Will Deacon <will.deacon@xxxxxxx> writes:
> [adding Rob, Mark, Arnd, Thierry and Hiroshi]
>
> On Wed, Aug 13, 2014 at 01:51:37AM +0100, Mitchel Humpherys wrote:
>> Generic IOMMU device tree bindings were recently added in
>> ["devicetree: Add generic IOMMU device tree bindings"]. Implement the
>> bindings in the ARM SMMU driver.
>>
>> See Documentation/devicetree/bindings/iommu/iommu.txt for the bindings
>> themselves.
>
> Could you look at moving the parsing code into a separate file please (maybe
> under lib/ ?). That way, other IOMMUs can use the same binding without the
> boilerplate having to be rewritten each time.
>
>> diff --git a/drivers/iommu/arm-smmu.c b/drivers/iommu/arm-smmu.c
>> index 63c6707fad..22e25f3172 100644
>> --- a/drivers/iommu/arm-smmu.c
>> +++ b/drivers/iommu/arm-smmu.c
>> @@ -538,25 +538,32 @@ static int insert_smmu_master(struct arm_smmu_device *smmu,
>> return 0;
>> }
>>
>> +struct iommus_entry {
>> + struct list_head list;
>> + struct device_node *node;
>> + u16 streamids[MAX_MASTER_STREAMIDS];
>> + int num_sids;
>> +};
>> +
>> static int register_smmu_master(struct arm_smmu_device *smmu,
>> - struct device *dev,
>> - struct of_phandle_args *masterspec)
>> + struct iommus_entry *entry)
>> {
>> int i;
>> struct arm_smmu_master *master;
>> + struct device *dev = smmu->dev;
>>
>> - master = find_smmu_master(smmu, masterspec->np);
>> + master = find_smmu_master(smmu, entry->node);
>> if (master) {
>> dev_err(dev,
>> "rejecting multiple registrations for master device %s\n",
>> - masterspec->np->name);
>> + entry->node->name);
>> return -EBUSY;
>> }
>>
>> - if (masterspec->args_count > MAX_MASTER_STREAMIDS) {
>> + if (entry->num_sids > MAX_MASTER_STREAMIDS) {
>> dev_err(dev,
>> "reached maximum number (%d) of stream IDs for master device %s\n",
>> - MAX_MASTER_STREAMIDS, masterspec->np->name);
>> + MAX_MASTER_STREAMIDS, entry->node->name);
>> return -ENOSPC;
>> }
>>
>> @@ -564,15 +571,58 @@ static int register_smmu_master(struct arm_smmu_device *smmu,
>> if (!master)
>> return -ENOMEM;
>>
>> - master->of_node = masterspec->np;
>> - master->cfg.num_streamids = masterspec->args_count;
>> + master->of_node = entry->node;
>> + master->cfg.num_streamids = entry->num_sids;
>>
>> for (i = 0; i < master->cfg.num_streamids; ++i)
>> - master->cfg.streamids[i] = masterspec->args[i];
>> + master->cfg.streamids[i] = entry->streamids[i];
>>
>> return insert_smmu_master(smmu, master);
>> }
>>
>> +static int arm_smmu_parse_iommus_properties(struct arm_smmu_device *smmu,
>> + int *num_masters)
>> +{
>> + struct of_phandle_args iommuspec;
>> + struct device_node *dn;
>> +
>> + for_each_node_with_property(dn, "iommus") {
>> + int arg_ind = 0;
>> + struct iommus_entry *entry, *n;
>> + LIST_HEAD(iommus);
You may want to use the macro, "of_property_for_each_phandle_with_args()"
to parse "iommus=".
https://patchwork.ozlabs.org/patch/354073/
>> +
>> + while (!of_parse_phandle_with_args(dn, "iommus", "#iommu-cells",
>> + arg_ind, &iommuspec)) {
>
> We need to check that the phandle does indeed point at one of our SMMUs
> here, in case we have a system with multiple IOMMU types, all using the
> generic binding.
>
>> + int i;
>> +
>> + list_for_each_entry(entry, &iommus, list)
>> + if (entry->node == dn)
>> + break;
>
> Oh, yuck, this is really nasty to parse...
>
>> + if (&entry->list == &iommus) {
>
> Where is entry initialised the first time around?
>
>> + entry = devm_kzalloc(smmu->dev, sizeof(*entry),
>> + GFP_KERNEL);
>> + if (!entry)
>> + return -ENOMEM;
>
> You need to of_node_put the guys you've got back from the phandle parsing
> code.
>
>> + entry->node = dn;
>> + list_add(&entry->list, &iommus);
>> + }
>> + entry->num_sids = iommuspec.args_count;
>> + for (i = 0; i < entry->num_sids; ++i)
>> + entry->streamids[i] = iommuspec.args[i];
>> + arg_ind++;
>
> Isn't this defined by #iommu-cells?
>
>> + }
>> +
>> + list_for_each_entry_safe(entry, n, &iommus, list) {
>
> Why the _safe variant? This is a local list, right?
>
> Will
--
To unsubscribe from this list: send the line "unsubscribe devicetree" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
