Hi Rob,
Thanks for your reply.
On 2016-09-27 10:57:56 -0500, Rob Herring wrote:
> On Tue, Sep 27, 2016 at 6:01 AM, Niklas Söderlund
> <niklas.soderlund@xxxxxxxxxxxx> wrote:
> > Hi Rob and Vinod,
> >
> > While testing on Renesas Koelsch board Vinods 'next' branch at
> >
> > git://git.kernel.org/pub/scm/linux/kernel/git/vkoul/slave-dma.git
> >
> > I ran into some problems with Robs patch 761ed4a (tty: serial_core:
> > convert uart_close to use tty_port_close) and my systemd based init
> > system. The kernel starts fine, mounts the NFS root and starts init but
> > then the serial console go silent and I'm not able to communicate with
> > the system.
>
> There's a fix later in Greg's tree that's not in this tree. If that
> doesn't fix it, can you try reverting the hunk that sets the
> tty_port.console flag.
I tried Greg's branch tty-next from
git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/tty.git
Which contains the commit I think you are referring to, a5a2b13 (serial:
core: fix potential NULL pointer dereference). Unfortunately it dose not
fix the problem I see.
I also tried your other suggestion by reverting the chunk that sets the
tty_port.console flag (on top of tty-next branch), like this:
@@ -2746,8 +2746,6 @@ int uart_add_one_port(struct uart_driver *drv, struct uart_port *uport)
uport->cons = drv->cons;
uport->minor = drv->tty_driver->minor_start + uport->line;
- port->console = uart_console(uport);
-
/*
* If this port is a console, then the spinlock is already
* initialised.
The problem still unfortunately persists :-(
I can still revert the commits a5a2b13 (serial: core: fix potential NULL
pointer dereference) and 761ed4a (tty: serial_core: convert uart_close
to use tty_port_close) from the top of the tty-next branch and the
system boots fine.
The configuration I'm using for all tests are shmobile_defconfig +
CONFIG_CGROUPS + CONFIG_SECCOMP.
--
Regards,
Niklas Söderlund
