Page 3 of 4

Re: [6/Sep/2013][RECOVERY]EXTENDED TWRP|v2.6.1.1

PostPosted: Fri Sep 06, 2013 1:30 pm
by Robbie P
Just thought that i would add that in the magldr test above, i had wp7 in nand, hence the need for a sd recovery.
Not sure what the results would be if using magldr with android in nand, but ppl would surely use nand recovery (a few may use sd recovery for whatever reason eg test new recovery before changing to nand and possibly having to increase recovery size). If using nand recovery, then wm6.5 would not be installed, so a check for haret is not needed in nand recovery. Also if using wp7, magldr is sure to be installed.
Also it was stock wm6.5 for eu model, Artemis, other custom and tmous wm versions might be different.
Hope this helps

Edit; Thinking a bit more about this, for nand-only android users, the sd version should surely be the preferred option since you would then have more space in nand for system if there were no recovery partition. No need to increase recovery partition size as and when larger versions arrive, also recovery could then be much larger if needed.

Edit2; Thinking again, to detect haret, you could add a line in twrp2/startup.txt
set mtype 2524
set ramaddr 0x11800000
set ramsize 0x1e400000
set initrd_offset 0x00a00000
set KERNEL zImage
set initrd initrd.gz
set cmdline "rel_path=twrp2"
set bootloader "HaRET"
boot


Then test for bootloader="HaRET" in twrp. Magldr doesn't read startup.txt anyway and clk just boots sboot afaik.
I had similar discussion with macs18max here http://forum.xda-developers.com/showpos ... tcount=351 to 357

Re: [6/Sep/2013][RECOVERY]EXTENDED TWRP|v2.6.1.1

PostPosted: Fri Sep 06, 2013 6:41 pm
by tytung
Robbie P wrote:Just thought that i would add that in the magldr test above, i had wp7 in nand, hence the need for a sd recovery.
Not sure what the results would be if using magldr with android in nand, but ppl would surely use nand recovery (a few may use sd recovery for whatever reason eg test new recovery before changing to nand and possibly having to increase recovery size). If using nand recovery, then wm6.5 would not be installed, so a check for haret is not needed in nand recovery. Also if using wp7, magldr is sure to be installed.
Also it was stock wm6.5 for eu model, Artemis, other custom and tmous wm versions might be different.
Hope this helps

Edit; Thinking a bit more about this, for nand-only android users, the sd version should surely be the preferred option since you would then have more space in nand for system if there were no recovery partition. No need to increase recovery partition size as and when larger versions arrive, also recovery could then be much larger if needed.

Edit2; Thinking again, to detect haret, you could add a line in twrp2/startup.txt

Then test for bootloader="HaRET" in twrp. Magldr doesn't read startup.txt anyway and clk just boots sboot afaik.
I had similar discussion with macs18max here http://forum.xda-developers.com/showpos ... tcount=351 to 357


There can be multiple key-value pairs in the cmdline.
set cmdline "rel_path=twrp2 bootloader=haret"

Re: [6/Sep/2013][RECOVERY]EXTENDED TWRP|v2.6.1.1

PostPosted: Fri Sep 06, 2013 7:34 pm
by twoxx
Hi!

Everytime i want to install a zip-file... twrp 2.6.1.1 makes a hot reboot.

I can do backups and restores, but no install

Re: [6/Sep/2013][RECOVERY]EXTENDED TWRP|v2.6.1.1

PostPosted: Fri Sep 06, 2013 8:06 pm
by Odysseus
twoxx wrote:Hi!

Everytime i want to install a zip-file... twrp 2.6.1.1 makes a hot reboot.

I can do backups and restores, but no install

What bootloader are you using, Black LK, cLK, MAGLDR, haret? Do you have it set up to correctly match your partitioning scheme? As in data on ext, native nand, or sd?

I suspect you don't have TWRP configured properly and that's why you're having issues.

Re: [6/Sep/2013][RECOVERY]EXTENDED TWRP|v2.6.1.1

PostPosted: Fri Sep 06, 2013 8:44 pm
by twoxx
Odysseus wrote:What bootloader are you using, Black LK, cLK, MAGLDR, haret? Do you have it set up to correctly match your partitioning scheme? As in data on ext, native nand, or sd?

I suspect you don't have TWRP configured properly and that's why you're having issues.



i use magldr. i set "data on ext" because i am using data on ext. what else do i have to configure? v 2.6.1.0 is working correctly. only v. 2.6.1.1 has issues.

Re: [6/Sep/2013][RECOVERY]EXTENDED TWRP|v2.6.1.1

PostPosted: Fri Sep 06, 2013 9:16 pm
by Odysseus
twoxx wrote:

i use magldr. i set "data on ext" because i am using data on ext. what else do i have to configure? v 2.6.1.0 is working correctly. only v. 2.6.1.1 has issues.


I switched to Black LK several months ago, so I'm unable to reproduce your issue. Are you sure you installed the extended TWRP and not the official version? When exactly is it crashing, when you select the ROM or during the unpacking and installation process?

The only other thing I can think of off the top of my head is perhaps your recovery partition is too small. The recommended size is 8MB though many have reported needing 10MB for it to work correctly. I hope this helps.

Ciao

Re: Feature Request

PostPosted: Sat Sep 07, 2013 2:30 pm
by agbommarito
Kokotas,

There is a feature in TWRP to check partitions every so many boots. Would it be possible to add a message on the display when a boot check occurs? This would give you a clue as to why the boot was taking so much longer than normal.

And thanks for a great product, by the way.

Sent from my Nexus 7 using Tapatalk 4

PostPosted: Sat Sep 07, 2013 9:18 pm
by NYLimited
agbommarito wrote:Kokotas,

There is a feature in TWRP to check partitions every so many boots. Would it be possible to add a message on the display when a boot check occurs? This would give you a clue as to why the boot was taking so much longer than normal


I may be mistaken but I think that setting is not *for* twrp. It is a setting for e2fsck which is not normally run by twrp.

If you check the rom's init.rc you should find where Tytung executes e2fsck before mounting the ext partition. The twrp setting - as I understand it - sets the value for this process via tune2fs and the "-C" parameter (upper case C, not lower case!) to force a check every X boots.

-C mount-count
Set the number of times the filesystem has been mounted. If set to a greater value than the max-mount-counts parameter set by the -c option, e2fsck(8) will check the filesystem at the next reboot.


You should also be able to locate the fsck log file in the device's root which has the result of the last check. I used to automatically email this file to myself on every boot.

The init process is much more limited in commands than a shell script and the message you are asking for would have to be coming during this process.

Keep in mind that all this can indeed make the boot process longer at times.

Re:

PostPosted: Sun Sep 08, 2013 3:44 am
by agbommarito
NYLimited wrote:
I may be mistaken but I think that setting is not *for* twrp. It is a setting for e2fsck which is not normally run by twrp.

If you check the rom's init.rc you should find where Tytung executes e2fsck before mounting the ext partition. The twrp setting - as I understand it - sets the value for this process via tune2fs and the "-C" parameter (upper case C, not lower case!) to force a check every X boots.



You should also be able to locate the fsck log file in the device's root which has the result of the last check. I used to automatically email this file to myself on every boot.

The init process is much more limited in commands than a shell script and the message you are asking for would have to be coming during this process.

Keep in mind that all this can indeed make the boot process longer at times.


Thanks for the explanation. I thought that TWRP was somehow involved since it has the option to modify the frequency at which the check occurs.

Re: [6/Sep/2013][RECOVERY]EXTENDED TWRP|v2.6.1.1

PostPosted: Sun Sep 08, 2013 12:56 pm
by andy
The newest v2.6.1.1 cann't install ROM. I have tried Tytung's cm4.2 and MIUI2.3