// SPDX-License-Identifier: GPL-2.0 OR BSD-3-Clause /* * Copyright (C) 2018-2024 Intel Corporation */ #include #include "iwl-trans.h" #include "iwl-fh.h" #include "iwl-context-info-gen3.h" #include "internal.h" #include "iwl-prph.h" static const struct dmi_system_id dmi_force_scu_active_approved_list[] = { { .ident = "DELL", .matches = { DMI_MATCH(DMI_SYS_VENDOR, "Dell Inc."), }, }, { .ident = "DELL", .matches = { DMI_MATCH(DMI_SYS_VENDOR, "Alienware"), }, }, /* keep last */ {} }; static bool iwl_is_force_scu_active_approved(void) { return !!dmi_check_system(dmi_force_scu_active_approved_list); } static void iwl_pcie_ctxt_info_dbg_enable(struct iwl_trans *trans, struct iwl_prph_scratch_hwm_cfg *dbg_cfg, u32 *control_flags) { enum iwl_fw_ini_allocation_id alloc_id = IWL_FW_INI_ALLOCATION_ID_DBGC1; struct iwl_fw_ini_allocation_tlv *fw_mon_cfg; u32 dbg_flags = 0; if (!iwl_trans_dbg_ini_valid(trans)) { struct iwl_dram_data *fw_mon = &trans->dbg.fw_mon; iwl_pcie_alloc_fw_monitor(trans, 0); if (fw_mon->size) { dbg_flags |= IWL_PRPH_SCRATCH_EDBG_DEST_DRAM; IWL_DEBUG_FW(trans, "WRT: Applying DRAM buffer destination\n"); dbg_cfg->hwm_base_addr = cpu_to_le64(fw_mon->physical); dbg_cfg->hwm_size = cpu_to_le32(fw_mon->size); } goto out; } fw_mon_cfg = &trans->dbg.fw_mon_cfg[alloc_id]; switch (le32_to_cpu(fw_mon_cfg->buf_location)) { case IWL_FW_INI_LOCATION_SRAM_PATH: dbg_flags |= IWL_PRPH_SCRATCH_EDBG_DEST_INTERNAL; IWL_DEBUG_FW(trans, "WRT: Applying SMEM buffer destination\n"); break; case IWL_FW_INI_LOCATION_NPK_PATH: dbg_flags |= IWL_PRPH_SCRATCH_EDBG_DEST_TB22DTF; IWL_DEBUG_FW(trans, "WRT: Applying NPK buffer destination\n"); break; case IWL_FW_INI_LOCATION_DRAM_PATH: if (trans->dbg.fw_mon_ini[alloc_id].num_frags) { struct iwl_dram_data *frag = &trans->dbg.fw_mon_ini[alloc_id].frags[0]; dbg_flags |= IWL_PRPH_SCRATCH_EDBG_DEST_DRAM; dbg_cfg->hwm_base_addr = cpu_to_le64(frag->physical); dbg_cfg->hwm_size = cpu_to_le32(frag->size); dbg_cfg->debug_token_config = cpu_to_le32(trans->dbg.ucode_preset); IWL_DEBUG_FW(trans, "WRT: Applying DRAM destination (debug_token_config=%u)\n", dbg_cfg->debug_token_config); IWL_DEBUG_FW(trans, "WRT: Applying DRAM destination (alloc_id=%u, num_frags=%u)\n", alloc_id, trans->dbg.fw_mon_ini[alloc_id].num_frags); } break; default: IWL_ERR(trans, "WRT: Invalid buffer destination\n"); } out: if (dbg_flags) *control_flags |= IWL_PRPH_SCRATCH_EARLY_DEBUG_EN | dbg_flags; } int iwl_pcie_ctxt_info_gen3_init(struct iwl_trans *trans, const struct fw_img *fw) { struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans); struct iwl_context_info_gen3 *ctxt_info_gen3; struct iwl_prph_scratch *prph_scratch; struct iwl_prph_scratch_ctrl_cfg *prph_sc_ctrl; struct iwl_prph_info *prph_info; u32 control_flags = 0; int ret; int cmdq_size = max_t(u32, IWL_CMD_QUEUE_SIZE, trans->cfg->min_txq_size); switch (trans_pcie->rx_buf_size) { case IWL_AMSDU_DEF: return -EINVAL; case IWL_AMSDU_2K: break; case IWL_AMSDU_4K: control_flags |= IWL_PRPH_SCRATCH_RB_SIZE_4K; break; case IWL_AMSDU_8K: control_flags |= IWL_PRPH_SCRATCH_RB_SIZE_4K; /* if firmware supports the ext size, tell it */ control_flags |= IWL_PRPH_SCRATCH_RB_SIZE_EXT_8K; break; case IWL_AMSDU_12K: control_flags |= IWL_PRPH_SCRATCH_RB_SIZE_4K; /* if firmware supports the ext size, tell it */ control_flags |= IWL_PRPH_SCRATCH_RB_SIZE_EXT_16K; break; } /* Allocate prph scratch */ prph_scratch = dma_alloc_coherent(trans->dev, sizeof(*prph_scratch), &trans_pcie->prph_scratch_dma_addr, GFP_KERNEL); if (!prph_scratch) return -ENOMEM; prph_sc_ctrl = &prph_scratch->ctrl_cfg; prph_sc_ctrl->version.version = 0; prph_sc_ctrl->version.mac_id = cpu_to_le16((u16)trans->hw_rev); prph_sc_ctrl->version.size = cpu_to_le16(sizeof(*prph_scratch) / 4); control_flags |= IWL_PRPH_SCRATCH_MTR_MODE; control_flags |= IWL_PRPH_MTR_FORMAT_256B & IWL_PRPH_SCRATCH_MTR_FORMAT; if (trans->trans_cfg->imr_enabled) control_flags |= IWL_PRPH_SCRATCH_IMR_DEBUG_EN; if (CSR_HW_REV_TYPE(trans->hw_rev) == IWL_CFG_MAC_TYPE_GL && iwl_is_force_scu_active_approved()) { control_flags |= IWL_PRPH_SCRATCH_SCU_FORCE_ACTIVE; IWL_DEBUG_FW(trans, "Context Info: Set SCU_FORCE_ACTIVE (0x%x) in control_flags\n", IWL_PRPH_SCRATCH_SCU_FORCE_ACTIVE); } /* initialize RX default queue */ prph_sc_ctrl->rbd_cfg.free_rbd_addr = cpu_to_le64(trans_pcie->rxq->bd_dma); iwl_pcie_ctxt_info_dbg_enable(trans, &prph_sc_ctrl->hwm_cfg, &control_flags); prph_sc_ctrl->control.control_flags = cpu_to_le32(control_flags); /* initialize the Step equalizer data */ prph_sc_ctrl->step_cfg.mbx_addr_0 = cpu_to_le32(trans->mbx_addr_0_step); prph_sc_ctrl->step_cfg.mbx_addr_1 = cpu_to_le32(trans->mbx_addr_1_step); /* allocate ucode sections in dram and set addresses */ ret = iwl_pcie_init_fw_sec(trans, fw, &prph_scratch->dram); if (ret) goto err_free_prph_scratch; /* Allocate prph information * currently we don't assign to the prph info anything, but it would get * assigned later * * We also use the second half of this page to give the device some * dummy TR/CR tail pointers - which shouldn't be necessary as we don't * use this, but the hardware still reads/writes there and we can't let * it go do that with a NULL pointer. */ BUILD_BUG_ON(sizeof(*prph_info) > PAGE_SIZE / 2); prph_info = dma_alloc_coherent(trans->dev, PAGE_SIZE, &trans_pcie->prph_info_dma_addr, GFP_KERNEL); if (!prph_info) { ret = -ENOMEM; goto err_free_prph_scratch; } /* Allocate context info */ ctxt_info_gen3 = dma_alloc_coherent(trans->dev, sizeof(*ctxt_info_gen3), &trans_pcie->ctxt_info_dma_addr, GFP_KERNEL); if (!ctxt_info_gen3) { ret = -ENOMEM; goto err_free_prph_info; } ctxt_info_gen3->prph_info_base_addr = cpu_to_le64(trans_pcie->prph_info_dma_addr); ctxt_info_gen3->prph_scratch_base_addr = cpu_to_le64(trans_pcie->prph_scratch_dma_addr); ctxt_info_gen3->prph_scratch_size = cpu_to_le32(sizeof(*prph_scratch)); ctxt_info_gen3->cr_head_idx_arr_base_addr = cpu_to_le64(trans_pcie->rxq->rb_stts_dma); ctxt_info_gen3->tr_tail_idx_arr_base_addr = cpu_to_le64(trans_pcie->prph_info_dma_addr + PAGE_SIZE / 2); ctxt_info_gen3->cr_tail_idx_arr_base_addr = cpu_to_le64(trans_pcie->prph_info_dma_addr + 3 * PAGE_SIZE / 4); ctxt_info_gen3->mtr_base_addr = cpu_to_le64(trans_pcie->txqs.txq[trans_pcie->txqs.cmd.q_id]->dma_addr); ctxt_info_gen3->mcr_base_addr = cpu_to_le64(trans_pcie->rxq->used_bd_dma); ctxt_info_gen3->mtr_size = cpu_to_le16(TFD_QUEUE_CB_SIZE(cmdq_size)); ctxt_info_gen3->mcr_size = cpu_to_le16(RX_QUEUE_CB_SIZE(trans->cfg->num_rbds)); trans_pcie->ctxt_info_gen3 = ctxt_info_gen3; trans_pcie->prph_info = prph_info; trans_pcie->prph_scratch = prph_scratch; /* Allocate IML */ trans_pcie->iml = dma_alloc_coherent(trans->dev, trans->iml_len, &trans_pcie->iml_dma_addr, GFP_KERNEL); if (!trans_pcie->iml) { ret = -ENOMEM; goto err_free_ctxt_info; } memcpy(trans_pcie->iml, trans->iml, trans->iml_len); iwl_enable_fw_load_int_ctx_info(trans); /* kick FW self load */ iwl_write64(trans, CSR_CTXT_INFO_ADDR, trans_pcie->ctxt_info_dma_addr); iwl_write64(trans, CSR_IML_DATA_ADDR, trans_pcie->iml_dma_addr); iwl_write32(trans, CSR_IML_SIZE_ADDR, trans->iml_len); iwl_set_bit(trans, CSR_CTXT_INFO_BOOT_CTRL, CSR_AUTO_FUNC_BOOT_ENA); return 0; err_free_ctxt_info: dma_free_coherent(trans->dev, sizeof(*trans_pcie->ctxt_info_gen3), trans_pcie->ctxt_info_gen3, trans_pcie->ctxt_info_dma_addr); trans_pcie->ctxt_info_gen3 = NULL; err_free_prph_info: dma_free_coherent(trans->dev, PAGE_SIZE, prph_info, trans_pcie->prph_info_dma_addr); err_free_prph_scratch: dma_free_coherent(trans->dev, sizeof(*prph_scratch), prph_scratch, trans_pcie->prph_scratch_dma_addr); return ret; } void iwl_pcie_ctxt_info_gen3_free(struct iwl_trans *trans, bool alive) { struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans); if (trans_pcie->iml) { dma_free_coherent(trans->dev, trans->iml_len, trans_pcie->iml, trans_pcie->iml_dma_addr); trans_pcie->iml_dma_addr = 0; trans_pcie->iml = NULL; } iwl_pcie_ctxt_info_free_fw_img(trans); if (alive) return; if (!trans_pcie->ctxt_info_gen3) return; /* ctxt_info_gen3 and prph_scratch are still needed for PNVM load */ dma_free_coherent(trans->dev, sizeof(*trans_pcie->ctxt_info_gen3), trans_pcie->ctxt_info_gen3, trans_pcie->ctxt_info_dma_addr); trans_pcie->ctxt_info_dma_addr = 0; trans_pcie->ctxt_info_gen3 = NULL; dma_free_coherent(trans->dev, sizeof(*trans_pcie->prph_scratch), trans_pcie->prph_scratch, trans_pcie->prph_scratch_dma_addr); trans_pcie->prph_scratch_dma_addr = 0; trans_pcie->prph_scratch = NULL; /* this is needed for the entire lifetime */ dma_free_coherent(trans->dev, PAGE_SIZE, trans_pcie->prph_info, trans_pcie->prph_info_dma_addr); trans_pcie->prph_info_dma_addr = 0; trans_pcie->prph_info = NULL; } static int iwl_pcie_load_payloads_continuously(struct iwl_trans *trans, const struct iwl_pnvm_image *pnvm_data, struct iwl_dram_data *dram) { u32 len, len0, len1; if (pnvm_data->n_chunks != UNFRAGMENTED_PNVM_PAYLOADS_NUMBER) { IWL_DEBUG_FW(trans, "expected 2 payloads, got %d.\n", pnvm_data->n_chunks); return -EINVAL; } len0 = pnvm_data->chunks[0].len; len1 = pnvm_data->chunks[1].len; if (len1 > 0xFFFFFFFF - len0) { IWL_DEBUG_FW(trans, "sizes of payloads overflow.\n"); return -EINVAL; } len = len0 + len1; dram->block = iwl_pcie_ctxt_info_dma_alloc_coherent(trans, len, &dram->physical); if (!dram->block) { IWL_DEBUG_FW(trans, "Failed to allocate PNVM DMA.\n"); return -ENOMEM; } dram->size = len; memcpy(dram->block, pnvm_data->chunks[0].data, len0); memcpy((u8 *)dram->block + len0, pnvm_data->chunks[1].data, len1); return 0; } static int iwl_pcie_load_payloads_segments (struct iwl_trans *trans, struct iwl_dram_regions *dram_regions, const struct iwl_pnvm_image *pnvm_data) { struct iwl_dram_data *cur_payload_dram = &dram_regions->drams[0]; struct iwl_dram_data *desc_dram = &dram_regions->prph_scratch_mem_desc; struct iwl_prph_scrath_mem_desc_addr_array *addresses; const void *data; u32 len; int i; /* allocate and init DRAM descriptors array */ len = sizeof(struct iwl_prph_scrath_mem_desc_addr_array); desc_dram->block = iwl_pcie_ctxt_info_dma_alloc_coherent (trans, len, &desc_dram->physical); if (!desc_dram->block) { IWL_DEBUG_FW(trans, "Failed to allocate PNVM DMA.\n"); return -ENOMEM; } desc_dram->size = len; memset(desc_dram->block, 0, len); /* allocate DRAM region for each payload */ dram_regions->n_regions = 0; for (i = 0; i < pnvm_data->n_chunks; i++) { len = pnvm_data->chunks[i].len; data = pnvm_data->chunks[i].data; if (iwl_pcie_ctxt_info_alloc_dma(trans, data, len, cur_payload_dram)) { iwl_trans_pcie_free_pnvm_dram_regions(dram_regions, trans->dev); return -ENOMEM; } dram_regions->n_regions++; cur_payload_dram++; } /* fill desc with the DRAM payloads addresses */ addresses = desc_dram->block; for (i = 0; i < pnvm_data->n_chunks; i++) { addresses->mem_descs[i] = cpu_to_le64(dram_regions->drams[i].physical); } return 0; } int iwl_trans_pcie_ctx_info_gen3_load_pnvm(struct iwl_trans *trans, const struct iwl_pnvm_image *pnvm_payloads, const struct iwl_ucode_capabilities *capa) { struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans); struct iwl_prph_scratch_ctrl_cfg *prph_sc_ctrl = &trans_pcie->prph_scratch->ctrl_cfg; struct iwl_dram_regions *dram_regions = &trans_pcie->pnvm_data; int ret = 0; /* only allocate the DRAM if not allocated yet */ if (trans->pnvm_loaded) return 0; if (WARN_ON(prph_sc_ctrl->pnvm_cfg.pnvm_size)) return -EBUSY; if (trans->trans_cfg->device_family < IWL_DEVICE_FAMILY_AX210) return 0; if (!pnvm_payloads->n_chunks) { IWL_DEBUG_FW(trans, "no payloads\n"); return -EINVAL; } /* save payloads in several DRAM sections */ if (fw_has_capa(capa, IWL_UCODE_TLV_CAPA_FRAGMENTED_PNVM_IMG)) { ret = iwl_pcie_load_payloads_segments(trans, dram_regions, pnvm_payloads); if (!ret) trans->pnvm_loaded = true; } else { /* save only in one DRAM section */ ret = iwl_pcie_load_payloads_continuously (trans, pnvm_payloads, &dram_regions->drams[0]); if (!ret) { dram_regions->n_regions = 1; trans->pnvm_loaded = true; } } return ret; } static inline size_t iwl_dram_regions_size(const struct iwl_dram_regions *dram_regions) { size_t total_size = 0; int i; for (i = 0; i < dram_regions->n_regions; i++) total_size += dram_regions->drams[i].size; return total_size; } static void iwl_pcie_set_pnvm_segments(struct iwl_trans *trans) { struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans); struct iwl_prph_scratch_ctrl_cfg *prph_sc_ctrl = &trans_pcie->prph_scratch->ctrl_cfg; struct iwl_dram_regions *dram_regions = &trans_pcie->pnvm_data; prph_sc_ctrl->pnvm_cfg.pnvm_base_addr = cpu_to_le64(dram_regions->prph_scratch_mem_desc.physical); prph_sc_ctrl->pnvm_cfg.pnvm_size = cpu_to_le32(iwl_dram_regions_size(dram_regions)); } static void iwl_pcie_set_continuous_pnvm(struct iwl_trans *trans) { struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans); struct iwl_prph_scratch_ctrl_cfg *prph_sc_ctrl = &trans_pcie->prph_scratch->ctrl_cfg; prph_sc_ctrl->pnvm_cfg.pnvm_base_addr = cpu_to_le64(trans_pcie->pnvm_data.drams[0].physical); prph_sc_ctrl->pnvm_cfg.pnvm_size = cpu_to_le32(trans_pcie->pnvm_data.drams[0].size); } void iwl_trans_pcie_ctx_info_gen3_set_pnvm(struct iwl_trans *trans, const struct iwl_ucode_capabilities *capa) { if (trans->trans_cfg->device_family < IWL_DEVICE_FAMILY_AX210) return; if (fw_has_capa(capa, IWL_UCODE_TLV_CAPA_FRAGMENTED_PNVM_IMG)) iwl_pcie_set_pnvm_segments(trans); else iwl_pcie_set_continuous_pnvm(trans); } int iwl_trans_pcie_ctx_info_gen3_load_reduce_power(struct iwl_trans *trans, const struct iwl_pnvm_image *payloads, const struct iwl_ucode_capabilities *capa) { struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans); struct iwl_prph_scratch_ctrl_cfg *prph_sc_ctrl = &trans_pcie->prph_scratch->ctrl_cfg; struct iwl_dram_regions *dram_regions = &trans_pcie->reduced_tables_data; int ret = 0; /* only allocate the DRAM if not allocated yet */ if (trans->reduce_power_loaded) return 0; if (trans->trans_cfg->device_family < IWL_DEVICE_FAMILY_AX210) return 0; if (WARN_ON(prph_sc_ctrl->reduce_power_cfg.size)) return -EBUSY; if (!payloads->n_chunks) { IWL_DEBUG_FW(trans, "no payloads\n"); return -EINVAL; } /* save payloads in several DRAM sections */ if (fw_has_capa(capa, IWL_UCODE_TLV_CAPA_FRAGMENTED_PNVM_IMG)) { ret = iwl_pcie_load_payloads_segments(trans, dram_regions, payloads); if (!ret) trans->reduce_power_loaded = true; } else { /* save only in one DRAM section */ ret = iwl_pcie_load_payloads_continuously (trans, payloads, &dram_regions->drams[0]); if (!ret) { dram_regions->n_regions = 1; trans->reduce_power_loaded = true; } } return ret; } static void iwl_pcie_set_reduce_power_segments(struct iwl_trans *trans) { struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans); struct iwl_prph_scratch_ctrl_cfg *prph_sc_ctrl = &trans_pcie->prph_scratch->ctrl_cfg; struct iwl_dram_regions *dram_regions = &trans_pcie->reduced_tables_data; prph_sc_ctrl->reduce_power_cfg.base_addr = cpu_to_le64(dram_regions->prph_scratch_mem_desc.physical); prph_sc_ctrl->reduce_power_cfg.size = cpu_to_le32(iwl_dram_regions_size(dram_regions)); } static void iwl_pcie_set_continuous_reduce_power(struct iwl_trans *trans) { struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans); struct iwl_prph_scratch_ctrl_cfg *prph_sc_ctrl = &trans_pcie->prph_scratch->ctrl_cfg; prph_sc_ctrl->reduce_power_cfg.base_addr = cpu_to_le64(trans_pcie->reduced_tables_data.drams[0].physical); prph_sc_ctrl->reduce_power_cfg.size = cpu_to_le32(trans_pcie->reduced_tables_data.drams[0].size); } void iwl_trans_pcie_ctx_info_gen3_set_reduce_power(struct iwl_trans *trans, const struct iwl_ucode_capabilities *capa) { if (trans->trans_cfg->device_family < IWL_DEVICE_FAMILY_AX210) return; if (fw_has_capa(capa, IWL_UCODE_TLV_CAPA_FRAGMENTED_PNVM_IMG)) iwl_pcie_set_reduce_power_segments(trans); else iwl_pcie_set_continuous_reduce_power(trans); }