Skip to contentSkip to conversation

iommu/amd: return devid as alias for ACPI HID devices

PATCH In review Arindam Nath 2 participants 4 messages Sep 18, 2018 – Sep 26, 2018 3 messages not in the archive IOMMU

lore
Share & tools
lore
#0 lore raw

PATCH 1/1 iommu/amd: return devid as alias for ACPI HID devices

ACPI HID devices do not actually have an alias for them in the IVRS. But dev_data->alias is still used for indexing into the IOMMU device table for devices being handled by the IOMMU. So for ACPI HID devices, we simply return the corresponding devid as an alias, as parsed from IVRS table.

---
 drivers/iommu/amd_iommu.c | 6 ++++++
 1 file changed, 6 insertions(+)
drivers/iommu/amd_iommu.c +6 -0
index 4e04fff23977..73e47d93e7a0 100644
@@ -246,7 +246,13 @@ static u16 get_alias(struct device *dev)
 
 	/* The callers make sure that get_device_id() does not fail here */
 	devid = get_device_id(dev);
+
+	/* For ACPI HID devices, we simply return the devid as such */
+	if (!dev_is_pci(dev))
+		return devid;
+
 	ivrs_alias = amd_iommu_alias_table[devid];
+
 	pci_for_each_dma_alias(pdev, __last_alias, &pci_alias);
 
 	if (ivrs_alias == pci_alias)
Signature · 1 line
2.17.1

A message that is not in the archive (sent off-list or not yet received): [email protected]

#2 lore raw

Hi Arindam,

drivers/iommu/amd_iommu.c static u16 get_alias(struct device *dev) · line 253

Quoted 9 lines · Arindam Nath
drivers/iommu/amd_iommu.c +5 -0
@@ -246,7 +246,13 @@ static u16 get_alias(struct device *dev)
 
 	/* The callers make sure that get_device_id() does not fail here */
 	devid = get_device_id(dev);
+
+	/* For ACPI HID devices, we simply return the devid as such */
+	if (!dev_is_pci(dev))
+		return devid;
+

Does this fix some bug? If there is no alias defined for the device-id, it should just return the correct devid even without that change.

Regards,

Joerg

A message that is not in the archive (sent off-list or not yet received): [email protected]

#4 lore raw

Hi Joerg,

Quoted 13 lines

-----Original Message-----
From: Joerg Roedel <[email protected]>
Sent: Tuesday, September 25, 2018 7:02 PM
To: Nath, Arindam <[email protected]>
Cc: iommu-cunTk1MwBs9QetFLy7KEm3xJsTq8ys+cHZ5vskTnxNA@public.gmane.org; Suthikulpanit, Suravee <[email protected]>
Subject: Re: [PATCH] iommu/amd: return devid as alias for ACPI HID devices

Hi Arindam,

Quoted 2 lines · Arindam Nath
(quoted diff) +0 -0
@@ -246,7 +246,13 @@ static u16 get_alias(struct device *dev)
 
Quoted 10 lines

drivers/iommu/amd_iommu.c static u16 get_alias(struct device *dev) · line 253

Quoted 7 lines
drivers/iommu/amd_iommu.c +5 -0
 	/* The callers make sure that get_device_id() does not fail here */
 	devid = get_device_id(dev);
+
+	/* For ACPI HID devices, we simply return the devid as such */
+	if (!dev_is_pci(dev))
+		return devid;
+

Does this fix some bug? If there is no alias defined for the device-id, it should just return the correct devid even without that change.

It does fix an issue where eMMC controller is an ACPI HID device and it fails to work with IOMMU enabled.

Since for the ACPI HID, there is no actual pci_dev backing it, there is a NULL pointer de-referencing inside pci_for_each_dma_alias() while trying to access pdev->bus->number.

Thanks,
Arindam

Quoted 4 lines

Regards,

Joerg

A message that is not in the archive (sent off-list or not yet received): BLUPR12MB05166F5680FA08277A8CC4F99C160-7LeqcoF/[email protected]

#6 lore raw
Quoted 6 lines · Arindam

It does fix an issue where eMMC controller is an ACPI HID device and it fails to work with IOMMU enabled.

Since for the ACPI HID, there is no actual pci_dev backing it, there is a NULL pointer de-referencing inside pci_for_each_dma_alias() while trying to access pdev->bus->number.

Okay, added a Fixes: tag and applied for v4.19, thanks Arindam.