* [PATCH v2 0/3] Add Ether support for R8A7791/Koelsch board
@ 2013-11-14 23:17 ` Sergei Shtylyov
0 siblings, 0 replies; 26+ messages in thread
From: Sergei Shtylyov @ 2013-11-14 22:17 UTC (permalink / raw)
To: linux-arm-kernel
Hello.
Resending with correct subject...
Here's the set of 3 patches against Simon Horman's 'renesas.git' repo,
'renesas-devel-v3.12-20131112' tag. Here we add the Ether support to the
R8A7791/Koelsch board. The patchset requires the 'sh_eth' driver R8A7791
support patch posted earlier in order to work.
[1/3] ARM: shmobile: r8a7791: add Ether clock
[2/3] ARM: shmobile: Koelsch: add Ether support
[3/3] ARM: shmobile: Koelsch: enable Ether in defconfig
WBR, Sergei
^ permalink raw reply [flat|nested] 26+ messages in thread
* [PATCH v2 1/3] ARM: shmobile: r8a7791: add Ether clock
2013-11-14 23:17 ` Sergei Shtylyov
@ 2013-11-14 23:19 ` Sergei Shtylyov
-1 siblings, 0 replies; 26+ messages in thread
From: Sergei Shtylyov @ 2013-11-14 22:19 UTC (permalink / raw)
To: linux-arm-kernel
Add support for R8A7791 Ether clock.
Signed-off-by: Sergei Shtylyov <sergei.shtylyov@cogentembedded.com>
---
Changes in version 2:
- changed Ether device name to "r8a779x-ether".
arch/arm/mach-shmobile/clock-r8a7791.c | 3 +++
1 file changed, 3 insertions(+)
Index: renesas/arch/arm/mach-shmobile/clock-r8a7791.c
=================================--- renesas.orig/arch/arm/mach-shmobile/clock-r8a7791.c
+++ renesas/arch/arm/mach-shmobile/clock-r8a7791.c
@@ -120,6 +120,7 @@ static struct clk *main_clks[] = {
/* MSTP */
enum {
+ MSTP813,
MSTP721, MSTP720,
MSTP719, MSTP718, MSTP715, MSTP714,
MSTP216, MSTP207, MSTP206,
@@ -129,6 +130,7 @@ enum {
};
static struct clk mstp_clks[MSTP_NR] = {
+ [MSTP813] = SH_CLK_MSTP32(&p_clk, SMSTPCR8, 13, 0), /* Ether */
[MSTP721] = SH_CLK_MSTP32(&p_clk, SMSTPCR7, 21, 0), /* SCIF0 */
[MSTP720] = SH_CLK_MSTP32(&p_clk, SMSTPCR7, 20, 0), /* SCIF1 */
[MSTP719] = SH_CLK_MSTP32(&p_clk, SMSTPCR7, 19, 0), /* SCIF2 */
@@ -180,6 +182,7 @@ static struct clk_lookup lookups[] = {
CLKDEV_DEV_ID("sh-sci.13", &mstp_clks[MSTP1106]), /* SCIFA4 */
CLKDEV_DEV_ID("sh-sci.14", &mstp_clks[MSTP1107]), /* SCIFA5 */
CLKDEV_DEV_ID("sh_cmt.0", &mstp_clks[MSTP124]),
+ CLKDEV_DEV_ID("r8a779x-ether", &mstp_clks[MSTP813]), /* Ether */
};
#define R8A7791_CLOCK_ROOT(e, m, p0, p1, p30, p31) \
^ permalink raw reply [flat|nested] 26+ messages in thread
* [PATCH v2 2/3] ARM: shmobile: Koelsch: add Ether support
2013-11-14 23:17 ` Sergei Shtylyov
@ 2013-11-14 23:21 ` Sergei Shtylyov
-1 siblings, 0 replies; 26+ messages in thread
From: Sergei Shtylyov @ 2013-11-14 22:21 UTC (permalink / raw)
To: linux-arm-kernel
Register Ether platform device and pin data on the Koelsch board.
Register platform fixup for Micrel KSZ8041 PHY, just like on the Lager board.
Signed-off-by: Sergei Shtylyov <sergei.shtylyov@cogentembedded.com>
---
Changes in version 2:
- added *if* (IS_ENABLED(CONFIG_PHYLIB)) around phy_register_fixup_for_id()
call;
- changed Ether device name to "r8a779x-ether".
arch/arm/mach-shmobile/board-koelsch.c | 63 ++++++++++++++++++++++++++++++++-
1 file changed, 62 insertions(+), 1 deletion(-)
Index: renesas/arch/arm/mach-shmobile/board-koelsch.c
=================================--- renesas.orig/arch/arm/mach-shmobile/board-koelsch.c
+++ renesas/arch/arm/mach-shmobile/board-koelsch.c
@@ -24,14 +24,31 @@
#include <linux/input.h>
#include <linux/kernel.h>
#include <linux/leds.h>
+#include <linux/phy.h>
+#include <linux/pinctrl/machine.h>
#include <linux/platform_data/gpio-rcar.h>
#include <linux/platform_device.h>
+#include <linux/sh_eth.h>
#include <mach/common.h>
+#include <mach/irqs.h>
#include <mach/r8a7791.h>
#include <mach/rcar-gen2.h>
#include <asm/mach-types.h>
#include <asm/mach/arch.h>
+/* Ether */
+static const struct sh_eth_plat_data ether_pdata __initconst = {
+ .phy = 0x1,
+ .edmac_endian = EDMAC_LITTLE_ENDIAN,
+ .phy_interface = PHY_INTERFACE_MODE_RMII,
+ .ether_link_active_low = 1,
+};
+
+static const struct resource ether_resources[] __initconst = {
+ DEFINE_RES_MEM(0xee700000, 0x400),
+ DEFINE_RES_IRQ(gic_spi(162)),
+};
+
/* LEDS */
static struct gpio_led koelsch_leds[] = {
{
@@ -71,11 +88,29 @@ static const struct gpio_keys_platform_d
.nbuttons = ARRAY_SIZE(gpio_buttons),
};
+static const struct pinctrl_map koelsch_pinctrl_map[] = {
+ /* Ether */
+ PIN_MAP_MUX_GROUP_DEFAULT("r8a779x-ether", "pfc-r8a7791",
+ "eth_link", "eth"),
+ PIN_MAP_MUX_GROUP_DEFAULT("r8a779x-ether", "pfc-r8a7791",
+ "eth_mdio", "eth"),
+ PIN_MAP_MUX_GROUP_DEFAULT("r8a779x-ether", "pfc-r8a7791",
+ "eth_rmii", "eth"),
+};
+
static void __init koelsch_add_standard_devices(void)
{
r8a7791_clock_init();
+
+ pinctrl_register_mappings(koelsch_pinctrl_map,
+ ARRAY_SIZE(koelsch_pinctrl_map));
r8a7791_pinmux_init();
+
r8a7791_add_standard_devices();
+ platform_device_register_resndata(&platform_bus, "r8a779x-ether", -1,
+ ether_resources,
+ ARRAY_SIZE(ether_resources),
+ ðer_pdata, sizeof(ether_pdata));
platform_device_register_data(&platform_bus, "leds-gpio", -1,
&koelsch_leds_pdata,
sizeof(koelsch_leds_pdata));
@@ -84,6 +119,32 @@ static void __init koelsch_add_standard_
sizeof(koelsch_keys_pdata));
}
+/*
+ * Ether LEDs on the Koelsch board are named LINK and ACTIVE which corresponds
+ * to non-default 01 setting of the Micrel KSZ8041 PHY control register 1 bits
+ * 14-15. We have to set them back to 01 from the default 00 value each time
+ * the PHY is reset. It's also important because the PHY's LED0 signal is
+ * connected to SoC's ETH_LINK signal and in the PHY's default mode it will
+ * bounce on and off after each packet, which we apparently want to avoid.
+ */
+static int koelsch_ksz8041_fixup(struct phy_device *phydev)
+{
+ u16 phyctrl1 = phy_read(phydev, 0x1e);
+
+ phyctrl1 &= ~0xc000;
+ phyctrl1 |= 0x4000;
+ return phy_write(phydev, 0x1e, phyctrl1);
+}
+
+static void __init koelsch_init(void)
+{
+ koelsch_add_standard_devices();
+
+ if (IS_ENABLED(CONFIG_PHYLIB))
+ phy_register_fixup_for_id("r8a779x-ether-ff:01",
+ koelsch_ksz8041_fixup);
+}
+
static const char * const koelsch_boards_compat_dt[] __initconst = {
"renesas,koelsch",
NULL,
@@ -92,7 +153,7 @@ static const char * const koelsch_boards
DT_MACHINE_START(KOELSCH_DT, "koelsch")
.smp = smp_ops(r8a7791_smp_ops),
.init_early = r8a7791_init_early,
- .init_machine = koelsch_add_standard_devices,
+ .init_machine = koelsch_init,
.init_time = rcar_gen2_timer_init,
.dt_compat = koelsch_boards_compat_dt,
MACHINE_END
^ permalink raw reply [flat|nested] 26+ messages in thread
* [PATCH v2 3/3] ARM: shmobile: Koelsch: enable Ether in defconfig
2013-11-14 23:17 ` Sergei Shtylyov
@ 2013-11-14 23:23 ` Sergei Shtylyov
-1 siblings, 0 replies; 26+ messages in thread
From: Sergei Shtylyov @ 2013-11-14 22:23 UTC (permalink / raw)
To: linux-arm-kernel
Enable the Ether driver in 'koelsch_defconfig' along with DHCP autoconfiguration
and NFS root.
Signed-off-by: Sergei Shtylyov <sergei.shtylyov@cogentembedded.com>
---
Changes in version 2:
- new patch in the series.
arch/arm/configs/koelsch_defconfig | 22 ++++++++++++++++++++++
1 file changed, 22 insertions(+)
Index: renesas/arch/arm/configs/koelsch_defconfig
=================================--- renesas.orig/arch/arm/configs/koelsch_defconfig
+++ renesas/arch/arm/configs/koelsch_defconfig
@@ -29,7 +29,27 @@ CONFIG_VFP=y
CONFIG_NEON=y
# CONFIG_CORE_DUMP_DEFAULT_ELF_HEADERS is not set
CONFIG_PM_RUNTIME=y
+CONFIG_NET=y
+CONFIG_INET=y
+CONFIG_IP_PNP=y
+CONFIG_IP_PNP_DHCP=y
CONFIG_UEVENT_HELPER_PATH="/sbin/hotplug"
+CONFIG_NETDEVICES=y
+# CONFIG_NET_VENDOR_ARC is not set
+# CONFIG_NET_CADENCE is not set
+# CONFIG_NET_VENDOR_BROADCOM is not set
+# CONFIG_NET_VENDOR_CIRRUS is not set
+# CONFIG_NET_VENDOR_FARADAY is not set
+# CONFIG_NET_VENDOR_INTEL is not set
+# CONFIG_NET_VENDOR_MARVELL is not set
+# CONFIG_NET_VENDOR_MICREL is not set
+# CONFIG_NET_VENDOR_NATSEMI is not set
+CONFIG_SH_ETH=y
+# CONFIG_NET_VENDOR_SEEQ is not set
+# CONFIG_NET_VENDOR_SMSC is not set
+# CONFIG_NET_VENDOR_STMICRO is not set
+# CONFIG_NET_VENDOR_VIA is not set
+# CONFIG_NET_VENDOR_WIZNET is not set
# CONFIG_INPUT_MOUSEDEV_PSAUX is not set
# CONFIG_INPUT_MOUSE is not set
# CONFIG_LEGACY_PTYS is not set
@@ -49,6 +69,8 @@ CONFIG_LEDS_CLASS=y
CONFIG_TMPFS=y
CONFIG_CONFIGFS_FS=y
# CONFIG_MISC_FILESYSTEMS is not set
+CONFIG_NFS_FS=y
+CONFIG_ROOT_NFS=y
# CONFIG_ENABLE_WARN_DEPRECATED is not set
# CONFIG_ENABLE_MUST_CHECK is not set
# CONFIG_ARM_UNWIND is not set
^ permalink raw reply [flat|nested] 26+ messages in thread
* Re: [PATCH v2 2/3] ARM: shmobile: Koelsch: add Ether support
2013-11-14 23:21 ` Sergei Shtylyov
@ 2013-11-14 23:06 ` Laurent Pinchart
-1 siblings, 0 replies; 26+ messages in thread
From: Laurent Pinchart @ 2013-11-14 23:06 UTC (permalink / raw)
To: linux-arm-kernel
Hi Sergei,
Thank you for the patch.
On Friday 15 November 2013 02:21:11 Sergei Shtylyov wrote:
> Register Ether platform device and pin data on the Koelsch board.
> Register platform fixup for Micrel KSZ8041 PHY, just like on the Lager
> board.
>
> Signed-off-by: Sergei Shtylyov <sergei.shtylyov@cogentembedded.com>
>
> ---
> Changes in version 2:
> - added *if* (IS_ENABLED(CONFIG_PHYLIB)) around phy_register_fixup_for_id()
> call;
> - changed Ether device name to "r8a779x-ether".
>
> arch/arm/mach-shmobile/board-koelsch.c | 63 ++++++++++++++++++++++++++++-
> 1 file changed, 62 insertions(+), 1 deletion(-)
>
> Index: renesas/arch/arm/mach-shmobile/board-koelsch.c
> =================================> --- renesas.orig/arch/arm/mach-shmobile/board-koelsch.c
> +++ renesas/arch/arm/mach-shmobile/board-koelsch.c
> @@ -24,14 +24,31 @@
> #include <linux/input.h>
> #include <linux/kernel.h>
> #include <linux/leds.h>
> +#include <linux/phy.h>
> +#include <linux/pinctrl/machine.h>
> #include <linux/platform_data/gpio-rcar.h>
> #include <linux/platform_device.h>
> +#include <linux/sh_eth.h>
> #include <mach/common.h>
> +#include <mach/irqs.h>
> #include <mach/r8a7791.h>
> #include <mach/rcar-gen2.h>
> #include <asm/mach-types.h>
> #include <asm/mach/arch.h>
>
> +/* Ether */
> +static const struct sh_eth_plat_data ether_pdata __initconst = {
> + .phy = 0x1,
> + .edmac_endian = EDMAC_LITTLE_ENDIAN,
> + .phy_interface = PHY_INTERFACE_MODE_RMII,
> + .ether_link_active_low = 1,
> +};
> +
> +static const struct resource ether_resources[] __initconst = {
> + DEFINE_RES_MEM(0xee700000, 0x400),
> + DEFINE_RES_IRQ(gic_spi(162)),
> +};
> +
> /* LEDS */
> static struct gpio_led koelsch_leds[] = {
> {
> @@ -71,11 +88,29 @@ static const struct gpio_keys_platform_d
> .nbuttons = ARRAY_SIZE(gpio_buttons),
> };
>
> +static const struct pinctrl_map koelsch_pinctrl_map[] = {
> + /* Ether */
> + PIN_MAP_MUX_GROUP_DEFAULT("r8a779x-ether", "pfc-r8a7791",
> + "eth_link", "eth"),
> + PIN_MAP_MUX_GROUP_DEFAULT("r8a779x-ether", "pfc-r8a7791",
> + "eth_mdio", "eth"),
> + PIN_MAP_MUX_GROUP_DEFAULT("r8a779x-ether", "pfc-r8a7791",
> + "eth_rmii", "eth"),
> +};
> +
> static void __init koelsch_add_standard_devices(void)
> {
> r8a7791_clock_init();
> +
> + pinctrl_register_mappings(koelsch_pinctrl_map,
> + ARRAY_SIZE(koelsch_pinctrl_map));
> r8a7791_pinmux_init();
> +
> r8a7791_add_standard_devices();
> + platform_device_register_resndata(&platform_bus, "r8a779x-ether", -1,
> + ether_resources,
> + ARRAY_SIZE(ether_resources),
> + ðer_pdata, sizeof(ether_pdata));
> platform_device_register_data(&platform_bus, "leds-gpio", -1,
> &koelsch_leds_pdata,
> sizeof(koelsch_leds_pdata));
> @@ -84,6 +119,32 @@ static void __init koelsch_add_standard_
> sizeof(koelsch_keys_pdata));
> }
>
> +/*
> + * Ether LEDs on the Koelsch board are named LINK and ACTIVE which
> corresponds + * to non-default 01 setting of the Micrel KSZ8041 PHY control
> register 1 bits + * 14-15. We have to set them back to 01 from the default
> 00 value each time + * the PHY is reset. It's also important because the
> PHY's LED0 signal is + * connected to SoC's ETH_LINK signal and in the
> PHY's default mode it will + * bounce on and off after each packet, which
> we apparently want to avoid. + */
> +static int koelsch_ksz8041_fixup(struct phy_device *phydev)
> +{
> + u16 phyctrl1 = phy_read(phydev, 0x1e);
> +
> + phyctrl1 &= ~0xc000;
> + phyctrl1 |= 0x4000;
> + return phy_write(phydev, 0x1e, phyctrl1);
> +}
> +
> +static void __init koelsch_init(void)
> +{
> + koelsch_add_standard_devices();
> +
> + if (IS_ENABLED(CONFIG_PHYLIB))
> + phy_register_fixup_for_id("r8a779x-ether-ff:01",
> + koelsch_ksz8041_fixup);
This is fine with board code, but will break when we'll switch to DT. Would it
be difficult to replace board code by using the existing micrel phy driver
(drivers/net/phy/micrel.c) which should support the ksz8041 already and
extending it with support for register 0x1e which doesn't seem to be handled ?
> +}
> +
> static const char * const koelsch_boards_compat_dt[] __initconst = {
> "renesas,koelsch",
> NULL,
> @@ -92,7 +153,7 @@ static const char * const koelsch_boards
> DT_MACHINE_START(KOELSCH_DT, "koelsch")
> .smp = smp_ops(r8a7791_smp_ops),
> .init_early = r8a7791_init_early,
> - .init_machine = koelsch_add_standard_devices,
> + .init_machine = koelsch_init,
> .init_time = rcar_gen2_timer_init,
> .dt_compat = koelsch_boards_compat_dt,
> MACHINE_END
--
Regards,
Laurent Pinchart
^ permalink raw reply [flat|nested] 26+ messages in thread
* [PATCH v2 2/3] ARM: shmobile: Koelsch: add Ether support
@ 2013-11-14 23:06 ` Laurent Pinchart
0 siblings, 0 replies; 26+ messages in thread
From: Laurent Pinchart @ 2013-11-14 23:06 UTC (permalink / raw)
To: linux-arm-kernel
Hi Sergei,
Thank you for the patch.
On Friday 15 November 2013 02:21:11 Sergei Shtylyov wrote:
> Register Ether platform device and pin data on the Koelsch board.
> Register platform fixup for Micrel KSZ8041 PHY, just like on the Lager
> board.
>
> Signed-off-by: Sergei Shtylyov <sergei.shtylyov@cogentembedded.com>
>
> ---
> Changes in version 2:
> - added *if* (IS_ENABLED(CONFIG_PHYLIB)) around phy_register_fixup_for_id()
> call;
> - changed Ether device name to "r8a779x-ether".
>
> arch/arm/mach-shmobile/board-koelsch.c | 63 ++++++++++++++++++++++++++++-
> 1 file changed, 62 insertions(+), 1 deletion(-)
>
> Index: renesas/arch/arm/mach-shmobile/board-koelsch.c
> ===================================================================
> --- renesas.orig/arch/arm/mach-shmobile/board-koelsch.c
> +++ renesas/arch/arm/mach-shmobile/board-koelsch.c
> @@ -24,14 +24,31 @@
> #include <linux/input.h>
> #include <linux/kernel.h>
> #include <linux/leds.h>
> +#include <linux/phy.h>
> +#include <linux/pinctrl/machine.h>
> #include <linux/platform_data/gpio-rcar.h>
> #include <linux/platform_device.h>
> +#include <linux/sh_eth.h>
> #include <mach/common.h>
> +#include <mach/irqs.h>
> #include <mach/r8a7791.h>
> #include <mach/rcar-gen2.h>
> #include <asm/mach-types.h>
> #include <asm/mach/arch.h>
>
> +/* Ether */
> +static const struct sh_eth_plat_data ether_pdata __initconst = {
> + .phy = 0x1,
> + .edmac_endian = EDMAC_LITTLE_ENDIAN,
> + .phy_interface = PHY_INTERFACE_MODE_RMII,
> + .ether_link_active_low = 1,
> +};
> +
> +static const struct resource ether_resources[] __initconst = {
> + DEFINE_RES_MEM(0xee700000, 0x400),
> + DEFINE_RES_IRQ(gic_spi(162)),
> +};
> +
> /* LEDS */
> static struct gpio_led koelsch_leds[] = {
> {
> @@ -71,11 +88,29 @@ static const struct gpio_keys_platform_d
> .nbuttons = ARRAY_SIZE(gpio_buttons),
> };
>
> +static const struct pinctrl_map koelsch_pinctrl_map[] = {
> + /* Ether */
> + PIN_MAP_MUX_GROUP_DEFAULT("r8a779x-ether", "pfc-r8a7791",
> + "eth_link", "eth"),
> + PIN_MAP_MUX_GROUP_DEFAULT("r8a779x-ether", "pfc-r8a7791",
> + "eth_mdio", "eth"),
> + PIN_MAP_MUX_GROUP_DEFAULT("r8a779x-ether", "pfc-r8a7791",
> + "eth_rmii", "eth"),
> +};
> +
> static void __init koelsch_add_standard_devices(void)
> {
> r8a7791_clock_init();
> +
> + pinctrl_register_mappings(koelsch_pinctrl_map,
> + ARRAY_SIZE(koelsch_pinctrl_map));
> r8a7791_pinmux_init();
> +
> r8a7791_add_standard_devices();
> + platform_device_register_resndata(&platform_bus, "r8a779x-ether", -1,
> + ether_resources,
> + ARRAY_SIZE(ether_resources),
> + ðer_pdata, sizeof(ether_pdata));
> platform_device_register_data(&platform_bus, "leds-gpio", -1,
> &koelsch_leds_pdata,
> sizeof(koelsch_leds_pdata));
> @@ -84,6 +119,32 @@ static void __init koelsch_add_standard_
> sizeof(koelsch_keys_pdata));
> }
>
> +/*
> + * Ether LEDs on the Koelsch board are named LINK and ACTIVE which
> corresponds + * to non-default 01 setting of the Micrel KSZ8041 PHY control
> register 1 bits + * 14-15. We have to set them back to 01 from the default
> 00 value each time + * the PHY is reset. It's also important because the
> PHY's LED0 signal is + * connected to SoC's ETH_LINK signal and in the
> PHY's default mode it will + * bounce on and off after each packet, which
> we apparently want to avoid. + */
> +static int koelsch_ksz8041_fixup(struct phy_device *phydev)
> +{
> + u16 phyctrl1 = phy_read(phydev, 0x1e);
> +
> + phyctrl1 &= ~0xc000;
> + phyctrl1 |= 0x4000;
> + return phy_write(phydev, 0x1e, phyctrl1);
> +}
> +
> +static void __init koelsch_init(void)
> +{
> + koelsch_add_standard_devices();
> +
> + if (IS_ENABLED(CONFIG_PHYLIB))
> + phy_register_fixup_for_id("r8a779x-ether-ff:01",
> + koelsch_ksz8041_fixup);
This is fine with board code, but will break when we'll switch to DT. Would it
be difficult to replace board code by using the existing micrel phy driver
(drivers/net/phy/micrel.c) which should support the ksz8041 already and
extending it with support for register 0x1e which doesn't seem to be handled ?
> +}
> +
> static const char * const koelsch_boards_compat_dt[] __initconst = {
> "renesas,koelsch",
> NULL,
> @@ -92,7 +153,7 @@ static const char * const koelsch_boards
> DT_MACHINE_START(KOELSCH_DT, "koelsch")
> .smp = smp_ops(r8a7791_smp_ops),
> .init_early = r8a7791_init_early,
> - .init_machine = koelsch_add_standard_devices,
> + .init_machine = koelsch_init,
> .init_time = rcar_gen2_timer_init,
> .dt_compat = koelsch_boards_compat_dt,
> MACHINE_END
--
Regards,
Laurent Pinchart
^ permalink raw reply [flat|nested] 26+ messages in thread
* [PATCH v2 0/3] Add Ether support for R8A7791/Koelsch board
@ 2013-11-14 23:17 ` Sergei Shtylyov
0 siblings, 0 replies; 26+ messages in thread
From: Sergei Shtylyov @ 2013-11-14 23:17 UTC (permalink / raw)
To: linux-arm-kernel
Hello.
Resending with correct subject...
Here's the set of 3 patches against Simon Horman's 'renesas.git' repo,
'renesas-devel-v3.12-20131112' tag. Here we add the Ether support to the
R8A7791/Koelsch board. The patchset requires the 'sh_eth' driver R8A7791
support patch posted earlier in order to work.
[1/3] ARM: shmobile: r8a7791: add Ether clock
[2/3] ARM: shmobile: Koelsch: add Ether support
[3/3] ARM: shmobile: Koelsch: enable Ether in defconfig
WBR, Sergei
^ permalink raw reply [flat|nested] 26+ messages in thread
* [PATCH v2 1/3] ARM: shmobile: r8a7791: add Ether clock
@ 2013-11-14 23:19 ` Sergei Shtylyov
0 siblings, 0 replies; 26+ messages in thread
From: Sergei Shtylyov @ 2013-11-14 23:19 UTC (permalink / raw)
To: linux-arm-kernel
Add support for R8A7791 Ether clock.
Signed-off-by: Sergei Shtylyov <sergei.shtylyov@cogentembedded.com>
---
Changes in version 2:
- changed Ether device name to "r8a779x-ether".
arch/arm/mach-shmobile/clock-r8a7791.c | 3 +++
1 file changed, 3 insertions(+)
Index: renesas/arch/arm/mach-shmobile/clock-r8a7791.c
===================================================================
--- renesas.orig/arch/arm/mach-shmobile/clock-r8a7791.c
+++ renesas/arch/arm/mach-shmobile/clock-r8a7791.c
@@ -120,6 +120,7 @@ static struct clk *main_clks[] = {
/* MSTP */
enum {
+ MSTP813,
MSTP721, MSTP720,
MSTP719, MSTP718, MSTP715, MSTP714,
MSTP216, MSTP207, MSTP206,
@@ -129,6 +130,7 @@ enum {
};
static struct clk mstp_clks[MSTP_NR] = {
+ [MSTP813] = SH_CLK_MSTP32(&p_clk, SMSTPCR8, 13, 0), /* Ether */
[MSTP721] = SH_CLK_MSTP32(&p_clk, SMSTPCR7, 21, 0), /* SCIF0 */
[MSTP720] = SH_CLK_MSTP32(&p_clk, SMSTPCR7, 20, 0), /* SCIF1 */
[MSTP719] = SH_CLK_MSTP32(&p_clk, SMSTPCR7, 19, 0), /* SCIF2 */
@@ -180,6 +182,7 @@ static struct clk_lookup lookups[] = {
CLKDEV_DEV_ID("sh-sci.13", &mstp_clks[MSTP1106]), /* SCIFA4 */
CLKDEV_DEV_ID("sh-sci.14", &mstp_clks[MSTP1107]), /* SCIFA5 */
CLKDEV_DEV_ID("sh_cmt.0", &mstp_clks[MSTP124]),
+ CLKDEV_DEV_ID("r8a779x-ether", &mstp_clks[MSTP813]), /* Ether */
};
#define R8A7791_CLOCK_ROOT(e, m, p0, p1, p30, p31) \
^ permalink raw reply [flat|nested] 26+ messages in thread
* [PATCH v2 2/3] ARM: shmobile: Koelsch: add Ether support
@ 2013-11-14 23:21 ` Sergei Shtylyov
0 siblings, 0 replies; 26+ messages in thread
From: Sergei Shtylyov @ 2013-11-14 23:21 UTC (permalink / raw)
To: linux-arm-kernel
Register Ether platform device and pin data on the Koelsch board.
Register platform fixup for Micrel KSZ8041 PHY, just like on the Lager board.
Signed-off-by: Sergei Shtylyov <sergei.shtylyov@cogentembedded.com>
---
Changes in version 2:
- added *if* (IS_ENABLED(CONFIG_PHYLIB)) around phy_register_fixup_for_id()
call;
- changed Ether device name to "r8a779x-ether".
arch/arm/mach-shmobile/board-koelsch.c | 63 ++++++++++++++++++++++++++++++++-
1 file changed, 62 insertions(+), 1 deletion(-)
Index: renesas/arch/arm/mach-shmobile/board-koelsch.c
===================================================================
--- renesas.orig/arch/arm/mach-shmobile/board-koelsch.c
+++ renesas/arch/arm/mach-shmobile/board-koelsch.c
@@ -24,14 +24,31 @@
#include <linux/input.h>
#include <linux/kernel.h>
#include <linux/leds.h>
+#include <linux/phy.h>
+#include <linux/pinctrl/machine.h>
#include <linux/platform_data/gpio-rcar.h>
#include <linux/platform_device.h>
+#include <linux/sh_eth.h>
#include <mach/common.h>
+#include <mach/irqs.h>
#include <mach/r8a7791.h>
#include <mach/rcar-gen2.h>
#include <asm/mach-types.h>
#include <asm/mach/arch.h>
+/* Ether */
+static const struct sh_eth_plat_data ether_pdata __initconst = {
+ .phy = 0x1,
+ .edmac_endian = EDMAC_LITTLE_ENDIAN,
+ .phy_interface = PHY_INTERFACE_MODE_RMII,
+ .ether_link_active_low = 1,
+};
+
+static const struct resource ether_resources[] __initconst = {
+ DEFINE_RES_MEM(0xee700000, 0x400),
+ DEFINE_RES_IRQ(gic_spi(162)),
+};
+
/* LEDS */
static struct gpio_led koelsch_leds[] = {
{
@@ -71,11 +88,29 @@ static const struct gpio_keys_platform_d
.nbuttons = ARRAY_SIZE(gpio_buttons),
};
+static const struct pinctrl_map koelsch_pinctrl_map[] = {
+ /* Ether */
+ PIN_MAP_MUX_GROUP_DEFAULT("r8a779x-ether", "pfc-r8a7791",
+ "eth_link", "eth"),
+ PIN_MAP_MUX_GROUP_DEFAULT("r8a779x-ether", "pfc-r8a7791",
+ "eth_mdio", "eth"),
+ PIN_MAP_MUX_GROUP_DEFAULT("r8a779x-ether", "pfc-r8a7791",
+ "eth_rmii", "eth"),
+};
+
static void __init koelsch_add_standard_devices(void)
{
r8a7791_clock_init();
+
+ pinctrl_register_mappings(koelsch_pinctrl_map,
+ ARRAY_SIZE(koelsch_pinctrl_map));
r8a7791_pinmux_init();
+
r8a7791_add_standard_devices();
+ platform_device_register_resndata(&platform_bus, "r8a779x-ether", -1,
+ ether_resources,
+ ARRAY_SIZE(ether_resources),
+ ðer_pdata, sizeof(ether_pdata));
platform_device_register_data(&platform_bus, "leds-gpio", -1,
&koelsch_leds_pdata,
sizeof(koelsch_leds_pdata));
@@ -84,6 +119,32 @@ static void __init koelsch_add_standard_
sizeof(koelsch_keys_pdata));
}
+/*
+ * Ether LEDs on the Koelsch board are named LINK and ACTIVE which corresponds
+ * to non-default 01 setting of the Micrel KSZ8041 PHY control register 1 bits
+ * 14-15. We have to set them back to 01 from the default 00 value each time
+ * the PHY is reset. It's also important because the PHY's LED0 signal is
+ * connected to SoC's ETH_LINK signal and in the PHY's default mode it will
+ * bounce on and off after each packet, which we apparently want to avoid.
+ */
+static int koelsch_ksz8041_fixup(struct phy_device *phydev)
+{
+ u16 phyctrl1 = phy_read(phydev, 0x1e);
+
+ phyctrl1 &= ~0xc000;
+ phyctrl1 |= 0x4000;
+ return phy_write(phydev, 0x1e, phyctrl1);
+}
+
+static void __init koelsch_init(void)
+{
+ koelsch_add_standard_devices();
+
+ if (IS_ENABLED(CONFIG_PHYLIB))
+ phy_register_fixup_for_id("r8a779x-ether-ff:01",
+ koelsch_ksz8041_fixup);
+}
+
static const char * const koelsch_boards_compat_dt[] __initconst = {
"renesas,koelsch",
NULL,
@@ -92,7 +153,7 @@ static const char * const koelsch_boards
DT_MACHINE_START(KOELSCH_DT, "koelsch")
.smp = smp_ops(r8a7791_smp_ops),
.init_early = r8a7791_init_early,
- .init_machine = koelsch_add_standard_devices,
+ .init_machine = koelsch_init,
.init_time = rcar_gen2_timer_init,
.dt_compat = koelsch_boards_compat_dt,
MACHINE_END
^ permalink raw reply [flat|nested] 26+ messages in thread
* [PATCH v2 3/3] ARM: shmobile: Koelsch: enable Ether in defconfig
@ 2013-11-14 23:23 ` Sergei Shtylyov
0 siblings, 0 replies; 26+ messages in thread
From: Sergei Shtylyov @ 2013-11-14 23:23 UTC (permalink / raw)
To: linux-arm-kernel
Enable the Ether driver in 'koelsch_defconfig' along with DHCP autoconfiguration
and NFS root.
Signed-off-by: Sergei Shtylyov <sergei.shtylyov@cogentembedded.com>
---
Changes in version 2:
- new patch in the series.
arch/arm/configs/koelsch_defconfig | 22 ++++++++++++++++++++++
1 file changed, 22 insertions(+)
Index: renesas/arch/arm/configs/koelsch_defconfig
===================================================================
--- renesas.orig/arch/arm/configs/koelsch_defconfig
+++ renesas/arch/arm/configs/koelsch_defconfig
@@ -29,7 +29,27 @@ CONFIG_VFP=y
CONFIG_NEON=y
# CONFIG_CORE_DUMP_DEFAULT_ELF_HEADERS is not set
CONFIG_PM_RUNTIME=y
+CONFIG_NET=y
+CONFIG_INET=y
+CONFIG_IP_PNP=y
+CONFIG_IP_PNP_DHCP=y
CONFIG_UEVENT_HELPER_PATH="/sbin/hotplug"
+CONFIG_NETDEVICES=y
+# CONFIG_NET_VENDOR_ARC is not set
+# CONFIG_NET_CADENCE is not set
+# CONFIG_NET_VENDOR_BROADCOM is not set
+# CONFIG_NET_VENDOR_CIRRUS is not set
+# CONFIG_NET_VENDOR_FARADAY is not set
+# CONFIG_NET_VENDOR_INTEL is not set
+# CONFIG_NET_VENDOR_MARVELL is not set
+# CONFIG_NET_VENDOR_MICREL is not set
+# CONFIG_NET_VENDOR_NATSEMI is not set
+CONFIG_SH_ETH=y
+# CONFIG_NET_VENDOR_SEEQ is not set
+# CONFIG_NET_VENDOR_SMSC is not set
+# CONFIG_NET_VENDOR_STMICRO is not set
+# CONFIG_NET_VENDOR_VIA is not set
+# CONFIG_NET_VENDOR_WIZNET is not set
# CONFIG_INPUT_MOUSEDEV_PSAUX is not set
# CONFIG_INPUT_MOUSE is not set
# CONFIG_LEGACY_PTYS is not set
@@ -49,6 +69,8 @@ CONFIG_LEDS_CLASS=y
CONFIG_TMPFS=y
CONFIG_CONFIGFS_FS=y
# CONFIG_MISC_FILESYSTEMS is not set
+CONFIG_NFS_FS=y
+CONFIG_ROOT_NFS=y
# CONFIG_ENABLE_WARN_DEPRECATED is not set
# CONFIG_ENABLE_MUST_CHECK is not set
# CONFIG_ARM_UNWIND is not set
^ permalink raw reply [flat|nested] 26+ messages in thread
* Re: [PATCH v2 3/3] ARM: shmobile: Koelsch: enable Ether in defconfig
2013-11-14 23:23 ` Sergei Shtylyov
@ 2013-11-15 5:29 ` Simon Horman
-1 siblings, 0 replies; 26+ messages in thread
From: Simon Horman @ 2013-11-15 5:29 UTC (permalink / raw)
To: linux-arm-kernel
On Fri, Nov 15, 2013 at 02:23:40AM +0300, Sergei Shtylyov wrote:
> Enable the Ether driver in 'koelsch_defconfig' along with DHCP autoconfiguration
> and NFS root.
>
> Signed-off-by: Sergei Shtylyov <sergei.shtylyov@cogentembedded.com>
Thanks.
I think it is also appropriate to remove these too.
To be hones their inclusion appears to have been an error on my part.
And if nfsroot is being enabled it seems to be a good opportunity
to get rid of them.
CONFIG_BLK_DEV_INITRD=y
CONFIG_INITRAMFS_SOURCE="../../initramfs/out/rootfs.arm.cpio"
CONFIG_INITRAMFS_COMPRESSION_GZIP=y
>
> ---
> Changes in version 2:
> - new patch in the series.
>
> arch/arm/configs/koelsch_defconfig | 22 ++++++++++++++++++++++
> 1 file changed, 22 insertions(+)
>
> Index: renesas/arch/arm/configs/koelsch_defconfig
> =================================> --- renesas.orig/arch/arm/configs/koelsch_defconfig
> +++ renesas/arch/arm/configs/koelsch_defconfig
> @@ -29,7 +29,27 @@ CONFIG_VFP=y
> CONFIG_NEON=y
> # CONFIG_CORE_DUMP_DEFAULT_ELF_HEADERS is not set
> CONFIG_PM_RUNTIME=y
> +CONFIG_NET=y
> +CONFIG_INET=y
> +CONFIG_IP_PNP=y
> +CONFIG_IP_PNP_DHCP=y
> CONFIG_UEVENT_HELPER_PATH="/sbin/hotplug"
> +CONFIG_NETDEVICES=y
> +# CONFIG_NET_VENDOR_ARC is not set
> +# CONFIG_NET_CADENCE is not set
> +# CONFIG_NET_VENDOR_BROADCOM is not set
> +# CONFIG_NET_VENDOR_CIRRUS is not set
> +# CONFIG_NET_VENDOR_FARADAY is not set
> +# CONFIG_NET_VENDOR_INTEL is not set
> +# CONFIG_NET_VENDOR_MARVELL is not set
> +# CONFIG_NET_VENDOR_MICREL is not set
> +# CONFIG_NET_VENDOR_NATSEMI is not set
> +CONFIG_SH_ETH=y
> +# CONFIG_NET_VENDOR_SEEQ is not set
> +# CONFIG_NET_VENDOR_SMSC is not set
> +# CONFIG_NET_VENDOR_STMICRO is not set
> +# CONFIG_NET_VENDOR_VIA is not set
> +# CONFIG_NET_VENDOR_WIZNET is not set
> # CONFIG_INPUT_MOUSEDEV_PSAUX is not set
> # CONFIG_INPUT_MOUSE is not set
> # CONFIG_LEGACY_PTYS is not set
> @@ -49,6 +69,8 @@ CONFIG_LEDS_CLASS=y
> CONFIG_TMPFS=y
> CONFIG_CONFIGFS_FS=y
> # CONFIG_MISC_FILESYSTEMS is not set
> +CONFIG_NFS_FS=y
> +CONFIG_ROOT_NFS=y
> # CONFIG_ENABLE_WARN_DEPRECATED is not set
> # CONFIG_ENABLE_MUST_CHECK is not set
> # CONFIG_ARM_UNWIND is not set
> --
> To unsubscribe from this list: send the line "unsubscribe linux-sh" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at http://vger.kernel.org/majordomo-info.html
>
^ permalink raw reply [flat|nested] 26+ messages in thread
* [PATCH v2 3/3] ARM: shmobile: Koelsch: enable Ether in defconfig
@ 2013-11-15 5:29 ` Simon Horman
0 siblings, 0 replies; 26+ messages in thread
From: Simon Horman @ 2013-11-15 5:29 UTC (permalink / raw)
To: linux-arm-kernel
On Fri, Nov 15, 2013 at 02:23:40AM +0300, Sergei Shtylyov wrote:
> Enable the Ether driver in 'koelsch_defconfig' along with DHCP autoconfiguration
> and NFS root.
>
> Signed-off-by: Sergei Shtylyov <sergei.shtylyov@cogentembedded.com>
Thanks.
I think it is also appropriate to remove these too.
To be hones their inclusion appears to have been an error on my part.
And if nfsroot is being enabled it seems to be a good opportunity
to get rid of them.
CONFIG_BLK_DEV_INITRD=y
CONFIG_INITRAMFS_SOURCE="../../initramfs/out/rootfs.arm.cpio"
CONFIG_INITRAMFS_COMPRESSION_GZIP=y
>
> ---
> Changes in version 2:
> - new patch in the series.
>
> arch/arm/configs/koelsch_defconfig | 22 ++++++++++++++++++++++
> 1 file changed, 22 insertions(+)
>
> Index: renesas/arch/arm/configs/koelsch_defconfig
> ===================================================================
> --- renesas.orig/arch/arm/configs/koelsch_defconfig
> +++ renesas/arch/arm/configs/koelsch_defconfig
> @@ -29,7 +29,27 @@ CONFIG_VFP=y
> CONFIG_NEON=y
> # CONFIG_CORE_DUMP_DEFAULT_ELF_HEADERS is not set
> CONFIG_PM_RUNTIME=y
> +CONFIG_NET=y
> +CONFIG_INET=y
> +CONFIG_IP_PNP=y
> +CONFIG_IP_PNP_DHCP=y
> CONFIG_UEVENT_HELPER_PATH="/sbin/hotplug"
> +CONFIG_NETDEVICES=y
> +# CONFIG_NET_VENDOR_ARC is not set
> +# CONFIG_NET_CADENCE is not set
> +# CONFIG_NET_VENDOR_BROADCOM is not set
> +# CONFIG_NET_VENDOR_CIRRUS is not set
> +# CONFIG_NET_VENDOR_FARADAY is not set
> +# CONFIG_NET_VENDOR_INTEL is not set
> +# CONFIG_NET_VENDOR_MARVELL is not set
> +# CONFIG_NET_VENDOR_MICREL is not set
> +# CONFIG_NET_VENDOR_NATSEMI is not set
> +CONFIG_SH_ETH=y
> +# CONFIG_NET_VENDOR_SEEQ is not set
> +# CONFIG_NET_VENDOR_SMSC is not set
> +# CONFIG_NET_VENDOR_STMICRO is not set
> +# CONFIG_NET_VENDOR_VIA is not set
> +# CONFIG_NET_VENDOR_WIZNET is not set
> # CONFIG_INPUT_MOUSEDEV_PSAUX is not set
> # CONFIG_INPUT_MOUSE is not set
> # CONFIG_LEGACY_PTYS is not set
> @@ -49,6 +69,8 @@ CONFIG_LEDS_CLASS=y
> CONFIG_TMPFS=y
> CONFIG_CONFIGFS_FS=y
> # CONFIG_MISC_FILESYSTEMS is not set
> +CONFIG_NFS_FS=y
> +CONFIG_ROOT_NFS=y
> # CONFIG_ENABLE_WARN_DEPRECATED is not set
> # CONFIG_ENABLE_MUST_CHECK is not set
> # CONFIG_ARM_UNWIND is not set
> --
> To unsubscribe from this list: send the line "unsubscribe linux-sh" in
> the body of a message to majordomo at vger.kernel.org
> More majordomo info at http://vger.kernel.org/majordomo-info.html
>
^ permalink raw reply [flat|nested] 26+ messages in thread
* Re: [PATCH v2 3/3] ARM: shmobile: Koelsch: enable Ether in defconfig
2013-11-15 5:29 ` Simon Horman
@ 2013-11-15 21:03 ` Sergei Shtylyov
-1 siblings, 0 replies; 26+ messages in thread
From: Sergei Shtylyov @ 2013-11-15 20:03 UTC (permalink / raw)
To: linux-arm-kernel
Hello.
On 11/15/2013 08:29 AM, Simon Horman wrote:
>> Enable the Ether driver in 'koelsch_defconfig' along with DHCP autoconfiguration
>> and NFS root.
>> Signed-off-by: Sergei Shtylyov <sergei.shtylyov@cogentembedded.com>
> Thanks.
> I think it is also appropriate to remove these too.
> To be hones their inclusion appears to have been an error on my part.
> And if nfsroot is being enabled it seems to be a good opportunity
> to get rid of them.
> CONFIG_BLK_DEV_INITRD=y
> CONFIG_INITRAMFS_SOURCE="../../initramfs/out/rootfs.arm.cpio"
> CONFIG_INITRAMFS_COMPRESSION_GZIP=y
Hm, where are you seeing these? Tag renesas-devel-v3.12-20131112 doesn't
contain anything similar...
WBR, Sergei
^ permalink raw reply [flat|nested] 26+ messages in thread
* Re: [PATCH v2 2/3] ARM: shmobile: Koelsch: add Ether support
2013-11-14 23:06 ` Laurent Pinchart
@ 2013-11-15 21:28 ` Sergei Shtylyov
-1 siblings, 0 replies; 26+ messages in thread
From: Sergei Shtylyov @ 2013-11-15 20:28 UTC (permalink / raw)
To: linux-arm-kernel
Hello.
On 11/15/2013 02:06 AM, Laurent Pinchart wrote:
>> Register Ether platform device and pin data on the Koelsch board.
>> Register platform fixup for Micrel KSZ8041 PHY, just like on the Lager
>> board.
>> Signed-off-by: Sergei Shtylyov <sergei.shtylyov@cogentembedded.com>
>> ---
>> Changes in version 2:
>> - added *if* (IS_ENABLED(CONFIG_PHYLIB)) around phy_register_fixup_for_id()
>> call;
>> - changed Ether device name to "r8a779x-ether".
>> arch/arm/mach-shmobile/board-koelsch.c | 63 ++++++++++++++++++++++++++++-
>> 1 file changed, 62 insertions(+), 1 deletion(-)
>> Index: renesas/arch/arm/mach-shmobile/board-koelsch.c
>> =================================>> --- renesas.orig/arch/arm/mach-shmobile/board-koelsch.c
>> +++ renesas/arch/arm/mach-shmobile/board-koelsch.c
[...]
>> @@ -84,6 +119,32 @@ static void __init koelsch_add_standard_
>> sizeof(koelsch_keys_pdata));
>> }
>>
>> +/*
>> + * Ether LEDs on the Koelsch board are named LINK and ACTIVE which
>> corresponds + * to non-default 01 setting of the Micrel KSZ8041 PHY control
>> register 1 bits + * 14-15. We have to set them back to 01 from the default
>> 00 value each time + * the PHY is reset. It's also important because the
>> PHY's LED0 signal is + * connected to SoC's ETH_LINK signal and in the
>> PHY's default mode it will + * bounce on and off after each packet, which
>> we apparently want to avoid. + */
>> +static int koelsch_ksz8041_fixup(struct phy_device *phydev)
>> +{
>> + u16 phyctrl1 = phy_read(phydev, 0x1e);
>> +
>> + phyctrl1 &= ~0xc000;
>> + phyctrl1 |= 0x4000;
>> + return phy_write(phydev, 0x1e, phyctrl1);
>> +}
>> +
>> +static void __init koelsch_init(void)
>> +{
>> + koelsch_add_standard_devices();
>> +
>> + if (IS_ENABLED(CONFIG_PHYLIB))
>> + phy_register_fixup_for_id("r8a779x-ether-ff:01",
>> + koelsch_ksz8041_fixup);
> This is fine with board code, but will break when we'll switch to DT. Would it
> be difficult to replace board code by using the existing micrel phy driver
> (drivers/net/phy/micrel.c) which should support the ksz8041 already and
The first issue here is KSZ8041 on the BOCK-W/Lager/Koelsch boards uses
undocumented PHY ID for some reason, so the current driver doesn't really
support it. :-)
> extending it with support for register 0x1e which doesn't seem to be handled ?
You probably didn't quite understand the purpose of the platform PHY
fixup. It's designed to do the board-specific changes, not the driver-specific
changes. In this case, the setting of the bits 14-15 of the PHY control
register 1 (w/address 0x1E) purely depends on the board schematics and simply
can't be selected by the PHY driver.
It could have been set by the PHY driver iff we would find a way to pass
the platform data to the PHY device (on the automatically probed MDIO bus).
We can also ignore ETH_LINK signal altogether in the 'sh_eth' driver
(because it would bounce on/off on each packet) and stop caring about the LED
function matching to what's designed for the board.
Also we can stop resetting PHY in the 'sh_eth' driver -- that should help
(the platform fixup was mainly added to work around this reset but DaveM
didn't take my simple 'sh_eth' patch), although not with PHY resets done via
phy_mii_ioctl()...
WBR, Sergei
^ permalink raw reply [flat|nested] 26+ messages in thread
* [PATCH v2 3/3] ARM: shmobile: Koelsch: enable Ether in defconfig
@ 2013-11-15 21:03 ` Sergei Shtylyov
0 siblings, 0 replies; 26+ messages in thread
From: Sergei Shtylyov @ 2013-11-15 21:03 UTC (permalink / raw)
To: linux-arm-kernel
Hello.
On 11/15/2013 08:29 AM, Simon Horman wrote:
>> Enable the Ether driver in 'koelsch_defconfig' along with DHCP autoconfiguration
>> and NFS root.
>> Signed-off-by: Sergei Shtylyov <sergei.shtylyov@cogentembedded.com>
> Thanks.
> I think it is also appropriate to remove these too.
> To be hones their inclusion appears to have been an error on my part.
> And if nfsroot is being enabled it seems to be a good opportunity
> to get rid of them.
> CONFIG_BLK_DEV_INITRD=y
> CONFIG_INITRAMFS_SOURCE="../../initramfs/out/rootfs.arm.cpio"
> CONFIG_INITRAMFS_COMPRESSION_GZIP=y
Hm, where are you seeing these? Tag renesas-devel-v3.12-20131112 doesn't
contain anything similar...
WBR, Sergei
^ permalink raw reply [flat|nested] 26+ messages in thread
* [PATCH v2 2/3] ARM: shmobile: Koelsch: add Ether support
@ 2013-11-15 21:28 ` Sergei Shtylyov
0 siblings, 0 replies; 26+ messages in thread
From: Sergei Shtylyov @ 2013-11-15 21:28 UTC (permalink / raw)
To: linux-arm-kernel
Hello.
On 11/15/2013 02:06 AM, Laurent Pinchart wrote:
>> Register Ether platform device and pin data on the Koelsch board.
>> Register platform fixup for Micrel KSZ8041 PHY, just like on the Lager
>> board.
>> Signed-off-by: Sergei Shtylyov <sergei.shtylyov@cogentembedded.com>
>> ---
>> Changes in version 2:
>> - added *if* (IS_ENABLED(CONFIG_PHYLIB)) around phy_register_fixup_for_id()
>> call;
>> - changed Ether device name to "r8a779x-ether".
>> arch/arm/mach-shmobile/board-koelsch.c | 63 ++++++++++++++++++++++++++++-
>> 1 file changed, 62 insertions(+), 1 deletion(-)
>> Index: renesas/arch/arm/mach-shmobile/board-koelsch.c
>> ===================================================================
>> --- renesas.orig/arch/arm/mach-shmobile/board-koelsch.c
>> +++ renesas/arch/arm/mach-shmobile/board-koelsch.c
[...]
>> @@ -84,6 +119,32 @@ static void __init koelsch_add_standard_
>> sizeof(koelsch_keys_pdata));
>> }
>>
>> +/*
>> + * Ether LEDs on the Koelsch board are named LINK and ACTIVE which
>> corresponds + * to non-default 01 setting of the Micrel KSZ8041 PHY control
>> register 1 bits + * 14-15. We have to set them back to 01 from the default
>> 00 value each time + * the PHY is reset. It's also important because the
>> PHY's LED0 signal is + * connected to SoC's ETH_LINK signal and in the
>> PHY's default mode it will + * bounce on and off after each packet, which
>> we apparently want to avoid. + */
>> +static int koelsch_ksz8041_fixup(struct phy_device *phydev)
>> +{
>> + u16 phyctrl1 = phy_read(phydev, 0x1e);
>> +
>> + phyctrl1 &= ~0xc000;
>> + phyctrl1 |= 0x4000;
>> + return phy_write(phydev, 0x1e, phyctrl1);
>> +}
>> +
>> +static void __init koelsch_init(void)
>> +{
>> + koelsch_add_standard_devices();
>> +
>> + if (IS_ENABLED(CONFIG_PHYLIB))
>> + phy_register_fixup_for_id("r8a779x-ether-ff:01",
>> + koelsch_ksz8041_fixup);
> This is fine with board code, but will break when we'll switch to DT. Would it
> be difficult to replace board code by using the existing micrel phy driver
> (drivers/net/phy/micrel.c) which should support the ksz8041 already and
The first issue here is KSZ8041 on the BOCK-W/Lager/Koelsch boards uses
undocumented PHY ID for some reason, so the current driver doesn't really
support it. :-)
> extending it with support for register 0x1e which doesn't seem to be handled ?
You probably didn't quite understand the purpose of the platform PHY
fixup. It's designed to do the board-specific changes, not the driver-specific
changes. In this case, the setting of the bits 14-15 of the PHY control
register 1 (w/address 0x1E) purely depends on the board schematics and simply
can't be selected by the PHY driver.
It could have been set by the PHY driver iff we would find a way to pass
the platform data to the PHY device (on the automatically probed MDIO bus).
We can also ignore ETH_LINK signal altogether in the 'sh_eth' driver
(because it would bounce on/off on each packet) and stop caring about the LED
function matching to what's designed for the board.
Also we can stop resetting PHY in the 'sh_eth' driver -- that should help
(the platform fixup was mainly added to work around this reset but DaveM
didn't take my simple 'sh_eth' patch), although not with PHY resets done via
phy_mii_ioctl()...
WBR, Sergei
^ permalink raw reply [flat|nested] 26+ messages in thread
* Re: [PATCH v2 2/3] ARM: shmobile: Koelsch: add Ether support
2013-11-15 21:28 ` Sergei Shtylyov
@ 2013-11-15 23:03 ` Laurent Pinchart
-1 siblings, 0 replies; 26+ messages in thread
From: Laurent Pinchart @ 2013-11-15 23:03 UTC (permalink / raw)
To: linux-arm-kernel
Hi Sergei,
On Saturday 16 November 2013 00:28:32 Sergei Shtylyov wrote:
> On 11/15/2013 02:06 AM, Laurent Pinchart wrote:
> >> Register Ether platform device and pin data on the Koelsch board.
> >> Register platform fixup for Micrel KSZ8041 PHY, just like on the Lager
> >> board.
> >>
> >> Signed-off-by: Sergei Shtylyov <sergei.shtylyov@cogentembedded.com>
> >>
> >> ---
> >> Changes in version 2:
> >> - added *if* (IS_ENABLED(CONFIG_PHYLIB)) around
> >> phy_register_fixup_for_id()
> >>
> >> call;
> >>
> >> - changed Ether device name to "r8a779x-ether".
> >>
> >> arch/arm/mach-shmobile/board-koelsch.c | 63 ++++++++++++++++++++++++-
> >> 1 file changed, 62 insertions(+), 1 deletion(-)
> >>
> >> Index: renesas/arch/arm/mach-shmobile/board-koelsch.c
> >> =================================> >> --- renesas.orig/arch/arm/mach-shmobile/board-koelsch.c
> >> +++ renesas/arch/arm/mach-shmobile/board-koelsch.c
>
> [...]
>
> >> @@ -84,6 +119,32 @@ static void __init koelsch_add_standard_
> >> sizeof(koelsch_keys_pdata));
> >> }
> >>
> >> +/*
> >> + * Ether LEDs on the Koelsch board are named LINK and ACTIVE which
> >> corresponds
> >> + * to non-default 01 setting of the Micrel KSZ8041 PHY control register
> >> 1 bits
> >> + * 14-15. We have to set them back to 01 from the default 00 value each
> >> time
> >> + * the PHY is reset. It's also important because the PHY's LED0 signal
> >> is
> >> + * connected to SoC's ETH_LINK signal and in the PHY's default mode it
> >> will
> >> + * bounce on and off after each packet, which we apparently want to
> >> avoid.
> >> + */
> >> +static int koelsch_ksz8041_fixup(struct phy_device *phydev)
> >> +{
> >> + u16 phyctrl1 = phy_read(phydev, 0x1e);
> >> +
> >> + phyctrl1 &= ~0xc000;
> >> + phyctrl1 |= 0x4000;
> >> + return phy_write(phydev, 0x1e, phyctrl1);
> >> +}
> >> +
> >> +static void __init koelsch_init(void)
> >> +{
> >> + koelsch_add_standard_devices();
> >> +
> >> + if (IS_ENABLED(CONFIG_PHYLIB))
> >> + phy_register_fixup_for_id("r8a779x-ether-ff:01",
> >> + koelsch_ksz8041_fixup);
> >
> > This is fine with board code, but will break when we'll switch to DT.
> > Would it be difficult to replace board code by using the existing micrel
> > phy driver (drivers/net/phy/micrel.c) which should support the ksz8041
> > already and
>
> The first issue here is KSZ8041 on the BOCK-W/Lager/Koelsch boards uses
> undocumented PHY ID for some reason, so the current driver doesn't really
> support it. :-)
Right. Let's teach the KSZ8041 driver about that then :-)
> > extending it with support for register 0x1e which doesn't seem to be
> > handled ?
>
> You probably didn't quite understand the purpose of the platform PHY fixup.
> It's designed to do the board-specific changes, not the driver-specific
> changes. In this case, the setting of the bits 14-15 of the PHY control
> register 1 (w/address 0x1E) purely depends on the board schematics and
> simply can't be selected by the PHY driver.
>
> It could have been set by the PHY driver iff we would find a way to pass the
> platform data to the PHY device (on the automatically probed MDIO bus).
That seems to me like the real issue we need to solve. The PHY clearly needs
platform (or DT) data, so we need a way to pass it to the driver. Once we get
that it shouldn't be difficult to add LED configuration information to the
platform data.
> We can also ignore ETH_LINK signal altogether in the 'sh_eth' driver
> (because it would bounce on/off on each packet) and stop caring about the
> LED function matching to what's designed for the board.
>
> Also we can stop resetting PHY in the 'sh_eth' driver -- that should help
> (the platform fixup was mainly added to work around this reset but DaveM
> didn't take my simple 'sh_eth' patch), although not with PHY resets done via
> phy_mii_ioctl()...
--
Regards,
Laurent Pinchart
^ permalink raw reply [flat|nested] 26+ messages in thread
* [PATCH v2 2/3] ARM: shmobile: Koelsch: add Ether support
@ 2013-11-15 23:03 ` Laurent Pinchart
0 siblings, 0 replies; 26+ messages in thread
From: Laurent Pinchart @ 2013-11-15 23:03 UTC (permalink / raw)
To: linux-arm-kernel
Hi Sergei,
On Saturday 16 November 2013 00:28:32 Sergei Shtylyov wrote:
> On 11/15/2013 02:06 AM, Laurent Pinchart wrote:
> >> Register Ether platform device and pin data on the Koelsch board.
> >> Register platform fixup for Micrel KSZ8041 PHY, just like on the Lager
> >> board.
> >>
> >> Signed-off-by: Sergei Shtylyov <sergei.shtylyov@cogentembedded.com>
> >>
> >> ---
> >> Changes in version 2:
> >> - added *if* (IS_ENABLED(CONFIG_PHYLIB)) around
> >> phy_register_fixup_for_id()
> >>
> >> call;
> >>
> >> - changed Ether device name to "r8a779x-ether".
> >>
> >> arch/arm/mach-shmobile/board-koelsch.c | 63 ++++++++++++++++++++++++-
> >> 1 file changed, 62 insertions(+), 1 deletion(-)
> >>
> >> Index: renesas/arch/arm/mach-shmobile/board-koelsch.c
> >> ===================================================================
> >> --- renesas.orig/arch/arm/mach-shmobile/board-koelsch.c
> >> +++ renesas/arch/arm/mach-shmobile/board-koelsch.c
>
> [...]
>
> >> @@ -84,6 +119,32 @@ static void __init koelsch_add_standard_
> >> sizeof(koelsch_keys_pdata));
> >> }
> >>
> >> +/*
> >> + * Ether LEDs on the Koelsch board are named LINK and ACTIVE which
> >> corresponds
> >> + * to non-default 01 setting of the Micrel KSZ8041 PHY control register
> >> 1 bits
> >> + * 14-15. We have to set them back to 01 from the default 00 value each
> >> time
> >> + * the PHY is reset. It's also important because the PHY's LED0 signal
> >> is
> >> + * connected to SoC's ETH_LINK signal and in the PHY's default mode it
> >> will
> >> + * bounce on and off after each packet, which we apparently want to
> >> avoid.
> >> + */
> >> +static int koelsch_ksz8041_fixup(struct phy_device *phydev)
> >> +{
> >> + u16 phyctrl1 = phy_read(phydev, 0x1e);
> >> +
> >> + phyctrl1 &= ~0xc000;
> >> + phyctrl1 |= 0x4000;
> >> + return phy_write(phydev, 0x1e, phyctrl1);
> >> +}
> >> +
> >> +static void __init koelsch_init(void)
> >> +{
> >> + koelsch_add_standard_devices();
> >> +
> >> + if (IS_ENABLED(CONFIG_PHYLIB))
> >> + phy_register_fixup_for_id("r8a779x-ether-ff:01",
> >> + koelsch_ksz8041_fixup);
> >
> > This is fine with board code, but will break when we'll switch to DT.
> > Would it be difficult to replace board code by using the existing micrel
> > phy driver (drivers/net/phy/micrel.c) which should support the ksz8041
> > already and
>
> The first issue here is KSZ8041 on the BOCK-W/Lager/Koelsch boards uses
> undocumented PHY ID for some reason, so the current driver doesn't really
> support it. :-)
Right. Let's teach the KSZ8041 driver about that then :-)
> > extending it with support for register 0x1e which doesn't seem to be
> > handled ?
>
> You probably didn't quite understand the purpose of the platform PHY fixup.
> It's designed to do the board-specific changes, not the driver-specific
> changes. In this case, the setting of the bits 14-15 of the PHY control
> register 1 (w/address 0x1E) purely depends on the board schematics and
> simply can't be selected by the PHY driver.
>
> It could have been set by the PHY driver iff we would find a way to pass the
> platform data to the PHY device (on the automatically probed MDIO bus).
That seems to me like the real issue we need to solve. The PHY clearly needs
platform (or DT) data, so we need a way to pass it to the driver. Once we get
that it shouldn't be difficult to add LED configuration information to the
platform data.
> We can also ignore ETH_LINK signal altogether in the 'sh_eth' driver
> (because it would bounce on/off on each packet) and stop caring about the
> LED function matching to what's designed for the board.
>
> Also we can stop resetting PHY in the 'sh_eth' driver -- that should help
> (the platform fixup was mainly added to work around this reset but DaveM
> didn't take my simple 'sh_eth' patch), although not with PHY resets done via
> phy_mii_ioctl()...
--
Regards,
Laurent Pinchart
^ permalink raw reply [flat|nested] 26+ messages in thread
* Re: [PATCH v2 2/3] ARM: shmobile: Koelsch: add Ether support
2013-11-15 23:03 ` Laurent Pinchart
@ 2013-11-16 1:32 ` Sergei Shtylyov
-1 siblings, 0 replies; 26+ messages in thread
From: Sergei Shtylyov @ 2013-11-16 0:32 UTC (permalink / raw)
To: linux-arm-kernel
On 11/16/2013 02:03 AM, Laurent Pinchart wrote:
>>>> Register Ether platform device and pin data on the Koelsch board.
>>>> Register platform fixup for Micrel KSZ8041 PHY, just like on the Lager
>>>> board.
>>>> Signed-off-by: Sergei Shtylyov <sergei.shtylyov@cogentembedded.com>
>>>> ---
>>>> Changes in version 2:
>>>> - added *if* (IS_ENABLED(CONFIG_PHYLIB)) around
>>>> phy_register_fixup_for_id()
>>>> call;
>>>> - changed Ether device name to "r8a779x-ether".
>>>> arch/arm/mach-shmobile/board-koelsch.c | 63 ++++++++++++++++++++++++-
>>>> 1 file changed, 62 insertions(+), 1 deletion(-)
>>>> Index: renesas/arch/arm/mach-shmobile/board-koelsch.c
>>>> =================================>>>> --- renesas.orig/arch/arm/mach-shmobile/board-koelsch.c
>>>> +++ renesas/arch/arm/mach-shmobile/board-koelsch.c
>> [...]
>>>> @@ -84,6 +119,32 @@ static void __init koelsch_add_standard_
>>>> sizeof(koelsch_keys_pdata));
>>>> }
>>>>
>>>> +/*
>>>> + * Ether LEDs on the Koelsch board are named LINK and ACTIVE which
>>>> corresponds
>>>> + * to non-default 01 setting of the Micrel KSZ8041 PHY control register
>>>> 1 bits
>>>> + * 14-15. We have to set them back to 01 from the default 00 value each
>>>> time
>>>> + * the PHY is reset. It's also important because the PHY's LED0 signal
>>>> is
>>>> + * connected to SoC's ETH_LINK signal and in the PHY's default mode it
>>>> will
>>>> + * bounce on and off after each packet, which we apparently want to
>>>> avoid.
>>>> + */
>>>> +static int koelsch_ksz8041_fixup(struct phy_device *phydev)
>>>> +{
>>>> + u16 phyctrl1 = phy_read(phydev, 0x1e);
>>>> +
>>>> + phyctrl1 &= ~0xc000;
>>>> + phyctrl1 |= 0x4000;
>>>> + return phy_write(phydev, 0x1e, phyctrl1);
>>>> +}
>>>> +
>>>> +static void __init koelsch_init(void)
>>>> +{
>>>> + koelsch_add_standard_devices();
>>>> +
>>>> + if (IS_ENABLED(CONFIG_PHYLIB))
>>>> + phy_register_fixup_for_id("r8a779x-ether-ff:01",
>>>> + koelsch_ksz8041_fixup);
>>> This is fine with board code, but will break when we'll switch to DT.
>>> Would it be difficult to replace board code by using the existing micrel
>>> phy driver (drivers/net/phy/micrel.c) which should support the ksz8041
>>> already and
>> The first issue here is KSZ8041 on the BOCK-W/Lager/Koelsch boards uses
>> undocumented PHY ID for some reason, so the current driver doesn't really
>> support it. :-)
> Right. Let's teach the KSZ8041 driver about that then :-)
Well, this seems at least easy. :-)
>>> extending it with support for register 0x1e which doesn't seem to be
>>> handled ?
>> You probably didn't quite understand the purpose of the platform PHY fixup.
>> It's designed to do the board-specific changes, not the driver-specific
>> changes. In this case, the setting of the bits 14-15 of the PHY control
>> register 1 (w/address 0x1E) purely depends on the board schematics and
>> simply can't be selected by the PHY driver.
>> It could have been set by the PHY driver iff we would find a way to pass the
>> platform data to the PHY device (on the automatically probed MDIO bus).
> That seems to me like the real issue we need to solve. The PHY clearly needs
> platform (or DT) data, so we need a way to pass it to the driver. Once we get
> that it shouldn't be difficult to add LED configuration information to the
> platform data.
Well, with the PHY driver being DT-enabled somehow, it wouldn't be
difficult (though how to make phylib's automatic probing coexist with DT
probing is not clear to me ATM). More difficult is to link the platform data
to an automatically probed device; there were attempts on linux-usb to link
the platform data to an USB device but that went nowhere IIRC... anyway, with
the PHY platform fixups already in place we need to care only about the DT
case really...
WBR, Sergei
^ permalink raw reply [flat|nested] 26+ messages in thread
* [PATCH v2 2/3] ARM: shmobile: Koelsch: add Ether support
@ 2013-11-16 1:32 ` Sergei Shtylyov
0 siblings, 0 replies; 26+ messages in thread
From: Sergei Shtylyov @ 2013-11-16 1:32 UTC (permalink / raw)
To: linux-arm-kernel
On 11/16/2013 02:03 AM, Laurent Pinchart wrote:
>>>> Register Ether platform device and pin data on the Koelsch board.
>>>> Register platform fixup for Micrel KSZ8041 PHY, just like on the Lager
>>>> board.
>>>> Signed-off-by: Sergei Shtylyov <sergei.shtylyov@cogentembedded.com>
>>>> ---
>>>> Changes in version 2:
>>>> - added *if* (IS_ENABLED(CONFIG_PHYLIB)) around
>>>> phy_register_fixup_for_id()
>>>> call;
>>>> - changed Ether device name to "r8a779x-ether".
>>>> arch/arm/mach-shmobile/board-koelsch.c | 63 ++++++++++++++++++++++++-
>>>> 1 file changed, 62 insertions(+), 1 deletion(-)
>>>> Index: renesas/arch/arm/mach-shmobile/board-koelsch.c
>>>> ===================================================================
>>>> --- renesas.orig/arch/arm/mach-shmobile/board-koelsch.c
>>>> +++ renesas/arch/arm/mach-shmobile/board-koelsch.c
>> [...]
>>>> @@ -84,6 +119,32 @@ static void __init koelsch_add_standard_
>>>> sizeof(koelsch_keys_pdata));
>>>> }
>>>>
>>>> +/*
>>>> + * Ether LEDs on the Koelsch board are named LINK and ACTIVE which
>>>> corresponds
>>>> + * to non-default 01 setting of the Micrel KSZ8041 PHY control register
>>>> 1 bits
>>>> + * 14-15. We have to set them back to 01 from the default 00 value each
>>>> time
>>>> + * the PHY is reset. It's also important because the PHY's LED0 signal
>>>> is
>>>> + * connected to SoC's ETH_LINK signal and in the PHY's default mode it
>>>> will
>>>> + * bounce on and off after each packet, which we apparently want to
>>>> avoid.
>>>> + */
>>>> +static int koelsch_ksz8041_fixup(struct phy_device *phydev)
>>>> +{
>>>> + u16 phyctrl1 = phy_read(phydev, 0x1e);
>>>> +
>>>> + phyctrl1 &= ~0xc000;
>>>> + phyctrl1 |= 0x4000;
>>>> + return phy_write(phydev, 0x1e, phyctrl1);
>>>> +}
>>>> +
>>>> +static void __init koelsch_init(void)
>>>> +{
>>>> + koelsch_add_standard_devices();
>>>> +
>>>> + if (IS_ENABLED(CONFIG_PHYLIB))
>>>> + phy_register_fixup_for_id("r8a779x-ether-ff:01",
>>>> + koelsch_ksz8041_fixup);
>>> This is fine with board code, but will break when we'll switch to DT.
>>> Would it be difficult to replace board code by using the existing micrel
>>> phy driver (drivers/net/phy/micrel.c) which should support the ksz8041
>>> already and
>> The first issue here is KSZ8041 on the BOCK-W/Lager/Koelsch boards uses
>> undocumented PHY ID for some reason, so the current driver doesn't really
>> support it. :-)
> Right. Let's teach the KSZ8041 driver about that then :-)
Well, this seems at least easy. :-)
>>> extending it with support for register 0x1e which doesn't seem to be
>>> handled ?
>> You probably didn't quite understand the purpose of the platform PHY fixup.
>> It's designed to do the board-specific changes, not the driver-specific
>> changes. In this case, the setting of the bits 14-15 of the PHY control
>> register 1 (w/address 0x1E) purely depends on the board schematics and
>> simply can't be selected by the PHY driver.
>> It could have been set by the PHY driver iff we would find a way to pass the
>> platform data to the PHY device (on the automatically probed MDIO bus).
> That seems to me like the real issue we need to solve. The PHY clearly needs
> platform (or DT) data, so we need a way to pass it to the driver. Once we get
> that it shouldn't be difficult to add LED configuration information to the
> platform data.
Well, with the PHY driver being DT-enabled somehow, it wouldn't be
difficult (though how to make phylib's automatic probing coexist with DT
probing is not clear to me ATM). More difficult is to link the platform data
to an automatically probed device; there were attempts on linux-usb to link
the platform data to an USB device but that went nowhere IIRC... anyway, with
the PHY platform fixups already in place we need to care only about the DT
case really...
WBR, Sergei
^ permalink raw reply [flat|nested] 26+ messages in thread
* Re: [PATCH v2 2/3] ARM: shmobile: Koelsch: add Ether support
2013-11-16 1:32 ` Sergei Shtylyov
@ 2013-11-16 14:11 ` Laurent Pinchart
-1 siblings, 0 replies; 26+ messages in thread
From: Laurent Pinchart @ 2013-11-16 14:11 UTC (permalink / raw)
To: linux-arm-kernel
Hi Sergei,
On Saturday 16 November 2013 04:32:42 Sergei Shtylyov wrote:
> On 11/16/2013 02:03 AM, Laurent Pinchart wrote:
> >>>> Register Ether platform device and pin data on the Koelsch board.
> >>>> Register platform fixup for Micrel KSZ8041 PHY, just like on the Lager
> >>>> board.
> >>>>
> >>>> Signed-off-by: Sergei Shtylyov <sergei.shtylyov@cogentembedded.com>
> >>>>
> >>>> ---
> >>>> Changes in version 2:
> >>>> - added *if* (IS_ENABLED(CONFIG_PHYLIB)) around
> >>>> phy_register_fixup_for_id() call;
> >>>>
> >>>> - changed Ether device name to "r8a779x-ether".
> >>>>
> >>>> arch/arm/mach-shmobile/board-koelsch.c | 63 +++++++++++++++++++++-
> >>>> 1 file changed, 62 insertions(+), 1 deletion(-)
> >>>>
> >>>> Index: renesas/arch/arm/mach-shmobile/board-koelsch.c
> >>>> =================================> >>>> --- renesas.orig/arch/arm/mach-shmobile/board-koelsch.c
> >>>> +++ renesas/arch/arm/mach-shmobile/board-koelsch.c
> >>
> >> [...]
> >>
> >>>> @@ -84,6 +119,32 @@ static void __init koelsch_add_standard_
> >>>> sizeof(koelsch_keys_pdata));
> >>>> }
> >>>>
> >>>> +/*
> >>>> + * Ether LEDs on the Koelsch board are named LINK and ACTIVE which
> >>>> corresponds
> >>>> + * to non-default 01 setting of the Micrel KSZ8041 PHY control
> >>>> register 1 bits
> >>>> + * 14-15. We have to set them back to 01 from the default 00 value
> >>>> each time
> >>>> + * the PHY is reset. It's also important because the PHY's LED0 signal
> >>>> is
> >>>> + * connected to SoC's ETH_LINK signal and in the PHY's default mode it
> >>>> will
> >>>> + * bounce on and off after each packet, which we apparently want to
> >>>> avoid.
> >>>> + */
> >>>> +static int koelsch_ksz8041_fixup(struct phy_device *phydev)
> >>>> +{
> >>>> + u16 phyctrl1 = phy_read(phydev, 0x1e);
> >>>> +
> >>>> + phyctrl1 &= ~0xc000;
> >>>> + phyctrl1 |= 0x4000;
> >>>> + return phy_write(phydev, 0x1e, phyctrl1);
> >>>> +}
> >>>> +
> >>>> +static void __init koelsch_init(void)
> >>>> +{
> >>>> + koelsch_add_standard_devices();
> >>>> +
> >>>> + if (IS_ENABLED(CONFIG_PHYLIB))
> >>>> + phy_register_fixup_for_id("r8a779x-ether-ff:01",
> >>>> + koelsch_ksz8041_fixup);
> >>>
> >>> This is fine with board code, but will break when we'll switch to DT.
> >>> Would it be difficult to replace board code by using the existing micrel
> >>> phy driver (drivers/net/phy/micrel.c) which should support the ksz8041
> >>> already and
> >>
> >> The first issue here is KSZ8041 on the BOCK-W/Lager/Koelsch boards uses
> >> undocumented PHY ID for some reason, so the current driver doesn't really
> >> support it. :-)
> >
> > Right. Let's teach the KSZ8041 driver about that then :-)
>
> Well, this seems at least easy. :-)
>
> >>> extending it with support for register 0x1e which doesn't seem to be
> >>> handled ?
> >>
> >> You probably didn't quite understand the purpose of the platform PHY
> >> fixup. It's designed to do the board-specific changes, not the driver-
> >> specific changes. In this case, the setting of the bits 14-15 of the PHY
> >> control register 1 (w/address 0x1E) purely depends on the board
> >> schematics and simply can't be selected by the PHY driver.
> >>
> >> It could have been set by the PHY driver iff we would find a way to pass
> >> the platform data to the PHY device (on the automatically probed MDIO
> >> bus).
> >
> > That seems to me like the real issue we need to solve. The PHY clearly
> > needs platform (or DT) data, so we need a way to pass it to the driver.
> > Once we get that it shouldn't be difficult to add LED configuration
> > information to the platform data.
>
> Well, with the PHY driver being DT-enabled somehow, it wouldn't be difficult
> (though how to make phylib's automatic probing coexist with DT probing is
> not clear to me ATM). More difficult is to link the platform data to an
> automatically probed device; there were attempts on linux-usb to link the
> platform data to an USB device but that went nowhere IIRC... anyway, with
> the PHY platform fixups already in place we need to care only about the DT
> case really...
I agree, let's focus on the DT case, board code can use platform fixups for
now.
--
Regards,
Laurent Pinchart
^ permalink raw reply [flat|nested] 26+ messages in thread
* [PATCH v2 2/3] ARM: shmobile: Koelsch: add Ether support
@ 2013-11-16 14:11 ` Laurent Pinchart
0 siblings, 0 replies; 26+ messages in thread
From: Laurent Pinchart @ 2013-11-16 14:11 UTC (permalink / raw)
To: linux-arm-kernel
Hi Sergei,
On Saturday 16 November 2013 04:32:42 Sergei Shtylyov wrote:
> On 11/16/2013 02:03 AM, Laurent Pinchart wrote:
> >>>> Register Ether platform device and pin data on the Koelsch board.
> >>>> Register platform fixup for Micrel KSZ8041 PHY, just like on the Lager
> >>>> board.
> >>>>
> >>>> Signed-off-by: Sergei Shtylyov <sergei.shtylyov@cogentembedded.com>
> >>>>
> >>>> ---
> >>>> Changes in version 2:
> >>>> - added *if* (IS_ENABLED(CONFIG_PHYLIB)) around
> >>>> phy_register_fixup_for_id() call;
> >>>>
> >>>> - changed Ether device name to "r8a779x-ether".
> >>>>
> >>>> arch/arm/mach-shmobile/board-koelsch.c | 63 +++++++++++++++++++++-
> >>>> 1 file changed, 62 insertions(+), 1 deletion(-)
> >>>>
> >>>> Index: renesas/arch/arm/mach-shmobile/board-koelsch.c
> >>>> ===================================================================
> >>>> --- renesas.orig/arch/arm/mach-shmobile/board-koelsch.c
> >>>> +++ renesas/arch/arm/mach-shmobile/board-koelsch.c
> >>
> >> [...]
> >>
> >>>> @@ -84,6 +119,32 @@ static void __init koelsch_add_standard_
> >>>> sizeof(koelsch_keys_pdata));
> >>>> }
> >>>>
> >>>> +/*
> >>>> + * Ether LEDs on the Koelsch board are named LINK and ACTIVE which
> >>>> corresponds
> >>>> + * to non-default 01 setting of the Micrel KSZ8041 PHY control
> >>>> register 1 bits
> >>>> + * 14-15. We have to set them back to 01 from the default 00 value
> >>>> each time
> >>>> + * the PHY is reset. It's also important because the PHY's LED0 signal
> >>>> is
> >>>> + * connected to SoC's ETH_LINK signal and in the PHY's default mode it
> >>>> will
> >>>> + * bounce on and off after each packet, which we apparently want to
> >>>> avoid.
> >>>> + */
> >>>> +static int koelsch_ksz8041_fixup(struct phy_device *phydev)
> >>>> +{
> >>>> + u16 phyctrl1 = phy_read(phydev, 0x1e);
> >>>> +
> >>>> + phyctrl1 &= ~0xc000;
> >>>> + phyctrl1 |= 0x4000;
> >>>> + return phy_write(phydev, 0x1e, phyctrl1);
> >>>> +}
> >>>> +
> >>>> +static void __init koelsch_init(void)
> >>>> +{
> >>>> + koelsch_add_standard_devices();
> >>>> +
> >>>> + if (IS_ENABLED(CONFIG_PHYLIB))
> >>>> + phy_register_fixup_for_id("r8a779x-ether-ff:01",
> >>>> + koelsch_ksz8041_fixup);
> >>>
> >>> This is fine with board code, but will break when we'll switch to DT.
> >>> Would it be difficult to replace board code by using the existing micrel
> >>> phy driver (drivers/net/phy/micrel.c) which should support the ksz8041
> >>> already and
> >>
> >> The first issue here is KSZ8041 on the BOCK-W/Lager/Koelsch boards uses
> >> undocumented PHY ID for some reason, so the current driver doesn't really
> >> support it. :-)
> >
> > Right. Let's teach the KSZ8041 driver about that then :-)
>
> Well, this seems at least easy. :-)
>
> >>> extending it with support for register 0x1e which doesn't seem to be
> >>> handled ?
> >>
> >> You probably didn't quite understand the purpose of the platform PHY
> >> fixup. It's designed to do the board-specific changes, not the driver-
> >> specific changes. In this case, the setting of the bits 14-15 of the PHY
> >> control register 1 (w/address 0x1E) purely depends on the board
> >> schematics and simply can't be selected by the PHY driver.
> >>
> >> It could have been set by the PHY driver iff we would find a way to pass
> >> the platform data to the PHY device (on the automatically probed MDIO
> >> bus).
> >
> > That seems to me like the real issue we need to solve. The PHY clearly
> > needs platform (or DT) data, so we need a way to pass it to the driver.
> > Once we get that it shouldn't be difficult to add LED configuration
> > information to the platform data.
>
> Well, with the PHY driver being DT-enabled somehow, it wouldn't be difficult
> (though how to make phylib's automatic probing coexist with DT probing is
> not clear to me ATM). More difficult is to link the platform data to an
> automatically probed device; there were attempts on linux-usb to link the
> platform data to an USB device but that went nowhere IIRC... anyway, with
> the PHY platform fixups already in place we need to care only about the DT
> case really...
I agree, let's focus on the DT case, board code can use platform fixups for
now.
--
Regards,
Laurent Pinchart
^ permalink raw reply [flat|nested] 26+ messages in thread
* Re: [PATCH v2 3/3] ARM: shmobile: Koelsch: enable Ether in defconfig
2013-11-15 21:03 ` Sergei Shtylyov
@ 2013-11-18 7:04 ` Simon Horman
-1 siblings, 0 replies; 26+ messages in thread
From: Simon Horman @ 2013-11-18 7:04 UTC (permalink / raw)
To: linux-arm-kernel
On Sat, Nov 16, 2013 at 12:03:23AM +0300, Sergei Shtylyov wrote:
> Hello.
>
> On 11/15/2013 08:29 AM, Simon Horman wrote:
>
> >>Enable the Ether driver in 'koelsch_defconfig' along with DHCP autoconfiguration
> >>and NFS root.
>
> >>Signed-off-by: Sergei Shtylyov <sergei.shtylyov@cogentembedded.com>
>
> >Thanks.
>
> >I think it is also appropriate to remove these too.
> >To be hones their inclusion appears to have been an error on my part.
> >And if nfsroot is being enabled it seems to be a good opportunity
> >to get rid of them.
>
> >CONFIG_BLK_DEV_INITRD=y
> >CONFIG_INITRAMFS_SOURCE="../../initramfs/out/rootfs.arm.cpio"
> >CONFIG_INITRAMFS_COMPRESSION_GZIP=y
>
> Hm, where are you seeing these? Tag renesas-devel-v3.12-20131112
> doesn't contain anything similar...
Sorry, please ignore the noise.
I must have been looking at a local version by mistake.
^ permalink raw reply [flat|nested] 26+ messages in thread
* [PATCH v2 3/3] ARM: shmobile: Koelsch: enable Ether in defconfig
@ 2013-11-18 7:04 ` Simon Horman
0 siblings, 0 replies; 26+ messages in thread
From: Simon Horman @ 2013-11-18 7:04 UTC (permalink / raw)
To: linux-arm-kernel
On Sat, Nov 16, 2013 at 12:03:23AM +0300, Sergei Shtylyov wrote:
> Hello.
>
> On 11/15/2013 08:29 AM, Simon Horman wrote:
>
> >>Enable the Ether driver in 'koelsch_defconfig' along with DHCP autoconfiguration
> >>and NFS root.
>
> >>Signed-off-by: Sergei Shtylyov <sergei.shtylyov@cogentembedded.com>
>
> >Thanks.
>
> >I think it is also appropriate to remove these too.
> >To be hones their inclusion appears to have been an error on my part.
> >And if nfsroot is being enabled it seems to be a good opportunity
> >to get rid of them.
>
> >CONFIG_BLK_DEV_INITRD=y
> >CONFIG_INITRAMFS_SOURCE="../../initramfs/out/rootfs.arm.cpio"
> >CONFIG_INITRAMFS_COMPRESSION_GZIP=y
>
> Hm, where are you seeing these? Tag renesas-devel-v3.12-20131112
> doesn't contain anything similar...
Sorry, please ignore the noise.
I must have been looking at a local version by mistake.
^ permalink raw reply [flat|nested] 26+ messages in thread
* Re: [PATCH v2 2/3] ARM: shmobile: Koelsch: add Ether support
2013-11-16 14:11 ` Laurent Pinchart
@ 2013-11-18 7:05 ` Simon Horman
-1 siblings, 0 replies; 26+ messages in thread
From: Simon Horman @ 2013-11-18 7:05 UTC (permalink / raw)
To: linux-arm-kernel
On Sat, Nov 16, 2013 at 03:11:23PM +0100, Laurent Pinchart wrote:
> Hi Sergei,
>
> On Saturday 16 November 2013 04:32:42 Sergei Shtylyov wrote:
> > On 11/16/2013 02:03 AM, Laurent Pinchart wrote:
> > >>>> Register Ether platform device and pin data on the Koelsch board.
> > >>>> Register platform fixup for Micrel KSZ8041 PHY, just like on the Lager
> > >>>> board.
> > >>>>
> > >>>> Signed-off-by: Sergei Shtylyov <sergei.shtylyov@cogentembedded.com>
> > >>>>
> > >>>> ---
> > >>>> Changes in version 2:
> > >>>> - added *if* (IS_ENABLED(CONFIG_PHYLIB)) around
> > >>>> phy_register_fixup_for_id() call;
> > >>>>
> > >>>> - changed Ether device name to "r8a779x-ether".
> > >>>>
> > >>>> arch/arm/mach-shmobile/board-koelsch.c | 63 +++++++++++++++++++++-
> > >>>> 1 file changed, 62 insertions(+), 1 deletion(-)
> > >>>>
> > >>>> Index: renesas/arch/arm/mach-shmobile/board-koelsch.c
> > >>>> =================================> > >>>> --- renesas.orig/arch/arm/mach-shmobile/board-koelsch.c
> > >>>> +++ renesas/arch/arm/mach-shmobile/board-koelsch.c
> > >>
> > >> [...]
> > >>
> > >>>> @@ -84,6 +119,32 @@ static void __init koelsch_add_standard_
> > >>>> sizeof(koelsch_keys_pdata));
> > >>>> }
> > >>>>
> > >>>> +/*
> > >>>> + * Ether LEDs on the Koelsch board are named LINK and ACTIVE which
> > >>>> corresponds
> > >>>> + * to non-default 01 setting of the Micrel KSZ8041 PHY control
> > >>>> register 1 bits
> > >>>> + * 14-15. We have to set them back to 01 from the default 00 value
> > >>>> each time
> > >>>> + * the PHY is reset. It's also important because the PHY's LED0 signal
> > >>>> is
> > >>>> + * connected to SoC's ETH_LINK signal and in the PHY's default mode it
> > >>>> will
> > >>>> + * bounce on and off after each packet, which we apparently want to
> > >>>> avoid.
> > >>>> + */
> > >>>> +static int koelsch_ksz8041_fixup(struct phy_device *phydev)
> > >>>> +{
> > >>>> + u16 phyctrl1 = phy_read(phydev, 0x1e);
> > >>>> +
> > >>>> + phyctrl1 &= ~0xc000;
> > >>>> + phyctrl1 |= 0x4000;
> > >>>> + return phy_write(phydev, 0x1e, phyctrl1);
> > >>>> +}
> > >>>> +
> > >>>> +static void __init koelsch_init(void)
> > >>>> +{
> > >>>> + koelsch_add_standard_devices();
> > >>>> +
> > >>>> + if (IS_ENABLED(CONFIG_PHYLIB))
> > >>>> + phy_register_fixup_for_id("r8a779x-ether-ff:01",
> > >>>> + koelsch_ksz8041_fixup);
> > >>>
> > >>> This is fine with board code, but will break when we'll switch to DT.
> > >>> Would it be difficult to replace board code by using the existing micrel
> > >>> phy driver (drivers/net/phy/micrel.c) which should support the ksz8041
> > >>> already and
> > >>
> > >> The first issue here is KSZ8041 on the BOCK-W/Lager/Koelsch boards uses
> > >> undocumented PHY ID for some reason, so the current driver doesn't really
> > >> support it. :-)
> > >
> > > Right. Let's teach the KSZ8041 driver about that then :-)
> >
> > Well, this seems at least easy. :-)
> >
> > >>> extending it with support for register 0x1e which doesn't seem to be
> > >>> handled ?
> > >>
> > >> You probably didn't quite understand the purpose of the platform PHY
> > >> fixup. It's designed to do the board-specific changes, not the driver-
> > >> specific changes. In this case, the setting of the bits 14-15 of the PHY
> > >> control register 1 (w/address 0x1E) purely depends on the board
> > >> schematics and simply can't be selected by the PHY driver.
> > >>
> > >> It could have been set by the PHY driver iff we would find a way to pass
> > >> the platform data to the PHY device (on the automatically probed MDIO
> > >> bus).
> > >
> > > That seems to me like the real issue we need to solve. The PHY clearly
> > > needs platform (or DT) data, so we need a way to pass it to the driver.
> > > Once we get that it shouldn't be difficult to add LED configuration
> > > information to the platform data.
> >
> > Well, with the PHY driver being DT-enabled somehow, it wouldn't be difficult
> > (though how to make phylib's automatic probing coexist with DT probing is
> > not clear to me ATM). More difficult is to link the platform data to an
> > automatically probed device; there were attempts on linux-usb to link the
> > platform data to an USB device but that went nowhere IIRC... anyway, with
> > the PHY platform fixups already in place we need to care only about the DT
> > case really...
>
> I agree, let's focus on the DT case, board code can use platform fixups for
> now.
FWIW, I agree with this approach.
^ permalink raw reply [flat|nested] 26+ messages in thread
* [PATCH v2 2/3] ARM: shmobile: Koelsch: add Ether support
@ 2013-11-18 7:05 ` Simon Horman
0 siblings, 0 replies; 26+ messages in thread
From: Simon Horman @ 2013-11-18 7:05 UTC (permalink / raw)
To: linux-arm-kernel
On Sat, Nov 16, 2013 at 03:11:23PM +0100, Laurent Pinchart wrote:
> Hi Sergei,
>
> On Saturday 16 November 2013 04:32:42 Sergei Shtylyov wrote:
> > On 11/16/2013 02:03 AM, Laurent Pinchart wrote:
> > >>>> Register Ether platform device and pin data on the Koelsch board.
> > >>>> Register platform fixup for Micrel KSZ8041 PHY, just like on the Lager
> > >>>> board.
> > >>>>
> > >>>> Signed-off-by: Sergei Shtylyov <sergei.shtylyov@cogentembedded.com>
> > >>>>
> > >>>> ---
> > >>>> Changes in version 2:
> > >>>> - added *if* (IS_ENABLED(CONFIG_PHYLIB)) around
> > >>>> phy_register_fixup_for_id() call;
> > >>>>
> > >>>> - changed Ether device name to "r8a779x-ether".
> > >>>>
> > >>>> arch/arm/mach-shmobile/board-koelsch.c | 63 +++++++++++++++++++++-
> > >>>> 1 file changed, 62 insertions(+), 1 deletion(-)
> > >>>>
> > >>>> Index: renesas/arch/arm/mach-shmobile/board-koelsch.c
> > >>>> ===================================================================
> > >>>> --- renesas.orig/arch/arm/mach-shmobile/board-koelsch.c
> > >>>> +++ renesas/arch/arm/mach-shmobile/board-koelsch.c
> > >>
> > >> [...]
> > >>
> > >>>> @@ -84,6 +119,32 @@ static void __init koelsch_add_standard_
> > >>>> sizeof(koelsch_keys_pdata));
> > >>>> }
> > >>>>
> > >>>> +/*
> > >>>> + * Ether LEDs on the Koelsch board are named LINK and ACTIVE which
> > >>>> corresponds
> > >>>> + * to non-default 01 setting of the Micrel KSZ8041 PHY control
> > >>>> register 1 bits
> > >>>> + * 14-15. We have to set them back to 01 from the default 00 value
> > >>>> each time
> > >>>> + * the PHY is reset. It's also important because the PHY's LED0 signal
> > >>>> is
> > >>>> + * connected to SoC's ETH_LINK signal and in the PHY's default mode it
> > >>>> will
> > >>>> + * bounce on and off after each packet, which we apparently want to
> > >>>> avoid.
> > >>>> + */
> > >>>> +static int koelsch_ksz8041_fixup(struct phy_device *phydev)
> > >>>> +{
> > >>>> + u16 phyctrl1 = phy_read(phydev, 0x1e);
> > >>>> +
> > >>>> + phyctrl1 &= ~0xc000;
> > >>>> + phyctrl1 |= 0x4000;
> > >>>> + return phy_write(phydev, 0x1e, phyctrl1);
> > >>>> +}
> > >>>> +
> > >>>> +static void __init koelsch_init(void)
> > >>>> +{
> > >>>> + koelsch_add_standard_devices();
> > >>>> +
> > >>>> + if (IS_ENABLED(CONFIG_PHYLIB))
> > >>>> + phy_register_fixup_for_id("r8a779x-ether-ff:01",
> > >>>> + koelsch_ksz8041_fixup);
> > >>>
> > >>> This is fine with board code, but will break when we'll switch to DT.
> > >>> Would it be difficult to replace board code by using the existing micrel
> > >>> phy driver (drivers/net/phy/micrel.c) which should support the ksz8041
> > >>> already and
> > >>
> > >> The first issue here is KSZ8041 on the BOCK-W/Lager/Koelsch boards uses
> > >> undocumented PHY ID for some reason, so the current driver doesn't really
> > >> support it. :-)
> > >
> > > Right. Let's teach the KSZ8041 driver about that then :-)
> >
> > Well, this seems at least easy. :-)
> >
> > >>> extending it with support for register 0x1e which doesn't seem to be
> > >>> handled ?
> > >>
> > >> You probably didn't quite understand the purpose of the platform PHY
> > >> fixup. It's designed to do the board-specific changes, not the driver-
> > >> specific changes. In this case, the setting of the bits 14-15 of the PHY
> > >> control register 1 (w/address 0x1E) purely depends on the board
> > >> schematics and simply can't be selected by the PHY driver.
> > >>
> > >> It could have been set by the PHY driver iff we would find a way to pass
> > >> the platform data to the PHY device (on the automatically probed MDIO
> > >> bus).
> > >
> > > That seems to me like the real issue we need to solve. The PHY clearly
> > > needs platform (or DT) data, so we need a way to pass it to the driver.
> > > Once we get that it shouldn't be difficult to add LED configuration
> > > information to the platform data.
> >
> > Well, with the PHY driver being DT-enabled somehow, it wouldn't be difficult
> > (though how to make phylib's automatic probing coexist with DT probing is
> > not clear to me ATM). More difficult is to link the platform data to an
> > automatically probed device; there were attempts on linux-usb to link the
> > platform data to an USB device but that went nowhere IIRC... anyway, with
> > the PHY platform fixups already in place we need to care only about the DT
> > case really...
>
> I agree, let's focus on the DT case, board code can use platform fixups for
> now.
FWIW, I agree with this approach.
^ permalink raw reply [flat|nested] 26+ messages in thread
end of thread, other threads:[~2013-11-18 7:05 UTC | newest]
Thread overview: 26+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2013-11-14 22:17 [PATCH v2 0/3] Add Ether support for R8A7791/Koelsch board Sergei Shtylyov
2013-11-14 23:17 ` Sergei Shtylyov
2013-11-14 22:19 ` [PATCH v2 1/3] ARM: shmobile: r8a7791: add Ether clock Sergei Shtylyov
2013-11-14 23:19 ` Sergei Shtylyov
2013-11-14 22:21 ` [PATCH v2 2/3] ARM: shmobile: Koelsch: add Ether support Sergei Shtylyov
2013-11-14 23:21 ` Sergei Shtylyov
2013-11-14 23:06 ` Laurent Pinchart
2013-11-14 23:06 ` Laurent Pinchart
2013-11-15 20:28 ` Sergei Shtylyov
2013-11-15 21:28 ` Sergei Shtylyov
2013-11-15 23:03 ` Laurent Pinchart
2013-11-15 23:03 ` Laurent Pinchart
2013-11-16 0:32 ` Sergei Shtylyov
2013-11-16 1:32 ` Sergei Shtylyov
2013-11-16 14:11 ` Laurent Pinchart
2013-11-16 14:11 ` Laurent Pinchart
2013-11-18 7:05 ` Simon Horman
2013-11-18 7:05 ` Simon Horman
2013-11-14 22:23 ` [PATCH v2 3/3] ARM: shmobile: Koelsch: enable Ether in defconfig Sergei Shtylyov
2013-11-14 23:23 ` Sergei Shtylyov
2013-11-15 5:29 ` Simon Horman
2013-11-15 5:29 ` Simon Horman
2013-11-15 20:03 ` Sergei Shtylyov
2013-11-15 21:03 ` Sergei Shtylyov
2013-11-18 7:04 ` Simon Horman
2013-11-18 7:04 ` Simon Horman
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.