[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <ZOSuz+16JD5E0k3z@nvidia.com>
Date: Tue, 22 Aug 2023 09:49:19 -0300
From: Jason Gunthorpe <jgg@...dia.com>
To: Michael Shavit <mshavit@...gle.com>
Cc: iommu@...ts.linux.dev, linux-arm-kernel@...ts.infradead.org,
linux-kernel@...r.kernel.org, nicolinc@...dia.com,
tina.zhang@...el.com, jean-philippe@...aro.org, will@...nel.org,
robin.murphy@....com, Dawei Li <set_pte_at@...look.com>,
Joerg Roedel <joro@...tes.org>,
"Kirill A. Shutemov" <kirill.shutemov@...ux.intel.com>,
Lu Baolu <baolu.lu@...ux.intel.com>,
Mark Brown <broonie@...nel.org>
Subject: Re: [RFC PATCH v2 1/9] iommu/arm-smmu-v3: group attached devices by
smmu
On Tue, Aug 22, 2023 at 06:56:57PM +0800, Michael Shavit wrote:
> Always insert a new master in the devices_list besides other masters
> that belong to the same smmu.
> This allows code to batch commands by SMMU when iterating over masters
> that a domain is attached to.
>
> Signed-off-by: Michael Shavit <mshavit@...gle.com>
> ---
>
> Changes in v2:
> - New commit
>
> drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.c | 22 ++++++++++++++++++---
> 1 file changed, 19 insertions(+), 3 deletions(-)
>
> diff --git a/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.c b/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.c
> index f17704c35858d..37b9223c145ba 100644
> --- a/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.c
> +++ b/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.c
> @@ -2382,6 +2382,24 @@ static void arm_smmu_detach_dev(struct arm_smmu_master *master)
> arm_smmu_write_ctx_desc(master, 0, NULL);
> }
>
> +static void arm_smmu_domain_device_list_add(struct arm_smmu_domain *smmu_domain,
> + struct arm_smmu_master *master)
> +{
> + struct arm_smmu_master *iter;
> + unsigned long flags;
> +
> + spin_lock_irqsave(&smmu_domain->devices_lock, flags);
> + if (list_empty(&smmu_domain->devices))
> + list_add(&master->domain_head, &smmu_domain->devices);
> + else {
> + list_for_each_entry(iter, &smmu_domain->devices, domain_head)
> + if (iter->smmu == master->smmu)
> + break;
> + list_add(&master->domain_head, &iter->domain_head);
> + }
IIRC you are not supposed to touch iter after the list_for_each. Like this:
list_for_each_entry(iter, &smmu_domain->devices, domain_head) {
if (iter->smmu == master->smmu) {
list_add(&master->domain_head, iter->domain_head);
goto out;
}
}
list_add(&master->domain_head, &smmu_domain->devices);
out:
spin_unlock_irqrestore(&smmu_domain->devices_lock, flags);
Jason
Powered by blists - more mailing lists