stored value will be left 0- all * caps/features are off */ reg_val = RREG32(sts_boot_dev_sts0_reg); if (reg_val & CPU_BOOT_DEV_STS0_ENABLED) { prop->fw_cpu_boot_dev_sts0_valid = true; prop->fw_preboot_cpu_boot_dev_sts0 = reg_val; } reg_val = RREG32(sts_boot_dev_sts1_reg); if (reg_val & CPU_BOOT_DEV_STS1_ENABLED) { prop->fw_cpu_boot_dev_sts1_valid = true; prop->fw_preboot_cpu_boot_dev_sts1 = reg_val; } prop->dynamic_fw_load = !!(prop->fw_preboot_cpu_boot_dev_sts0 & CPU_BOOT_DEV_STS0_FW_LD_COM_EN); /* initialize FW loader once we know what load protocol is used */ hdev->asic_funcs->init_firmware_loader(hdev); dev_dbg(hdev->dev, "Attempting %s FW load\n", prop->dynamic_fw_load ? "dynamic" : "legacy"); return 0; } static int hl_fw_static_read_device_fw_version(struct hl_device *hdev, enum hl_fw_component fwc) { struct asic_fixed_properties *prop = &hdev->asic_prop; struct fw_load_mgr *fw_loader = &hdev->fw_loader; struct static_fw_load_mgr *static_loader; char *dest, *boot_ver, *preboot_ver; u32 ver_off, limit; const char *name; char btl_ver[32]; static_loader = &hdev->fw_loader.static_loader; switch (fwc) { case FW_COMP_BOOT_FIT: ver_off = RREG32(static_loader->boot_fit_version_offset_reg); dest = prop->uboot_ver; name = "Boot-fit"; limit = static_loader->boot_fit_version_max_off; break; case FW_COMP_PREBOOT: ver_off = RREG32(static_loader->preboot_version_offset_reg); dest = prop->preboot_ver; name = "Preboot"; limit = static_loader->preboot_version_max_off; break; default: dev_warn(hdev->dev, "Undefined FW component: %d\n", fwc); return -EIO; } ver_off &= static_loader->sram_offset_mask; if (ver_off < limit) { memcpy_fromio(dest, hdev->pcie_bar[fw_loader->sram_bar_id] + ver_off, VERSION_MAX_LEN); } else { dev_err(hdev->dev, "%s version offset (0x%x) is above SRAM\n", name, ver_off); strscpy(dest, "unavailable", VERSION_MAX_LEN); return -EIO; } if (fwc == FW_COMP_BOOT_FIT) { boot_ver = extract_fw_ver_from_str(prop->uboot_ver); if (boot_ver) { dev_info(hdev->dev, "boot-fit version %s\n", boot_ver); kfree(boot_ver); } } else if (fwc == FW_COMP_PREBOOT) { preboot_ver = strnstr(prop->preboot_ver, "Preboot", VERSION_MAX_LEN); if (preboot_ver && preboot_ver != prop->preboot_ver) { strscpy(btl_ver, prop->preboot_ver, min((int) (preboot_ver - prop->preboot_ver), 31)); dev_info(hdev->dev, "%s\n", btl_ver); } preboot_ver = extract_fw_ver_from_str(prop->preboot_ver); if (preboot_ver) { dev_info(hdev->dev, "preboot version %s\n", preboot_ver); kfree(preboot_ver); } } return 0; } /** * hl_fw_preboot_update_state - update internal data structures during * handshake with preboot * * * @hdev: pointer to the habanalabs device structure * * @return 0 on success, otherwise non-zero error code */ static void hl_fw_preboot_update_state(struct hl_device *hdev) { struct asic_fixed_properties *prop = &hdev->asic_prop; u32 cpu_boot_dev_sts0, cpu_boot_dev_sts1; cpu_boot_dev_sts0 = prop->fw_preboot_cpu_boot_dev_sts0; cpu_boot_dev_sts1 = prop->fw_preboot_cpu_boot_dev_sts1; /* We read boot_dev_sts registers multiple times during boot: * 1. preboot - a. Check whether the security status bits are valid * b. Check whether fw security is enabled * c. Check whether hard reset is done by preboot * 2. boot cpu - a. Fetch boot cpu security status * b. Check whether hard reset is done by boot cpu * 3. FW application - a. Fetch fw application security status * b. Check whether hard reset is done by fw app */ prop->hard_reset_done_by_fw = !!(cpu_boot_dev_sts0 & CPU_BOOT_DEV_STS0_FW_HARD_RST_EN); dev_dbg(hdev->dev, "Firmware preboot boot device status0 %#x\n", cpu_boot_dev_sts0); dev_dbg(hdev->dev, "Firmware preboot boot device status1 %#x\n", cpu_boot_dev_sts1); dev_dbg(hdev->dev, "Firmware preboot hard-reset is %s\n", prop->hard_reset_done_by_fw ? "enabled" : "disabled"); dev_dbg(hdev->dev, "firmware-level security is %s\n", prop->fw_security_enabled ? "enabled" : "disabled"); dev_dbg(hdev->dev, "GIC controller is %s\n", prop->gic_interrupts_enable ? "enabled" : "disabled"); } static int hl_fw_static_read_preboot_status(struct hl_device *hdev) { int rc; rc = hl_fw_static_read_device_fw_version(hdev, FW_COMP_PREBOOT); if (rc) return rc; return 0; } int hl_fw_read_preboot_status(struct hl_device *hdev, u32 cpu_boot_status_reg, u32 sts_boot_dev_sts0_reg, u32 sts_boot_dev_sts1_reg, u32 boot_err0_reg, u32 boot_err1_reg, u32 timeout) { int rc; if (!(hdev->fw_components & FW_TYPE_PREBOOT_CPU)) return 0; /* * In order to determine boot method (static VS dymanic) we need to * read the boot caps register */ rc = hl_fw_read_preboot_caps(hdev, cpu_boot_status_reg, sts_boot_dev_sts0_reg, sts_boot_dev_sts1_reg, boot_err0_reg, boot_err1_reg, timeout); if (rc) return rc; hl_fw_preboot_update_state(hdev); /* no need to read preboot status in dynamic load */ if (hdev->asic_prop.dynamic_fw_load) return 0; return hl_fw_static_read_preboot_status(hdev); } /* associate string with COMM status */ static char *hl_dynamic_fw_status_str[COMMS_STS_INVLD_LAST] = { [COMMS_STS_NOOP] = "NOOP", [COMMS_STS_ACK] = "ACK", [COMMS_STS_OK] = "OK", [COMMS_STS_ERR] = "ERR", [COMMS_STS_VALID_ERR] = "VALID_ERR", [COMMS_STS_TIMEOUT_ERR] = "TIMEOUT_ERR", }; /** * hl_fw_dynamic_report_error_status - report error status * * @hdev: pointer to the habanalabs device structure * @status: value of FW status register * @expected_status: the expected status */ static void hl_fw_dynamic_report_error_status(struct hl_device *hdev, u32 status, enum comms_sts expected_status) { enum comms_sts comm_status = FIELD_GET(COMMS_STATUS_STATUS_MASK, status); if (comm_status < COMMS_STS_INVLD_LAST) dev_err(hdev->dev, "Device status %s, expected status: %s\n", hl_dynamic_fw_status_str[comm_status], hl_dynamic_fw_status_str[expected_status]); else dev_err(hdev->dev, "Device status unknown %d, expected status: %s\n", comm_status, hl_dynamic_fw_status_str[expected_status]); } /** * hl_fw_dynamic_send_cmd - send LKD to FW cmd * * @hdev: pointer to the habanalabs device structure * @fw_loader: managing structure for loading device's FW * @cmd: LKD to FW cmd code * @size: size of next FW component to be loaded (0 if not necessary) * * LDK to FW exact command layout is defined at struct comms_command. * note: the size argument is used only when the next FW component should be * loaded, otherwise it shall be 0. the size is used by the FW in later * protocol stages and when sending only indicating the amount of memory * to be allocated by the FW to receive the next boot component. */ static void hl_fw_dynamic_send_cmd(struct hl_device *hdev, struct fw_load_mgr *fw_loader, enum comms_cmd cmd, unsigned int size) { struct cpu_dyn_regs *dyn_regs; u32 val; dyn_regs = &fw_loader->dynamic_loader.comm_desc.cpu_dyn_regs; val = FIELD_PREP(COMMS_COMMAND_CMD_MASK, cmd); val |= FIELD_PREP(COMMS_COMMAND_SIZE_MASK, size); WREG32(le32_to_cpu(dyn_regs->kmd_msg_to_cpu), val); } /** * hl_fw_dynamic_extract_fw_response - update the FW response * * @hdev: pointer to the habanalabs device structure * @fw_loader: managing structure for loading device's FW * @response: FW response * @status: the status read from CPU status register * * @return 0 on success, otherwise non-zero error code */ static int hl_fw_dynamic_extract_fw_response(struct hl_device *hdev, struct fw_load_mgr *fw_loader, struct fw_response *response, u32 status) { response->status = FIELD_GET(COMMS_STATUS_STATUS_MASK, status); response->ram_offset = FIELD_GET(COMMS_STATUS_OFFSET_MASK, status) << COMMS_STATUS_OFFSET_ALIGN_SHIFT; response->ram_type = FIELD_GET(COMMS_STATUS_RAM_TYPE_MASK, status); if ((response->ram_type != COMMS_SRAM) && (response->ram_type != COMMS_DRAM)) { dev_err(hdev->dev, "FW status: invalid RAM type %u\n", response->ram_type); return -EIO; } return 0; } /** * hl_fw_dynamic_wait_for_status - wait for status in dynamic FW load * * @hdev: pointer to the habanalabs device structure * @fw_loader: managing structure for loading device's FW * @expected_status: expected status to wait for * @timeout: timeout for status wait * * @return 0 on success, otherwise non-zero error code * * waiting for status from FW include polling the FW status register until * expected status is received or timeout occurs (whatever occurs first). */ static int hl_fw_dynamic_wait_for_status(struct hl_device *hdev, struct fw_load_mgr *fw_loader, enum comms_sts expected_status, u32 timeout) { struct cpu_dyn_regs *dyn_regs; u32 status; int rc; dyn_regs = &fw_loader->dynamic_loader.comm_desc.cpu_dyn_regs; /* Wait for expected status */ rc = hl_poll_timeout( hdev, le32_to_cpu(dyn_regs->cpu_cmd_status_to_host), status, FIELD_GET(COMMS_STATUS_STATUS_MASK, status) == expected_status, hdev->fw_poll_interval_usec, timeout); if (rc) { hl_fw_dynamic_report_error_status(hdev, status, expected_status); return -EIO; } /* * skip storing FW response for NOOP to preserve the actual desired * FW status */ if (expected_status == COMMS_STS_NOOP) return 0; rc = hl_fw_dynamic_extract_fw_response(hdev, fw_loader, &fw_loader->dynamic_loader.response, status); return rc; } /** * hl_fw_dynamic_send_clear_cmd - send clear command to FW * * @hdev: pointer to the habanalabs device structure * @fw_loader: managing structure for loading device's FW * * @return 0 on success, otherwise non-zero error code * * after command cycle between LKD to FW CPU (i.e. LKD got an expected status * from FW) we need to clear the CPU status register in order to avoid garbage * between command cycles. * This is done by sending clear command and polling the CPU to LKD status * register to hold the status NOOP */ static int hl_fw_dynamic_send_clear_cmd(struct hl_device *hdev, struct fw_load_mgr *fw_loader) { hl_fw_dynamic_send_cmd(hdev, fw_loader, COMMS_CLR_STS, 0); return hl_fw_dynamic_wait_for_status(hdev, fw_loader, COMMS_STS_NOOP, fw_loader->cpu_timeout); } /** * hl_fw_dynamic_send_protocol_cmd - send LKD to FW cmd and wait for ACK * * @hdev: pointer to the habanalabs device structure * @fw_loader: managing structure for loading device's FW * @cmd: LKD to FW cmd code * @size: size of next FW component to be loaded (0 if not necessary) * @wait_ok: if true also wait for OK response from FW * @timeout: timeout for status wait * * @return 0 on success, otherwise non-zero error code * * brief: * when sending protocol command we have the following steps: * - send clear (clear command and verify clear status register) * - send the actual protocol command * - wait for ACK on the protocol command * - send clear * - send NOOP * if, in addition, the specific protocol command should wait for OK then: * - wait for OK * - send clear * - send NOOP * * NOTES: * send clear: this is necessary in order to clear the status register to avoid * leftovers between command * NOOP command: necessary to avoid loop on the clear command by the FW */ int hl_fw_dynamic_send_protocol_cmd(struct hl_device *hdev, struct fw_load_mgr *fw_loader, enum comms_cmd cmd, unsigned int size, bool wait_ok, u32 timeout) { int rc; /* first send clear command to clean former commands */ rc = hl_fw_dynamic_send_clear_cmd(hdev, fw_loader); /* send the actual command */ hl_fw_dynamic_send_cmd(hdev, fw_loader, cmd, size); /* wait for ACK for the command */ rc = hl_fw_dynamic_wait_for_status(hdev, fw_loader, COMMS_STS_ACK, timeout); if (rc) return rc; /* clear command to prepare for NOOP command */ rc = hl_fw_dynamic_send_clear_cmd(hdev, fw_loader); if (rc) return rc; /* send the actual NOOP command */ hl_fw_dynamic_send_cmd(hdev, fw_loader, COMMS_NOOP, 0); if (!wait_ok) return 0; rc = hl_fw_dynamic_wait_for_status(hdev, fw_loader, COMMS_STS_OK, timeout); if (rc) return rc; /* clear command to prepare for NOOP command */ rc = hl_fw_dynamic_send_clear_cmd(hdev, fw_loader); if (rc) return rc; /* send the actual NOOP command */ hl_fw_dynamic_send_cmd(hdev, fw_loader, COMMS_NOOP, 0); return 0; } /** * hl_fw_compat_crc32 - CRC compatible with FW * * @data: pointer to the data * @size: size of the data * * @return the CRC32 result * * NOTE: kernel's CRC32 differ's from standard CRC32 calculation. * in order to be aligned we need to flip the bits of both the input * initial CRC and kernel's CRC32 result. * in addition both sides use initial CRC of 0, */ static u32 hl_fw_compat_crc32(u8 *data, size_t size) { return ~crc32_le(~((u32)0), data, size); } /** * hl_fw_dynamic_validate_memory_bound - validate memory bounds for memory * transfer (image or descriptor) between * host and FW * * @hdev: pointer to the habanalabs device structure * @addr: device address of memory transfer * @size: memory transter size * @region: PCI memory region * * @return 0 on success, otherwise non-zero error code */ static int hl_fw_dynamic_validate_memory_bound(struct hl_device *hdev, u64 addr, size_t size, struct pci_mem_region *region) { u64 end_addr; /* now make sure that the memory transfer is within region's bounds */ end_addr = addr + size; if (end_addr >= region->region_base + region->region_size) { dev_err(hdev->dev, "dynamic FW load: memory transfer end address out of memory region bounds. addr: %llx\n", end_addr); return -EIO; } /* * now make sure memory transfer is within predefined BAR bounds. * this is to make sure we do not need to set the bar (e.g. for DRAM * memory transfers) */ if (end_addr >= region->region_base - region->offset_in_bar + region->bar_size) { dev_err(hdev->dev, "FW image beyond PCI BAR bounds\n"); return -EIO; } return 0; } /** * hl_fw_dynamic_validate_descriptor - validate FW descriptor * * @hdev: pointer to the habanalabs device structure * @fw_loader: managing structure for loading device's FW * @fw_desc: the descriptor form FW * * @return 0 on success, otherwise non-zero error code */ static int hl_fw_dynamic_validate_descriptor(struct hl_device *hdev, struct fw_load_mgr *fw_loader, struct lkd_fw_comms_desc *fw_desc) { struct pci_mem_region *region; enum pci_region region_id; size_t data_size; u32 data_crc32; u8 *data_ptr; u64 addr; int rc; if (le32_to_cpu(fw_desc->header.magic) != HL_COMMS_DESC_MAGIC) { dev_err(hdev->dev, "Invalid magic for dynamic FW descriptor (%x)\n", fw_desc->header.magic); return -EIO; } if (fw_desc->header.version != HL_COMMS_DESC_VER) { dev_err(hdev->dev, "Invalid version for dynamic FW descriptor (%x)\n", fw_desc->header.version); return -EIO; } /* * calc CRC32 of data without header. * note that no alignment/stride address issues here as all structures * are 64 bit padded */ data_size = sizeof(struct lkd_fw_comms_desc) - sizeof(struct comms_desc_header); data_ptr = (u8 *)fw_desc + sizeof(struct comms_desc_header); if (le16_to_cpu(fw_desc->header.size) != data_size) { dev_err(hdev->dev, "Invalid descriptor size 0x%x, expected size 0x%zx\n", le16_to_cpu(fw_desc->header.size), data_size); return -EIO; } data_crc32 = hl_fw_compat_crc32(data_ptr, data_size); if (data_crc32 != le32_to_cpu(fw_desc->header.crc32)) { dev_err(hdev->dev, "CRC32 mismatch for dynamic FW descriptor (%x:%x)\n", data_crc32, fw_desc->header.crc32); return -EIO; } /* find memory region to which to copy the image */ addr = le64_to_cpu(fw_desc->img_addr); region_id = hl_get_pci_memory_region(hdev, addr); if ((region_id != PCI_REGION_SRAM) && ((region_id != PCI_REGION_DRAM))) { dev_err(hdev->dev, "Invalid region to copy FW image address=%llx\n", addr); return -EIO; } region = &hdev->pci_mem_region[region_id]; /* store the region for the copy stage */ fw_loader->dynamic_loader.image_region = region; /* * here we know that the start address is valid, now make sure that the * image is within region's bounds */ rc = hl_fw_dynamic_validate_memory_bound(hdev, addr, fw_loader->dynamic_loader.fw_image_size, region); if (rc) { dev_err(hdev->dev, "invalid mem transfer request for FW image\n"); return rc; } /* here we can mark the descriptor as valid as the content has been validated */ fw_loader->dynamic_loader.fw_desc_valid = true; return 0; } static int hl_fw_dynamic_validate_response(struct hl_device *hdev, struct fw_response *response, struct pci_mem_region *region) { u64 device_addr; int rc; device_addr = region->region_base + response->ram_offset; /* * validate that the descriptor is within region's bounds * Note that as the start address was supplied according to the RAM * type- testing only the end address is enough */ rc = hl_fw_dynamic_validate_memory_bound(hdev, device_addr, sizeof(struct lkd_fw_comms_desc), region); return rc; } /** * hl_fw_dynamic_read_and_validate_descriptor - read and validate FW descriptor * * @hdev: pointer to the habanalabs device structure * @fw_loader: managing structure for loading device's FW * * @return 0 on success, otherwise non-zero error code */ static int hl_fw_dynamic_read_and_validate_descriptor(struct hl_device *hdev, struct fw_load_mgr *fw_loader) { struct lkd_fw_comms_desc *fw_desc; struct pci_mem_region *region; struct fw_response *response; enum pci_region region_id; void __iomem *src; int rc; fw_desc = &fw_loader->dynamic_loader.comm_desc; response = &fw_loader->dynamic_loader.response; region_id = (response->ram_type == COMMS_SRAM) ? PCI_REGION_SRAM : PCI_REGION_DRAM; region = &hdev->pci_mem_region[region_id]; rc = hl_fw_dynamic_validate_response(hdev, response, region); if (rc) { dev_err(hdev->dev, "invalid mem transfer request for FW descriptor\n"); return rc; } /* * extract address to copy the descriptor from * in addition, as the descriptor value is going to be over-ridden by new data- we mark it * as invalid. * it will be marked again as valid once validated */ fw_loader->dynamic_loader.fw_desc_valid = false; src = hdev->pcie_bar[region->bar_id] + region->offset_in_bar + response->ram_offset; memcpy_fromio(fw_desc, src, sizeof(struct lkd_fw_comms_desc)); return hl_fw_dynamic_validate_descriptor(hdev, fw_loader, fw_desc); } /** * hl_fw_dynamic_request_descriptor - handshake with CPU to get FW descriptor * * @hdev: pointer to the habanalabs device structure * @fw_loader: managing structure for loading device's FW * @next_image_size: size to allocate for next FW component * * @return 0 on success, otherwise non-zero error code */ static int hl_fw_dynamic_request_descriptor(struct hl_device *hdev, struct fw_load_mgr *fw_loader, size_t next_image_size) { int rc; rc = hl_fw_dynamic_send_protocol_cmd(hdev, fw_loader, COMMS_PREP_DESC, next_image_size, true, fw_loader->cpu_timeout); if (rc) return rc; return hl_fw_dynamic_read_and_validate_descriptor(hdev, fw_loader); } /** * hl_fw_dynamic_read_device_fw_version - read FW version to exposed properties * * @hdev: pointer to the habanalabs device structure * @fwc: the firmware component * @fw_version: fw component's version string */ static void hl_fw_dynamic_read_device_fw_version(struct hl_device *hdev, enum hl_fw_component fwc, const char *fw_version) { struct asic_fixed_properties *prop = &hdev->asic_prop; char *preboot_ver, *boot_ver; char btl_ver[32]; switch (fwc) { case FW_COMP_BOOT_FIT: strscpy(prop->uboot_ver, fw_version, VERSION_MAX_LEN); boot_ver = extract_fw_ver_from_str(prop->uboot_ver); if (boot_ver) { dev_info(hdev->dev, "boot-fit version %s\n", boot_ver); kfree(boot_ver); } break; case FW_COMP_PREBOOT: strscpy(prop->preboot_ver, fw_version, VERSION_MAX_LEN); preboot_ver = strnstr(prop->preboot_ver, "Preboot", VERSION_MAX_LEN); if (preboot_ver && preboot_ver != prop->preboot_ver) { strscpy(btl_ver, prop->preboot_ver, min((int) (preboot_ver - prop->preboot_ver), 31)); dev_info(hdev->dev, "%s\n", btl_ver); } preboot_ver = extract_fw_ver_from_str(prop->preboot_ver); if (preboot_ver) { dev_info(hdev->dev, "preboot version %s\n", preboot_ver); kfree(preboot_ver); } break; default: dev_warn(hdev->dev, "Undefined FW component: %d\n", fwc); return; } } /** * hl_fw_dynamic_copy_image - copy image to memory allocated by the FW * * @hdev: pointer to the habanalabs device structure * @fw: fw descriptor * @fw_loader: managing structure for loading device's FW */ static int hl_fw_dynamic_copy_image(struct hl_device *hdev, const struct firmware *fw, struct fw_load_mgr *fw_loader) { struct lkd_fw_comms_desc *fw_desc; struct pci_mem_region *region; void __iomem *dest; u64 addr; int rc; fw_desc = &fw_loader->dynamic_loader.comm_desc; addr = le64_to_cpu(fw_desc->img_addr); /* find memory region to which to copy the image */ region = fw_loader->dynamic_loader.image_region; dest = hdev->pcie_bar[region->bar_id] + region->offset_in_bar + (addr - region->region_base); rc = hl_fw_copy_fw_to_device(hdev, fw, dest, fw_loader->boot_fit_img.src_off, fw_loader->boot_fit_img.copy_size); return rc; } /** * hl_fw_dynamic_copy_msg - copy msg to memory allocated by the FW * * @hdev: pointer to the habanalabs device structure * @msg: message * @fw_loader: managing structure for loading device's FW */ static int hl_fw_dynamic_copy_msg(struct hl_device *hdev, struct lkd_msg_comms *msg, struct fw_load_mgr *fw_loader) { struct lkd_fw_comms_desc *fw_desc; struct pci_mem_region *region; void __iomem *dest; u64 addr; int rc; fw_desc = &fw_loader->dynamic_loader.comm_desc; addr = le64_to_cpu(fw_desc->img_addr); /* find memory region to which to copy the image */ region = fw_loader->dynamic_loader.image_region; dest = hdev->pcie_bar[region->bar_id] + region->offset_in_bar + (addr - region->region_base); rc = hl_fw_copy_msg_to_device(hdev, msg, dest, 0, 0); return rc; } /** * hl_fw_boot_fit_update_state - update internal data structures after boot-fit * is loaded * * @hdev: pointer to the habanalabs device structure * @cpu_boot_dev_sts0_reg: register holding CPU boot dev status 0 * @cpu_boot_dev_sts1_reg: register holding CPU boot dev status 1 * * @return 0 on success, otherwise non-zero error code */ static void hl_fw_boot_fit_update_state(struct hl_device *hdev, u32 cpu_boot_dev_sts0_reg, u32 cpu_boot_dev_sts1_reg) { struct asic_fixed_properties *prop = &hdev->asic_prop; hdev->fw_loader.fw_comp_loaded |= FW_TYPE_BOOT_CPU; /* Read boot_cpu status bits */ if (prop->fw_preboot_cpu_boot_dev_sts0 & CPU_BOOT_DEV_STS0_ENABLED) { prop->fw_bootfit_cpu_boot_dev_sts0 = RREG32(cpu_boot_dev_sts0_reg); prop->hard_reset_done_by_fw = !!(prop->fw_bootfit_cpu_boot_dev_sts0 & CPU_BOOT_DEV_STS0_FW_HARD_RST_EN); dev_dbg(hdev->dev, "Firmware boot CPU status0 %#x\n", prop->fw_bootfit_cpu_boot_dev_sts0); } if (prop->fw_cpu_boot_dev_sts1_valid) { prop->fw_bootfit_cpu_boot_dev_sts1 = RREG32(cpu_boot_dev_sts1_reg); dev_dbg(hdev->dev, "Firmware boot CPU status1 %#x\n", prop->fw_bootfit_cpu_boot_dev_sts1); } dev_dbg(hdev->dev, "Firmware boot CPU hard-reset is %s\n", prop->hard_reset_done_by_fw ? "enabled" : "disabled"); } static void hl_fw_dynamic_update_linux_interrupt_if(struct hl_device *hdev) { struct cpu_dyn_regs *dyn_regs = &hdev->fw_loader.dynamic_loader.comm_desc.cpu_dyn_regs; /* Check whether all 3 interrupt interfaces are set, if not use a * single interface */ if (!hdev->asic_prop.gic_interrupts_enable && !(hdev->asic_prop.fw_app_cpu_boot_dev_sts0 & CPU_BOOT_DEV_STS0_MULTI_IRQ_POLL_EN)) { dyn_regs->gic_host_halt_irq = dyn_regs->gic_host_pi_upd_irq; dyn_regs->gic_host_ints_irq = dyn_regs->gic_host_pi_upd_irq; dev_warn(hdev->dev, "Using a single interrupt interface towards cpucp"); } } /** * hl_fw_dynamic_load_image - load FW image using dynamic protocol * * @hdev: pointer to the habanalabs device structure * @fw_loader: managing structure for loading device's FW * @load_fwc: the FW component to be loaded * @img_ld_timeout: image load timeout * * @return 0 on success, otherwise non-zero error code */ static int hl_fw_dynamic_load_image(struct hl_device *hdev, struct fw_load_mgr *fw_loader, enum hl_fw_component load_fwc, u32 img_ld_timeout) { enum hl_fw_component cur_fwc; const struct firmware *fw; char *fw_name; int rc = 0; /* * when loading image we have one of 2 scenarios: * 1. current FW component is preboot and we want to load boot-fit * 2. current FW component is boot-fit and we want to load linux */ if (load_fwc == FW_COMP_BOOT_FIT) { cur_fwc = FW_COMP_PREBOOT; fw_name = fw_loader->boot_fit_img.image_name; } else { cur_fwc = FW_COMP_BOOT_FIT; fw_name = fw_loader->linux_img.image_name; } /* request FW in order to communicate to FW the size to be allocated */ rc = hl_request_fw(hdev, &fw, fw_name); if (rc) return rc; /* store the image size for future validation */ fw_loader->dynamic_loader.fw_image_size = fw->size; rc = hl_fw_dynamic_request_descriptor(hdev, fw_loader, fw->size); if (rc) goto release_fw; /* read preboot version */ hl_fw_dynamic_read_device_fw_version(hdev, cur_fwc, fw_loader->dynamic_loader.comm_desc.cur_fw_ver); /* update state according to boot stage */ if (cur_fwc == FW_COMP_BOOT_FIT) { struct cpu_dyn_regs *dyn_regs; dyn_regs = &fw_loader->dynamic_loader.comm_desc.cpu_dyn_regs; hl_fw_boot_fit_update_state(hdev, le32_to_cpu(dyn_regs->cpu_boot_dev_sts0), le32_to_cpu(dyn_regs->cpu_boot_dev_sts1)); } /* copy boot fit to space allocated by FW */ rc = hl_fw_dynamic_copy_image(hdev, fw, fw_loader); if (rc) goto release_fw; rc = hl_fw_dynamic_send_protocol_cmd(hdev, fw_loader, COMMS_DATA_RDY, 0, true, fw_loader->cpu_timeout); if (rc) goto release_fw; rc = hl_fw_dynamic_send_protocol_cmd(hdev, fw_loader, COMMS_EXEC, 0, false, img_ld_timeout); release_fw: hl_release_firmware(fw); return rc; } static int hl_fw_dynamic_wait_for_boot_fit_active(struct hl_device *hdev, struct fw_load_mgr *fw_loader) { struct dynamic_fw_load_mgr *dyn_loader; u32 status; int rc; dyn_loader = &fw_loader->dynamic_loader; /* * Make sure CPU boot-loader is running * Note that the CPU_BOOT_STATUS_SRAM_AVAIL is generally set by Linux * yet there is a debug scenario in which we loading uboot (without Linux) * which at later stage is relocated to DRAM. In this case we expect * uboot to set the CPU_BOOT_STATUS_SRAM_AVAIL and so we add it to the * poll flags */ rc = hl_poll_timeout( hdev, le32_to_cpu(dyn_loader->comm_desc.cpu_dyn_regs.cpu_boot_status), status, (status == CPU_BOOT_STATUS_READY_TO_BOOT) || (status == CPU_BOOT_STATUS_SRAM_AVAIL), hdev->fw_poll_interval_usec, dyn_loader->wait_for_bl_timeout); if (rc) { dev_err(hdev->dev, "failed to wait for boot\n"); return rc; } dev_dbg(hdev->dev, "uboot status = %d\n", status); return 0; } static int hl_fw_dynamic_wait_for_linux_active(struct hl_device *hdev, struct fw_load_mgr *fw_loader) { struct dynamic_fw_load_mgr *dyn_loader; u32 status; int rc; dyn_loader = &fw_loader->dynamic_loader; /* Make sure CPU linux is running */ rc = hl_poll_timeout( hdev, le32_to_cpu(dyn_loader->comm_desc.cpu_dyn_regs.cpu_boot_status), status, (status == CPU_BOOT_STATUS_SRAM_AVAIL), hdev->fw_poll_interval_usec, fw_loader->cpu_timeout); if (rc) { dev_err(hdev->dev, "failed to wait for Linux\n"); return rc; } dev_dbg(hdev->dev, "Boot status = %d\n", status); return 0; } /** * hl_fw_linux_update_state - update internal data structures after Linux * is loaded. * Note: Linux initialization is comprised mainly * of two stages - loading kernel (SRAM_AVAIL) * & loading ARMCP. * Therefore reading boot device status in any of * these stages might result in different values. * * @hdev: pointer to the habanalabs device structure * @cpu_boot_dev_sts0_reg: register holding CPU boot dev status 0 * @cpu_boot_dev_sts1_reg: register holding CPU boot dev status 1 * * @return 0 on success, otherwise non-zero error code */ static void hl_fw_linux_update_state(struct hl_device *hdev, u32 cpu_boot_dev_sts0_reg, u32 cpu_boot_dev_sts1_reg) { struct asic_fixed_properties *prop = &hdev->asic_prop; hdev->fw_loader.fw_comp_loaded |= FW_TYPE_LINUX; /* Read FW application security bits */ if (prop->fw_cpu_boot_dev_sts0_valid) { prop->fw_app_cpu_boot_dev_sts0 = RREG32(cpu_boot_dev_sts0_reg); prop->hard_reset_done_by_fw = !!(prop->fw_app_cpu_boot_dev_sts0 & CPU_BOOT_DEV_STS0_FW_HARD_RST_EN); if (prop->fw_app_cpu_boot_dev_sts0 & CPU_BOOT_DEV_STS0_GIC_PRIVILEGED_EN) prop->gic_interrupts_enable = false; dev_dbg(hdev->dev, "Firmware application CPU status0 %#x\n", prop->fw_app_cpu_boot_dev_sts0); dev_dbg(hdev->dev, "GIC controller is %s\n", prop->gic_interrupts_enable ? "enabled" : "disabled"); } if (prop->fw_cpu_boot_dev_sts1_valid) { prop->fw_app_cpu_boot_dev_sts1 = RREG32(cpu_boot_dev_sts1_reg); dev_dbg(hdev->dev, "Firmware application CPU status1 %#x\n", prop->fw_app_cpu_boot_dev_sts1); } dev_dbg(hdev->dev, "Firmware application CPU hard-reset is %s\n", prop->hard_reset_done_by_fw ? "enabled" : "disabled"); dev_info(hdev->dev, "Successfully loaded firmware to device\n"); } /** * hl_fw_dynamic_send_msg - send a COMMS message with attached data * * @hdev: pointer to the habanalabs device structure * @fw_loader: managing structure for loading device's FW * @msg_type: message type * @data: data to be sent * * @return 0 on success, otherwise non-zero error code */ static int hl_fw_dynamic_send_msg(struct hl_device *hdev, struct fw_load_mgr *fw_loader, u8 msg_type, void *data) { struct lkd_msg_comms msg; int rc; memset(&msg, 0, sizeof(msg)); /* create message to be sent */ msg.header.type = msg_type; msg.header.size = cpu_to_le16(sizeof(struct comms_msg_header)); msg.header.magic = cpu_to_le32(HL_COMMS_MSG_MAGIC); switch (msg_type) { case HL_COMMS_RESET_CAUSE_TYPE: msg.reset_cause = *(__u8 *) data; break; default: dev_err(hdev->dev, "Send COMMS message - invalid message type %u\n", msg_type); return -EINVAL; } rc = hl_fw_dynamic_request_descriptor(hdev, fw_loader, sizeof(struct lkd_msg_comms)); if (rc) return rc; /* copy message to space allocated by FW */ rc = hl_fw_dynamic_copy_msg(hdev, &msg, fw_loader); if (rc) return rc; rc = hl_fw_dynamic_send_protocol_cmd(hdev, fw_loader, COMMS_DATA_RDY, 0, true, fw_loader->cpu_timeout); if (rc) return rc; rc = hl_fw_dynamic_send_protocol_cmd(hdev, fw_loader, COMMS_EXEC, 0, true, fw_loader->cpu_timeout); if (rc) return rc; return 0; } /** * hl_fw_dynamic_init_cpu - initialize the device CPU using dynamic protocol * * @hdev: pointer to the habanalabs device structure * @fw_loader: managing structure for loading device's FW * * @return 0 on success, otherwise non-zero error code * * brief: the dynamic protocol is master (LKD) slave (FW CPU) protocol. * the communication is done using registers: * - LKD command register * - FW status register * the protocol is race free. this goal is achieved by splitting the requests * and response to known synchronization points between the LKD and the FW. * each response to LKD request is known and bound to a predefined timeout. * in case of timeout expiration without the desired status from FW- the * protocol (and hence the boot) will fail. */ static int hl_fw_dynamic_init_cpu(struct hl_device *hdev, struct fw_load_mgr *fw_loader) { struct cpu_dyn_regs *dyn_regs; int rc; dev_info(hdev->dev, "Loading firmware to device, may take some time...\n"); /* initialize FW descriptor as invalid */ fw_loader->dynamic_loader.fw_desc_valid = false; /* * In this stage, "cpu_dyn_regs" contains only LKD's hard coded values! * It will be updated from FW after hl_fw_dynamic_request_descriptor(). */ dyn_regs = &fw_loader->dynamic_loader.comm_desc.cpu_dyn_regs; rc = hl_fw_dynamic_send_protocol_cmd(hdev, fw_loader, COMMS_RST_STATE, 0, true, fw_loader->cpu_timeout); if (rc) goto protocol_err; if (hdev->reset_info.curr_reset_cause) { rc = hl_fw_dynamic_send_msg(hdev, fw_loader, HL_COMMS_RESET_CAUSE_TYPE, &hdev->reset_info.curr_reset_cause); if (rc) goto protocol_err; /* Clear current reset cause */ hdev->reset_info.curr_reset_cause = HL_RESET_CAUSE_UNKNOWN; } if (!(hdev->fw_components & FW_TYPE_BOOT_CPU)) { rc = hl_fw_dynamic_request_descriptor(hdev, fw_loader, 0); if (rc) goto protocol_err; /* read preboot version */ hl_fw_dynamic_read_device_fw_version(hdev, FW_COMP_PREBOOT, fw_loader->dynamic_loader.comm_desc.cur_fw_ver); return 0; } /* load boot fit to FW */ rc = hl_fw_dynamic_load_image(hdev, fw_loader, FW_COMP_BOOT_FIT, fw_loader->boot_fit_timeout); if (rc) { dev_err(hdev->dev, "failed to load boot fit\n"); goto protocol_err; } /* * when testing FW load (without Linux) on PLDM we don't want to * wait until boot fit is active as it may take several hours. * instead, we load the bootfit and let it do all initializations in * the background. */ if (hdev->pldm && !(hdev->fw_components & FW_TYPE_LINUX)) return 0; rc = hl_fw_dynamic_wait_for_boot_fit_active(hdev, fw_loader); if (rc) goto protocol_err; /* Enable DRAM scrambling before Linux boot and after successful * UBoot */ hdev->asic_funcs->init_cpu_scrambler_dram(hdev); if (!(hdev->fw_components & FW_TYPE_LINUX)) { dev_info(hdev->dev, "Skip loading Linux F/W\n"); return 0; } if (fw_loader->skip_bmc) { rc = hl_fw_dynamic_send_protocol_cmd(hdev, fw_loader, COMMS_SKIP_BMC, 0, true, fw_loader->cpu_timeout); if (rc) { dev_err(hdev->dev, "failed to load boot fit\n"); goto protocol_err; } } /* load Linux image to FW */ rc = hl_fw_dynamic_load_image(hdev, fw_loader, FW_COMP_LINUX, fw_loader->cpu_timeout); if (rc) { dev_err(hdev->dev, "failed to load Linux\n"); goto protocol_err; } rc = hl_fw_dynamic_wait_for_linux_active(hdev, fw_loader); if (rc) goto protocol_err; hl_fw_linux_update_state(hdev, le32_to_cpu(dyn_regs->cpu_boot_dev_sts0), le32_to_cpu(dyn_regs->cpu_boot_dev_sts1)); hl_fw_dynamic_update_linux_interrupt_if(hdev); return 0; protocol_err: if (fw_loader->dynamic_loader.fw_desc_valid) fw_read_errors(hdev, le32_to_cpu(dyn_regs->cpu_boot_err0), le32_to_cpu(dyn_regs->cpu_boot_err1), le32_to_cpu(dyn_regs->cpu_boot_dev_sts0), le32_to_cpu(dyn_regs->cpu_boot_dev_sts1)); return rc; } /** * hl_fw_static_init_cpu - initialize the device CPU using static protocol * * @hdev: pointer to the habanalabs device structure * @fw_loader: managing structure for loading device's FW * * @return 0 on success, otherwise non-zero error code */ static int hl_fw_static_init_cpu(struct hl_device *hdev, struct fw_load_mgr *fw_loader) { u32 cpu_msg_status_reg, cpu_timeout, msg_to_cpu_reg, status; u32 cpu_boot_dev_status0_reg, cpu_boot_dev_status1_reg; struct static_fw_load_mgr *static_loader; u32 cpu_boot_status_reg; int rc; if (!(hdev->fw_components & FW_TYPE_BOOT_CPU)) return 0; /* init common loader parameters */ cpu_timeout = fw_loader->cpu_timeout; /* init static loader parameters */ static_loader = &fw_loader->static_loader; cpu_msg_status_reg = static_loader->cpu_cmd_status_to_host_reg; msg_to_cpu_reg = static_loader->kmd_msg_to_cpu_reg; cpu_boot_dev_status0_reg = static_loader->cpu_boot_dev_status0_reg; cpu_boot_dev_status1_reg = static_loader->cpu_boot_dev_status1_reg; cpu_boot_status_reg = static_loader->cpu_boot_status_reg; dev_info(hdev->dev, "Going to wait for device boot (up to %lds)\n", cpu_timeout / USEC_PER_SEC); /* Wait for boot FIT request */ rc = hl_poll_timeout( hdev, cpu_boot_status_reg, status, status == CPU_BOOT_STATUS_WAITING_FOR_BOOT_FIT, hdev->fw_poll_interval_usec, fw_loader->boot_fit_timeout); if (rc) { dev_dbg(hdev->dev, "No boot fit request received, resuming boot\n"); } else { rc = hdev->asic_funcs->load_boot_fit_to_device(hdev); if (rc) goto out; /* Clear device CPU message status */ WREG32(cpu_msg_status_reg, CPU_MSG_CLR); /* Signal device CPU that boot loader is ready */ WREG32(msg_to_cpu_reg, KMD_MSG_FIT_RDY); /* Poll for CPU device ack */ rc = hl_poll_timeout( hdev, cpu_msg_status_reg, status, status == CPU_MSG_OK, hdev->fw_poll_interval_usec, fw_loader->boot_fit_timeout); if (rc) { dev_err(hdev->dev, "Timeout waiting for boot fit load ack\n"); goto out; } /* Clear message */ WREG32(msg_to_cpu_reg, KMD_MSG_NA); } /* * Make sure CPU boot-loader is running * Note that the CPU_BOOT_STATUS_SRAM_AVAIL is generally set by Linux * yet there is a debug scenario in which we loading uboot (without Linux) * which at later stage is relocated to DRAM. In this case we expect * uboot to set the CPU_BOOT_STATUS_SRAM_AVAIL and so we add it to the * poll flags */ rc = hl_poll_timeout( hdev, cpu_boot_status_reg, status, (status == CPU_BOOT_STATUS_DRAM_RDY) || (status == CPU_BOOT_STATUS_NIC_FW_RDY) || (status == CPU_BOOT_STATUS_READY_TO_BOOT) || (status == CPU_BOOT_STATUS_SRAM_AVAIL), hdev->fw_poll_interval_usec, cpu_timeout); dev_dbg(hdev->dev, "uboot status = %d\n", status); /* Read U-Boot version now in case we will later fail */ hl_fw_static_read_device_fw_version(hdev, FW_COMP_BOOT_FIT); /* update state according to boot stage */ hl_fw_boot_fit_update_state(hdev, cpu_boot_dev_status0_reg, cpu_boot_dev_status1_reg); if (rc) { detect_cpu_boot_status(hdev, status); rc = -EIO; goto out; } /* Enable DRAM scrambling before Linux boot and after successful * UBoot */ hdev->asic_funcs->init_cpu_scrambler_dram(hdev); if (!(hdev->fw_components & FW_TYPE_LINUX)) { dev_info(hdev->dev, "Skip loading Linux F/W\n"); rc = 0; goto out; } if (status == CPU_BOOT_STATUS_SRAM_AVAIL) { rc = 0; goto out; } dev_info(hdev->dev, "Loading firmware to device, may take some time...\n"); rc = hdev->asic_funcs->load_firmware_to_device(hdev); if (rc) goto out; if (fw_loader->skip_bmc) { WREG32(msg_to_cpu_reg, KMD_MSG_SKIP_BMC); rc = hl_poll_timeout( hdev, cpu_boot_status_reg, status, (status == CPU_BOOT_STATUS_BMC_WAITING_SKIPPED), hdev->fw_poll_interval_usec, cpu_timeout); if (rc) { dev_err(hdev->dev, "Failed to get ACK on skipping BMC, %d\n", status); WREG32(msg_to_cpu_reg, KMD_MSG_NA); rc = -EIO; goto out; } } WREG32(msg_to_cpu_reg, KMD_MSG_FIT_RDY); rc = hl_poll_timeout( hdev, cpu_boot_status_reg, status, (status == CPU_BOOT_STATUS_SRAM_AVAIL), hdev->fw_poll_interval_usec, cpu_timeout); /* Clear message */ WREG32(msg_to_cpu_reg, KMD_MSG_NA); if (rc) { if (status == CPU_BOOT_STATUS_FIT_CORRUPTED) dev_err(hdev->dev, "Device reports FIT image is corrupted\n"); else dev_err(hdev->dev, "Failed to load firmware to device, %d\n", status); rc = -EIO; goto out; } rc = fw_read_errors(hdev, fw_loader->static_loader.boot_err0_reg, fw_loader->static_loader.boot_err1_reg, cpu_boot_dev_status0_reg, cpu_boot_dev_status1_reg); if (rc) return rc; hl_fw_linux_update_state(hdev, cpu_boot_dev_status0_reg, cpu_boot_dev_status1_reg); return 0; out: fw_read_errors(hdev, fw_loader->static_loader.boot_err0_reg, fw_loader->static_loader.boot_err1_reg, cpu_boot_dev_status0_reg, cpu_boot_dev_status1_reg); return rc; } /** * hl_fw_init_cpu - initialize the device CPU * * @hdev: pointer to the habanalabs device structure * * @return 0 on success, otherwise non-zero error code * * perform necessary initializations for device's CPU. takes into account if * init protocol is static or dynamic. */ int hl_fw_init_cpu(struct hl_device *hdev) { struct asic_fixed_properties *prop = &hdev->asic_prop; struct fw_load_mgr *fw_loader = &hdev->fw_loader; return prop->dynamic_fw_load ? hl_fw_dynamic_init_cpu(hdev, fw_loader) : hl_fw_static_init_cpu(hdev, fw_loader); } void hl_fw_set_pll_profile(struct hl_device *hdev) { hl_fw_set_frequency(hdev, hdev->asic_prop.clk_pll_index, hdev->asic_prop.max_freq_value); } int hl_fw_get_clk_rate(struct hl_device *hdev, u32 *cur_clk, u32 *max_clk) { long value; if (!hl_device_operational(hdev, NULL)) return -ENODEV; if (!hdev->pdev) { *cur_clk = 0; *max_clk = 0; return 0; } value = hl_fw_get_frequency(hdev, hdev->asic_prop.clk_pll_index, false); if (value < 0) { dev_err(hdev->dev, "Failed to retrieve device max clock %ld\n", value); return value; } *max_clk = (value / 1000 / 1000); value = hl_fw_get_frequency(hdev, hdev->asic_prop.clk_pll_index, true); if (value < 0) { dev_err(hdev->dev, "Failed to retrieve device current clock %ld\n", value); return value; } *cur_clk = (value / 1000 / 1000); return 0; } long hl_fw_get_frequency(struct hl_device *hdev, u32 pll_index, bool curr) { struct cpucp_packet pkt; u32 used_pll_idx; u64 result; int rc; rc = get_used_pll_index(hdev, pll_index, &used_pll_idx); if (rc) return rc; memset(&pkt, 0, sizeof(pkt)); if (curr) pkt.ctl = cpu_to_le32(CPUCP_PACKET_FREQUENCY_CURR_GET << CPUCP_PKT_CTL_OPCODE_SHIFT); else pkt.ctl = cpu_to_le32(CPUCP_PACKET_FREQUENCY_GET << CPUCP_PKT_CTL_OPCODE_SHIFT); pkt.pll_index = cpu_to_le32((u32)used_pll_idx); rc = hdev->asic_funcs->send_cpu_message(hdev, (u32 *) &pkt, sizeof(pkt), 0, &result); if (rc) { dev_err(hdev->dev, "Failed to get frequency of PLL %d, error %d\n", used_pll_idx, rc); return rc; } return (long) result; } void hl_fw_set_frequency(struct hl_device *hdev, u32 pll_index, u64 freq) { struct cpucp_packet pkt; u32 used_pll_idx; int rc; rc = get_used_pll_index(hdev, pll_index, &used_pll_idx); if (rc) return; memset(&pkt, 0, sizeof(pkt)); pkt.ctl = cpu_to_le32(CPUCP_PACKET_FREQUENCY_SET << CPUCP_PKT_CTL_OPCODE_SHIFT); pkt.pll_index = cpu_to_le32((u32)used_pll_idx); pkt.value = cpu_to_le64(freq); rc = hdev->asic_funcs->send_cpu_message(hdev, (u32 *) &pkt, sizeof(pkt), 0, NULL); if (rc) dev_err(hdev->dev, "Failed to set frequency to PLL %d, error %d\n", used_pll_idx, rc); } long hl_fw_get_max_power(struct hl_device *hdev) { struct cpucp_packet pkt; u64 result; int rc; memset(&pkt, 0, sizeof(pkt)); pkt.ctl = cpu_to_le32(CPUCP_PACKET_MAX_POWER_GET << CPUCP_PKT_CTL_OPCODE_SHIFT); rc = hdev->asic_funcs->send_cpu_message(hdev, (u32 *) &pkt, sizeof(pkt), 0, &result); if (rc) { dev_err(hdev->dev, "Failed to get max power, error %d\n", rc); return rc; } return result; } void hl_fw_set_max_power(struct hl_device *hdev) { struct cpucp_packet pkt; int rc; /* TODO: remove this after simulator supports this packet */ if (!hdev->pdev) return; memset(&pkt, 0, sizeof(pkt)); pkt.ctl = cpu_to_le32(CPUCP_PACKET_MAX_POWER_SET << CPUCP_PKT_CTL_OPCODE_SHIFT); pkt.value = cpu_to_le64(hdev->max_power); rc = hdev->asic_funcs->send_cpu_message(hdev, (u32 *) &pkt, sizeof(pkt), 0, NULL); if (rc) dev_err(hdev->dev, "Failed to set max power, error %d\n", rc); }