[PATCH 01/14] NVMe: Fix setup of affinity hint for unallocated queues

From: Alexander Gordeev
Date: Tue Jan 28 2014 - 03:37:58 EST


IRQ affinity hints are attempted to setup for some or all
queues which have not yet been allocated: either on device
probe or resume. This update moves the setup after all
queues are successfully created.

Signed-off-by: Alexander Gordeev <agordeev@xxxxxxxxxx>
---
drivers/block/nvme-core.c | 12 ++++++------
1 files changed, 6 insertions(+), 6 deletions(-)

diff --git a/drivers/block/nvme-core.c b/drivers/block/nvme-core.c
index b59a93a..3dfb0d4 100644
--- a/drivers/block/nvme-core.c
+++ b/drivers/block/nvme-core.c
@@ -1956,12 +1956,6 @@ static int nvme_setup_io_queues(struct nvme_dev *dev)
}
spin_unlock(&dev_list_lock);

- cpu = cpumask_first(cpu_online_mask);
- for (i = 0; i < nr_io_queues; i++) {
- irq_set_affinity_hint(dev->entry[i].vector, get_cpu_mask(cpu));
- cpu = cpumask_next(cpu, cpu_online_mask);
- }
-
q_depth = min_t(int, NVME_CAP_MQES(readq(&dev->bar->cap)) + 1,
NVME_Q_DEPTH);
for (i = dev->queue_count - 1; i < nr_io_queues; i++) {
@@ -1986,6 +1980,12 @@ static int nvme_setup_io_queues(struct nvme_dev *dev)
}
}

+ cpu = cpumask_first(cpu_online_mask);
+ for (i = 0; i < nr_io_queues; i++) {
+ irq_set_affinity_hint(dev->entry[i].vector, get_cpu_mask(cpu));
+ cpu = cpumask_next(cpu, cpu_online_mask);
+ }
+
return 0;

free_queues:
--
1.7.7.6

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/