Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]


Groups > linux.kernel > #1281672 > unrolled thread

[PATCH] blk-mq: Reuse hardware context cpumask for tags

Started byRaghavendra K T <raghavendra.kt@linux.vnet.ibm.com>
First post2015-12-02 12:30 +0100
Last post2015-12-03 18:00 +0100
Articles 3 — 3 participants

Back to article view | Back to linux.kernel


Contents

  [PATCH] blk-mq: Reuse hardware context cpumask for tags Raghavendra K T <raghavendra.kt@linux.vnet.ibm.com> - 2015-12-02 12:30 +0100
    Re: [PATCH] blk-mq: Reuse hardware context cpumask for tags Sagi Grimberg <sagig@dev.mellanox.co.il> - 2015-12-03 11:10 +0100
    Re: [PATCH] blk-mq: Reuse hardware context cpumask for tags Jens Axboe <axboe@kernel.dk> - 2015-12-03 18:00 +0100

#1281672 — [PATCH] blk-mq: Reuse hardware context cpumask for tags

FromRaghavendra K T <raghavendra.kt@linux.vnet.ibm.com>
Date2015-12-02 12:30 +0100
Subject[PATCH] blk-mq: Reuse hardware context cpumask for tags
Message-ID<qBdLK-6vD-57@gated-at.bofh.it>
hctx->cpumask is already populated and let the tag cpumask follow that
instead of going through a new for loop.

Signed-off-by: Raghavendra K T <raghavendra.kt@linux.vnet.ibm.com>
---
 block/blk-mq.c | 9 +--------
 1 file changed, 1 insertion(+), 8 deletions(-)

 Nish had suggested to put cpumask_copy after WARN_ON (instead of before).

diff --git a/block/blk-mq.c b/block/blk-mq.c
index 6d6f8fe..6ada3b4 100644
--- a/block/blk-mq.c
+++ b/block/blk-mq.c
@@ -1854,6 +1854,7 @@ static void blk_mq_map_swqueue(struct request_queue *q,
 		hctx->tags = set->tags[i];
 		WARN_ON(!hctx->tags);
 
+		cpumask_copy(hctx->tags->cpumask, hctx->cpumask);
 		/*
 		 * Set the map size to the number of mapped software queues.
 		 * This is more accurate and more efficient than looping
@@ -1867,14 +1868,6 @@ static void blk_mq_map_swqueue(struct request_queue *q,
 		hctx->next_cpu = cpumask_first(hctx->cpumask);
 		hctx->next_cpu_batch = BLK_MQ_CPU_WORK_BATCH;
 	}
-
-	queue_for_each_ctx(q, ctx, i) {
-		if (!cpumask_test_cpu(i, online_mask))
-			continue;
-
-		hctx = q->mq_ops->map_queue(q, i);
-		cpumask_set_cpu(i, hctx->tags->cpumask);
-	}
 }
 
 static void queue_set_hctx_shared(struct request_queue *q, bool shared)
-- 
1.7.11.7

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

[toc] | [next] | [standalone]


#1282892

FromSagi Grimberg <sagig@dev.mellanox.co.il>
Date2015-12-03 11:10 +0100
Message-ID<qByZP-3zg-9@gated-at.bofh.it>
In reply to#1281672
Looks good,

Reviewed-by: Sagi Grimberg <sagig@mellanox.com>
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

[toc] | [prev] | [next] | [standalone]


#1283194

FromJens Axboe <axboe@kernel.dk>
Date2015-12-03 18:00 +0100
Message-ID<qBFoD-7HE-25@gated-at.bofh.it>
In reply to#1281672
On 12/02/2015 04:27 AM, Raghavendra K T wrote:
> hctx->cpumask is already populated and let the tag cpumask follow that
> instead of going through a new for loop.

Applied for 4.5, thanks.

-- 
Jens Axboe

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

[toc] | [prev] | [standalone]


Back to top | Article view | linux.kernel


csiph-web