Unverified Commit 7e550774 authored by openeuler-ci-bot's avatar openeuler-ci-bot Committed by Gitee
Browse files

!9572 dma-mapping: benchmark: handle NUMA_NO_NODE correctly 

Merge Pull Request from: @ci-robot 
 
PR sync from: Liu Shixin <liushixin2@huawei.com>
https://mailweb.openeuler.org/hyperkitty/list/kernel@openeuler.org/message/EXTJAWMZVDF3CWKA27R3J27INN2XFTXI/ 
 
https://gitee.com/src-openeuler/kernel/issues/IA7DCL 
 
Link:https://gitee.com/openeuler/kernel/pulls/9572

 

Reviewed-by: default avatarKefeng Wang <wangkefeng.wang@huawei.com>
Signed-off-by: default avatarJialin Zhang <zhangjialin11@huawei.com>
parents a095618e 17c50c73

kernel/dma/map_benchmark.c

+1 −2
Original line number Diff line number Diff line
@@ -121,7 +121,6 @@ static int do_map_benchmark(struct map_benchmark_data *map) 
	struct task_struct **tsk;

	int threads = map->bparam.threads;

	int node = map->bparam.node;

	const cpumask_t *cpu_mask = cpumask_of_node(node);

	u64 loops;

	int ret = 0;

	int i;
@@ -142,7 +141,7 @@ static int do_map_benchmark(struct map_benchmark_data *map) 
		}



		if (node != NUMA_NO_NODE)

			kthread_bind_mask(tsk[i], cpu_mask);

			kthread_bind_mask(tsk[i], cpumask_of_node(node));

	}



	/* clear the old value in the previous benchmark */