Skip to contentSkip to conversation

ARM: shmobile: koelsch: Add support HSCIF1

PATCH New Nobuhiro Iwamatsu 5 participants 10 messages Jun 10, 2014 – Jun 16, 2014

Subsystems (2)
lore
Share & tools
lore
#0 lore raw

PATCH v2 1/1 ARM: shmobile: koelsch: Add support HSCIF1

Koelsch can use HSCIF1 insetead of SCIF1 as serial port. But if we want to use, we will need to disable the GPIO key and remodeling of the board. Therefore status of hscif1 does not set the "okay".

---
V2: Remove hscif1_clk from renesas,groups. This pin does not use.

 arch/arm/boot/dts/r8a7791-koelsch.dts | 11 +++++++++++
 1 file changed, 11 insertions(+)
arch/arm/boot/dts/r8a7791-koelsch.dts +11 -0
index b2e6616..0992673 100644
@@ -22,6 +22,7 @@
 	aliases {
 		serial6 = &scif0;
 		serial7 = &scif1;
+		serial16 = &hscif1;
 	};
 
 	chosen {
@@ -258,6 +259,11 @@
 		renesas,function = "scif1";
 	};
 
+	hscif1_pins: serial2 {
+		renesas,groups = "hscif1_data", "hscif1_ctrl";
+		renesas,function = "hscif1";
+	};
+
 	ether_pins: ether {
 		renesas,groups = "eth_link", "eth_mdio", "eth_rmii";
 		renesas,function = "eth";
@@ -334,6 +340,11 @@
 	status = "okay";
 };
 
+&hscif1 {
+	pinctrl-0 = <&hscif1_pins>;
+	pinctrl-names = "default";
+};
+
 &sdhi0 {
 	pinctrl-0 = <&sdhi0_pins>;
 	pinctrl-names = "default";
Signature · 1 line
2.0.0
#1 lore raw

PATCH v2 1/1 ARM: shmobile: koelsch: Add support HSCIF1

Hi Iwamatsu-san,

Quoted 7 lines · Nobuhiro Iwamatsu

Koelsch can use HSCIF1 insetead of SCIF1 as serial port. But if we want to use, we will need to disable the GPIO key and remodeling of the board. Therefore status of hscif1 does not set the "okay".

---
V2: Remove hscif1_clk from renesas,groups. This pin does not use.

Thanks for updating the patch. This looks fine to me.

Cheers,

/ magnus

#2 lore raw

PATCH v2 1/1 ARM: shmobile: koelsch: Add support HSCIF1

Quoted 8 lines · Magnus Damm

Hi Iwamatsu-san,

Quoted 4 lines · Nobuhiro Iwamatsu

Koelsch can use HSCIF1 insetead of SCIF1 as serial port. But if we want to use, we will need to disable the GPIO key and remodeling of the board. Therefore status of hscif1 does not set the "okay".

Quoted 7 lines
Quoted 3 lines

---
V2: Remove hscif1_clk from renesas,groups. This pin does not use.

Thanks for updating the patch. This looks fine to me.

Thanks, I will queue this up.

#3 lore raw

PATCH v2 1/1 ARM: shmobile: koelsch: Add support HSCIF1

Hi Simon,

Quoted 9 lines · On Tuesday 10 June 2014 16:00:48 Simon Horman wrote:
Quoted 8 lines · Magnus Damm
Quoted 7 lines · Nobuhiro Iwamatsu

Koelsch can use HSCIF1 insetead of SCIF1 as serial port. But if we want to use, we will need to disable the GPIO key and remodeling of the board. Therefore status of hscif1 does not set the "okay".

---

Quoted 7 lines
Quoted 5 lines
Quoted 1 line

V2: Remove hscif1_clk from renesas,groups. This pin does not use.

Thanks for updating the patch. This looks fine to me.

Thanks, I will queue this up.

Sorry for the late reply, but I have some doubts about this. The HRX0 and HTX0 signals are connected to push buttons and used as GPIOs. Beside, they're not routed to serial port transceiver. If we want to use HSCIF0 on those pins on Koelsch and extension board must be connected to the EXIO C connector, and two resistors and two capacitors should be unsoldered. I thus don't think the Koeslch DT should contain pinmux data for the HSCIF0 device. If an extension board is used and HSCIF0 needed, then the DT file for the extension board should define pinmuxing for the serial port.

Signature · 3 lines
Regards,

Laurent Pinchart
#4 lore raw

PATCH v2 1/1 ARM: shmobile: koelsch: Add support HSCIF1

Quoted 20 lines · Laurent Pinchart

Hi Simon,

Quoted 9 lines · On Tuesday 10 June 2014 16:00:48 Simon Horman wrote:
Quoted 8 lines · Magnus Damm
Quoted 7 lines · Nobuhiro Iwamatsu

Koelsch can use HSCIF1 insetead of SCIF1 as serial port. But if we want to use, we will need to disable the GPIO key and remodeling of the board. Therefore status of hscif1 does not set the "okay".

---

Quoted 7 lines
Quoted 5 lines
Quoted 1 line

V2: Remove hscif1_clk from renesas,groups. This pin does not use.

Thanks for updating the patch. This looks fine to me.

Thanks, I will queue this up.

Quoted 9 lines

Sorry for the late reply, but I have some doubts about this. The HRX0 and HTX0 signals are connected to push buttons and used as GPIOs. Beside, they're not routed to serial port transceiver. If we want to use HSCIF0 on those pins on Koelsch and extension board must be connected to the EXIO C connector, and two resistors and two capacitors should be unsoldered. I thus don't think the Koeslch DT should contain pinmux data for the HSCIF0 device. If an extension board is used and HSCIF0 needed, then the DT file for the extension board should define pinmuxing for the serial port.

If you would like me to revert this patch I am happy to consider doing so.

#5 lore raw

PATCH v2 1/1 ARM: shmobile: koelsch: Add support HSCIF1

Hi Simon,

Quoted 30 lines · On Friday 13 June 2014 16:43:07 Simon Horman wrote:
Quoted 19 lines · Laurent Pinchart
Quoted 10 lines · On Tuesday 10 June 2014 16:00:48 Simon Horman wrote:
Quoted 9 lines · Magnus Damm
Quoted 8 lines · Nobuhiro Iwamatsu

Koelsch can use HSCIF1 insetead of SCIF1 as serial port. But if we want
to use, we will need to disable the GPIO key and remodeling of the board. Therefore status of hscif1 does not set the "okay".

---

Quoted 7 lines
Quoted 5 lines
Quoted 1 line

V2: Remove hscif1_clk from renesas,groups. This pin does not use.

Thanks for updating the patch. This looks fine to me.

Thanks, I will queue this up.

Quoted 9 lines

Sorry for the late reply, but I have some doubts about this. The HRX0 and HTX0 signals are connected to push buttons and used as GPIOs. Beside, they're not routed to serial port transceiver. If we want to use HSCIF0 on those pins on Koelsch and extension board must be connected to the EXIO C connector, and two resistors and two capacitors should be unsoldered. I thus don't think the Koeslch DT should contain pinmux data for the HSCIF0 device. If an extension board is used and HSCIF0 needed, then the DT file for the extension board should define pinmuxing for the serial port.

Quoted 1 line

If you would like me to revert this patch I am happy to consider doing so.

That's one option, but I'd like to get feedback on my point of view before reverting it. Who knows, I might be wrong ;-)

Signature · 3 lines
Regards,

Laurent Pinchart
#6 lore raw

PATCH v2 1/1 ARM: shmobile: koelsch: Add support HSCIF1

Quoted 21 lines · Laurent Pinchart
Quoted 20 lines · On Friday 13 June 2014 16:43:07 Simon Horman wrote:
Quoted 19 lines · Laurent Pinchart
Quoted 17 lines · On Tuesday 10 June 2014 16:00:48 Simon Horman wrote:
Quoted 6 lines · Magnus Damm
Quoted 5 lines · Nobuhiro Iwamatsu

Koelsch can use HSCIF1 insetead of SCIF1 as serial port. But if we want
to use, we will need to disable the GPIO key and remodeling of the board. Therefore status of hscif1 does not set the "okay".

Quoted 8 lines
Quoted 4 lines

---

V2: Remove hscif1_clk from renesas,groups. This pin does not use.

Thanks for updating the patch. This looks fine to me.

Thanks, I will queue this up.

Quoted 9 lines
Quoted 9 lines
Quoted 9 lines

Sorry for the late reply, but I have some doubts about this. The HRX0 and HTX0 signals are connected to push buttons and used as GPIOs. Beside, they're not routed to serial port transceiver. If we want to use HSCIF0 on those pins on Koelsch and extension board must be connected to the EXIO C connector, and two resistors and two capacitors should be unsoldered. I thus don't think the Koeslch DT should contain pinmux data for the HSCIF0 device. If an extension board is used and HSCIF0 needed, then the DT file for the extension board should define pinmuxing for the serial port.

You mean HRX1, HTX1, and HSCIF1?

HTX1 and HRX1 can be connected to the DEBUG1 transceiver by stuffing zero-Ohm resisters to R834 and R835.

About your comment about the resistors and capacitors, which applies to both HSCIF0 and HSCIF1: probably the serial port will still work if these are present.

Quoted 4 lines
Quoted 1 line

If you would like me to revert this patch I am happy to consider doing so.

That's one option, but I'd like to get feedback on my point of view before reverting it. Who knows, I might be wrong ;-)

Nevertheless, I agree with Laurent.

HSCIF1 (and HSCIF0) cannot be used on the Koelsch board as-is, hence they should not be present in the plain Koelsch DTS. They should be put in an overlay
DTS instead.

Gr{oetje,eeting}s,

Geert

--
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- geert at linux-m68k.org

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

#7 lore raw

PATCH v2 1/1 ARM: shmobile: koelsch: Add support HSCIF1

Hi Geert,

Quoted 34 lines · On Friday 13 June 2014 13:16:34 Geert Uytterhoeven wrote:
Quoted 23 lines · Laurent Pinchart
Quoted 22 lines · On Friday 13 June 2014 16:43:07 Simon Horman wrote:
Quoted 21 lines · Laurent Pinchart
Quoted 19 lines · On Tuesday 10 June 2014 16:00:48 Simon Horman wrote:
Quoted 10 lines · Magnus Damm
Quoted 9 lines · Nobuhiro Iwamatsu

Koelsch can use HSCIF1 insetead of SCIF1 as serial port. But if we want to use, we will need to disable the GPIO key and remodeling of the board. Therefore status of hscif1 does not set the "okay".

Signed-off-by: Nobuhiro Iwamatsu
<[email protected]>
---

Quoted 6 lines
Quoted 2 lines

V2: Remove hscif1_clk from renesas,groups. This pin does not use.

Thanks for updating the patch. This looks fine to me.

Thanks, I will queue this up.

Quoted 9 lines
Quoted 9 lines
Quoted 9 lines

Sorry for the late reply, but I have some doubts about this. The HRX0 and HTX0 signals are connected to push buttons and used as GPIOs. Beside, they're not routed to serial port transceiver. If we want to use HSCIF0 on those pins on Koelsch and extension board must be connected to the EXIO C connector, and two resistors and two capacitors should be unsoldered. I thus don't think the Koeslch DT should contain pinmux data for the HSCIF0 device. If an extension board is used and HSCIF0 needed, then the DT file for the extension board should define pinmuxing for the serial port.

Quoted 1 line

You mean HRX1, HTX1, and HSCIF1?

*sigh*... :-)

Quoted 6 lines

HTX1 and HRX1 can be connected to the DEBUG1 transceiver by stuffing zero-Ohm resisters to R834 and R835.

About your comment about the resistors and capacitors, which applies to both HSCIF0 and HSCIF1: probably the serial port will still work if these are present.

With a 4.7?F capacitive load on the transmission line, assuming a drive strength of 40mA (which is pretty high), switching between 0V and 3.3V takes 388?s, corresponding to 2579 Hz. You'll have to reduce the baud rate :-)

Quoted 11 lines
Quoted 5 lines
Quoted 2 lines

If you would like me to revert this patch I am happy to consider doing so.

That's one option, but I'd like to get feedback on my point of view before reverting it. Who knows, I might be wrong ;-)

Nevertheless, I agree with Laurent.

HSCIF1 (and HSCIF0) cannot be used on the Koelsch board as-is, hence they should not be present in the plain Koelsch DTS. They should be put in an overlay DTS instead.

Signature · 3 lines
Regards,

Laurent Pinchart
#8 lore raw

PATCH v2 1/1 ARM: shmobile: koelsch: Add support HSCIF1

Hi Laurent,

Quoted 7 lines · Laurent Pinchart
Quoted 3 lines

About your comment about the resistors and capacitors, which applies to both HSCIF0 and HSCIF1: probably the serial port will still work if these are present.

With a 4.7?F capacitive load on the transmission line, assuming a drive strength of 40mA (which is pretty high), switching between 0V and 3.3V takes 388?s, corresponding to 2579 Hz. You'll have to reduce the baud rate :-)

Right, 4.7?F is a bit high. So at least the capacitor has to go for today's typical serial speeds.

Gr{oetje,eeting}s,

Geert

--
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- geert at linux-m68k.org

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

#9 lore raw

PATCH v2 1/1 ARM: shmobile: koelsch: Add support HSCIF1

In reply to Geert Uytterhoeven
Quoted 47 lines · Geert Uytterhoeven
Quoted 21 lines · Laurent Pinchart
Quoted 20 lines · On Friday 13 June 2014 16:43:07 Simon Horman wrote:
Quoted 19 lines · Laurent Pinchart
Quoted 17 lines · On Tuesday 10 June 2014 16:00:48 Simon Horman wrote:
Quoted 6 lines · Magnus Damm
Quoted 5 lines · Nobuhiro Iwamatsu

Koelsch can use HSCIF1 insetead of SCIF1 as serial port. But if we want
to use, we will need to disable the GPIO key and remodeling of the board. Therefore status of hscif1 does not set the "okay".

Quoted 8 lines
Quoted 4 lines

---

V2: Remove hscif1_clk from renesas,groups. This pin does not use.

Thanks for updating the patch. This looks fine to me.

Thanks, I will queue this up.

Quoted 9 lines
Quoted 9 lines
Quoted 9 lines

Sorry for the late reply, but I have some doubts about this. The HRX0 and HTX0 signals are connected to push buttons and used as GPIOs. Beside, they're not routed to serial port transceiver. If we want to use HSCIF0 on those pins on Koelsch and extension board must be connected to the EXIO C connector, and two resistors and two capacitors should be unsoldered. I thus don't think the Koeslch DT should contain pinmux data for the HSCIF0 device. If an extension board is used and HSCIF0 needed, then the DT file for the extension board should define pinmuxing for the serial port.

You mean HRX1, HTX1, and HSCIF1?

HTX1 and HRX1 can be connected to the DEBUG1 transceiver by stuffing zero-Ohm resisters to R834 and R835.

About your comment about the resistors and capacitors, which applies to both HSCIF0 and HSCIF1: probably the serial port will still work if these are present.

Quoted 4 lines
Quoted 1 line

If you would like me to revert this patch I am happy to consider doing so.

That's one option, but I'd like to get feedback on my point of view before reverting it. Who knows, I might be wrong ;-)

Quoted 6 lines

Nevertheless, I agree with Laurent.

HSCIF1 (and HSCIF0) cannot be used on the Koelsch board as-is, hence they should not be present in the plain Koelsch DTS. They should be put in an overlay
DTS instead.

Thanks, I have reverted this patch and will probably subsequently drop it and the revert when I rebase my branches on top of v3.16-rc1 (due to ARM-SoC process requirements).

Lets revisit this if necessary.