aboutsummaryrefslogtreecommitdiff
path: root/drivers/infiniband/hw/ehca/ehca_irq.c
diff options
context:
space:
mode:
Diffstat (limited to 'drivers/infiniband/hw/ehca/ehca_irq.c')
-rw-r--r--drivers/infiniband/hw/ehca/ehca_irq.c350
1 files changed, 151 insertions, 199 deletions
diff --git a/drivers/infiniband/hw/ehca/ehca_irq.c b/drivers/infiniband/hw/ehca/ehca_irq.c
index b5ca94c6b8d..8615d7cf7e0 100644
--- a/drivers/infiniband/hw/ehca/ehca_irq.c
+++ b/drivers/infiniband/hw/ehca/ehca_irq.c
@@ -41,6 +41,9 @@
* POSSIBILITY OF SUCH DAMAGE.
*/
+#include <linux/slab.h>
+#include <linux/smpboot.h>
+
#include "ehca_classes.h"
#include "ehca_irq.h"
#include "ehca_iverbs.h"
@@ -99,7 +102,7 @@ static void print_error_data(struct ehca_shca *shca, void *data,
return;
ehca_err(&shca->ib_device,
- "QP 0x%x (resource=%lx) has errors.",
+ "QP 0x%x (resource=%llx) has errors.",
qp->ib_qp.qp_num, resource);
break;
}
@@ -108,21 +111,21 @@ static void print_error_data(struct ehca_shca *shca, void *data,
struct ehca_cq *cq = (struct ehca_cq *)data;
ehca_err(&shca->ib_device,
- "CQ 0x%x (resource=%lx) has errors.",
+ "CQ 0x%x (resource=%llx) has errors.",
cq->cq_number, resource);
break;
}
default:
ehca_err(&shca->ib_device,
- "Unknown error type: %lx on %s.",
+ "Unknown error type: %llx on %s.",
type, shca->ib_device.name);
break;
}
- ehca_err(&shca->ib_device, "Error data is available: %lx.", resource);
+ ehca_err(&shca->ib_device, "Error data is available: %llx.", resource);
ehca_err(&shca->ib_device, "EHCA ----- error data begin "
"---------------------------------------------------");
- ehca_dmp(rblock, length, "resource=%lx", resource);
+ ehca_dmp(rblock, length, "resource=%llx", resource);
ehca_err(&shca->ib_device, "EHCA ----- error data end "
"----------------------------------------------------");
@@ -152,7 +155,7 @@ int ehca_error_data(struct ehca_shca *shca, void *data,
if (ret == H_R_STATE)
ehca_err(&shca->ib_device,
- "No error data is available: %lx.", resource);
+ "No error data is available: %llx.", resource);
else if (ret == H_SUCCESS) {
int length;
@@ -164,7 +167,7 @@ int ehca_error_data(struct ehca_shca *shca, void *data,
print_error_data(shca, data, rblock, length);
} else
ehca_err(&shca->ib_device,
- "Error data could not be fetched: %lx", resource);
+ "Error data could not be fetched: %llx", resource);
ehca_free_fw_ctrlblock(rblock);
@@ -178,6 +181,10 @@ static void dispatch_qp_event(struct ehca_shca *shca, struct ehca_qp *qp,
{
struct ib_event event;
+ /* PATH_MIG without the QP ever having been armed is false alarm */
+ if (event_type == IB_EVENT_PATH_MIG && !qp->mig_armed)
+ return;
+
event.device = &shca->ib_device;
event.event = event_type;
@@ -204,6 +211,8 @@ static void qp_event_callback(struct ehca_shca *shca, u64 eqe,
read_lock(&ehca_qp_idr_lock);
qp = idr_find(&ehca_qp_idr, token);
+ if (qp)
+ atomic_inc(&qp->nr_events);
read_unlock(&ehca_qp_idr_lock);
if (!qp)
@@ -223,6 +232,8 @@ static void qp_event_callback(struct ehca_shca *shca, u64 eqe,
if (fatal && qp->ext_type == EQPT_SRQBASE)
dispatch_qp_event(shca, qp, IB_EVENT_QP_LAST_WQE_REACHED);
+ if (atomic_dec_and_test(&qp->nr_events))
+ wake_up(&qp->wait_completion);
return;
}
@@ -351,36 +362,48 @@ static void notify_port_conf_change(struct ehca_shca *shca, int port_num)
*old_attr = new_attr;
}
+/* replay modify_qp for sqps -- return 0 if all is well, 1 if AQP1 destroyed */
+static int replay_modify_qp(struct ehca_sport *sport)
+{
+ int aqp1_destroyed;
+ unsigned long flags;
+
+ spin_lock_irqsave(&sport->mod_sqp_lock, flags);
+
+ aqp1_destroyed = !sport->ibqp_sqp[IB_QPT_GSI];
+
+ if (sport->ibqp_sqp[IB_QPT_SMI])
+ ehca_recover_sqp(sport->ibqp_sqp[IB_QPT_SMI]);
+ if (!aqp1_destroyed)
+ ehca_recover_sqp(sport->ibqp_sqp[IB_QPT_GSI]);
+
+ spin_unlock_irqrestore(&sport->mod_sqp_lock, flags);
+
+ return aqp1_destroyed;
+}
+
static void parse_ec(struct ehca_shca *shca, u64 eqe)
{
u8 ec = EHCA_BMASK_GET(NEQE_EVENT_CODE, eqe);
u8 port = EHCA_BMASK_GET(NEQE_PORT_NUMBER, eqe);
u8 spec_event;
struct ehca_sport *sport = &shca->sport[port - 1];
- unsigned long flags;
switch (ec) {
case 0x30: /* port availability change */
if (EHCA_BMASK_GET(NEQE_PORT_AVAILABILITY, eqe)) {
- int suppress_event;
- /* replay modify_qp for sqps */
- spin_lock_irqsave(&sport->mod_sqp_lock, flags);
- suppress_event = !sport->ibqp_sqp[IB_QPT_GSI];
- if (sport->ibqp_sqp[IB_QPT_SMI])
- ehca_recover_sqp(sport->ibqp_sqp[IB_QPT_SMI]);
- if (!suppress_event)
- ehca_recover_sqp(sport->ibqp_sqp[IB_QPT_GSI]);
- spin_unlock_irqrestore(&sport->mod_sqp_lock, flags);
-
- /* AQP1 was destroyed, ignore this event */
- if (suppress_event)
- break;
+ /* only replay modify_qp calls in autodetect mode;
+ * if AQP1 was destroyed, the port is already down
+ * again and we can drop the event.
+ */
+ if (ehca_nr_ports < 0)
+ if (replay_modify_qp(sport))
+ break;
sport->port_state = IB_PORT_ACTIVE;
dispatch_port_event(shca, port, IB_EVENT_PORT_ACTIVE,
"is active");
- ehca_query_sma_attr(shca, port,
- &sport->saved_attr);
+ ehca_query_sma_attr(shca, port, &sport->saved_attr);
} else {
sport->port_state = IB_PORT_DOWN;
dispatch_port_event(shca, port, IB_EVENT_PORT_ERR,
@@ -459,13 +482,13 @@ void ehca_tasklet_neq(unsigned long data)
struct ehca_eqe *eqe;
u64 ret;
- eqe = (struct ehca_eqe *)ehca_poll_eq(shca, &shca->neq);
+ eqe = ehca_poll_eq(shca, &shca->neq);
while (eqe) {
if (!EHCA_BMASK_GET(NEQE_COMPLETION_EVENT, eqe->entry))
parse_ec(shca, eqe->entry);
- eqe = (struct ehca_eqe *)ehca_poll_eq(shca, &shca->neq);
+ eqe = ehca_poll_eq(shca, &shca->neq);
}
ret = hipz_h_reset_event(shca->ipz_hca_handle,
@@ -494,7 +517,7 @@ static inline void process_eqe(struct ehca_shca *shca, struct ehca_eqe *eqe)
struct ehca_cq *cq;
eqe_value = eqe->entry;
- ehca_dbg(&shca->ib_device, "eqe_value=%lx", eqe_value);
+ ehca_dbg(&shca->ib_device, "eqe_value=%llx", eqe_value);
if (EHCA_BMASK_GET(EQE_COMPLETION_EVENT, eqe_value)) {
ehca_dbg(&shca->ib_device, "Got completion event");
token = EHCA_BMASK_GET(EQE_CQ_TOKEN, eqe_value);
@@ -527,12 +550,11 @@ void ehca_process_eq(struct ehca_shca *shca, int is_irq)
{
struct ehca_eq *eq = &shca->eq;
struct ehca_eqe_cache_entry *eqe_cache = eq->eqe_cache;
- u64 eqe_value;
- unsigned long flags;
+ u64 eqe_value, ret;
int eqe_cnt, i;
int eq_empty = 0;
- spin_lock_irqsave(&eq->irq_spinlock, flags);
+ spin_lock(&eq->irq_spinlock);
if (is_irq) {
const int max_query_cnt = 100;
int query_cnt = 0;
@@ -552,8 +574,7 @@ void ehca_process_eq(struct ehca_shca *shca, int is_irq)
eqe_cnt = 0;
do {
u32 token;
- eqe_cache[eqe_cnt].eqe =
- (struct ehca_eqe *)ehca_poll_eq(shca, eq);
+ eqe_cache[eqe_cnt].eqe = ehca_poll_eq(shca, eq);
if (!eqe_cache[eqe_cnt].eqe)
break;
eqe_value = eqe_cache[eqe_cnt].eqe->entry;
@@ -579,8 +600,13 @@ void ehca_process_eq(struct ehca_shca *shca, int is_irq)
ehca_dbg(&shca->ib_device,
"No eqe found for irq event");
goto unlock_irq_spinlock;
- } else if (!is_irq)
+ } else if (!is_irq) {
+ ret = hipz_h_eoi(eq->ist);
+ if (ret != H_SUCCESS)
+ ehca_err(&shca->ib_device,
+ "bad return code EOI -rc = %lld\n", ret);
ehca_dbg(&shca->ib_device, "deadman found %x eqe", eqe_cnt);
+ }
if (unlikely(eqe_cnt == EHCA_EQE_CACHE_SIZE))
ehca_dbg(&shca->ib_device, "too many eqes for one irq event");
/* enable irq for new packets */
@@ -612,14 +638,14 @@ void ehca_process_eq(struct ehca_shca *shca, int is_irq)
goto unlock_irq_spinlock;
do {
struct ehca_eqe *eqe;
- eqe = (struct ehca_eqe *)ehca_poll_eq(shca, &shca->eq);
+ eqe = ehca_poll_eq(shca, &shca->eq);
if (!eqe)
break;
process_eqe(shca, eqe);
} while (1);
unlock_irq_spinlock:
- spin_unlock_irqrestore(&eq->irq_spinlock, flags);
+ spin_unlock(&eq->irq_spinlock);
}
void ehca_tasklet_eq(unsigned long data)
@@ -627,27 +653,30 @@ void ehca_tasklet_eq(unsigned long data)
ehca_process_eq((struct ehca_shca*)data, 1);
}
-static inline int find_next_online_cpu(struct ehca_comp_pool *pool)
+static int find_next_online_cpu(struct ehca_comp_pool *pool)
{
int cpu;
unsigned long flags;
WARN_ON_ONCE(!in_interrupt());
- if (ehca_debug_level)
- ehca_dmp(&cpu_online_map, sizeof(cpumask_t), "");
+ if (ehca_debug_level >= 3)
+ ehca_dmp(cpu_online_mask, cpumask_size(), "");
spin_lock_irqsave(&pool->last_cpu_lock, flags);
- cpu = next_cpu(pool->last_cpu, cpu_online_map);
- if (cpu == NR_CPUS)
- cpu = first_cpu(cpu_online_map);
- pool->last_cpu = cpu;
+ do {
+ cpu = cpumask_next(pool->last_cpu, cpu_online_mask);
+ if (cpu >= nr_cpu_ids)
+ cpu = cpumask_first(cpu_online_mask);
+ pool->last_cpu = cpu;
+ } while (!per_cpu_ptr(pool->cpu_comp_tasks, cpu)->active);
spin_unlock_irqrestore(&pool->last_cpu_lock, flags);
return cpu;
}
static void __queue_comp_task(struct ehca_cq *__cq,
- struct ehca_cpu_comp_task *cct)
+ struct ehca_cpu_comp_task *cct,
+ struct task_struct *thread)
{
unsigned long flags;
@@ -658,7 +687,7 @@ static void __queue_comp_task(struct ehca_cq *__cq,
__cq->nr_callbacks++;
list_add_tail(&__cq->entry, &cct->cq_list);
cct->cq_jobs++;
- wake_up(&cct->wait_queue);
+ wake_up_process(thread);
} else
__cq->nr_callbacks++;
@@ -670,6 +699,7 @@ static void queue_comp_task(struct ehca_cq *__cq)
{
int cpu_id;
struct ehca_cpu_comp_task *cct;
+ struct task_struct *thread;
int cq_jobs;
unsigned long flags;
@@ -677,7 +707,8 @@ static void queue_comp_task(struct ehca_cq *__cq)
BUG_ON(!cpu_online(cpu_id));
cct = per_cpu_ptr(pool->cpu_comp_tasks, cpu_id);
- BUG_ON(!cct);
+ thread = *per_cpu_ptr(pool->cpu_comp_threads, cpu_id);
+ BUG_ON(!cct || !thread);
spin_lock_irqsave(&cct->task_lock, flags);
cq_jobs = cct->cq_jobs;
@@ -685,28 +716,25 @@ static void queue_comp_task(struct ehca_cq *__cq)
if (cq_jobs > 0) {
cpu_id = find_next_online_cpu(pool);
cct = per_cpu_ptr(pool->cpu_comp_tasks, cpu_id);
- BUG_ON(!cct);
+ thread = *per_cpu_ptr(pool->cpu_comp_threads, cpu_id);
+ BUG_ON(!cct || !thread);
}
-
- __queue_comp_task(__cq, cct);
+ __queue_comp_task(__cq, cct, thread);
}
static void run_comp_task(struct ehca_cpu_comp_task *cct)
{
struct ehca_cq *cq;
- unsigned long flags;
-
- spin_lock_irqsave(&cct->task_lock, flags);
while (!list_empty(&cct->cq_list)) {
cq = list_entry(cct->cq_list.next, struct ehca_cq, entry);
- spin_unlock_irqrestore(&cct->task_lock, flags);
+ spin_unlock_irq(&cct->task_lock);
comp_event_callback(cq);
if (atomic_dec_and_test(&cq->nr_events))
wake_up(&cq->wait_completion);
- spin_lock_irqsave(&cct->task_lock, flags);
+ spin_lock_irq(&cct->task_lock);
spin_lock(&cq->task_lock);
cq->nr_callbacks--;
if (!cq->nr_callbacks) {
@@ -715,159 +743,76 @@ static void run_comp_task(struct ehca_cpu_comp_task *cct)
}
spin_unlock(&cq->task_lock);
}
-
- spin_unlock_irqrestore(&cct->task_lock, flags);
}
-static int comp_task(void *__cct)
+static void comp_task_park(unsigned int cpu)
{
- struct ehca_cpu_comp_task *cct = __cct;
- int cql_empty;
- DECLARE_WAITQUEUE(wait, current);
-
- set_current_state(TASK_INTERRUPTIBLE);
- while (!kthread_should_stop()) {
- add_wait_queue(&cct->wait_queue, &wait);
-
- spin_lock_irq(&cct->task_lock);
- cql_empty = list_empty(&cct->cq_list);
- spin_unlock_irq(&cct->task_lock);
- if (cql_empty)
- schedule();
- else
- __set_current_state(TASK_RUNNING);
-
- remove_wait_queue(&cct->wait_queue, &wait);
+ struct ehca_cpu_comp_task *cct = per_cpu_ptr(pool->cpu_comp_tasks, cpu);
+ struct ehca_cpu_comp_task *target;
+ struct task_struct *thread;
+ struct ehca_cq *cq, *tmp;
+ LIST_HEAD(list);
- spin_lock_irq(&cct->task_lock);
- cql_empty = list_empty(&cct->cq_list);
- spin_unlock_irq(&cct->task_lock);
- if (!cql_empty)
- run_comp_task(__cct);
+ spin_lock_irq(&cct->task_lock);
+ cct->cq_jobs = 0;
+ cct->active = 0;
+ list_splice_init(&cct->cq_list, &list);
+ spin_unlock_irq(&cct->task_lock);
- set_current_state(TASK_INTERRUPTIBLE);
+ cpu = find_next_online_cpu(pool);
+ target = per_cpu_ptr(pool->cpu_comp_tasks, cpu);
+ thread = *per_cpu_ptr(pool->cpu_comp_threads, cpu);
+ spin_lock_irq(&target->task_lock);
+ list_for_each_entry_safe(cq, tmp, &list, entry) {
+ list_del(&cq->entry);
+ __queue_comp_task(cq, target, thread);
}
- __set_current_state(TASK_RUNNING);
-
- return 0;
-}
-
-static struct task_struct *create_comp_task(struct ehca_comp_pool *pool,
- int cpu)
-{
- struct ehca_cpu_comp_task *cct;
-
- cct = per_cpu_ptr(pool->cpu_comp_tasks, cpu);
- spin_lock_init(&cct->task_lock);
- INIT_LIST_HEAD(&cct->cq_list);
- init_waitqueue_head(&cct->wait_queue);
- cct->task = kthread_create(comp_task, cct, "ehca_comp/%d", cpu);
-
- return cct->task;
+ spin_unlock_irq(&target->task_lock);
}
-static void destroy_comp_task(struct ehca_comp_pool *pool,
- int cpu)
+static void comp_task_stop(unsigned int cpu, bool online)
{
- struct ehca_cpu_comp_task *cct;
- struct task_struct *task;
- unsigned long flags_cct;
-
- cct = per_cpu_ptr(pool->cpu_comp_tasks, cpu);
-
- spin_lock_irqsave(&cct->task_lock, flags_cct);
+ struct ehca_cpu_comp_task *cct = per_cpu_ptr(pool->cpu_comp_tasks, cpu);
- task = cct->task;
- cct->task = NULL;
+ spin_lock_irq(&cct->task_lock);
cct->cq_jobs = 0;
-
- spin_unlock_irqrestore(&cct->task_lock, flags_cct);
-
- if (task)
- kthread_stop(task);
+ cct->active = 0;
+ WARN_ON(!list_empty(&cct->cq_list));
+ spin_unlock_irq(&cct->task_lock);
}
-static void __cpuinit take_over_work(struct ehca_comp_pool *pool, int cpu)
+static int comp_task_should_run(unsigned int cpu)
{
struct ehca_cpu_comp_task *cct = per_cpu_ptr(pool->cpu_comp_tasks, cpu);
- LIST_HEAD(list);
- struct ehca_cq *cq;
- unsigned long flags_cct;
-
- spin_lock_irqsave(&cct->task_lock, flags_cct);
-
- list_splice_init(&cct->cq_list, &list);
-
- while (!list_empty(&list)) {
- cq = list_entry(cct->cq_list.next, struct ehca_cq, entry);
-
- list_del(&cq->entry);
- __queue_comp_task(cq, per_cpu_ptr(pool->cpu_comp_tasks,
- smp_processor_id()));
- }
-
- spin_unlock_irqrestore(&cct->task_lock, flags_cct);
+ return cct->cq_jobs;
}
-static int __cpuinit comp_pool_callback(struct notifier_block *nfb,
- unsigned long action,
- void *hcpu)
+static void comp_task(unsigned int cpu)
{
- unsigned int cpu = (unsigned long)hcpu;
- struct ehca_cpu_comp_task *cct;
+ struct ehca_cpu_comp_task *cct = this_cpu_ptr(pool->cpu_comp_tasks);
+ int cql_empty;
- switch (action) {
- case CPU_UP_PREPARE:
- case CPU_UP_PREPARE_FROZEN:
- ehca_gen_dbg("CPU: %x (CPU_PREPARE)", cpu);
- if (!create_comp_task(pool, cpu)) {
- ehca_gen_err("Can't create comp_task for cpu: %x", cpu);
- return NOTIFY_BAD;
- }
- break;
- case CPU_UP_CANCELED:
- case CPU_UP_CANCELED_FROZEN:
- ehca_gen_dbg("CPU: %x (CPU_CANCELED)", cpu);
- cct = per_cpu_ptr(pool->cpu_comp_tasks, cpu);
- kthread_bind(cct->task, any_online_cpu(cpu_online_map));
- destroy_comp_task(pool, cpu);
- break;
- case CPU_ONLINE:
- case CPU_ONLINE_FROZEN:
- ehca_gen_dbg("CPU: %x (CPU_ONLINE)", cpu);
- cct = per_cpu_ptr(pool->cpu_comp_tasks, cpu);
- kthread_bind(cct->task, cpu);
- wake_up_process(cct->task);
- break;
- case CPU_DOWN_PREPARE:
- case CPU_DOWN_PREPARE_FROZEN:
- ehca_gen_dbg("CPU: %x (CPU_DOWN_PREPARE)", cpu);
- break;
- case CPU_DOWN_FAILED:
- case CPU_DOWN_FAILED_FROZEN:
- ehca_gen_dbg("CPU: %x (CPU_DOWN_FAILED)", cpu);
- break;
- case CPU_DEAD:
- case CPU_DEAD_FROZEN:
- ehca_gen_dbg("CPU: %x (CPU_DEAD)", cpu);
- destroy_comp_task(pool, cpu);
- take_over_work(pool, cpu);
- break;
+ spin_lock_irq(&cct->task_lock);
+ cql_empty = list_empty(&cct->cq_list);
+ if (!cql_empty) {
+ __set_current_state(TASK_RUNNING);
+ run_comp_task(cct);
}
-
- return NOTIFY_OK;
+ spin_unlock_irq(&cct->task_lock);
}
-static struct notifier_block comp_pool_callback_nb __cpuinitdata = {
- .notifier_call = comp_pool_callback,
- .priority = 0,
+static struct smp_hotplug_thread comp_pool_threads = {
+ .thread_should_run = comp_task_should_run,
+ .thread_fn = comp_task,
+ .thread_comm = "ehca_comp/%u",
+ .cleanup = comp_task_stop,
+ .park = comp_task_park,
};
int ehca_create_comp_pool(void)
{
- int cpu;
- struct task_struct *task;
+ int cpu, ret = -ENOMEM;
if (!ehca_scaling_code)
return 0;
@@ -877,42 +822,49 @@ int ehca_create_comp_pool(void)
return -ENOMEM;
spin_lock_init(&pool->last_cpu_lock);
- pool->last_cpu = any_online_cpu(cpu_online_map);
+ pool->last_cpu = cpumask_any(cpu_online_mask);
pool->cpu_comp_tasks = alloc_percpu(struct ehca_cpu_comp_task);
- if (pool->cpu_comp_tasks == NULL) {
- kfree(pool);
- return -EINVAL;
- }
+ if (!pool->cpu_comp_tasks)
+ goto out_pool;
- for_each_online_cpu(cpu) {
- task = create_comp_task(pool, cpu);
- if (task) {
- kthread_bind(task, cpu);
- wake_up_process(task);
- }
+ pool->cpu_comp_threads = alloc_percpu(struct task_struct *);
+ if (!pool->cpu_comp_threads)
+ goto out_tasks;
+
+ for_each_present_cpu(cpu) {
+ struct ehca_cpu_comp_task *cct;
+
+ cct = per_cpu_ptr(pool->cpu_comp_tasks, cpu);
+ spin_lock_init(&cct->task_lock);
+ INIT_LIST_HEAD(&cct->cq_list);
}
- register_hotcpu_notifier(&comp_pool_callback_nb);
+ comp_pool_threads.store = pool->cpu_comp_threads;
+ ret = smpboot_register_percpu_thread(&comp_pool_threads);
+ if (ret)
+ goto out_threads;
- printk(KERN_INFO "eHCA scaling code enabled\n");
+ pr_info("eHCA scaling code enabled\n");
+ return ret;
- return 0;
+out_threads:
+ free_percpu(pool->cpu_comp_threads);
+out_tasks:
+ free_percpu(pool->cpu_comp_tasks);
+out_pool:
+ kfree(pool);
+ return ret;
}
void ehca_destroy_comp_pool(void)
{
- int i;
-
if (!ehca_scaling_code)
return;
- unregister_hotcpu_notifier(&comp_pool_callback_nb);
+ smpboot_unregister_percpu_thread(&comp_pool_threads);
- for (i = 0; i < NR_CPUS; i++) {
- if (cpu_online(i))
- destroy_comp_task(pool, i);
- }
+ free_percpu(pool->cpu_comp_threads);
free_percpu(pool->cpu_comp_tasks);
kfree(pool);
}