Menu
▾
▴
simpad-general — This list is to discuss general problems of linux with SIMpad
You can subscribe to this list here.
| 2003 |
Jan
(27) |
Feb
(16) |
Mar
(107) |
Apr
(67) |
May
(227) |
Jun
(174) |
Jul
(128) |
Aug
(68) |
Sep
(64) |
Oct
(25) |
Nov
(78) |
Dec
(18) |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 2004 |
Jan
(24) |
Feb
(26) |
Mar
(20) |
Apr
(26) |
May
(22) |
Jun
(11) |
Jul
(31) |
Aug
(11) |
Sep
(69) |
Oct
(5) |
Nov
(11) |
Dec
(16) |
| 2005 |
Jan
(10) |
Feb
(6) |
Mar
(2) |
Apr
(19) |
May
(4) |
Jun
(18) |
Jul
|
Aug
(3) |
Sep
(8) |
Oct
(15) |
Nov
(7) |
Dec
|
| 2006 |
Jan
|
Feb
|
Mar
(2) |
Apr
(3) |
May
|
Jun
|
Jul
|
Aug
(1) |
Sep
(1) |
Oct
|
Nov
|
Dec
|
| 2007 |
Jan
|
Feb
|
Mar
|
Apr
|
May
(1) |
Jun
|
Jul
|
Aug
|
Sep
|
Oct
|
Nov
|
Dec
|
|
From: Juergen M. <jue...@fr...> - 2003-03-14 10:01:45
|
This are old debug printk from mq200.c which can be removed. I will test the config with consol output. Best regards Juergen Chris Martin wrote: >Hi, > >I don't know if this is a hint in the right direction or not. But I noticed >in minicom that at suspend the following is written to the console: > >suspend3 >done > >Could it be possible, that the kernel was trying to output this at a time >were the display was no longer present and was hanging there? Maybe, it has >something to do with the fact that /dev was mounted as a loop device and the >touchscreen device was being jerked out!? > >I have no idea. But it would be nice to have the boot messages on the >screen. Especially for those booting OPIE for the first time, since this takes a >bit longer (jffs partition activation, /etc /dev image activation, ssh key >generation etc.) and they may think that nothing is happening and turn the device >off. That would not de good. > >Chris > > > >>Thats really interesting. I will test this behaviour. >>That was a real hard bug. >> >>Good job. >> >>Juergen >> >> > > > |
|
From: Chris M. <chr...@gm...> - 2003-03-14 08:49:43
|
Hi, I don't know if this is a hint in the right direction or not. But I noticed in minicom that at suspend the following is written to the console: suspend3 done Could it be possible, that the kernel was trying to output this at a time were the display was no longer present and was hanging there? Maybe, it has something to do with the fact that /dev was mounted as a loop device and the touchscreen device was being jerked out!? I have no idea. But it would be nice to have the boot messages on the screen. Especially for those booting OPIE for the first time, since this takes a bit longer (jffs partition activation, /etc /dev image activation, ssh key generation etc.) and they may think that nothing is happening and turn the device off. That would not de good. Chris > Thats really interesting. I will test this behaviour. > That was a real hard bug. > > Good job. > > Juergen -- +++ GMX - Mail, Messaging & more http://www.gmx.net +++ Bitte lächeln! Fotogalerie online mit GMX ohne eigene Homepage! |
|
From: Juergen M. <jue...@fr...> - 2003-03-14 01:48:32
|
Thats really interesting. I will test this behaviour. That was a real hard bug. Good job. Juergen Chris Martin wrote: >Hi guys, > >I found the problem with the power switch. I can't make any sense of why >this was the problem, but it was solved after setting the console to sa1100 >serial. I no longer have boot messages, but I can turn the simpad off AND on with >the power button now! >Whew, that was a battle!!! > >Peter, thanks for all your help. You don't need to try your kernel without >nfs root for me. It was your kernel config that gave some clues to where I >needed to troubleshoot. > >I will put up the kernel image as a 'SP1' for the current opie image. Just >in case there are some people out there just wanting to try the images and not >build their own kernel. > >Now I can get to the other important topics. > >A very happy, >Chris > > > |
|
From: Chris M. <chr...@gm...> - 2003-03-13 20:58:54
|
Hi guys, I found the problem with the power switch. I can't make any sense of why this was the problem, but it was solved after setting the console to sa1100 serial. I no longer have boot messages, but I can turn the simpad off AND on with the power button now! Whew, that was a battle!!! Peter, thanks for all your help. You don't need to try your kernel without nfs root for me. It was your kernel config that gave some clues to where I needed to troubleshoot. I will put up the kernel image as a 'SP1' for the current opie image. Just in case there are some people out there just wanting to try the images and not build their own kernel. Now I can get to the other important topics. A very happy, Chris -- +++ GMX - Mail, Messaging & more http://www.gmx.net +++ Bitte lächeln! Fotogalerie online mit GMX ohne eigene Homepage! |
|
From: Chris M. <chr...@gm...> - 2003-03-13 19:44:30
|
Hi, Well, I have tried almost everything (standard simpad.conf, Peters conf without nfs root). I can't find the problem :-( The only thing I have not done was Peters console configuration. I will try that next. But, in the mean time could you do me a favor Peter? Could you load a kernel without nfs root and see if the problem is reproducible? I'm about to go mad!! Peter are you sure you haven't changed any source in your kernel? I remember you mentioning that you found an ouch in the switch code. Did you change something to fix that? Thankful for any tips, Chris P.S. About my problems with mounting loop devices in the past dealt with Busybox. I replaced the Busybox version of mount with the original binary and everything was fine afterward. And my consumptions about initrd were most likely imaginary. I have taken initrd out of my kernel with no further problems. -- +++ GMX - Mail, Messaging & more http://www.gmx.net +++ Bitte lächeln! Fotogalerie online mit GMX ohne eigene Homepage! |
|
From: Peter P. <Pet...@em...> - 2003-03-12 18:21:21
|
On Wed, 12 Mar 2003 13:39:09 +0100 (MET) Chris Martin <chr...@gm...> wrote: > I was suprised myself, but in my first attempts to get my own kernel running > on my device, it only booted correctly after including "initrd". I haven't > been able to make sense of it. I'll try a kernel without initrd again and see > what happens. What does 'not booted correctly' mean? It is hard to debug something based on that statement. > No, special reason. I was having problems with "mount -o loop" (by the way > Messi seems to have the same problem at the time) and this was one of my > "debugging" attempts. I should compile it back into the kernel to save space on > the CRAMFS. I could then take out the "insmod" for loop out of my > jffs2-partition.sh. Well, it runs here again with the compiled on module. Guess I am more lucky than you. ;) What were the probelms exactly? > By the way, I have changed my handling of the jffs2 partition. I will be > mounting it to /mnt/jffs instead of /usr/local to stay conform to the X11 image. > I just have a symlink from /usr/local to /mnt/jffs so I can keep my paths on > the opie image. This would at least avoid erasing the jffs2-partition after > installing the opie image even if the x11-image was used previously. I will > keep the possibility to force a re-creation of jffs by way of "touch > /usr/local/erase_jffs2" or later also "touch /mnt/jffs/erase_jffs2". > Another change is to link /usr/lib/ipkg to /usr/local/lib/ipkg, because ipkg > has to be able to write to its status files. The same will have to be done > with all "dest" s defined in the /etc/ipkg.conf. These changes will be in the > next opie-image release ;-) Actually I would prefer to have /lib/modules on the jffs. It is not so much fun to create a new cramfs if you want to debug/add kernel modules (which I have to do to get my prism2-based wlan-crad to work). -Peter |
|
From: Juergen M. <jue...@fr...> - 2003-03-12 18:16:28
|
Chris Martin wrote: >Me again, > >I know I'm probably getting on your nerves, but I may have found something. >One difference between my kernel config and the original simpad.conf is the >fact that I have ext2 compiled as a module and the original has it compiled >in the kernel. Suspicion tells me that this could be the problem, because I >have /etc and /dev mounted as loop devices and these are formatted as ext2. So, >my question, could it be that the suspend is not finishing (like Juergen had >mentioned) because the ext2 module is having a problem. This would explain >why someone elses kernel is working, because their ext2 support is compiled >into the kernel and not as a module. >This may be a wild theory, but maybe one of you guys are more knowledgable >about kernel modules than me ;-) > >With kind regards ;-) >Chris > > > Try it without as module. |
|
From: Juergen M. <jue...@fr...> - 2003-03-12 18:07:55
|
Oliver Hohl wrote:
>Hello,
>
>currently I am trying to bring together a selfmade kernel and an existin=
g cram
>image from your project site. The things I am using so far are the follo=
wing:
>
>* I use the kernel sources linux-2.4.18-rmk7-jpm3 which produce a runnab=
le=20
>kernel for our t-sinus pad.=20
>* I downloaded the cram file cram-2.4.18-rmk7-jpm3b from your project si=
te.
>* I am using the simpadtools-1 also from your project site.
>
>Now the thing I want to do is using our self compiled kernel with this c=
ram
>disk, which should work if i understood everything right from the docume=
ntation
>in the simpadtool directory.
>Since you write there that I should use the flashimage script to combine=
the
>kernel file and the cram file, i tried to use that script. It combines b=
oth
>files, and when i transfer it via tftp it also writes the new file into =
flash.
>But when I try to boot from flash, the kernel does not start but the sim=
pad
>reboots directly. The following output is from the simpad boot sequence:=
>
>-> f =
Boothdr at 0x04080000 -> 0x31415926 =
Loading executable with header 0xe1a00000 =
[*********************\] 19512kB/sec =
Done =
=
=
PBL V1.4 20.07.2001 =
Booting... =
=3D=3D=3D=
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D =
Siemens Switzerland Bootloader V1.2 01.03.01 =
Running @206Mhz Processor ID 0x6901b118 =
Press ESC for monitor !
> Occupied Memory: [0x=
00004000-0x00008000],[0x01f00000-0x01f74af8] Flash Id 0x00=
000018 at virt adr 0x04000000 Size 0x01000000 No 32B=
it Flash detected 0x30003000 =
FlashImage Checksum 0x04040014-0x04061a20=3D0x5f2dd29f Bootloader Image =
OK Fabdata Flash @0x04061a24 Ram @0x01f21a10 FABDATA: C421-CZBGE =
MSC2=3D0x211154e0 :: MQ200 - Revision ID=3D2 ... =
Initialise PCMCIA Interface... done. =
Codec UCB1200 ID:0x1005 Ver.5 Dev.0 Su=
ppl.1 IO 0x00000000 PCMCIA Card Attached... Man-ID:=
0x140 Ositech Trumpcard:Four o=
f Diamonds Ethernet 400022-004 A14D02003E97 SMC CHIP ID/REVIS=
ION 0x3341 @ 0x0800 SMC9194: S=
MC91C94(r:1) at 0x300 IRQ:0 INTF:TP MEM:4608b A!
> DDR: 00:c0:e3:00:3e:97
>
>Searching for server...
>
>
> =20
>
It seems to me that you didn't set /etc/bootptab correctly. It sould=20
look like this:
=2Ekernel:\
:sm=3D255.255.255.0:\
:ht=3Dethernet:\
:bf=3Dtestkernel:
simpad:tc=3D.kernel:ha=3D00c0e3290097:ip=3D192.168.0.100:
Best regards
Juergen
|
|
From: Chris M. <chr...@gm...> - 2003-03-12 16:47:12
|
Me again, I know I'm probably getting on your nerves, but I may have found something. One difference between my kernel config and the original simpad.conf is the fact that I have ext2 compiled as a module and the original has it compiled in the kernel. Suspicion tells me that this could be the problem, because I have /etc and /dev mounted as loop devices and these are formatted as ext2. So, my question, could it be that the suspend is not finishing (like Juergen had mentioned) because the ext2 module is having a problem. This would explain why someone elses kernel is working, because their ext2 support is compiled into the kernel and not as a module. This may be a wild theory, but maybe one of you guys are more knowledgable about kernel modules than me ;-) With kind regards ;-) Chris -- +++ GMX - Mail, Messaging & more http://www.gmx.net +++ Bitte lächeln! Fotogalerie online mit GMX ohne eigene Homepage! |
|
From: Chris M. <chr...@gm...> - 2003-03-12 16:33:04
|
Hi Peter, o.k. I took out initrd and put loop in the kernel. Device boots fine, but still the same problem with the power switch. Could you send me your kernel config. I did a diff between my config and the original simpad.conf. It looks like this: diff simpad.conf .config 432c432 < # CONFIG_AIRO_CS is not set --- > CONFIG_AIRO_CS=m 462c462,466 < # CONFIG_NET_PCMCIA_RADIO is not set --- > CONFIG_NET_PCMCIA_RADIO=y > # CONFIG_PCMCIA_RAYCS is not set > # CONFIG_PCMCIA_NETWAVE is not set > CONFIG_PCMCIA_WAVELAN=m > # CONFIG_AIRONET4500_CS is not set 688c692 < # CONFIG_REISERFS_FS is not set --- > CONFIG_REISERFS_FS=m 696,697c700,701 < # CONFIG_EXT3_FS is not set < # CONFIG_JBD is not set --- > CONFIG_EXT3_FS=m > CONFIG_JBD=m 704c708 < CONFIG_JFFS_FS=m --- > CONFIG_JFFS_FS=y 728c732 < CONFIG_EXT2_FS=y --- > CONFIG_EXT2_FS=m I can't see anything suspicious in there! It shouldn't effect the switches if wireless, ext3, reiserfs are installed (whether as a module or not). And it should not be of any effect if jffs and ext2 are compiled as modules or not. Or do I see this wrong? Please help. This is a big annoyance not being able to turn the device back on after suspend. It's even more annoying when you know that someone else has my same CRAMFS working with just another kernel arghh!! Regards, Chris -- +++ GMX - Mail, Messaging & more http://www.gmx.net +++ Bitte lächeln! Fotogalerie online mit GMX ohne eigene Homepage! |
|
From: Chris M. <chr...@gm...> - 2003-03-12 15:48:07
|
Hi, I may have an idea why my switches don't work. Do I have to apply the patch-2.4.18-rmk7-jpm3 to the linux-2.4.18-rmk7-jpm3 from sourceforge or is it already applied? I never applied this patch since I took for granted that this was already applied and should only be applied to a standard kernel from arm-linux.org. Regards, Chris -- +++ GMX - Mail, Messaging & more http://www.gmx.net +++ Bitte lächeln! Fotogalerie online mit GMX ohne eigene Homepage! |
|
From: Chris M. <chr...@gm...> - 2003-03-12 15:25:37
|
Hi,
I've done this many times, so I'm sure it works for me.
Here's my directory structure:
/home/simpad
/home/simpad/simpadtools
/home/simpad/kernel
/home/simpad/rootdisk
Build the kernel (make dep clean zImage modules)
install the modules into rootdisk (make modules_install)
[make sure INSTALL_MOD_PATH is set to your rootdisk path]
from /home/simpad run:
./simpadtools/mkcramfs rootdisk cram (you're left with a new CRAMFS image
named 'cram' including your modules)
./simpadtools/flashimage (you're left with a boot image combined of zImage
and CRAMFS)
If you later only need to install a new kernel (without new modules) just
run ./simpadtools/kernelflash
(you're then left with the boot sector)
But, it looks to me you're problem is the fact that the presence of the Four
of Diamonds network card is forcing the bootloader to try and load a new
image over the net. Try taking the card out of the slot before reboot.
Regards,
Chris Martin
> Hello,
>
> currently I am trying to bring together a selfmade kernel and an existing
> cram
> image from your project site. The things I am using so far are the
> following:
>
> * I use the kernel sources linux-2.4.18-rmk7-jpm3 which produce a runnable
>
> kernel for our t-sinus pad.
> * I downloaded the cram file cram-2.4.18-rmk7-jpm3b from your project
> site.
> * I am using the simpadtools-1 also from your project site.
>
> Now the thing I want to do is using our self compiled kernel with this
> cram
> disk, which should work if i understood everything right from the
> documentation
> in the simpadtool directory.
> Since you write there that I should use the flashimage script to combine
> the
> kernel file and the cram file, i tried to use that script. It combines
> both
> files, and when i transfer it via tftp it also writes the new file into
> flash.
> But when I try to boot from flash, the kernel does not start but the
> simpad
> reboots directly. The following output is from the simpad boot sequence:
>
> -> f
> Boothdr at 0x04080000 -> 0x31415926
> Loading executable with header 0xe1a00000
> [*********************\] 19512kB/sec
> Done
>
>
> PBL V1.4 20.07.2001
> Booting...
>
> ================================================
Siemens Switzerland
> Bootloader V1.2 01.03.01 Running
@206Mhz
> Processor ID 0x6901b118 Press ESC
for
> monitor !
> Occupied Memory:
> [0x00004000-0x00008000],[0x01f00000-0x01f74af8] Flash Id
> 0x00000018 at virt adr 0x04000000 Size 0x01000000 No
32Bit
> Flash detected 0x30003000
> FlashImage Checksum 0x04040014-0x04061a20=0x5f2dd29f Bootloader Image OK
> Fabdata Flash @0x04061a24 Ram @0x01f21a10 FABDATA: C421-CZBGE
> MSC2=0x211154e0 :: MQ200 - Revision ID=2 ...
> Initialise PCMCIA Interface... done.
> Codec UCB1200 ID:0x1005 Ver.5 Dev.0 Suppl.1 IO 0x00000000
> PCMCIA Card Attached... Man-ID: 0x140
> Ositech Trumpcard:Four of Diamonds Ethernet
> 400022-004 A14D02003E97 SMC CHIP ID/REVISION 0x3341 @ 0x0800
> SMC9194: SMC91C94(r:1) at 0x300 IRQ:0
> INTF:TP MEM:4608b A!
> DDR: 00:c0:e3:00:3e:97
>
> Searching for server...
>
>
> Now my question is, is there anything I am doing wrong to combine the
> kernel
> with the cram file, or is this not possible at all, or is there something
> wrong
> in the flashimg script, since it seems to write the cram file somewhere
> over
> the kernel, at least as far is I understand the following part from that
> script
>
> cp $kernel /tmp/$$.tmp || exit 1
> # Compensate 20 bytes
> dd if=/dev/zero bs=1c count=1004 > filler
> cat filler $cram | dd of=/tmp/$$.tmp bs=1k seek=1023
>
> For me it looks like its writing the cram file into the kernel starting at
>
> block 1023, no matter what size the kernel is. Is this correct? and if
> not,
> how can I tell the kernel at what address to find the cram image, if I
> append
> it to the kernel at the real end?
>
> I would really appreciate any help, since I'm stuck at that point once
> again.
>
> Yours,
>
> Oliver Hohl
>
>
>
> -------------------------------------------------------
> This SF.net email is sponsored by:Crypto Challenge is now open!
> Get cracking and register here for some mind boggling fun and
> the chance of winning an Apple iPod:
> http://ads.sourceforge.net/cgi-bin/redirect.pl?thaw0031en
> _______________________________________________
> Simpad-general mailing list
> Sim...@li...
> https://lists.sourceforge.net/lists/listinfo/simpad-general
>
--
+++ GMX - Mail, Messaging & more http://www.gmx.net +++
Bitte lächeln! Fotogalerie online mit GMX ohne eigene Homepage!
|
|
From: Oliver H. <ho...@te...> - 2003-03-12 15:07:55
|
Hello,
currently I am trying to bring together a selfmade kernel and an existing cram
image from your project site. The things I am using so far are the following:
* I use the kernel sources linux-2.4.18-rmk7-jpm3 which produce a runnable
kernel for our t-sinus pad.
* I downloaded the cram file cram-2.4.18-rmk7-jpm3b from your project site.
* I am using the simpadtools-1 also from your project site.
Now the thing I want to do is using our self compiled kernel with this cram
disk, which should work if i understood everything right from the documentation
in the simpadtool directory.
Since you write there that I should use the flashimage script to combine the
kernel file and the cram file, i tried to use that script. It combines both
files, and when i transfer it via tftp it also writes the new file into flash.
But when I try to boot from flash, the kernel does not start but the simpad
reboots directly. The following output is from the simpad boot sequence:
-> f Boothdr at 0x04080000 -> 0x31415926 Loading executable with header 0xe1a00000 [*********************\] 19512kB/sec Done PBL V1.4 20.07.2001 Booting... ================================================ Siemens Switzerland Bootloader V1.2 01.03.01 Running @206Mhz Processor ID 0x6901b118 Press ESC for monitor !
Occupied Memory: [0x00004000-0x00008000],[0x01f00000-0x01f74af8] Flash Id 0x00000018 at virt adr 0x04000000 Size 0x01000000 No 32Bit Flash detected 0x30003000 FlashImage Checksum 0x04040014-0x04061a20=0x5f2dd29f Bootloader Image OK Fabdata Flash @0x04061a24 Ram @0x01f21a10 FABDATA: C421-CZBGE MSC2=0x211154e0 :: MQ200 - Revision ID=2 ... Initialise PCMCIA Interface... done. Codec UCB1200 ID:0x1005 Ver.5 Dev.0 Suppl.1 IO 0x00000000 PCMCIA Card Attached... Man-ID: 0x140 Ositech Trumpcard:Four of Diamonds Ethernet 400022-004 A14D02003E97 SMC CHIP ID/REVISION 0x3341 @ 0x0800 SMC9194: SMC91C94(r:1) at 0x300 IRQ:0 INTF:TP MEM:4608b A!
DDR: 00:c0:e3:00:3e:97
Searching for server...
Now my question is, is there anything I am doing wrong to combine the kernel
with the cram file, or is this not possible at all, or is there something wrong
in the flashimg script, since it seems to write the cram file somewhere over
the kernel, at least as far is I understand the following part from that script
cp $kernel /tmp/$$.tmp || exit 1
# Compensate 20 bytes
dd if=/dev/zero bs=1c count=1004 > filler
cat filler $cram | dd of=/tmp/$$.tmp bs=1k seek=1023
For me it looks like its writing the cram file into the kernel starting at
block 1023, no matter what size the kernel is. Is this correct? and if not,
how can I tell the kernel at what address to find the cram image, if I append
it to the kernel at the real end?
I would really appreciate any help, since I'm stuck at that point once again.
Yours,
Oliver Hohl
|
|
From: Chris M. <chr...@gm...> - 2003-03-12 12:39:18
|
Hi Peter, Nice to know that the switches work (at least the Power switch) for you with my opie image. That definately means that I'm having a kernel problem. Could you do a diff on our kernel configs (mine and yours). Maybe I can see what I'm missing. > I can't such anything strange. Maybe just two questions: > - why initrd? I was suprised myself, but in my first attempts to get my own kernel running on my device, it only booted correctly after including "initrd". I haven't been able to make sense of it. I'll try a kernel without initrd again and see what happens. > - why loop as a module? No, special reason. I was having problems with "mount -o loop" (by the way Messi seems to have the same problem at the time) and this was one of my "debugging" attempts. I should compile it back into the kernel to save space on the CRAMFS. I could then take out the "insmod" for loop out of my jffs2-partition.sh. By the way, I have changed my handling of the jffs2 partition. I will be mounting it to /mnt/jffs instead of /usr/local to stay conform to the X11 image. I just have a symlink from /usr/local to /mnt/jffs so I can keep my paths on the opie image. This would at least avoid erasing the jffs2-partition after installing the opie image even if the x11-image was used previously. I will keep the possibility to force a re-creation of jffs by way of "touch /usr/local/erase_jffs2" or later also "touch /mnt/jffs/erase_jffs2". Another change is to link /usr/lib/ipkg to /usr/local/lib/ipkg, because ipkg has to be able to write to its status files. The same will have to be done with all "dest" s defined in the /etc/ipkg.conf. These changes will be in the next opie-image release ;-) Chris -- +++ GMX - Mail, Messaging & more http://www.gmx.net +++ Bitte lächeln! Fotogalerie online mit GMX ohne eigene Homepage! |
|
From: Juergen M. <jue...@fr...> - 2003-03-12 00:35:27
|
> > >My next questions will probably be: >How is the LED turned on and off? > This is set by the kernel. See linux-2.4.18-rmk7-jpm3/arch/arm/mach-sa1100/leds.c >How do I set the time and date so that they are correct after resume and >reboot? > This are two different things: After suspend/resume the time is still wright. After reboot you have to set the time again, because the cpu has ereased everything also the rtc Under WinCE after reset the controlpanel for setting the clock appears. >When is the bootloader with NE2000 support gonna be released? > I have talked to the guy who have written the original one and he told me that he has already a new one. I still waiting for it. I think he is busy at the moment because of the Cbit > >Thanx for your help! Keep up the good work guys! > >Regards, >Chris Martin > > > -J. |
|
From: Juergen M. <jue...@fr...> - 2003-03-12 00:22:35
|
Hi everybody. There are new version out of the following applications: -dillo-0.7.1 (Webbrowser) -vlc 0.5.1 (videolan client) (The ipaq version works fine on the simpad) I will add it to the next cram-x11. Best regards Juergen |
|
From: Peter P. <Pet...@em...> - 2003-03-11 22:23:26
|
On Tue, 11 Mar 2003 11:07:29 +0100 (MET) Chris Martin <chr...@gm...> wrote: > Sorry, forgot my kernel config file. I can't such anything strange. Maybe just two questions: - why initrd? - why loop as a module? -Peter |
|
From: Peter P. <Pet...@em...> - 2003-03-11 22:15:17
|
On Tue, 11 Mar 2003 11:01:39 +0100 (MET) Chris Martin <chr...@gm...> wrote: > Hi all, > > now that I finally got OPIE working on my SIMpad (Image and CRAMFS at > sourceforge ;-)) I need to concentrate on getting the switches (or at least switch > #1) working properly. Your problem is most probably not the switches. See below. Regarding switches and opie. I found the following in the standard debian-port of opie. I think that might be used to configure switche-support. Still I have no idea how opie interfaces to the switchd. And I did not find the package in your feed-list? [gretel:ppregler!506] apt-cache show op-button-settings-fb Package: op-button-settings-fb Source: opie-settings-fb Version: 0.9.1-0.snapshot.20030303-1 Description: Device-buttons settings dialog For the Opie environment. > My problem is: > With switchd running I can suspend my SIMpad but I can't get it to wake up > with the switch again. Needless to say I have to reset to get it back again. > Oh, yeah another thing - the green LED stays on too, only the system is > suspended and the display turned off. ?! It works well for me with your cram (but my kernel), i.e. I can suspend and wake it up with the correct button. That is actually not an opie functionality but switchd plus kernel support. And I managed to get it to the state you describe. Looks like a kernel problem. I have also seen an ooops in the opie-application world-clock. [...] > My next questions will probably be: > How is the LED turned on and off? Done automatically by the kernel. As I said there is a more serious case. Please use a serial console and see if you get any ooopses. > How do I set the time and date so that they are correct after resume and > reboot? > When is the bootloader with NE2000 support gonna be released? > > Thanx for your help! Keep up the good work guys! Thank you for the opie. -Peter |
|
From: Chris M. <chr...@gm...> - 2003-03-11 10:07:39
|
Sorry, forgot my kernel config file. --- Weitergeleitete Nachricht / Forwarded Message --- Date: Tue, 11 Mar 2003 11:01:39 +0100 (MET) From: Chris Martin <chr...@gm...> To: sim...@li... Subject: Switches > Hi all, > > now that I finally got OPIE working on my SIMpad (Image and CRAMFS at > sourceforge ;-)) I need to concentrate on getting the switches (or at > least switch > #1) working properly. My problem is: > With switchd running I can suspend my SIMpad but I can't get it to wake up > with the switch again. Needless to say I have to reset to get it back > again. > Oh, yeah another thing - the green LED stays on too, only the system is > suspended and the display turned off. ?! > > I noticed that /etc/init.d/switches looks for "sa1100_switches" in > /proc/misc. Well, my /proc/misc looks like this: > > ### /proc/misc ### > > 63 switches > 14 touchscreen/ucb1x00 > 135 rtc > > > So I changed my /etc/init.d/switches to look like this: > > ### /etc/init.d/switches ### > > #!/bin/sh > > > case "$1" in > start) > > if [ -n "`grep switches /proc/misc`" ] ; then > echo -n "Starting switch daemon: " > mknod /dev/switches c 10 `grep switches /proc/misc | sed -e "s/^ > //g" | cut -d\ -f1` > start-stop-daemon --start --quiet --pidfile /var/run/switchd.pid > --exec /sbin/switchd > echo > fi > > ;; > > stop) > > echo -n "Shutting down switch daemon: " > start-stop-daemon --stop --quiet --oknodo --pidfile > /var/run/switchd.pid > rm /dev/switches > > ;; > > esac > > The original exec line "daemon /sbin/switchd" didn't work for me since > there > is no "daemon" prog in the familiar 0.6.1 dist. So I used > "start-stop-daemon" instead. > But, I think the problem is somewhere else. > > My /etc/switches/1 looks like this: > > ### /etc/switches/1 ### > > #!/bin/sh > > case "$1" in > > pressed) > > echo > /proc/sys/pm/suspend > ;; > > esac > > It's basically the original script minus the "xrefresh", which does not > exist for opie. > > Just in case there's a problem with my kernel setup, I attached my kernel > config. > > My next questions will probably be: > How is the LED turned on and off? > How do I set the time and date so that they are correct after resume and > reboot? > When is the bootloader with NE2000 support gonna be released? > > Thanx for your help! Keep up the good work guys! > > Regards, > Chris Martin > > -- > +++ GMX - Mail, Messaging & more http://www.gmx.net +++ > Bitte lächeln! Fotogalerie online mit GMX ohne eigene Homepage! > > -- +++ GMX - Mail, Messaging & more http://www.gmx.net +++ Bitte lächeln! Fotogalerie online mit GMX ohne eigene Homepage! |
|
From: Chris M. <chr...@gm...> - 2003-03-11 10:01:50
|
Hi all,
now that I finally got OPIE working on my SIMpad (Image and CRAMFS at
sourceforge ;-)) I need to concentrate on getting the switches (or at least switch
#1) working properly. My problem is:
With switchd running I can suspend my SIMpad but I can't get it to wake up
with the switch again. Needless to say I have to reset to get it back again.
Oh, yeah another thing - the green LED stays on too, only the system is
suspended and the display turned off. ?!
I noticed that /etc/init.d/switches looks for "sa1100_switches" in
/proc/misc. Well, my /proc/misc looks like this:
### /proc/misc ###
63 switches
14 touchscreen/ucb1x00
135 rtc
So I changed my /etc/init.d/switches to look like this:
### /etc/init.d/switches ###
#!/bin/sh
case "$1" in
start)
if [ -n "`grep switches /proc/misc`" ] ; then
echo -n "Starting switch daemon: "
mknod /dev/switches c 10 `grep switches /proc/misc | sed -e "s/^
//g" | cut -d\ -f1`
start-stop-daemon --start --quiet --pidfile /var/run/switchd.pid
--exec /sbin/switchd
echo
fi
;;
stop)
echo -n "Shutting down switch daemon: "
start-stop-daemon --stop --quiet --oknodo --pidfile /var/run/switchd.pid
rm /dev/switches
;;
esac
The original exec line "daemon /sbin/switchd" didn't work for me since there
is no "daemon" prog in the familiar 0.6.1 dist. So I used
"start-stop-daemon" instead.
But, I think the problem is somewhere else.
My /etc/switches/1 looks like this:
### /etc/switches/1 ###
#!/bin/sh
case "$1" in
pressed)
echo > /proc/sys/pm/suspend
;;
esac
It's basically the original script minus the "xrefresh", which does not
exist for opie.
Just in case there's a problem with my kernel setup, I attached my kernel
config.
My next questions will probably be:
How is the LED turned on and off?
How do I set the time and date so that they are correct after resume and
reboot?
When is the bootloader with NE2000 support gonna be released?
Thanx for your help! Keep up the good work guys!
Regards,
Chris Martin
--
+++ GMX - Mail, Messaging & more http://www.gmx.net +++
Bitte lächeln! Fotogalerie online mit GMX ohne eigene Homepage!
|
|
From: Peter P. <Pet...@em...> - 2003-03-01 19:13:37
|
On Thu, 27 Feb 2003 07:13:17 +0100 jue...@fr... wrote: > The wake-up problem of the pcmcia is well known. > The only solution by now is to make it as module. I am a bit confused now. I told Walter that the sleep code has still some problem with usbnet (at that time my guess was DMA) and suggested that pcmcia might need some pm-callback. He told me that his NE2000 pcmcia works well after a sleep. What is the 'well-known' wake-up problem of the pmcia bridge? -Peter |
|
From: <jue...@fr...> - 2003-02-27 06:24:37
|
The wake-up problem of the pcmcia is well known. The only solution by now is to make it as module. -J. >-- Original Message -- >From: Peter Pregler <Pet...@em...> >To: SIMpad list <sim...@li...> >Subject: Re: [Simpad-general] oops in switches code :( >Date: Wed, 26 Feb 2003 20:35:12 +0100 > > >On Wed, 26 Feb 2003 11:16:47 +0100 >Albert Huitsing <al...@hu...> wrote: > >> >> I've a Xircom ethernet card... works fine until after wake-up of the simpad >:-( >> and a remove and re-insert doesn't help (very strange), a reboot *does= * >help >> I didn't have much time to investigate it further... > >Thanx for that info. As I said I saw the same with usbnet. There a >restart of the driver does also not help. I need a reboot too. But I >thought the problem is related to the sleep-code of the usbnet driver >since I patched and debugged that stuff and it is still rather >experimental. Seems the problem is somewhere else more to the core of >the system. Might be a timing issue somewhere, I do not get any packet >loss. But only long delays in waves. > >-Peter > > >------------------------------------------------------- >This SF.net email is sponsored by: Scholarships for Techies! >Can't afford IT training? All 2003 ictp students receive scholarships. >Get hands-on training in Microsoft, Cisco, Sun, Linux/UNIX, and more. >www.ictp.com/training/sourceforge.asp >_______________________________________________ >Simpad-general mailing list >Sim...@li... >https://lists.sourceforge.net/lists/listinfo/simpad-general sunrise premiumsurf - geben Sie sich nicht mit weniger zufrieden http://internet.sunrise.ch/de/internet/int_sps.asp |
|
From: Peter P. <Pet...@em...> - 2003-02-26 19:35:56
|
On Wed, 26 Feb 2003 11:16:47 +0100 Albert Huitsing <al...@hu...> wrote: > > I've a Xircom ethernet card... works fine until after wake-up of the simpad :-( > and a remove and re-insert doesn't help (very strange), a reboot *does* help > I didn't have much time to investigate it further... Thanx for that info. As I said I saw the same with usbnet. There a restart of the driver does also not help. I need a reboot too. But I thought the problem is related to the sleep-code of the usbnet driver since I patched and debugged that stuff and it is still rather experimental. Seems the problem is somewhere else more to the core of the system. Might be a timing issue somewhere, I do not get any packet loss. But only long delays in waves. -Peter |
|
From: <jue...@fr...> - 2003-02-25 19:09:25
|
Hi Peter, when I look at your ksymoops output then I think the problem can be in the ucb-switch driver a problem with irq. I have to check it. But at the moment my notebook is out of order. I hope it will not takes= to long to let it fix. I will check it as soon as possible. Best regards Juergen >-- Original Message -- >From: Peter Pregler <Pet...@em...> >To: SIMpad list <sim...@li...> >Subject: [Simpad-general] oops in switches code :( >Date: Sun, 23 Feb 2003 23:49:48 +0100 > > >Hi, > >the ooops in the switches code is a bit nastier than I thought. Only >nice thing is that the trace is always the same. I can unreliable >produce it by pressing random switches a bit faster. But it is not a >definite sequence of keypresses. Seems to be race if I put some load on >the system. :( > >The code in switches_ucb1x00_handler and switches_event seems to be >okay. Also the last/this values are okay after the invert-patch I sent >erlier to account for the different wiring of the switches. I have no >idea why the crash occurs in the kmalloc. Looks like something points to= >the void in some other part and kills the malloc-stuff as a side-effect.= >Strange thing is that it always affects the same routines. So it >should/could be related to the switches code. Anyone has a better eye >than me? And could someone try to reproduce this with a non-nfs root. >Note, you need to set the console to serial (should be default for the >current development kernels in future anyway!) to see the ooops. > >-Peter. > >-- >N'oubliez pas de me rendre mon crayon. >------------------------------- >Email: Pet...@em... > >Attachment: ooops-1.txt > sunrise premiumsurf - geben Sie sich nicht mit weniger zufrieden http://internet.sunrise.ch/de/internet/int_sps.asp |
|
From: <jue...@fr...> - 2003-02-25 19:01:17
|
It is normally /dev/ts major number 10 minor number 14 When you make a ls -l of /dev(touchscreen then you will realize that /dev= ts and /dev/touchscreen/ucb1x00 have the same number. //dev/fb0 is for the display. /dev/H3600_ts was for the iPaq and is not used by us. cat /proc/misc is also usefull Best regards Juergen >-- Original Message -- >From: Chris Martin <chr...@gm...> >To: sim...@li... >Subject: [Simpad-general] Frambuffer device >Date: Mon, 24 Feb 2003 14:12:07 +0100 (MET) > > >Hi, > >which device is being used for the touchscreen? With the iPAQ it is H360= 0_ts >and H3600_tsraw. Which devices are bieng used on the SIMpad? >/dev/touchscreen/ucb1x00 or /dev/fb0 or /dev/H3600_ts or /dev/ts and /de= v/tsraw >etc.... All >these devices are under /dev, but I don't know which is really being use= d. >I >would like to get OPIE working on the SIMpad. I get a calibration screen= , >but >the screen isn't accepting input. I think it's a device thing. > >Thanx, >Chris > >-- >+++ GMX - Mail, Messaging & more http://www.gmx.net +++ >Bitte l=E4cheln! Fotogalerie online mit GMX ohne eigene Homepage! > > > >------------------------------------------------------- >This sf.net email is sponsored by:ThinkGeek >Welcome to geek heaven. >http://thinkgeek.com/sf >_______________________________________________ >Simpad-general mailing list >Sim...@li... >https://lists.sourceforge.net/lists/listinfo/simpad-general sunrise premiumsurf - geben Sie sich nicht mit weniger zufrieden http://internet.sunrise.ch/de/internet/int_sps.asp |
7 messages has been excluded from this view by a project administrator.