On Fri, Oct 5, 2012 at 10:34 PM, Peter Maydell <[email protected]> wrote: > On 5 October 2012 01:08, Peter Crosthwaite > <[email protected]> wrote: >> From: Peter A. G. Crosthwaite <[email protected]> >> >> Removed the explicit SSI mux and wired the CS line directly up to the SSI >> devices. >> >> Signed-off-by: Peter A. G. Crosthwaite <[email protected]> >> + dev = ssi_create_slave(bus, "ssi-sd"); >> if (gpio_dev[GPIO_D]) { >> - qdev_connect_gpio_out(gpio_dev[GPIO_D], 0, select_pin); >> + qdev_connect_gpio_out(gpio_dev[GPIO_D], 0, >> + qdev_get_gpio_in(dev, 0)); >> } >> >> - bus = qdev_get_child_bus(mux, "ssi0"); >> - ssi_create_slave(bus, "ssi-sd"); >> - >> - bus = qdev_get_child_bus(mux, "ssi1"); >> dev = ssi_create_slave(bus, "ssd0323"); >> + if (gpio_dev[GPIO_D]) { >> + qdev_connect_gpio_out(gpio_dev[GPIO_D], 0, >> + qdev_get_gpio_in(dev, 0)); >> + } > > ...this seems to be trying to wire up two devices to the same > GPIO output ? >
Reimplemented using qemu_irq_split() Regards, Peter > -- PMM
