On Tue, 17 Sept 2024 at 22:39, Conor Dooley <conor@xxxxxxxxxx> wrote: > > On Tue, Sep 17, 2024 at 08:39:48PM +0200, Erez wrote: > > On Tue, 17 Sept 2024 at 19:32, Krzysztof Kozlowski <krzk@xxxxxxxxxx> wrote: > > > > > > On 17/09/2024 19:24, Erez wrote: > > > >>>>>> > > > >>>>>> It does not look like you tested the bindings, at least after quick > > > >>>>>> look. Please run `make dt_binding_check` (see > > > >>>>> > > > >>>>> I run "make dt_binding_check" on kernel 6.6. > > > >>>> > > > >>>> Yeah, we are no on kernel 6.6. You can run it also on kernel v4.1 - > > > >>>> still does not matter. > > > >>>> > > > >>>> Don't develop on ancient code because then you ask us to review same > > > >>>> broken stuff we already fixed. > > > >>> > > > >>> I test with Beaglebone black for testing, it is difficult to run the > > > >>> last vanille version. > > > >>> I did backport the spi-nor driver. > > > >>> As for "make dt_binding_check" on last kernel, it need to upgrade the tools, > > > >>> and I did not think it could change that much. > > > >>> > > > >> > > > >> Well, it is possible to build kernel on small embedded board, but that's > > > >> quite cumbersone, slow and inefficient, considering that it's just > > > >> easier to cross compile. But anyway, binding check does not even need > > > >> cross compilation. > > > >> > > > >> Sorry, the code is obviously wrong, there is no such thing as u32, so > > > >> you did not test it. I provided link which explains how to test it. You > > > >> must do it on latest mainline kernel. Just like you must develop and > > > >> generate patches on latest mainline kernel, because this is where we > > > >> apply the patches. We do not apply them to v6.6. > > > > > > > > The patches are based on the lastest mainline kernel. > > > > I do not understand why you think otherwise. > > > > > > Because you wrote: > > > "I run "make dt_binding_check" on kernel 6.6." > > > > > > The command is either part of build process or final check process > > > (static analyzers etc). If you say you did this on v6.6, you got such > > > response. > > > > I know you are NOT a service. > > The device tree is not the focus of my work. > > It should not be broken like that. > > > > I install dt-schema_2023.11-3_all.deb > > with Debian trixie > > I get: > > > > l > > SCHEMA Documentation/devicetree/bindings/processed-schema.json > > Traceback (most recent call last): > > File "/usr/bin/dt-mk-schema", line 8, in <module> > > sys.exit(main()) > > ^^^^^^ > > File "/usr/lib/python3/dist-packages/dtschema/mk_schema.py", line 28, in main > > schemas = dtschema.DTValidator(args.schemas).schemas > > ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ > > File "/usr/lib/python3/dist-packages/dtschema/validator.py", line > > 363, in __init__ > > self.make_property_type_cache() > > File "/usr/lib/python3/dist-packages/dtschema/validator.py", line > > 420, in make_property_type_cache > > self.props, self.pat_props = get_prop_types(self.schemas) > > ^^^^^^^^^^^^^^^^^^^^^^^^^^^^ > > File "/usr/lib/python3/dist-packages/dtschema/validator.py", line > > 187, in get_prop_types > > del props[r'^[a-z][a-z0-9\-]*$'] > > ~~~~~^^^^^^^^^^^^^^^^^^^^^^^ > > KeyError: '^[a-z][a-z0-9\\-]*$' > > make[2]: *** [Documentation/devicetree/bindings/Makefile:64: > > Documentation/devicetree/bindings/processed-schema.json] Error 1 > > make[2]: *** Deleting file > > 'Documentation/devicetree/bindings/processed-schema.json' > > make[1]: *** [/home/builder/kernel/Makefile:1435: dt_binding_schemas] Error 2 > > Have you considered that this might be because of the invalid types you > used? I remove the types. Anyway, scripts should report on errors, not crash. This is the purpose of syntax scripts, to help us, developers find out errors. Erez :-)
