Messages in this thread | | | From | Hiroshi Doyu <> | Date | Tue, 24 Apr 2012 10:59:48 +0200 | Subject | Re: [PATCH 2/3] iommu/tegra: smmu: Add device tree support for SMMU |
| |
From: Stephen Warren <swarren@wwwdotorg.org> Subject: Re: [PATCH 2/3] iommu/tegra: smmu: Add device tree support for SMMU Date: Mon, 23 Apr 2012 20:27:28 +0200 Message-ID: <4F959F10.9010908@wwwdotorg.org>
> On 04/23/2012 05:58 AM, Hiroshi DOYU wrote: > > Add device tree support for Tegra30 IOMMU(SMMU). > > > diff --git a/drivers/iommu/tegra-smmu.c b/drivers/iommu/tegra-smmu.c > > > + err = of_parse_dma_window(dev->of_node, "dma-window", 0, NULL, > > + &base, &size); > > This patch also depends on "dt: Add general DMA window parser" that you > sent earlier. > > > + asids = be32_to_cpup(of_get_property(dev->of_node, "asids", NULL)); > > + if (!asids) > > + return -ENODEV; > > What if of_get_property() fails?
Ok, need to be checked. I'll fix.
> BTW, does the # ASIDs vary? I wonder if it's worth representing it in > the device tree or not. If this driver has a chance of working > unmodified on some future chip just by updating this DT property without > any code changes, then it seems reasonable to have it in DT.
Yes, ASID should be passed from DT.
| |