182 replies to this topic

[malakudi]

Last time I checked, DMM's enigma2 still lucks many features that are available on OpenPLI for ages. Descent working teletext subtitles is just one of them. Auto language settings, descent media player (no need for EMC etc), descent graph multiepg etc still missing from DMM's enigma2. When DMM's was still open, I tried to add the Netmed EPG fix I also added on OpenPLi. Friend of mine also tried to update greek translation. There was no response from nobody at DMM.

 

DMM offers worse or overpriced hardware and worse closed source enigma2. Even all this talk about ubifs on DMM is years late. I had ubifs on my Xtrend ET9000 for ages, when DMM was still only using 2.6.18 and they only cared to update drivers just to break clone boxes.

 

DMM is now irrelevant in the Linux enigma2 scene. We have next generation boxes from VU+ and Gigablue and waiting others with their products. DMM hasn't announced anything.

 

Since you are in this scene for very long time, you should be supporting some of the open source movement and not supporting a company that screwed its open source herritage.

[Lost in Space]

Thanks for the advice, but I decide when and what i support on my own.

 

And to be honest the teletext now in the  DMM 2.0 images is far beyond of the old one in OpenPLi :-)

 

But if subtitles make your horny (sorry for the wording) - I agree :-)

Edited by gutemine, 15 April 2013 - 15:37.

[gorski]

Making light of a serious argument, never entering a dialogue, never dealing with arguments one by one, as charged - seems really poor to me.

 

Either weak or arrogant, possibly both - but this will be the end of DMM!

[luis67]

Thanks, because that's the only thing we can try for the moment.
 
The created backup image you are flashing is NOT > 60MB - because then flashing on a 500hd is likely to fail.
 
With standard images this can't happen, but with backups it is possible, if the base images is too big (eg more then 55MB used, then due to the 10% ubifs overhead you are reaching the flash limit).

I tried with opendreambox latest image and the result is the same.
FIRST STAGE 1.0/400 {BO23456}
[m

2ND STAGE OK, build #84 (2011-09-01)

....
0.031 - BCM board setup
0.032 - fp init
0.032 - ca init
0.523 - load config
0.550 - config loaded.
* woke up from power
* press [S] to enter setup
lcd: BOOT #84
- read config file: /boot/secondstage.conf
detected jffs2 blocksize: 16kb
+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
++++++++++++++++++++++UUUUUUUUUUUUUUUUUUUUUUUUUUUUU
dreamupd initialized
Scanning JFFS2 FS: . | / - \ | / - \ | / - \ | done.

find_inode failed for name=secondstage.conf

load: Failed to find inode

- failed to open /boot/secondstage.conf
- failed to open /cf/autorun.bat
lcd: /boot/bootlogo-dm500hd.elf.gz[    0.601000] Adjust partition complete size from 4000000 to 3f00000 to avoid overlap with BBT reserved space
[    0.602000] Adjust partition root size from 3c00000 to 3b00000 to avoid overlap with BBT reserved space
[    0.603000] Creating 4 MTD partitions on "brcmnand.0":
[    0.604000] 0x000000000000-0x000003f00000 : "complete"
[    0.609000] 0x000000000000-0x000000040000 : "loader"
[    0.611000] 0x000000040000-0x000000400000 : "boot"
[    0.613000] 0x000000400000-0x000003f00000 : "root"
[    0.621000] UBI: attaching mtd3 to ubi0
[    0.621000] UBI: physical eraseblock size:   16384 bytes (16 KiB)
[    0.622000] UBI: logical eraseblock size:    15360 bytes
[    0.623000] UBI: smallest flash I/O unit:    512
[    0.623000] UBI: VID header offset:          512 (aligned 512)
[    0.624000] UBI: data offset:                1024
[    0.674000] ata1: SATA link down (SStatus 4 SControl 300)
[    0.696000] hub 1-0:1.0: over-current condition on port 1
[    0.898000] hub 1-0:1.0: over-current condition on port 2
[    1.000000] ata2: SATA link down (SStatus 4 SControl 300)
[    1.101000] hub 2-0:1.0: over-current condition on port 1
[    1.217000] UBI: max. sequence number:       2155
[    1.234000] UBI: attached mtd3 to ubi0
[    1.235000] UBI: MTD device name:            "root"
[    1.235000] UBI: MTD device size:            59 MiB
[    1.236000] UBI: number of good PEBs:        3776
[    1.236000] UBI: number of bad PEBs:         0
[    1.237000] UBI: number of corrupted PEBs:   0
[    1.238000] UBI: max. allowed volumes:       89
[    1.238000] UBI: wear-leveling threshold:    4096
[    1.239000] UBI: number of internal volumes: 1
[    1.239000] UBI: number of user volumes:     1
[    1.240000] UBI: available PEBs:             0
[    1.240000] UBI: total number of reserved PEBs: 3776
[    1.241000] UBI: number of PEBs reserved for bad PEB handling: 37
[    1.242000] UBI: max/mean erase counter: 2/0
[    1.242000] UBI: image sequence number:  140662478
[    1.243000] UBI: background thread "ubi_bgt0d" started, PID 50
[    1.250000] UBIFS: recovery needed
[    1.297000] UBIFS: recovery completed
[    1.297000] UBIFS: mounted UBI device 0, volume 0, name "rootfs"
[    1.298000] UBIFS: file system size:   56908800 bytes (55575 KiB, 54 MiB, 3705 LEBs)
[    1.299000] UBIFS: journal size:       8017920 bytes (7830 KiB, 7 MiB, 522 LEBs)
[    1.300000] UBIFS: media format:       w4/r0 (latest is w4/r0)
[    1.301000] UBIFS: default compressor: zlib
[    1.301000] UBIFS: reserved for root:  0 bytes (0 KiB)
[    1.303000] VFS: Mounted root (ubifs filesystem) on device 0:13.
[    1.304000] devtmpfs: mounted
[    1.305000] Freeing unused kernel memory: 220k freed
[    1.307000] Kernel panic - not syncing: No init found.  Try passing init= option to kernel. See Linux Documentation/init.txt for guidance.
[    1.309000] Rebooting in 180 seconds..

Edited by luis67, 15 April 2013 - 16:51.

[Lost in Space]

I don't have to deal with your arguments as I already wrote now 3 times that this is a Plugin support thread and we become OT here, if you don't contribute to the plugin and want to discuss your arguments you are welcome to open your own thread.

 

@luis67

 

Regarding the boot proplems - we start to run out of options, the only remaining thing to try to make ubifs work on your box would be to flash an orginal Image and convert a virgin PLi instead of a backup - and you didn't answer my question how big your backup image is.

Edited by gutemine, 15 April 2013 - 17:39.

[gorski]

No, you misunderstood again: not mine, but malakudi's carefully crafted arguments, which depict the real logic of things behind all this, something many of us agree with, wholeheartedly.

 

And it would be really foolish of DMM if they wouldn't take them seriously, as they run deep and wide!

[Lost in Space]

Maybe we have this problem because I prefere facts and I am very sceptic on people who always believe other people's arguments :-)

 

Oriiginal JTR would not exist if I would not have taken the freedom to try out things where I was told are impossible due to the many improvements people have done to enigma2 on other boxes. And then suddenly this works on good old crappy dreamboxes too - almost unchanged.

 

Sometimes beeing a disbeliever has its advantages ...

Edited by gutemine, 15 April 2013 - 19:05.

[gorski]

Nonsense! Those arguments could have been told by anybody! Never mind who is saying it. Listen to the logic of the thing itself!

 

Well done on the JTR! As I always say - give credit where credit is due!

 

But this is not at dispute here! And there you're very "thin"!

[Lost in Space]

I'm pretty good in avoiding useless discussions and saving my time for some real work :-)

 

And strangely this works quite well to usually get what I want while others are still disputing ....

Edited by gutemine, 15 April 2013 - 19:38.

[WanWizard]

Gentlepeople. Last warning. Back on topic please.

[Lost in Space]

I try hard, really hard, belive me.

 

But you could cut out all the OT disucssion to an extra thread, because otherwise it is hard to stop.

Edited by gutemine, 15 April 2013 - 19:41.

[luis67]

I don't have to deal with your arguments as I already wrote now 3 times that this is a Plugin support thread and we become OT here, if you don't contribute to the plugin and want to discuss your arguments you are welcome to open your own thread.
 
@luis67
 
Regarding the boot proplems - we start to run out of options, the only remaining thing to try to make ubifs work on your box would be to flash an orginal Image and convert a virgin PLi instead of a backup - and you didn't answer my question how big your backup image is.

42 mb is virgin and 59 mb is backup open pli image,i tried with vigrin and backup and the result is the same.

[bacicciosat]

Hola gutemine, nice to see you around.

The world is beautiful because is various.

[Lost in Space]

well, if the 42MB are not working then I'm sorry as it works nicely on my 500hd.

 

Can somebody else confirm this Problem on his box ?

 

And yes, before movies had colors and language the people insisted that it was already great ...

[gweny]

sorry, but is it not easier to write a patch-file and bbapend-file for sources on git?

so all people can compile their own images with ubifs and didn't have to flash images x times to make an image with ubifs-support on their boxes... 

[Lost in Space]

Off course this would be easier, and with the templates of the other boxes and the build script of the DMM OE this would be probably less then 1h work (Ghost from DMM did it in half a day, but he had to start from scratch and also patch the Broadcom NAND drivers and test it extnesively, but this is done and they are open source too so there is nothing to complain) And Ghost also only had the proove from the Marsu Plugin that the boxes survice the UBIFS cure without any harm. And I testend and collected the needed patches for block2mtd driver which allows much easier ubifs booting and extracting then the nandsim hassle you are currently using. Again all open source and all patches publically available and except one which just streamlines the argument passing to block2mtd that I did myself all are signed off by the mtd developers. So there is no single piece of non-open source involved in the ubifs implementation, the newer kernel and newer drivers are just an additional benefit to have the latest and greatest available and not suffer from 6 months old stuff with lots of already fixed problems.

 

But you have to understand how JTR for PLi came to life - we had arguments if the new drivers and kernel would work stable with the current enigma2 from OpenPLi, if ubifs would be really faster and stable the current lzo compressed jffs2,... and instead of continuing disputing on these matters I simply decided to allow everybody to try out and test himself.

 

There were 3 threads within 2 days where people failed to try the new drivers because they didn't understand the kernel driver dependency, and as I wrote in all these threads how it needs to be properly done and nobody took my advice I created a Plugin which can do the entire job for all boxes just with a few remote clicks.

 

And so far nobody complained that his box is slower or crashes 2x a day ...

 

So the technical discussion should be now settled, and only the doing remains.

Which is not on my agenda, as I don't do any OE development or check in anything to a public git.

 

So I'm sorry if I can't help you on this issue any further,

 

PS: Technically I could change JTR so that it kicks out the ssl, drivers and kernel from the PLi Images on a permanent basis and updates these parts from the DMM Feed so that these images would be updateable as normal, but this would be inpolite to both involved parties and it would be nonsense if it is only 1h of work to do it properly. On the other hand if PLi is not giving the Dreamboxes any new driver and kernel either then there should be no problem wich such ugly converted images either, except that you have to be carefull when adding drivers at a later stage from the PLi feed, because they will maybe not match the newer kernel.

 

gutemine

Edited by gutemine, 16 April 2013 - 08:24.

[Dutchdude]

@ Gutemine,

 

Is it possible to add this "hold" option in JTR Se ?

Or is there a possibility to enable this on a installed image?

OpenPli 3 runs better than ever on my 8000

 

The only remaining thing I could offer is to add also the driver, ssl and kernel hold option to the Special edition as the original JTR has, which allows you to do Softwareupgrades without overwriting the things that JTR replaced.

[Lost in Space]

Well it is just changing in opkg status file the package status flag from installed to hold then it will be ignored on updates.

kernel image dreambox secondstage bootlogo  and dvb modules are normaly the most important packages.

If you really need this you can do this with the opkg command yourself after booting the converted image:

 

opkg flag flag=hold kernel-image

opkg flag flag=hold dreambox-bootlogo
opkg flag flag=hold dreambox-secondstage
opkg flag flag=hold dreambox-dvb-modules

 

Probably in PLi it will be the same ...
From then on these packages will not be replaced anymore.

Try it out and do an opkg update && opkg upgrade afterwards to see if it works :-)

 

If it is tested and prooven to work I can add it to the Plugin ...

Edited by gutemine, 18 April 2013 - 16:01.

[Dutchdude]

Hi,

 

in Openpli some package names are different.

I've run:

 

opkg flag flag=hold kernel-image

opkg flag flag=hold openpli-bootlogo
opkg flag flag=hold dreambox-secondstage-dm8000
opkg flag flag=hold dreambox-dvb-modules-dm8000

 

What is the flag option to "unhold"?

Will do an opkg upgrade this weekend during my servicewindow

 

thx

[Lost in Space]

Thanks for checking and posting the correct commands.

 

opkg help tells you all possible flags:

 

        <flag>=hold|noprune|user|ok|installed|unpacked (one per invocation)
 

There is no unhold - installed is the normal status as I already told you in my previous post.

 

So opkg flag flag=installed packagename should undo the holding.

 

But as I already said try it out and see if it works.

 

Original JTR Plugin for converting non-Dreambox images uses more or less the same trick to keep them running so it should do the job. It is not perfect as you still could get problems if installing kernel modules from feed after the conversion, but most of the time they will simply not load. So it is always good advice to install everything needed in the DMM Image before it does the /lib/module donation.

Edited by gutemine, 18 April 2013 - 18:21.