Case 2 -> none physical port, virtual console Case 1 -> 00:04: ttyS0 at I/O 0x3f8 (irq = 4, base_baud = 115200) is a 16550A Alexander -----Ursprüngliche Nachricht----- Von: linux-serial-owner@xxxxxxxxxxxxxxx [mailto:linux-serial-owner@xxxxxxxxxxxxxxx] Im Auftrag von Peter Hurley Gesendet: Mittwoch, 6. Mai 2015 13:38 An: Hänel-Baas, Alexander Cc: linux-serial@xxxxxxxxxxxxxxx Betreff: Re: AW: AW: 8250_pci -> change device name On 05/06/2015 07:32 AM, Hänel-Baas, Alexander wrote: > Hi Peter > > In case 2 the SIO-UART controller are not present (unplugged, > physically not connected), and then the tty enumeration set the HSUART from ttys2 to ttyS0 and so the kernel message was printed physically to the wrong com port. To what port did you want the kernel message to print? > Look the kernel message: > Case 1: SIO-UART and PCI-HSUART are present > 00:04: ttyS0 at I/O 0x3f8 (irq = 4, base_baud = 115200) is a 16550A <-- Kernel message > 00:05: ttyS1 at I/O 0x2f8 (irq = 3, base_baud = 115200) is a 16550A > 0000:00:1e.3: ttyS2 at MMIO 0x50705000 (irq = 18, base_baud = 2764800) > is a 16550A <-- free to use > 0000:00:1e.4: ttyS3 at MMIO 0x50707000 (irq = 16, base_baud = 2764800) > is a 16550A > > Case 2: only PCI-HSUART are present > 0000:00:1e.3: ttyS0 at MMIO 0x50705000 (irq = 18, base_baud = 2764800) > is a 16550A <-- not free now with kernel message include > 0000:00:1e.4: ttyS1 at MMIO 0x50707000 (irq = 16, base_baud = 2764800) > is a 16550A > > We switch on the system in this 2 different cases. Case 1 is the service case and we can see when the kernel starts,... and so on. > Case 1 the normal use without connected SIO-UART Controller. > > And so my idea was to change the device name for the pci uart. > > I hope this help you to understand my little problem. > > Alexander > > -----Ursprüngliche Nachricht----- > Von: Peter Hurley [mailto:peter@xxxxxxxxxxxxxxxxxx] > Gesendet: Mittwoch, 6. Mai 2015 12:14 > An: Hänel-Baas, Alexander > Cc: Greg KH; linux-serial@xxxxxxxxxxxxxxx > Betreff: Re: AW: 8250_pci -> change device name > > On 05/06/2015 02:34 AM, Hänel-Baas, Alexander wrote: >> Hi, >> >> thanks four your reply. >> >>> The right way to handle this would be to create a symlink in a udev script (or rc.d scripts if you don't have/use udev). >>> Regards, >>> Peter Hurley >> >>> Why not just change it in userspace using a tool like udev? Don't force the kernel to do something like this. >>> thanks, >>> greg k-h >> >> I have played with udev and yes i can create a symlink when the 8250_pci.ko is load. >> But this is not the right solution for my problem: >> >> My System has a SIO-Uart pluggable over the PLC-Bus and a PCI ValleyView HSUART connected over the pci-bus. >> >> So the tty enumeration look like this: >> SIO-UART -> ttyS0 -> this is the kernel console >> -> ttyS1 >> PCI-HSUART -> ttyS2 -> this is the console that can use by the userspace program >> -> ttyS3 >> >> Case 2: the SIO-Uart is not connected. >> PCI-HSUART -> ttyS0 -> and now the kernel prompt message and the userspace program too >> -> ttyS1 > > It's not clear to me what you want to have happen in case 2. > > >> In this case I need a free tty device for the PCI-HSUART and so I hope I can changed the device name only for the 8250_pci device to do this. >> >> Any ideas what can I do? >> >> Regards, >> Alexander >> >> > -- To unsubscribe from this list: send the line "unsubscribe linux-serial" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html ��.n��������+%������w��{.n�����{��ǫ����{ay� ʇڙ���f���h������_�(�階�ݢj"��������G����?���&��
