On Tue, Mar 12, 2024 at 9:24 AM Daniel Gimpelevich <daniel@xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx> wrote: > > That link is to a question that the submitter asked, and nobody > responded to it. It seems that this patch stems from an incomplete > reading of the kernel documentation. Those docs say: > > > SPP (Standard Parallel Port) functions modify so-called SPP registers: > > data, status, and control. The hardware may not actually have > > registers exactly like that, but the PC does and this interface is > > modelled after common PC implementations. Other low-level drivers may > > be able to emulate most of the functionality. > > So, the PARPORT_MODE_PCSPP flag denotes the availability of the SPP port > functions, not any fields in a struct. Hello Daniel, thanks for your reply. I still don't quite understand what it would mean for a driver to lack PARPORT_MODE_PCSPP. Is the idea that a parallel port could support EPP or ECP without supporting SPP? If that's the case then in my opinion the documentation should still be rewritten, but in a way to clarify that the SPP functions [1] are not available without the flag, and the flag does not imply low latency. -Alex [1] Namely: read_data, write_data, read_status, read_control, write_control, frob_control, enable_irq, disable_irq, data_forward, and data_reverse. See https://docs.kernel.org/driver-api/parport-lowlevel.html
