tegra2: Use cmd_link_o_target in board Makefiles

[Patch series relating to SD/MMC support on Tegra boards.]

Stephen Warren wrote at Wednesday, October 12, 2011 4:53 PM:

Quoted 1 line

[Patch series relating to SD/MMC support on Tegra boards.]

Does this patch series look good?

Since these are my first U-Boot patches, I'm not sure exactly who would pick these up and apply them (I assume the "ARM custodians", who are CC'd) nor what I should expect in terms of when they'd be applied.

Thanks.

--
nvpublic

-----Original Message-----
From: sjg at google.com [mailto:sjg at google.com] On Behalf Of Simon Glass
Sent: Tuesday, October 18, 2011 11:31 AM
To: Stephen Warren
Cc: albert.u.boot at aribaud.net; afleming at gmail.com; Tom Warren; U-Boot Mailing List; Mike Frysinger
Subject: Re: [PATCH v6 1/5] tegra2: Use cmd_link_o_target in board Makefiles

Hi Tom,

Quoted 8 lines · Stephen Warren

Stephen Warren wrote at Wednesday, October 12, 2011 4:53 PM:

Quoted 1 line

[Patch series relating to SD/MMC support on Tegra boards.]

Does this patch series look good?

Since these are my first U-Boot patches, I'm not sure exactly who would pick these up and apply them (I assume the "ARM custodians", who are CC'd) nor what I should expect in terms of when they'd be applied.

This patch set looks good to me, and is the next series that needs to be applied to move Tegra along. I have based my latest series 'tegra2:
Tidy up boot path' on your series.

Albert are you happy to apply Stephen's patch series?

Regards,
Simon

Quoted 7 lines

Thanks.

--
nvpublic

s/Hi Tom/Hi Stephen/

Quoted 37 lines

-----Original Message-----
From: sjg at google.com [mailto:sjg at google.com] On Behalf Of Simon Glass
Sent: Tuesday, October 18, 2011 11:31 AM
To: Stephen Warren
Cc: albert.u.boot at aribaud.net; afleming at gmail.com; Tom Warren; U-Boot Mailing List; Mike Frysinger
Subject: Re: [PATCH v6 1/5] tegra2: Use cmd_link_o_target in board Makefiles

Hi Tom,

Quoted 8 lines · Stephen Warren

Stephen Warren wrote at Wednesday, October 12, 2011 4:53 PM:

Quoted 1 line

[Patch series relating to SD/MMC support on Tegra boards.]

Does this patch series look good?

Since these are my first U-Boot patches, I'm not sure exactly who would pick these up and apply them (I assume the "ARM custodians", who are CC'd) nor what I should expect in terms of when they'd be applied.

This patch set looks good to me, and is the next series that needs to be applied to move Tegra along. I have based my latest series 'tegra2:
Tidy up boot path' ?on your series.

Albert are you happy to apply Stephen's patch series?

Regards,
Simon

Quoted 7 lines

Thanks.

--
nvpublic

Quoted 1 line

s/Hi Tom/Hi Stephen/

Sorry! And my question was really directed to Albert.

Regards,
Simon

Going up my backlog... for 2/4 and 4/5, as their are touching MMC driver code, I'd like Andy to ack them as the the MMC custodian, then I can apply the whole set (pending the MAINTAINER addition).

Quoted 3 lines · Albert ARIBAUD

Going up my backlog... for 2/4 and 4/5, as their are touching MMC driver code, I'd like Andy to ack them as the the MMC custodian, then I can apply the whole set (pending the MAINTAINER addition).

i don't think this particular patch needs that much acknowledgement. but at any rate, this patch is obsolete now as a newer one has already been merged by Wolfgang.
-mike

This centralizes knowledge of MMC clocking into the MMC driver. This also removes clock setup from the board files, which will simplify later changes that modify the Harmony board to support the correct set of MMC controllers.

Signed-off-by Stephen Warren <[email protected]>

Acked-by Simon Glass <[email protected]>

Tested-by Simon Glass <[email protected]>

---
 board/nvidia/common/board.c |   13 +------------
 drivers/mmc/tegra2_mmc.c    |   12 +++++++++---
 2 files changed, 10 insertions(+), 15 deletions(-)

board/nvidia/common/board.c +1 -12

index d13537d..370a259 100644
@@ -102,16 +102,6 @@ static void pin_mux_uart(void)
 
 #ifdef CONFIG_TEGRA2_MMC
 /*
- * Routine: clock_init_mmc
- * Description: init the PLL and clocks for the SDMMC controllers
- */
-static void clock_init_mmc(void)
-{
-	clock_start_periph_pll(PERIPH_ID_SDMMC4, CLOCK_ID_PERIPH, 20000000);
-	clock_start_periph_pll(PERIPH_ID_SDMMC3, CLOCK_ID_PERIPH, 20000000);
-}
-
-/*
  * Routine: pin_mux_mmc
  * Description: setup the pin muxes/tristate values for the SDMMC(s)
  */

@@ -157,8 +147,7 @@ int board_init(void)
 int board_mmc_init(bd_t *bd)
 {
 	debug("board_mmc_init called\n");
-	/* Enable clocks, muxes, etc. for SDMMC controllers */
-	clock_init_mmc();
+	/* Enable muxes, etc. for SDMMC controllers */
 	pin_mux_mmc();
 	gpio_config_mmc();
 

drivers/mmc/tegra2_mmc.c +9 -4

index 9e741f2..78b1190 100644
@@ -435,14 +435,22 @@ static int mmc_core_init(struct mmc *mmc)
 
 static int tegra2_mmc_initialize(int dev_index, int bus_width)
 {
+	struct mmc_host *host;
 	struct mmc *mmc;
 
 	debug(" mmc_initialize called\n");
 
+	host = &mmc_host[dev_index];
+
+	host->clock = 0;
+	tegra2_get_setup(host, dev_index);
+
+	clock_start_periph_pll(host->mmc_id, CLOCK_ID_PERIPH, 20000000);
+
 	mmc = &mmc_dev[dev_index];
 
 	sprintf(mmc->name, "Tegra2 SD/MMC");
-	mmc->priv = &mmc_host[dev_index];
+	mmc->priv = host;
 	mmc->send_cmd = mmc_send_cmd;
 	mmc->set_ios = mmc_set_ios;
 	mmc->init = mmc_core_init;

@@ -465,8 +473,6 @@ static int tegra2_mmc_initialize(int dev_index, int bus_width)
 	mmc->f_min = 375000;
 	mmc->f_max = 48000000;
 
-	mmc_host[dev_index].clock = 0;
-	tegra2_get_setup(&mmc_host[dev_index], dev_index);
 	mmc_register(mmc);
 
 	return 0;
--

1.7.0.4

Pass the GPIO numbers for power and card detect to tegra2_mmc_init(), and modify that function to perform all required GPIO initialization. This removes the need for board files to perform these operations.

Move board_mmc_getcd() into tegra2_mmc.c now that the driver knows which GPIOs to use.

Update affected call-sites in seaboard.c and harmony.c. Note that this change should make all SD ports work on Harmony, since the required GPIO setup is now being performed.

v4: Fix prototype of tegra2_mmc_init() in board.h to match driver change.
Remove prototype of gpio_config_mmc() from board.h

Signed-off-by Stephen Warren <[email protected]>

Tested-by Simon Glass <[email protected]>

Acked-by Simon Glass <[email protected]>

---
 board/nvidia/common/board.h      |    3 +-
 board/nvidia/harmony/harmony.c   |   27 ++---------------------
 board/nvidia/seaboard/seaboard.c |   33 +----------------------------
 drivers/mmc/tegra2_mmc.c         |   42 ++++++++++++++++++++++++++++++++-----
 drivers/mmc/tegra2_mmc.h         |    4 ++-
 5 files changed, 45 insertions(+), 64 deletions(-)

board/nvidia/common/board.h +1 -2

index 344e702..35acbca 100644
@@ -26,7 +26,6 @@
 
 void tegra2_start(void);
 void gpio_config_uart(void);
-void gpio_config_mmc(void);
-int tegra2_mmc_init(int dev_index, int bus_width);
+int tegra2_mmc_init(int dev_index, int bus_width, int pwr_gpio, int cd_gpio);
 
 #endif	/* BOARD_H */

board/nvidia/harmony/harmony.c +3 -24

index f2c3867..3cbe820 100644
@@ -25,6 +25,7 @@
 #include <asm/io.h>
 #include <asm/arch/tegra2.h>
 #include <asm/arch/pinmux.h>
+#include <asm/gpio.h>
 #ifdef CONFIG_TEGRA2_MMC
 #include <mmc.h>
 #endif

@@ -72,15 +73,6 @@ static void pin_mux_mmc(void)
 	pinmux_tristate_disable(PINGRP_ATC);
 }
 
-/*
- * Routine: gpio_config_mmc
- * Description: Set GPIOs for SD card
- */
-void gpio_config_mmc(void)
-{
-	/* Not implemented for now */
-}
-
 /* this is a weak define that we are overriding */
 int board_mmc_init(bd_t *bd)
 {

@@ -88,29 +80,16 @@ int board_mmc_init(bd_t *bd)
 
 	/* Enable muxes, etc. for SDMMC controllers */
 	pin_mux_mmc();
-	gpio_config_mmc();
 
 	debug("board_mmc_init: init SD slot J26\n");
 	/* init dev 0, SD slot J26, with 4-bit bus */
 	/* The board has an 8-bit bus, but 8-bit doesn't work yet */
-	tegra2_mmc_init(0, 4);
+	tegra2_mmc_init(0, 4, GPIO_PI6, GPIO_PH2);
 
 	debug("board_mmc_init: init SD slot J5\n");
 	/* init dev 2, SD slot J5, with 4-bit bus */
-	tegra2_mmc_init(2, 4);
+	tegra2_mmc_init(2, 4, GPIO_PT3, GPIO_PI5);
 
 	return 0;
 }
-
-/* this is a weak define that we are overriding */
-int board_mmc_getcd(u8 *cd, struct mmc *mmc)
-{
-	debug("board_mmc_getcd called\n");
-	/*
-	 * Hard-code CD presence for now. Need to add GPIO inputs
-	 * for Harmony
-	 */
-	*cd = 1;
-	return 0;
-}
 #endif

board/nvidia/seaboard/seaboard.c +2 -31

index 22a0e69..356d616 100644
@@ -87,19 +87,6 @@ static void pin_mux_mmc(void)
 	pinmux_tristate_disable(PINGRP_ATC);
 }
 
-/*
- * Routine: gpio_config_mmc
- * Description: Set GPIOs for SDMMC3 SDIO slot.
- */
-void gpio_config_mmc(void)
-{
-	/* Set EN_VDDIO_SD (GPIO I6) */
-	gpio_direction_output(GPIO_PI6, 1);
-
-	/* Config pin as GPI for Card Detect (GPIO I5) */
-	gpio_direction_input(GPIO_PI5);
-}
-
 /* this is a weak define that we are overriding */
 int board_mmc_init(bd_t *bd)
 {

@@ -107,31 +94,15 @@ int board_mmc_init(bd_t *bd)
 
 	/* Enable muxes, etc. for SDMMC controllers */
 	pin_mux_mmc();
-	gpio_config_mmc();
 
 	debug("board_mmc_init: init eMMC\n");
 	/* init dev 0, eMMC chip, with 4-bit bus */
 	/* The board has an 8-bit bus, but 8-bit doesn't work yet */
-	tegra2_mmc_init(0, 4);
+	tegra2_mmc_init(0, 4, -1, -1);
 
 	debug("board_mmc_init: init SD slot\n");
 	/* init dev 1, SD slot, with 4-bit bus */
-	tegra2_mmc_init(1, 4);
-
-	return 0;
-}
-
-/* this is a weak define that we are overriding */
-int board_mmc_getcd(u8 *cd, struct mmc *mmc)
-{
-	debug("board_mmc_getcd called\n");
-	*cd = 1;			/* Assume card is inserted, or eMMC */
-
-	if (IS_SD(mmc)) {
-		/* Seaboard SDMMC3 = SDIO3_CD = GPIO_PI5 */
-		if (gpio_get_value(GPIO_PI5))
-			*cd = 0;
-	}
+	tegra2_mmc_init(1, 4, GPIO_PI6, GPIO_PI5);
 
 	return 0;
 }

drivers/mmc/tegra2_mmc.c +36 -6

index 78b1190..3de9c5d 100644
@@ -21,6 +21,7 @@
 
 #include <common.h>
 #include <mmc.h>
+#include <asm/gpio.h>
 #include <asm/io.h>
 #include <asm/arch/clk_rst.h>
 #include <asm/arch/clock.h>

@@ -433,20 +434,37 @@ static int mmc_core_init(struct mmc *mmc)
 	return 0;
 }
 
-static int tegra2_mmc_initialize(int dev_index, int bus_width)
+int tegra2_mmc_init(int dev_index, int bus_width, int pwr_gpio, int cd_gpio)
 {
 	struct mmc_host *host;
+	char gpusage[12]; /* "SD/MMCn PWR" or "SD/MMCn CD" */
 	struct mmc *mmc;
 
-	debug(" mmc_initialize called\n");
+	debug(" tegra2_mmc_init: index %d, bus width %d "
+		"pwr_gpio %d cd_gpio %d\n",
+		dev_index, bus_width, pwr_gpio, cd_gpio);
 
 	host = &mmc_host[dev_index];
 
 	host->clock = 0;
+	host->pwr_gpio = pwr_gpio;
+	host->cd_gpio = cd_gpio;
 	tegra2_get_setup(host, dev_index);
 
 	clock_start_periph_pll(host->mmc_id, CLOCK_ID_PERIPH, 20000000);
 
+	if (host->pwr_gpio >= 0) {
+		sprintf(gpusage, "SD/MMC%d PWR", dev_index);
+		gpio_request(host->pwr_gpio, gpusage);
+		gpio_direction_output(host->pwr_gpio, 1);
+	}
+
+	if (host->cd_gpio >= 0) {
+		sprintf(gpusage, "SD/MMC%d CD", dev_index);
+		gpio_request(host->cd_gpio, gpusage);
+		gpio_direction_input(host->cd_gpio);
+	}
+
 	mmc = &mmc_dev[dev_index];
 
 	sprintf(mmc->name, "Tegra2 SD/MMC");

@@ -478,9 +496,21 @@ static int tegra2_mmc_initialize(int dev_index, int bus_width)
 	return 0;
 }
 
-int tegra2_mmc_init(int dev_index, int bus_width)
+/* this is a weak define that we are overriding */
+int board_mmc_getcd(u8 *cd, struct mmc *mmc)
 {
-	debug(" tegra2_mmc_init: index %d, bus width %d\n",
-		dev_index, bus_width);
-	return tegra2_mmc_initialize(dev_index, bus_width);
+	struct mmc_host *host = (struct mmc_host *)mmc->priv;
+
+	debug("board_mmc_getcd called\n");
+
+	*cd = 1; /* Assume card is inserted, or eMMC */
+
+	if (IS_SD(mmc)) {
+		if (host->cd_gpio >= 0) {
+			if (gpio_get_value(host->cd_gpio))
+				*cd = 0;
+		}
+	}
+
+	return 0;
 }

drivers/mmc/tegra2_mmc.h +3 -2

index 28698e0..72f587b 100644
@@ -74,9 +74,11 @@ struct mmc_host {
 	unsigned int clock;	/* Current clock (MHz) */
 	unsigned int base;	/* Base address, SDMMC1/2/3/4 */
 	enum periph_id mmc_id;	/* Peripheral ID: PERIPH_ID_... */
+	int pwr_gpio;		/* Power GPIO */
+	int cd_gpio;		/* Change Detect GPIO */
 };
 
-int tegra2_mmc_init(int dev_index, int bus_width);
+int tegra2_mmc_init(int dev_index, int bus_width, int pwr_gpio, int cd_gpio);
 
 #endif	/* __ASSEMBLY__ */
 #endif	/* __TEGRA2_MMC_H_ */
--

1.7.0.4

Ventana is a board which is very similar to Seaboard. Support it by re-using board/nvidia/seaboard/seaboard.c with minor run-time conditionals.

v5: Makefile: Use cmd_link_o_target, remove unused clean/distclean targets.
v6: Make gpio_config_uart_seaboard() static.

Signed-off-by Stephen Warren <[email protected]>

---

?board/nvidia/seaboard/seaboard.c | ? 11 ++++++-
?board/nvidia/ventana/Makefile ? ?| ? 49 +++++++++++++++++++++++++++++++++
?boards.cfg ? ? ? ? ? ? ? ? ? ? ? | ? ?1 +
?include/configs/ventana.h ? ? ? ?| ? 55 ++++++++++++++++++++++++++++++++++++++

?4 files changed, 114 insertions(+), 2 deletions(-) ?create mode 100644 board/nvidia/ventana/Makefile ?create mode 100644 include/configs/ventana.h

Quoted 7 lines

Ventana is a board which is very similar to Seaboard. Support it by re-using board/nvidia/seaboard/seaboard.c with minor run-time conditionals.

v5: Makefile: Use cmd_link_o_target, remove unused clean/distclean targets.
v6: Make gpio_config_uart_seaboard() static.

Signed-off-by Stephen Warren<[email protected]>

Acked-by Simon Glass<[email protected]>

Quoted 8 lines

---
  board/nvidia/seaboard/seaboard.c |   11 ++++++-
  board/nvidia/ventana/Makefile    |   49 +++++++++++++++++++++++++++++++++
  boards.cfg                       |    1 +
  include/configs/ventana.h        |   55 ++++++++++++++++++++++++++++++++++++++
  4 files changed, 114 insertions(+), 2 deletions(-)

create mode 100644 board/nvidia/ventana/Makefile create mode 100644 include/configs/ventana.h