On Tue, Mar 14, 2023 at 08:09:45PM +0300, Serge Semin wrote:
> On Tue, Mar 14, 2023 at 06:07:01PM +0300, Serge Semin wrote:
> > On Tue, Mar 14, 2023 at 09:10:19AM -0500, Rob Herring wrote:
> > >
> > > On Tue, 14 Mar 2023 01:50:48 +0300, Serge Semin wrote:
> > > > Indeed the maximum DMA burst length can be programmed not only for DW
> > > > xGMACs, Allwinner EMACs and Spear SoC GMAC, but in accordance with [1]
> > > > for Generic DW *MAC IP-cores. Moreover the STMMAC set of drivers parse
> > > > the property and then apply the configuration for all supported DW MAC
> > > > devices. All of that makes the property being available for all IP-cores
> > > > the bindings supports. Let's make sure the PBL-related properties are
> > > > validated for all of them by the common DW MAC DT schema.
> > > >
> > > > [1] DesignWare Cores Ethernet MAC Universal Databook, Revision 3.73a,
> > > > October 2013, p. 380.
> > > >
> > > > Signed-off-by: Serge Semin <Sergey.Semin@xxxxxxxxxxxxxxxxxxxx>
> > > > Reviewed-by: Rob Herring <robh@xxxxxxxxxx>
> > > >
> > > > ---
> > > >
> > > > Changelog v1:
> > > > - Use correct syntax of the JSON pointers, so the later would begin
> > > > with a '/' after the '#'.
> > > > ---
> > > > .../devicetree/bindings/net/snps,dwmac.yaml | 77 +++++++------------
> > > > 1 file changed, 26 insertions(+), 51 deletions(-)
> > > >
> > >
> > > My bot found errors running 'make DT_CHECKER_FLAGS=-m dt_binding_check'
> > > on your patch (DT_CHECKER_FLAGS is new in v5.13):
> > >
> > > yamllint warnings/errors:
> > >
> > > dtschema/dtc warnings/errors:
> > > /builds/robherring/dt-review-ci/linux/Documentation/devicetree/bindings/net/mediatek-dwmac.example.dtb: ethernet@1101c000: snps,txpbl:0:0: 1 is not one of [2, 4, 8]
> > > From schema: /builds/robherring/dt-review-ci/linux/Documentation/devicetree/bindings/net/snps,dwmac.yaml
> > > /builds/robherring/dt-review-ci/linux/Documentation/devicetree/bindings/net/mediatek-dwmac.example.dtb: ethernet@1101c000: snps,rxpbl:0:0: 1 is not one of [2, 4, 8]
> > > From schema: /builds/robherring/dt-review-ci/linux/Documentation/devicetree/bindings/net/snps,dwmac.yaml
> > > /builds/robherring/dt-review-ci/linux/Documentation/devicetree/bindings/net/mediatek-dwmac.example.dtb: ethernet@1101c000: snps,txpbl:0:0: 1 is not one of [2, 4, 8]
> > > From schema: /builds/robherring/dt-review-ci/linux/Documentation/devicetree/bindings/net/mediatek-dwmac.yaml
> > > /builds/robherring/dt-review-ci/linux/Documentation/devicetree/bindings/net/mediatek-dwmac.example.dtb: ethernet@1101c000: snps,rxpbl:0:0: 1 is not one of [2, 4, 8]
> > > From schema: /builds/robherring/dt-review-ci/linux/Documentation/devicetree/bindings/net/mediatek-dwmac.yaml
> > > /builds/robherring/dt-review-ci/linux/Documentation/devicetree/bindings/net/mediatek-dwmac.example.dtb: ethernet@1101c000: Unevaluated properties are not allowed ('interrupt-names', 'interrupts', 'mac-address', 'phy-mode', 'reg', 'snps,reset-delays-us', 'snps,reset-gpio', 'snps,rxpbl', 'snps,txpbl' were unexpected)
> > > From schema: /builds/robherring/dt-review-ci/linux/Documentation/devicetree/bindings/net/mediatek-dwmac.yaml
> >
> > Oops, on rebasing my work from older kernel I missed that the PBL
> > properties constraints have already been extended. I'll drop the next
> > patch in the series then and fix this one so the already defined
> > constraints would be preserved.
>
> BTW it's strange I didn't have that bug spotted during my
> dt_binding_check run...
Perhaps because you set DT_SCHEMA_FILES?
