On 17 December 2017 at 05:57, Adam Baxter <volta...@voltagex.org> wrote: > Hi, > I realise this isn't strictly a qemu issue but I can't find any reference > to this kernel panic happening on real hardware (I can't currently get > serial out working on my Pi, so I can't test there) > > When running qemu 2.10 on Debian, with the command line > > qemu-system-arm -machine raspi2 -curses -serial stdio -kernel zImage > -append "rw earlyprintk loglevel=8 console=ttyAMA0,115200 > dwc_otg.lpm_enable=0 root=ram0" -dtb bcm2710-rpi-3-b.dtb > > I get a kernel panic: WARNING: CPU: 0 PID: 1 at > drivers/tty/serial/serial_core.c:463 uart_get_baud_rate+0xfc/0x160
That's just a warning. The kernel panic is later, because the init process died for some reason. Looking at all the warnings and backtraces in the log that seem to be problems with clock setup, my guess is that QEMU's model of the clock devices for this board are missing or wrong, and the kernel then fails to configure the clocktree, which in turn means the serial port isn't configured right, and then init dies because it can't open the thing it's trying to output to. thanks -- PMM
