Introducing ubiboot N9 (multiboot OS loader) - Page 12 - maemo.org

Active Topics

The N900 is now 15 years old! (6)
to Nokia N900 by Macros - 1 day, 2 hrs ago
F(x)tec Pro1x (QWERTY) for sale (0)
to Buy & Sell by aln00000 - 1 day, 15 hrs ago
n770 youtube video (0)
to General by nmdv - 5 days, 14 hrs ago
more...

Page 12 of 140

Thread Tools

Hurrian	2013-03-11 , 09:32
Posts: 1,225 \| Thanked: 1,905 times \| Joined on Feb 2011 @ Quezon City, Philippines	#111

The Linux kernel can boot from extended partitions just fine.
Bootloaders like GRUB may have problems being installed to them, but we don't use that.

Actually, it may even be possible to partition the N9/50 with GPT, making the number of partitions made on it a non-issue.

__________________
N9 PR 1.3 Open Mode + kernel-plus for Harmattan
@kenweknot, working on Glacier for Nemo.

Quote & Reply |

juiceme	2013-03-11 , 09:39
Community Council \| Posts: 4,920 \| Thanked: 12,867 times \| Joined on May 2012 @ Southerrn Finland	#112

coderus, are you still having problems mounting the ubifs root when booting?

I am sure the install method goes exactly as I describe in the README document, as I tried it following my own instructions down to letter:

Just this morning, I helped a friend on the other side of country to install ubiboot and Nitdroid on his device.
The guy in question is total noob, so what we did is I instructed him to open a ssh connection for me, and then I created shared VNC session and remonte-operated his device staright from clean-flashing it to installing ubiboot & jellybean.

For this reason, I belive your device might have faulty NAND flash, and so unable to mount the ubifs root.

It is possible that people who have used their NAND for swap have burned it so badly it cannot be mounted

Quote & Reply |

	peterleinchen	2013-03-11 , 16:41
	Posts: 4,118 \| Thanked: 8,901 times \| Joined on Aug 2010 @ Ruhrgebiet, Germany	#113

Thanks theyosh, Hurrian and juiceme for your input.

Theyosh may be right, that Nemo may not be possible to boot from p7 instead of p4 (opposed to Hurrians -fully agreed from side- answer)?

@juiceme
of course I modified that func and edited some more passages where p4 was mentioned (even in kexec_load). And also modified cmdline of Nemo boot partition.

The kernel of Nemo bzImage is already located as link inside tarball, so there is no need to copy anything after untaring. Right?

I checked everything once again. And I do not get it. Grrr... :angry:
I have attached all my mods in a zip file.
Nitdroid and all kernels of Harmattan work fine.
If you could check that once again, please?

Attached Files

mods.zip (13.4 KB, 166 views)

__________________
SIM-Switcher, automated SIM switching with a Double (Dual) SIM adapter
--
Thank you all for voting me into the Community Council 2014-2016!

Please consider your membership / supporting Maemo e.V. and help to spread this by following/copying this link to your TMO signature:
[MC eV] Maemo Community eV membership application, http://talk.maemo.org/showthread.php?t=94257

editsignature, http://talk.maemo.org/profile.php?do=editsignature

Quote & Reply |

juiceme	2013-03-11 , 21:32
Community Council \| Posts: 4,920 \| Thanked: 12,867 times \| Joined on May 2012 @ Southerrn Finland	#114

Originally Posted by peterleinchen

Thanks theyosh, Hurrian and juiceme for your input.

Theyosh may be right, that Nemo may not be possible to boot from p7 instead of p4 (opposed to Hurrians -fully agreed from side- answer)?

@juiceme
of course I modified that func and edited some more passages where p4 was mentioned (even in kexec_load). And also modified cmdline of Nemo boot partition.

The kernel of Nemo bzImage is already located as link inside tarball, so there is no need to copy anything after untaring. Right?

I checked everything once again. And I do not get it. Grrr... :angry:
I have attached all my mods in a zip file.
Nitdroid and all kernels of Harmattan work fine.
If you could check that once again, please?

On checking your modificatinos they seem to be OK.
How exactly does your nemo behave when you boot it, when you select the kernel does it boot after that, and stay hung or does it boot back to ubiboot?

Could you try to boot it manually, meaning: boot to the repair console and log in to the device. Then copy the /proc/original_cmdline and change the partition and give path to the kernel, and give kexec_load;

something like this:

Code:

OCMD=$(cat /proc/original_cmdline); TMP1=$(echo "$OCMD" | sed -e "s/root\=\/dev\/mmcblk0p2/root\=\/dev\/mmcblk0p7/") ; TMP2=$(echo "$TMP1" | sed -e "s/ init\=\/sbin\/preinit/init\=\/sbin\/init/") ; FCMD="\"$TMP2\"" ; kexec -l --type=zImage --command-line=$FCMD /mnt/7/boot/bzImage ; echo $?

If the return value is 0, then the kernel and command line was loaded correctly, and you can try to boot it with command;

Code:

/usr/bin/disable_pm ; kexec -e ; echo $? ; sleep 10

There is one thing still that could affect it, maybe on the command line you need to specify that the boot partition is extended partition, similarily as on booting ubifs partition you need to specify the partition label?

Quote & Reply |

	peterleinchen	2013-03-11 , 22:33
	Posts: 4,118 \| Thanked: 8,901 times \| Joined on Aug 2010 @ Ruhrgebiet, Germany	#115

Just came to "similar" conclusion

No, not at least at your level regarding kernel tweaking ...

But according to your question: after selection the menu fades out and then the screen stays blank.
I have added a few log outputs, most important to you/me will be the result of eval --> it is 1.
You will find the used selections (cmdline) in log.txt.
But without knowing where the output comes from not so good, or?

So important message comes from load_nemo func

3 _ /boot/Nemo/boot/bzImage _ _
3 _ /boot/Nemo/boot/bzImage _ "init=/sbin/init root=/dev/mmcblk0p7 rootwait rootflags=errors=remount-ro rootfstype=ext4 rw mtdoops.mtddev=log mtdoops.record_size=65536 console=tty0 mtdparts=omap2-onenand:1024k(bootloader),2816k@1024k(config),1638 4k@3840k(kernel),2048k@20224k(log),469248k@22272k( var),32768k@491520k(moslo) bootreason=sw_rst bootmode=normal nolo=2.3.6 product_name=RM-696 dbi-type=V5JW dbi-vendor=1 dbi-size=73 dbi-uid=01234xxxxxxxxxxxxxxx imei=35xxxxxxxxxxxxx g_nokia.iSerialNumber=35xxxxxxxxxxxxx serialnumber=CTKxxxxxx omap_sec.kci=958 gpio=cmt_apeslpx:93:sysfsutput,cmt_en:23:sysfs:i nit_activeutput,cmt_rst_rq:88:sysfs:init_active: output,sleep_ind:92:sysfsutput omapdss.def_disp_enabled=1 vhash=xxxx93xxxxxded180a3xxxxx42c0d09b0dxxxxx product_model=N9 hwid=1601" _ /boot/Nemo/boot/bzImage
1

selected_os _ selected_kernel _ f_command_line _ bootkernel
selected_os _ selected_kernel _ f_command_line _ bootkernel
$?

What is the option to set for loading from logical instead of primary?

Thank you for caring.

--edit
had to modify your manual script to match my current situation. And struggled with virtual kb and not yet installed any text editor and BT and and and and
But got it up and did see something about func not implemented and something with proc_atags. now batt shutdown (N9s and mine also

)
Good night.

Last edited by peterleinchen; 2013-03-11 at 23:23.

Quote & Reply |

	peterleinchen	2013-03-12 , 08:31
	Posts: 4,118 \| Thanked: 8,901 times \| Joined on Aug 2010 @ Ruhrgebiet, Germany	#116

Good morning.

Seems I hit now another real serious problem.
After shutting down yesterday I attached the N9 to charger without looking at it. This morning I detached (again without looking) and it did not start up. No LED at all. Even after decoupling battery no lights/vibrates/starts.

Seems it started ubi, but did not start charging and battery drained fully. Have to find a charger now (and may flasher tools). Will report back.

P.S.: I had noticed some strange restart behaviour before, but had similar with u-boot and did not care.

Quote & Reply |

	coderus	2013-03-12 , 09:36
	Posts: 6,436 \| Thanked: 12,701 times \| Joined on Nov 2011 @ Ängelholm, Sweden	#117

@juiceme issue still exists for me. i tried to create ubi partition only 20MB, it before first bad area, but still not loading. is there any other way to set fixed size for ubi partiton?

__________________
Telegram | Openrepos | GitHub | Revolut donations

Quote & Reply |

juiceme	2013-03-12 , 12:00
Community Council \| Posts: 4,920 \| Thanked: 12,867 times \| Joined on May 2012 @ Southerrn Finland	#118

Originally Posted by peterleinchen

Good morning.

Seems I hit now another real serious problem.
After shutting down yesterday I attached the N9 to charger without looking at it. This morning I detached (again without looking) and it did not start up. No LED at all. Even after decoupling battery no lights/vibrates/starts.

Seems it started ubi, but did not start charging and battery drained fully. Have to find a charger now (and may flasher tools). Will report back.

P.S.: I had noticed some strange restart behaviour before, but had similar with u-boot and did not care.

Hi!

Yes, this behaviour can be explained:
(this is a long posting but full of content, so forgive me)

The N9 (like most other modern handhelds) is powered by LiIon/LiPo battery, and charging the battery is a delicate process. This is made more complicated by the fact that the power is sourced from USB line which has it's own specific quirks; For the USB power to be available both ends need to negotiate and announce each other, only then can the source start providing full current to the sink. If the negotiation fails, the power draw is limited to 200mA as a safety feature.

How this is handled in N9: When USB power is applied, even from a dummy chager, the N9 will boot up the kernel and install the BME drivers to handle the charging. When it is "switched off" the user expects it to be dead, but it cannot be for purpose of charging. This is why, if the device is switched off and connected to a charger, it in reality boots up, but it "pretends to be off", it is in the ACT_DEAD state where it runs but user should think it is off.

Unfortunately ubiboot comes there on the way, and the device which normally would boot the stock Harmattan kernel in ACT_DEAD mode now boots to ubiboot. Ubiboot sees that the USB is connected and enters the repair console, exporting directories over USB and waiting for telnet connection.

This is actually a bug in ubiboot I know of but have not corrected yet, what it should do is to check the startup flags, and if NOLO instructed it to "play dead", then it should start up harmattan in ACT_DEAD mode.

Reason why it's not corrected is that I only now thought about it, since I never charge my own device when it is in poweroff state. (I actually tend not to ever power it off...)

Now, when all power has been drained from the N9 battery, it cannot any longer boot up so much that it would get to charging state.
(actually this is also true of Lumias, and propably other smartphones too, at least I know my son's Nexus7 has trouble charging when it is totally drained)

Luckily, there is simple solution how to fix it, you need to use the flasher to jump-start the battery to 10% capacity, and then connect it to charger to top up the charge.

Instructions can be found, for example here: http://talk.maemo.org/showthread.php?t=87709

Quote & Reply |

The Following 2 Users Say Thank You to juiceme For This Useful Post:
fri.K, michaelmhk

juiceme	2013-03-12 , 12:06
Community Council \| Posts: 4,920 \| Thanked: 12,867 times \| Joined on May 2012 @ Southerrn Finland	#119

Originally Posted by coderus

@juiceme issue still exists for me. i tried to create ubi partition only 20MB, it before first bad area, but still not loading. is there any other way to set fixed size for ubi partiton?

I have not tried to resize it, but I am fairly certain at least the whole hard partition size cannot be changed as the use of the partitions is determined by NOLO.
However it really might be possible to resize the filesystem size smaller than the actual partition size.
You might want to consult Jonni for example, as he should know better if it is possible and give some advice on that...

Quote & Reply |

	peterleinchen	2013-03-12 , 12:31
	Posts: 4,118 \| Thanked: 8,901 times \| Joined on Aug 2010 @ Ruhrgebiet, Germany	#120

Originally Posted by juiceme

Hi!

Yes, this behaviour can be explained:
(this is a long posting but full of content, so forgive me)

Yep, thanks for explaining in detail (not that I was not aware of that, but it is good to reload to memory

)

Originally Posted by juiceme

This is actually a bug in ubiboot I know of but have not corrected yet, what it should do is to check the startup flags, and if NOLO instructed it to "play dead", then it should start up harmattan in ACT_DEAD mode.

That'll be nice, because I sometimes just connect to charger without even knowing it is off. Also for sleeping or charging purpose sometimes realy switch off (normally it is up and just flight mode, as you do).
Maybe add that to README or OP?

Oh. And I was able to start it up easily. Just connected charger, did see ubi screen and it was charging a bit. But ust for a minute or so, then the batt LED went off. so I assume no more power to batt. I detached it, started normally , booted to ubi then selected Harmattan kernel and immediately connected charger again. Then it booted up completely and everything fine (without need for flasher).

Thanks.

Any ideas regarding my p7 problem? (Will post the complete error log tonight, hopefully.)