2021-07-12 00:15:55

by Dmitry Osipenko

[permalink] [raw]
Subject: [PATCH v1] iommu/tegra-smmu: Change debugfs directory name

Change debugfs directory name to "smmu", which is a much more obvious name
than the generic name of the memory controller device-tree node.

Signed-off-by: Dmitry Osipenko <[email protected]>
---
drivers/iommu/tegra-smmu.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/iommu/tegra-smmu.c b/drivers/iommu/tegra-smmu.c
index 0a281833f611..093c270b9245 100644
--- a/drivers/iommu/tegra-smmu.c
+++ b/drivers/iommu/tegra-smmu.c
@@ -1141,7 +1141,7 @@ struct tegra_smmu *tegra_smmu_probe(struct device *dev,

tegra_smmu_ahb_enable();

- err = iommu_device_sysfs_add(&smmu->iommu, dev, NULL, dev_name(dev));
+ err = iommu_device_sysfs_add(&smmu->iommu, dev, NULL, "smmu");
if (err)
return ERR_PTR(err);

--
2.32.0


2021-07-14 10:53:58

by Joerg Roedel

[permalink] [raw]
Subject: Re: [PATCH v1] iommu/tegra-smmu: Change debugfs directory name

On Mon, Jul 12, 2021 at 03:13:41AM +0300, Dmitry Osipenko wrote:
> - err = iommu_device_sysfs_add(&smmu->iommu, dev, NULL, dev_name(dev));
> + err = iommu_device_sysfs_add(&smmu->iommu, dev, NULL, "smmu");

Are you sure no one is relying on the old name so that this change
breaks ABI? Also this change means that there is always be only one SMMU
per system. Is that guaranteed too?

Regards,

Joerg

2021-07-14 11:21:08

by Dmitry Osipenko

[permalink] [raw]
Subject: Re: [PATCH v1] iommu/tegra-smmu: Change debugfs directory name

14.07.2021 13:52, Joerg Roedel пишет:
> On Mon, Jul 12, 2021 at 03:13:41AM +0300, Dmitry Osipenko wrote:
>> - err = iommu_device_sysfs_add(&smmu->iommu, dev, NULL, dev_name(dev));
>> + err = iommu_device_sysfs_add(&smmu->iommu, dev, NULL, "smmu");
>
> Are you sure no one is relying on the old name so that this change
> breaks ABI?

IIUC, Thierry and Jon have a testing suite that uses the old name, but
it shouldn't be a problem to support the new name in addition to the old
name since it's internal/private testing suite.

The reason I'm proposing this change is because it's not obvious where
the SMMU debug is located when you look at the debugfs directory, it
takes some effort to find it.

> Also this change means that there is always be only one SMMU
> per system. Is that guaranteed too?

A single SMMU is guaranteed here. Only latest Tegra SoCs have two SMMUs
and those are ARM SMMUs, while this is a legacy Tegra SMMU here.