! LaTeX Error: File `lmodern.sty' not found.

! LaTeX Error: File `lmodern.sty' not found.

install lmodern to fix:.

sudo apt-get install lmodern

USB Host support!

Dear all,
today I've reached USB host support for OTG and USB Host2 port too!
This was done adapting and correcting the Daniel Mack RFC for MX3 EHCI support.
Wait some more for a deep description on changes :p

Bye!
Alberto!

Updates, NOR and NAND flash support.

Modifications on mx3fb driver are not to be accepted because of the maintainer won't add machine specific code in that file.. Nevermind, I keep the patch in my hand waiting for another solution :)

In those days I worked on NOR & NAND flash support:
NOR flash device is mounted on the CPU credit card and it works well with physmap mapping driver.
NAND flash device is mounted on the motherboard and supporting it was a bit harder:
In tree there is a NAND flash driver for MXC architecture mxc_nand.c derived from original freescale driver and corrected over coding style and kernel way of doing by Sascha Hauer . Unfortunately this driver do not support flash devices with large page size (2kB wide, over 512 B of small page size devices) and upgrading it was over my capability..
The solution came from a graceful guy called Vladimir Barinov from linux-mtd list where I posted the problem. He gave me the right patch that merged with my work make the NAND flash device working!

So now I can store data in all memory devices mounted on the board! Great!

Alberto!

Framebuffer OK!! explanation

OOK what was the problem and what I've understood:
First of all I built with success the OSELAS.BSP-Phytec-phyCORE-10 following the great guide OSELAS.BSP-Phytec-phyCORE-i.MX31-Quickstart.pdf, remember, I had already built the OSELAS toolchain optimized for arm-1136jfs-linux-gnueabi.

Booting with the new filesystem (after some adjustment to /etc/network/interfaces to have eth0 statically assigned ip and /etc/modules to do not try to load unwanted modules) was fine, but fbset was still not working!
From the logo printed on the screen the problem with mx3fb driver was for sure not only in timings parameters of the video mode but also some other timings settings.

Fortunately the driver mx3fb is derived from the Freescale mxcfb making comparison simpler :)
With a deep debugging and comparison between the two drivers: mx3fb and mxcfb (from Atmark version of 2.6.26 kernel) I've found this:

in ipu_sdc.c:ipu_sdc_init_panel

#ifdef CONFIG_MACH_ARMADILLO5X0
old_conf = (div |
((div >> 4) - 1) << 24 | /* DISP3_IF_CLK_DOWN_WR */
((div >> 4) - 2) << 14); /* DISP3_IF_CLK_UP_WR */
mx3fb_write_reg(mx3fb, old_conf, DI_DISP3_TIME_CONF);
#else
/*
* DISP3_IF_CLK_DOWN_WR is half the divider value and 2 fraction bits
* fewer. Subtract 1 extra from DISP3_IF_CLK_DOWN_WR based on timing
* debug. DISP3_IF_CLK_UP_WR is 0
*/
mx3fb_write_reg(mx3fb, (((div / 8) - 1) << 22) | div, DI_DISP3_TIME_CONF);
#endif

The DI_DISP3_TIME_CONF sdc register contain timing configuration for the pixel writing process (from lower bit):

• DISP3_IF_CLK_PER_WR: 12 bit (8 bit for integer part and 4 bit for fractional) sets the divider for HSP_CLOCK (ipu_clk) to obtain the right pixel period.
• DISP3_IF_CLK_UP_WR: 10 bit (6 bit for integer part and 4 bit for fractional) sets the rising edge position for display write access.
• DISP3_IF_CLK_DOWN_WR: 10 bit (6 bit for integer part and 4 bit for fractional) sets the falling edge position for display write access.

The Atmark armadillo 500 board have a Video DAC (ADV7125) that transforms digital synchronous display signals in analog signals to drive a standard VGA connector for standard display devices. Those different timing configurations was due Video DAC timing specifications and so this piece of code must be ported to mx3fb driver in sdc_init_panel function!

Yes! first timing misconfiguration solved!

But the screen was still over margins, not well proportioned and last but not least there was a shadowed band of a few pixels in width on the entire height of the screen!
All those misconfiguration was solved calibrating the timings mode parameters moving the image on right over the shadowed band. This maintaining the timings constraints of the VGA displays.

How work the VGA standard display:

Horizontal sync signal:
__   ______________________________   ___________
 |_|                              |_|
 |A|
 |---------------B----------------|

Data:
        ______________________          ________
________|        VIDEO         |________| VIDEO (next line)
   |-C-|----------D-----------|-E-|

A: Horizontal sync pulse.
B: Scanline period made of:
C: Left border (Back porch) waiting for current stability.
D: Valid Video data.
E: Right border (Front porch) cutting off current instability.

Vertical timings definition are similar.
Now for a resolution of 640x480 at 60Hz refresh rate (VGA standard) are defined those frequencies:

Clock frequency 25.175 MHz (Pixel frequency)
Line  frequency 31469 Hz   -> 800 pixel per line.
Field frequency 59.94 Hz   -> 525 lines.

Timing mode parameters must meet those constraints but we can decide amounts of left and upper margins panning the valid image onto a well displayed position!

After many tests, the better result was achieved with this configuration:

{    /* 640x480 @ 60 Hz */
   .name         = "CRT-VGA",
   .refresh      = 60,
   .xres         = 640,
   .yres         = 480,
   .pixclock     = 39721, //Period in picos
   .left_margin  = 35,
   .right_margin = 115,
   .upper_margin = 43,
   .lower_margin = 1,
   .hsync_len    = 10,
   .vsync_len    = 1,
   .sync         = FB_SYNC_OE_ACT_HIGH,
   .vmode        = FB_VMODE_NONINTERLACED,
   .flag         = 0,
},

Same process for the resolution 800x600 with 56Hz of refresh rate results in this video mode:

{    /* 800x600 @ 56 Hz */
      .name         = "CRT-SVGA",
      .refresh      = 56,
      .xres         = 800,
      .yres         = 600,
      .pixclock     = 30000,
      .left_margin  = 30,
      .right_margin = 108,
      .upper_margin = 13,
      .lower_margin = 10,
      .hsync_len    = 10,
      .vsync_len    = 1,
      .sync         = FB_SYNC_OE_ACT_HIGH | FB_SYNC_HOR_HIGH_ACT |
                      FB_SYNC_VERT_HIGH_ACT,
      .vmode        = FB_VMODE_NONINTERLACED,
      .flag         = 0,
  },

Here a well formed information repository on how VGA work.

Next, USB!

Framebuffer OK!!

Five days spent solving framebuffers problems, but damn! I've got it! :)
Next post will be the full explanation..
Bye!

Framebuffer calibration!

I'm trying to calibrate the frame buffer modes.
Basically I tried to apply the same modes from the old Atmark implementation for CRT-VGA and CRT-SVGA in drivers/video/mxc/mxcfb.c to the new mx3fb.c driver in 2.6.30 kernel but the screen look out of borders in left and up. Also words printed have soft yellow shadows and the logo seems to be printed in 16 colours not 16 bpp :)

fbset is the utility made for me now (here the man page) it give me the ability to change geometry and timings of the screen but seems not work!
I've tried for so much time, but no results and every time recompiling the kernel to see changes of adjusting parameters is not a praticable way.

Now I'm trying to build the OSELAS BSP for their i.MX31 board. Thanks to Pengutronix!
Hoping this will help me :)

Bye!

Nice Day Today! :)

Guys, I am proud to announce that the console-image from openembedded is working!
I've got full serial login and VGA login prompt!

Next work:

• RTC not found? maybe shoud be only enabled from kernel config.
• Adjust VGA output: It print over screen and seems to have only 16 colours not 16bpp!
• Add USB support: Mouse! Keyboard!

After that i will focus on Android!

Openembedded and a sutable distro

Here all the steps to build a valid root filesystem where I can deal with all my work since here!

I choose Openembedded as build system to build the linux distribution, first for curiosity (in so many place it is mentioned an so many people tell me about), last for no other known choices! (Sorry at anyone that fight to open the way for your own embedded distribuition build system)

Openembedded main page is here and following the Getting Started guide i had been able to create a valid Openembedded enviroment in my workstation:

Base dir
$ mkdir -p /media/dati/oe/stuff/build/conf
$ cd /media/dati/oe/stuff/

For BitBake
$ svn co svn://svn.berlios.de/bitbake/branches/bitbake-1.8/ bitbake

For Openembedded
$ git clone git://git.openembedded.net/openembedded

Next I created the BitBake configuration file locale.conf:

$ cd /media/dati/oe/stuff/
$ cp openembedded/conf/local.conf.sample build/conf/local.conf

Editing it for me it result: (without comments..):

TOPDIR = "/media/dati/oe"
DL_DIR = "${TOPDIR}/sources"

BBFILES = "${TOPDIR}/stuff/openembedded/recipes/*/*.bb"

BBMASK = ""

PREFERRED_PROVIDERS = "virtual/qte:qte virtual/libqpe:libqpe-opie"
PREFERRED_PROVIDERS += " virtual/libsdl:libsdl-x11"
PREFERRED_PROVIDERS += " virtual/${TARGET_PREFIX}gcc-initial:gcc-cross-initial"
PREFERRED_PROVIDERS += " virtual/${TARGET_PREFIX}gcc-intermediate:gcc-cross-intermediate"
PREFERRED_PROVIDERS += " virtual/${TARGET_PREFIX}gcc:gcc-cross"
PREFERRED_PROVIDERS += " virtual/${TARGET_PREFIX}g++:gcc-cross"

MACHINE = "armadillo500"

TARGET_OS = "linux"

DISTRO = "angstrom-2008.1"

GLIBC_GENERATE_LOCALES = "en_GB.UTF-8 it_DE.UTF-8"

IMAGE_FSTYPES = "ext2"

CACHE = "${TOPDIR}/cache"

BBINCLUDELOGS = "yes"

And next the patch to openembedded: 0001-Initial-Armadillo-500-support-for-OpenEmbedded.patch

Applying this the command:

$ cd /media/dati/oe/stuff/
$ bitbake helloworld-image

Produce at /media/dati/oe/tmp/deploy/glibc/images/armadillo500 the file Angstrom-helloworld-image-glibc-ipk-2009.X-test-20090506-armadillo500.rootfs.ext2
that mounted as loopback filesystem on /nfsarmadilloroot:

$ sudo mount Angstrom-helloworld-image-glibc-ipk-2009.X-test-20090506-armadillo500.rootfs.ext2 /nfsarmadilloroot -t ext2 -o loop

Work great as rootfilesystem!

That's all folks :)

Next is a console-image..
Openembedded take so much time building! Sure.. it build also so much software not needed.. nevermind I'm not here to penetrate in that build process. :P

Angsrom Login!!!

Yes YES YESSSSS!
I've got the Angstrom Login prompt on the VGA screen! Yes the problem suggested was true, new kernel, wrong version of glibc.
Next the entire process to have a working Angstrom helloworld-image :)

Revision Point.

Last post in this blog was weeks ago!
I've been far from internet for a while but also not far from my thesis!
What I worked out:

1. Ethernet support (SMSC 9118) with nfs root capability.
2. MMC/SD card device support (SDHC).

Since now I worked on Frame buffer support with discrete success (Logo displayed on the screen), but THE problem is not solved! THE problem is that with the latest kernel, the distro given from Atmark crashes in early init process! with no chances of debugging!
The only things that I know are the lists of files that the init process try to load from the nfs root before it crash without outputs.

I think it is a conflict between the new kernel and the precompiled C libraries stored in the root directory.
So now I'm working for building a compatible distro from Openembedded hoping this process is painless :p

Bye!

Early Armadillo 500 support (the last?)

Ok, after reviewing mechanism this is the patch that give early support to the Atmark Armadillo 500 dev board.
The patch apply to the latests kernel source obtained by

git pull

in the root local git repository.

0001-Initial-support-for-Atmark-Armadillo-500-developing.patch

Author and name on Git configuration

From: http://www.kernel.org/pub/software/scm/git/docs/gittutorial.html

$ git config --global user.name "Alberto Panizzo"
$ git config --global user.email maramaopercheseimorto@gmail.com

Early Armadillo 500 support

Ok! Now I am ready to post the first official Early Atmark Armadillo 500 developing board support patch!

This patch add the Armadillo 500 initial support. Enabling the mxc uart console via make gconfig (by default using make armadillo500_defconfig) we can see the console output through Kernel panic - not syncing: VFS: Unable to mount root fs on unknown-block(1,0).

To see debug strings on the console you have to apply the patch "Clock ipg_clk/ipg_per_clk synchronization" due to resolve clock the synchronization issue.

Patches attached because of changes made by this blog to the code style. (Indentation, ...)

0001-Initial-support-for-Atmark-Armadillo-500-developing.patch
0002-Add-armadillo500_defconfig-file-in-the-tree.patch
0003-Clock-ipg_clk-ipg_per_clk-synchronization.patch

Clock syncronization issue!

Time spent to analyse the imx console driver to found where is the bug that produce scrambling chars on the serial debug and the problem was not there!

The problem is ipg_clk/ipg_per_clk synchronization!
ipg_per_clk clock signal has two possible clock source decided by bit 24 (PERCS) of CCMR register.
On my armadillo 500 board the clock source was set to usb_pll instead of ipg_clk resulting on not synchronized clocks signals on the UART device! producing wired characters on the output.

This patch correct the issue, for me and also for all iMX31 boards (if misconfigured)

diff --git a/arch/arm/mach-mx3/clock.c b/arch/arm/mach-mx3/clock.c
index ca46f48..f1a581d 100644
--- a/arch/arm/mach-mx3/clock.c
+++ b/arch/arm/mach-mx3/clock.c
@@ -581,6 +581,11 @@ int __init mx31_clocks_init(unsigned long fref)
MX32, but still required to be set */
MXC_CCM_CGR2);

+ /* Ensure ipg_per_clk clock is generated by ipg_clk and not by
+ usb_pll (CCMR:24 = 1) to do not turn off UART after disabling
+ usb_pll and mantain ipg_clk and ipg_per_clk syncronization */
+ __raw_writel(__raw_readl(MXC_CCM_CCMR) | (1 << 24), MXC_CCM_CCMR);
+
usb_pll_disable(&usb_pll_clk);

pr_info("Clock input source is %ld\n", clk_get_rate(&ckih_clk));

Very Early Support!

Ook some time working and first code work fine.
This is the first debugging patch to the kernel vanilla 2.6.29 and the result is very early debug strings from /arch/arm/boot/compressed/head.S and /arch/arm/kernel/head-common.S b/arch/arm/kernel/head-common.S printed on the serial port. And an Alive message that state all is gone ok before call start_kernel().
Ehm..! no debug output is shown after start_kernel call..
Next is porting ttymcx console driver to support printk() message printing.

The patch:

diff --git a/arch/arm/boot/compressed/head.S b/arch/arm/boot/compressed/head.S
index b371fba..5d08227 100644
--- a/arch/arm/boot/compressed/head.S
+++ b/arch/arm/boot/compressed/head.S
@@ -17,6 +17,9 @@
* 100% relocatable. Any attempt to do so will result in a crash.
* Please select one of the following when turning on debugging.
*/
+
+#define DEBUG 1
+
#ifdef DEBUG

#if defined(CONFIG_DEBUG_ICEDCC)
diff --git a/arch/arm/kernel/head-common.S b/arch/arm/kernel/head-common.S
index 991952c..5d43cfa 100644
--- a/arch/arm/kernel/head-common.S
+++ b/arch/arm/kernel/head-common.S
@@ -50,7 +50,9 @@ __mmap_switched:
1: cmp r6, r7
strcc fp, [r6],#4
bcc 1b
-
+
+ bl __alive
+
ldmia r3, {r4, r5, r6, r7, sp}
str r9, [r4] @ Save processor ID
str r1, [r5] @ Save machine type
@@ -69,6 +71,16 @@ ENDPROC(__mmap_switched)
* and hope for the best (useful if bootloader fails to pass a proper
* machine ID for example).
*/
+__alive:
+ mov r10,r0
+ adr r0, str_a
+ bl printascii
+ mov r0,r10
+ mov pc, lr
+str_a: .asciz "\n Alive!\n"
+ .align
+ENDPROC(__alive)
+
__error_p:
#ifdef CONFIG_DEBUG_LL
adr r0, str_p1
diff --git a/arch/arm/mach-mx3/Kconfig b/arch/arm/mach-mx3/Kconfig
index d623558..4ac4337 100644
--- a/arch/arm/mach-mx3/Kconfig
+++ b/arch/arm/mach-mx3/Kconfig
@@ -64,4 +64,12 @@ config MACH_QONG
Include support for Dave/DENX QongEVB-LITE platform. This includes
specific configurations for the board and its peripherals.

+config MACH_ARMADILLO500
+ bool "Support Atmark Armadillo-500 Development Base Board"
+ select ARCH_MX31
+ default n
+ help
+ Include support for Atmark Armadillo-500 platform. This includes
+ specific configurations for the board and its peripherals.
+
endif
diff --git a/arch/arm/mach-mx3/Makefile b/arch/arm/mach-mx3/Makefile
index 272c8a9..4659598 100644
--- a/arch/arm/mach-mx3/Makefile
+++ b/arch/arm/mach-mx3/Makefile
@@ -14,3 +14,4 @@ obj-$(CONFIG_MACH_MX31_3DS) += mx31pdk.o
obj-$(CONFIG_MACH_MX31MOBOARD) += mx31moboard.o mx31moboard-devboard.o \
mx31moboard-marxbot.o
obj-$(CONFIG_MACH_QONG) += qong.o
+obj-$(CONFIG_MACH_ARMADILLO500) += armadillo5x0.o
diff --git a/arch/arm/mach-mx3/armadillo5x0.c b/arch/arm/mach-mx3/armadillo5x0.c
new file mode 100644
index 0000000..ae4e35e
--- /dev/null
+++ b/arch/arm/mach-mx3/armadillo5x0.c
@@ -0,0 +1,117 @@
+/*
+ * armadillo500.c
+ *
+ * Copyright 2009 Alberto Panizzo
+ * updates in http://alberdroid.blogspot.com/
+ *
+ * Based on Atmark Techno, Inc. armadillo 500 BSP 2008 All Rights Reserved.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston,
+ * MA 02110-1301, USA.
+ */
+
+#include
+#include
+#include
+
+/* Define Machine specific Memory regions */
+#include
+/* All machine types definitions */
+#include
+/* Machine descriptor definition */
+#include
+/* Time keepeing definitions */
+#include
+/* Memory mapping functions */
+#include
+/* Define map_desc */
+#include
+
+/* Platform functions */
+#include
+#include
+
+
+
+static void __init armadillo5x0_init(void)
+{
+}
+
+/*!
+ * This structure defines static mappings for the i.MX31ADS board.
+ */
+static struct map_desc armadillo5x0_io_desc[] __initdata = {
+ /*{//Internal RAM
+ .virtual = MX31_IRAM_BASE_ADDR_VIRT,
+ .pfn = __phys_to_pfn(IRAM_BASE_ADDR),
+ .length = IRAM_SIZE,
+ .type = MT_NONSHARED_DEVICE},
+ },*//* {//ROM PATCH
+ .virtual = ROMP_BASE_ADDR_VIRT,
+ .pfn = __phys_to_pfn(ROMP_BASE_ADDR),
+ .length = ROMP_SIZE,
+ .type = MT_DEVICE_NONSHARED
+ },*/ {//Shared Perpheral Bus Arbiter
+ .virtual = SPBA0_BASE_ADDR_VIRT,
+ .pfn = __phys_to_pfn(SPBA0_BASE_ADDR),
+ .length = SPBA0_SIZE,
+ .type = MT_DEVICE_NONSHARED
+ }, {//NOR Flash memory (32 M)
+ .virtual = CS0_BASE_ADDR_VIRT,
+ .pfn = __phys_to_pfn(CS0_BASE_ADDR),
+ .length = CS0_SIZE,
+ .type = MT_DEVICE
+ }, {//Extended Bus Region
+ .virtual = CS4_BASE_ADDR_VIRT,
+ .pfn = __phys_to_pfn(CS4_BASE_ADDR),
+ .length = CS4_SIZE,
+ .type = MT_DEVICE
+ },
+};
+
+/*!
+ * Set up static virtual mappings.
+ */
+static void __init armadillo5x0_map_io(void)
+{
+ /*
+ * Maps:
+ * - NAND, SDRAM, WEIM, M3IF, EMI controllers
+ * - AVIC (Vectored Interrupt controller)
+ * - AIPS1 and AIPS2
+ **/
+ mxc_map_io();
+ iotable_init(armadillo5x0_io_desc, ARRAY_SIZE(armadillo5x0_io_desc));
+}
+
+static void __init armadillo5x0_timer_init(void)
+{
+ mx31_clocks_init(26000000);
+}
+
+static struct sys_timer armadillo5x0_timer = {
+ .init = armadillo5x0_timer_init,
+};
+
+MACHINE_START(ARMADILLO5X0, "Armadillo-500")
+ /* Maintainer: Atmark Techno, Inc. */
+ .phys_io = AIPS1_BASE_ADDR,
+ .io_pg_offst = ((AIPS1_BASE_ADDR_VIRT) >> 18) & 0xfffc,
+ .boot_params = PHYS_OFFSET + 0x00000100,
+ //.map_io = armadillo5x0_map_io,
+ //.init_irq = mxc_init_irq,
+ //.timer = &armadillo5x0_timer,
+ //.init_machine = armadillo5x0_init,
+MACHINE_END
diff --git a/arch/arm/plat-mxc/include/mach/board-armadillo5x0.h b/arch/arm/plat-mxc/include/mach/board-armadillo5x0.h
new file mode 100644
index 0000000..b16897b
--- /dev/null
+++ b/arch/arm/plat-mxc/include/mach/board-armadillo5x0.h
@@ -0,0 +1,29 @@
+/*
+ * Copyright 2005-2007 Alberto Panizzo .
+ * All Rights Reserved.
+ */
+
+/*
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+#ifndef __ASM_ARCH_MXC_BOARD_ARMADILLO5X0_H__
+#define __ASM_ARCH_MXC_BOARD_ARMADILLO5X0_H__
+
+#include
+
+/*
+ * Other memory mapping regions
+ */
+
+#define CS0_BASE_ADDR_VIRT 0xE0000000
+#define CS0_SIZE SZ_128M
+
+/* mandatory for CONFIG_LL_DEBUG */
+
+#define MXC_LL_UART_PADDR UART1_BASE_ADDR
+#define MXC_LL_UART_VADDR AIPS1_IO_ADDRESS(UART1_BASE_ADDR)
+
+#endif
diff --git a/arch/arm/plat-mxc/include/mach/debug-macro.S b/arch/arm/plat-mxc/include/mach/debug-macro.S
index 4f77314..d3440ab 100644
--- a/arch/arm/plat-mxc/include/mach/debug-macro.S
+++ b/arch/arm/plat-mxc/include/mach/debug-macro.S
@@ -34,6 +34,9 @@
#ifdef CONFIG_MACH_QONG
#include
#endif
+#ifdef CONFIG_MACH_ARMADILLO500
+#include
+#endif
.macro addruart,rx
mrc p15, 0, \rx, c1, c0
tst \rx, #1 @ MMU enabled?

Compiling:

apply the patch...
$ export ARCH=arm
$ export CROSS_COMPILE=/media/dati/android/cupcake/prebuilt/linux-x86/toolchain/arm-eabi-4.2.1/bin/arm-eabi-
$ make allnoconfig
$ make gconfig
->System Type>ARMSystemType>Freescale MXC/iMX based=y
->Freescale MXC Implementation > Support Atmark Armadillo-500 Development Board=y
$ make
-> download to the board and boot!

iMX Linux Boot Process description online

Within my links on right side now there is the pdf document that describe the entire Linux boot process adapted for arm i.MX31: after the bootloader work through the execution of /sbin/init!

Taste it! Bye!

Very deep focus on ARM Boot process

Here Peter Pearse make a very deep and up-to-date focus analysis of SMP ARM Boot process:
http://www.linux-arm.org/LinuxBootLoader/SMPBoot
Thanks a lot to him :)

With little adjusting Voilà! the single core ARM Boot process!
The context where I have to work and without I will not understand nothing :p

Kernel XIP

What is XIP?? Only for memo.. eXecute In Place.
Ref: http://elinux.org/Kernel_XIP

What Run First

Form Arm-Linux Mailing list, posted by Russell King
http://lists.arm.linux.org.uk/lurker/message/20011005.154003.5d698445.html

What runs first depends on what type of kernel you start running.

Compressed kernel (aka zImage):

1. the startup code in arch/arm/boot/compressed/head.S (start label)
   (and maybe some other architecture specific code in other head-*.S files in that directory).
   
2. Decompressor proper (via decompress_kernel) in arch/arm/boot/compressed/misc.c which gunzips the kernel.
   
3. We then relocate the kernel if required, and call it.
   

Uncompressed kernel (aka Image):
This is where we join the uncompressed kernel startup.

4. arch/arm/kernel/head-armv.S (Edit head.S and head-*.S ) describes this, and includes such things as:
   page table initialization and processor cache initialization
   (via support stuff in arch/arm/mm/proc-*.S).
5. We then proceed to start_kernel in init/main.c, and everything follows from that point on.
   
6. The only other thing to note is that the various drivers are initialized via a table built by the linker from the __initcall() and module_init() statements scattered throughout the kernel source. The code which calls them can be found in init/main.c: do_initcalls().
   

All begin by main.c

All the time I spent since now was for looking to understand the real order in booting/init procedure (not counting time spent for diffs over different kernel source and bsp's)

Here I found an effective utile documentation: http://www.linux-arm.org/LinuxKernel/LinuxNewPlatformPort

In Kernel Boot section we can find that the function init/main.c:start_kernel() starts the kernel over all architecture, and give the real order of the initialization sequence.

Sorry it it was obvious I am a sort of Davide that fight with Golia in this work :)

Anyway every tutorial/docs I've found is related to old kernel version..

C Preprocessor and Macro, symbol ##

It is easy to see in kernel sources within macros the symbol ## but what it mean??
It is used to Concatenate "token" of text source: when the macro is expanded first the preprocessor substitute the variables name with their values and then concatenate tokens divided by ##

#define init_func(prefix_name, arg) prefix_name##_init(arg)

so calling:

init_func(armadillo, null)

expands in:

armadillo_init(null)

fox!

Reference: http://gcc.gnu.org/onlinedocs/cpp/Concatenation.html#Concatenation

Fermo per un pò.. ma riprendo!

Dopo gli eventi della settimana scorsa ho ripreso il lavoro..
Ora mi stò concentrando sulla creazione del supporto ad armadillo 500 per la versione corrente del kernel: 2.6.29 Questo perché:

1. Atmark con la loro bsp si sono strafregati dello sviluppo mailine del kernel..
2. -> La bsp risultate è difficile da portare su kernel recenti.
3. Se devo fare lo sforzo di rivedere l'intera bsp, cosa mi trattiene da farlo fatto bene in modo che possa convivere con tutte le altre??
4. Per Android è già presente la versione del kernel 2.6.29..! skeggie! :)
   

Compilare il kernel Atrmark 2.6.26 per armadillo 500

Ecco le maggiori info :)
Cercando in rete ho scoperto che su armadillo 500-FX è già ststo portato android con kernel 2.6.26.
Dato che le due macchine sono così simili a livello hardware mi è parso così strano che il kernel più aggiornato disponibile per armadillo 500 sia il solo 2.6.18.. Quindi ho cercato tra i repository dell'armadillo 500-FX i sorgenti del kernel con l'intenzione di portarlo su armadillo-500.
Il kernel disponibile era proprio il 2.6.26 e con gioia ho scoperto che esisteva anche il file di configurazione per armadillo 500! nonché tutti i sorgenti specifici della CPU board e driver della board di sviluppo!

Bene! Allora i passi per il deploiment:
1
Scaricare e decomprimere i sorgenti del kernel dai repository atmark:

$ wget http://download.atmark-techno.com/armadillo-500-fx/source/kernel/linux-2.6.26-at3.tar.gz
$ tar -xvf linux-2.6.26-at3.tar.gz

2
Configurare e compilare i sorgenti per armadillo con l'aiuto di android:

$ cd linux-2.6.26-at3
$ make armadillo500_defconfig
$ export ARCH=arm
$ export CROSS_COMPILE=[basePathAndroid]/cupcake/prebuilt/linux-x86/toolchain/arm-eabi-4.2.1/bin/arm-eabi-
$ make

Nota: ho usato la toolchain di cupcake, credo che anche con il branch master non abbia problemi..
3
Deploiment nella scheda:
In ./arch/arm/boot sono stati creati i file immagine del kernel: Image e compresso zImage
quindi copiando zImage nella cartella /tftp e scaricandolo nella flash attraverso hermit come nei post precedenti (indicando il nome di file zImage eh!) al reboot tutto funziona correttamente :)

Nota: il comando tftpdl da hermit della scheda, all'attivazione della periferica di rete è come se alcune volte non avesse il link attivato.. rimanendo in attesa della risposta dal server per il download.
Workaround: se da workstation eseguo il ping al'indirizzo impostato per la board il download dell'immagine, prima in attesa, parte contemporaneamente.

Linux 2.6.26 done!!!

Yeeeee!! spulciando tra i repository atmark ho trovato i sorgenti del kernel 2.6.26! i quali supportano l'armadillo 500!
Compilato il tutto con la toolchain arm-eabi di android e scaricato nella board funziona!!

Dettagli alla prossima volta!

Boot dell'ultima distribuzine armadilliana!

Oggi ho terminato la traduzione dei manuali pdf Atmark, i quali contengono si abbastanza informazioni per utilizzare l'Armadillo ma poco o nulla sulle modifiche al kernel per supportare la board.. Spero ci sia qualcosa in più in giro nel kernel tree!
Comunque, mi sono concentrato a far girare il codice più aggiornato possibile -> kernel e distribuzione scaricabili dall'area di download Atmark dedicata.

Cosa ho fatto:

1. Ho aggiornato il kernel nella flash (Unici device di boot supportati dal bootloader sono la flash interna e la compactflash che non ho..).
2. Ho Installato la debian-armadillo distribuzione in una partizione della mia SD card utilizzandola come rootfilesystem.

1
Ci sono tre modi per scrivere nella flash della scheda:

1. Con Hermit-at installato nella Workstation.
2. Con Shoehorn-at installato nella Workstation.
3. Da bootloader (Hermit-at nel device) attraverso il protocollo tftp.
   

Le prime due modalità richiedono l'installazione nella workstation delle relative utility cosa possibile compilando i sorgenti oppure installando i pacchetti debian .deb tramite dpkg.
Visto che lavoro su una Ubuntu non volevo scombinare il tutto ed allora ho optato per la terza!

La terza modalità prevede la presenza di un server tftp raggiungibile nella rete. Nel post precedente descrivo come eseguirne uno nella mia macchina e per controllarne il debug eseguo da shell:

$ inetd -d.

Scaricata l'ultima versione del kernel da http://download.atmark-techno.com/armadillo-500/image/linux-a500-1.05.bin.gz nella cartella /tftp sono pronto per l'aggiornamento:
Da una nuova shell eseguo il minicom (Impostazioni: rate 115200 bps, data 8bit, 1bit stop, No parity, No flowcontrol ) (Il jumper JP1 è cortocircuitato così da farmi presentare il menu del bootloader).
Dal menu del bootloader eseguo:

hermit> tftpdl 192.168.0.10 192.168.0.2 --kernel=linux-a500-1.05.bin.gz
(tftpdl deviceIp workstationIp --regionName=fileToDownload)

Ok! aggiornato il kernel!
Con il seguente comando dico al bootloader di eseguire il boot da flash:

hermit> setbootdevice flash

2
La debian-armadillo si trova qui http://download.atmark-techno.com/armadillo-500/debian/ in 4 file .tgz da estrarre in un'unica cartella.
Dato che in kernel c'è il supporto per i device MMC/SD allora decido di sfruttare la mia schedina da 1 GB ->

• Cancello le partizioni preesistenti.
• Creo un'unica partizione ext2.
• Ci estraggo i debian-armadillo files.
• Rootfs pronto!
  

Da menu del bootloader non devo far altro che definire le opzioni di boot del kernel ed eseguire il boot:

hermit> clearenv
hermit> setenv console=ttymcx0 root=/dev/mmcblk0p1 rootdelay=1
hermit> boot

Risultati

• Nella console seriale ho una shell completamente funzionante.
• L'uscita vga mi presenta la console virtuale tty1
• Settando le impostazioni di rete, il controller SMSC funziona perfettamente.
• Non ci sono moduli in /lib/modules .. ma comunque ciò che serve credo sia tutti in kernel (ma nel caso volessi supporti per espansioni ...! ricompilare il kernel o installare moduli aggiuntivi)
  

Server tftp con ubuntu

Link di riferimento:
http://mylinuxmind.myblog.it/tag/ubuntu%20tftp
Per il download di file nella flash della scheda tramite bootloader è utile creare un server tftp.

Installare il demone:

sudo apt-get install tftpd-hpa

Creiamo una cartella per lo scambio dei dati e diamo accesso in lettura e scrittura a chi vuole:

sudo mkdir /tftp
sudo chmod a+r /tftp
sudo chmod a+w /tftp

Ora modifichiamo il file di configurazione dei servizi /etc/inet.conf (Ubuntu 8.04 hardy)
la riga tftp deve essere modificata in:

tftp dgram udp wait root /usr/sbin/in.tftpd /usr/sbin/in.tftpd -s /tftp

dove ogni campo significa:
#servicename sockettype protocol wait/nowait user serverprogram serverprogarguments
l'argomento -s indica al dempne tftpd di utilizzare la directory /tftp come root directory, se si vuole più debug aggiungere -v.

con:

sudo inetd

attiveremo anche il server tftp, specificando l'opzione -d verrà visualizzato il debug dei servizi su shell.

Alle prese con traduzioni...

Dopo una giornata alla ricerca di vari automatismi per la traduzione della documentazione dal Giapponese all'Inglese ho trovato questa serie di passi che rendono il tutto abbastanza semplice..
Essenzialmente mi servo della capacità di Google translate di tradurre intere pagine web (basta metterci il link al posto del testo..) e di ZOHO Viewer uno strumento di condivisione dei documenti capace di visualizzare i file pdf.

Tips and Tricks:
Google Translate non riesce a tradurre più di 5-6 pagine di pdf alla volta! -> se si carica l'intero file verrà tradotto pressocé il solo indice..
Pdftk ci viene in aiuto con burst per suddividere il file pdf nelle sue singole pagine:

pdftk nomefile.pdf burst

e merge per fondere singole pagine in unici file pdf:

pdftk nome1.pdf nome2.pdf ... cat output merge.pdf

Allora, suddivisi i file di documentazione in gruppi di 6 pagine, stò trascrivendo le informazioni iportanti della scheda che pubblicherò :)

Ah.. chissà se mi rispondono alla Atmark per la richiesta di documentazione in inglese..! o faranno la stessa cosa???

Porting del Kernel Android 2.6.27 su Armadillo 500

Link di riferimento:
Specifiche della board: http://www.atmark-techno.com/en/products/armadillo/a500/specs
Specifiche del processore: http://www.arm.com/products/CPUs/ARM1136JF-S.html
Area di download sw armadillo: http://download.atmark-techno.com/armadillo-500/

Bene Bene! Prima fase reale del mio progetto: Porting del kernel più aggiornato possibile nella board di sviluppo Atmark Armadillo 500! Questa monta un processore Freescale i.MX31L con set di istruzioni ARMv6.
Ho ordinato il mio fantastico libro Building Embedded Linux Systems, Second Edition che spero illuminerà la mia strada ma nell'attesa che arrivi cerco di individuare le modifiche fatte al kernel "vanilla" per supportare la scheda nella versione presente sul sito atmark.

Importante: Questi Giapponesi scrivono solo in giapponese! Servirà la mail che ho inviato per la richiesta di doc inglese?

Altro: Il kernel a disposizione è un 2.6.18 debian etch1 sssoooo old! :)

Sul repository locale del kernel

Nella cartella "master/prebuilt/android-arm/kernel/" troviamo un bellissimo file "PREBUILT" che recita:

To rebuild the emulator kernel:

% git clone git://android.kernel.org/kernel/common.git kernel
% cd kernel
% export ARCH=arm
% export CROSS_COMPILE=arm-eabi-
% make goldfish_defconfig
% make

If kernel/common.git is mapped into your repo client, you can
just built it from there instead of re-cloning it.

ehm.. un pò da adattare dato che non menziona il branch android-goldfish-2.6.27 (nel branch common del kernel non c'è il file goldfish_defconfig) e che la toolchain di crosscompilazione non è (per ora) nel path d'esecuzione principale (non abbiamo modificato PATH).
Ma almeno abbiam trovato delle informazioni :)

Il nostro nuovo Android sull'emulatore!!

Bene bene, compilato il Kernel 2.6.27 per la macchina goldfish eseguiamo il tutto sull'emulatore!
Raccimoliamo tutti i file di output dei processi di build necessari che sono:

• Kernel, l'immagine compressa:
  alberdroid/kernel-goldfish/.repo/manifests/arch/arm/boot/zImage
• Android, i file immagine per ramdisk userdata e system nella cartella:
  alberdroid/master/out/target/product/generic

Io ho copiato la zImage nella cartella di output di android:

$ cd PATH_GIUSTO/alberdroid/master/out/target/product/generic/
$ cp PATH_GIUSTO/alberdroid/kernel-goldfish/.repo/manifests/arch/arm/boot/zImage .

e via all'emulatore!

$ emulator -system . -kernel zImage -show-kernel

Così l'emulatore cercherà i file ramdisk.img userdata.img e system.img nella cartella corrente (-system .), utilizzerà zImage come immagine del kernel (..non utilizzare l'immagine Image ma quella compressa zImage) e nella shell corrente mostrerà l'output del kernel!

Kernel per Android..

Link di riferimento:
http://groups.google.com/group/android-kernel
Allora, la sottocartella kernel esiste in master! e perché non viene compilata????
Urc! solita ricerca in rete e vien fuori che la versione che ho non è propriamente la versione per goldfish -> attiviamo il branch giusto!
..consiglio, creiamo un'altra directory per il nostro kernel:

$ mkdir -p alberdroid/kernel-goldfish
$ cd alberdroid/kernel-goldfish
$ repo init -u git://android.git.kernel.org/kernel/common.git -b android-goldfish-2.6.27
$ repo sync

Solito tempo che passa.....
poi:
-
Edit: perché ha salvato i sorgenti in .repo/manifests ??
cmq, dal precedente codice shell

$ cd .repo/manifests

e si continua con:
-

$ export ARCH=arm
$ export CROSS_COMPILE=PATHGIUSTO/alberdroid/master/prebuilt/linux-x86/toolchain/arm-eabi-4.2.1/bin/arm-eabi-
$ make goldfish_defconfig
$ make

Verrà così creato un buon kernel per l'emulatore :)

Prima Build!

Dopo il sync con i sorgenti via al processo di make!
se si vuol compilare il sistema per l'emulatore goldfish (emulatore arm v5) allora dalla directory madre eseguiamo:

$ cd ~/alberdroid/master
$ make

Il processo è abbastanza lungo (più di un'ora nel mio portatile) e necessita di molta memoria! 1GB di ram non basta.....

Ma al termine, con il comando:

$ emulator -system out/target/product/generic/ \
-kernel prebuilt/android-arm/kernel/kernel-qemu -show-kernel

Vedremo apparire il nostro caro emulatore arm con android funzionante!!
Chiaro.. dovremmo avere l'emulatore installato correttamente.. rimando all'installzione dell'sdk per le applicazioni di android.. ma notiamo una cosa fondamentale: Il kernel!
In quasi due ore di compilazione il processo make non ha compilato il kernel! ho dovuto usarne uno che stà nella cartella prebuilt!

Ma siamo contenti d'aver compilato tutto il resto dai..! Il prossimo post sarà dedicato al kernel!

1.1 Download dei sorgenti!

Link di riferimento:
http://source.android.com/download

Allora, il progetto opensource di Android utilizza Git per il Version Management ed è stato sviluppato uno script "repo" per automatizzare molte operazioni.
Qui un manuale sull'utilizzo di Git e repo.

1-Download di Repo

repo è uno script eseguibile e per questo è utile posizionarlo in una cartella presente nel search path di bash, si consiglia quindi di creare una cartella ~/bin, aggiungerla alla variabile d'ambiente PATH scaricare quindi lo script e renderlo eseguibile:

$ cd ~
$ mkdir bin

nel file .bashrc inserire:

PATH=$PATH:/home/NOMEUTENTE/bin
export PATH

poi..

$ curl http://android.git.kernel.org/repo >~/bin/repo
$ chmod a+x ~/bin/repo

Repo è uno script in continuo sviluppo, eseguendolo verrà stampata un'informazione relativa alla presenza di versioni più aggiornate se necessario.

2-Inizializzazione dei sorgenti

Creiamo una cartella che conterrà i sorgenti di android.. attenzione che i sorgenti non compressi sono circa 2,1GB mentre dopo una build verranno occupati in tutto circa 4 GB (6GB a detta del sito ufficiale).

$ mkdir -p alberdroid/master
$ cd alberdroid/master

ed inizializziamo il repository locale con:

$ repo init -u git://android.git.kernel.org/platform/manifest.git

La cartella master è stata creata perché con questo comando verrà scaricato il barnch "master" con "git branch" otterremo i branch disponibili e con

$ repo init -u git://android.git.kernel.org/platform/manifest.git -b BRANCHNAME

inizializzeremo il branch voluto.

3-Download dei sorgenti

dalla cartella alberdroid/master eseguire:

$ cd alberdroid/master
$ repo sync

Per sincronizzare singoli progetti..

$ repo sync NOME1 NOME2 ..

Capitolo Primo, l'inizializzazione!

Link di riferimento:
http://source.android.com/download

Bene, la mia distribuzione è la Ubuntu 8.04 su Pentium M (32bit) aggiornata il più possiblie :p

I pacchetti richiesti sono:

• Git 1.5.4 or newer and the GNU Privacy Guard.
• JDK 5.0, update 12 or higher. Java 6 is not supported, because of incompatibilities with @Override. (Ma se me la fa installare dopo con java6-sdk??)
• flex, bison, gperf, libsdl-dev, libesd0-dev, libwxgtk2.6-dev (optional), build-essential, zip, curl.
  

e quindi si procede con:

$ sudo apt-get install git-core gnupg sun-java6-jdk flex bison gperf libsdl-dev libesd0-dev libwxgtk2.6-dev build-essential zip curl libncurses5-dev zlib1g-dev

oook procediamo con il download...

Presentazione!

Ciao! Sono Alberto aka Alberdroid per voi :)
Il questo Blog leggerete delle mie peripezie nell'avventura: Porting di Android lella board di sviluppo ... (Puntini da riempire più avanti :p).

Beh... presentazione breve ma efficace.. in sostanza, stay tuned!