Hi Bryan,
On 11/4/2026 9:53 PM, Bryan O'Donoghue wrote:
On 03/11/2026 15:29, Vikash Garodia wrote:
diff --git a/drivers/iommu/of_iommu.c b/drivers/iommu/of_iommu.c
index
5c266511db78910d6fd221af5437cecff4931c22..0428ceb50dec58f11872b4fb64e283fa8d9046a3 100644
--- a/drivers/iommu/of_iommu.c
+++ b/drivers/iommu/of_iommu.c
@@ -185,11 +185,118 @@ iommu_resv_region_get_type(struct device *dev,
if (start == phys->start && end == phys->end)
return IOMMU_RESV_DIRECT;
- dev_warn(dev, "treating non-direct mapping [%pr] -> [%pap-%pap]
as reservation\n", phys,
- &start, &end);
+ dev_err(dev, "non-direct mapping [%pr] -> [%pap-%pap] is not
supported\n", phys,
+ &start, &end);
return IOMMU_RESV_RESERVED;
}
Same feedback as @ the last version of this.
I still think this behaviour change should have its own patch that
preceds this patch so that the logical change is clearly explained and
understood as a standalone change.
Supporting iommu-ranges != not mapping misaligned ranges.
Otherwise this change goes under the radar of the new functionality
addition.
Ack. Will push this out as a separate patch.
Regards,
Vikash
---
bod
