Applied, thanks!

On Tue, Oct 18, 2011 at 09:32:27AM +0200, Koen Kooi wrote:

Applied, thanks!

Thanks.

And seems that meta-ti list started working as well!

--
Denys

Op 17 okt. 2011, om 22:06 heeft Dmytriyenko, Denys het volgende geschreven:

And here's corresponding u-boot patch.

-----Original Message-----
From: Dmytriyenko, Denys
Sent: Monday, October 17, 2011 9:45 AM
To: Koen Kooi; meta-ti@...
Cc: Kooi, Koen; Maupin, Chase; Kridner, Jason
Subject: RE: [meta-ti][PATCHES] linux-ti33x-psp

First 3 patches pushed with a follow up patch to fix XSERVER for the

evm.

Thanks!

I need to do some more testing on 0004- and don't want to mess up things

before the image deadline today.

Sounds fair, especially since I cannot really test beaglebone myself.

Denys


________________________________________
From: Koen Kooi [koen@...]
Sent: Monday, October 17, 2011 9:20 AM
To: meta-ti@...
Cc: Kooi, Koen; Maupin, Chase; Kridner, Jason; Dmytriyenko, Denys
Subject: Re: [meta-ti][PATCHES] linux-ti33x-psp

Op 17 okt. 2011, om 11:29 heeft Dmytriyenko, Denys het volgende
geschreven:

Well, there seems to be some delivery problems with meta-ti ML at the

moment (working with their admins to resolve), so I'm sending the patches
here for now.

Summary:

Introduce TI33x SOC and AM335x EVM machine with corresponding kernel

recipe

for PSP tree on Arago.

The second version of the patches add BeagleBone support and obsoletes

Joel's

initial recipe.

First 3 patches pushed with a follow up patch to fix XSERVER for the evm.
I need to do some more testing on 0004- and don't want to mess up things
before the image deadline today.

Denys


________________________________________
From: Dmytriyenko, Denys
Sent: Sunday, October 16, 2011 5:54 PM
To: Kooi, Koen; Koen Kooi
Cc: Maupin, Chase; Kridner, Jason
Subject: RE: [meta-ti][PATCHES] linux-ti33x-psp

Koen,

There is a "tipspkernel" distro feature, that I added last time, which

can be used to quickly switch between pure PSP and patched kernels. I'll
be sending an updated set of patches shortly for linux-ti33x recipes to
utilize that for applying LED and other patches on top of PSP kernel.

Denys


________________________________________
From: Dmytriyenko, Denys
Sent: Friday, October 14, 2011 9:48 PM
To: Kooi, Koen; Koen Kooi
Cc: Maupin, Chase
Subject: RE: [meta-ti][PATCHES] linux-ti33x-psp

the daily call (which doesn't feature denys)

On one hand, nobody invited me.
On the other hand, at this point I'm useless with BeagleBone, as there

is no board available for me... And I don't really have any time to fiddle
with it anyway.

the top of the psp tree is "questionable" to say at least looking at

the commits, too many "really fix crash now" and "really, really fix crash
now".

:) Well, sometimes it happens. But it shouldn't become a rule.


Anyway, I'll try combining the recipes.

Denys


________________________________________
From: Kooi, Koen
Sent: Friday, October 14, 2011 5:30 PM
To: Dmytriyenko, Denys; Koen Kooi
Cc: Maupin, Chase
Subject: RE: [meta-ti][PATCHES] linux-ti33x-psp

AIUI bone support is "upstream" now in the psp kernel, so I think we

only need the LED patch on top of that for bone. As I said on the daily
call (which doesn't feature denys), the top of the psp tree is
"questionable" to say at least looking at the commits, too many "really
fix crash now" and "really, really fix crash now".

Between now and RTM I'd only like to see a handfull of patches being

different in SRC_URI between bone and evm.

Texas Instruments Limited, 800 Pavilion Drive, Northampton, NN4 7YL.

Registered in England & Wales under company number 00574102

________________________________________

From: Dmytriyenko, Denys
Sent: 14 October 2011 23:05
To: Kooi, Koen; Koen Kooi
Cc: Maupin, Chase
Subject: RE: [meta-ti][PATCHES] linux-ti33x-psp

Would SRC_URI_beaglebone = "github/joel" in my recipe work for you? I

can send a patch for that as a follow up, i.e merging joel's into mine...

And yes, Bill wants to have working EVM in couple days (early next

week). Still lots of things to do...

-----Original Message-----
From: Kooi, Koen
Sent: Friday, October 14, 2011 4:54 PM
To: Dmytriyenko, Denys; Koen Kooi
Cc: Maupin, Chase
Subject: RE: [meta-ti][PATCHES] linux-ti33x-psp

At this point I think I prefer a single recipe for both (as in, kill

the

joel one), but if it's better to get the evm supported ASAP I can merge
these as-is.

Texas Instruments Limited, 800 Pavilion Drive, Northampton, NN4
7YL. Registered in England & Wales under company number 00574102

________________________________________

From: Dmytriyenko, Denys
Sent: 14 October 2011 22:24
To: Koen Kooi; Kooi, Koen
Cc: Maupin, Chase
Subject: [meta-ti][PATCHES] linux-ti33x-psp

Koen,

Latest PSP kernel for ti33x/am335x-evm - should not conflict with

Joel's,

but need to unify/merge later.

BTW, I'm in between setting up meta-ti trees, so I'm sending the

patches

instead for now...

Denys

<0001-linux-omap-remove-unused-.inc-files.patch><0002-linux-ti33x-psp-

add-TI33x-PSP-kernel-with-am335x-evm.patch><0003-am335x-evm-add-machine-
config-with-corresponding-ti3.patch><0004-linux-am335x-joel-remove-
beaglebone-specific-recipe-.patch>

<0001-u-boot-2011.10rc-update-to-cover-ti33x-SOC-boards-no.patch>

_______________________________________________
meta-ti mailing list
meta-ti@...
https://lists.yoctoproject.org/listinfo/meta-ti

Hi,

Do we want commit mails to go to this list as well? If so I'll poke Graeme to start sending them.

regards,

Koen

On Tue, Oct 18, 2011 at 04:53:36PM +0200, Koen Kooi wrote:

Hi,

Do we want commit mails to go to this list as well? If so I'll poke Graeme to start sending them.

We are going to be mvong the tree anyway. Should probably send commit emails
from a master tree...

--
Denys

From: Denys Dmytriyenko <denys@...>

* Tree moved to arago-project.org
* Remove i2c releated voltage change writes that were a nop

Signed-off-by: Denys Dmytriyenko <denys@...>
---
recipes-bsp/u-boot/u-boot_2011.10rc.bb | 6 +++---
1 files changed, 3 insertions(+), 3 deletions(-)

diff --git a/recipes-bsp/u-boot/u-boot_2011.10rc.bb b/recipes-bsp/u-boot/u-boot_2011.10rc.bb
index ad104d6..272ed70 100644
--- a/recipes-bsp/u-boot/u-boot_2011.10rc.bb
+++ b/recipes-bsp/u-boot/u-boot_2011.10rc.bb
@@ -4,14 +4,14 @@ require u-boot.inc
COMPATIBLE_MACHINE = "(ti33x)"
DEFAULT_PREFERENCE_ti33x = "99"
PV = "2011.09+git"
-PR = "r10"
+PR = "r11"

# SPL build
UBOOT_BINARY = "u-boot.img"
UBOOT_IMAGE = "u-boot-${MACHINE}-${PV}-${PR}.img"
UBOOT_SYMLINK = "u-boot-${MACHINE}.img"

-SRC_URI = "git://github.com/trini/u-boot.git;protocol=git;branch=int_am335xpsp_04.06.00.01-v2011.09-for-sdk-05.03.00.00 \
+SRC_URI = "git://arago-project.org/git/projects/u-boot-am33x.git;protocol=git;branch=int_am335xpsp_04.06.00.01-v2011.09-for-sdk-05.03.00.00 \
file://2011.09git/0001-am335x_evm-boot-kernel-from-ext2-3-filesystem.patch \
file://2011.09git/0002-am335x_evm-set-bootdelay-to-0.patch \
file://2011.09git/0003-am335x-evm-make-rootfs-RO-on-boot.patch \
@@ -21,7 +21,7 @@ SRC_URI = "git://github.com/trini/u-boot.git;protocol=git;branch=int_am335xpsp_0
file://2011.09git/0004-am335x-evm-fall-back-to-bone-if-header.config-is-emp.patch \
"

-SRCREV = "3fb65523c5994718b56a3b4b18ddf1ef3ffc02dc"
+SRCREV = "4b41772ae4ef8fbea0b3e93af169a115e3ad6d12"

LIC_FILES_CHKSUM = "file://COPYING;md5=1707d6db1d42237583f50183a5651ecb"

--
1.7.0.4

[added meta-ti list]


Op 20 okt. 2011, om 03:54 heeft Joel A Fernandes het volgende geschreven:

Currently added for following:
* User LEDs (added by Koen Kooi <k-kooi@...>)
* BeagleBone tester

Build and run time tested.

Signed-off-by: Joel A Fernandes <joelagnel@...>
---
v2 changes:
Replaced correct patch

I don't like this, at all. Please do your changes as a seperate patch, not as a change to an existing patch. And if you update a patch in the future please use -B -C to git format-patch so it can detect the rename and produce a real diff instead of a add+delete pair.

regards,

Koen

...pin-mux-and-init-for-beaglebone-specific-.patch | 132 ++++++++++++++++++++
...5x-evm-hack-in-LED-support-for-beaglebone.patch | 109 ----------------
recipes-kernel/linux/linux-ti33x-psp_3.0+3.1rc.bb | 4 +-
3 files changed, 134 insertions(+), 111 deletions(-)
create mode 100644 recipes-kernel/linux/linux-ti33x-psp-3.0+3.1rc/0001-am335x-Add-pin-mux-and-init-for-beaglebone-specific-.patch
delete mode 100644 recipes-kernel/linux/linux-ti33x-psp-3.0+3.1rc/0001-am335x-evm-hack-in-LED-support-for-beaglebone.patch

diff --git a/recipes-kernel/linux/linux-ti33x-psp-3.0+3.1rc/0001-am335x-Add-pin-mux-and-init-for-beaglebone-specific-.patch b/recipes-kernel/linux/linux-ti33x-psp-3.0+3.1rc/0001-am335x-Add-pin-mux-and-init-for-beaglebone-specific-.patch
new file mode 100644
index 0000000..a501c8a
--- /dev/null
+++ b/recipes-kernel/linux/linux-ti33x-psp-3.0+3.1rc/0001-am335x-Add-pin-mux-and-init-for-beaglebone-specific-.patch
@@ -0,0 +1,132 @@
+From 5be793dc9de7e694d9256396aecfdc136c8e2cf6 Mon Sep 17 00:00:00 2001
+From: Joel A Fernandes <joelagnel@...>
+Date: Mon, 17 Oct 2011 02:00:39 -0400
+Subject: [PATCH] am335x: Add pin mux and init for beaglebone specific devices
+
+Currently added for following:
+ * User LEDs (added by Koen Kooi <k-kooi@...>)
+ * BeagleBone tester
+
+Signed-off-by: Joel A Fernandes <joelagnel@...>
+---
+ arch/arm/mach-omap2/board-am335xevm.c | 69 +++++++++++++++++++++++++++++++++
+ arch/arm/mach-omap2/mux33xx.c | 2 +-
+ 2 files changed, 70 insertions(+), 1 deletions(-)
+
+diff --git a/arch/arm/mach-omap2/board-am335xevm.c b/arch/arm/mach-omap2/board-am335xevm.c
+index f625510..6beaa23 100644
+--- a/arch/arm/mach-omap2/board-am335xevm.c
++++ b/arch/arm/mach-omap2/board-am335xevm.c
+@@ -17,6 +17,7 @@
+ #include <linux/i2c.h>
+ #include <linux/i2c/at24.h>
+ #include <linux/gpio.h>
++#include <linux/leds.h>
+ #include <linux/spi/spi.h>
+ #include <linux/spi/flash.h>
+ #include <linux/mtd/mtd.h>
+@@ -590,6 +591,61 @@ static struct pinmux_config usb1_pin_mux[] = {
+ {NULL, 0},
+ };
+
++/* LEDS - gpio1_21 -> gpio1_24 */
++
++#define BEAGLEBONE_USR1_LED GPIO_TO_PIN(1, 21)
++#define BEAGLEBONE_USR2_LED GPIO_TO_PIN(1, 22)
++#define BEAGLEBONE_USR3_LED GPIO_TO_PIN(1, 23)
++#define BEAGLEBONE_USR4_LED GPIO_TO_PIN(1, 24)
++
++static struct gpio_led gpio_leds[] = {
++ {
++ .name = "beaglebone::usr0",
++ .default_trigger = "heartbeat",
++ .gpio = BEAGLEBONE_USR1_LED,
++ },
++ {
++ .name = "beaglebone::usr1",
++ .default_trigger = "mmc0",
++ .gpio = BEAGLEBONE_USR2_LED,
++ },
++ {
++ .name = "beaglebone::usr2",
++ .gpio = BEAGLEBONE_USR3_LED,
++ },
++ {
++ .name = "beaglebone::usr3",
++ .gpio = BEAGLEBONE_USR4_LED,
++ },
++};
++
++static struct gpio_led_platform_data gpio_led_info = {
++ .leds = gpio_leds,
++ .num_leds = ARRAY_SIZE(gpio_leds),
++};
++
++static struct platform_device leds_gpio = {
++ .name = "leds-gpio",
++ .id = -1,
++ .dev = {
++ .platform_data = &gpio_led_info,
++ },
++};
++
++static struct platform_device *bone_devices[] __initdata = {
++ &leds_gpio,
++};
++
++static struct pinmux_config bone_pin_mux[] = {
++ /* User LED gpios (gpio1_21 to gpio1_24) */
++ {"gpmc_a5.rgmii2_td0", OMAP_MUX_MODE7 | AM33XX_PIN_OUTPUT},
++ {"gpmc_a6.rgmii2_tclk", OMAP_MUX_MODE7 | AM33XX_PIN_OUTPUT},
++ {"gpmc_a7.rgmii2_rclk", OMAP_MUX_MODE7 | AM33XX_PIN_OUTPUT},
++ {"gpmc_a8.rgmii2_rd3", OMAP_MUX_MODE7 | AM33XX_PIN_OUTPUT},
++ /* Grounding gpio1_6 (pin 3 Conn A) signals bone tester to start diag tests */
++ {"gpmc_ad6.gpio1_6", OMAP_MUX_MODE7 | AM33XX_PIN_INPUT_PULLUP},
++};
++
+ /* Module pin mux for eCAP0 */
+ static struct pinmux_config ecap0_pin_mux[] = {
+ {"ecap0_in_pwm0_out.gpio0_7", AM33XX_PIN_OUTPUT},
+@@ -698,6 +754,18 @@ static void tsc_init(int evm_id, int profile)
+ pr_err("failed to register touchscreen device\n");
+ }
+
++/*
++ * Initialization of BeagleBone specific devices
++ */
++static void bone_init(int evm_id, int profil )
++{
++ int err;
++ setup_pin_mux(bone_pin_mux);
++ err = platform_add_devices(bone_devices, ARRAY_SIZE(bone_devices));
++ if (err)
++ pr_err("failed to register LEDS\n");
++}
++
+ static void rgmii1_init(int evm_id, int profile)
+ {
+ setup_pin_mux(rgmii1_pin_mux);
+@@ -1103,6 +1171,7 @@ static struct evm_dev_cfg beaglebone_dev_cfg[] = {
+ {usb0_init, DEV_ON_BASEBOARD, PROFILE_NONE},
+ {usb1_init, DEV_ON_BASEBOARD, PROFILE_NONE},
+ {mmc0_init, DEV_ON_BASEBOARD, PROFILE_NONE},
++ {bone_init, DEV_ON_BASEBOARD, PROFILE_ALL},
+ {NULL, 0, 0},
+ };
+
+diff --git a/arch/arm/mach-omap2/mux33xx.c b/arch/arm/mach-omap2/mux33xx.c
+index 0286c4f..f09e827 100644
+--- a/arch/arm/mach-omap2/mux33xx.c
++++ b/arch/arm/mach-omap2/mux33xx.c
+@@ -49,7 +49,7 @@ static struct omap_mux __initdata am33xx_muxmodes[] = {
+ NULL, NULL, NULL, NULL),
+ _AM33XX_MUXENTRY(GPMC_AD6, 0,
+ "gpmc_ad6", "mmc1_dat6", NULL, NULL,
+- NULL, NULL, NULL, NULL),
++ NULL, NULL, NULL, "gpio1_6"),
+ _AM33XX_MUXENTRY(GPMC_AD7, 0,
+ "gpmc_ad7", "mmc1_dat7", NULL, NULL,
+ NULL, NULL, NULL, NULL),
+--
+1.7.1
+
diff --git a/recipes-kernel/linux/linux-ti33x-psp-3.0+3.1rc/0001-am335x-evm-hack-in-LED-support-for-beaglebone.patch b/recipes-kernel/linux/linux-ti33x-psp-3.0+3.1rc/0001-am335x-evm-hack-in-LED-support-for-beaglebone.patch
deleted file mode 100644
index b043ce9..0000000
--- a/recipes-kernel/linux/linux-ti33x-psp-3.0+3.1rc/0001-am335x-evm-hack-in-LED-support-for-beaglebone.patch
+++ /dev/null
@@ -1,109 +0,0 @@
-From f658d7975f2e7232786190002ddd24d6b92ba965 Mon Sep 17 00:00:00 2001
-From: Denys Dmytriyenko <denys@...>
-Date: Mon, 17 Oct 2011 02:00:39 -0400
-Subject: [PATCH] am335x-evm: hack in LED support for beaglebone
-
-Signed-off-by: Koen Kooi <koen@...>
-Signed-off-by: Denys Dmytriyenko <denys@...>
----
- arch/arm/mach-omap2/board-am335xevm.c | 63 +++++++++++++++++++++++++++++++++
- 1 files changed, 63 insertions(+), 0 deletions(-)
-
-diff --git a/arch/arm/mach-omap2/board-am335xevm.c b/arch/arm/mach-omap2/board-am335xevm.c
-index 87eec7b..3a3c55b 100644
---- a/arch/arm/mach-omap2/board-am335xevm.c
-+++ b/arch/arm/mach-omap2/board-am335xevm.c
-@@ -17,6 +17,7 @@
- #include <linux/i2c.h>
- #include <linux/i2c/at24.h>
- #include <linux/gpio.h>
-+#include <linux/leds.h>
- #include <linux/spi/spi.h>
- #include <linux/spi/flash.h>
- #include <linux/mtd/mtd.h>
-@@ -588,6 +589,58 @@ static struct pinmux_config usb1_pin_mux[] = {
- {NULL, 0},
- };
-
-+/* LEDS - gpio1_21 -> gpio1_24 */
-+
-+#define BEAGLEBONE_USR1_LED GPIO_TO_PIN(1, 21)
-+#define BEAGLEBONE_USR2_LED GPIO_TO_PIN(1, 22)
-+#define BEAGLEBONE_USR3_LED GPIO_TO_PIN(1, 23)
-+#define BEAGLEBONE_USR4_LED GPIO_TO_PIN(1, 24)
-+
-+static struct gpio_led gpio_leds[] = {
-+ {
-+ .name = "beaglebone::usr0",
-+ .default_trigger = "heartbeat",
-+ .gpio = BEAGLEBONE_USR1_LED,
-+ },
-+ {
-+ .name = "beaglebone::usr1",
-+ .default_trigger = "mmc0",
-+ .gpio = BEAGLEBONE_USR2_LED,
-+ },
-+ {
-+ .name = "beaglebone::usr2",
-+ .gpio = BEAGLEBONE_USR3_LED,
-+ },
-+ {
-+ .name = "beaglebone::usr3",
-+ .gpio = BEAGLEBONE_USR4_LED,
-+ },
-+};
-+
-+static struct gpio_led_platform_data gpio_led_info = {
-+ .leds = gpio_leds,
-+ .num_leds = ARRAY_SIZE(gpio_leds),
-+};
-+
-+static struct platform_device leds_gpio = {
-+ .name = "leds-gpio",
-+ .id = -1,
-+ .dev = {
-+ .platform_data = &gpio_led_info,
-+ },
-+};
-+
-+static struct platform_device *bone_devices[] __initdata = {
-+ &leds_gpio,
-+};
-+
-+static struct pinmux_config boneled_pin_mux[] = {
-+ {"gpmc_a5.rgmii2_td0", OMAP_MUX_MODE7 | AM33XX_PIN_OUTPUT}, // gpio 21
-+ {"gpmc_a6.rgmii2_tclk", OMAP_MUX_MODE7 | AM33XX_PIN_OUTPUT}, // gpio22
-+ {"gpmc_a7.rgmii2_rclk", OMAP_MUX_MODE7 | AM33XX_PIN_OUTPUT}, // gpio23
-+ {"gpmc_a8.rgmii2_rd3", OMAP_MUX_MODE7 | AM33XX_PIN_OUTPUT}, // gpio 24
-+};
-+
- /* Module pin mux for eCAP0 */
- static struct pinmux_config ecap0_pin_mux[] = {
- {"ecap0_in_pwm0_out.gpio0_7", AM33XX_PIN_OUTPUT},
-@@ -696,6 +749,15 @@ static void tsc_init(int evm_id, int profile)
- pr_err("failed to register touchscreen device\n");
- }
-
-+static void bone_leds_init(int evm_id, int profil )
-+{
-+ int err;
-+ setup_pin_mux(boneled_pin_mux);
-+ err = platform_add_devices(bone_devices, ARRAY_SIZE(bone_devices));
-+ if (err)
-+ pr_err("failed to register LEDS\n");
-+}
-+
- static void rgmii1_init(int evm_id, int profile)
- {
- setup_pin_mux(rgmii1_pin_mux);
-@@ -1101,6 +1163,7 @@ static struct evm_dev_cfg beaglebone_dev_cfg[] = {
- {usb0_init, DEV_ON_BASEBOARD, PROFILE_NONE},
- {usb1_init, DEV_ON_BASEBOARD, PROFILE_NONE},
- {mmc0_init, DEV_ON_BASEBOARD, PROFILE_NONE},
-+ {bone_leds_init, DEV_ON_BASEBOARD, PROFILE_ALL},
- {NULL, 0, 0},
- };
-
---
-1.7.0.4
-
diff --git a/recipes-kernel/linux/linux-ti33x-psp_3.0+3.1rc.bb b/recipes-kernel/linux/linux-ti33x-psp_3.0+3.1rc.bb
index e42b8a0..ba40bdb 100644
--- a/recipes-kernel/linux/linux-ti33x-psp_3.0+3.1rc.bb
+++ b/recipes-kernel/linux/linux-ti33x-psp_3.0+3.1rc.bb
@@ -11,7 +11,7 @@ MULTI_CONFIG_BASE_SUFFIX = ""

BRANCH = "master"
SRCREV = "a95ceb11b5858feae89895d14d7f7bc41cca9fd6"
-MACHINE_KERNEL_PR_append = "d+gitr${SRCREV}"
+MACHINE_KERNEL_PR_append = "e+gitr${SRCREV}"

COMPATIBLE_MACHINE = "(ti33x)"

@@ -26,7 +26,7 @@ SRC_URI += "git://arago-project.org/git/projects/linux-am33x.git;protocol=git;br

PATCHES_OVER_PSP = " \
file://0001-f_rndis-HACK-around-undefined-variables.patch \
- file://0001-am335x-evm-hack-in-LED-support-for-beaglebone.patch \
+ file://0001-am335x-Add-pin-mux-and-init-for-beaglebone-specific-.patch \
file://0001-mach-types-Add-new-machine-type-beaglebone-to-mach-t.patch \
file://0002-Kconfig-Add-support-for-beaglebone-machine-id.patch \
file://0003-am335xevm-Use-new-beaglebone-machine-id.patch \
--
1.7.0.4


_______________________________________________
Angstrom-distro-devel mailing list
Angstrom-distro-devel@...
http://lists.linuxtogo.org/cgi-bin/mailman/listinfo/angstrom-distro-devel

[added meta-ti list]

Op 20 okt. 2011, om 05:52 heeft Joel A Fernandes het volgende geschreven:

Pick up uEnv.txt/user.txt U-boot environment files installed in /boot

This doesn't match what the patch does :/

Signed-off-by: Joel A Fernandes <joelagnel@...>
---
classes/sdcard_image.bbclass | 4 +++-
1 files changed, 3 insertions(+), 1 deletions(-)

diff --git a/classes/sdcard_image.bbclass b/classes/sdcard_image.bbclass
index ed9d779..c535ad2 100644
--- a/classes/sdcard_image.bbclass
+++ b/classes/sdcard_image.bbclass
@@ -86,8 +86,10 @@ IMAGE_CMD_sdimg () {
suffix=bin
fi

+ cp -v ${IMAGE_ROOTFS}/boot/{user.txt,uEnv.txt} ${WORKDIR}/tmp-mnt-boot || true
+
if [ -e ${IMAGE_ROOTFS}/boot/u-boot.$suffix ] ; then
- cp -v ${IMAGE_ROOTFS}/boot/{u-boot.$suffix,user.txt,uEnv.txt} ${WORKDIR}/tmp-mnt-boot || true
+ cp -v ${IMAGE_ROOTFS}/boot/{u-boot.$suffix} ${WORKDIR}/tmp-mnt-boot || true

Your patch is effectively a noop, I don't see the value in applying it.

regards,

Koen

On Thu, Oct 20, 2011 at 12:40 AM, Koen Kooi <koen@...> wrote:

[added meta-ti list]

Op 20 okt. 2011, om 05:52 heeft Joel A Fernandes het volgende geschreven:

Pick up uEnv.txt/user.txt U-boot environment files installed in /boot

This doesn't match what the patch does :/

Yes it does

Signed-off-by: Joel A Fernandes <joelagnel@...>
---
classes/sdcard_image.bbclass |    4 +++-
1 files changed, 3 insertions(+), 1 deletions(-)

diff --git a/classes/sdcard_image.bbclass b/classes/sdcard_image.bbclass
index ed9d779..c535ad2 100644
--- a/classes/sdcard_image.bbclass
+++ b/classes/sdcard_image.bbclass
@@ -86,8 +86,10 @@ IMAGE_CMD_sdimg () {
              suffix=bin
      fi

+     cp -v ${IMAGE_ROOTFS}/boot/{user.txt,uEnv.txt} ${WORKDIR}/tmp-mnt-boot || true
+
      if [ -e ${IMAGE_ROOTFS}/boot/u-boot.$suffix ] ; then
-             cp -v ${IMAGE_ROOTFS}/boot/{u-boot.$suffix,user.txt,uEnv.txt} ${WORKDIR}/tmp-mnt-boot || true
+             cp -v ${IMAGE_ROOTFS}/boot/{u-boot.$suffix} ${WORKDIR}/tmp-mnt-boot || true

Your patch is effectively a noop, I don't see the value in applying it.

There is quite a lot of value in applying it.

I suggest you try to build an image (cloud9 / ti-hw-bringup) and
notice that there is no uEnv.txt in the boot partition.

Thanks,
Joel

Op 20 okt. 2011, om 08:17 heeft Joel A Fernandes het volgende geschreven:

On Thu, Oct 20, 2011 at 12:40 AM, Koen Kooi <koen@...> wrote:

[added meta-ti list]

Op 20 okt. 2011, om 05:52 heeft Joel A Fernandes het volgende geschreven:

Pick up uEnv.txt/user.txt U-boot environment files installed in /boot

This doesn't match what the patch does :/

Yes it does

"pick up" implies you're adding something new instead of fixing it. Please write better commit messages.

Signed-off-by: Joel A Fernandes <joelagnel@...>
---
classes/sdcard_image.bbclass | 4 +++-
1 files changed, 3 insertions(+), 1 deletions(-)

diff --git a/classes/sdcard_image.bbclass b/classes/sdcard_image.bbclass
index ed9d779..c535ad2 100644
--- a/classes/sdcard_image.bbclass
+++ b/classes/sdcard_image.bbclass
@@ -86,8 +86,10 @@ IMAGE_CMD_sdimg () {
suffix=bin
fi

+ cp -v ${IMAGE_ROOTFS}/boot/{user.txt,uEnv.txt} ${WORKDIR}/tmp-mnt-boot || true
+
if [ -e ${IMAGE_ROOTFS}/boot/u-boot.$suffix ] ; then
- cp -v ${IMAGE_ROOTFS}/boot/{u-boot.$suffix,user.txt,uEnv.txt} ${WORKDIR}/tmp-mnt-boot || true
+ cp -v ${IMAGE_ROOTFS}/boot/{u-boot.$suffix} ${WORKDIR}/tmp-mnt-boot || true

Your patch is effectively a noop, I don't see the value in applying it.

There is quite a lot of value in applying it.

I suggest you try to build an image (cloud9 / ti-hw-bringup) and
notice that there is no uEnv.txt in the boot partition.

If you're fixing the bug that {1,2,3} will only pick 1, but forget 2,3 then user.txt is still broken. If you're saying uboot.suffix is not in /boot, we need to fix that. Either way, I'm not supposed to be guessing that, you should explain that in the commit message or with comments in the code or both.

regards,

Koen

On Thu, Oct 20, 2011 at 12:38 AM, Koen Kooi <koen@...> wrote:

[added meta-ti list]


Op 20 okt. 2011, om 03:54 heeft Joel A Fernandes het volgende geschreven:

Currently added for following:
 * User LEDs (added by Koen Kooi <k-kooi@...>)
 * BeagleBone tester

Build and run time tested.

Signed-off-by: Joel A Fernandes <joelagnel@...>
---
v2 changes:
Replaced correct patch

I don't like this, at all. Please do your changes as a seperate patch, not as a change to an existing patch. And if you  update a patch in the future please use -B -C to git format-patch so it can detect the rename and produce a real diff instead of a add+delete pair.

O.K.

Thanks,
Joel

Signed-off-by: Joel A Fernandes <joelagnel@...>
---
classes/sdcard_image.bbclass |    4 +++-
1 files changed, 3 insertions(+), 1 deletions(-)

diff --git a/classes/sdcard_image.bbclass b/classes/sdcard_image.bbclass
index ed9d779..c535ad2 100644
--- a/classes/sdcard_image.bbclass
+++ b/classes/sdcard_image.bbclass
@@ -86,8 +86,10 @@ IMAGE_CMD_sdimg () {
              suffix=bin
      fi

+     cp -v ${IMAGE_ROOTFS}/boot/{user.txt,uEnv.txt} ${WORKDIR}/tmp-mnt-boot || true
+
      if [ -e ${IMAGE_ROOTFS}/boot/u-boot.$suffix ] ; then
-             cp -v ${IMAGE_ROOTFS}/boot/{u-boot.$suffix,user.txt,uEnv.txt} ${WORKDIR}/tmp-mnt-boot || true
+             cp -v ${IMAGE_ROOTFS}/boot/{u-boot.$suffix} ${WORKDIR}/tmp-mnt-boot || true

Your patch is effectively a noop, I don't see the value in applying it.

There is quite a lot of value in applying it.

I suggest you try to build an image (cloud9 / ti-hw-bringup) and
notice that there is no uEnv.txt in the boot partition.

If you're fixing the bug that {1,2,3} will only pick 1, but forget 2,3 then user.txt is still broken. If you're saying uboot.suffix is not in /boot, we need to fix that. Either way, I'm not supposed to be guessing that, you should explain that in the commit message or with comments in the code or both.

Please let the sdcard class copy the U-boot environment files (as it
originally was) from /boot to the boot partition and not make it
conditional on the presence of anything else.

Joel

Op 20 okt. 2011, om 08:54 heeft Joel A Fernandes het volgende geschreven:

Updated the USER LED patch to include BeagleBone tester specific initialization/code.

Build and run time tested.

Signed-off-by: Joel A Fernandes <joelagnel@...>
---
v3:
* formatted patch with -B -C
* Modified commit summary

applied, thanks!

Op 20 okt. 2011, om 08:59 heeft Joel A Fernandes het volgende geschreven:

Fixed a bug where the U-boot environment files are not copied to the boot partition if they
were installed in /boot by other recipes.

We should copy the uEnv.txt and user.txt always without placing a condition for the presence
of anything else in /boot. This is the original behavior of sdcard_image before the recent
wave of changes to it.

Signed-off-by: Joel A Fernandes <joelagnel@...>
---
v2:
Updated commit summary

applied, thanks!

From: Denys Dmytriyenko <denys@...>

Updated to latest version to pick up the following changes:
* 9158d7f am335x: Really correct NAND partiton layout
* 4723773 am335x: Wrap spl_board_init in a CONFIG_SPL_BOARD_INIT check
* 40c310c Change CPU to 720MHz and bump VDD1 to 1.26V
* 9c154ee Add missing i2c0 pin mux configuration in spl_board_init

Signed-off-by: Denys Dmytriyenko <denys@...>
---
recipes-bsp/u-boot/u-boot_2011.10rc.bb | 4 ++--
1 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/recipes-bsp/u-boot/u-boot_2011.10rc.bb b/recipes-bsp/u-boot/u-boot_2011.10rc.bb
index 3f6e570..888c5bf 100644
--- a/recipes-bsp/u-boot/u-boot_2011.10rc.bb
+++ b/recipes-bsp/u-boot/u-boot_2011.10rc.bb
@@ -4,7 +4,7 @@ require u-boot.inc
COMPATIBLE_MACHINE = "(ti33x)"
DEFAULT_PREFERENCE_ti33x = "99"
PV = "2011.09+git"
-PR = "r12"
+PR = "r13"

# SPL build
UBOOT_BINARY = "u-boot.img"
@@ -21,7 +21,7 @@ SRC_URI = "git://arago-project.org/git/projects/u-boot-am33x.git;protocol=git;br
file://2011.09git/0004-am335x-evm-fall-back-to-bone-if-header.config-is-emp.patch \
"

-SRCREV = "066e4c106677e9243bd7ae26e638f8d98e1f0606"
+SRCREV = "9158d7f37d596b867039345cb2e3f8f18f54fe46"

LIC_FILES_CHKSUM = "file://COPYING;md5=1707d6db1d42237583f50183a5651ecb"

--
1.7.0.4

Op 21 okt. 2011, om 03:04 heeft Joel A Fernandes het volgende geschreven:

As we're not going to be allocating a new machine ID for BeagleBone

That's not what I heard yesterday.

And please follow the instructions at http://git.angstrom-distribution.org/cgi-bin/cgit.cgi/meta-texasinstruments/tree/README

, and as these patches
have been verified to cause the kernel to hang and thus make newer images unbootable,
lets drop it from the recipe but lets optionally leave the patches itself undeleted for now.

Signed-off-by: Joel A Fernandes <joelagnel@...>
---
Build and run time tested

recipes-bsp/u-boot/u-boot_2011.10rc.bb | 2 --
1 files changed, 0 insertions(+), 2 deletions(-)

diff --git a/recipes-bsp/u-boot/u-boot_2011.10rc.bb b/recipes-bsp/u-boot/u-boot_2011.10rc.bb
index 3f6e570..35c84da 100644
--- a/recipes-bsp/u-boot/u-boot_2011.10rc.bb
+++ b/recipes-bsp/u-boot/u-boot_2011.10rc.bb
@@ -15,8 +15,6 @@ SRC_URI = "git://arago-project.org/git/projects/u-boot-am33x.git;protocol=git;br
file://2011.09git/0001-am335x_evm-boot-kernel-from-ext2-3-filesystem.patch \
file://2011.09git/0002-am335x_evm-set-bootdelay-to-0.patch \
file://2011.09git/0003-am335x-evm-make-rootfs-RO-on-boot.patch \
- file://2011.09git/0001-mach-types-Add-new-beaglebone-machine-type.patch \
- file://2011.09git/0002-evm-If-beaglebone-is-detected-pass-the-MACH_TYPE_BEA.patch \
file://2011.09git/0003-am335x_evm-single-byte-address-EEPROM-for-board-iden.patch \
file://2011.09git/0004-am335x-evm-fall-back-to-bone-if-header.config-is-emp.patch \
"
--
1.7.0.4


_______________________________________________
Angstrom-distro-devel mailing list
Angstrom-distro-devel@...
http://lists.linuxtogo.org/cgi-bin/mailman/listinfo/angstrom-distro-devel

Op 21 okt. 2011, om 03:04 heeft Joel A Fernandes het volgende geschreven:

This patch fixes the case where the EEPROM is programmed but we still use the
fall back case to detect the bone board. Now that EEPROMs are programmed, we should
correctly detect the bone board instead of relying on fall back to bone mode.

PR bumped

Can you please rebase this on top of Denys' earlier patch that he sent to the meta-ti mailiinglist?

Signed-off-by: Joel A Fernandes <joelagnel@...>
---
* Build and run time tested
* This also verifies Jason's 8 bit EEPROM read patch

...Read-name-field-in-EEPROM-header-for-Beag.patch | 28 ++++++++++++++++++++
recipes-bsp/u-boot/u-boot_2011.10rc.bb | 3 +-
2 files changed, 30 insertions(+), 1 deletions(-)
create mode 100644 recipes-bsp/u-boot/u-boot/2011.09git/0005-am335x-evm-Read-name-field-in-EEPROM-header-for-Beag.patch

diff --git a/recipes-bsp/u-boot/u-boot/2011.09git/0005-am335x-evm-Read-name-field-in-EEPROM-header-for-Beag.patch b/recipes-bsp/u-boot/u-boot/2011.09git/0005-am335x-evm-Read-name-field-in-EEPROM-header-for-Beag.patch
new file mode 100644
index 0000000..ef901a9
--- /dev/null
+++ b/recipes-bsp/u-boot/u-boot/2011.09git/0005-am335x-evm-Read-name-field-in-EEPROM-header-for-Beag.patch
@@ -0,0 +1,28 @@
+From 186250ee6abca3f080b5321504a1c5881ffbf7c6 Mon Sep 17 00:00:00 2001
+From: Joel A Fernandes <joelagnel@...>
+Date: Thu, 20 Oct 2011 19:46:53 -0500
+Subject: [PATCH] am335x-evm: Read name field in EEPROM header for BeagleBoard bone detection
+
+Signed-off-by: Joel A Fernandes <joelagnel@...>
+---
+ board/ti/am335x/evm.c | 4 ++++
+ 1 files changed, 4 insertions(+), 0 deletions(-)
+
+diff --git a/board/ti/am335x/evm.c b/board/ti/am335x/evm.c
+index 37e348f..b63dcbe 100755
+--- a/board/ti/am335x/evm.c
++++ b/board/ti/am335x/evm.c
+@@ -501,6 +501,10 @@ int board_init(void)
+ detect_daughter_board_profile();
+ } else if (!strncmp("SKU#03", header.config, 6)) {
+ board_id = IPP_BOARD;
++ } else if (!strncmp("A335BONE", header.name, 8)) {
++ board_id = BONE_BOARD;
++ profile = 1; /* profile 0 is internally considered as 1 */
++ daughter_board_connected = 0;
+ } else {
+ printf("Did not find a recognized configuration, "
+ "assuming BeagleBone\n");
+--
+1.7.1
+
diff --git a/recipes-bsp/u-boot/u-boot_2011.10rc.bb b/recipes-bsp/u-boot/u-boot_2011.10rc.bb
index 35c84da..7054c11 100644
--- a/recipes-bsp/u-boot/u-boot_2011.10rc.bb
+++ b/recipes-bsp/u-boot/u-boot_2011.10rc.bb
@@ -4,7 +4,7 @@ require u-boot.inc
COMPATIBLE_MACHINE = "(ti33x)"
DEFAULT_PREFERENCE_ti33x = "99"
PV = "2011.09+git"
-PR = "r12"
+PR = "r13"

# SPL build
UBOOT_BINARY = "u-boot.img"
@@ -17,6 +17,7 @@ SRC_URI = "git://arago-project.org/git/projects/u-boot-am33x.git;protocol=git;br
file://2011.09git/0003-am335x-evm-make-rootfs-RO-on-boot.patch \
file://2011.09git/0003-am335x_evm-single-byte-address-EEPROM-for-board-iden.patch \
file://2011.09git/0004-am335x-evm-fall-back-to-bone-if-header.config-is-emp.patch \
+ file://2011.09git/0005-am335x-evm-Read-name-field-in-EEPROM-header-for-Beag.patch \
"

SRCREV = "066e4c106677e9243bd7ae26e638f8d98e1f0606"
--
1.7.0.4


_______________________________________________
Angstrom-distro-devel mailing list
Angstrom-distro-devel@...
http://lists.linuxtogo.org/cgi-bin/mailman/listinfo/angstrom-distro-devel

On Fri, Oct 21, 2011 at 09:16:44AM +0200, Koen Kooi wrote:

Op 21 okt. 2011, om 05:01 heeft Chris Whittenburg het volgende geschreven:

Not sure if this is the right place to report this issue, but I've run
into it a couple times lately...

I think the md5 hash for ti_cgt_c6000_6.1.17_setup_linux_x86.bin has
changed from the one in the recipe cgt6x_6.1.17.bb

Perhaps TI changed the install file?

TI did in fact change the file :( Updating the checksums will cause people
who already downloaded the old file (e.g. me) to go thru pains without a
guarantee that TI won't change the file again.

[switch to meta-ti]

Was this discussed with CGT people? Can we ensure it doesn't happen in the
future?

--
Denys

Op 21 okt. 2011, om 07:56 heeft Denys Dmytriyenko het volgende geschreven:

From: Denys Dmytriyenko <denys@...>

Updated to latest version to pick up the following changes:
* 9158d7f am335x: Really correct NAND partiton layout

That one broke a patch, but I massaged it to work again :)

Hi,

I'm currently trying to build an image for the TI 8148 EVM board. For
that I added the meta-ti layer to the standard Yocto/Edison environment.

meta-ti has under conf/machine a dm8148-evm.conf. So I wrote dm8148-evm
as MACHINE in my local.conf.

Build stopped at building u-boot_git.bb and William Mills pointed me to
u-boot_2010.06-psp.bb for the 8148 EVM board. I tried that but then the
build process complained about the SRCREV. Looking at the recipe I see:

require u-boot.inc

FILESPATHPKG =. "u-boot-psp-git:"

COMPATIBLE_MACHINE =
"am387x-evm|am389x-evm|c6a814x-evm|c6a816x-evm|dm814x-evm"

SRC_URI =
"git://arago-project.org/git/projects/u-boot-omap3.git;branch=${BRANCH};protocol=git"

BRANCH_ti814x = "ti81xx-master"
SRCREV_pn-${PN}_ti814x = "5fcf46a405fe8e8a59a04d3cebdafd39ac0c4bd0"
LIC_FILES_CHKSUM_pn-${PN}_ti814x =
"file://COPYING;md5=4c6cde5df68eff615d36789dc18edd3b"

Questions: I can solve my SRCREV problem when I substite the ti814x in
BRANCH... and SRCREV with my actual dm8148-evm. That makes me ask, what
the purpose of the COMPATIBLE_MACHINE variable is?

If this what I (beginner level) found is a bug, then where do I report
it or submit a fix for it?

The recipe built u-boot.bin now, but I don't see an MLO file in my image
directory. What did I do wrong?

After the image compiled I put it on an SD card and tried to boot with
it, but the kernel runs into a panic:

USB Video Class driver (v1.0.0)
OMAP Watchdog Timer Rev 0x00: initial timeout 60 sec
usbcore: registered new interface driver usbhid
usbhid: USB HID core driver
Unable to handle kernel NULL pointer dereference at virtual address 00000002
pgd = c0004000
[00000002] *pgd=00000000
Internal error: Oops: 5 [#1]
last sysfs file:
Modules linked in:
CPU: 0 Not tainted (2.6.37+ #2)
PC is at strcmp+0xc/0x40
LR is at omap_mbox_get+0x3c/0x1d0
pc : [<c01cf034>] lr : [<c00566c4>] psr: a0000013
sp : de83df28 ip : de83df38 fp : de83df34
r10: 00000000 r9 : 00000000 r8 : 00000000
r7 : 00000013 r6 : c0458048 r5 : c0503af8 r4 : c04de2e4
r3 : 00000064 r2 : 00000076 r1 : c0458048 r0 : 00000002
Flags: NzCv IRQs on FIQs on Mode SVC_32 ISA ARM Segment kernel
Control: 10c5387d Table: 80004019 DAC: 00000017
Process swapper (pid: 1, stack limit = 0xde83c2e8)
Stack: (0xde83df28 to 0xde83e000)
df20: de83df5c de83df38 c00566c4 c01cf034 c02cc038
c01cf034
df40: 00000000 c0503a9c c0503a98 00000000 de83df7c de83df60 c02cfb0c
c0056694
df60: 00000000 00000000 c0029c0c c0024034 de83dfa4 de83df80 c0024070
c02cf988
df80: de83c000 c0029c0c c0024034 00000000 00000013 00000000 de83dfdc
de83dfa8
dfa0: c00343b8 c0024040 de83dfc4 00000176 c04ea918 c006154c c0029c0c
c0029cb0
dfc0: c006154c 00000013 00000000 00000000 de83dff4 de83dfe0 c0008cf4
c0034304
dfe0: 00000000 c0008c40 00000000 de83dff8 c006154c c0008c4c 00010960
4a408a08
Backtrace:
[<c01cf028>] (strcmp+0x0/0x40) from [<c00566c4>] (omap_mbox_get+0x3c/0x1d0)
[<c0056688>] (omap_mbox_get+0x0/0x1d0) from [<c02cfb0c>]
(notify_shm_drv_setup+0x190/0x268)
r6:00000000 r5:c0503a98 r4:c0503a9c
[<c02cf97c>] (notify_shm_drv_setup+0x0/0x268) from [<c0024070>]
(notify_init+0x3c/0x2b0)
r5:c0024034 r4:c0029c0c
[<c0024034>] (notify_init+0x0/0x2b0) from [<c00343b8>]
(do_one_initcall+0xc0/0x194)
r8:00000000 r7:00000013 r6:00000000 r5:c0024034 r4:c0029c0c
r3:de83c000
[<c00342f8>] (do_one_initcall+0x0/0x194) from [<c0008cf4>]
(kernel_init+0xb4/0x164)
r9:00000000 r8:00000000 r7:00000013 r6:c006154c r5:c0029cb0
r4:c0029c0c
[<c0008c40>] (kernel_init+0x0/0x164) from [<c006154c>] (do_exit+0x0/0x61c)
r5:c0008c40 r4:00000000
Code: e89da800 e1a0c00d e92dd800 e24cb004 (e4d03001)
---[ end trace e6ffc3d1c2d89a51 ]---
Kernel panic - not syncing: Attempted to kill init!

Now I start wondering if I'm using the right kernel for this board.
meta-ti/recipes-kernel/linux/lists recipes for a lot of kernels, but
according to the conf/machine/include/ti814x.inc I should have used the
right one. So how can I debug this kernel panic?

Regards
Rainer
--
Dipl.-Inf. (FH) Rainer Koenig
Project Manager Linux Clients
Dept. PDG WPS R&D SW OSE

Fujitsu Technology Solutions
Bürgermeister-Ullrich-Str. 100
86199 Augsburg
Germany

Telephone: +49-821-804-3321
Telefax: +49-821-804-2131
Mail: mailto:Rainer.Koenig@...

Internet ts.fujtsu.com
Company Details ts.fujitsu.com/imprint.html

Op 21 okt. 2011, om 11:07 heeft Rainer Koenig het volgende geschreven:

Hi,

I'm currently trying to build an image for the TI 8148 EVM board. For
that I added the meta-ti layer to the standard Yocto/Edison environment.

Please follow the instructions detailed in the meta-ti README: http://git.angstrom-distribution.org/cgi-bin/cgit.cgi/meta-texasinstruments/tree/README

Yocto/Poky lacks support for various meta-ti (and OE-classic) constructs needed for a successfull build. Following the instructions above you will get a proper setup.

regards,

Koen