[PATCH] blk-mq: check if all HW queues are mapped to cpu

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



From: Ming Lin <ming.l@xxxxxxxxxxxxxxx>

Suppose that a system has 8 logical CPUs(4 cores with hyperthread)
and that 5 hardware queues are provided by a block driver.
With the current algorithm this will lead to the following assignment
of logical CPU to hardware queue:

HWQ 0: 0 1
HWQ 1: 2 3
HWQ 2: 4 5
HWQ 3: 6 7
HWQ 4: (none)

One way to fix it is to change the algorithm so the assignment may be:

HWQ 0: 0 1
HWQ 1: 2 3
HWQ 2: 4 5
HWQ 3: 6
HWQ 4: 7

This patch only checks if all HW queues are mapped to cpu.
blk_mq_init_queue() will fail if not all HW queues are mapped.
And it also fails when hardware queue numbers > cpu numbers.

Signed-off-by: Ming Lin <ming.l@xxxxxxxxxxxxxxxx>
---
 block/blk-mq-cpumap.c | 27 +++++++++++++++++++++++----
 1 file changed, 23 insertions(+), 4 deletions(-)

diff --git a/block/blk-mq-cpumap.c b/block/blk-mq-cpumap.c
index 8764c24..5a88ac4 100644
--- a/block/blk-mq-cpumap.c
+++ b/block/blk-mq-cpumap.c
@@ -36,10 +36,17 @@ int blk_mq_update_queue_map(unsigned int *map, unsigned int nr_queues,
 {
 	unsigned int i, nr_cpus, nr_uniq_cpus, queue, first_sibling;
 	cpumask_var_t cpus;
+	int *queue_map;
 
-	if (!alloc_cpumask_var(&cpus, GFP_ATOMIC))
+	queue_map = kzalloc(sizeof(int) * nr_queues, GFP_KERNEL);
+	if (!queue_map)
 		return 1;
 
+	if (!alloc_cpumask_var(&cpus, GFP_ATOMIC)) {
+		kfree(queue_map);
+		return 1;
+	}
+
 	cpumask_clear(cpus);
 	nr_cpus = nr_uniq_cpus = 0;
 	for_each_cpu(i, online_mask) {
@@ -54,7 +61,7 @@ int blk_mq_update_queue_map(unsigned int *map, unsigned int nr_queues,
 	for_each_possible_cpu(i) {
 		if (!cpumask_test_cpu(i, online_mask)) {
 			map[i] = 0;
-			continue;
+			goto mapped;
 		}
 
 		/*
@@ -65,7 +72,7 @@ int blk_mq_update_queue_map(unsigned int *map, unsigned int nr_queues,
 		if (nr_queues >= nr_cpus || nr_cpus == nr_uniq_cpus) {
 			map[i] = cpu_to_queue_index(nr_cpus, nr_queues, queue);
 			queue++;
-			continue;
+			goto mapped;
 		}
 
 		/*
@@ -80,10 +87,22 @@ int blk_mq_update_queue_map(unsigned int *map, unsigned int nr_queues,
 			queue++;
 		} else
 			map[i] = map[first_sibling];
+
+mapped:
+		/* i is cpu, map[i] is queue index */
+		queue_map[map[i]] = 1;
 	}
 
 	free_cpumask_var(cpus);
-	return 0;
+
+	/* check if all queues are mapped to cpu */
+	for (i = 0; i < nr_queues; i++)
+		if (!queue_map[i])
+			break;
+
+	kfree(queue_map);
+
+	return i != nr_queues;
 }
 
 unsigned int *blk_mq_make_queue_map(struct blk_mq_tag_set *set)
-- 
1.9.1

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



[Index of Archives]     [Linux RAID]     [Linux SCSI]     [Linux ATA RAID]     [IDE]     [Linux Wireless]     [Linux Kernel]     [ATH6KL]     [Linux Bluetooth]     [Linux Netdev]     [Kernel Newbies]     [Security]     [Git]     [Netfilter]     [Bugtraq]     [Yosemite News]     [MIPS Linux]     [ARM Linux]     [Linux Security]     [Device Mapper]

  Powered by Linux