On Wed, 2021-10-06 at 11:58 +0300, Leon Romanovsky wrote: > On Thu, Sep 30, 2021 at 06:28:12PM -0700, David E. Box wrote: > > Intel Platform Monitoring Technology (PMT) support is indicated by presence > > of an Intel defined PCIe DVSEC structure with a PMT ID. However DVSEC > > structures may also be used by Intel to indicate support for other > > capabilities unrelated to PMT. The Out Of Band Management Services Module > > (OOBMSM) is an example of a device that can have both PMT and non-PMT > > capabilities. In order to support these capabilities it is necessary to > > modify the intel_pmt driver to handle the creation of platform devices more > > generically. To that end the following changes are made. > > > > Convert the driver and child drivers from MFD to the Auxiliary Bus. This > > architecture is more suitable anyway since the driver partitions a > > multifunctional PCIe device. This also moves the driver out of the MFD > > subsystem and into platform/x86/intel. > > > > Before, devices were named by their capability (e.g. pmt_telemetry). > > Instead, generically name them by their capability ID (e.g. > > intel_extended_cap.2). This allows the IDs to be created automatically, > > minimizing the code needed to support future capabilities. However, to > > ensure that unsupported devices aren't created, use an allow list to > > specify supported capabilities. Along these lines, rename the driver from > > intel_pmt to intel_extended_caps to better reflect the purpose. > > > > Signed-off-by: David E. Box <david.e.box@xxxxxxxxxxxxxxx> > > --- > > > > V1: New patch. However incorporates some elements of [1] which was > > dropped. Namely enumerating features generically and creating an > > allow list. Also cleans up probe by moving some code to functions > > and using a bool instead of an int to track whether a device was > > added. > > > > [1] https://lore.kernel.org/all/20210922213007.2738388-3-david.e.box@xxxxxxxxxxxxxxx/ > > <...> > > > +static int extended_caps_pci_probe(struct pci_dev *pdev, const struct pci_device_id *id) > > +{ > > + struct extended_caps_platform_info *info; > > + bool have_devices = false; > > + unsigned long quirks = 0; > > + int ret; > > + > > + ret = pcim_enable_device(pdev); > > + if (ret) > > + return ret; > > + > > + info = (struct extended_caps_platform_info *)id->driver_data; > > pci_get_drvdata() in all places and no need to cast void *. This is coming from the id not the pdev. The data here is type kernel_ulong_t. > > > + if (info) > > + quirks = info->quirks; > > + > > + have_devices |= extended_caps_walk_dvsec(pdev, quirks); > > + > > + if (info && (info->quirks & EXT_CAPS_QUIRK_NO_DVSEC)) > > + have_devices |= extended_caps_walk_header(pdev, quirks, info->capabilities); > > + > > + if (!have_devices) > > + return -ENODEV; > > + > > + return 0; > > +} > > <...> > > > -static struct platform_driver pmt_telem_driver = { > > - .driver = { > > - .name = TELEM_DEV_NAME, > > - }, > > - .remove = pmt_telem_remove, > > - .probe = pmt_telem_probe, > > +static const struct auxiliary_device_id pmt_telem_aux_id_table[] = { > > + { .name = "intel_extended_caps.2", }, > > Why "2" in the name? This is being changed to a string for the next version after similar comment. Thanks. David > > Thanks > > > + {}, > > +}; > > +MODULE_DEVICE_TABLE(auxiliary, pmt_telem_aux_id_table); > > + > > +static struct auxiliary_driver pmt_telem_aux_driver = { > > + .id_table = pmt_telem_aux_id_table, > > + .remove = pmt_telem_remove, > > + .probe = pmt_telem_probe, > > }; > >
