Messages in this thread | | | From | Andy Shevchenko <> | Date | Mon, 22 Apr 2024 08:00:17 +0300 | Subject | Re: [PATCH 1/4] iommu: Resolve fwspec ops automatically |
| |
Fri, Apr 19, 2024 at 05:55:59PM +0100, Robin Murphy kirjoitti: > There's no real need for callers to resolve ops from a fwnode in order > to then pass both to iommu_fwspec_init() - it's simpler and more sensible > for that to resolve the ops itself. This in turn means we can centralise > the notion of checking for a present driver, and enforce that fwspecs > aren't allocated unless and until we know they will be usable. > > Also we've grown a generic fwnode_handle_get() since this code was first > written, so may as well clear up that ugly mismatch while we're in here.
..
> +++ b/drivers/iommu/mtk_iommu_v1.c
> if (!fwspec) { > - ret = iommu_fwspec_init(dev, &args->np->fwnode, &mtk_iommu_v1_ops); > + ret = iommu_fwspec_init(dev, &args->np->fwnode);
I'm wondering, while at it, if can avoid direct dereference of fwnode by using of_fwnode_handle().
> if (ret) > return ret;
..
> +++ b/drivers/iommu/of_iommu.c
> - ret = iommu_fwspec_init(dev, fwnode, ops); > + ret = iommu_fwspec_init(dev, &iommu_spec->np->fwnode);
Ditto.
> + if (ret == -EPROBE_DEFER) > + return driver_deferred_probe_check_state(dev); > if (ret) > return ret;
..
> --- a/drivers/iommu/tegra-smmu.c > +++ b/drivers/iommu/tegra-smmu.c
> - err = iommu_fwspec_init(dev, &dev->of_node->fwnode, ops); > + err = iommu_fwspec_init(dev, &dev->of_node->fwnode);
Ditto.
> if (err < 0) { > dev_err(dev, "failed to initialize fwspec: %d\n", err); > return err;
-- With Best Regards, Andy Shevchenko
| |