Hi Chris,
On Sun, Jul 29, 2018 at 1:11 PM Geert Uytterhoeven <geert@xxxxxxxxxxxxxx> wrote:
> On Fri, Jul 27, 2018 at 11:09 PM Chris Brandt <chris.brandt@xxxxxxxxxxx> wrote:
> > Some devices with SCIx_SH4_SCIF_REGTYPE have no space between registers.
> > Use the register area size to determine the spacing between register.
> >
> > Signed-off-by: Chris Brandt <chris.brandt@xxxxxxxxxxx>
> > ---
> > drivers/tty/serial/sh-sci.c | 22 +++++++++++++---------
> > 1 file changed, 13 insertions(+), 9 deletions(-)
> >
> > diff --git a/drivers/tty/serial/sh-sci.c b/drivers/tty/serial/sh-sci.c
> > index 6ff6f2bf3b9b..138296ec9a7d 100644
> > --- a/drivers/tty/serial/sh-sci.c
> > +++ b/drivers/tty/serial/sh-sci.c
> > @@ -348,15 +348,15 @@ static const struct sci_port_params sci_port_params[SCIx_NR_REGTYPES] = {
> > [SCIx_SH4_SCIF_REGTYPE] = {
> > .regs = {
> > [SCSMR] = { 0x00, 16 },
> > - [SCBRR] = { 0x04, 8 },
> > - [SCSCR] = { 0x08, 16 },
> > - [SCxTDR] = { 0x0c, 8 },
> > - [SCxSR] = { 0x10, 16 },
> > - [SCxRDR] = { 0x14, 8 },
> > - [SCFCR] = { 0x18, 16 },
> > - [SCFDR] = { 0x1c, 16 },
> > - [SCSPTR] = { 0x20, 16 },
> > - [SCLSR] = { 0x24, 16 },
> > + [SCBRR] = { 0x02, 8 },
> > + [SCSCR] = { 0x04, 16 },
> > + [SCxTDR] = { 0x06, 8 },
> > + [SCxSR] = { 0x08, 16 },
> > + [SCxRDR] = { 0x0a, 8 },
> > + [SCFCR] = { 0x0c, 16 },
> > + [SCFDR] = { 0x0e, 16 },
> > + [SCSPTR] = { 0x10, 16 },
> > + [SCLSR] = { 0x12, 16 },
> > },
> > .fifosize = 16,
> > .overrun_reg = SCLSR,
> > @@ -2939,6 +2939,10 @@ static int sci_init_single(struct platform_device *dev,
> > port->regshift = 1;
> > }
> >
> > + if (p->regtype == SCIx_SH4_SCIF_REGTYPE)
> > + if (sci_port->reg_size >= 0x20)
> > + port->regshift = 1;
>
> I've accidentally discovered this breaks SCIF on SH7751R2D (QEMU), as the
> board code doesn't fill in regtype, so it is 0 = SCIx_PROBE_REGTYPE.
> The proper (default) regtype of SCIx_SH4_SCIF_REGTYPE is derived by
> sci_probe_regmap(). However, that value is never fed back to sci_init_single(),
> as plat_sci_port is const. So regshift ends up being wrong.
>
> I made it work by changing the check to:
>
> - if (p->regtype == SCIx_SH4_SCIF_REGTYPE)
> + if (p->regtype == SCIx_SH4_SCIF_REGTYPE ||
> + (p->regtype == SCIx_PROBE_REGTYPE && port->type == PORT_SCIF))
>
> Perhaps there's a better way?
Like (whitespace-damaged-gmail):
--- a/drivers/tty/serial/sh-sci.c
+++ b/drivers/tty/serial/sh-sci.c
@@ -3340,7 +3340,7 @@ static int sci_init_single(struct platform_device *dev,
{
struct uart_port *port = &sci_port->port;
const struct resource *res;
- unsigned int i;
+ unsigned int i, regtype;
int ret;
sci_port->cfg = p;
@@ -3381,6 +3381,7 @@ static int sci_init_single(struct platform_device *dev,
if (unlikely(sci_port->params == NULL))
return -EINVAL;
+ regtype = sci_port->params - sci_port_params;
switch (p->type) {
case PORT_SCIFB:
sci_port->rx_trigger = 48;
@@ -3435,7 +3436,7 @@ static int sci_init_single(struct platform_device *dev,
port->regshift = 1;
}
- if (p->regtype == SCIx_SH4_SCIF_REGTYPE)
+ if (regtype == SCIx_SH4_SCIF_REGTYPE)
if (sci_port->reg_size >= 0x20)
port->regshift = 1;
Gr{oetje,eeting}s,
Geert
--
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- geert@xxxxxxxxxxxxxx
In personal conversations with technical people, I call myself a hacker. But
when I'm talking to journalists I just say "programmer" or something like that.
-- Linus Torvalds
--
To unsubscribe from this list: send the line "unsubscribe devicetree" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
