diff mbox series

mmc: davinci_mmc: Cleanup to use dt in U-boot and static platdata in SPL

Message ID 20200522020228.29265-1-faiz_abbas@ti.com
State Accepted
Commit c78ae11e07d84c10ad5ee91593e49fcf20b07359
Headers show
Series mmc: davinci_mmc: Cleanup to use dt in U-boot and static platdata in SPL | expand

Commit Message

Faiz Abbas May 22, 2020, 2:02 a.m. UTC
Cleanup this driver to use dt in U-boot and static platdata in SPL.
This requires the following steps:

1. Move all platdata assignment from probe() to ofdata_to_platdata().
   This function is only called in U-boot.
2. Replicate all the platdata assignment being done in
   ofdata_to_platdata() in the omapl138 board file. This data is used in
   the SPL case where SPL_OF_CONTROL is not enabled.
3. Remove SPL_OF_CONTROL and related configs from omapl138_lcdk_defconfig

This cleanup effectively reverts 3ef94715cc ('mmc: davinci: fix mmc boot in SPL')

Signed-off-by: Faiz Abbas <faiz_abbas at ti.com>
---

This patch fixes things being broken by Simon's series disabling
libfdt when using of-platdata:
https://patchwork.ozlabs.org/project/uboot/cover/20191107155318.103300-1-sjg at chromium.org/


 .../mach-davinci/include/mach/sdmmc_defs.h    |  7 +++
 board/davinci/da8xxevm/omapl138_lcdk.c        | 12 ++++
 configs/omapl138_lcdk_defconfig               |  4 --
 drivers/mmc/davinci_mmc.c                     | 63 ++++++++-----------
 4 files changed, 45 insertions(+), 41 deletions(-)

Comments

Bartosz Golaszewski May 25, 2020, 12:03 p.m. UTC | #1
pt., 22 maj 2020 o 04:02 Faiz Abbas <faiz_abbas at ti.com> napisa?(a):
>
> Cleanup this driver to use dt in U-boot and static platdata in SPL.
> This requires the following steps:
>
> 1. Move all platdata assignment from probe() to ofdata_to_platdata().
>    This function is only called in U-boot.
> 2. Replicate all the platdata assignment being done in
>    ofdata_to_platdata() in the omapl138 board file. This data is used in
>    the SPL case where SPL_OF_CONTROL is not enabled.
> 3. Remove SPL_OF_CONTROL and related configs from omapl138_lcdk_defconfig
>
> This cleanup effectively reverts 3ef94715cc ('mmc: davinci: fix mmc boot in SPL')
>
> Signed-off-by: Faiz Abbas <faiz_abbas at ti.com>

Tested-by: Bartosz Golaszewski <bgolaszewski at baylibre.com>
Simon Glass May 25, 2020, 2:57 p.m. UTC | #2
Hi Faiz,

On Thu, 21 May 2020 at 20:02, Faiz Abbas <faiz_abbas at ti.com> wrote:
>
> Cleanup this driver to use dt in U-boot and static platdata in SPL.
> This requires the following steps:
>
> 1. Move all platdata assignment from probe() to ofdata_to_platdata().
>    This function is only called in U-boot.
> 2. Replicate all the platdata assignment being done in
>    ofdata_to_platdata() in the omapl138 board file. This data is used in
>    the SPL case where SPL_OF_CONTROL is not enabled.
> 3. Remove SPL_OF_CONTROL and related configs from omapl138_lcdk_defconfig
>
> This cleanup effectively reverts 3ef94715cc ('mmc: davinci: fix mmc boot in SPL')
>
> Signed-off-by: Faiz Abbas <faiz_abbas at ti.com>
> ---
>
> This patch fixes things being broken by Simon's series disabling
> libfdt when using of-platdata:
> https://patchwork.ozlabs.org/project/uboot/cover/20191107155318.103300-1-sjg at chromium.org/
>
>
>  .../mach-davinci/include/mach/sdmmc_defs.h    |  7 +++
>  board/davinci/da8xxevm/omapl138_lcdk.c        | 12 ++++
>  configs/omapl138_lcdk_defconfig               |  4 --
>  drivers/mmc/davinci_mmc.c                     | 63 ++++++++-----------
>  4 files changed, 45 insertions(+), 41 deletions(-)

Could this use of-platdata instead?

Regards,
Simon
Faiz Abbas May 25, 2020, 3:33 p.m. UTC | #3
Hi Simon,

On 25/05/20 8:27 pm, Simon Glass wrote:
> Hi Faiz,
> 
> On Thu, 21 May 2020 at 20:02, Faiz Abbas <faiz_abbas at ti.com> wrote:
>>
>> Cleanup this driver to use dt in U-boot and static platdata in SPL.
>> This requires the following steps:
>>
>> 1. Move all platdata assignment from probe() to ofdata_to_platdata().
>>    This function is only called in U-boot.
>> 2. Replicate all the platdata assignment being done in
>>    ofdata_to_platdata() in the omapl138 board file. This data is used in
>>    the SPL case where SPL_OF_CONTROL is not enabled.
>> 3. Remove SPL_OF_CONTROL and related configs from omapl138_lcdk_defconfig
>>
>> This cleanup effectively reverts 3ef94715cc ('mmc: davinci: fix mmc boot in SPL')
>>
>> Signed-off-by: Faiz Abbas <faiz_abbas at ti.com>
>> ---
>>
>> This patch fixes things being broken by Simon's series disabling
>> libfdt when using of-platdata:
>> https://patchwork.ozlabs.org/project/uboot/cover/20191107155318.103300-1-sjg at chromium.org/
>>
>>
>>  .../mach-davinci/include/mach/sdmmc_defs.h    |  7 +++
>>  board/davinci/da8xxevm/omapl138_lcdk.c        | 12 ++++
>>  configs/omapl138_lcdk_defconfig               |  4 --
>>  drivers/mmc/davinci_mmc.c                     | 63 ++++++++-----------
>>  4 files changed, 45 insertions(+), 41 deletions(-)
> 
> Could this use of-platdata instead?
> 

We would need to implement address translation in of-platdata to do that. I prefer merging your
fixes with this for now and leave of-platdata up for the future.

Thanks,
Faiz
Simon Glass May 25, 2020, 3:38 p.m. UTC | #4
Hi Faiz,

On Mon, 25 May 2020 at 09:33, Faiz Abbas <faiz_abbas at ti.com> wrote:
>
> Hi Simon,
>
> On 25/05/20 8:27 pm, Simon Glass wrote:
> > Hi Faiz,
> >
> > On Thu, 21 May 2020 at 20:02, Faiz Abbas <faiz_abbas at ti.com> wrote:
> >>
> >> Cleanup this driver to use dt in U-boot and static platdata in SPL.
> >> This requires the following steps:
> >>
> >> 1. Move all platdata assignment from probe() to ofdata_to_platdata().
> >>    This function is only called in U-boot.
> >> 2. Replicate all the platdata assignment being done in
> >>    ofdata_to_platdata() in the omapl138 board file. This data is used in
> >>    the SPL case where SPL_OF_CONTROL is not enabled.
> >> 3. Remove SPL_OF_CONTROL and related configs from omapl138_lcdk_defconfig
> >>
> >> This cleanup effectively reverts 3ef94715cc ('mmc: davinci: fix mmc boot in SPL')
> >>
> >> Signed-off-by: Faiz Abbas <faiz_abbas at ti.com>
> >> ---
> >>
> >> This patch fixes things being broken by Simon's series disabling
> >> libfdt when using of-platdata:
> >> https://patchwork.ozlabs.org/project/uboot/cover/20191107155318.103300-1-sjg at chromium.org/
> >>
> >>
> >>  .../mach-davinci/include/mach/sdmmc_defs.h    |  7 +++
> >>  board/davinci/da8xxevm/omapl138_lcdk.c        | 12 ++++
> >>  configs/omapl138_lcdk_defconfig               |  4 --
> >>  drivers/mmc/davinci_mmc.c                     | 63 ++++++++-----------
> >>  4 files changed, 45 insertions(+), 41 deletions(-)
> >
> > Could this use of-platdata instead?
> >
>
> We would need to implement address translation in of-platdata to do that. I prefer merging your
> fixes with this for now and leave of-platdata up for the future.

OK, yes, we can't wait for non-existent features.

There is a bit of a queue of functionality lining up for of-platdata at present.

Regards,
Simon
Simon Glass May 26, 2020, 4:54 p.m. UTC | #5
Hi Faiz,

I have ended up with an omap L138 lcdk board. Do you have instructions
on how to get it booting from uSD and how to create the card image?  I
have tried various sources but nothing works so far. Also, are you
able to send me a test image I can 'dd' onto the board?

Regards,
Simon

On Mon, 25 May 2020 at 09:38, Simon Glass <sjg at chromium.org> wrote:
>
> Hi Faiz,
>
> On Mon, 25 May 2020 at 09:33, Faiz Abbas <faiz_abbas at ti.com> wrote:
> >
> > Hi Simon,
> >
> > On 25/05/20 8:27 pm, Simon Glass wrote:
> > > Hi Faiz,
> > >
> > > On Thu, 21 May 2020 at 20:02, Faiz Abbas <faiz_abbas at ti.com> wrote:
> > >>
> > >> Cleanup this driver to use dt in U-boot and static platdata in SPL.
> > >> This requires the following steps:
> > >>
> > >> 1. Move all platdata assignment from probe() to ofdata_to_platdata().
> > >>    This function is only called in U-boot.
> > >> 2. Replicate all the platdata assignment being done in
> > >>    ofdata_to_platdata() in the omapl138 board file. This data is used in
> > >>    the SPL case where SPL_OF_CONTROL is not enabled.
> > >> 3. Remove SPL_OF_CONTROL and related configs from omapl138_lcdk_defconfig
> > >>
> > >> This cleanup effectively reverts 3ef94715cc ('mmc: davinci: fix mmc boot in SPL')
> > >>
> > >> Signed-off-by: Faiz Abbas <faiz_abbas at ti.com>
> > >> ---
> > >>
> > >> This patch fixes things being broken by Simon's series disabling
> > >> libfdt when using of-platdata:
> > >> https://patchwork.ozlabs.org/project/uboot/cover/20191107155318.103300-1-sjg at chromium.org/
> > >>
> > >>
> > >>  .../mach-davinci/include/mach/sdmmc_defs.h    |  7 +++
> > >>  board/davinci/da8xxevm/omapl138_lcdk.c        | 12 ++++
> > >>  configs/omapl138_lcdk_defconfig               |  4 --
> > >>  drivers/mmc/davinci_mmc.c                     | 63 ++++++++-----------
> > >>  4 files changed, 45 insertions(+), 41 deletions(-)
> > >
> > > Could this use of-platdata instead?
> > >
> >
> > We would need to implement address translation in of-platdata to do that. I prefer merging your
> > fixes with this for now and leave of-platdata up for the future.
>
> OK, yes, we can't wait for non-existent features.
>
> There is a bit of a queue of functionality lining up for of-platdata at present.
>
> Regards,
> Simon
Faiz Abbas May 26, 2020, 5:37 p.m. UTC | #6
Simon,

On 26/05/20 10:24 pm, Simon Glass wrote:
> Hi Faiz,
> 
> I have ended up with an omap L138 lcdk board. Do you have instructions
> on how to get it booting from uSD and how to create the card image?  I
> have tried various sources but nothing works so far. Also, are you
> able to send me a test image I can 'dd' onto the board?
> 

The commands in board/davinci/da8xxevm/README.da850 don't work?

Try the following:

 2 - using the "dd" command
        $ dd if=u-boot.ais of=/dev/mmcblk0 seek=117 bs=512 conv=fsync

Thanks,
Faiz
Simon Glass May 26, 2020, 6:38 p.m. UTC | #7
Hi Faiz,

On Tue, 26 May 2020 at 11:37, Faiz Abbas <faiz_abbas at ti.com> wrote:
>
> Simon,
>
> On 26/05/20 10:24 pm, Simon Glass wrote:
> > Hi Faiz,
> >
> > I have ended up with an omap L138 lcdk board. Do you have instructions
> > on how to get it booting from uSD and how to create the card image?  I
> > have tried various sources but nothing works so far. Also, are you
> > able to send me a test image I can 'dd' onto the board?
> >
>
> The commands in board/davinci/da8xxevm/README.da850 don't work?
>
> Try the following:
>
>  2 - using the "dd" command
>         $ dd if=u-boot.ais of=/dev/mmcblk0 seek=117 bs=512 conv=fsync

Thanks! I didn't make the connection with da850 or think to look in
board/ for docs.

Regards,
Simon
diff mbox series

Patch

diff --git a/arch/arm/mach-davinci/include/mach/sdmmc_defs.h b/arch/arm/mach-davinci/include/mach/sdmmc_defs.h
index 46f6391aa2..f95a607e52 100644
--- a/arch/arm/mach-davinci/include/mach/sdmmc_defs.h
+++ b/arch/arm/mach-davinci/include/mach/sdmmc_defs.h
@@ -152,6 +152,13 @@  struct davinci_mmc {
 	struct mmc_config cfg;
 };
 
+#define DAVINCI_MAX_BLOCKS	(32)
+struct davinci_mmc_plat {
+	struct davinci_mmc_regs *reg_base;	/* Register base address */
+	struct mmc_config cfg;
+	struct mmc mmc;
+};
+
 int davinci_mmc_init(bd_t *bis, struct davinci_mmc *host);
 
 #endif /* _SDMMC_DEFS_H */
diff --git a/board/davinci/da8xxevm/omapl138_lcdk.c b/board/davinci/da8xxevm/omapl138_lcdk.c
index adb56c6c87..84603cb117 100644
--- a/board/davinci/da8xxevm/omapl138_lcdk.c
+++ b/board/davinci/da8xxevm/omapl138_lcdk.c
@@ -368,8 +368,20 @@  U_BOOT_DEVICE(omapl138_uart) = {
 	.platdata = &serial_pdata,
 };
 
+static const struct davinci_mmc_plat mmc_platdata = {
+	.reg_base = (struct davinci_mmc_regs *)DAVINCI_MMC_SD0_BASE,
+	.cfg = {
+		.f_min = 200000,
+		.f_max = 25000000,
+		.voltages = MMC_VDD_32_33 | MMC_VDD_33_34,
+		.host_caps = MMC_MODE_4BIT,
+		.b_max = DAVINCI_MAX_BLOCKS,
+		.name = "da830-mmc",
+	},
+};
 U_BOOT_DEVICE(omapl138_mmc) = {
 	.name = "davinci_mmc",
+	.platdata = &mmc_platdata,
 };
 
 void spl_board_init(void)
diff --git a/configs/omapl138_lcdk_defconfig b/configs/omapl138_lcdk_defconfig
index 50cf09c7f1..b0a58de03d 100644
--- a/configs/omapl138_lcdk_defconfig
+++ b/configs/omapl138_lcdk_defconfig
@@ -40,16 +40,13 @@  CONFIG_CMD_MTDPARTS=y
 CONFIG_CMD_DIAG=y
 CONFIG_CMD_UBI=y
 CONFIG_OF_CONTROL=y
-CONFIG_SPL_OF_CONTROL=y
 CONFIG_DEFAULT_DEVICE_TREE="da850-lcdk"
-CONFIG_SPL_OF_PLATDATA=y
 CONFIG_ENV_IS_IN_NAND=y
 CONFIG_SYS_RELOC_GD_ENV_ADDR=y
 CONFIG_NET_RANDOM_ETHADDR=y
 CONFIG_DM=y
 CONFIG_SPL_DM=y
 CONFIG_SPL_DM_SEQ_ALIAS=y
-CONFIG_SPL_OF_TRANSLATE=y
 CONFIG_DA8XX_GPIO=y
 CONFIG_DM_I2C=y
 CONFIG_SYS_I2C_DAVINCI=y
@@ -82,4 +79,3 @@  CONFIG_USB_MUSB_HOST=y
 CONFIG_USB_MUSB_DA8XX=y
 CONFIG_USB_MUSB_PIO_ONLY=y
 CONFIG_USB_STORAGE=y
-# CONFIG_SPL_OF_LIBFDT is not set
diff --git a/drivers/mmc/davinci_mmc.c b/drivers/mmc/davinci_mmc.c
index 2408a687d2..4ef9f7cc8b 100644
--- a/drivers/mmc/davinci_mmc.c
+++ b/drivers/mmc/davinci_mmc.c
@@ -18,7 +18,6 @@ 
 #include <asm-generic/gpio.h>
 #include <linux/delay.h>
 
-#define DAVINCI_MAX_BLOCKS	(32)
 #define WATCHDOG_COUNT		(100000)
 
 #define get_val(addr)		REG(addr)
@@ -34,12 +33,6 @@  struct davinci_mmc_priv {
 	struct gpio_desc cd_gpio;       /* Card Detect GPIO */
 	struct gpio_desc wp_gpio;       /* Write Protect GPIO */
 };
-
-struct davinci_mmc_plat
-{
-	struct mmc_config cfg;
-	struct mmc mmc;
-};
 #endif
 
 /* Set davinci clock prescalar value based on the required clock in HZ */
@@ -487,43 +480,16 @@  static int davinci_mmc_probe(struct udevice *dev)
 	struct mmc_uclass_priv *upriv = dev_get_uclass_priv(dev);
 	struct davinci_mmc_plat *plat = dev_get_platdata(dev);
 	struct davinci_mmc_priv *priv = dev_get_priv(dev);
- 	struct mmc_config *cfg = &plat->cfg;
-#ifdef CONFIG_SPL_BUILD
-	int ret;
-#endif
-
-	cfg->f_min = 200000;
-	cfg->f_max = 25000000;
-	cfg->voltages = MMC_VDD_32_33 | MMC_VDD_33_34,
-	cfg->host_caps = MMC_MODE_4BIT, /* DA850 supports only 4-bit SD/MMC */
-	cfg->b_max = DAVINCI_MAX_BLOCKS;
-	cfg->name = "da830-mmc";
 
-	priv->reg_base = (struct davinci_mmc_regs *)dev_read_addr(dev);
+	priv->reg_base = plat->reg_base;
 	priv->input_clk = clk_get(DAVINCI_MMCSD_CLKID);
-
 #if CONFIG_IS_ENABLED(DM_GPIO)
 	/* These GPIOs are optional */
 	gpio_request_by_name(dev, "cd-gpios", 0, &priv->cd_gpio, GPIOD_IS_IN);
 	gpio_request_by_name(dev, "wp-gpios", 0, &priv->wp_gpio, GPIOD_IS_IN);
 #endif
-
 	upriv->mmc = &plat->mmc;
 
-#ifdef CONFIG_SPL_BUILD
-	/*
-	 * FIXME This is a temporary workaround to enable the driver model in
-	 * SPL on omapl138-lcdk. For some reason the bind() callback is not
-	 * being called in SPL for MMC which breaks the mmc boot - the hack
-	 * is to call mmc_bind() from probe(). We also don't have full DT
-	 * support in SPL, hence the hard-coded base register address.
-	 */
-	priv->reg_base = (struct davinci_mmc_regs *)DAVINCI_MMC_SD0_BASE;
-	ret = mmc_bind(dev, &plat->mmc, &plat->cfg);
-	if (ret)
-		return ret;
-#endif
-
 	return davinci_dm_mmc_init(dev);
 }
 
@@ -534,21 +500,44 @@  static int davinci_mmc_bind(struct udevice *dev)
 	return mmc_bind(dev, &plat->mmc, &plat->cfg);
 }
 
+#if CONFIG_IS_ENABLED(OF_CONTROL)
+static int davinci_mmc_ofdata_to_platdata(struct udevice *dev)
+{
+	struct davinci_mmc_plat *plat = dev_get_platdata(dev);
+	struct mmc_config *cfg = &plat->cfg;
+
+	plat->reg_base = (struct davinci_mmc_regs *)dev_read_addr(dev);
+	cfg->f_min = 200000;
+	cfg->f_max = 25000000;
+	cfg->voltages = MMC_VDD_32_33 | MMC_VDD_33_34,
+	cfg->host_caps = MMC_MODE_4BIT, /* DA850 supports only 4-bit SD/MMC */
+	cfg->b_max = DAVINCI_MAX_BLOCKS;
+	cfg->name = "da830-mmc";
+
+	return 0;
+}
+
 static const struct udevice_id davinci_mmc_ids[] = {
 	{ .compatible = "ti,da830-mmc" },
 	{},
 };
-
+#endif
 U_BOOT_DRIVER(davinci_mmc_drv) = {
 	.name = "davinci_mmc",
 	.id		= UCLASS_MMC,
+#if CONFIG_IS_ENABLED(OF_CONTROL)
 	.of_match	= davinci_mmc_ids,
+	.platdata_auto_alloc_size = sizeof(struct davinci_mmc_plat),
+	.ofdata_to_platdata = davinci_mmc_ofdata_to_platdata,
+#endif
 #if CONFIG_BLK
 	.bind		= davinci_mmc_bind,
 #endif
 	.probe = davinci_mmc_probe,
 	.ops = &davinci_mmc_ops,
-	.platdata_auto_alloc_size = sizeof(struct davinci_mmc_plat),
 	.priv_auto_alloc_size = sizeof(struct davinci_mmc_priv),
+#if !CONFIG_IS_ENABLED(OF_CONTROL)
+	.flags	= DM_FLAG_PRE_RELOC,
+#endif
 };
 #endif