Commit 972c7759 authored Jul 27, 2021 by Zhen Lei Committed by Zheng Zengkai Jul 28, 2021

iommu/arm-smmu-v3: Use command queue batching helpers to improve performance



hulk inclusion
category: feature
bugzilla: 174251
CVE: NA

-------------------------------------------------------------------------

The obvious key to the performance optimization of commit 587e6c10
("iommu/arm-smmu-v3: Reduce contention during command-queue insertion") is
to allow multiple cores to insert commands in parallel after a brief mutex
contention.

Obviously, inserting as many commands at a time as possible can reduce the
number of times the mutex contention participates, thereby improving the
overall performance. At least it reduces the number of calls to function
arm_smmu_cmdq_issue_cmdlist().

Therefore, use command queue batching helpers to insert multiple commands
at a time.

Signed-off-by: Zhen Lei <thunder.leizhen@huawei.com>
Reviewed-by: Hanjun Guo <guohanjun@huawei.com>
Signed-off-by: Zheng Zengkai <zhengzengkai@huawei.com>

parent 4ba103df

drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.c

+3 −2

Original line number	Diff line number	Diff line
		@@ -1964,15 +1964,16 @@ static int arm_smmu_atc_inv_master(struct arm_smmu_master *master, unsigned int
		{
		int i;
		struct arm_smmu_cmdq_ent cmd;
		struct arm_smmu_cmdq_batch cmds = {};

		arm_smmu_atc_inv_to_cmd(ssid, 0, 0, &cmd);

		for (i = 0; i < master->num_streams; i++) {
		cmd.atc.sid = master->streams[i].id;
		arm_smmu_cmdq_issue_cmd(master->smmu, &cmd);
		arm_smmu_cmdq_batch_add(master->smmu, &cmds, &cmd);
		}

		return arm_smmu_cmdq_issue_sync(master->smmu);
		return arm_smmu_cmdq_batch_submit(master->smmu, &cmds);
		}

		int arm_smmu_atc_inv_domain(struct arm_smmu_domain *smmu_domain, int ssid,