All of lore.kernel.org
 help / color / mirror / Atom feed
* [U-Boot] [PATCH] mx7dsabresd: Remove CONFIG_DM_MMC to avoid regression
@ 2017-10-11 14:23 Fabio Estevam
  2017-10-11 14:27 ` Otavio Salvador
  0 siblings, 1 reply; 25+ messages in thread
From: Fabio Estevam @ 2017-10-11 14:23 UTC (permalink / raw)
  To: u-boot

Since U-Boot 2017.09 it is no longer possible to boot a kernel from SD
card:

U-Boot 2017.11-rc1-00149-g38c8107-dirty (Oct 11 2017 - 09:55:22 -0300)          
                                                                                
CPU:   Freescale i.MX7D rev1.2 1000 MHz (running at 792 MHz)                    
CPU:   Commercial temperature grade (0C to 95C) at 35C                          
Reset cause: POR                                                                
Model: Freescale i.MX7 SabreSD Board                                            
Board: i.MX7D SABRESD in non-secure mode                                        
DRAM:  1 GiB                                                                    
MMC:   MMC/SD: Reset never completed.                                           
                                                                                
MMC Device 0 not found                                                          
*** Warning - No MMC card found, using default environment                      

While a proper fix is not available, remove CONFIG_DM_MMC and register
the USDHC1 port in the old way, so that this board can boot a kernel
again.

Signed-off-by: Fabio Estevam <fabio.estevam@nxp.com>
---
 board/freescale/mx7dsabresd/mx7dsabresd.c | 42 +++++++++++++++++++++++++++++++
 configs/mx7dsabresd_defconfig             |  1 -
 2 files changed, 42 insertions(+), 1 deletion(-)

diff --git a/board/freescale/mx7dsabresd/mx7dsabresd.c b/board/freescale/mx7dsabresd/mx7dsabresd.c
index 5819b18..12f7c58 100644
--- a/board/freescale/mx7dsabresd/mx7dsabresd.c
+++ b/board/freescale/mx7dsabresd/mx7dsabresd.c
@@ -29,6 +29,9 @@ DECLARE_GLOBAL_DATA_PTR;
 #define UART_PAD_CTRL  (PAD_CTL_DSE_3P3V_49OHM | \
 	PAD_CTL_PUS_PU100KOHM | PAD_CTL_HYS)
 
+#define USDHC_PAD_CTRL (PAD_CTL_DSE_3P3V_32OHM | PAD_CTL_SRE_SLOW | \
+	PAD_CTL_HYS | PAD_CTL_PUE | PAD_CTL_PUS_PU47KOHM)
+
 #define ENET_PAD_CTRL  (PAD_CTL_PUS_PU100KOHM | PAD_CTL_DSE_3P3V_49OHM)
 #define ENET_PAD_CTRL_MII  (PAD_CTL_DSE_3P3V_32OHM)
 
@@ -191,6 +194,18 @@ static iomux_v3_cfg_t const fec1_pads[] = {
 	MX7D_PAD_GPIO1_IO11__ENET1_MDC | MUX_PAD_CTRL(ENET_PAD_CTRL_MII),
 };
 
+static iomux_v3_cfg_t const usdhc1_pads[] = {
+	MX7D_PAD_SD1_CLK__SD1_CLK | MUX_PAD_CTRL(USDHC_PAD_CTRL),
+	MX7D_PAD_SD1_CMD__SD1_CMD | MUX_PAD_CTRL(USDHC_PAD_CTRL),
+	MX7D_PAD_SD1_DATA0__SD1_DATA0 | MUX_PAD_CTRL(USDHC_PAD_CTRL),
+	MX7D_PAD_SD1_DATA1__SD1_DATA1 | MUX_PAD_CTRL(USDHC_PAD_CTRL),
+	MX7D_PAD_SD1_DATA2__SD1_DATA2 | MUX_PAD_CTRL(USDHC_PAD_CTRL),
+	MX7D_PAD_SD1_DATA3__SD1_DATA3 | MUX_PAD_CTRL(USDHC_PAD_CTRL),
+
+	MX7D_PAD_SD1_CD_B__GPIO5_IO0 | MUX_PAD_CTRL(USDHC_PAD_CTRL),
+	MX7D_PAD_SD1_RESET_B__GPIO5_IO2 | MUX_PAD_CTRL(USDHC_PAD_CTRL),
+};
+
 static void setup_iomux_fec(void)
 {
 	imx_iomux_v3_setup_multiple_pads(fec1_pads, ARRAY_SIZE(fec1_pads));
@@ -218,6 +233,33 @@ int mmc_map_to_kernel_blk(int dev_no)
 	return dev_no;
 }
 
+#define USDHC1_CD_GPIO	IMX_GPIO_NR(5, 0)
+#define USDHC1_PWR_GPIO	IMX_GPIO_NR(5, 2)
+
+static struct fsl_esdhc_cfg usdhc_cfg[3] = {
+	{USDHC1_BASE_ADDR, 0, 4},
+};
+
+int board_mmc_getcd(struct mmc *mmc)
+{
+	return !gpio_get_value(USDHC1_CD_GPIO);
+}
+
+int board_mmc_init(bd_t *bis)
+{
+	imx_iomux_v3_setup_multiple_pads(usdhc1_pads, ARRAY_SIZE(usdhc1_pads));
+	gpio_request(USDHC1_CD_GPIO, "usdhc1_cd");
+	gpio_direction_input(USDHC1_CD_GPIO);
+	gpio_request(USDHC1_PWR_GPIO, "usdhc1_pwr");
+	gpio_direction_output(USDHC1_PWR_GPIO, 0);
+	udelay(500);
+	gpio_direction_output(USDHC1_PWR_GPIO, 1);
+	usdhc_cfg[0].sdhc_clk = mxc_get_clock(MXC_ESDHC_CLK);
+	return fsl_esdhc_initialize(bis, &usdhc_cfg[0]);
+
+	return 0;
+}
+
 #ifdef CONFIG_FEC_MXC
 int board_eth_init(bd_t *bis)
 {
diff --git a/configs/mx7dsabresd_defconfig b/configs/mx7dsabresd_defconfig
index 26250fd..9e07a94 100644
--- a/configs/mx7dsabresd_defconfig
+++ b/configs/mx7dsabresd_defconfig
@@ -44,7 +44,6 @@ CONFIG_DFU_RAM=y
 CONFIG_DM_GPIO=y
 CONFIG_DM_74X164=y
 CONFIG_DM_I2C=y
-CONFIG_DM_MMC=y
 CONFIG_SPI_FLASH=y
 CONFIG_SPI_FLASH_EON=y
 CONFIG_PHYLIB=y
-- 
2.7.4

^ permalink raw reply related	[flat|nested] 25+ messages in thread

* [U-Boot] [PATCH] mx7dsabresd: Remove CONFIG_DM_MMC to avoid regression
  2017-10-11 14:23 [U-Boot] [PATCH] mx7dsabresd: Remove CONFIG_DM_MMC to avoid regression Fabio Estevam
@ 2017-10-11 14:27 ` Otavio Salvador
  2017-10-11 15:38   ` Fabio Estevam
  0 siblings, 1 reply; 25+ messages in thread
From: Otavio Salvador @ 2017-10-11 14:27 UTC (permalink / raw)
  To: u-boot

On Wed, Oct 11, 2017 at 11:23 AM, Fabio Estevam <fabio.estevam@nxp.com> wrote:
> Since U-Boot 2017.09 it is no longer possible to boot a kernel from SD
> card:
>
> U-Boot 2017.11-rc1-00149-g38c8107-dirty (Oct 11 2017 - 09:55:22 -0300)
>
> CPU:   Freescale i.MX7D rev1.2 1000 MHz (running at 792 MHz)
> CPU:   Commercial temperature grade (0C to 95C) at 35C
> Reset cause: POR
> Model: Freescale i.MX7 SabreSD Board
> Board: i.MX7D SABRESD in non-secure mode
> DRAM:  1 GiB
> MMC:   MMC/SD: Reset never completed.
>
> MMC Device 0 not found
> *** Warning - No MMC card found, using default environment
>
> While a proper fix is not available, remove CONFIG_DM_MMC and register
> the USDHC1 port in the old way, so that this board can boot a kernel
> again.
>
> Signed-off-by: Fabio Estevam <fabio.estevam@nxp.com>

I understand your goal but I think we ought to focus on fixing the
root cause of it instead of moving things back out of DM. I even think
we could postpone a little the release until a fix is made but revert
out from DM seems wrong.

-- 
Otavio Salvador                             O.S. Systems
http://www.ossystems.com.br        http://code.ossystems.com.br
Mobile: +55 (53) 9981-7854            Mobile: +1 (347) 903-9750

^ permalink raw reply	[flat|nested] 25+ messages in thread

* [U-Boot] [PATCH] mx7dsabresd: Remove CONFIG_DM_MMC to avoid regression
  2017-10-11 14:27 ` Otavio Salvador
@ 2017-10-11 15:38   ` Fabio Estevam
  2017-10-11 20:29     ` [U-Boot] [PATCH] imx: mx7dsabresd: include BLK for MMC/eMMC support of driver model Eric Nelson
  0 siblings, 1 reply; 25+ messages in thread
From: Fabio Estevam @ 2017-10-11 15:38 UTC (permalink / raw)
  To: u-boot

On Wed, Oct 11, 2017 at 11:27 AM, Otavio Salvador
<otavio.salvador@ossystems.com.br> wrote:

> I understand your goal but I think we ought to focus on fixing the
> root cause of it instead of moving things back out of DM. I even think
> we could postpone a little the release until a fix is made but revert
> out from DM seems wrong.

Sure, patches are welcome!

However if the fix does not come in soon, I would really like to apply
this one so that U-Boot mainline can be used again on this board.

^ permalink raw reply	[flat|nested] 25+ messages in thread

* [U-Boot] [PATCH] imx: mx7dsabresd: include BLK for MMC/eMMC support of driver model
  2017-10-11 15:38   ` Fabio Estevam
@ 2017-10-11 20:29     ` Eric Nelson
  2017-10-11 20:41       ` Jagan Teki
                         ` (3 more replies)
  0 siblings, 4 replies; 25+ messages in thread
From: Eric Nelson @ 2017-10-11 20:29 UTC (permalink / raw)
  To: u-boot

Change-Id: I1bdfffe782a61a4c688f1bb56e85448024cd497b
Signed-off-by: Eric Nelson <eric@nelint.com>
---
 configs/mx7dsabresd_defconfig | 1 -
 1 file changed, 1 deletion(-)

diff --git a/configs/mx7dsabresd_defconfig b/configs/mx7dsabresd_defconfig
index 795c4f2..144fb50 100644
--- a/configs/mx7dsabresd_defconfig
+++ b/configs/mx7dsabresd_defconfig
@@ -38,7 +38,6 @@ CONFIG_CMD_EXT4=y
 CONFIG_CMD_EXT4_WRITE=y
 CONFIG_CMD_FAT=y
 CONFIG_OF_CONTROL=y
-# CONFIG_BLK is not set
 CONFIG_DFU_MMC=y
 CONFIG_DFU_RAM=y
 CONFIG_DM_GPIO=y
-- 
2.7.4

^ permalink raw reply related	[flat|nested] 25+ messages in thread

* [U-Boot] [PATCH] imx: mx7dsabresd: include BLK for MMC/eMMC support of driver model
  2017-10-11 20:29     ` [U-Boot] [PATCH] imx: mx7dsabresd: include BLK for MMC/eMMC support of driver model Eric Nelson
@ 2017-10-11 20:41       ` Jagan Teki
  2017-10-11 20:50       ` Fabio Estevam
                         ` (2 subsequent siblings)
  3 siblings, 0 replies; 25+ messages in thread
From: Jagan Teki @ 2017-10-11 20:41 UTC (permalink / raw)
  To: u-boot

On Thu, Oct 12, 2017 at 1:59 AM, Eric Nelson <eric@nelint.com> wrote:
> Change-Id: I1bdfffe782a61a4c688f1bb56e85448024cd497b
> Signed-off-by: Eric Nelson <eric@nelint.com>

Reviewed-by: Jagan Teki <jagan@openedev.com>

-- 
Jagan Teki
Free Software Engineer | www.openedev.com
U-Boot, Linux | Upstream Maintainer
Hyderabad, India.

^ permalink raw reply	[flat|nested] 25+ messages in thread

* [U-Boot] [PATCH] imx: mx7dsabresd: include BLK for MMC/eMMC support of driver model
  2017-10-11 20:29     ` [U-Boot] [PATCH] imx: mx7dsabresd: include BLK for MMC/eMMC support of driver model Eric Nelson
  2017-10-11 20:41       ` Jagan Teki
@ 2017-10-11 20:50       ` Fabio Estevam
  2017-10-11 20:51         ` Tom Rini
  2017-10-11 21:14         ` Eric Nelson
  2017-10-11 21:34         ` Eric Nelson
  2017-10-11 21:45       ` [U-Boot] " Eric Nelson
  3 siblings, 2 replies; 25+ messages in thread
From: Fabio Estevam @ 2017-10-11 20:50 UTC (permalink / raw)
  To: u-boot

Hi Eric,

That was the fix I was waiting for, thanks!

On Wed, Oct 11, 2017 at 5:29 PM, Eric Nelson <eric@nelint.com> wrote:

Please add a commit log and explain that this fixes a regression.

> Change-Id: I1bdfffe782a61a4c688f1bb56e85448024cd497b

Please remove this line.

Also, when you send a v2, please do the same change in
mx7dsabresd_secure_defconfig.

> Signed-off-by: Eric Nelson <eric@nelint.com>

Now I can boot a kernel successfully, thanks:

Tested-by: Fabio Estevam <fabio.estevam@nxp.com>

^ permalink raw reply	[flat|nested] 25+ messages in thread

* [U-Boot] [PATCH] imx: mx7dsabresd: include BLK for MMC/eMMC support of driver model
  2017-10-11 20:50       ` Fabio Estevam
@ 2017-10-11 20:51         ` Tom Rini
  2017-10-11 21:40           ` Eric Nelson
  2017-10-11 21:14         ` Eric Nelson
  1 sibling, 1 reply; 25+ messages in thread
From: Tom Rini @ 2017-10-11 20:51 UTC (permalink / raw)
  To: u-boot

On Wed, Oct 11, 2017 at 05:50:04PM -0300, Fabio Estevam wrote:
> Hi Eric,
> 
> That was the fix I was waiting for, thanks!
> 
> On Wed, Oct 11, 2017 at 5:29 PM, Eric Nelson <eric@nelint.com> wrote:
> 
> Please add a commit log and explain that this fixes a regression.
> 
> > Change-Id: I1bdfffe782a61a4c688f1bb56e85448024cd497b
> 
> Please remove this line.
> 
> Also, when you send a v2, please do the same change in
> mx7dsabresd_secure_defconfig.
> 
> > Signed-off-by: Eric Nelson <eric@nelint.com>
> 
> Now I can boot a kernel successfully, thanks:
> 
> Tested-by: Fabio Estevam <fabio.estevam@nxp.com>

Is there perhaps a dependency here we need to enforce via Kconfig so
this isn't a game of whack-a-mole?  Thanks!

-- 
Tom
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 819 bytes
Desc: not available
URL: <http://lists.denx.de/pipermail/u-boot/attachments/20171011/468d3322/attachment.sig>

^ permalink raw reply	[flat|nested] 25+ messages in thread

* [U-Boot] [PATCH] imx: mx7dsabresd: include BLK for MMC/eMMC support of driver model
  2017-10-11 20:50       ` Fabio Estevam
  2017-10-11 20:51         ` Tom Rini
@ 2017-10-11 21:14         ` Eric Nelson
  1 sibling, 0 replies; 25+ messages in thread
From: Eric Nelson @ 2017-10-11 21:14 UTC (permalink / raw)
  To: u-boot

Hi Fabio,

On 10/11/2017 01:50 PM, Fabio Estevam wrote:
> Hi Eric,
> 
> That was the fix I was waiting for, thanks!
> 

Glad to hear it.

> On Wed, Oct 11, 2017 at 5:29 PM, Eric Nelson <eric@nelint.com> wrote:
> 
> Please add a commit log and explain that this fixes a regression.
> 

Okay. If you insist ;)

>> Change-Id: I1bdfffe782a61a4c688f1bb56e85448024cd497b
> 
> Please remove this line.
> 

Oops. I think we need to refresh checkpatch. I think the
version in the kernel tree checks for that.

> Also, when you send a v2, please do the same change in
> mx7dsabresd_secure_defconfig.
> 

Okay.

>> Signed-off-by: Eric Nelson <eric@nelint.com>
> 
> Now I can boot a kernel successfully, thanks:
> 
> Tested-by: Fabio Estevam <fabio.estevam@nxp.com>
> 

Great.

^ permalink raw reply	[flat|nested] 25+ messages in thread

* [PATCH V2] imx: mx7dsabresd: include BLK for MMC/eMMC support of driver model
@ 2017-10-11 21:34         ` Eric Nelson
  0 siblings, 0 replies; 25+ messages in thread
From: Eric Nelson @ 2017-10-11 21:34 UTC (permalink / raw)
  To: linux-rtc
  Cc: a.zummo, alexandre.belloni, robh+dt, mark.rutland, devicetree,
	otavio.salvador, peng.fan, Eric Nelson

Commit 6fbbcfd introduced device-tree support for MMC devices on
the mx7sabresd boards and didn't include BLK, which requires BLK.

Commit 8ae5bb3 did the same for secure boot.

Fix both by allowing blk-uclass (BLK) support.

Tested-by: Fabio Estevam <festevam@gmail.com>
Signed-off-by: Eric Nelson <eric@nelint.com>
---
V2 includes the updated to mx7dsabresd_secure_defconfig
 configs/mx7dsabresd_defconfig        | 1 -
 configs/mx7dsabresd_secure_defconfig | 1 -
 2 files changed, 2 deletions(-)

diff --git a/configs/mx7dsabresd_defconfig b/configs/mx7dsabresd_defconfig
index 795c4f2..144fb50 100644
--- a/configs/mx7dsabresd_defconfig
+++ b/configs/mx7dsabresd_defconfig
@@ -38,7 +38,6 @@ CONFIG_CMD_EXT4=y
 CONFIG_CMD_EXT4_WRITE=y
 CONFIG_CMD_FAT=y
 CONFIG_OF_CONTROL=y
-# CONFIG_BLK is not set
 CONFIG_DFU_MMC=y
 CONFIG_DFU_RAM=y
 CONFIG_DM_GPIO=y
diff --git a/configs/mx7dsabresd_secure_defconfig b/configs/mx7dsabresd_secure_defconfig
index bd68831..d1af138 100644
--- a/configs/mx7dsabresd_secure_defconfig
+++ b/configs/mx7dsabresd_secure_defconfig
@@ -40,7 +40,6 @@ CONFIG_CMD_EXT4=y
 CONFIG_CMD_EXT4_WRITE=y
 CONFIG_CMD_FAT=y
 CONFIG_OF_CONTROL=y
-# CONFIG_BLK is not set
 CONFIG_DFU_MMC=y
 CONFIG_DFU_RAM=y
 CONFIG_DM_GPIO=y
-- 
2.7.4

^ permalink raw reply related	[flat|nested] 25+ messages in thread

* [PATCH V2] imx: mx7dsabresd: include BLK for MMC/eMMC support of driver model
@ 2017-10-11 21:34         ` Eric Nelson
  0 siblings, 0 replies; 25+ messages in thread
From: Eric Nelson @ 2017-10-11 21:34 UTC (permalink / raw)
  To: linux-rtc-u79uwXL29TY76Z2rM5mHXA
  Cc: a.zummo-BfzFCNDTiLLj+vYz1yj4TQ,
	alexandre.belloni-wi1+55ScJUtKEb57/3fJTNBPR1lH4CV8,
	robh+dt-DgEjT+Ai2ygdnm+yROfE0A, mark.rutland-5wv7dgnIgG8,
	devicetree-u79uwXL29TY76Z2rM5mHXA,
	otavio.salvador-fKevB0iiKLMBZ+LybsDmbA, peng.fan-3arQi8VN3Tc,
	Eric Nelson

Commit 6fbbcfd introduced device-tree support for MMC devices on
the mx7sabresd boards and didn't include BLK, which requires BLK.

Commit 8ae5bb3 did the same for secure boot.

Fix both by allowing blk-uclass (BLK) support.

Tested-by: Fabio Estevam <festevam-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>
Signed-off-by: Eric Nelson <eric-SeqgQ6RdavfQT0dZR+AlfA@public.gmane.org>
---
V2 includes the updated to mx7dsabresd_secure_defconfig
 configs/mx7dsabresd_defconfig        | 1 -
 configs/mx7dsabresd_secure_defconfig | 1 -
 2 files changed, 2 deletions(-)

diff --git a/configs/mx7dsabresd_defconfig b/configs/mx7dsabresd_defconfig
index 795c4f2..144fb50 100644
--- a/configs/mx7dsabresd_defconfig
+++ b/configs/mx7dsabresd_defconfig
@@ -38,7 +38,6 @@ CONFIG_CMD_EXT4=y
 CONFIG_CMD_EXT4_WRITE=y
 CONFIG_CMD_FAT=y
 CONFIG_OF_CONTROL=y
-# CONFIG_BLK is not set
 CONFIG_DFU_MMC=y
 CONFIG_DFU_RAM=y
 CONFIG_DM_GPIO=y
diff --git a/configs/mx7dsabresd_secure_defconfig b/configs/mx7dsabresd_secure_defconfig
index bd68831..d1af138 100644
--- a/configs/mx7dsabresd_secure_defconfig
+++ b/configs/mx7dsabresd_secure_defconfig
@@ -40,7 +40,6 @@ CONFIG_CMD_EXT4=y
 CONFIG_CMD_EXT4_WRITE=y
 CONFIG_CMD_FAT=y
 CONFIG_OF_CONTROL=y
-# CONFIG_BLK is not set
 CONFIG_DFU_MMC=y
 CONFIG_DFU_RAM=y
 CONFIG_DM_GPIO=y
-- 
2.7.4

--
To unsubscribe from this list: send the line "unsubscribe devicetree" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

^ permalink raw reply related	[flat|nested] 25+ messages in thread

* [U-Boot] [PATCH] imx: mx7dsabresd: include BLK for MMC/eMMC support of driver model
  2017-10-11 20:51         ` Tom Rini
@ 2017-10-11 21:40           ` Eric Nelson
  0 siblings, 0 replies; 25+ messages in thread
From: Eric Nelson @ 2017-10-11 21:40 UTC (permalink / raw)
  To: u-boot

Hi Tom,

On 10/11/2017 01:51 PM, Tom Rini wrote:
> On Wed, Oct 11, 2017 at 05:50:04PM -0300, Fabio Estevam wrote:
>> Hi Eric,
>>
>> That was the fix I was waiting for, thanks!
>>
>> On Wed, Oct 11, 2017 at 5:29 PM, Eric Nelson <eric@nelint.com> wrote:
>>
>> Please add a commit log and explain that this fixes a regression.
>>
>>> Change-Id: I1bdfffe782a61a4c688f1bb56e85448024cd497b
>>
>> Please remove this line.
>>
>> Also, when you send a v2, please do the same change in
>> mx7dsabresd_secure_defconfig.
>>
>>> Signed-off-by: Eric Nelson <eric@nelint.com>
>>
>> Now I can boot a kernel successfully, thanks:
>>
>> Tested-by: Fabio Estevam <fabio.estevam@nxp.com>
> 
> Is there perhaps a dependency here we need to enforce via Kconfig so
> this isn't a game of whack-a-mole?  Thanks!
> 

We have a "default Y if DM" for it, which should be enough.

http://git.denx.de/?p=u-boot.git;a=blob;f=drivers/block/Kconfig;h=26760895f99dd53370f9077f5b7213a1a6f241fe;hb=HEAD#l3

Commit 6fbbcfd explicitly disabled it when converting to
DM_MMC (secure_defconfig followed suit in commit 8ae5bb3), and
that's where the trouble came in.

Hmmm. A quick search shows that quite a few configurations
have this choice (CONFIG_DM_MMC without CONFIG_BLK):

~u-boot/configs$ git grep -l CONFIG_DM_MMC=y | xargs grep BLK
am335x_boneblack_vboot_defconfig:# CONFIG_BLK is not set
am335x_evm_defconfig:# CONFIG_BLK is not set
am335x_hs_evm_defconfig:# CONFIG_BLK is not set
am43xx_evm_defconfig:# CONFIG_BLK is not set
am43xx_evm_usbhost_boot_defconfig:# CONFIG_BLK is not set
am43xx_hs_evm_defconfig:# CONFIG_BLK is not set
am57xx_evm_defconfig:# CONFIG_BLK is not set
am57xx_hs_evm_defconfig:# CONFIG_BLK is not set
k2g_evm_defconfig:# CONFIG_BLK is not set
k2g_hs_evm_defconfig:# CONFIG_BLK is not set
ls1012aqds_qspi_defconfig:# CONFIG_BLK is not set
ls1012ardb_qspi_defconfig:# CONFIG_BLK is not set
mx6slevk_defconfig:# CONFIG_BLK is not set
mx6slevk_spinor_defconfig:# CONFIG_BLK is not set
mx6sllevk_defconfig:# CONFIG_BLK is not set
mx6sllevk_plugin_defconfig:# CONFIG_BLK is not set
mx6sxsabreauto_defconfig:# CONFIG_BLK is not set
mx6ull_14x14_evk_defconfig:# CONFIG_BLK is not set
mx6ull_14x14_evk_plugin_defconfig:# CONFIG_BLK is not set
mx7ulp_evk_defconfig:# CONFIG_BLK is not set
mx7ulp_evk_plugin_defconfig:# CONFIG_BLK is not set
omap3_logic_defconfig:# CONFIG_BLK is not set
pic32mzdask_defconfig:# CONFIG_BLK is not set

Are all of these broken?

I don't have any of the other boards, and am not in a
position to say whether there's a legitimate use case
for DM_MMC without BLK.

^ permalink raw reply	[flat|nested] 25+ messages in thread

* Re: [PATCH V2] imx: mx7dsabresd: include BLK for MMC/eMMC support of driver model
@ 2017-10-11 21:41           ` Eric Nelson
  0 siblings, 0 replies; 25+ messages in thread
From: Eric Nelson @ 2017-10-11 21:41 UTC (permalink / raw)
  To: linux-rtc
  Cc: a.zummo, alexandre.belloni, robh+dt, mark.rutland, devicetree,
	otavio.salvador, peng.fan

Please disregard.

I chose the wrong entry in my command-line history and
sent this to the wrong list.

On 10/11/2017 02:34 PM, Eric Nelson wrote:
> Commit 6fbbcfd introduced device-tree support for MMC devices on
> the mx7sabresd boards and didn't include BLK, which requires BLK.
> 
> Commit 8ae5bb3 did the same for secure boot.
> 
> Fix both by allowing blk-uclass (BLK) support.
> 
> Tested-by: Fabio Estevam <festevam@gmail.com>
> Signed-off-by: Eric Nelson <eric@nelint.com>
> ---
> V2 includes the updated to mx7dsabresd_secure_defconfig
>   configs/mx7dsabresd_defconfig        | 1 -
>   configs/mx7dsabresd_secure_defconfig | 1 -
>   2 files changed, 2 deletions(-)
> 
> diff --git a/configs/mx7dsabresd_defconfig b/configs/mx7dsabresd_defconfig
> index 795c4f2..144fb50 100644
> --- a/configs/mx7dsabresd_defconfig
> +++ b/configs/mx7dsabresd_defconfig
> @@ -38,7 +38,6 @@ CONFIG_CMD_EXT4=y
>   CONFIG_CMD_EXT4_WRITE=y
>   CONFIG_CMD_FAT=y
>   CONFIG_OF_CONTROL=y
> -# CONFIG_BLK is not set
>   CONFIG_DFU_MMC=y
>   CONFIG_DFU_RAM=y
>   CONFIG_DM_GPIO=y
> diff --git a/configs/mx7dsabresd_secure_defconfig b/configs/mx7dsabresd_secure_defconfig
> index bd68831..d1af138 100644
> --- a/configs/mx7dsabresd_secure_defconfig
> +++ b/configs/mx7dsabresd_secure_defconfig
> @@ -40,7 +40,6 @@ CONFIG_CMD_EXT4=y
>   CONFIG_CMD_EXT4_WRITE=y
>   CONFIG_CMD_FAT=y
>   CONFIG_OF_CONTROL=y
> -# CONFIG_BLK is not set
>   CONFIG_DFU_MMC=y
>   CONFIG_DFU_RAM=y
>   CONFIG_DM_GPIO=y
> 

^ permalink raw reply	[flat|nested] 25+ messages in thread

* Re: [PATCH V2] imx: mx7dsabresd: include BLK for MMC/eMMC support of driver model
@ 2017-10-11 21:41           ` Eric Nelson
  0 siblings, 0 replies; 25+ messages in thread
From: Eric Nelson @ 2017-10-11 21:41 UTC (permalink / raw)
  To: linux-rtc-u79uwXL29TY76Z2rM5mHXA
  Cc: a.zummo-BfzFCNDTiLLj+vYz1yj4TQ,
	alexandre.belloni-wi1+55ScJUtKEb57/3fJTNBPR1lH4CV8,
	robh+dt-DgEjT+Ai2ygdnm+yROfE0A, mark.rutland-5wv7dgnIgG8,
	devicetree-u79uwXL29TY76Z2rM5mHXA,
	otavio.salvador-fKevB0iiKLMBZ+LybsDmbA, peng.fan-3arQi8VN3Tc

Please disregard.

I chose the wrong entry in my command-line history and
sent this to the wrong list.

On 10/11/2017 02:34 PM, Eric Nelson wrote:
> Commit 6fbbcfd introduced device-tree support for MMC devices on
> the mx7sabresd boards and didn't include BLK, which requires BLK.
> 
> Commit 8ae5bb3 did the same for secure boot.
> 
> Fix both by allowing blk-uclass (BLK) support.
> 
> Tested-by: Fabio Estevam <festevam-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>
> Signed-off-by: Eric Nelson <eric-SeqgQ6RdavfQT0dZR+AlfA@public.gmane.org>
> ---
> V2 includes the updated to mx7dsabresd_secure_defconfig
>   configs/mx7dsabresd_defconfig        | 1 -
>   configs/mx7dsabresd_secure_defconfig | 1 -
>   2 files changed, 2 deletions(-)
> 
> diff --git a/configs/mx7dsabresd_defconfig b/configs/mx7dsabresd_defconfig
> index 795c4f2..144fb50 100644
> --- a/configs/mx7dsabresd_defconfig
> +++ b/configs/mx7dsabresd_defconfig
> @@ -38,7 +38,6 @@ CONFIG_CMD_EXT4=y
>   CONFIG_CMD_EXT4_WRITE=y
>   CONFIG_CMD_FAT=y
>   CONFIG_OF_CONTROL=y
> -# CONFIG_BLK is not set
>   CONFIG_DFU_MMC=y
>   CONFIG_DFU_RAM=y
>   CONFIG_DM_GPIO=y
> diff --git a/configs/mx7dsabresd_secure_defconfig b/configs/mx7dsabresd_secure_defconfig
> index bd68831..d1af138 100644
> --- a/configs/mx7dsabresd_secure_defconfig
> +++ b/configs/mx7dsabresd_secure_defconfig
> @@ -40,7 +40,6 @@ CONFIG_CMD_EXT4=y
>   CONFIG_CMD_EXT4_WRITE=y
>   CONFIG_CMD_FAT=y
>   CONFIG_OF_CONTROL=y
> -# CONFIG_BLK is not set
>   CONFIG_DFU_MMC=y
>   CONFIG_DFU_RAM=y
>   CONFIG_DM_GPIO=y
> 

--
To unsubscribe from this list: send the line "unsubscribe devicetree" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

^ permalink raw reply	[flat|nested] 25+ messages in thread

* [U-Boot] [PATCH V2] imx: mx7dsabresd: include BLK for MMC/eMMC support of driver model
  2017-10-11 20:29     ` [U-Boot] [PATCH] imx: mx7dsabresd: include BLK for MMC/eMMC support of driver model Eric Nelson
                         ` (2 preceding siblings ...)
  2017-10-11 21:34         ` Eric Nelson
@ 2017-10-11 21:45       ` Eric Nelson
  2017-10-11 21:49         ` Tom Rini
  3 siblings, 1 reply; 25+ messages in thread
From: Eric Nelson @ 2017-10-11 21:45 UTC (permalink / raw)
  To: u-boot

Commit 6fbbcfd introduced device-tree support for MMC devices on
the mx7sabresd boards and didn't include BLK, which requires BLK.

Commit 8ae5bb3 did the same for secure boot.

Fix both by allowing blk-uclass (BLK) support.

Tested-by: Fabio Estevam <festevam@gmail.com>
Signed-off-by: Eric Nelson <eric@nelint.com>
---
V2 includes the updated to mx7dsabresd_secure_defconfig
 configs/mx7dsabresd_defconfig        | 1 -
 configs/mx7dsabresd_secure_defconfig | 1 -
 2 files changed, 2 deletions(-)

diff --git a/configs/mx7dsabresd_defconfig b/configs/mx7dsabresd_defconfig
index 795c4f2..144fb50 100644
--- a/configs/mx7dsabresd_defconfig
+++ b/configs/mx7dsabresd_defconfig
@@ -38,7 +38,6 @@ CONFIG_CMD_EXT4=y
 CONFIG_CMD_EXT4_WRITE=y
 CONFIG_CMD_FAT=y
 CONFIG_OF_CONTROL=y
-# CONFIG_BLK is not set
 CONFIG_DFU_MMC=y
 CONFIG_DFU_RAM=y
 CONFIG_DM_GPIO=y
diff --git a/configs/mx7dsabresd_secure_defconfig b/configs/mx7dsabresd_secure_defconfig
index bd68831..d1af138 100644
--- a/configs/mx7dsabresd_secure_defconfig
+++ b/configs/mx7dsabresd_secure_defconfig
@@ -40,7 +40,6 @@ CONFIG_CMD_EXT4=y
 CONFIG_CMD_EXT4_WRITE=y
 CONFIG_CMD_FAT=y
 CONFIG_OF_CONTROL=y
-# CONFIG_BLK is not set
 CONFIG_DFU_MMC=y
 CONFIG_DFU_RAM=y
 CONFIG_DM_GPIO=y
-- 
2.7.4

^ permalink raw reply related	[flat|nested] 25+ messages in thread

* [U-Boot] [PATCH V2] imx: mx7dsabresd: include BLK for MMC/eMMC support of driver model
  2017-10-11 21:45       ` [U-Boot] " Eric Nelson
@ 2017-10-11 21:49         ` Tom Rini
  2017-10-11 21:53           ` Fabio Estevam
  0 siblings, 1 reply; 25+ messages in thread
From: Tom Rini @ 2017-10-11 21:49 UTC (permalink / raw)
  To: u-boot

On Wed, Oct 11, 2017 at 02:45:25PM -0700, Eric Nelson wrote:

> Commit 6fbbcfd introduced device-tree support for MMC devices on
> the mx7sabresd boards and didn't include BLK, which requires BLK.
> 
> Commit 8ae5bb3 did the same for secure boot.
> 
> Fix both by allowing blk-uclass (BLK) support.
> 
> Tested-by: Fabio Estevam <festevam@gmail.com>
> Signed-off-by: Eric Nelson <eric@nelint.com>
> ---
> V2 includes the updated to mx7dsabresd_secure_defconfig
>  configs/mx7dsabresd_defconfig        | 1 -
>  configs/mx7dsabresd_secure_defconfig | 1 -
>  2 files changed, 2 deletions(-)
> 
> diff --git a/configs/mx7dsabresd_defconfig b/configs/mx7dsabresd_defconfig
> index 795c4f2..144fb50 100644
> --- a/configs/mx7dsabresd_defconfig
> +++ b/configs/mx7dsabresd_defconfig
> @@ -38,7 +38,6 @@ CONFIG_CMD_EXT4=y
>  CONFIG_CMD_EXT4_WRITE=y
>  CONFIG_CMD_FAT=y
>  CONFIG_OF_CONTROL=y
> -# CONFIG_BLK is not set
>  CONFIG_DFU_MMC=y
>  CONFIG_DFU_RAM=y
>  CONFIG_DM_GPIO=y
> diff --git a/configs/mx7dsabresd_secure_defconfig b/configs/mx7dsabresd_secure_defconfig
> index bd68831..d1af138 100644
> --- a/configs/mx7dsabresd_secure_defconfig
> +++ b/configs/mx7dsabresd_secure_defconfig
> @@ -40,7 +40,6 @@ CONFIG_CMD_EXT4=y
>  CONFIG_CMD_EXT4_WRITE=y
>  CONFIG_CMD_FAT=y
>  CONFIG_OF_CONTROL=y
> -# CONFIG_BLK is not set
>  CONFIG_DFU_MMC=y
>  CONFIG_DFU_RAM=y
>  CONFIG_DM_GPIO=y

It sounds like BLK shouldn't be default y if DM_MMC but rather selected
by DM_MMC.  Yes?

-- 
Tom
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 819 bytes
Desc: not available
URL: <http://lists.denx.de/pipermail/u-boot/attachments/20171011/e04eae64/attachment.sig>

^ permalink raw reply	[flat|nested] 25+ messages in thread

* [U-Boot] [PATCH V2] imx: mx7dsabresd: include BLK for MMC/eMMC support of driver model
  2017-10-11 21:49         ` Tom Rini
@ 2017-10-11 21:53           ` Fabio Estevam
  2017-10-11 21:55             ` Tom Rini
  0 siblings, 1 reply; 25+ messages in thread
From: Fabio Estevam @ 2017-10-11 21:53 UTC (permalink / raw)
  To: u-boot

On Wed, Oct 11, 2017 at 6:49 PM, Tom Rini <trini@konsulko.com> wrote:

> It sounds like BLK shouldn't be default y if DM_MMC but rather selected
> by DM_MMC.  Yes?

What about this?

--- a/drivers/mmc/Kconfig
+++ b/drivers/mmc/Kconfig
@@ -13,6 +13,7 @@ config MMC
 config DM_MMC
        bool "Enable MMC controllers using Driver Model"
        depends on DM
+       select BLK
        help
          This enables the MultiMediaCard (MMC) uclass which supports MMC and
          Secure Digital I/O (SDIO) cards. Both removable (SD, micro-SD, etc.)

^ permalink raw reply	[flat|nested] 25+ messages in thread

* [U-Boot] [PATCH V2] imx: mx7dsabresd: include BLK for MMC/eMMC support of driver model
  2017-10-11 21:53           ` Fabio Estevam
@ 2017-10-11 21:55             ` Tom Rini
  2017-10-11 22:07               ` Fabio Estevam
  0 siblings, 1 reply; 25+ messages in thread
From: Tom Rini @ 2017-10-11 21:55 UTC (permalink / raw)
  To: u-boot

On Wed, Oct 11, 2017 at 06:53:13PM -0300, Fabio Estevam wrote:
> On Wed, Oct 11, 2017 at 6:49 PM, Tom Rini <trini@konsulko.com> wrote:
> 
> > It sounds like BLK shouldn't be default y if DM_MMC but rather selected
> > by DM_MMC.  Yes?
> 
> What about this?
> 
> --- a/drivers/mmc/Kconfig
> +++ b/drivers/mmc/Kconfig
> @@ -13,6 +13,7 @@ config MMC
>  config DM_MMC
>         bool "Enable MMC controllers using Driver Model"
>         depends on DM
> +       select BLK
>         help
>           This enables the MultiMediaCard (MMC) uclass which supports MMC and
>           Secure Digital I/O (SDIO) cards. Both removable (SD, micro-SD, etc.)

Yes, I think that's it, along with removing the default y if DM_MMC from
the BLK entry.  Thanks!

-- 
Tom
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 819 bytes
Desc: not available
URL: <http://lists.denx.de/pipermail/u-boot/attachments/20171011/90746733/attachment.sig>

^ permalink raw reply	[flat|nested] 25+ messages in thread

* [U-Boot] [PATCH V2] imx: mx7dsabresd: include BLK for MMC/eMMC support of driver model
  2017-10-11 21:55             ` Tom Rini
@ 2017-10-11 22:07               ` Fabio Estevam
  2017-10-11 22:17                 ` Tom Rini
  0 siblings, 1 reply; 25+ messages in thread
From: Fabio Estevam @ 2017-10-11 22:07 UTC (permalink / raw)
  To: u-boot

On Wed, Oct 11, 2017 at 6:55 PM, Tom Rini <trini@konsulko.com> wrote:

> Yes, I think that's it, along with removing the default y if DM_MMC from
> the BLK entry.  Thanks!

Ok, if I do as suggested:

--- a/drivers/block/Kconfig
+++ b/drivers/block/Kconfig
@@ -1,7 +1,6 @@
 config BLK
        bool "Support block devices"
        depends on DM
-       default y if DM_MMC
        help
          Enable support for block devices, such as SCSI, MMC and USB
          flash sticks. These provide a block-level interface which permits
diff --git a/drivers/mmc/Kconfig b/drivers/mmc/Kconfig
index 94050836..f4c953c 100644
--- a/drivers/mmc/Kconfig
+++ b/drivers/mmc/Kconfig
@@ -13,6 +13,7 @@ config MMC
 config DM_MMC
        bool "Enable MMC controllers using Driver Model"
        depends on DM
+       select BLK
        help
          This enables the MultiMediaCard (MMC) uclass which supports MMC and
          Secure Digital I/O (SDIO) cards. Both removable (SD, micro-SD, etc.)

Then mx7dsabresd_defconfig build and boots fine.

However, mx6slevk_defconfig fails to build like this:

  CC      common/usb_storage.o
common/usb_storage.c: In function ‘usb_stor_probe_device’:
common/usb_storage.c:208:30: error: ‘struct usb_device’ has no member
named ‘dev’
  data = dev_get_platdata(udev->dev);
                              ^
common/usb_storage.c:218:32: error: ‘struct usb_device’ has no member
named ‘dev’
   ret = blk_create_devicef(udev->dev, "usb_storage_blk", str,
                                ^
I can fix it by doing:

--- a/configs/mx6slevk_defconfig
+++ b/configs/mx6slevk_defconfig
@@ -44,6 +44,7 @@ CONFIG_DM_REGULATOR_FIXED=y
 CONFIG_DM_REGULATOR_GPIO=y
 CONFIG_DM_THERMAL=y
 CONFIG_USB=y
+CONFIG_DM_USB=y
 CONFIG_USB_STORAGE=y
 CONFIG_USB_HOST_ETHER=y
 CONFIG_USB_ETHER_ASIX=y

Is there a better fix for this breakage?

^ permalink raw reply related	[flat|nested] 25+ messages in thread

* [U-Boot] [PATCH V2] imx: mx7dsabresd: include BLK for MMC/eMMC support of driver model
  2017-10-11 22:07               ` Fabio Estevam
@ 2017-10-11 22:17                 ` Tom Rini
  2017-10-11 22:28                   ` Fabio Estevam
  2017-10-12 16:33                   ` Tom Rini
  0 siblings, 2 replies; 25+ messages in thread
From: Tom Rini @ 2017-10-11 22:17 UTC (permalink / raw)
  To: u-boot

On Wed, Oct 11, 2017 at 07:07:04PM -0300, Fabio Estevam wrote:
> On Wed, Oct 11, 2017 at 6:55 PM, Tom Rini <trini@konsulko.com> wrote:
> 
> > Yes, I think that's it, along with removing the default y if DM_MMC from
> > the BLK entry.  Thanks!
> 
> Ok, if I do as suggested:
> 
> --- a/drivers/block/Kconfig
> +++ b/drivers/block/Kconfig
> @@ -1,7 +1,6 @@
>  config BLK
>         bool "Support block devices"
>         depends on DM
> -       default y if DM_MMC
>         help
>           Enable support for block devices, such as SCSI, MMC and USB
>           flash sticks. These provide a block-level interface which permits
> diff --git a/drivers/mmc/Kconfig b/drivers/mmc/Kconfig
> index 94050836..f4c953c 100644
> --- a/drivers/mmc/Kconfig
> +++ b/drivers/mmc/Kconfig
> @@ -13,6 +13,7 @@ config MMC
>  config DM_MMC
>         bool "Enable MMC controllers using Driver Model"
>         depends on DM
> +       select BLK
>         help
>           This enables the MultiMediaCard (MMC) uclass which supports MMC and
>           Secure Digital I/O (SDIO) cards. Both removable (SD, micro-SD, etc.)
> 
> Then mx7dsabresd_defconfig build and boots fine.
> 
> However, mx6slevk_defconfig fails to build like this:
> 
>   CC      common/usb_storage.o
> common/usb_storage.c: In function ‘usb_stor_probe_device’:
> common/usb_storage.c:208:30: error: ‘struct usb_device’ has no member
> named ‘dev’
>   data = dev_get_platdata(udev->dev);
>                               ^
> common/usb_storage.c:218:32: error: ‘struct usb_device’ has no member
> named ‘dev’
>    ret = blk_create_devicef(udev->dev, "usb_storage_blk", str,
>                                 ^
> I can fix it by doing:
> 
> --- a/configs/mx6slevk_defconfig
> +++ b/configs/mx6slevk_defconfig
> @@ -44,6 +44,7 @@ CONFIG_DM_REGULATOR_FIXED=y
>  CONFIG_DM_REGULATOR_GPIO=y
>  CONFIG_DM_THERMAL=y
>  CONFIG_USB=y
> +CONFIG_DM_USB=y
>  CONFIG_USB_STORAGE=y
>  CONFIG_USB_HOST_ETHER=y
>  CONFIG_USB_ETHER_ASIX=y
> 
> Is there a better fix for this breakage?

Hmmm.  So, if you have DM and DM_MMC, you need BLK.  If you have DM and
BLK and USB, you also need DM_USB (and vice-versa, if you have DM and DM_USB and
not DM_MMC, things will break too, on another platform I bet).  So,
DM_USB should also select BLK I think.  The case like mx6slkevk where
you have DM_USB supported by not enabled are a bug of sorts where it
needs to also be enabling DM_USB.  I think I'll need to do a world build
to see what fails in this case, unless you want to do it?  Thanks!

-- 
Tom
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 819 bytes
Desc: not available
URL: <http://lists.denx.de/pipermail/u-boot/attachments/20171011/1521f14b/attachment.sig>

^ permalink raw reply	[flat|nested] 25+ messages in thread

* [U-Boot] [PATCH V2] imx: mx7dsabresd: include BLK for MMC/eMMC support of driver model
  2017-10-11 22:17                 ` Tom Rini
@ 2017-10-11 22:28                   ` Fabio Estevam
  2017-10-12 16:33                   ` Tom Rini
  1 sibling, 0 replies; 25+ messages in thread
From: Fabio Estevam @ 2017-10-11 22:28 UTC (permalink / raw)
  To: u-boot

On Wed, Oct 11, 2017 at 7:17 PM, Tom Rini <trini@konsulko.com> wrote:

> Hmmm.  So, if you have DM and DM_MMC, you need BLK.  If you have DM and
> BLK and USB, you also need DM_USB (and vice-versa, if you have DM and DM_USB and
> not DM_MMC, things will break too, on another platform I bet).  So,
> DM_USB should also select BLK I think.  The case like mx6slkevk where
> you have DM_USB supported by not enabled are a bug of sorts where it
> needs to also be enabling DM_USB.  I think I'll need to do a world build
> to see what fails in this case, unless you want to do it?  Thanks!

Please go ahead with the build. I am about to leave for holidays.

Thanks

^ permalink raw reply	[flat|nested] 25+ messages in thread

* [U-Boot] [PATCH V2] imx: mx7dsabresd: include BLK for MMC/eMMC support of driver model
  2017-10-11 22:17                 ` Tom Rini
  2017-10-11 22:28                   ` Fabio Estevam
@ 2017-10-12 16:33                   ` Tom Rini
  2017-10-12 18:35                     ` Fabio Estevam
  1 sibling, 1 reply; 25+ messages in thread
From: Tom Rini @ 2017-10-12 16:33 UTC (permalink / raw)
  To: u-boot

On Wed, Oct 11, 2017 at 06:17:12PM -0400, Tom Rini wrote:
> On Wed, Oct 11, 2017 at 07:07:04PM -0300, Fabio Estevam wrote:
> > On Wed, Oct 11, 2017 at 6:55 PM, Tom Rini <trini@konsulko.com> wrote:
> > 
> > > Yes, I think that's it, along with removing the default y if DM_MMC from
> > > the BLK entry.  Thanks!
> > 
> > Ok, if I do as suggested:
> > 
> > --- a/drivers/block/Kconfig
> > +++ b/drivers/block/Kconfig
> > @@ -1,7 +1,6 @@
> >  config BLK
> >         bool "Support block devices"
> >         depends on DM
> > -       default y if DM_MMC
> >         help
> >           Enable support for block devices, such as SCSI, MMC and USB
> >           flash sticks. These provide a block-level interface which permits
> > diff --git a/drivers/mmc/Kconfig b/drivers/mmc/Kconfig
> > index 94050836..f4c953c 100644
> > --- a/drivers/mmc/Kconfig
> > +++ b/drivers/mmc/Kconfig
> > @@ -13,6 +13,7 @@ config MMC
> >  config DM_MMC
> >         bool "Enable MMC controllers using Driver Model"
> >         depends on DM
> > +       select BLK
> >         help
> >           This enables the MultiMediaCard (MMC) uclass which supports MMC and
> >           Secure Digital I/O (SDIO) cards. Both removable (SD, micro-SD, etc.)
> > 
> > Then mx7dsabresd_defconfig build and boots fine.
> > 
> > However, mx6slevk_defconfig fails to build like this:
> > 
> >   CC      common/usb_storage.o
> > common/usb_storage.c: In function ‘usb_stor_probe_device’:
> > common/usb_storage.c:208:30: error: ‘struct usb_device’ has no member
> > named ‘dev’
> >   data = dev_get_platdata(udev->dev);
> >                               ^
> > common/usb_storage.c:218:32: error: ‘struct usb_device’ has no member
> > named ‘dev’
> >    ret = blk_create_devicef(udev->dev, "usb_storage_blk", str,
> >                                 ^
> > I can fix it by doing:
> > 
> > --- a/configs/mx6slevk_defconfig
> > +++ b/configs/mx6slevk_defconfig
> > @@ -44,6 +44,7 @@ CONFIG_DM_REGULATOR_FIXED=y
> >  CONFIG_DM_REGULATOR_GPIO=y
> >  CONFIG_DM_THERMAL=y
> >  CONFIG_USB=y
> > +CONFIG_DM_USB=y
> >  CONFIG_USB_STORAGE=y
> >  CONFIG_USB_HOST_ETHER=y
> >  CONFIG_USB_ETHER_ASIX=y
> > 
> > Is there a better fix for this breakage?
> 
> Hmmm.  So, if you have DM and DM_MMC, you need BLK.  If you have DM and
> BLK and USB, you also need DM_USB (and vice-versa, if you have DM and DM_USB and
> not DM_MMC, things will break too, on another platform I bet).  So,
> DM_USB should also select BLK I think.  The case like mx6slkevk where
> you have DM_USB supported by not enabled are a bug of sorts where it
> needs to also be enabling DM_USB.  I think I'll need to do a world build
> to see what fails in this case, unless you want to do it?  Thanks!

OK, disregard what I had been saying.  At this point, it's a matter of
correcting and testing boards to have either DM_MMC (and DM_USB and BLK)
on, or having DM_MMC off (because they want USB gadget support).

-- 
Tom
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 819 bytes
Desc: not available
URL: <http://lists.denx.de/pipermail/u-boot/attachments/20171012/ccff4eb2/attachment.sig>

^ permalink raw reply	[flat|nested] 25+ messages in thread

* [U-Boot] [PATCH V2] imx: mx7dsabresd: include BLK for MMC/eMMC support of driver model
  2017-10-12 16:33                   ` Tom Rini
@ 2017-10-12 18:35                     ` Fabio Estevam
  2017-10-12 18:38                       ` Tom Rini
  0 siblings, 1 reply; 25+ messages in thread
From: Fabio Estevam @ 2017-10-12 18:35 UTC (permalink / raw)
  To: u-boot

Tom,

On Thu, Oct 12, 2017 at 1:33 PM, Tom Rini <trini@konsulko.com> wrote:

> OK, disregard what I had been saying.  At this point, it's a matter of
> correcting and testing boards to have either DM_MMC (and DM_USB and BLK)
> on, or having DM_MMC off (because they want USB gadget support).

Here is what I tried:
https://pastebin.com/yYEdC5fA

but then am335x_hs_evm_defconfig fails like this:

  LD      spl/drivers/built-in.o
  LD      spl/u-boot-spl
/usr/bin/arm-linux-gnueabi-ld.bfd: u-boot-spl section `.rodata' will
not fit in region `.sram'
/usr/bin/arm-linux-gnueabi-ld.bfd: region `.sram' overflowed by 7212 bytes
scripts/Makefile.spl:358: recipe for target 'spl/u-boot-spl' failed
make[1]: *** [spl/u-boot-spl] Error 1
Makefile:1407: recipe for target 'spl/u-boot-spl' failed
make: *** [spl/u-boot-spl] Error

Suggestions?

^ permalink raw reply	[flat|nested] 25+ messages in thread

* [U-Boot] [PATCH V2] imx: mx7dsabresd: include BLK for MMC/eMMC support of driver model
  2017-10-12 18:35                     ` Fabio Estevam
@ 2017-10-12 18:38                       ` Tom Rini
  2017-10-12 18:39                         ` Fabio Estevam
  0 siblings, 1 reply; 25+ messages in thread
From: Tom Rini @ 2017-10-12 18:38 UTC (permalink / raw)
  To: u-boot

On Thu, Oct 12, 2017 at 03:35:38PM -0300, Fabio Estevam wrote:
> Tom,
> 
> On Thu, Oct 12, 2017 at 1:33 PM, Tom Rini <trini@konsulko.com> wrote:
> 
> > OK, disregard what I had been saying.  At this point, it's a matter of
> > correcting and testing boards to have either DM_MMC (and DM_USB and BLK)
> > on, or having DM_MMC off (because they want USB gadget support).
> 
> Here is what I tried:
> https://pastebin.com/yYEdC5fA
> 
> but then am335x_hs_evm_defconfig fails like this:
> 
>   LD      spl/drivers/built-in.o
>   LD      spl/u-boot-spl
> /usr/bin/arm-linux-gnueabi-ld.bfd: u-boot-spl section `.rodata' will
> not fit in region `.sram'
> /usr/bin/arm-linux-gnueabi-ld.bfd: region `.sram' overflowed by 7212 bytes
> scripts/Makefile.spl:358: recipe for target 'spl/u-boot-spl' failed
> make[1]: *** [spl/u-boot-spl] Error 1
> Makefile:1407: recipe for target 'spl/u-boot-spl' failed
> make: *** [spl/u-boot-spl] Error
> 
> Suggestions?

Yes, just work on the imx* targets.  For am33xx/am43xx/am57xx we don't
want to swap things around as we then get rid of USB gadget support.
It's a bit tricky atm to get all cases covered as, iirc, there's
something outstanding for the USB gadget case for DM.

-- 
Tom
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 819 bytes
Desc: not available
URL: <http://lists.denx.de/pipermail/u-boot/attachments/20171012/bee18484/attachment.sig>

^ permalink raw reply	[flat|nested] 25+ messages in thread

* [U-Boot] [PATCH V2] imx: mx7dsabresd: include BLK for MMC/eMMC support of driver model
  2017-10-12 18:38                       ` Tom Rini
@ 2017-10-12 18:39                         ` Fabio Estevam
  2017-10-12 18:49                           ` Tom Rini
  0 siblings, 1 reply; 25+ messages in thread
From: Fabio Estevam @ 2017-10-12 18:39 UTC (permalink / raw)
  To: u-boot

On Thu, Oct 12, 2017 at 3:38 PM, Tom Rini <trini@konsulko.com> wrote:

> Yes, just work on the imx* targets.  For am33xx/am43xx/am57xx we don't
> want to swap things around as we then get rid of USB gadget support.
> It's a bit tricky atm to get all cases covered as, iirc, there's
> something outstanding for the USB gadget case for DM.

Ok, so in this case, should I fix this by touching only the imx6 defconfigs?

^ permalink raw reply	[flat|nested] 25+ messages in thread

* [U-Boot] [PATCH V2] imx: mx7dsabresd: include BLK for MMC/eMMC support of driver model
  2017-10-12 18:39                         ` Fabio Estevam
@ 2017-10-12 18:49                           ` Tom Rini
  0 siblings, 0 replies; 25+ messages in thread
From: Tom Rini @ 2017-10-12 18:49 UTC (permalink / raw)
  To: u-boot

On Thu, Oct 12, 2017 at 03:39:56PM -0300, Fabio Estevam wrote:
> On Thu, Oct 12, 2017 at 3:38 PM, Tom Rini <trini@konsulko.com> wrote:
> 
> > Yes, just work on the imx* targets.  For am33xx/am43xx/am57xx we don't
> > want to swap things around as we then get rid of USB gadget support.
> > It's a bit tricky atm to get all cases covered as, iirc, there's
> > something outstanding for the USB gadget case for DM.
> 
> Ok, so in this case, should I fix this by touching only the imx6 defconfigs?

Yes, thanks.

-- 
Tom
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 819 bytes
Desc: not available
URL: <http://lists.denx.de/pipermail/u-boot/attachments/20171012/fae31843/attachment.sig>

^ permalink raw reply	[flat|nested] 25+ messages in thread

end of thread, other threads:[~2017-10-12 18:49 UTC | newest]

Thread overview: 25+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2017-10-11 14:23 [U-Boot] [PATCH] mx7dsabresd: Remove CONFIG_DM_MMC to avoid regression Fabio Estevam
2017-10-11 14:27 ` Otavio Salvador
2017-10-11 15:38   ` Fabio Estevam
2017-10-11 20:29     ` [U-Boot] [PATCH] imx: mx7dsabresd: include BLK for MMC/eMMC support of driver model Eric Nelson
2017-10-11 20:41       ` Jagan Teki
2017-10-11 20:50       ` Fabio Estevam
2017-10-11 20:51         ` Tom Rini
2017-10-11 21:40           ` Eric Nelson
2017-10-11 21:14         ` Eric Nelson
2017-10-11 21:34       ` [PATCH V2] " Eric Nelson
2017-10-11 21:34         ` Eric Nelson
2017-10-11 21:41         ` Eric Nelson
2017-10-11 21:41           ` Eric Nelson
2017-10-11 21:45       ` [U-Boot] " Eric Nelson
2017-10-11 21:49         ` Tom Rini
2017-10-11 21:53           ` Fabio Estevam
2017-10-11 21:55             ` Tom Rini
2017-10-11 22:07               ` Fabio Estevam
2017-10-11 22:17                 ` Tom Rini
2017-10-11 22:28                   ` Fabio Estevam
2017-10-12 16:33                   ` Tom Rini
2017-10-12 18:35                     ` Fabio Estevam
2017-10-12 18:38                       ` Tom Rini
2017-10-12 18:39                         ` Fabio Estevam
2017-10-12 18:49                           ` Tom Rini

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.