* [U-Boot] [PATCH v2 1/4] tegra2: Seaboard's MMC has an 8-bit bus
@ 2011-10-04 17:32 Stephen Warren
2011-10-04 17:32 ` [U-Boot] [PATCH v2 2/4] tegra2: Call gpio_request for SDMMC GPIOs Stephen Warren
` (3 more replies)
0 siblings, 4 replies; 12+ messages in thread
From: Stephen Warren @ 2011-10-04 17:32 UTC (permalink / raw)
To: u-boot
Modify board_mmc_init to configure SDMMC4 for an 8-bit bus, since that's
what is wired up on the board.
Signed-off-by: Stephen Warren <swarren@nvidia.com>
---
Note: These patches build on various Tegra2-related patches recently posted
by Simon Glass; see v2 of the 6-entry patchset shown at:
http://patchwork.ozlabs.org/project/uboot/list/?submitter=6170&state=1&q=tegra2
board/nvidia/common/board.c | 4 ++--
1 files changed, 2 insertions(+), 2 deletions(-)
diff --git a/board/nvidia/common/board.c b/board/nvidia/common/board.c
index d13537d..8033612 100644
--- a/board/nvidia/common/board.c
+++ b/board/nvidia/common/board.c
@@ -163,8 +163,8 @@ int board_mmc_init(bd_t *bd)
gpio_config_mmc();
debug("board_mmc_init: init eMMC\n");
- /* init dev 0, eMMC chip, with 4-bit bus */
- tegra2_mmc_init(0, 4);
+ /* init dev 0, eMMC chip, with 8-bit bus */
+ tegra2_mmc_init(0, 8);
debug("board_mmc_init: init SD slot\n");
/* init dev 1, SD slot, with 4-bit bus */
--
1.7.0.4
^ permalink raw reply related [flat|nested] 12+ messages in thread* [U-Boot] [PATCH v2 2/4] tegra2: Call gpio_request for SDMMC GPIOs 2011-10-04 17:32 [U-Boot] [PATCH v2 1/4] tegra2: Seaboard's MMC has an 8-bit bus Stephen Warren @ 2011-10-04 17:32 ` Stephen Warren 2011-10-04 23:56 ` Simon Glass 2011-10-04 17:32 ` [U-Boot] [PATCH v2 3/4] tegra2: Enable second SD port (J5) on Harmony Stephen Warren ` (2 subsequent siblings) 3 siblings, 1 reply; 12+ messages in thread From: Stephen Warren @ 2011-10-04 17:32 UTC (permalink / raw) To: u-boot Without this, the GPIO_CNF register will not be programmed, and hence the GPIO signals will not reach the pins; the pinmux's configured function will be routed to the pins instead. v2: s/SDMMC4/SDMMC3/ in strings passed to gpio_request(). Signed-off-by: Stephen Warren <swarren@nvidia.com> --- board/nvidia/seaboard/seaboard.c | 2 ++ 1 files changed, 2 insertions(+), 0 deletions(-) diff --git a/board/nvidia/seaboard/seaboard.c b/board/nvidia/seaboard/seaboard.c index bc67d0f..1310e3d 100644 --- a/board/nvidia/seaboard/seaboard.c +++ b/board/nvidia/seaboard/seaboard.c @@ -62,9 +62,11 @@ void gpio_config_uart(void) void gpio_config_mmc(void) { /* Set EN_VDDIO_SD (GPIO I6) */ + gpio_request(GPIO_PI6, "SDMMC3 power"); gpio_direction_output(GPIO_PI6, 1); /* Config pin as GPI for Card Detect (GPIO I5) */ + gpio_request(GPIO_PI5, "SDMMC3 card detect"); gpio_direction_input(GPIO_PI5); } -- 1.7.0.4 ^ permalink raw reply related [flat|nested] 12+ messages in thread
* [U-Boot] [PATCH v2 2/4] tegra2: Call gpio_request for SDMMC GPIOs 2011-10-04 17:32 ` [U-Boot] [PATCH v2 2/4] tegra2: Call gpio_request for SDMMC GPIOs Stephen Warren @ 2011-10-04 23:56 ` Simon Glass 0 siblings, 0 replies; 12+ messages in thread From: Simon Glass @ 2011-10-04 23:56 UTC (permalink / raw) To: u-boot On Tue, Oct 4, 2011 at 10:32 AM, Stephen Warren <swarren@nvidia.com> wrote: > Without this, the GPIO_CNF register will not be programmed, and hence the > GPIO signals will not reach the pins; the pinmux's configured function > will be routed to the pins instead. > > v2: s/SDMMC4/SDMMC3/ in strings passed to gpio_request(). > > Signed-off-by: Stephen Warren <swarren@nvidia.com> Acked-by: Simon Glass <sjg@chromium.org> > --- > ?board/nvidia/seaboard/seaboard.c | ? ?2 ++ > ?1 files changed, 2 insertions(+), 0 deletions(-) > > diff --git a/board/nvidia/seaboard/seaboard.c b/board/nvidia/seaboard/seaboard.c > index bc67d0f..1310e3d 100644 > --- a/board/nvidia/seaboard/seaboard.c > +++ b/board/nvidia/seaboard/seaboard.c > @@ -62,9 +62,11 @@ void gpio_config_uart(void) > ?void gpio_config_mmc(void) > ?{ > ? ? ? ?/* Set EN_VDDIO_SD (GPIO I6) */ > + ? ? ? gpio_request(GPIO_PI6, "SDMMC3 power"); > ? ? ? ?gpio_direction_output(GPIO_PI6, 1); > > ? ? ? ?/* Config pin as GPI for Card Detect (GPIO I5) */ > + ? ? ? gpio_request(GPIO_PI5, "SDMMC3 card detect"); > ? ? ? ?gpio_direction_input(GPIO_PI5); > ?} > > -- > 1.7.0.4 > > ^ permalink raw reply [flat|nested] 12+ messages in thread
* [U-Boot] [PATCH v2 3/4] tegra2: Enable second SD port (J5) on Harmony 2011-10-04 17:32 [U-Boot] [PATCH v2 1/4] tegra2: Seaboard's MMC has an 8-bit bus Stephen Warren 2011-10-04 17:32 ` [U-Boot] [PATCH v2 2/4] tegra2: Call gpio_request for SDMMC GPIOs Stephen Warren @ 2011-10-04 17:32 ` Stephen Warren 2011-10-05 14:39 ` Simon Glass 2011-11-03 0:16 ` Andy Fleming 2011-10-04 17:32 ` [U-Boot] [PATCH v2 4/4] tegra2: Add support for Ventana Stephen Warren 2011-10-04 20:52 ` [U-Boot] [PATCH v2 1/4] tegra2: Seaboard's MMC has an 8-bit bus Tom Warren 3 siblings, 2 replies; 12+ messages in thread From: Stephen Warren @ 2011-10-04 17:32 UTC (permalink / raw) To: u-boot Seaboard uses SDMMC4, SDMMC3. Harmony uses SDMMC4, SDMMC2. Move board_init_mmc and gpio_config_mmc into board-specific files, so boards can choose which ports to set up. Split clock_init_mmc and pin_mux_mmc into per-port functions, and export them for use by boards. Signed-off-by: Stephen Warren <swarren@nvidia.com> --- board/nvidia/common/board.c | 77 ++++++++++++++++++++++++-------------- board/nvidia/common/board.h | 8 +++- board/nvidia/harmony/harmony.c | 56 +++++++++++++++++++++++++-- board/nvidia/seaboard/seaboard.c | 23 +++++++++++ drivers/mmc/tegra2_mmc.c | 18 +++++++++ 5 files changed, 148 insertions(+), 34 deletions(-) diff --git a/board/nvidia/common/board.c b/board/nvidia/common/board.c index 8033612..6c09a4c 100644 --- a/board/nvidia/common/board.c +++ b/board/nvidia/common/board.c @@ -102,20 +102,37 @@ static void pin_mux_uart(void) #ifdef CONFIG_TEGRA2_MMC /* - * Routine: clock_init_mmc - * Description: init the PLL and clocks for the SDMMC controllers + * Routine: clock_init_mmc4 + * Description: init the PLL and clocks for SDMMC4 controller */ -static void clock_init_mmc(void) +void clock_init_mmc4(void) { clock_start_periph_pll(PERIPH_ID_SDMMC4, CLOCK_ID_PERIPH, 20000000); +} + +/* + * Routine: clock_init_mmc3 + * Description: init the PLL and clocks for SDMMC3 controller + */ +void clock_init_mmc3(void) +{ 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) + * Routine: clock_init_mmc2 + * Description: init the PLL and clocks for SDMMC2 controller */ -static void pin_mux_mmc(void) +void clock_init_mmc2(void) +{ + clock_start_periph_pll(PERIPH_ID_SDMMC2, CLOCK_ID_PERIPH, 20000000); +} + +/* + * Routine: pin_mux_mmc4 + * Description: setup the pin muxes/tristate values for SDMMC4 + */ +void pin_mux_mmc4(void) { /* SDMMC4: config 3, x8 on 2nd set of pins */ pinmux_set_func(PINGRP_ATB, PMUX_FUNC_SDIO4); @@ -125,7 +142,14 @@ static void pin_mux_mmc(void) pinmux_tristate_disable(PINGRP_ATB); pinmux_tristate_disable(PINGRP_GMA); pinmux_tristate_disable(PINGRP_GME); +} +/* + * Routine: pin_mux_mmc3 + * Description: setup the pin muxes/tristate values for SDMMC3 + */ +void pin_mux_mmc3(void) +{ /* SDMMC3: SDIO3_CLK, SDIO3_CMD, SDIO3_DAT[3:0] */ pinmux_set_func(PINGRP_SDB, PMUX_FUNC_SDIO3); pinmux_set_func(PINGRP_SDC, PMUX_FUNC_SDIO3); @@ -135,6 +159,25 @@ static void pin_mux_mmc(void) pinmux_tristate_disable(PINGRP_SDD); pinmux_tristate_disable(PINGRP_SDB); } + +/* + * Routine: pin_mux_mmc2 + * Description: setup the pin muxes/tristate values for SDMMC2 + */ +void pin_mux_mmc2(void) +{ + /* SDMMC2: SDIO2_CLK, SDIO2_CMD, SDIO2_DAT[7:0] */ + pinmux_set_func(PINGRP_DTA, PMUX_FUNC_SDIO2); + pinmux_set_func(PINGRP_DTD, PMUX_FUNC_SDIO2); + + pinmux_tristate_disable(PINGRP_DTA); + pinmux_tristate_disable(PINGRP_DTD); + + /* For power GPIO PI6 */ + pinmux_tristate_disable(PINGRP_ATA); + /* For power GPIO PT3 */ + pinmux_tristate_disable(PINGRP_DTB); +} #endif /* @@ -152,28 +195,6 @@ int board_init(void) return 0; } -#ifdef CONFIG_TEGRA2_MMC -/* this is a weak define that we are overriding */ -int board_mmc_init(bd_t *bd) -{ - debug("board_mmc_init called\n"); - /* Enable clocks, muxes, etc. for SDMMC controllers */ - clock_init_mmc(); - pin_mux_mmc(); - gpio_config_mmc(); - - debug("board_mmc_init: init eMMC\n"); - /* init dev 0, eMMC chip, with 8-bit bus */ - tegra2_mmc_init(0, 8); - - debug("board_mmc_init: init SD slot\n"); - /* init dev 1, SD slot, with 4-bit bus */ - tegra2_mmc_init(1, 4); - - return 0; -} -#endif - #ifdef CONFIG_BOARD_EARLY_INIT_F int board_early_init_f(void) { diff --git a/board/nvidia/common/board.h b/board/nvidia/common/board.h index 344e702..eee475d 100644 --- a/board/nvidia/common/board.h +++ b/board/nvidia/common/board.h @@ -26,7 +26,13 @@ void tegra2_start(void); void gpio_config_uart(void); -void gpio_config_mmc(void); +void clock_init_mmc4(void); +void clock_init_mmc3(void); +void clock_init_mmc2(void); +void pin_mux_mmc4(void); +void pin_mux_mmc3(void); +void pin_mux_mmc2(void); int tegra2_mmc_init(int dev_index, int bus_width); +int tegra2_mmc_index(struct mmc *mmc); #endif /* BOARD_H */ diff --git a/board/nvidia/harmony/harmony.c b/board/nvidia/harmony/harmony.c index cbb30d6..228ae2e 100644 --- a/board/nvidia/harmony/harmony.c +++ b/board/nvidia/harmony/harmony.c @@ -23,10 +23,13 @@ #include <common.h> #include <asm/io.h> +#include <asm/arch/clock.h> #include <asm/arch/tegra2.h> +#include <asm/gpio.h> #ifdef CONFIG_TEGRA2_MMC #include <mmc.h> #endif +#include "../common/board.h" /* * Routine: gpio_config_uart @@ -43,18 +46,61 @@ void gpio_config_uart(void) */ void gpio_config_mmc(void) { - /* Not implemented for now */ + /* Set SDMMC4 power (GPIO I6) */ + gpio_request(GPIO_PI6, "SDMMC4 power"); + gpio_direction_output(GPIO_PI6, 1); + + /* Config pin as GPI for SDMMC4 Card Detect (GPIO H2) */ + gpio_request(GPIO_PH2, "SDMMC4 card detect"); + gpio_direction_input(GPIO_PH2); + + /* Set SDMMC2 power (GPIO T3) */ + gpio_request(GPIO_PT3, "SDMMC2 power"); + gpio_direction_output(GPIO_PT3, 1); + + /* Config pin as GPI for SDMMC2 Card Detect (GPIO I5) */ + gpio_request(GPIO_PI5, "SDMMC2 card detect"); + gpio_direction_input(GPIO_PI5); +} + +/* this is a weak define that we are overriding */ +int board_mmc_init(bd_t *bd) +{ + debug("board_mmc_init called\n"); + /* Enable clocks, muxes, etc. for SDMMC controllers */ + clock_init_mmc4(); + clock_init_mmc2(); + pin_mux_mmc4(); + pin_mux_mmc2(); + gpio_config_mmc(); + + debug("board_mmc_init: init SD slot J26\n"); + /* init dev 0, SD slot J26, with 8-bit bus */ + tegra2_mmc_init(0, 8); + + debug("board_mmc_init: init SD slot J5\n"); + /* init dev 2, SD slot J5, with 8-bit bus */ + tegra2_mmc_init(2, 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"); - /* - * Hard-code CD presence for now. Need to add GPIO inputs - * for Harmony - */ *cd = 1; + + if (tegra2_mmc_index(mmc) == 0) { + /* Harmony SDMMC4 = SDIO4_CD = GPIO_PH2 */ + if (gpio_get_value(GPIO_PH2)) + *cd = 0; + } else { + /* Harmony SDMMC2 = SDIO2_CD = GPIO_PI5 */ + if (gpio_get_value(GPIO_PI5)) + *cd = 0; + } + return 0; } #endif diff --git a/board/nvidia/seaboard/seaboard.c b/board/nvidia/seaboard/seaboard.c index 1310e3d..0667eaa 100644 --- a/board/nvidia/seaboard/seaboard.c +++ b/board/nvidia/seaboard/seaboard.c @@ -28,6 +28,7 @@ #ifdef CONFIG_TEGRA2_MMC #include <mmc.h> #endif +#include "../common/board.h" /* * Routine: gpio_config_uart @@ -71,6 +72,28 @@ void gpio_config_mmc(void) } /* this is a weak define that we are overriding */ +int board_mmc_init(bd_t *bd) +{ + debug("board_mmc_init called\n"); + /* Enable clocks, muxes, etc. for SDMMC controllers */ + clock_init_mmc4(); + clock_init_mmc3(); + pin_mux_mmc4(); + pin_mux_mmc3(); + gpio_config_mmc(); + + debug("board_mmc_init: init eMMC\n"); + /* init dev 0, eMMC chip, with 8-bit bus */ + tegra2_mmc_init(0, 8); + + 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"); diff --git a/drivers/mmc/tegra2_mmc.c b/drivers/mmc/tegra2_mmc.c index 9e741f2..223dbf3 100644 --- a/drivers/mmc/tegra2_mmc.c +++ b/drivers/mmc/tegra2_mmc.c @@ -478,3 +478,21 @@ int tegra2_mmc_init(int dev_index, int bus_width) dev_index, bus_width); return tegra2_mmc_initialize(dev_index, bus_width); } + +int tegra2_mmc_index(struct mmc *mmc) +{ + struct mmc_host *host = (struct mmc_host *)mmc->priv; + + switch (host->base) { + case TEGRA2_SDMMC3_BASE: + return 1; + case TEGRA2_SDMMC2_BASE: + return 2; + case TEGRA2_SDMMC1_BASE: + return 3; + case TEGRA2_SDMMC4_BASE: + default: + return 0; + } +} + -- 1.7.0.4 ^ permalink raw reply related [flat|nested] 12+ messages in thread
* [U-Boot] [PATCH v2 3/4] tegra2: Enable second SD port (J5) on Harmony 2011-10-04 17:32 ` [U-Boot] [PATCH v2 3/4] tegra2: Enable second SD port (J5) on Harmony Stephen Warren @ 2011-10-05 14:39 ` Simon Glass 2011-10-05 15:46 ` Stephen Warren 2011-11-03 0:16 ` Andy Fleming 1 sibling, 1 reply; 12+ messages in thread From: Simon Glass @ 2011-10-05 14:39 UTC (permalink / raw) To: u-boot Hi Stephen, On Tue, Oct 4, 2011 at 10:32 AM, Stephen Warren <swarren@nvidia.com> wrote: > Seaboard uses SDMMC4, SDMMC3. Harmony uses SDMMC4, SDMMC2. Move > board_init_mmc and gpio_config_mmc into board-specific files, so boards > can choose which ports to set up. Split clock_init_mmc and pin_mux_mmc > into per-port functions, and export them for use by boards. > > Signed-off-by: Stephen Warren <swarren@nvidia.com> I am not thrilled with this patch. It seems to create a lot of new functions, moves knowledge of MMC clock speed into the board file and means that we will always init MMC no matter what the CONFIG settings say. ^ permalink raw reply [flat|nested] 12+ messages in thread
* [U-Boot] [PATCH v2 3/4] tegra2: Enable second SD port (J5) on Harmony 2011-10-05 14:39 ` Simon Glass @ 2011-10-05 15:46 ` Stephen Warren 2011-10-07 4:20 ` Simon Glass 0 siblings, 1 reply; 12+ messages in thread From: Stephen Warren @ 2011-10-05 15:46 UTC (permalink / raw) To: u-boot Simon Glass wrote at Wednesday, October 05, 2011 8:39 AM: > Hi Stephen, > > On Tue, Oct 4, 2011 at 10:32 AM, Stephen Warren <swarren@nvidia.com> wrote: > > Seaboard uses SDMMC4, SDMMC3. Harmony uses SDMMC4, SDMMC2. Move > > board_init_mmc and gpio_config_mmc into board-specific files, so boards > > can choose which ports to set up. Split clock_init_mmc and pin_mux_mmc > > into per-port functions, and export them for use by boards. > > > > Signed-off-by: Stephen Warren <swarren@nvidia.com> > > I am not thrilled with this patch. It seems to create a lot of new > functions, moves knowledge of MMC clock speed into the board file By "board file", do you mean board.c (which already had and still has this knowledge) or harmony.c/seaboard.c? > and means that we will always init MMC no matter what the CONFIG > settings say. All the MMC init calls are already only enabled based on config settings; they're wrapped inside: #ifdef CONFIG_TEGRA2_MMC > From a boot time perspective we want to init only those periphs that > we need to use to boot. If we set up clocks and pinmuxes in the board > init, then this is not possible. We had a discussion on the list a few > months back about how to minimize this and the answer given was that > we should only init peripherals when they are used. Well, nothing in my patch changes anything related to that. I wasn't subscribed then, so I don't know exactly what was discussed. My take: It seems reasonable to defer /driver/ initialization to when a device is first used, but I'm not convinced it's reasonable to defer system-level initialization such as pinmux/gpio-ownership. By either way, that's beyond what these patches aim to do. > Looking ahead to fdt, I think the control point for what gets inited > and what the pinmux settings are will be in the device tree. My > understanding of the concept is that drivers will look at their fdt > node and init accordingly. Sure. > We already have a function: > > int tegra2_mmc_init(int dev_index, int bus_width) > > Could this not be enhanced to set the pinmux and clocks? It seems from > your patch that the pinmux is always fixed but if not it could be a > parameter. Yes, it does make perfect sense to pass the GPIOs into this function. I will rework the patches to do that. Note that the aim of my patches wasn't to re-factor the driver, but to follow existing practices though. This would then move the CD handling into the MMC driver. Note however that the CD function isn't actually used for Tegra; simply removing it may be the best change at this point? Perhaps the clock initialization should be moved into the driver too? For the pinmux, I'm not convinced the MMC driver should be controlling this. It remains to be seen exactly what the DT binding is for pinmux. The pinmux information may or may not be related to the MMC nodes in the DT. I wouldn't want to make the MMC driver control the pinmux until that question was answered. Note that there certainly are different pinmuxing options for the MMC ports, or at least some of them, so the mux settings are board-specific. It just so happens that for ports that multiple currently-supported Boards use, all boards happen to use the same pinmux settings for that port. In the future, I could see say pin_mux_mmc4 being split into e.g. pin_mux_mmc4_xxx and pin_mux_mmc4_yyy or even just making the pinmux calls directly from individual board files if we end up supporting a lot of options. -- nvpublic ^ permalink raw reply [flat|nested] 12+ messages in thread
* [U-Boot] [PATCH v2 3/4] tegra2: Enable second SD port (J5) on Harmony 2011-10-05 15:46 ` Stephen Warren @ 2011-10-07 4:20 ` Simon Glass 0 siblings, 0 replies; 12+ messages in thread From: Simon Glass @ 2011-10-07 4:20 UTC (permalink / raw) To: u-boot Hi Stephen, Have been on hols, sorry for the delay. On Wed, Oct 5, 2011 at 8:46 AM, Stephen Warren <swarren@nvidia.com> wrote: > Simon Glass wrote at Wednesday, October 05, 2011 8:39 AM: >> Hi Stephen, >> >> On Tue, Oct 4, 2011 at 10:32 AM, Stephen Warren <swarren@nvidia.com> wrote: >> > Seaboard uses SDMMC4, SDMMC3. Harmony uses SDMMC4, SDMMC2. Move >> > board_init_mmc and gpio_config_mmc into board-specific files, so boards >> > can choose which ports to set up. Split clock_init_mmc and pin_mux_mmc >> > into per-port functions, and export them for use by boards. >> > >> > Signed-off-by: Stephen Warren <swarren@nvidia.com> >> >> I am not thrilled with this patch. It seems to create a lot of new >> functions, moves knowledge of MMC clock speed into the board file > > By "board file", do you mean board.c (which already had and still has this > knowledge) or harmony.c/seaboard.c? Yes, I'm afraid I am comparing the newer code in the Chromium tree to your patch. It is true that we moved the clock setting into the driver in our tree, since it already has to do it in response to MMC events, and it seems wrong to have this 20MHz thing sitting in the board. > >> and means that we will always init MMC no matter what the CONFIG >> settings say. > > All the MMC init calls are already only enabled based on config settings; > they're wrapped inside: > > #ifdef CONFIG_TEGRA2_MMC Yes - but I suppose if the code were in the driver then you wouldn't need any #ifdefs - the code would just not be included in the build. We will end up with a lot of #ifdefs in the board files (board.c, seaboard.c etc.) if we go down this route. > >> From a boot time perspective we want to init only those periphs that >> we need to use to boot. If we set up clocks and pinmuxes in the board >> init, then this is not possible. We had a discussion on the list a few >> months back about how to minimize this and the answer given was that >> we should only init peripherals when they are used. > > Well, nothing in my patch changes anything related to that. > > I wasn't subscribed then, so I don't know exactly what was discussed. My > take: It seems reasonable to defer /driver/ initialization to when a > device is first used, but I'm not convinced it's reasonable to defer > system-level initialization such as pinmux/gpio-ownership. By either way, > that's beyond what these patches aim to do. No you are right - I was thinking of the Chromium code base. > >> Looking ahead to fdt, I think the control point for what gets inited >> and what the pinmux settings are will be in the device tree. My >> understanding of the concept is that drivers will look at their fdt >> node and init accordingly. > > Sure. > >> We already have a function: >> >> int tegra2_mmc_init(int dev_index, int bus_width) >> >> Could this not be enhanced to set the pinmux and clocks? It seems from >> your patch that the pinmux is always fixed but if not it could be a >> parameter. > > Yes, it does make perfect sense to pass the GPIOs into this function. I > will rework the patches to do that. Note that the aim of my patches wasn't > to re-factor the driver, but to follow existing practices though. This > would then move the CD handling into the MMC driver. Note however that > the CD function isn't actually used for Tegra; simply removing it may be > the best change at this point? > > Perhaps the clock initialization should be moved into the driver too? > Not sure about removing CD - but certainly if the GPIO is available to the driver then it might as well implement it. > For the pinmux, I'm not convinced the MMC driver should be controlling > this. It remains to be seen exactly what the DT binding is for pinmux. > The pinmux information may or may not be related to the MMC nodes in the > DT. I wouldn't want to make the MMC driver control the pinmux until that > question was answered. Yes that's the big question. If pinmux is driver-related then it will make just-in-time device init a lot easier. I agree it is better to resolve this later when all is known. > > Note that there certainly are different pinmuxing options for the MMC > ports, or at least some of them, so the mux settings are board-specific. > It just so happens that for ports that multiple currently-supported > Boards use, all boards happen to use the same pinmux settings for that > port. In the future, I could see say pin_mux_mmc4 being split into e.g. > pin_mux_mmc4_xxx and pin_mux_mmc4_yyy or even just making the pinmux calls > directly from individual board files if we end up supporting a lot of > options. OK I see - but I feel that having lots of pin_mux_mmcx_yyy functions is not great. I depends on where this goes in the kernel, as you say, but I have to express a preference for telling the driver what setting is required. Regards, Simon > > -- > nvpublic > > ^ permalink raw reply [flat|nested] 12+ messages in thread
* [U-Boot] [PATCH v2 3/4] tegra2: Enable second SD port (J5) on Harmony 2011-10-04 17:32 ` [U-Boot] [PATCH v2 3/4] tegra2: Enable second SD port (J5) on Harmony Stephen Warren 2011-10-05 14:39 ` Simon Glass @ 2011-11-03 0:16 ` Andy Fleming 2011-11-03 15:20 ` Stephen Warren 1 sibling, 1 reply; 12+ messages in thread From: Andy Fleming @ 2011-11-03 0:16 UTC (permalink / raw) To: u-boot I'm currently assuming this patch has been superseded by the patches you submitted on the 31st? Andy On Tue, Oct 4, 2011 at 12:32 PM, Stephen Warren <swarren@nvidia.com> wrote: > Seaboard uses SDMMC4, SDMMC3. Harmony uses SDMMC4, SDMMC2. Move > board_init_mmc and gpio_config_mmc into board-specific files, so boards > can choose which ports to set up. Split clock_init_mmc and pin_mux_mmc > into per-port functions, and export them for use by boards. > > Signed-off-by: Stephen Warren <swarren@nvidia.com> > --- > ?board/nvidia/common/board.c ? ? ?| ? 77 ++++++++++++++++++++++++-------------- > ?board/nvidia/common/board.h ? ? ?| ? ?8 +++- > ?board/nvidia/harmony/harmony.c ? | ? 56 +++++++++++++++++++++++++-- > ?board/nvidia/seaboard/seaboard.c | ? 23 +++++++++++ > ?drivers/mmc/tegra2_mmc.c ? ? ? ? | ? 18 +++++++++ > ?5 files changed, 148 insertions(+), 34 deletions(-) > > diff --git a/board/nvidia/common/board.c b/board/nvidia/common/board.c > index 8033612..6c09a4c 100644 > --- a/board/nvidia/common/board.c > +++ b/board/nvidia/common/board.c > @@ -102,20 +102,37 @@ static void pin_mux_uart(void) > > ?#ifdef CONFIG_TEGRA2_MMC > ?/* > - * Routine: clock_init_mmc > - * Description: init the PLL and clocks for the SDMMC controllers > + * Routine: clock_init_mmc4 > + * Description: init the PLL and clocks for SDMMC4 controller > ?*/ > -static void clock_init_mmc(void) > +void clock_init_mmc4(void) > ?{ > ? ? ? ?clock_start_periph_pll(PERIPH_ID_SDMMC4, CLOCK_ID_PERIPH, 20000000); > +} > + > +/* > + * Routine: clock_init_mmc3 > + * Description: init the PLL and clocks for SDMMC3 controller > + */ > +void clock_init_mmc3(void) > +{ > ? ? ? ?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) > + * Routine: clock_init_mmc2 > + * Description: init the PLL and clocks for SDMMC2 controller > ?*/ > -static void pin_mux_mmc(void) > +void clock_init_mmc2(void) > +{ > + ? ? ? clock_start_periph_pll(PERIPH_ID_SDMMC2, CLOCK_ID_PERIPH, 20000000); > +} > + > +/* > + * Routine: pin_mux_mmc4 > + * Description: setup the pin muxes/tristate values for SDMMC4 > + */ > +void pin_mux_mmc4(void) > ?{ > ? ? ? ?/* SDMMC4: config 3, x8 on 2nd set of pins */ > ? ? ? ?pinmux_set_func(PINGRP_ATB, PMUX_FUNC_SDIO4); > @@ -125,7 +142,14 @@ static void pin_mux_mmc(void) > ? ? ? ?pinmux_tristate_disable(PINGRP_ATB); > ? ? ? ?pinmux_tristate_disable(PINGRP_GMA); > ? ? ? ?pinmux_tristate_disable(PINGRP_GME); > +} > > +/* > + * Routine: pin_mux_mmc3 > + * Description: setup the pin muxes/tristate values for SDMMC3 > + */ > +void pin_mux_mmc3(void) > +{ > ? ? ? ?/* SDMMC3: SDIO3_CLK, SDIO3_CMD, SDIO3_DAT[3:0] */ > ? ? ? ?pinmux_set_func(PINGRP_SDB, PMUX_FUNC_SDIO3); > ? ? ? ?pinmux_set_func(PINGRP_SDC, PMUX_FUNC_SDIO3); > @@ -135,6 +159,25 @@ static void pin_mux_mmc(void) > ? ? ? ?pinmux_tristate_disable(PINGRP_SDD); > ? ? ? ?pinmux_tristate_disable(PINGRP_SDB); > ?} > + > +/* > + * Routine: pin_mux_mmc2 > + * Description: setup the pin muxes/tristate values for SDMMC2 > + */ > +void pin_mux_mmc2(void) > +{ > + ? ? ? /* SDMMC2: SDIO2_CLK, SDIO2_CMD, SDIO2_DAT[7:0] */ > + ? ? ? pinmux_set_func(PINGRP_DTA, PMUX_FUNC_SDIO2); > + ? ? ? pinmux_set_func(PINGRP_DTD, PMUX_FUNC_SDIO2); > + > + ? ? ? pinmux_tristate_disable(PINGRP_DTA); > + ? ? ? pinmux_tristate_disable(PINGRP_DTD); > + > + ? ? ? /* For power GPIO PI6 */ > + ? ? ? pinmux_tristate_disable(PINGRP_ATA); > + ? ? ? /* For power GPIO PT3 */ > + ? ? ? pinmux_tristate_disable(PINGRP_DTB); > +} > ?#endif > > ?/* > @@ -152,28 +195,6 @@ int board_init(void) > ? ? ? ?return 0; > ?} > > -#ifdef CONFIG_TEGRA2_MMC > -/* this is a weak define that we are overriding */ > -int board_mmc_init(bd_t *bd) > -{ > - ? ? ? debug("board_mmc_init called\n"); > - ? ? ? /* Enable clocks, muxes, etc. for SDMMC controllers */ > - ? ? ? clock_init_mmc(); > - ? ? ? pin_mux_mmc(); > - ? ? ? gpio_config_mmc(); > - > - ? ? ? debug("board_mmc_init: init eMMC\n"); > - ? ? ? /* init dev 0, eMMC chip, with 8-bit bus */ > - ? ? ? tegra2_mmc_init(0, 8); > - > - ? ? ? debug("board_mmc_init: init SD slot\n"); > - ? ? ? /* init dev 1, SD slot, with 4-bit bus */ > - ? ? ? tegra2_mmc_init(1, 4); > - > - ? ? ? return 0; > -} > -#endif > - > ?#ifdef CONFIG_BOARD_EARLY_INIT_F > ?int board_early_init_f(void) > ?{ > diff --git a/board/nvidia/common/board.h b/board/nvidia/common/board.h > index 344e702..eee475d 100644 > --- a/board/nvidia/common/board.h > +++ b/board/nvidia/common/board.h > @@ -26,7 +26,13 @@ > > ?void tegra2_start(void); > ?void gpio_config_uart(void); > -void gpio_config_mmc(void); > +void clock_init_mmc4(void); > +void clock_init_mmc3(void); > +void clock_init_mmc2(void); > +void pin_mux_mmc4(void); > +void pin_mux_mmc3(void); > +void pin_mux_mmc2(void); > ?int tegra2_mmc_init(int dev_index, int bus_width); > +int tegra2_mmc_index(struct mmc *mmc); > > ?#endif /* BOARD_H */ > diff --git a/board/nvidia/harmony/harmony.c b/board/nvidia/harmony/harmony.c > index cbb30d6..228ae2e 100644 > --- a/board/nvidia/harmony/harmony.c > +++ b/board/nvidia/harmony/harmony.c > @@ -23,10 +23,13 @@ > > ?#include <common.h> > ?#include <asm/io.h> > +#include <asm/arch/clock.h> > ?#include <asm/arch/tegra2.h> > +#include <asm/gpio.h> > ?#ifdef CONFIG_TEGRA2_MMC > ?#include <mmc.h> > ?#endif > +#include "../common/board.h" > > ?/* > ?* Routine: gpio_config_uart > @@ -43,18 +46,61 @@ void gpio_config_uart(void) > ?*/ > ?void gpio_config_mmc(void) > ?{ > - ? ? ? /* Not implemented for now */ > + ? ? ? /* Set SDMMC4 power (GPIO I6) */ > + ? ? ? gpio_request(GPIO_PI6, "SDMMC4 power"); > + ? ? ? gpio_direction_output(GPIO_PI6, 1); > + > + ? ? ? /* Config pin as GPI for SDMMC4 Card Detect (GPIO H2) */ > + ? ? ? gpio_request(GPIO_PH2, "SDMMC4 card detect"); > + ? ? ? gpio_direction_input(GPIO_PH2); > + > + ? ? ? /* Set SDMMC2 power (GPIO T3) */ > + ? ? ? gpio_request(GPIO_PT3, "SDMMC2 power"); > + ? ? ? gpio_direction_output(GPIO_PT3, 1); > + > + ? ? ? /* Config pin as GPI for SDMMC2 Card Detect (GPIO I5) */ > + ? ? ? gpio_request(GPIO_PI5, "SDMMC2 card detect"); > + ? ? ? gpio_direction_input(GPIO_PI5); > +} > + > +/* this is a weak define that we are overriding */ > +int board_mmc_init(bd_t *bd) > +{ > + ? ? ? debug("board_mmc_init called\n"); > + ? ? ? /* Enable clocks, muxes, etc. for SDMMC controllers */ > + ? ? ? clock_init_mmc4(); > + ? ? ? clock_init_mmc2(); > + ? ? ? pin_mux_mmc4(); > + ? ? ? pin_mux_mmc2(); > + ? ? ? gpio_config_mmc(); > + > + ? ? ? debug("board_mmc_init: init SD slot J26\n"); > + ? ? ? /* init dev 0, SD slot J26, with 8-bit bus */ > + ? ? ? tegra2_mmc_init(0, 8); > + > + ? ? ? debug("board_mmc_init: init SD slot J5\n"); > + ? ? ? /* init dev 2, SD slot J5, with 8-bit bus */ > + ? ? ? tegra2_mmc_init(2, 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"); > - ? ? ? /* > - ? ? ? ?* Hard-code CD presence for now. Need to add GPIO inputs > - ? ? ? ?* for Harmony > - ? ? ? ?*/ > ? ? ? ?*cd = 1; > + > + ? ? ? if (tegra2_mmc_index(mmc) == 0) { > + ? ? ? ? ? ? ? /* Harmony SDMMC4 = SDIO4_CD = GPIO_PH2 */ > + ? ? ? ? ? ? ? if (gpio_get_value(GPIO_PH2)) > + ? ? ? ? ? ? ? ? ? ? ? *cd = 0; > + ? ? ? } else { > + ? ? ? ? ? ? ? /* Harmony SDMMC2 = SDIO2_CD = GPIO_PI5 */ > + ? ? ? ? ? ? ? if (gpio_get_value(GPIO_PI5)) > + ? ? ? ? ? ? ? ? ? ? ? *cd = 0; > + ? ? ? } > + > ? ? ? ?return 0; > ?} > ?#endif > diff --git a/board/nvidia/seaboard/seaboard.c b/board/nvidia/seaboard/seaboard.c > index 1310e3d..0667eaa 100644 > --- a/board/nvidia/seaboard/seaboard.c > +++ b/board/nvidia/seaboard/seaboard.c > @@ -28,6 +28,7 @@ > ?#ifdef CONFIG_TEGRA2_MMC > ?#include <mmc.h> > ?#endif > +#include "../common/board.h" > > ?/* > ?* Routine: gpio_config_uart > @@ -71,6 +72,28 @@ void gpio_config_mmc(void) > ?} > > ?/* this is a weak define that we are overriding */ > +int board_mmc_init(bd_t *bd) > +{ > + ? ? ? debug("board_mmc_init called\n"); > + ? ? ? /* Enable clocks, muxes, etc. for SDMMC controllers */ > + ? ? ? clock_init_mmc4(); > + ? ? ? clock_init_mmc3(); > + ? ? ? pin_mux_mmc4(); > + ? ? ? pin_mux_mmc3(); > + ? ? ? gpio_config_mmc(); > + > + ? ? ? debug("board_mmc_init: init eMMC\n"); > + ? ? ? /* init dev 0, eMMC chip, with 8-bit bus */ > + ? ? ? tegra2_mmc_init(0, 8); > + > + ? ? ? 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"); > diff --git a/drivers/mmc/tegra2_mmc.c b/drivers/mmc/tegra2_mmc.c > index 9e741f2..223dbf3 100644 > --- a/drivers/mmc/tegra2_mmc.c > +++ b/drivers/mmc/tegra2_mmc.c > @@ -478,3 +478,21 @@ int tegra2_mmc_init(int dev_index, int bus_width) > ? ? ? ? ? ? ? ?dev_index, bus_width); > ? ? ? ?return tegra2_mmc_initialize(dev_index, bus_width); > ?} > + > +int tegra2_mmc_index(struct mmc *mmc) > +{ > + ? ? ? struct mmc_host *host = (struct mmc_host *)mmc->priv; > + > + ? ? ? switch (host->base) { > + ? ? ? case TEGRA2_SDMMC3_BASE: > + ? ? ? ? ? ? ? return 1; > + ? ? ? case TEGRA2_SDMMC2_BASE: > + ? ? ? ? ? ? ? return 2; > + ? ? ? case TEGRA2_SDMMC1_BASE: > + ? ? ? ? ? ? ? return 3; > + ? ? ? case TEGRA2_SDMMC4_BASE: > + ? ? ? default: > + ? ? ? ? ? ? ? return 0; > + ? ? ? } > +} > + > -- > 1.7.0.4 > > _______________________________________________ > U-Boot mailing list > U-Boot at lists.denx.de > http://lists.denx.de/mailman/listinfo/u-boot > ^ permalink raw reply [flat|nested] 12+ messages in thread
* [U-Boot] [PATCH v2 3/4] tegra2: Enable second SD port (J5) on Harmony 2011-11-03 0:16 ` Andy Fleming @ 2011-11-03 15:20 ` Stephen Warren 0 siblings, 0 replies; 12+ messages in thread From: Stephen Warren @ 2011-11-03 15:20 UTC (permalink / raw) To: u-boot Andy Fleming wrote at Wednesday, November 02, 2011 6:17 PM: > I'm currently assuming this patch has been superseded by the patches > you submitted on the 31st? That looks correct, yes. This whole v2 series was replaced by v7 IIRC. > On Tue, Oct 4, 2011 at 12:32 PM, Stephen Warren <swarren@nvidia.com> wrote: > > Seaboard uses SDMMC4, SDMMC3. Harmony uses SDMMC4, SDMMC2. Move > > board_init_mmc and gpio_config_mmc into board-specific files, so boards > > can choose which ports to set up. Split clock_init_mmc and pin_mux_mmc > > into per-port functions, and export them for use by boards. > > > > Signed-off-by: Stephen Warren <swarren@nvidia.com> -- nvpublic ^ permalink raw reply [flat|nested] 12+ messages in thread
* [U-Boot] [PATCH v2 4/4] tegra2: Add support for Ventana 2011-10-04 17:32 [U-Boot] [PATCH v2 1/4] tegra2: Seaboard's MMC has an 8-bit bus Stephen Warren 2011-10-04 17:32 ` [U-Boot] [PATCH v2 2/4] tegra2: Call gpio_request for SDMMC GPIOs Stephen Warren 2011-10-04 17:32 ` [U-Boot] [PATCH v2 3/4] tegra2: Enable second SD port (J5) on Harmony Stephen Warren @ 2011-10-04 17:32 ` Stephen Warren 2011-10-04 20:52 ` [U-Boot] [PATCH v2 1/4] tegra2: Seaboard's MMC has an 8-bit bus Tom Warren 3 siblings, 0 replies; 12+ messages in thread From: Stephen Warren @ 2011-10-04 17:32 UTC (permalink / raw) To: u-boot 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. Signed-off-by: Stephen Warren <swarren@nvidia.com> --- board/nvidia/seaboard/seaboard.c | 11 ++++++- board/nvidia/ventana/Makefile | 55 ++++++++++++++++++++++++++++++++++++++ boards.cfg | 1 + include/configs/ventana.h | 55 ++++++++++++++++++++++++++++++++++++++ 4 files changed, 120 insertions(+), 2 deletions(-) create mode 100644 board/nvidia/ventana/Makefile create mode 100644 include/configs/ventana.h diff --git a/board/nvidia/seaboard/seaboard.c b/board/nvidia/seaboard/seaboard.c index 0667eaa..58ce256 100644 --- a/board/nvidia/seaboard/seaboard.c +++ b/board/nvidia/seaboard/seaboard.c @@ -31,10 +31,10 @@ #include "../common/board.h" /* - * Routine: gpio_config_uart + * Routine: gpio_config_uart_seaboard * Description: Force GPIO_PI3 low on Seaboard so UART4 works. */ -void gpio_config_uart(void) +void gpio_config_uart_seaboard(void) { int gp = GPIO_PI3; struct gpio_ctlr *gpio = (struct gpio_ctlr *)NV_PA_GPIO_BASE; @@ -55,6 +55,13 @@ void gpio_config_uart(void) writel(val, &bank->gpio_dir_out[GPIO_PORT(gp)]); } +void gpio_config_uart(void) +{ + if (machine_is_ventana()) + return; + gpio_config_uart_seaboard(); +} + #ifdef CONFIG_TEGRA2_MMC /* * Routine: gpio_config_mmc diff --git a/board/nvidia/ventana/Makefile b/board/nvidia/ventana/Makefile new file mode 100644 index 0000000..029673f --- /dev/null +++ b/board/nvidia/ventana/Makefile @@ -0,0 +1,55 @@ +# +# (C) Copyright 2010,2011 +# NVIDIA Corporation <www.nvidia.com> +# +# +# See file CREDITS for list of people who contributed to this +# project. +# +# This program is free software; you can redistribute it and/or +# modify it under the terms of the GNU General Public License as +# published by the Free Software Foundation; either version 2 of +# the License, or (at your option) any later version. +# +# This program is distributed in the hope that it will be useful, +# but WITHOUT ANY WARRANTY; without even the implied warranty of +# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the +# GNU General Public License for more details. +# +# You should have received a copy of the GNU General Public License +# along with this program; if not, write to the Free Software +# Foundation, Inc., 59 Temple Place, Suite 330, Boston, +# MA 02111-1307 USA +# + +include $(TOPDIR)/config.mk + +ifneq ($(OBJTREE),$(SRCTREE)) +$(shell mkdir -p $(obj)../common) +endif + +LIB = $(obj)lib$(BOARD).o + +COBJS += ../seaboard/seaboard.o +COBJS += ../common/board.o + +SRCS := $(COBJS:.o=.c) +OBJS := $(addprefix $(obj),$(COBJS)) + +$(LIB): $(obj).depend $(OBJS) + $(AR) $(ARFLAGS) $@ $(OBJS) + +clean: + rm -f $(OBJS) + +distclean: clean + rm -f $(LIB) core *.bak $(obj).depend + +######################################################################### + +# defines $(obj).depend target +include $(SRCTREE)/rules.mk + +sinclude $(obj).depend + +######################################################################### diff --git a/boards.cfg b/boards.cfg index e5cb91b..e1ab6f9 100644 --- a/boards.cfg +++ b/boards.cfg @@ -188,6 +188,7 @@ s5pc210_universal arm armv7 universal_c210 samsung smdkv310 arm armv7 smdkv310 samsung s5pc2xx harmony arm armv7 harmony nvidia tegra2 seaboard arm armv7 seaboard nvidia tegra2 +ventana arm armv7 ventana nvidia tegra2 u8500_href arm armv7 u8500 st-ericsson u8500 actux1_4_16 arm ixp actux1 - - actux1:FLASH2X2 actux1_8_16 arm ixp actux1 - - actux1:FLASH1X8 diff --git a/include/configs/ventana.h b/include/configs/ventana.h new file mode 100644 index 0000000..afd6ff6 --- /dev/null +++ b/include/configs/ventana.h @@ -0,0 +1,55 @@ +/* + * (C) Copyright 2010,2011 + * NVIDIA Corporation <www.nvidia.com> + * + * See file CREDITS for list of people who contributed to this + * project. + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as + * published by the Free Software Foundation; either version 2 of + * the License, or (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write to the Free Software + * Foundation, Inc., 59 Temple Place, Suite 330, Boston, + * MA 02111-1307 USA + */ + +#ifndef __CONFIG_H +#define __CONFIG_H + +#include <asm/sizes.h> +#include "tegra2-common.h" + +/* High-level configuration options */ +#define TEGRA2_SYSMEM "mem=384M at 0M nvmem=128M at 384M mem=512M at 512M" +#define V_PROMPT "Tegra2 (Ventana) # " +#define CONFIG_TEGRA2_BOARD_STRING "NVIDIA Ventana" + +/* Board-specific serial config */ +#define CONFIG_SERIAL_MULTI +#define CONFIG_TEGRA2_ENABLE_UARTD +#define CONFIG_SYS_NS16550_COM1 NV_PA_APB_UARTD_BASE + +#define CONFIG_MACH_TYPE MACH_TYPE_VENTANA +#define CONFIG_SYS_BOARD_ODMDATA 0x300d8011 /* lp1, 1GB */ + +#define CONFIG_BOARD_EARLY_INIT_F + +/* SD/MMC */ +#define CONFIG_MMC +#define CONFIG_GENERIC_MMC +#define CONFIG_TEGRA2_MMC +#define CONFIG_CMD_MMC + +#define CONFIG_DOS_PARTITION +#define CONFIG_EFI_PARTITION +#define CONFIG_CMD_EXT2 +#define CONFIG_CMD_FAT +#endif /* __CONFIG_H */ -- 1.7.0.4 ^ permalink raw reply related [flat|nested] 12+ messages in thread
* [U-Boot] [PATCH v2 1/4] tegra2: Seaboard's MMC has an 8-bit bus 2011-10-04 17:32 [U-Boot] [PATCH v2 1/4] tegra2: Seaboard's MMC has an 8-bit bus Stephen Warren ` (2 preceding siblings ...) 2011-10-04 17:32 ` [U-Boot] [PATCH v2 4/4] tegra2: Add support for Ventana Stephen Warren @ 2011-10-04 20:52 ` Tom Warren 2011-10-04 21:14 ` Stephen Warren 3 siblings, 1 reply; 12+ messages in thread From: Tom Warren @ 2011-10-04 20:52 UTC (permalink / raw) To: u-boot Stephen, > -----Original Message----- > From: Stephen Warren [mailto:swarren at nvidia.com] > Sent: Tuesday, October 04, 2011 10:33 AM > To: U-Boot Mailing List > Cc: Tom Warren; Simon Glass; Stephen Warren > Subject: [PATCH v2 1/4] tegra2: Seaboard's MMC has an 8-bit bus > > Modify board_mmc_init to configure SDMMC4 for an 8-bit bus, since that's > what is wired up on the board. > [Tom Warren] Did you try this and then do a mmcinfo to see what you get back? On my Seaboard T20-A03, initializing with 8-bit arg gets you 1-bit back on mmcinfo; using 4-bit (as I had it) gets you back 4-bit. Although it's wired for all 8 data bits, I'm not sure the eMMC controller or device or I/F handles 8 bits correctly. Need to do some more research here, but we should leave this @ 4-bit for now IMO. Tom > Signed-off-by: Stephen Warren <swarren@nvidia.com> > --- > Note: These patches build on various Tegra2-related patches recently posted > by Simon Glass; see v2 of the 6-entry patchset shown at: > http://patchwork.ozlabs.org/project/uboot/list/?submitter=6170&state=1&q=teg > ra2 > > board/nvidia/common/board.c | 4 ++-- > 1 files changed, 2 insertions(+), 2 deletions(-) > > diff --git a/board/nvidia/common/board.c b/board/nvidia/common/board.c > index d13537d..8033612 100644 > --- a/board/nvidia/common/board.c > +++ b/board/nvidia/common/board.c > @@ -163,8 +163,8 @@ int board_mmc_init(bd_t *bd) > gpio_config_mmc(); > > debug("board_mmc_init: init eMMC\n"); > - /* init dev 0, eMMC chip, with 4-bit bus */ > - tegra2_mmc_init(0, 4); > + /* init dev 0, eMMC chip, with 8-bit bus */ > + tegra2_mmc_init(0, 8); > > debug("board_mmc_init: init SD slot\n"); > /* init dev 1, SD slot, with 4-bit bus */ > -- > 1.7.0.4 ----------------------------------------------------------------------------------- This email message is for the sole use of the intended recipient(s) and may contain confidential information. Any unauthorized review, use, disclosure or distribution is prohibited. If you are not the intended recipient, please contact the sender by reply email and destroy all copies of the original message. ----------------------------------------------------------------------------------- ^ permalink raw reply [flat|nested] 12+ messages in thread
* [U-Boot] [PATCH v2 1/4] tegra2: Seaboard's MMC has an 8-bit bus 2011-10-04 20:52 ` [U-Boot] [PATCH v2 1/4] tegra2: Seaboard's MMC has an 8-bit bus Tom Warren @ 2011-10-04 21:14 ` Stephen Warren 0 siblings, 0 replies; 12+ messages in thread From: Stephen Warren @ 2011-10-04 21:14 UTC (permalink / raw) To: u-boot Tom Warren wrote at Tuesday, October 04, 2011 2:52 PM: > Stephen Warren wrote at Tuesday, October 04, 2011 10:33 AM: > > Modify board_mmc_init to configure SDMMC4 for an 8-bit bus, since that's > > what is wired up on the board. > > > [Tom Warren] Did you try this and then do a mmcinfo to see what you get back? On my Seaboard T20-A03, > initializing with 8-bit arg gets you 1-bit back on mmcinfo; using 4-bit (as I had it) gets you back 4- > bit. Although it's wired for all 8 data bits, I'm not sure the eMMC controller or device or I/F > handles 8 bits correctly. Need to do some more research here, but we should leave this @ 4-bit for now > IMO. Ah yes, I see the same issue. I had only validated that running "mmc part" and ext2load still operated correctly. I see the exact same issue on Harmony's J26 port, which I'd also marked as 8-bit. I guess the Tegra MMC driver doesn't support 8-bit mode or something. I'll take a quick look and see if there's anything obvious. If not, I'll just drop that patch from the series. Thanks for taking a look. -- nvpublic ^ permalink raw reply [flat|nested] 12+ messages in thread
end of thread, other threads:[~2011-11-03 15:20 UTC | newest] Thread overview: 12+ messages (download: mbox.gz / follow: Atom feed) -- links below jump to the message on this page -- 2011-10-04 17:32 [U-Boot] [PATCH v2 1/4] tegra2: Seaboard's MMC has an 8-bit bus Stephen Warren 2011-10-04 17:32 ` [U-Boot] [PATCH v2 2/4] tegra2: Call gpio_request for SDMMC GPIOs Stephen Warren 2011-10-04 23:56 ` Simon Glass 2011-10-04 17:32 ` [U-Boot] [PATCH v2 3/4] tegra2: Enable second SD port (J5) on Harmony Stephen Warren 2011-10-05 14:39 ` Simon Glass 2011-10-05 15:46 ` Stephen Warren 2011-10-07 4:20 ` Simon Glass 2011-11-03 0:16 ` Andy Fleming 2011-11-03 15:20 ` Stephen Warren 2011-10-04 17:32 ` [U-Boot] [PATCH v2 4/4] tegra2: Add support for Ventana Stephen Warren 2011-10-04 20:52 ` [U-Boot] [PATCH v2 1/4] tegra2: Seaboard's MMC has an 8-bit bus Tom Warren 2011-10-04 21:14 ` Stephen Warren
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.