On Fri, 21 Sep 2007, Russell King wrote: > I suspect the correct fix for this is to take the fix in the ARM tree > for serial_core's console initialisation, and ensure that zs.c handles > the power management of the port (enabling/disabling transmitter) > in the power management callback. Hmm, it sounds a bit hackish to me. The serial part is from the 1970s -- nobody thought of power management back then. And even though the DECstation is about 20 years younger, it does not provide any means for power management either -- it is a workstation/server class system after all. > You can find the fix in the mbox archive: > > http://ftp.arm.linux.org.uk/pub/linux/arm/kernel/git-cut/arm:devel.mbox Hmm, nothing of relevance there. I see no problem with implementing the ->pm() call as something like: void zs_pm(struct uart_port uport, unsigned int state, unsigned int oldstate) { struct zs_port *zport = to_zport(uport); if (state < 3) zport->regs[5] |= TxENAB; else zport->regs[5] &= ~TxENAB; write_zsreg(zport, R5, zport->regs[5]); } itself, but it looks more like a workaround than a fix to me. I'll have to think about it and also double-check it is safe. The SCC is about as complicated as you can get with a serial port. Thanks for the hint though. Maciej - 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