scsi: lpfc: Resize cpu maps structures based on possible cpus

The work done to date utilized the number of present cpus when sizing
per-cpu structures. Structures should have been sized based on the max
possible cpu count.

Convert the driver over to possible cpu count for sizing allocation.

Signed-off-by: Dick Kennedy <dick.kennedy@broadcom.com>
Signed-off-by: James Smart <jsmart2021@gmail.com>
Reviewed-by: Hannes Reinecke <hare@suse.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>

4 files changed, 51 insertions, 41 deletions

diff --git a/drivers/scsi/lpfc/lpfc_attr.c b/drivers/scsi/lpfc/lpfc_attr.c
index 2864cb53b1e8..a114965a376c 100644
--- a/drivers/scsi/lpfc/lpfc_attr.c
+++ b/drivers/scsi/lpfc/lpfc_attr.c
@@ -5176,16 +5176,22 @@ lpfc_fcp_cpu_map_show(struct device *dev, struct device_attribute *attr,
 	case 1:
 		len += snprintf(buf + len, PAGE_SIZE-len,
 				"fcp_cpu_map: HBA centric mapping (%d): "
-				"%d online CPUs\n",
-				phba->cfg_fcp_cpu_map,
-				phba->sli4_hba.num_online_cpu);
+				"%d of %d CPUs online from %d possible CPUs\n",
+				phba->cfg_fcp_cpu_map, num_online_cpus(),
+				num_present_cpus(),
+				phba->sli4_hba.num_possible_cpu);
 		break;
 	}
 
-	while (phba->sli4_hba.curr_disp_cpu < phba->sli4_hba.num_present_cpu) {
+	while (phba->sli4_hba.curr_disp_cpu <
+	       phba->sli4_hba.num_possible_cpu) {
 		cpup = &phba->sli4_hba.cpu_map[phba->sli4_hba.curr_disp_cpu];
 
-		if (cpup->irq == LPFC_VECTOR_MAP_EMPTY) {
+		if (!cpu_present(phba->sli4_hba.curr_disp_cpu))
+			len += snprintf(buf + len, PAGE_SIZE - len,
+					"CPU %02d not present\n",
+					phba->sli4_hba.curr_disp_cpu);
+		else if (cpup->irq == LPFC_VECTOR_MAP_EMPTY) {
 			if (cpup->hdwq == LPFC_VECTOR_MAP_EMPTY)
 				len += snprintf(
 					buf + len, PAGE_SIZE - len,
@@ -5225,14 +5231,15 @@ lpfc_fcp_cpu_map_show(struct device *dev, struct device_attribute *attr,
 
 		/* display max number of CPUs keeping some margin */
 		if (phba->sli4_hba.curr_disp_cpu <
-				phba->sli4_hba.num_present_cpu &&
+				phba->sli4_hba.num_possible_cpu &&
 				(len >= (PAGE_SIZE - 64))) {
-			len += snprintf(buf + len, PAGE_SIZE-len, "more...\n");
+			len += snprintf(buf + len,
+					PAGE_SIZE - len, "more...\n");
 			break;
 		}
 	}
 
-	if (phba->sli4_hba.curr_disp_cpu == phba->sli4_hba.num_present_cpu)
+	if (phba->sli4_hba.curr_disp_cpu == phba->sli4_hba.num_possible_cpu)
 		phba->sli4_hba.curr_disp_cpu = 0;
 
 	return len;
diff --git a/drivers/scsi/lpfc/lpfc_init.c b/drivers/scsi/lpfc/lpfc_init.c
index 05919480e430..8ba2861db7b6 100644
--- a/drivers/scsi/lpfc/lpfc_init.c
+++ b/drivers/scsi/lpfc/lpfc_init.c
@@ -6373,8 +6373,8 @@ lpfc_sli4_driver_resource_setup(struct lpfc_hba *phba)
 	u32 if_type;
 	u32 if_fam;
 
-	phba->sli4_hba.num_online_cpu = num_online_cpus();
 	phba->sli4_hba.num_present_cpu = lpfc_present_cpu;
+	phba->sli4_hba.num_possible_cpu = num_possible_cpus();
 	phba->sli4_hba.curr_disp_cpu = 0;
 
 	/* Get all the module params for configuring this host */
@@ -6796,7 +6796,7 @@ lpfc_sli4_driver_resource_setup(struct lpfc_hba *phba)
 		goto out_free_fcf_rr_bmask;
 	}
 
-	phba->sli4_hba.cpu_map = kcalloc(phba->sli4_hba.num_present_cpu,
+	phba->sli4_hba.cpu_map = kcalloc(phba->sli4_hba.num_possible_cpu,
 					sizeof(struct lpfc_vector_map_info),
 					GFP_KERNEL);
 	if (!phba->sli4_hba.cpu_map) {
@@ -6868,8 +6868,8 @@ lpfc_sli4_driver_resource_unset(struct lpfc_hba *phba)
 
 	/* Free memory allocated for msi-x interrupt vector to CPU mapping */
 	kfree(phba->sli4_hba.cpu_map);
+	phba->sli4_hba.num_possible_cpu = 0;
 	phba->sli4_hba.num_present_cpu = 0;
-	phba->sli4_hba.num_online_cpu = 0;
 	phba->sli4_hba.curr_disp_cpu = 0;
 
 	/* Free memory allocated for fast-path work queue handles */
@@ -10519,15 +10519,14 @@ lpfc_find_cpu_handle(struct lpfc_hba *phba, uint16_t id, int match)
 	int cpu;
 
 	/* Find the desired phys_id for the specified EQ */
-	cpup = phba->sli4_hba.cpu_map;
-	for (cpu = 0; cpu < phba->sli4_hba.num_present_cpu; cpu++) {
+	for_each_present_cpu(cpu) {
+		cpup = &phba->sli4_hba.cpu_map[cpu];
 		if ((match == LPFC_FIND_BY_EQ) &&
 		    (cpup->irq != LPFC_VECTOR_MAP_EMPTY) &&
 		    (cpup->eq == id))
 			return cpu;
 		if ((match == LPFC_FIND_BY_HDWQ) && (cpup->hdwq == id))
 			return cpu;
-		cpup++;
 	}
 	return 0;
 }
@@ -10545,11 +10544,10 @@ lpfc_find_eq_handle(struct lpfc_hba *phba, uint16_t hdwq)
 	int cpu;
 
 	/* Find the desired phys_id for the specified EQ */
-	cpup = phba->sli4_hba.cpu_map;
-	for (cpu = 0; cpu < phba->sli4_hba.num_present_cpu; cpu++) {
+	for_each_present_cpu(cpu) {
+		cpup = &phba->sli4_hba.cpu_map[cpu];
 		if (cpup->hdwq == hdwq)
 			return cpup->eq;
-		cpup++;
 	}
 	return 0;
 }
@@ -10569,15 +10567,13 @@ lpfc_find_hyper(struct lpfc_hba *phba, int cpu,
 	struct lpfc_vector_map_info *cpup;
 	int idx;
 
-	cpup = phba->sli4_hba.cpu_map;
-	for (idx = 0; idx < phba->sli4_hba.num_present_cpu; idx++) {
+	for_each_present_cpu(idx) {
+		cpup = &phba->sli4_hba.cpu_map[idx];
 		/* Does the cpup match the one we are looking for */
 		if ((cpup->phys_id == phys_id) &&
 		    (cpup->core_id == core_id) &&
-		    (cpu != idx)) {
+		    (cpu != idx))
 			return 1;
-		}
-		cpup++;
 	}
 	return 0;
 }
@@ -10608,7 +10604,7 @@ lpfc_cpu_affinity_check(struct lpfc_hba *phba, int vectors)
 	/* Init cpu_map array */
 	memset(phba->sli4_hba.cpu_map, 0xff,
 	       (sizeof(struct lpfc_vector_map_info) *
-	       phba->sli4_hba.num_present_cpu));
+	       phba->sli4_hba.num_possible_cpu));
 
 	max_phys_id = 0;
 	min_phys_id = 0xffff;
@@ -10617,8 +10613,8 @@ lpfc_cpu_affinity_check(struct lpfc_hba *phba, int vectors)
 	phys_id = 0;
 
 	/* Update CPU map with physical id and core id of each CPU */
-	cpup = phba->sli4_hba.cpu_map;
-	for (cpu = 0; cpu < phba->sli4_hba.num_present_cpu; cpu++) {
+	for_each_present_cpu(cpu) {
+		cpup = &phba->sli4_hba.cpu_map[cpu];
 #ifdef CONFIG_X86
 		cpuinfo = &cpu_data(cpu);
 		cpup->phys_id = cpuinfo->phys_proc_id;
@@ -10645,8 +10641,6 @@ lpfc_cpu_affinity_check(struct lpfc_hba *phba, int vectors)
 			max_core_id = cpup->core_id;
 		if (cpup->core_id < min_core_id)
 			min_core_id = cpup->core_id;
-
-		cpup++;
 	}
 
 	for_each_possible_cpu(i) {
diff --git a/drivers/scsi/lpfc/lpfc_nvmet.c b/drivers/scsi/lpfc/lpfc_nvmet.c
index 0d296aee2d82..0b27e8c5ae32 100644
--- a/drivers/scsi/lpfc/lpfc_nvmet.c
+++ b/drivers/scsi/lpfc/lpfc_nvmet.c
@@ -1194,9 +1194,9 @@ lpfc_nvmet_cleanup_io_context(struct lpfc_hba *phba)
 
 	/* Cycle the the entire CPU context list for every MRQ */
 	for (i = 0; i < phba->cfg_nvmet_mrq; i++) {
-		for (j = 0; j < phba->sli4_hba.num_present_cpu; j++) {
+		for_each_present_cpu(j) {
+			infop = lpfc_get_ctx_list(phba, j, i);
 			__lpfc_nvmet_clean_io_for_cpu(phba, infop);
-			infop++; /* next */
 		}
 	}
 	kfree(phba->sli4_hba.nvmet_ctx_info);
@@ -1211,14 +1211,14 @@ lpfc_nvmet_setup_io_context(struct lpfc_hba *phba)
 	union lpfc_wqe128 *wqe;
 	struct lpfc_nvmet_ctx_info *last_infop;
 	struct lpfc_nvmet_ctx_info *infop;
-	int i, j, idx;
+	int i, j, idx, cpu;
 
 	lpfc_printf_log(phba, KERN_INFO, LOG_NVME,
 			"6403 Allocate NVMET resources for %d XRIs\n",
 			phba->sli4_hba.nvmet_xri_cnt);
 
 	phba->sli4_hba.nvmet_ctx_info = kcalloc(
-		phba->sli4_hba.num_present_cpu * phba->cfg_nvmet_mrq,
+		phba->sli4_hba.num_possible_cpu * phba->cfg_nvmet_mrq,
 		sizeof(struct lpfc_nvmet_ctx_info), GFP_KERNEL);
 	if (!phba->sli4_hba.nvmet_ctx_info) {
 		lpfc_printf_log(phba, KERN_ERR, LOG_INIT,
@@ -1246,13 +1246,12 @@ lpfc_nvmet_setup_io_context(struct lpfc_hba *phba)
 	 * of the IO completion. Thus a context that was allocated for MRQ A
 	 * whose IO completed on CPU B will be freed to cpuB/mrqA.
 	 */
-	infop = phba->sli4_hba.nvmet_ctx_info;
-	for (i = 0; i < phba->sli4_hba.num_present_cpu; i++) {
+	for_each_possible_cpu(i) {
 		for (j = 0; j < phba->cfg_nvmet_mrq; j++) {
+			infop = lpfc_get_ctx_list(phba, i, j);
 			INIT_LIST_HEAD(&infop->nvmet_ctx_list);
 			spin_lock_init(&infop->nvmet_ctx_list_lock);
 			infop->nvmet_ctx_list_cnt = 0;
-			infop++;
 		}
 	}
 
@@ -1262,8 +1261,10 @@ lpfc_nvmet_setup_io_context(struct lpfc_hba *phba)
 	 * MRQ 1 cycling thru CPUs 0 - X, and so on.
 	 */
 	for (j = 0; j < phba->cfg_nvmet_mrq; j++) {
-		last_infop = lpfc_get_ctx_list(phba, 0, j);
-		for (i = phba->sli4_hba.num_present_cpu - 1;  i >= 0; i--) {
+		last_infop = lpfc_get_ctx_list(phba,
+					       cpumask_first(cpu_present_mask),
+					       j);
+		for (i = phba->sli4_hba.num_possible_cpu - 1;  i >= 0; i--) {
 			infop = lpfc_get_ctx_list(phba, i, j);
 			infop->nvmet_ctx_next_cpu = last_infop;
 			last_infop = infop;
@@ -1274,6 +1275,7 @@ lpfc_nvmet_setup_io_context(struct lpfc_hba *phba)
 	 * received command on a per xri basis.
 	 */
 	idx = 0;
+	cpu = cpumask_first(cpu_present_mask);
 	for (i = 0; i < phba->sli4_hba.nvmet_xri_cnt; i++) {
 		ctx_buf = kzalloc(sizeof(*ctx_buf), GFP_KERNEL);
 		if (!ctx_buf) {
@@ -1327,7 +1329,7 @@ lpfc_nvmet_setup_io_context(struct lpfc_hba *phba)
 		 * is MRQidx will be associated with CPUidx. This association
 		 * can change on the fly.
 		 */
-		infop = lpfc_get_ctx_list(phba, idx, idx);
+		infop = lpfc_get_ctx_list(phba, cpu, idx);
 		spin_lock(&infop->nvmet_ctx_list_lock);
 		list_add_tail(&ctx_buf->list, &infop->nvmet_ctx_list);
 		infop->nvmet_ctx_list_cnt++;
@@ -1335,11 +1337,18 @@ lpfc_nvmet_setup_io_context(struct lpfc_hba *phba)
 
 		/* Spread ctx structures evenly across all MRQs */
 		idx++;
-		if (idx >= phba->cfg_nvmet_mrq)
+		if (idx >= phba->cfg_nvmet_mrq) {
 			idx = 0;
+			cpu = cpumask_first(cpu_present_mask);
+			continue;
+		}
+		cpu = cpumask_next(cpu, cpu_present_mask);
+		if (cpu == nr_cpu_ids)
+			cpu = cpumask_first(cpu_present_mask);
+
 	}
 
-	for (i = 0; i < phba->sli4_hba.num_present_cpu; i++) {
+	for_each_present_cpu(i) {
 		for (j = 0; j < phba->cfg_nvmet_mrq; j++) {
 			infop = lpfc_get_ctx_list(phba, i, j);
 			lpfc_printf_log(phba, KERN_INFO, LOG_NVME | LOG_INIT,
@@ -1839,7 +1848,7 @@ lpfc_nvmet_replenish_context(struct lpfc_hba *phba,
 	else
 		get_infop = current_infop->nvmet_ctx_next_cpu;
 
-	for (i = 0; i < phba->sli4_hba.num_present_cpu; i++) {
+	for (i = 0; i < phba->sli4_hba.num_possible_cpu; i++) {
 		if (get_infop == current_infop) {
 			get_infop = get_infop->nvmet_ctx_next_cpu;
 			continue;
diff --git a/drivers/scsi/lpfc/lpfc_sli4.h b/drivers/scsi/lpfc/lpfc_sli4.h
index 20566c506e5f..1e3d7f534eaa 100644
--- a/drivers/scsi/lpfc/lpfc_sli4.h
+++ b/drivers/scsi/lpfc/lpfc_sli4.h
@@ -890,7 +890,7 @@ struct lpfc_sli4_hba {
 
 	/* CPU to vector mapping information */
 	struct lpfc_vector_map_info *cpu_map;
-	uint16_t num_online_cpu;
+	uint16_t num_possible_cpu;
 	uint16_t num_present_cpu;
 	uint16_t curr_disp_cpu;
 	struct lpfc_eq_intr_info __percpu *eq_info;