Each IRQ source of liointc may be routed to different IRQ source of
cpuintc, for implementing this, bit mapping between liointc and cpuintc
is required, and the mapping information is used for liointc IRQ routing
register setting. It seems that interrupt-map can not pass the mapping
to driver in current driver/of code, so the mapping is used in a
non-standard way(of cause, underscore style may be changed in dts and
driver).
IMO, hardcode routing completely in driver is not flexible and not
recommended, and if possible, keep current map unchanged please.
Jianmin
Thanks.
On 2023/8/31 上午9:47, Binbin Zhou wrote:
cc Jianmin Lv.
Hi all:
Jianmin knows Loongson interrupt controllers well, he may have other
suggestions.
Thanks.
Binbin
On Wed, Aug 30, 2023 at 11:31 PM Jiaxun Yang <jiaxun.yang@xxxxxxxxxxx> wrote:
在 2023/8/30 22:35, Krzysztof Kozlowski 写道:
What's the best way, in your opinion, to overhaul this property? As we don't
really care backward compatibility of DTBs on those systems we can just
redesign it.
Deprecate the property in the bindings, allow driver to work with or
without it and finally drop it entirely from DTS.
I'd love to have such configuration flexibility so I'd be sad to see it go.
+ Huacai and Binbin, what's your opinion?
If dropping such functionality in kernel is a must go, we can hardcode
to route all downstream interrupt to the first pin that passed to DT.
Thanks
- Jiaxun
Best regards,
Krzysztof
