On Fri, Jun 21, 2024 at 8:47 PM Robin Murphy <robin.murphy@xxxxxxx> wrote: > > We no longer have a notion of partially-initialised fwspecs existing, > and we also no longer need to use an iommu_ops pointer to return status > to of_dma_configure(). Clean up the remains of those, which lends itself > to clarifying the logic around the dma_range_map allocation as well. ... > + if (!err && dev->bus) > + err = iommu_probe_device(dev); > > + if (err && err != -EPROBE_DEFER) > + dev_dbg(dev, "Adding to IOMMU failed: %d\n", err); Hmm... I'm wondering if dev_err_probe() can be used here. > return err; ... > + dev_dbg(dev, "device is%sbehind an iommu\n", > + !ret ? " " : " not "); Why not a positive test? -- With Best Regards, Andy Shevchenko
