Pdaxrom For C3000 Beta2

[desertrat]

hum... i have a problem with one lib  i have this error:

error while loading shared libraries: libstdc++.so.5: cannot open shared object file: No such file or directory.

i found this post: https://www.oesf.org/forums/index.php?showtopic=15693 (pinned)

but the ipk doesn't work and libstdc is already installed... what can i do?

After you have downloaded the attachment from that link you need to rename it to:

libstdcxx5-compat_5.0.7_armv5tel.ipk

(note the first underscore has become a hyphen)

Then you can install as normal.


-- cheers

[iamasmith]

FYI: If you are experiencing problems with suspend not happening on the hdd then take note.

The FIRST thing to be put out of service by the kernel during a suspend cycle is the display, it doesn't necessarily follow that if your screen is blank that your machine is in full suspend.

There is a telltale action that occurs upon full suspend if your Z if plugged in and charging, that is the charging light will flicker off and back on when the Z hits full suspend.

I have a theory about the suspend issue but I need to talk to ThC about it, since making my initial comments about syslogd I have observed the issue with other cards plugged in and usb devices connected. - I just built myself a new Kernel and these problems are gone (I think) - I do need to discuss it with ThC though, he may already have these options set.

- Andy
[div align=\"right\"][a href=\"index.php?act=findpost&pid=110658\"][{POST_SNAPBACK}][/a][/div]

Arrgh, I'm getting the error again - even with the patches and config options set , ok, back to the Kernel source then  - man I'm going to get this fixed even if I have to write a lock on suspend so that the Z won't suspend until Slot: 1 is suspended.

[pgas]

After you have downloaded the attachment from that link you need to rename it to:

libstdcxx5-compat_5.0.7_armv5tel.ipk

(note the first underscore has become a hyphen)

I have edited the post,
Thanks desertrat, I can't believe that nobody spotted this one!.

As a general rule, the forum replace - by _ in the name of the attachements, so for all packages downloaded from this forum, you should replace the _ by - in the package name (before the version number)

[pyknite]

hum... i have a problem with one lib  i have this error:

error while loading shared libraries: libstdc++.so.5: cannot open shared object file: No such file or directory.

i found this post: https://www.oesf.org/forums/index.php?showtopic=15693 (pinned)

but the ipk doesn't work and libstdc is already installed... what can i do?

After you have downloaded the attachment from that link you need to rename it to:

libstdcxx5-compat_5.0.7_armv5tel.ipk

(note the first underscore has become a hyphen)

Then you can install as normal.


-- cheers
[div align=\"right\"][a href=\"index.php?act=findpost&pid=110670\"][{POST_SNAPBACK}][/a][/div]

thx to you and pgas

[cht900]

Hi,

So here it is, thanks to recent help from iamasmith who fixed out my concerns about power management

Greate job~ Nice to see this new version of pdaXrom beta2~  

I have a question... Does beta2 support power management function of CF Wifi? Could I activate wifi power saving mode to extend duration of battery??  

Thanks~~

[Holo]

/usr/bin/usbfunction.py line 21:

Code: [Select]

STORAGE_OPTS = ["/dec/mmdca1", "/dev/hda1"]I guess what the second item of array have wrong value. The right string is "/dec/hdc1" (for CF).

/usb/bin/looknfeel.py lines from 1014 to 1018:

Code: [Select]

 if self.PANEL_LEFT_MARGIN != "" and int(self.PANEL_LEFT_MARGIN) >= 1 and int(self.PANEL_LEFT_MARGIN) <= 600:
    write_str = write_str + " --margin-start \"" + str(self.PANEL_LEFT_MARGIN)+ "\""

  if self.PANEL_RIGHT_MARGIN != "" and int(self.PANEL_RIGHT_MARGIN) >= 1 and int(self.PANEL_RIGHT_MARGIN) <= 600:
    write_str = write_str + " --margin-end \"" + str(self.PANEL_RIGHT_MARGIN)+ "\""If you try to ask the matchbox-panel for help ("matchbox-panel --help") you can see what it havn't "--margin-start" or "--margin-end" options.  So when you set panel margins in "Look-n-Feel" dialog you will be confused with startx script result.  

/etc/fstab line 1:

Code: [Select]

/dev/hda1    /mnt/cf  auto    noauto,owner    0  0It's the same problem as in 'usbfunction.py'. /dev/hda1 is not a compact flash device. I have replaced "hda1" to "hdc1".

That's all, folks!  

[Meanie]

/usr/bin/usbfunction.py line 21:

Code: [Select]

STORAGE_OPTS = ["/dec/mmdca1", "/dev/hda1"]I guess what the second item of array have wrong value. The right string is "/dec/hdc1" (for CF).

/usb/bin/looknfeel.py lines from 1014 to 1018:

Code: [Select]

 if self.PANEL_LEFT_MARGIN != "" and int(self.PANEL_LEFT_MARGIN) >= 1 and int(self.PANEL_LEFT_MARGIN) <= 600:
    write_str = write_str + " --margin-start \"" + str(self.PANEL_LEFT_MARGIN)+ "\""

  if self.PANEL_RIGHT_MARGIN != "" and int(self.PANEL_RIGHT_MARGIN) >= 1 and int(self.PANEL_RIGHT_MARGIN) <= 600:
    write_str = write_str + " --margin-end \"" + str(self.PANEL_RIGHT_MARGIN)+ "\""If you try to ask the matchbox-panel for help ("matchbox-panel --help") you can see what it havn't "--margin-start" or "--margin-end" options.  So when you set panel margins in "Look-n-Feel" dialog you will be confused with startx script result.   

/etc/fstab line 1:

Code: [Select]

/dev/hda1    /mnt/cf  auto    noauto,owner    0  0It's the same problem as in 'usbfunction.py'. /dev/hda1 is not a compact flash device. I have replaced "hda1" to "hdc1".

That's all, folks! 
[div align=\"right\"][a href=\"index.php?act=findpost&pid=110719\"][{POST_SNAPBACK}][/a][/div]

actually, it depends. the external cf can be either /dev/hda or /dev/hdc. this is dynamically mapped depending on how you booted, ie with the cf inserted or not.
the first disk detected is always hda and the second disk hdc
when u boot without a cf inserted, the internal microdrive will be hda and your cf will be hdc
if you boot with the cf inserted, your removable cf will be hda and the internal disk will be hdc
thus hardcoding hda or hdc is not a good idea. the script should check which one is the internal drive dynamically, thus adding an entry in fstab is a bad idea unless you also write a script to dynamically update fstab during bootup.
given that the pivot boot pivots to the internal disk and in order for it to be booted, the cf card has to be removed during boot, but this feature might and probably will change.
i personally think that it is a good idea to allow the pivot to either boot from the internal disk and ignore the inserted cf, or boot off the cf instead for dual booting. this should be a configurable parameter.

[Jondalar]

Hi Guys,

I am playing around with the rom for a little while and I like it nearly better than my last Opie or GPE install ...

What came to my attention - using the Wireless/LAN config tool works untill I press "OKAY" then the tools seems to reset all cardslots, incl. the Harddrive.

I got the feeling it does something like "CARDCTL STOP" on each of them and then wants to do a "CARDCTL START", which doesn't work as the whole system is on the Harddrive (which has been "stopped"). I might be wrong with the correct phrases up there, but I definately need to reboot the system after storing my network config. It is okay for now, as I simply did it once and then things are fine, but it gets quite annoying when I am on a trip next week and for each and every HotSpot I need to reboot.

Anyone Else seeing this ?

Config:

C3000
pdaXrom C3000 beta2
Netgear MA701 W-Lan (Zomax Airrunner, running PRISM-Hostap CS)

Hints/Help is more than welcome,

Cheer - J*

[ThC]

/usr/bin/usbfunction.py line 21:

Code: [Select]

STORAGE_OPTS = ["/dec/mmdca1", "/dev/hda1"]I guess what the second item of array have wrong value. The right string is "/dec/hdc1" (for CF).

/usb/bin/looknfeel.py lines from 1014 to 1018:

Code: [Select]

 if self.PANEL_LEFT_MARGIN != "" and int(self.PANEL_LEFT_MARGIN) >= 1 and int(self.PANEL_LEFT_MARGIN) <= 600:
    write_str = write_str + " --margin-start \"" + str(self.PANEL_LEFT_MARGIN)+ "\""

  if self.PANEL_RIGHT_MARGIN != "" and int(self.PANEL_RIGHT_MARGIN) >= 1 and int(self.PANEL_RIGHT_MARGIN) <= 600:
    write_str = write_str + " --margin-end \"" + str(self.PANEL_RIGHT_MARGIN)+ "\""If you try to ask the matchbox-panel for help ("matchbox-panel --help") you can see what it havn't "--margin-start" or "--margin-end" options.  So when you set panel margins in "Look-n-Feel" dialog you will be confused with startx script result.   

/etc/fstab line 1:

Code: [Select]

/dev/hda1    /mnt/cf  auto    noauto,owner    0  0It's the same problem as in 'usbfunction.py'. /dev/hda1 is not a compact flash device. I have replaced "hda1" to "hdc1".

That's all, folks! 
[div align=\"right\"][a href=\"index.php?act=findpost&pid=110719\"][{POST_SNAPBACK}][/a][/div]

actually, it depends. the external cf can be either /dev/hda or /dev/hdc. this is dynamically mapped depending on how you booted, ie with the cf inserted or not.
the first disk detected is always hda and the second disk hdc
when u boot without a cf inserted, the internal microdrive will be hda and your cf will be hdc
if you boot with the cf inserted, your removable cf will be hda and the internal disk will be hdc
thus hardcoding hda or hdc is not a good idea. the script should check which one is the internal drive dynamically, thus adding an entry in fstab is a bad idea unless you also write a script to dynamically update fstab during bootup.
given that the pivot boot pivots to the internal disk and in order for it to be booted, the cf card has to be removed during boot, but this feature might and probably will change.
i personally think that it is a good idea to allow the pivot to either boot from the internal disk and ignore the inserted cf, or boot off the cf instead for dual booting. this should be a configurable parameter.
[div align=\"right\"][a href=\"index.php?act=findpost&pid=110781\"][{POST_SNAPBACK}][/a][/div]

Actually, as pivot_root init script is not managing external CF slot, CF will be hdc in all cases so he's right and it's my fault forgetting this one

I may do a multi-boot script but I'm not sure of it's utility as sharp/cacko, pdaXrom and OZ uses different libc/kernels, we'll have problems dual-booting to any of these.
A soluion would be to build a "special" OZ version against pdaXrom libc and kernel so we would be able to have a pdaXrom/OZ dual boot but I have to finish my work on the pdaXrom builder before starting to learn using BitBake.
If someone is interrested in helping me and allready know how to use a different libc / kernel in BitBake and have time to do so, please contact me and we'll talk about the possibility of makeing a dual-boot install (and, why not, haveing both on the internal MD, each one having its own 2Gb partition) or a dual-boot image and a special OZ cramfs to put on a CF ...

@Jondalar : Edit /usr/bin/network.py
line 1071 : replace "/usr/sbin/cardctl eject " by "/usr/sbin/cardctl eject 0"
line 1075 : replace "/usr/sbin/cardctl insert " by "/usr/sbin/cardctl insert 0"

[Jondalar]

@Jondalar : Edit /usr/bin/network.py
line 1071 : replace "/usr/sbin/cardctl eject " by "/usr/sbin/cardctl eject 0"
line 1075 : replace "/usr/sbin/cardctl insert " by "/usr/sbin/cardctl insert 0"
[div align=\"right\"][a href=\"index.php?act=findpost&pid=110784\"][{POST_SNAPBACK}][/a][/div]

That did it! Thanks THC!

J*

[fastscirocco]

Is this still in Beta testing or is there a new version coming out?

I really want to switch from the stock rom to PdaXrom, but due to
my "newbie" status I won't be able to debug.

Is this rom ready to be used out of the box?  


Thanks

[ThC]

Hi,
I'm taking some rest right now and working a very little bit on remaking my builder environment as I lost everything recently ... beside, some personal problems I did put aside for a while have to be resolved so it'll take a while untill next release...

btw, this rom is ready to be used out of the box, you'll just have to modify the network.py file as described above as it seem to be the only really annoying bug so far (there are a few others bug which one can live with untill a next release, depending on his needings/capacities) and most of the bugs have been fixed in this topic so you may be able to reproduce fixes on your install

I think the better thing would be to backup your datas and try it by yourself so you can see if you can live with its bugs or if you prefer going back  

Btw I may consider making a fixed image, based on fixes provided in these lines if I get some time to do so, but don't want to make any promise as I'm not sure to have time

[pyknite]

just two question:

1) it's possible to map the cancel button to have the right click? (like the fn key)

2) it's possible to make a right click with a long pression of the stylus on the screen? (i try to configure this in matchbox with the mouse setting, but it disable the left click  )

thx ++

[anonuk]

Is there a version of the dual boot for the 3100? I would love to try it...

Here is the output of sfdisk -l  in case you still wanted it:

Code: [Select]

Disk /dev/hdc: 7936 cylinders, 16 heads, 63 sectors/track
Units = cylinders of 516096 bytes, blocks of 1024 bytes, counting from 0

Device Boot       Start    End    #cyls        #blocks            Id          System
/dev/hdc1               0+  19          20-        10048+           83         Linux
/dev/hdc2               20   39          20         10080             83         Linux
/dev/hdc3               40  7935    7896         3979584          c          Win95 FAT32 (LBA)
/dev/hdc4               0    -               0                 0            0          Empty
Hope that helps, let me know if there is a way I can try this on my 3100....

[ThC]

@anonuk : there is still no "dual boot" available as I'm still only dreaming about it and will have to recompile a whole OZ dist against pdaXrom kernel/libs so they can be dual booted well (well ... I guess I could use an actual snapshot but there would be a lot of non-working things then I guess) ...

btw, the actual pivot_root image *should* work on a 3100 with minor mods in the updater script... so if it's the pivot_root you want to test, let me know and I'll send you an updated updater script (keep in mind I won't be able to test it to any extent as I don't have any 3100 to test with)

ps : I'm also thinking of (and talking with iamasmith about this) makeing a pdaXrom/OpenBSD dualboot if enough people gets some interrest into it ...