On 24/07/2025 09:02, Ivaylo Ivanov wrote: > On 7/24/25 09:56, Krzysztof Kozlowski wrote: >> On 23/07/2025 10:21, Ivaylo Ivanov wrote: >>> On 7/23/25 11:15, Krzysztof Kozlowski wrote: >>>> On Tue, Jul 22, 2025 at 03:10:36PM +0300, Ivaylo Ivanov wrote: >>>>> Some device trees, like the exynos2200 one, configure the root node >>>>> with #address-cells and #size-cells set to 2. However, the usi binding >>>>> expects 32 bit address space only. Allow these determining properties to >>>> So if USI expects 32 bit, then why do we allow 64? >>>> >>>> Switching this to 2 means you use 64-bit addressing for children >>> I don't, but the main point was to avoid defining ranges for every single usi >> I do not understand your "I don't", because you do. > > I meant it in the "I don't _need_ to explicitly use that, but it's _nice_ to have" > way, so I don't have to clutter the nodes with address translations in ranges. It is not nice to have. The address space should not grow above the device limits or even above the needs (sometime ago Rob explicitly asked for that). Changing to 64-bit just because you do not want to add ranges property is not correct, because it misses the main point: what is the address space? Changing to 64-bit because that's the address space would be fine, but that was not argued here. I did not check in the datasheets, but I assume these devices want 32-bit address space and that's how it should stay. Best regards, Krzysztof
