Re: [PATCH RFC 3/3] MIPS: AR7: ensure the port type's FCR value is used

[James Hogan <james.hogan@mips.com>]

[-- Attachment #1: Type: text/plain, Size: 2051 bytes --]

On Thu, Dec 28, 2017 at 04:38:29PM +0100, Jonas Gorski wrote:
> On 14 November 2017 at 12:02, James Hogan <james.hogan@mips.com> wrote:
> > On Sun, Oct 29, 2017 at 04:27:21PM +0100, Jonas Gorski wrote:
> >> Since commit aef9a7bd9b67 ("serial/uart/8250: Add tunable RX interrupt
> >> trigger I/F of FIFO buffers"), the port's default FCR value isn't used
> >> in serial8250_do_set_termios anymore, but copied over once in
> >> serial8250_config_port and then modified as needed.
> >>
> >> Unfortunately, serial8250_config_port will never be called if the port
> >> is shared between kernel and userspace, and the port's flag doesn't have
> >> UPF_BOOT_AUTOCONF, which would trigger a serial8250_config_port as well.
> >>
> >> This causes garbled output from userspace:
> >>
> >> [    5.220000] random: procd urandom read with 49 bits of entropy available
> >> ers
> >>    [kee
> >>
> >> Fix this by forcing it to be configured on boot, resulting in the
> >> expected output:
> >>
> >> [    5.250000] random: procd urandom read with 50 bits of entropy available
> >> Press the [f] key and hit [enter] to enter failsafe mode
> >> Press the [1], [2], [3] or [4] key and hit [enter] to select the debug level
> >>
> >> Fixes: aef9a7bd9b67 ("serial/uart/8250: Add tunable RX interrupt trigger I/F of FIFO buffers")
> >> Signed-off-by: Jonas Gorski <jonas.gorski@gmail.com>
> >> ---
> >> I'm not sure if this is just AR7's issue, or if this points to a general
> >> issue for UARTs used as kernel console and login console with the "fixed"
> >> commit.
> >
> > Thanks. Given nobody seems to have objected, I've applied to my
> > mips-fixes branch, with stable tag for 3.17+.
> 
> Hmm, I don't see it in
> https://git.kernel.org/pub/scm/linux/kernel/git/jhogan/mips.git/log/?h=mips-fixes
> - did you maybe forget to push?

Ralf reappeared so I delegated the patch to him in patchwork, and he has
apparently marked it as accepted.

But perhaps that only just happened and he hasn't pushed yet. Ralf?

Cheers
James

[-- Attachment #2: Digital signature --]
[-- Type: application/pgp-signature, Size: 833 bytes --]