[PATCH] OMAP4: PandaBoard: Adding DVI support

[PATCH] OMAP4: PandaBoard: Adding DVI support

[Raghuveer Murthy]

Adding DVI support to OMAP4 PandaBoard.

PandaBoard uses TFP410 DVI Framer chip
http://focus.ti.com/lit/ds/symlink/tfp410.pdf

The TFP410 gets its power enable and display data over GPIO lines muxed
in from OMAP4430. PandaBoard supports other LCD displays through expansion
connectors, following board rework. This will disable the DVI interface.
However, the existing mux settings remain the same

PandaBoard additionally supports display over HDMI interface. It is
mutually exclusive to display over DVI. Hence the mux settings need to be
configured seperately, as and when HDMI is enabled

Also, I2C3 bus used for reading EDID data from DVI Monitors is
registered here. Since the design is similar to BeagleBoard, the code
for the same is taken from the kernel.org commit e3333f48dd5cb21
(omap: Adding beagle i2c eeprom driver to read EDID)

Reviewed-by: Manjunath G Kondaiah <manjugk@ti.com>
Reviewed-by: Anand Gadiyar <gadiyar@ti.com> 
Reviewed-by: Nishanth Menon <nm@ti.com>
Reviewed-by: Sumit Semwal <sumit.semwal@ti.com>
Signed-off-by: Raghuveer Murthy <raghuveer.murthy@ti.com>
---

Base
====
url =
git://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux-2.6.git
branch "master"
commit 100b33c8bd8a3235fd0b7948338d6cbb3db3c63d
tag 2.6.38-rc4

Applies over:
=============
1.
v10 of OMAP2,3 DSS2 HWMOD
https://patchwork.kernel.org/patch/500191/

2.
v3 of DSS2: Generalize clock names
https://patchwork.kernel.org/patch/520191/

3.
v3 of DSS2: OMAP4 DSS HWMOD :
https://patchwork.kernel.org/patch/511211/

4.
OMAP: DSS2: Common IRQ handler for all OMAPs
https://patchwork.kernel.org/patch/526241/

5.
OMAP4: DSS2: Adding fclk support for DPI interface
https://patchwork.kernel.org/patch/529381/

6.
OMAP: DSS2: Adding dss_features for independent core clk divider
https://patchwork.kernel.org/patch/529561/
=================================================================

 arch/arm/mach-omap2/board-omap4panda.c |  137 +++++++++++++++++++++++++++++++-
 1 files changed, 136 insertions(+), 1 deletions(-)

diff --git a/arch/arm/mach-omap2/board-omap4panda.c b/arch/arm/mach-omap2/board-omap4panda.c
index e944025..b09d239 100644
--- a/arch/arm/mach-omap2/board-omap4panda.c
+++ b/arch/arm/mach-omap2/board-omap4panda.c
@@ -37,6 +37,7 @@
 #include <plat/common.h>
 #include <plat/usb.h>
 #include <plat/mmc.h>
+#include <plat/panel-generic-dpi.h>
 #include "timer-gp.h"
 
 #include "hsmmc.h"
@@ -76,6 +77,60 @@ static struct platform_device *panda_devices[] __initdata = {
 	&leds_gpio,
 };
 
+/* Display DVI */
+#define PANDA_DVI_TFP410_POWER_DOWN_GPIO	0
+
+static int panda_enable_dvi(struct omap_dss_device *dssdev)
+{
+	gpio_set_value(dssdev->reset_gpio, 1);
+	return 0;
+}
+
+static void panda_disable_dvi(struct omap_dss_device *dssdev)
+{
+	gpio_set_value(dssdev->reset_gpio, 0);
+}
+
+/* Using generic display panel */
+static struct panel_generic_dpi_data dvi_panel = {
+	.name			= "generic",
+	.platform_enable	= panda_enable_dvi,
+	.platform_disable	= panda_disable_dvi,
+};
+
+struct omap_dss_device panda_dvi_device = {
+	.type			= OMAP_DISPLAY_TYPE_DPI,
+	.name			= "dvi",
+	.driver_name		= "generic_dpi_panel",
+	.data			= &dvi_panel,
+	.phy.dpi.data_lines	= 24,
+	.reset_gpio		= PANDA_DVI_TFP410_POWER_DOWN_GPIO,
+	.channel		= OMAP_DSS_CHANNEL_LCD2,
+};
+
+int __init panda_dvi_init(void)
+{
+	int r;
+
+	/* Requesting TFP410 DVI GPIO and disabling it, at bootup */
+	r = gpio_request_one(panda_dvi_device.reset_gpio,
+				GPIOF_OUT_INIT_LOW, "DVI PD");
+	if (r)
+		pr_err("Failed to get DVI powerdown GPIO\n");
+
+	return r;
+}
+
+static struct omap_dss_device *panda_dss_devices[] = {
+	&panda_dvi_device,
+};
+
+static struct omap_dss_board_info panda_dss_data = {
+	.num_devices	= ARRAY_SIZE(panda_dss_devices),
+	.devices	= panda_dss_devices,
+	.default_device	= &panda_dvi_device,
+};
+
 static void __init omap4_panda_init_irq(void)
 {
 	omap2_init_common_infrastructure();
@@ -375,6 +430,17 @@ static struct i2c_board_info __initdata omap4_panda_i2c_boardinfo[] = {
 		.platform_data = &omap4_panda_twldata,
 	},
 };
+
+/*
+ * Display monitor features are burnt in their EEPROM as EDID data. The EEPROM
+ * is connected as I2C slave device, and can be accessed at address 0x50
+ */
+static struct i2c_board_info __initdata panda_i2c_eeprom[] = {
+	{
+		I2C_BOARD_INFO("eeprom", 0x50),
+	},
+};
+
 static int __init omap4_panda_i2c_init(void)
 {
 	/*
@@ -384,13 +450,76 @@ static int __init omap4_panda_i2c_init(void)
 	omap_register_i2c_bus(1, 400, omap4_panda_i2c_boardinfo,
 			ARRAY_SIZE(omap4_panda_i2c_boardinfo));
 	omap_register_i2c_bus(2, 400, NULL, 0);
-	omap_register_i2c_bus(3, 400, NULL, 0);
+	/*
+	 * Bus 3 is attached to the DVI port where devices like the pico DLP
+	 * projector don't work reliably with 400kHz
+	 */
+	omap_register_i2c_bus(3, 100, panda_i2c_eeprom,
+					ARRAY_SIZE(panda_i2c_eeprom));
 	omap_register_i2c_bus(4, 400, NULL, 0);
 	return 0;
 }
 
 #ifdef CONFIG_OMAP_MUX
 static struct omap_board_mux board_mux[] __initdata = {
+	/* gpio 0 - TFP410 PD */
+	OMAP4_MUX(KPD_COL1, OMAP_PIN_OUTPUT | OMAP_MUX_MODE3),
+	/* dispc2_data23 */
+	OMAP4_MUX(USBB2_ULPITLL_STP, OMAP_PIN_OUTPUT | OMAP_MUX_MODE5),
+	/* dispc2_data22 */
+	OMAP4_MUX(USBB2_ULPITLL_DIR, OMAP_PIN_OUTPUT | OMAP_MUX_MODE5),
+	/* dispc2_data21 */
+	OMAP4_MUX(USBB2_ULPITLL_NXT, OMAP_PIN_OUTPUT | OMAP_MUX_MODE5),
+	/* dispc2_data20 */
+	OMAP4_MUX(USBB2_ULPITLL_DAT0, OMAP_PIN_OUTPUT | OMAP_MUX_MODE5),
+	/* dispc2_data19 */
+	OMAP4_MUX(USBB2_ULPITLL_DAT1, OMAP_PIN_OUTPUT | OMAP_MUX_MODE5),
+	/* dispc2_data18 */
+	OMAP4_MUX(USBB2_ULPITLL_DAT2, OMAP_PIN_OUTPUT | OMAP_MUX_MODE5),
+	/* dispc2_data15 */
+	OMAP4_MUX(USBB2_ULPITLL_DAT3, OMAP_PIN_OUTPUT | OMAP_MUX_MODE5),
+	/* dispc2_data14 */
+	OMAP4_MUX(USBB2_ULPITLL_DAT4, OMAP_PIN_OUTPUT | OMAP_MUX_MODE5),
+	/* dispc2_data13 */
+	OMAP4_MUX(USBB2_ULPITLL_DAT5, OMAP_PIN_OUTPUT | OMAP_MUX_MODE5),
+	/* dispc2_data12 */
+	OMAP4_MUX(USBB2_ULPITLL_DAT6, OMAP_PIN_OUTPUT | OMAP_MUX_MODE5),
+	/* dispc2_data11 */
+	OMAP4_MUX(USBB2_ULPITLL_DAT7, OMAP_PIN_OUTPUT | OMAP_MUX_MODE5),
+	/* dispc2_data10 */
+	OMAP4_MUX(DPM_EMU3, OMAP_PIN_OUTPUT | OMAP_MUX_MODE5),
+	/* dispc2_data9 */
+	OMAP4_MUX(DPM_EMU4, OMAP_PIN_OUTPUT | OMAP_MUX_MODE5),
+	/* dispc2_data16 */
+	OMAP4_MUX(DPM_EMU5, OMAP_PIN_OUTPUT | OMAP_MUX_MODE5),
+	/* dispc2_data17 */
+	OMAP4_MUX(DPM_EMU6, OMAP_PIN_OUTPUT | OMAP_MUX_MODE5),
+	/* dispc2_hsync */
+	OMAP4_MUX(DPM_EMU7, OMAP_PIN_OUTPUT | OMAP_MUX_MODE5),
+	/* dispc2_pclk */
+	OMAP4_MUX(DPM_EMU8, OMAP_PIN_OUTPUT | OMAP_MUX_MODE5),
+	/* dispc2_vsync */
+	OMAP4_MUX(DPM_EMU9, OMAP_PIN_OUTPUT | OMAP_MUX_MODE5),
+	/* dispc2_de */
+	OMAP4_MUX(DPM_EMU10, OMAP_PIN_OUTPUT | OMAP_MUX_MODE5),
+	/* dispc2_data8 */
+	OMAP4_MUX(DPM_EMU11, OMAP_PIN_OUTPUT | OMAP_MUX_MODE5),
+	/* dispc2_data7 */
+	OMAP4_MUX(DPM_EMU12, OMAP_PIN_OUTPUT | OMAP_MUX_MODE5),
+	/* dispc2_data6 */
+	OMAP4_MUX(DPM_EMU13, OMAP_PIN_OUTPUT | OMAP_MUX_MODE5),
+	/* dispc2_data5 */
+	OMAP4_MUX(DPM_EMU14, OMAP_PIN_OUTPUT | OMAP_MUX_MODE5),
+	/* dispc2_data4 */
+	OMAP4_MUX(DPM_EMU15, OMAP_PIN_OUTPUT | OMAP_MUX_MODE5),
+	/* dispc2_data3 */
+	OMAP4_MUX(DPM_EMU16, OMAP_PIN_OUTPUT | OMAP_MUX_MODE5),
+	/* dispc2_data2 */
+	OMAP4_MUX(DPM_EMU17, OMAP_PIN_OUTPUT | OMAP_MUX_MODE5),
+	/* dispc2_data1 */
+	OMAP4_MUX(DPM_EMU18, OMAP_PIN_OUTPUT | OMAP_MUX_MODE5),
+	/* dispc2_data0 */
+	OMAP4_MUX(DPM_EMU19, OMAP_PIN_OUTPUT | OMAP_MUX_MODE5),
 	{ .reg_offset = OMAP_MUX_TERMINATOR },
 };
 #else
@@ -400,6 +529,7 @@ static struct omap_board_mux board_mux[] __initdata = {
 static void __init omap4_panda_init(void)
 {
 	int package = OMAP_PACKAGE_CBS;
+	int err;
 
 	if (omap_rev() == OMAP4430_REV_ES1_0)
 		package = OMAP_PACKAGE_CBL;
@@ -411,6 +541,11 @@ static void __init omap4_panda_init(void)
 	omap4_twl6030_hsmmc_init(mmc);
 	omap4_ehci_init();
 	usb_musb_init(&musb_board_data);
+
+	/* Enabling DVI Display */
+	err = panda_dvi_init();
+	if (!err)
+		omap_display_init(&panda_dss_data);
 }
 
 static void __init omap4_panda_map_io(void)
-- 
1.7.0.4

[PATCH] OMAP4: PandaBoard: Adding DVI support

[Gadiyar, Anand]

On Wed, Feb 9, 2011 at 7:25 PM, Raghuveer Murthy
<raghuveer.murthy@ti.com> wrote:
> Adding DVI support to OMAP4 PandaBoard.
>
> PandaBoard uses TFP410 DVI Framer chip
> http://focus.ti.com/lit/ds/symlink/tfp410.pdf
>
> The TFP410 gets its power enable and display data over GPIO lines muxed
> in from OMAP4430. PandaBoard supports other LCD displays through expansion
> connectors, following board rework. This will disable the DVI interface.
> However, the existing mux settings remain the same
>
> PandaBoard additionally supports display over HDMI interface. It is
> mutually exclusive to display over DVI. Hence the mux settings need to be
> configured seperately, as and when HDMI is enabled
>
> Also, I2C3 bus used for reading EDID data from DVI Monitors is
> registered here. Since the design is similar to BeagleBoard, the code
> for the same is taken from the kernel.org commit e3333f48dd5cb21
> (omap: Adding beagle i2c eeprom driver to read EDID)
>
> Reviewed-by: Manjunath G Kondaiah <manjugk@ti.com>
> Reviewed-by: Anand Gadiyar <gadiyar@ti.com>

I've also tested it on the Panda, and gone through this and most
of the other patches in the series. So if you like, you could
consider this an:

Acked-by: Anand Gadiyar <gadiyar@ti.com>

> Reviewed-by: Nishanth Menon <nm@ti.com>
> Reviewed-by: Sumit Semwal <sumit.semwal@ti.com>
> Signed-off-by: Raghuveer Murthy <raghuveer.murthy@ti.com>
> ---
>
> Base
> ====
> url =
> git://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux-2.6.git
> branch "master"
> commit 100b33c8bd8a3235fd0b7948338d6cbb3db3c63d
> tag 2.6.38-rc4
>

To make it easier for someone to test, I've extracted this
and the 6 dependent series from patchwork, and hosted
them in a branch on my devel tree. They are available
against v2.6.38-rc4 here:

git://dev.omapzoom.org/pub/scm/anand/linux-omap-usb.git

in the display-patches-for-v2.6.38-rc4 branch if someone
wants to take a look.

- Anand

[PATCH] OMAP4: PandaBoard: Adding DVI support

[Robert Nelson]

On Wed, Feb 9, 2011 at 8:08 AM, Gadiyar, Anand <gadiyar@ti.com> wrote:
> On Wed, Feb 9, 2011 at 7:25 PM, Raghuveer Murthy
> <raghuveer.murthy@ti.com> wrote:
>> Adding DVI support to OMAP4 PandaBoard.
>>
>> PandaBoard uses TFP410 DVI Framer chip
>> http://focus.ti.com/lit/ds/symlink/tfp410.pdf
>>
>> The TFP410 gets its power enable and display data over GPIO lines muxed
>> in from OMAP4430. PandaBoard supports other LCD displays through expansion
>> connectors, following board rework. This will disable the DVI interface.
>> However, the existing mux settings remain the same
>>
>> PandaBoard additionally supports display over HDMI interface. It is
>> mutually exclusive to display over DVI. Hence the mux settings need to be
>> configured seperately, as and when HDMI is enabled
>>
>> Also, I2C3 bus used for reading EDID data from DVI Monitors is
>> registered here. Since the design is similar to BeagleBoard, the code
>> for the same is taken from the kernel.org commit e3333f48dd5cb21
>> (omap: Adding beagle i2c eeprom driver to read EDID)
>>
>> Reviewed-by: Manjunath G Kondaiah <manjugk@ti.com>
>> Reviewed-by: Anand Gadiyar <gadiyar@ti.com>
>
> I've also tested it on the Panda, and gone through this and most
> of the other patches in the series. So if you like, you could
> consider this an:
>
> Acked-by: Anand Gadiyar <gadiyar@ti.com>
>
>> Reviewed-by: Nishanth Menon <nm@ti.com>
>> Reviewed-by: Sumit Semwal <sumit.semwal@ti.com>
>> Signed-off-by: Raghuveer Murthy <raghuveer.murthy@ti.com>
>> ---
>>
>> Base
>> ====
>> url =
>> git://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux-2.6.git
>> branch "master"
>> commit 100b33c8bd8a3235fd0b7948338d6cbb3db3c63d
>> tag 2.6.38-rc4
>>
>
> To make it easier for someone to test, I've extracted this
> and the 6 dependent series from patchwork, and hosted
> them in a branch on my devel tree. They are available
> against v2.6.38-rc4 here:
>
> git://dev.omapzoom.org/pub/scm/anand/linux-omap-usb.git
>
> in the display-patches-for-v2.6.38-rc4 branch if someone
> wants to take a look.

These work great on my Panda A1, (mainline 2.6.38-rc4 base) haven't
seen any regressions on my Beagle C4 with the same patchset/image..

Regards,

-- 
Robert Nelson
http://www.rcn-ee.com/

[PATCH] OMAP4: PandaBoard: Adding DVI support

[Robert Nelson]

On Wed, Feb 9, 2011 at 1:07 PM, Robert Nelson <robertcnelson@gmail.com> wrote:
> On Wed, Feb 9, 2011 at 8:08 AM, Gadiyar, Anand <gadiyar@ti.com> wrote:
>> On Wed, Feb 9, 2011 at 7:25 PM, Raghuveer Murthy
>> <raghuveer.murthy@ti.com> wrote:
>>> Adding DVI support to OMAP4 PandaBoard.
>>>
>>> PandaBoard uses TFP410 DVI Framer chip
>>> http://focus.ti.com/lit/ds/symlink/tfp410.pdf
>>>
>>> The TFP410 gets its power enable and display data over GPIO lines muxed
>>> in from OMAP4430. PandaBoard supports other LCD displays through expansion
>>> connectors, following board rework. This will disable the DVI interface.
>>> However, the existing mux settings remain the same
>>>
>>> PandaBoard additionally supports display over HDMI interface. It is
>>> mutually exclusive to display over DVI. Hence the mux settings need to be
>>> configured seperately, as and when HDMI is enabled
>>>
>>> Also, I2C3 bus used for reading EDID data from DVI Monitors is
>>> registered here. Since the design is similar to BeagleBoard, the code
>>> for the same is taken from the kernel.org commit e3333f48dd5cb21
>>> (omap: Adding beagle i2c eeprom driver to read EDID)
>>>
>>> Reviewed-by: Manjunath G Kondaiah <manjugk@ti.com>
>>> Reviewed-by: Anand Gadiyar <gadiyar@ti.com>
>>
>> I've also tested it on the Panda, and gone through this and most
>> of the other patches in the series. So if you like, you could
>> consider this an:
>>
>> Acked-by: Anand Gadiyar <gadiyar@ti.com>
>>
>>> Reviewed-by: Nishanth Menon <nm@ti.com>
>>> Reviewed-by: Sumit Semwal <sumit.semwal@ti.com>
>>> Signed-off-by: Raghuveer Murthy <raghuveer.murthy@ti.com>
>>> ---
>>>
>>> Base
>>> ====
>>> url =
>>> git://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux-2.6.git
>>> branch "master"
>>> commit 100b33c8bd8a3235fd0b7948338d6cbb3db3c63d
>>> tag 2.6.38-rc4
>>>
>>
>> To make it easier for someone to test, I've extracted this
>> and the 6 dependent series from patchwork, and hosted
>> them in a branch on my devel tree. They are available
>> against v2.6.38-rc4 here:
>>
>> git://dev.omapzoom.org/pub/scm/anand/linux-omap-usb.git
>>
>> in the display-patches-for-v2.6.38-rc4 branch if someone
>> wants to take a look.
>
> These work great on my Panda A1, (mainline 2.6.38-rc4 base) haven't
> seen any regressions on my Beagle C4 with the same patchset/image..

for the panda patches:

Tested-by: Robert Nelson <robertcnelson@gmail.com>



-- 
Robert Nelson
http://www.rcn-ee.com/

[PATCH] OMAP4: PandaBoard: Adding DVI support

[Bryan Wu]

On Wed, Feb 9, 2011 at 10:08 PM, Gadiyar, Anand <gadiyar@ti.com> wrote:
> On Wed, Feb 9, 2011 at 7:25 PM, Raghuveer Murthy
> <raghuveer.murthy@ti.com> wrote:
>> Adding DVI support to OMAP4 PandaBoard.
>>
>> PandaBoard uses TFP410 DVI Framer chip
>> http://focus.ti.com/lit/ds/symlink/tfp410.pdf
>>
>> The TFP410 gets its power enable and display data over GPIO lines muxed
>> in from OMAP4430. PandaBoard supports other LCD displays through expansion
>> connectors, following board rework. This will disable the DVI interface.
>> However, the existing mux settings remain the same
>>
>> PandaBoard additionally supports display over HDMI interface. It is
>> mutually exclusive to display over DVI. Hence the mux settings need to be
>> configured seperately, as and when HDMI is enabled
>>
>> Also, I2C3 bus used for reading EDID data from DVI Monitors is
>> registered here. Since the design is similar to BeagleBoard, the code
>> for the same is taken from the kernel.org commit e3333f48dd5cb21
>> (omap: Adding beagle i2c eeprom driver to read EDID)
>>
>> Reviewed-by: Manjunath G Kondaiah <manjugk@ti.com>
>> Reviewed-by: Anand Gadiyar <gadiyar@ti.com>
>
> I've also tested it on the Panda, and gone through this and most
> of the other patches in the series. So if you like, you could
> consider this an:
>
> Acked-by: Anand Gadiyar <gadiyar@ti.com>
>
>> Reviewed-by: Nishanth Menon <nm@ti.com>
>> Reviewed-by: Sumit Semwal <sumit.semwal@ti.com>
>> Signed-off-by: Raghuveer Murthy <raghuveer.murthy@ti.com>
>> ---
>>
>> Base
>> ====
>> url =
>> git://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux-2.6.git
>> branch "master"
>> commit 100b33c8bd8a3235fd0b7948338d6cbb3db3c63d
>> tag 2.6.38-rc4
>>
>
> To make it easier for someone to test, I've extracted this
> and the 6 dependent series from patchwork, and hosted
> them in a branch on my devel tree. They are available
> against v2.6.38-rc4 here:
>
> git://dev.omapzoom.org/pub/scm/anand/linux-omap-usb.git
>
> in the display-patches-for-v2.6.38-rc4 branch if someone
> wants to take a look.

I fetched this branch which contains 35 patches on top of 2.6.38-rc4.
After fixing some conflict, I applied these 35 patches on our Ubuntu
ti-omap-dev kernel. With built-in the DSS2 driver and GENERIC_DPI
driver, kernel boots fine on my Panda with Ubuntu GUI. But I have to
set the kernel boot command args with omapfb.mode=dvi:1024x768MR-24 at 60
omapdss.def_disp=dvi, pointed out by Sebastien. Without this bootargs,
no graphic at all. so as Robert concerned, EDID detection doesn't
work. Does that related to I2C driver? I2C driver always tell me
timeout in dmesg.

Thanks a lot,
-- 
Bryan Wu <bryan.wu@canonical.com>
Kernel Developer ? ?+86.138-1617-6545 Mobile
Ubuntu Kernel Team
Canonical Ltd. ? ? ?www.canonical.com
Ubuntu - Linux for human beings | www.ubuntu.com

[PATCH] OMAP4: PandaBoard: Adding DVI support

[Koen Kooi]

Op 11 feb 2011, om 13:12 heeft Bryan Wu het volgende geschreven:

> On Wed, Feb 9, 2011 at 10:08 PM, Gadiyar, Anand <gadiyar@ti.com> wrote:
>> On Wed, Feb 9, 2011 at 7:25 PM, Raghuveer Murthy
>> <raghuveer.murthy@ti.com> wrote:
>>> Adding DVI support to OMAP4 PandaBoard.
>>> 
>>> PandaBoard uses TFP410 DVI Framer chip
>>> http://focus.ti.com/lit/ds/symlink/tfp410.pdf
>>> 
>>> The TFP410 gets its power enable and display data over GPIO lines muxed
>>> in from OMAP4430. PandaBoard supports other LCD displays through expansion
>>> connectors, following board rework. This will disable the DVI interface.
>>> However, the existing mux settings remain the same
>>> 
>>> PandaBoard additionally supports display over HDMI interface. It is
>>> mutually exclusive to display over DVI. Hence the mux settings need to be
>>> configured seperately, as and when HDMI is enabled
>>> 
>>> Also, I2C3 bus used for reading EDID data from DVI Monitors is
>>> registered here. Since the design is similar to BeagleBoard, the code
>>> for the same is taken from the kernel.org commit e3333f48dd5cb21
>>> (omap: Adding beagle i2c eeprom driver to read EDID)
>>> 
>>> Reviewed-by: Manjunath G Kondaiah <manjugk@ti.com>
>>> Reviewed-by: Anand Gadiyar <gadiyar@ti.com>
>> 
>> I've also tested it on the Panda, and gone through this and most
>> of the other patches in the series. So if you like, you could
>> consider this an:
>> 
>> Acked-by: Anand Gadiyar <gadiyar@ti.com>
>> 
>>> Reviewed-by: Nishanth Menon <nm@ti.com>
>>> Reviewed-by: Sumit Semwal <sumit.semwal@ti.com>
>>> Signed-off-by: Raghuveer Murthy <raghuveer.murthy@ti.com>
>>> ---
>>> 
>>> Base
>>> ====
>>> url =
>>> git://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux-2.6.git
>>> branch "master"
>>> commit 100b33c8bd8a3235fd0b7948338d6cbb3db3c63d
>>> tag 2.6.38-rc4
>>> 
>> 
>> To make it easier for someone to test, I've extracted this
>> and the 6 dependent series from patchwork, and hosted
>> them in a branch on my devel tree. They are available
>> against v2.6.38-rc4 here:
>> 
>> git://dev.omapzoom.org/pub/scm/anand/linux-omap-usb.git
>> 
>> in the display-patches-for-v2.6.38-rc4 branch if someone
>> wants to take a look.
> 
> I fetched this branch which contains 35 patches on top of 2.6.38-rc4.
> After fixing some conflict, I applied these 35 patches on our Ubuntu
> ti-omap-dev kernel. With built-in the DSS2 driver and GENERIC_DPI
> driver, kernel boots fine on my Panda with Ubuntu GUI. But I have to
> set the kernel boot command args with omapfb.mode=dvi:1024x768MR-24 at 60
> omapdss.def_disp=dvi, pointed out by Sebastien. Without this bootargs,
> no graphic at all. so as Robert concerned, EDID detection doesn't
> work. Does that related to I2C driver? I2C driver always tell me
> timeout in dmesg.

AIUI the i2c bus is only hooked up to the HDMI port, so unless you plug in something there, no dice.

regards,

Koen

[PATCH] OMAP4: PandaBoard: Adding DVI support

[Jan, Sebastien]

On Fri, Feb 11, 2011 at 4:38 PM, Koen Kooi <koen@dominion.thruhere.net> wrote:
>> I fetched this branch which contains 35 patches on top of 2.6.38-rc4.
>> After fixing some conflict, I applied these 35 patches on our Ubuntu
>> ti-omap-dev kernel. With built-in the DSS2 driver and GENERIC_DPI
>> driver, kernel boots fine on my Panda with Ubuntu GUI. But I have to
>> set the kernel boot command args with omapfb.mode=dvi:1024x768MR-24 at 60
>> omapdss.def_disp=dvi, pointed out by Sebastien. Without this bootargs,
>> no graphic at all. so as Robert concerned, EDID detection doesn't
>> work. Does that related to I2C driver? I2C driver always tell me
>> timeout in dmesg.
>
> AIUI the i2c bus is only hooked up to the HDMI port, so unless you plug in something there, no dice.

Are you sure? According to the pandaboard user manual, there seem to
be an i2c connection (i2c3): page 31, fig 9:
http://pandaboard.org/sites/default/files/board_reference/A1/Panda_Board_Spec_DOC-21010_REV0_6.pdf

[PATCH] OMAP4: PandaBoard: Adding DVI support

[Koen Kooi]

Op 25 feb 2011, om 19:57 heeft Jan, Sebastien het volgende geschreven:

> On Fri, Feb 11, 2011 at 4:38 PM, Koen Kooi <koen@dominion.thruhere.net> wrote:
>>> I fetched this branch which contains 35 patches on top of 2.6.38-rc4.
>>> After fixing some conflict, I applied these 35 patches on our Ubuntu
>>> ti-omap-dev kernel. With built-in the DSS2 driver and GENERIC_DPI
>>> driver, kernel boots fine on my Panda with Ubuntu GUI. But I have to
>>> set the kernel boot command args with omapfb.mode=dvi:1024x768MR-24 at 60
>>> omapdss.def_disp=dvi, pointed out by Sebastien. Without this bootargs,
>>> no graphic at all. so as Robert concerned, EDID detection doesn't
>>> work. Does that related to I2C driver? I2C driver always tell me
>>> timeout in dmesg.
>> 
>> AIUI the i2c bus is only hooked up to the HDMI port, so unless you plug in something there, no dice.
> 
> Are you sure? According to the pandaboard user manual, there seem to
> be an i2c connection (i2c3): page 31, fig 9:
> http://pandaboard.org/sites/default/files/board_reference/A1/Panda_Board_Spec_DOC-21010_REV0_6.pdf

And so it does, it's nice to have a DDC bus on each port. I guess I got confused by the fact no kernel is using that currently :)

regards,

Koen