David Gibson <david@xxxxxxxxxxxxxxxxxxxxx> writes: > On Mon, Sep 17, 2018 at 04:41:52PM +1000, Stewart Smith wrote: >> Rob Herring <robh@xxxxxxxxxx> writes: >> > On Thu, Sep 13, 2018 at 11:50 PM Stewart Smith <stewart@xxxxxxxxxxxxx> wrote: >> >> >> >> Hi! >> >> >> >> It appears that 32b9c61307629ac76c6ac0bead6f926d579b3d2c has introduced >> >> a couple of regressions in the DTS output: >> >> >> >> For a start, I get >> >> compatible = "ibm,powernv\0ibm,firenze"; >> >> Rather than >> >> compatible = "ibm,powernv", "ibm,firenze"; >> >> (which is what I'd expect, and what's in the dts) >> > >> > I guess this happens when there are no markers. I guess our choices >> > are handle this in the output or add markers on input. >> >> I should look into that for what we produce from firmware too. > > Uh.. what? > > Markers an strictly a dtc internal concept, they have no presence in > the dtb file. Ahh, I had just assumed there was an extension to it rather than looking it up and being sensible/intelligent. I'll go back in my corner. -- Stewart Smith OPAL Architect, IBM.
