Hi Geert,
On Friday 06 Jan 2017 11:18:32 Geert Uytterhoeven wrote:
> On Wed, Jan 4, 2017 at 12:06 AM, Laurent Pinchart wrote:
> > Even though most of its registers are 8-bit wide, the IRDA has two
> > 16-bit registers that make it a 16-bit peripheral and not a 8-bit
> > peripheral with addresses shifted by one. Fix the memory resource size
> > and the platform data regshift value.
> >
> > Signed-off-by: Laurent Pinchart
> > <laurent.pinchart+renesas@xxxxxxxxxxxxxxxx>
> > ---
> >
> > arch/sh/kernel/cpu/sh3/setup-sh770x.c | 3 +--
> > 1 file changed, 1 insertion(+), 2 deletions(-)
> >
> > diff --git a/arch/sh/kernel/cpu/sh3/setup-sh770x.c
> > b/arch/sh/kernel/cpu/sh3/setup-sh770x.c index e1e54258b822..084a91e6027e
> > 100644
> > --- a/arch/sh/kernel/cpu/sh3/setup-sh770x.c
> > +++ b/arch/sh/kernel/cpu/sh3/setup-sh770x.c
> > @@ -157,11 +157,10 @@ static struct platform_device scif1_device = {
> > static struct plat_sci_port scif2_platform_data = {
> > .type = PORT_IRDA,
> > .ops = &sh770x_sci_port_ops,
> > - .regshift = 1,
>
> According to my sh7707 and sh7709 datasheets. regshift = 1 is correct.
>
> SCIx_IRDA_REGTYPE uses:
>
> [SCSMR] = { 0x00, 8 },
> [SCBRR] = { 0x01, 8 },
> [SCSCR] = { 0x02, 8 },
> [SCxTDR] = { 0x03, 8 },
> [SCxSR] = { 0x04, 8 },
> [SCxRDR] = { 0x05, 8 },
> [SCFCR] = { 0x06, 8 },
> [SCFDR] = { 0x07, 16 },
>
> While the datasheet says:
>
> SCSMR1 H'A4000140 8 bits
> SCBRR1 H'A4000142 8 bits
> SCSCR1 H'A4000144 8 bits
> SCFTDR1 H'A4000146 8 bits
> SCSSR1 H'A4000148 16 bits
> SCFRDR1 H'A400014A 8 bits
> SCFCR1 H'A400014C 8 bits
> SCFDR1 H'A400014E 16 bits
>
> So you do need regshift = 1 to handle the gaps.
>
> Note that SCSSR1 is a 16-bit registers, while SCIx_IRDA_REGTYPE
> declares it as 8-bit, so it may not work at all...
>
> Ah, you're fixing all this in "[PATCH 19/19] serial: sh-sci: Compute the
> regshift value for SCI ports".
> I think that part should be moved to this patch, to not break bisection.
Yes, I realized that after sending the series. If you think we need to care
about bisection here (given that it seems that sh-sci hasn't been tested on SH
since at least v3.1...) we can fix that.
> > static struct resource scif2_resources[] = {
> > - DEFINE_RES_MEM(0xa4000140, 0x10),
> > + DEFINE_RES_MEM(0xa4000140, 0x20),
> > DEFINE_RES_IRQ(evt2irq(0x880)),
> > };
>
> According to the register list above, resource size 0x10 is correct.
I'll fix this.
--
Regards,
Laurent Pinchart
--
To unsubscribe from this list: send the line "unsubscribe linux-serial" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
