[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20170628095700.GE14532@8bytes.org>
Date: Wed, 28 Jun 2017 11:57:00 +0200
From: Joerg Roedel <joro@...tes.org>
To: Jacob Pan <jacob.jun.pan@...ux.intel.com>
Cc: iommu@...ts.linux-foundation.org,
LKML <linux-kernel@...r.kernel.org>,
David Woodhouse <dwmw2@...radead.org>,
"Liu, Yi L" <yi.l.liu@...el.com>,
Lan Tianyu <tianyu.lan@...el.com>,
"Tian, Kevin" <kevin.tian@...el.com>,
Raj Ashok <ashok.raj@...el.com>,
Alex Williamson <alex.williamson@...hat.com>,
Jean Delvare <khali@...ux-fr.org>, Liu@...tes.org,
Yi L <yi.l.liu@...ux.intel.com>
Subject: Re: [PATCH 1/9] iommu: Introduce bind_pasid_table API function
On Tue, Jun 27, 2017 at 12:47:55PM -0700, Jacob Pan wrote:
> +int iommu_bind_pasid_table(struct iommu_domain *domain, struct device *dev,
> + struct pasid_table_info *pasidt_binfo)
> +{
> + if (unlikely(!domain->ops->bind_pasid_table))
> + return -EINVAL;
I think its better to return -ENODEV here, like other iommu-api
functions do when a callback is NULL.
> +enum iommu_model {
> + IOMMU_MODEL_INTEL_VTD,
> + IOMMU_MODEL_ARM_SMMU,
> +};
AMD IOMMU also supports shared virtual memory.
Regards,
Joerg
Powered by blists - more mailing lists