mx6: Allow mx6 to access the IPUv3 registers

Rename CONFIG_VIDEO_MX5 as this driver can also be used on mx6.

Signed-off-by Fabio Estevam <[email protected]>

---
 drivers/video/Makefile      |    2 +-
 drivers/video/cfb_console.c |    2 +-
 include/configs/mx51evk.h   |    2 +-
 include/configs/mx53loco.h  |    2 +-
 include/configs/vision2.h   |    2 +-
 5 files changed, 5 insertions(+), 5 deletions(-)

The following erratum :

"ENGcm08316
IPU: Clarification regarding the bypass mode registers setup for display and camera interfaces"

,only applies to mx51, so restrict its usage for this SoC only.

Signed-off-by Fabio Estevam <[email protected]>

---
 drivers/video/ipu_common.c |    2 ++
 1 files changed, 2 insertions(+), 0 deletions(-)

The registers accessed inside clk_ipu_enable/disable are not present on MX6, so make sure they only run on MX51 and MX53.

Signed-off-by Fabio Estevam <[email protected]>

---
 drivers/video/ipu_common.c |    5 ++++-
 1 files changed, 4 insertions(+), 1 deletions(-)

Rename MXC_CCM_BASE to CCM_BASE_ADDR as this is already defined for MX6.

Signed-off-by Fabio Estevam <[email protected]>

---
 drivers/video/ipu_common.c |    2 +-
 1 files changed, 1 insertions(+), 1 deletions(-)

Do not hardcode the ipu_clk frequency and let the board file pass this value.

Signed-off-by Fabio Estevam <[email protected]>

---
 drivers/video/ipu_common.c |    2 +-
 include/configs/mx51evk.h  |    1 +
 include/configs/mx53loco.h |    1 +
 include/configs/vision2.h  |    1 +
 4 files changed, 4 insertions(+), 1 deletions(-)

Allow MX6Q I2C3 to work.

Cc Heiko Schocher<[email protected]>

Signed-off-by Fabio Estevam<[email protected]>

---
  drivers/i2c/mxc_i2c.c |    2 ++
  1 files changed, 2 insertions(+), 0 deletions(-)

drivers/i2c/mxc_i2c.c +0 -0

index c88ac7c..d3c323e 100644

This conflicts with the patch that Stefano acked on May 6.

[PATCH 1/3] mxc_i2c: specify i2c base address in config file

The following platforms had their config files changed flea3, imx31_phycore, mx35pdk, mx53ard, mx53evk, mx53smd

Signed-off-by Troy Kisky<[email protected]>

Hi Troy,

Quoted 3 lines · Troy Kisky

This conflicts with the patch that Stefano acked on May 6.

Quoted 6 lines

[PATCH 1/3] mxc_i2c: specify i2c base address in config file

The following platforms had their config files changed flea3, imx31_phycore, mx35pdk, mx53ard, mx53evk, mx53smd

Signed-off-by Troy Kisky<[email protected]>

Ok, I can rework this patch after your commit gets in.

From: Eric Nelson <[email protected]>

Add ldb_clk for use in parenting the pixel clock.

Signed-off-by Eric Nelson <[email protected]>

Signed-off-by Fabio Estevam <[email protected]>

---
 drivers/video/ipu_common.c |   12 ++++++++++--
 drivers/video/ipu_disp.c   |    3 ++-
 2 files changed, 12 insertions(+), 3 deletions(-)

Add splaschscreen support.

Quoted 2 lines · Fabio Estevam

Add splaschscreen support.

Here and in the subject a typo? splaschscreen -> splashscreen?

Hi Dirk,

Quoted 6 lines · Dirk Behme

Quoted 2 lines · Fabio Estevam

Add splaschscreen support.

Here and in the subject a typo? splaschscreen -> splashscreen?

Thanks, I will fix this and rework this patch according to your I2C suggestion.

Maybe you could check [1]

(quoted diff) +1 -1

-#define CONFIG_SYS_I2C_SLAVE ? ? ? ? ? ?0xfe
+#define CONFIG_SYS_I2C_SLAVE ? ?0x1f

then? I'm no I2C expert and I wasn't sure what to use for CONFIG_SYS_I2C_SLAVE.

Hi Dirk,

On Sat, Jun 2, 2012 at 3:52 PM, Dirk Behme<[email protected]> wrote:

Quoted 7 lines

Maybe you could check [1]

(quoted diff) +1 -1

-#define CONFIG_SYS_I2C_SLAVE            0xfe
+#define CONFIG_SYS_I2C_SLAVE    0x1f

then? I'm no I2C expert and I wasn't sure what to use for CONFIG_SYS_I2C_SLAVE.

Yes, 0x1f is the correct address for the Hannstar LVDS I2C slave address.

If 0x1f is the Hannstar LVDS address, then you definitely should NOT use 0x1f for the

CONFIG_SYS_I2C_SLAVE value.

CONFIG_SYS_I2C_SLAVE is meant to define which address the processor will respond to
in a multi-master bus environment. However, as mxc_i2c does not support slave mode
the entire discussion is moot.

Add splaschscreen support.

It was used a Hannstar 1024 x 768 LVDS panel that can be connected to the mx6qsabrelite board.

Signed-off-by Fabio Estevam <[email protected]>

---
I understand this patch will need to be improved, but I am posting it now, so that others can possibly test it.

Series includes one patch to fix a register name in imx6/crm_regs.h and a second that's a re-base of Fabio's patch from 5/31.

        http://patchwork.ozlabs.org/patch/162206/

Note that I'm not sure whether this should have been based on u-boot-video because it's video-related or u-boot imx because the bulk of the patch set is board-specific.

Adds support for the Hannstar 1024 x 768 LVDS panel (Freescale part number MCIMX-LVDS1) to SABRE-Lite board.

This commit is a rebase Fabio Estevan's patch from 5/31 to

Hi Eric,

Quoted 4 lines · Eric Nelson

Adds support for the Hannstar 1024 x 768 LVDS panel (Freescale part number MCIMX-LVDS1) to SABRE-Lite board.

This commit is a rebase Fabio Estevan's patch from 5/31 to

s/Estevan/Estevam

board/freescale/mx6qsabrelite/mx6qsabrelite.c int board_init(void) · line 490

Quoted 5 lines

board/freescale/mx6qsabrelite/mx6qsabrelite.c +5 -0

+int board_late_init(void)
+{
+       setenv("stdout", "serial");
+       return 0;
+}

Please see my previous comment on v1 about this.

Wolfgang nacked this method and Stefano has proposed a nice way for keeping the console in serial. Check mx53loco and mx51evk.

Hi Fabio,

Quoted 12 lines · Fabio Estevam

Hi Eric,

<snip >

board/freescale/mx6qsabrelite/mx6qsabrelite.c int board_init(void) · line 490

Quoted 5 lines

board/freescale/mx6qsabrelite/mx6qsabrelite.c +5 -0

+int board_late_init(void)
+{
+ setenv("stdout", "serial");
+ return 0;
+}

Please see my previous comment on v1 about this.

Quoted 3 lines

Wolfgang nacked this method and Stefano has proposed a nice way for keeping the console in serial. Check mx53loco and mx51evk.

Thanks. I'll look into it.

Like this?

        http://lists.denx.de/pipermail/u-boot/2012-August/129967.html

Hi Eric,

Quoted 4 lines · Eric Nelson

Adds support for the Hannstar 1024 x 768 LVDS panel (Freescale part number MCIMX-LVDS1) to SABRE-Lite board.

This commit is a rebase Fabio Estevan's patch from 5/31 to

s/Estevan/Estevam

board/freescale/mx6qsabrelite/mx6qsabrelite.c int board_init(void) · line 490

Quoted 5 lines

board/freescale/mx6qsabrelite/mx6qsabrelite.c +5 -0

+int board_late_init(void)
+{
+       setenv("stdout", "serial");
+       return 0;
+}

Please see my previous comment on v1 about this.

Wolfgang nacked this method and Stefano has proposed a nice way for keeping the console in serial. Check mx53loco and mx51evk.

Hi Fabio and Stefano,

While trying to understand this, I wonder whether the use of overwrite_console() addresses Wolfgang's original concern.

I think the suggestion was to set the preboot variable to change things:

#define CONFIG_PREBOOT "setenv stdout serial"

The has the benefit of allowing saveenv to completely overwrite the decision.

Am I over-thinking this?

Quoted 8 lines · Eric Nelson

Hi Fabio and Stefano,

While trying to understand this, I wonder whether the use of overwrite_console() addresses Wolfgang's original concern.

I think the suggestion was to set the preboot variable to change things:

Wolfgang's first concern was to avoid to hardcode variable inside the code, as it was done in the original mx51 / mx53 code (and not only in i.MX boards).

Quoted 1 line

Quoted 1 line

#define CONFIG_PREBOOT "setenv stdout serial"

This makes sense if you want maintain the possibility that u-boot output is displayed on the LCD. Is this really the case ?

In most cases, we want to display a splashscreen, while the console is still managed by serial line.

If this is the case, adding overwrite_console() tells video subsystem to not change the stdout variable, and then it is not required to set it back to serial neither in code nor with the help of the preboot variable.

Quoted 4 lines

The has the benefit of allowing saveenv to completely overwrite the decision.

Am I over-thinking this?

It depends if you want also U-Boot's output on the display or only a picture.

Adds support for the Hannstar 1024 x 768 LVDS panel (Freescale part number MCIMX-LVDS1) to SABRE-Lite board.

This commit is a rebase Fabio Estevan's patch from 5/31 to u-boot-video/master:

    http://patchwork.ozlabs.org/patch/162206/

Modifications include:
removal of i2c setup (unneeded)
cleanup of lcd_iomux to use struct mxc_ccm_reg and anatop_regs and associated constants

Signed-off-by Eric Nelson <[email protected]>

---

 arch/arm/include/asm/arch-mx6/crm_regs.h      |    4 +
 board/freescale/mx6qsabrelite/mx6qsabrelite.c |   90 +++++++++++++++++++++++++
 include/configs/mx6qsabrelite.h               |   14 ++++-
 3 files changed, 107 insertions(+), 1 deletions(-)

arch/arm/include/asm/arch-mx6/crm_regs.h +4 -0

index 8388e38..cffc0a1 100644
@@ -294,6 +294,10 @@ struct mxc_ccm_reg {
 #define MXC_CCM_CHSCCDR_IPU1_DI0_CLK_SEL_MASK		(0x7)
 #define MXC_CCM_CHSCCDR_IPU1_DI0_CLK_SEL_OFFSET		0
 
+#define CHSCCDR_CLK_SEL_LDB_DI0				3
+#define CHSCCDR_PODF_DIVIDE_BY_3			2
+#define CHSCCDR_IPU_PRE_CLK_540M_PFD			5
+
 /* Define the bits in register CSCDR2 */
 #define MXC_CCM_CSCDR2_ECSPI_CLK_PODF_MASK		(0x3F << 19)
 #define MXC_CCM_CSCDR2_ECSPI_CLK_PODF_OFFSET		19

board/freescale/mx6qsabrelite/mx6qsabrelite.c +27 -0

index 4b4e89b..1632e7b 100644
@@ -36,6 +36,9 @@
 #include <micrel.h>
 #include <miiphy.h>
 #include <netdev.h>
+#include <linux/fb.h>
+#include <ipu_pixfmt.h>
+#include <asm/arch/crm_regs.h>
 DECLARE_GLOBAL_DATA_PTR;
 
 #define UART_PAD_CTRL  (PAD_CTL_PKE | PAD_CTL_PUE |	       \

@@ -195,6 +198,10 @@ static iomux_v3_cfg_t button_pads[] = {
 	MX6Q_PAD_GPIO_18__GPIO_7_13	| MUX_PAD_CTRL(BUTTON_PAD_CTRL),
 };
 
+iomux_v3_cfg_t lcd_gpio[] = {
+	MX6Q_PAD_SD1_CMD__GPIO_1_18 | MUX_PAD_CTRL(NO_PAD_CTRL),
+};
+
 static void setup_iomux_enet(void)
 {
 	gpio_direction_output(IMX_GPIO_NR(3, 23), 0);

@@ -372,10 +379,84 @@ int setup_sata(void)
 }
 #endif
 
+static struct fb_videomode lvds_xga = {
+	.name           = "Hannstar-XGA",
+	.refresh        = 60,
+	.xres           = 1024,
+	.yres           = 768,
+	.pixclock       = 15385,
+	.left_margin    = 220,
+	.right_margin   = 40,
+	.upper_margin   = 21,
+	.lower_margin   = 7,
+	.hsync_len      = 60,
+	.vsync_len      = 10,
+	.sync           = FB_SYNC_EXT,
+	.vmode          = FB_VMODE_NONINTERLACED
+};
+
+void lcd_iomux(void)
+{
+	struct mxc_ccm_reg *mxc_ccm = (struct mxc_ccm_reg *)CCM_BASE_ADDR;
+	struct anatop_regs *anatop = (struct anatop_regs *)ANATOP_BASE_ADDR;

board/freescale/mx6qsabrelite/mx6qsabrelite.c +10 -0

+
+	/* set LDB0, LDB1 clk select to 011/011 */
+	reg = readl(&mxc_ccm->cs2cdr);
+	reg &= ~(MXC_CCM_CS2CDR_LDB_DI0_CLK_SEL_MASK
+		 |MXC_CCM_CS2CDR_LDB_DI1_CLK_SEL_MASK);
+	reg |= (3<<MXC_CCM_CS2CDR_LDB_DI0_CLK_SEL_OFFSET)
+	      |(3<<MXC_CCM_CS2CDR_LDB_DI1_CLK_SEL_OFFSET);
+	writel(reg, &mxc_ccm->cs2cdr);
+
+	reg = readl(&mxc_ccm->cscmr2);

board/freescale/mx6qsabrelite/mx6qsabrelite.c +11 -0

+}
+
+void lcd_enable(void)
+{
+
+	int ret = ipuv3_fb_init(&lvds_xga, 0, IPU_PIX_FMT_RGB666);
+	if (ret)
+		printf("LCD cannot be configured: %d\n", ret);
+}
+
 int board_early_init_f(void)
 {
 	setup_iomux_uart();
 	setup_buttons();
+	lcd_iomux();
 
 	return 0;
 }

@@ -396,9 +477,18 @@ int board_init(void)
 	setup_sata();
 #endif
 

include/configs/mx6qsabrelite.h +2 -1

@@ -39,9 +39,10 @@
 #define CONFIG_REVISION_TAG
 
 /* Size of malloc() pool */
-#define CONFIG_SYS_MALLOC_LEN	       (CONFIG_ENV_SIZE + 2 * 1024 * 1024)
+#define CONFIG_SYS_MALLOC_LEN	       (10 * 1024 * 1024)
 
 #define CONFIG_BOARD_EARLY_INIT_F
+#define CONFIG_BOARD_LATE_INIT
 #define CONFIG_MISC_INIT_R
 #define CONFIG_MXC_GPIO
 

@@ -121,6 +122,17 @@
 /* Miscellaneous commands */
 #define CONFIG_CMD_BMODE
 

On 18/09/2012 01:14, Eric Nelson wrote:

Hi Eric,

Quoted 7 lines

Adds support for the Hannstar 1024 x 768 LVDS panel (Freescale part number MCIMX-LVDS1) to SABRE-Lite board.

This commit is a rebase Fabio Estevan's patch from 5/31 to u-boot-video/master:

     http://patchwork.ozlabs.org/patch/162206/

Quoted 5 lines

Modifications include:
removal of i2c setup (unneeded)
cleanup of lcd_iomux to use struct mxc_ccm_reg and anatop_regs and associated constants

All this stuff should not be part of the commit message, because it is more or less a changelog. Should you also include Fabio's signed-off ?

Quoted 79 lines

Signed-off-by Eric Nelson<[email protected]>

---

  arch/arm/include/asm/arch-mx6/crm_regs.h      |    4 +
  board/freescale/mx6qsabrelite/mx6qsabrelite.c |   90 +++++++++++++++++++++++++
  include/configs/mx6qsabrelite.h               |   14 ++++-
  3 files changed, 107 insertions(+), 1 deletions(-)

arch/arm/include/asm/arch-mx6/crm_regs.h +4 -0

index 8388e38..cffc0a1 100644
@@ -294,6 +294,10 @@ struct mxc_ccm_reg {
  #define MXC_CCM_CHSCCDR_IPU1_DI0_CLK_SEL_MASK		(0x7)
  #define MXC_CCM_CHSCCDR_IPU1_DI0_CLK_SEL_OFFSET		0
 
+#define CHSCCDR_CLK_SEL_LDB_DI0				3
+#define CHSCCDR_PODF_DIVIDE_BY_3			2
+#define CHSCCDR_IPU_PRE_CLK_540M_PFD			5
+
  /* Define the bits in register CSCDR2 */
  #define MXC_CCM_CSCDR2_ECSPI_CLK_PODF_MASK		(0x3F<<  19)
  #define MXC_CCM_CSCDR2_ECSPI_CLK_PODF_OFFSET		19

board/freescale/mx6qsabrelite/mx6qsabrelite.c +34 -0

index 4b4e89b..1632e7b 100644
@@ -36,6 +36,9 @@
  #include<micrel.h>
  #include<miiphy.h>
  #include<netdev.h>
+#include<linux/fb.h>
+#include<ipu_pixfmt.h>
+#include<asm/arch/crm_regs.h>
  DECLARE_GLOBAL_DATA_PTR;
 
  #define UART_PAD_CTRL  (PAD_CTL_PKE | PAD_CTL_PUE |	       \

@@ -195,6 +198,10 @@ static iomux_v3_cfg_t button_pads[] = {
  	MX6Q_PAD_GPIO_18__GPIO_7_13	| MUX_PAD_CTRL(BUTTON_PAD_CTRL),
  };
 
+iomux_v3_cfg_t lcd_gpio[] = {
+	MX6Q_PAD_SD1_CMD__GPIO_1_18 | MUX_PAD_CTRL(NO_PAD_CTRL),
+};
+
  static void setup_iomux_enet(void)
  {
  	gpio_direction_output(IMX_GPIO_NR(3, 23), 0);

@@ -372,10 +379,84 @@ int setup_sata(void)
  }
  #endif
 
+static struct fb_videomode lvds_xga = {
+	.name           = "Hannstar-XGA",
+	.refresh        = 60,
+	.xres           = 1024,
+	.yres           = 768,
+	.pixclock       = 15385,
+	.left_margin    = 220,
+	.right_margin   = 40,
+	.upper_margin   = 21,
+	.lower_margin   = 7,
+	.hsync_len      = 60,
+	.vsync_len      = 10,
+	.sync           = FB_SYNC_EXT,
+	.vmode          = FB_VMODE_NONINTERLACED
+};
+
+void lcd_iomux(void)
+{
+	struct mxc_ccm_reg *mxc_ccm = (struct mxc_ccm_reg *)CCM_BASE_ADDR;
+	struct anatop_regs *anatop = (struct anatop_regs *)ANATOP_BASE_ADDR;
+
+	int reg;
+	/* Turn on GPIO backlight */
+	imx_iomux_v3_setup_multiple_pads(lcd_gpio, ARRAY_SIZE(lcd_gpio));
+	gpio_direction_output(18, 1);
+
+	/* Turn on LDB0,IPU,IPU DI0 clocks */

board/freescale/mx6qsabrelite/mx6qsabrelite.c int setup_sata(void)

Quoted 3 lines

board/freescale/mx6qsabrelite/mx6qsabrelite.c +3 -0

+	reg = __raw_readl(&mxc_ccm->CCGR3);
+	reg |= 0x300F;
+	writel(reg,&mxc_ccm->CCGR3);

I think we can add constants for these - at least for these constants, you could drop the not useful defines with offset like MXC_CCM_CCGR5_CGx_OFFSET. There is already a patch (not yet merged) for MX5, we should then doing the same for MX6.

board/freescale/mx6qsabrelite/mx6qsabrelite.c int setup_sata(void)

Quoted 4 lines

board/freescale/mx6qsabrelite/mx6qsabrelite.c +4 -0

+
+	/* set PFD3_FRAC to 0x13 == 455 MHz (480*18)/0x13 */
+	writel(0x00003F00,&anatop->pfd_480_clr);
+	writel(0x00001300,&anatop->pfd_480_set);

Add constants for these. They are not already defined, but they are added when needed, see for example ehci-mx6.c

Quoted 14 lines

board/freescale/mx6qsabrelite/mx6qsabrelite.c +14 -0

+
+	/* set LDB0, LDB1 clk select to 011/011 */
+	reg = readl(&mxc_ccm->cs2cdr);
+	reg&= ~(MXC_CCM_CS2CDR_LDB_DI0_CLK_SEL_MASK
+		 |MXC_CCM_CS2CDR_LDB_DI1_CLK_SEL_MASK);
+	reg |= (3<<MXC_CCM_CS2CDR_LDB_DI0_CLK_SEL_OFFSET)
+	      |(3<<MXC_CCM_CS2CDR_LDB_DI1_CLK_SEL_OFFSET);
+	writel(reg,&mxc_ccm->cs2cdr);
+
+	reg = readl(&mxc_ccm->cscmr2);
+	reg |= MXC_CCM_CSCMR2_LDB_DI0_IPU_DIV;
+	writel(reg,&mxc_ccm->cscmr2);
+
+	reg = readl(&mxc_ccm->chsccdr);

board/freescale/mx6qsabrelite/mx6qsabrelite.c int setup_sata(void) · line 435

Quoted 6 lines

board/freescale/mx6qsabrelite/mx6qsabrelite.c +6 -0

+	reg&= ~(MXC_CCM_CHSCCDR_IPU1_DI0_PRE_CLK_SEL_MASK
+		|MXC_CCM_CHSCCDR_IPU1_DI0_PODF_MASK
+		|MXC_CCM_CHSCCDR_IPU1_DI0_CLK_SEL_MASK);
+	/* derive clock from LDB_DI0 */
+	/* divide by 3 */
+	/* derive clock from 540M PFD */

Wrong style for a multiline comment

Quoted 4 lines

board/freescale/mx6qsabrelite/mx6qsabrelite.c +4 -0

+	reg |= (CHSCCDR_CLK_SEL_LDB_DI0
+		<<MXC_CCM_CHSCCDR_IPU1_DI0_CLK_SEL_OFFSET)
+	      |(CHSCCDR_PODF_DIVIDE_BY_3
+		<<MXC_CCM_CHSCCDR_IPU1_DI0_PODF_OFFSET)

board/freescale/mx6qsabrelite/mx6qsabrelite.c int setup_sata(void) · line 444

Quoted 5 lines

board/freescale/mx6qsabrelite/mx6qsabrelite.c +5 -0

+	      |(CHSCCDR_IPU_PRE_CLK_540M_PFD
+		<<MXC_CCM_CHSCCDR_IPU1_DI0_PRE_CLK_SEL_OFFSET);
+	writel(reg,&mxc_ccm->chsccdr);
+
+	writel(0x201, IOMUXC_BASE_ADDR + 0x8);	/* 16 bpp */

This is GPR2, right ? Nobody can easy understand. Use structure (not offset) and constants to set it.

Quoted 28 lines

board/freescale/mx6qsabrelite/mx6qsabrelite.c +14 -0

+}
+
+void lcd_enable(void)
+{
+
+	int ret = ipuv3_fb_init(&lvds_xga, 0, IPU_PIX_FMT_RGB666);
+	if (ret)
+		printf("LCD cannot be configured: %d\n", ret);
+}
+
  int board_early_init_f(void)
  {
  	setup_iomux_uart();
  	setup_buttons();
+	lcd_iomux();
 
  	return 0;
  }

@@ -396,9 +477,18 @@ int board_init(void)
  	setup_sata();
  #endif
 
+#ifdef CONFIG_VIDEO
+	lcd_enable();
+#endif
         return 0;
  }
 

board/freescale/mx6qsabrelite/mx6qsabrelite.c int board_init(void) · line 489

Quoted 4 lines

board/freescale/mx6qsabrelite/mx6qsabrelite.c +4 -0

+int board_late_init(void)
+{
+	setenv("stdout", "serial");
+	return 0;

This is wrong, we found a better way to do this. You should add a overwrite_console() function, see for example the MX5 boards.

Quoted 18 lines

include/configs/mx6qsabrelite.h +2 -1

@@ -39,9 +39,10 @@
  #define CONFIG_REVISION_TAG
 
  /* Size of malloc() pool */
-#define CONFIG_SYS_MALLOC_LEN	       (CONFIG_ENV_SIZE + 2 * 1024 * 1024)
+#define CONFIG_SYS_MALLOC_LEN	       (10 * 1024 * 1024)
 
  #define CONFIG_BOARD_EARLY_INIT_F
+#define CONFIG_BOARD_LATE_INIT
  #define CONFIG_MISC_INIT_R
  #define CONFIG_MXC_GPIO
 

@@ -121,6 +122,17 @@
  /* Miscellaneous commands */
  #define CONFIG_CMD_BMODE
 

include/configs/mx6qsabrelite.h line 134

Quoted 10 lines

include/configs/mx6qsabrelite.h +10 -0

+/* Framebuffer and LCD */
+#define CONFIG_VIDEO
+#define CONFIG_VIDEO_IPUV3
+#define CONFIG_CFB_CONSOLE
+#define CONFIG_VGA_AS_SINGLE_DEVICE
+#define CONFIG_VIDEO_BMP_RLE8
+#define CONFIG_SPLASH_SCREEN
+#define CONFIG_BMP_16BPP
+#define CONFIG_VIDEO_LOGO
+#define CONFIG_IPUV3_CLK 260000000

...adding also CONFIG_SYS_CONSOLE_OVERWRITE_ROUTINE

Best regards,
Stefano Babic

Do you have a reference to the patch so I can follow precedent?

Quoted 2 lines · Eric Nelson

Do you have a reference to the patch so I can follow precedent?

See http://patchwork.ozlabs.org/patch/177403/

Check changes in .h.

Quoted 8 lines · Stefano Babic

Quoted 2 lines · Eric Nelson

Do you have a reference to the patch so I can follow precedent?

See http://patchwork.ozlabs.org/patch/177403/

Check changes in .h.

Should I remove the un-named macros?
MXC_CCM_CCGRx_CGy_OFFSET
MXC_CCM_CCGRx_CGy_MASK

It appears that only arch/arm/cpu/armv7/mx6/clock.c is using them.

Adjust the IPUv3 registers, so that the IPUv3 driver can be extended for mx6 as well.

Signed-off-by Fabio Estevam <[email protected]>

---
This patch series apply against Anatolij's tree.

 arch/arm/include/asm/arch-mx5/imx-regs.h |    2 --
 arch/arm/include/asm/arch-mx6/imx-regs.h |    3 +++
 drivers/video/ipu_regs.h                 |   12 +++++++++++-
 3 files changed, 14 insertions(+), 3 deletions(-)