Re: More than 7 partitions on the SD

2009-03-10 Thread Andy Green
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

Somebody in the thread at some point said:
| Il giorno lun, 09/03/2009 alle 23.48 +0200, Timo Juhani Lindfors ha
| scritto:
| Pietro \m0nt0\ Montorfano mont...@gmail.com writes:
| Sooo, how it's that possible?? Why can't we simply patch the driver
| which make the mmcblk device to let us create all the devices?
| drivers/mmc/block.c has the limit:
|
| /*
|  * max 8 partitions per card
|  */
| #define MMC_SHIFT   3
| #define MMC_NUM_MINORS  (256  MMC_SHIFT)
|
| Feel free to change but I doubt this will be accepted upstream.
|
| Well this could be a solution, openmoko may compile that (mmcblk) as a
| module (need to verify that) and we or I can provide an opk that

I don't think this will happen in order to scratch this pretty unusual itch.

One way you can solve it for yourself is by rewriting (with dd) the
first sector that contains the partition table and rebooting.  So you
can have two views on to the rest of the storage exposing just the
partitions you want in each view.  That'd get you 14 partitions and the
ability to share eg /home between the two views and so all OSes.

- -Andy
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org

iEYEARECAAYFAkm2D0UACgkQOjLpvpq7dMqzsACfWLdYbmGna8SJCtjbXcZBDymz
nXEAoIWexd4ybdCAL0f1HeE/FhtgQTMK
=hECL
-END PGP SIGNATURE-

___
support mailing list
support@lists.openmoko.org
https://lists.openmoko.org/mailman/listinfo/support


Re: [FSO 5.1] WSOD on resume

2009-03-10 Thread Andy Green
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

Somebody in the thread at some point said:

| I've just flashed Qi, 2.6.28 and FSO 5.1.

Is there a git hash associated with this 2.6.28 kernel at all?

| When the FR resumes, the screen goes white and then slowly fades to black.

- -Andy
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org

iEYEARECAAYFAkm2hiMACgkQOjLpvpq7dMqFjwCeNwqA99ZSNubgdg/X9r0alqnc
rE0AnRr3cSdEFQJHkJgAVaV/QmoaL0IC
=KGWA
-END PGP SIGNATURE-

___
support mailing list
support@lists.openmoko.org
https://lists.openmoko.org/mailman/listinfo/support


Re: SD Card read/write errors and NSOD

2009-03-09 Thread Andy Green
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

Somebody in the thread at some point said:
| I'm experiencing read/write error on my 8GB SD card Kingston C08G
| SDC4/8GB 07 (wiki reports this card shoud work!) with 2.6.28 shr-testing
| kernel and with 2.6.24 from 2008.12, followed some times by NSOD (Noise
| Screen of Dead).
|
| This does not happen when GSM is off.
|
| I found a similiar problem on http://docs.openmoko.org/trac/ticket/2217
| , the difference is that I have the problem with 2.6.24 kernel too! (I
| followed the suggestion to down the glamo clock and the slow memory
| without success)
|
| Before updating the ticket with my details I'd like to know if there is
| kernel version suited to test this better, and have community feedback
| to know it this problem affects ather peoples too.

What does the noise look like, is it also the top-down block of static
random pixels described on the trac

https://docs.openmoko.org/trac/ticket/2217

or jazzy moving pixels, or all over the screen, or in a defined
invalidated area not redrawn or what?

- -Andy
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org

iEYEARECAAYFAkm1MIEACgkQOjLpvpq7dModpgCfQTnW+FJ++9ru7D1dnMzvcuxi
wRMAoIBXzdcdkGWNkR9FqdqaDUTYvgAR
=xNc2
-END PGP SIGNATURE-

___
support mailing list
support@lists.openmoko.org
https://lists.openmoko.org/mailman/listinfo/support


Re: SD Card read/write errors and NSOD

2009-03-09 Thread Andy Green
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

Somebody in the thread at some point said:
| 2009/3/9 Andy Green a...@openmoko.com mailto:a...@openmoko.com
|
| [...]
| What does the noise look like, is it also the top-down block of
static
| random pixels described on the trac
|
| https://docs.openmoko.org/trac/ticket/2217
|
| or jazzy moving pixels, or all over the screen, or in a defined
| invalidated area not redrawn or what?
|
|
| The screen is full of random pixels, but not uniform as in the trac
| ticket snapshot.

Hm that guy had an encrypted FS which was suspected as the source of the
entropy, because it's actually real tough to just happen across good
random-looking data coming out of nowhere.  So this also makes it sound
like SD Card blocks end up in the framebuffer region, in your case
they'll full of more normally distributed file contents maybe.

| Touching the TS sometimes update a random part of the screen with other
| random pixels.
| I'll try to reproduce and take a snapshot asap.

Is the random part of the screen always a linear set of vertical lines?
~ Ie, never a 64 x 64 or such pixel square in its own world but always a
linear block of framebuffer memory?


- -Andy
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org

iEYEARECAAYFAkm1N7MACgkQOjLpvpq7dMp9+gCbBDgaAWgYUPWU5Z01x1vVzoGe
DtoAn1CBJrI2mYOPfu227tF2ai5qXjy8
=wl7+
-END PGP SIGNATURE-

___
support mailing list
support@lists.openmoko.org
https://lists.openmoko.org/mailman/listinfo/support


Re: More than 7 partitions on the SD

2009-03-09 Thread Andy Green
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

Somebody in the thread at some point said:

| Hi, i've bought a 4GB uSD and i've partitioned it to try and test a lot
| of distro, done 10 partitions (1 for kernel and 1 extended for logical
| partitions) booted and only 7 partitions shown, no prob, i've read

| P.S. I'd like to buy an 8GB uSD but i can't live with only 7
| partitions :D

This is a limit in the MMC code in Linux, it doesn't make sense for us
to patch around that.

It's good to try a lot of distros but at any one time, surely only two
or three will be keepers.

- -Andy
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org

iEYEARECAAYFAkm1hzIACgkQOjLpvpq7dMp9RQCgiEvrinkXWKOJuUHgQpOfxqwY
5lUAmwTBewAhIpiHEW01qvScXBxhnpyN
=9Uoq
-END PGP SIGNATURE-

___
support mailing list
support@lists.openmoko.org
https://lists.openmoko.org/mailman/listinfo/support


Re: [debian/fso]: time stuck at suspend after resume

2009-03-03 Thread Andy Green
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

Somebody in the thread at some point said:
|  Does hwclock also show the correct date?
|
| hwclock otoh shows the correct
| time.
|
|
| Which kernel version are you
| running?
|
| 2.6.28-20090105.git69b2aa26
|
| At least the andy-tracking branch suffers from this bug:
|
| i don't think it is the same bug.
| i repeat again:
|
| - hwclock always shows correct time
| - date seems to stop counting when suspending and to start again on
| resume, thus, if suspended 2 hours, date is 2 hours behind hwclock

This RTC resume code was rewritten (since Jan 5th IIRC) by the new
pcf50633 stuff, it's worth trying current andy-tracking or stable.

- -Andy
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org

iEYEARECAAYFAkmtEDcACgkQOjLpvpq7dMqCrQCfb/4RbNTLEgwc199Z5mhnGkQK
r5AAn2e42pcyTg5A2ANdPX6osfJXHmdy
=+8x+
-END PGP SIGNATURE-

___
support mailing list
support@lists.openmoko.org
https://lists.openmoko.org/mailman/listinfo/support


Re: [debian/fso]: time stuck at suspend after resume

2009-03-02 Thread Andy Green
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

Somebody in the thread at some point said:
| On Mon, Mar 02, 2009 at 12:32:51PM +0100, arne anka wrote:
| when i resume the fr, the time zhone shows (and the date command) is
the
| one when the fr suspended (ie, suspending the fr at 16:43 and
resuming at
| 20:14, zhone and date bot show 16:43) -- hwclock otoh shows the correct
| time.
|
|Does hwclock also show the correct date? Which kernel version are you
| running? At least the andy-tracking branch suffers from this bug:
|
https://lists.openmoko.org/pipermail/openmoko-kernel/2009-March/009147.html
|
|If the date and time is wrong (1970-01-01) on powerup without
network or
| GSM connectivity, and if the problem magically disappears in two days
| (2009-03-04), chances are you've run into this bug.)

Thanks for the patch about this, it is on andy-tracking now.

- -Andy


-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org

iEYEARECAAYFAkms04IACgkQOjLpvpq7dMpiVgCgj3KcB+RvfEutKQRj/KM3WGeD
YTEAn3suzQX1MKvDoci+baPEa6avQp1D
=2cUT
-END PGP SIGNATURE-

___
support mailing list
support@lists.openmoko.org
https://lists.openmoko.org/mailman/listinfo/support


Re: Qi causes WSOD (was Re: [2008.testing] WSOD)

2009-02-23 Thread Andy Green
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

Somebody in the thread at some point said:
| 2009/2/22 Jeffrey Ratcliffe jeffrey.ratcli...@gmail.com:
| The only other thing I did around the same time was to replace uBoot
with Qi.
|
| To test this, I booted into NOR, thereby using uBoot, and then booted
| the 2.4.24 image again (which I had reflashed). Resume then works
| without WSOD.
|
| i.e. Qi, for me at least, causes WSOD.
|
| The suspend process seems different, too - with uBoot, the fade to
| black is gradual, but with Qi, it is more or less instantaneous, with
| a click, as if a relay was switching.

Suspend and resume on 2.6.24 only works by luck, it's not telling us much.

But on community list, I see people reporting the same thing with
2.6.28-rc4.  Why are people shipping 2.6.28-rc4 and not 2.6.29-rc2 in
our stable branch?  No idea.

However, Qi works perfectly fine with suspend and resume here on recent
andy-tracking and worked fine ever since the original WSOD fixes.

| I'll get the friend of mine to redo the SD/GPS fix to make sure that
| has nothing to do with it, but I doubt it.

I wouldn't bother, I doubt it has anything to do with it.

- -Andy

-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org

iEYEARECAAYFAkmjTgAACgkQOjLpvpq7dMoATQCfQY7omEOzkfdC/eL0zijzkIjD
GiAAmwRr5XH+oz2z9yBASdrLh8d2EVWI
=Wg2b
-END PGP SIGNATURE-

___
support mailing list
support@lists.openmoko.org
https://lists.openmoko.org/mailman/listinfo/support


Re: Qi causes WSOD (was Re: [2008.testing] WSOD)

2009-02-23 Thread Andy Green
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

Somebody in the thread at some point said:
| On Feb 24, 2009 2:31am, Andy Green a...@openmoko.com wrote:
|   Suspend and resume on 2.6.24 only works by luck, it's not telling us
| much.
|
| OK. You are saying that with uBoot+2.6.24, it wasn't suspending
| properly, hence resume worked, but with Qi+2.6.24, suspend worked, but
| resume is broken.

What I mean is that it's known 2.6.24 kernel stuff is broken for suspend
/ resume in a deep way, if it happens to work it's basically chance.  So
hearing that it doesn't work properly under some conditions don't give a
good clue, since it can break under different timing or interrupts or
whatever happening.  2.6.28+ stuff has this fixed, so misbehaviour on
resume means a smaller set of possibilities and we can start to imagine
what can be happening.

| Why does Qi/uBoot affect this?

I think it's to do with the sequencing of power / init to the LCM.  In
U-Boot, we bring up the Glamo and LCM to show the boot logo.  In Qi, we
do not bring up the Glamo LCD controller nor program the LCM until the
Linux drivers start.

We know that the guts of the LCM are not entirely under control since
the last WSOD business, there's some people reporting monochrome display
after resume as well.  But it varies by physical LCM, I run Qi for
months every build and I don't see any of it on my test GTA02.

| What distro do I need to flash to test this properly (i.e. which diistro
| is using 2.6.28 or 2.6.29)? FSO? I understand that I can't simply flash
| 2.6.29 onto 2008.12 as the userspace file structure is different.
|
| Which kernel is Debian using 2.6.24?

This FSO v5 thing, really anything with 2.6.28 should also work well
with 2.6.29 stuff.

- -Andy
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org

iEYEARECAAYFAkmjkhAACgkQOjLpvpq7dMrDUACfR+WZ2Xfw8r2Xwa4RURhD/n2o
Fs4An0t1vpnXqVj5n6SVqz6LeFZKwv2w
=+82c
-END PGP SIGNATURE-

___
support mailing list
support@lists.openmoko.org
https://lists.openmoko.org/mailman/listinfo/support


Re: [debian/fso] 2.6.28-20090105.git69b2aa26 weird jiffs2 messages in dmesg

2009-02-21 Thread Andy Green
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

Somebody in the thread at some point said:
| It seems that it takes a long long while to mount the jffs2 partition
| you have.  The first thing CPU 0 stall is just a comment that you seem
| to be stuck for two minutes, then the second thing is a diagnostic
| talking about what you are stuck on.  It seems you are stuck on mounting
| the jffs2 partition.
|
| Lastly the jffs2 warning about summaries... IIRC this is to do with how
| the jffs2 filesystem image was composed when it was built there was
| something about slow mount and broken summaries when the jffs2 image was
| created with wrong block size as I recall.
|
| ok ...
| so, what to do about it?

Dunno what you can do about it, but the guy who owns the process for
generating the jffs2 image for Debian may get a hint from this to change
how he generates it.  jffs2 mount should be a few seconds not 120.

Particularly as I recall the parameters for generating GTA01 and GTA02
images differ due to NAND block differences.

- -Andy
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org

iEUEARECAAYFAkmgHSEACgkQOjLpvpq7dMpRvgCeNH/9MURwGqCUfwz4BkXK1yD9
rVAAmMtyHNuXwFOIEJhnWGusE4io46s=
=ctYF
-END PGP SIGNATURE-

___
support mailing list
support@lists.openmoko.org
https://lists.openmoko.org/mailman/listinfo/support


Re: [debian/fso] 2.6.28-20090105.git69b2aa26 weird jiffs2 messages in dmesg

2009-02-21 Thread Andy Green
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

Somebody in the thread at some point said:
| Dunno what you can do about it, but the guy who owns the process for
| generating the jffs2 image for Debian may get a hint from this to change
| how he generates it.  jffs2 mount should be a few seconds not 120.
|
| oh, that's no flashed image or so.
| it is the jiffs the fr came with. months ago i deleted everything and
| copied my files over (the most crucial parts of debian: /sbin/ /lib/
...)
|  from teh sd install.
| with 2.6.24 i have never seen those errors. wouldn't that rather be some
| kernel (2.6.28) related problem?

Maybe, but the actual error here is to do with the format of the jffs2
partition.  It can be broken in kernel but I think we would have heard
about it before, 2.6.28 is old now in kernel terms.

- -Andy
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org

iEYEARECAAYFAkmgI9sACgkQOjLpvpq7dMq8hQCfcEOQ4srllKrAKCOjberbitNe
+8EAnA6NrUqJxfjkWeBUrpqOgHcTXjiN
=l802
-END PGP SIGNATURE-

___
support mailing list
support@lists.openmoko.org
https://lists.openmoko.org/mailman/listinfo/support


Re: [debian/fso] 2.6.28-20090105.git69b2aa26 weird jiffs2 messages in dmesg

2009-02-21 Thread Andy Green
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

Somebody in the thread at some point said:
| Maybe, but the actual error here is to do with the format of the jffs2
| partition.  It can be broken in kernel but I think we would have heard
| about it before, 2.6.28 is old now in kernel terms.
|
| well, so the best i can do is backup my data and recreate/reformat the
| jiffs2 fs in flash -- with which parameters exactly?

What I would try is a precooked jffs2 image with summaries, download it
and then DFU it in via NOR U-Boot.

Eg,

http://downloads.freesmartphone.org/fso-unstable/images/om-gta02/

Actually what I would really do is run from uSD on a normal ext3
filesystem, but that's not to the point :-)

- -Andy
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org

iEYEARECAAYFAkmgnWwACgkQOjLpvpq7dMp7AACfXm7WW43DeYR1zYodGL0yk+xI
Z6kAnRGyGXul0HaXVLy1MoAdid0ZAsvq
=7lbr
-END PGP SIGNATURE-

___
support mailing list
support@lists.openmoko.org
https://lists.openmoko.org/mailman/listinfo/support


Re: [FSO] Milestone 5, uBoot and SD, any success?

2009-02-11 Thread Andy Green
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

Somebody in the thread at some point said:
| On Wed, Feb 11, 2009 at 12:11 AM, Marc Bantle openm...@rcie.de wrote:
|
| Until current rootfs' get fixed, I dropped the 'ro' option in
| the boot options.
|
| Thanks to all of you for you help. But I tried something else : Marek
| Lindner just provided the steps for fixing the image. Here it is:
|
|  Not sure whether somebody was able to properly fix it or not. We
(Chris and me)
|  just ran into this issue and created a fix for the mount script.
|  The /etc/init.d/checkroot.sh file in line 24 is the reason for the
problem. It
|  complains about a wrong file descriptor here: exec 90 /etc/fstab
|  I removed the 2 exec lines (line 24 and line 85) and added a  
/etc/fstab
|  at the end of line 84.
|
| It worked for me.

It'd be great if we can get that change or something like it in the
upstream that these images come from.

- -Andy



-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org

iEYEARECAAYFAkmSpuEACgkQOjLpvpq7dMrjcwCfRjc484sHPqQw0BoCUyL5NkUS
lsAAn0J7duvLyNFWrKAu/HiepQTYE8iD
=Hv7T
-END PGP SIGNATURE-

___
support mailing list
support@lists.openmoko.org
https://lists.openmoko.org/mailman/listinfo/support


Re: [FSO] Milestone 5, uBoot and SD, any success?

2009-02-11 Thread Andy Green
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

Somebody in the thread at some point said:

| Rename the file from checkroot.sh to checkroot, and it achieves the same
| effect (it is run in a shell instead of being sourced).
|
| As I said previously, this was an upstream OE bug which made it's way
| into FSO MS5 and has since been fixed upstream on OE.dev

Great, I'll just tell people to update their rootfs then.

- -Andy
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org

iEYEARECAAYFAkmTpcYACgkQOjLpvpq7dMq6nQCdEQwX0IIuRj259bxXAGush6ku
3pwAn0gZkNznoAVxwV8OIkjPg3jGvuII
=WCx/
-END PGP SIGNATURE-

___
support mailing list
support@lists.openmoko.org
https://lists.openmoko.org/mailman/listinfo/support


Re: qi not working? or is it android?

2009-02-07 Thread Andy Green
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

Somebody in the thread at some point said:
| Hi, all:
|
| Just flashed qi (GTA02), and I get the flash, the vibration, and then this
| message:
|
| s3c2410_udc: debugfs dir creation failed -19
| power_supply bat: driver failed to report `status' properly
|
| trying to boot android (the second message doesn't seem to stop
booting the
| OM, but the first message does)
|
| Any idea what's up? Any way to get rid of qi, which gave me the same
message
| about the power_supply before booting qtopia . . .

You're mistaken, Qi does not issue these messages but the kernel.

Neither of them represent anything fatal so the problem is somewhere
completely different.

- -Andy
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org

iEYEARECAAYFAkmNrZgACgkQOjLpvpq7dMriYACeJCnS1aR9FmjmApxhLGbrPDbJ
j1kAnRNvFm6kxMvVRGcjdT+O2f3U48tT
=+cog
-END PGP SIGNATURE-

___
support mailing list
support@lists.openmoko.org
https://lists.openmoko.org/mailman/listinfo/support


Re: qi not working? or is it android?

2009-02-07 Thread Andy Green
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

Somebody in the thread at some point said:

| Yes, I suspect that the first message (dbugfs dir creation failed) is
| coming from the android kernel and that the power_supply message was
| appearing before qi -- I think I'm just now noticing it because of the
fact
| that I don't see the rest of the scrolling boot . . .
|
| But I'll confirm that I flashed the right qi. Is it a good idea to
just keep
| that as the bootloader?

Yeah, Qi is doing fine loading and booting Linux as it should, there's
nothing to blame it for here.

If you see the backlight come up you're out of Qi and the kernel is
running, unless the issue involves kernel commandline it's most likely a
rootfs or kernel related issue then.  Some rootfs cannot cope with
mounting the thing ro which is Qi policy, maybe that's the problem.

I don't think you have the wrong Qi either, I guess it would not start
at all.

- -Andy
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org

iEYEARECAAYFAkmN5OQACgkQOjLpvpq7dMpIXwCfeHx3sPRF8Bz9KReMYX3zB2J3
5OcAnjQ73Ae7YE6DDFqkcJgWddqPb4yH
=LA9n
-END PGP SIGNATURE-

___
support mailing list
support@lists.openmoko.org
https://lists.openmoko.org/mailman/listinfo/support


Re: [FSO] Wifi in M5 no longer functioning

2009-02-05 Thread Andy Green
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

Somebody in the thread at some point said:

[4.965000] AR6000 connected event on freq 2437 with bssid my AP MAC
address  listenInterval=100, beaconInterval = 100, beaconIeLen = 22
assocReqLen=59 assocRespLen =59
[4.965000] Network: Infrastructure

looks OK

| eth0  AR6000 802.11g  ESSID:my ESSID
|   Mode:Managed  Frequency:2.437 GHz  Access Point: my AP MAC
| address
|   Bit Rate=65.535 Mb/s   Tx-Power=0 dBm   Sensitivity=0/3

Bitrate is broken.

| eth0  Link encap:Ethernet  HWaddr 00:12:CF:8F:33:91
|   inet6 addr: fe80::212:cfff:fe8f:3391/64 Scope:Link
|   UP BROADCAST RUNNING MULTICAST  MTU:1500  Metric:1
|   RX packets:125 errors:0 dropped:0 overruns:0 frame:0
|   TX packets:40 errors:0 dropped:0 overruns:0 carrier:0
|   collisions:0 txqueuelen:1000
|   RX bytes:144316 (140.9 KiB)  TX bytes:4497 (4.3 KiB)
|
| eth0:avah Link encap:Ethernet  HWaddr 00:12:CF:8F:33:91
|   inet addr:169.254.6.255  Bcast:169.254.255.255  Mask:255.255.0.0
|   UP BROADCAST RUNNING MULTICAST  MTU:1500  Metric:1

Wow that's new, avahi secondary network interface.

This has a link-local address only.

| loLink encap:Local Loopback
|   inet addr:127.0.0.1  Mask:255.0.0.0
|   inet6 addr: ::1/128 Scope:Host
|   UP LOOPBACK RUNNING  MTU:16436  Metric:1
|   RX packets:1 errors:0 dropped:0 overruns:0 frame:0
|   TX packets:1 errors:0 dropped:0 overruns:0 carrier:0
|   collisions:0 txqueuelen:0
|   RX bytes:104 (104.0 B)  TX bytes:104 (104.0 B)
|
| pan0  Link encap:Ethernet  HWaddr AA:92:6C:B1:E1:A9
|   BROADCAST MULTICAST  MTU:1500  Metric:1
|   RX packets:0 errors:0 dropped:0 overruns:0 frame:0
|   TX packets:0 errors:0 dropped:0 overruns:0 carrier:0
|   collisions:0 txqueuelen:0
|   RX bytes:0 (0.0 B)  TX bytes:0 (0.0 B)
|
| usb0  Link encap:Ethernet  HWaddr 6E:F1:73:29:F6:4E
|   inet addr:192.168.0.202  Bcast:192.168.0.255  Mask:255.255.255.0
|   inet6 addr: fe80::6cf1:73ff:fe29:f64e/64 Scope:Link
|   UP BROADCAST RUNNING MULTICAST  MTU:1500  Metric:1
|   RX packets:638 errors:0 dropped:0 overruns:0 frame:0
|   TX packets:510 errors:0 dropped:0 overruns:0 carrier:0
|   collisions:0 txqueuelen:1000
|   RX bytes:46204 (45.1 KiB)  TX bytes:154823 (151.1 KiB)
|
|
| It seems to look like it's connected, but I still can't ping google, the
| AP, or access the internet.

usb0 is up too.

But, the missing info is your routing table

route -n

Will the PC you are connected to act as gateway on the USB Ethernet
connection?  Or is the only way out to the Internet through WLAN?

Can you ping even if you don't try to use DNS?

What network range does the AP subnet use... 192.168.0.x again?

- -Andy
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org

iEYEARECAAYFAkmKrMsACgkQOjLpvpq7dMrSxQCfZE1Zs7aIQkBDtSKdbc+AY4iN
dq0AnRDVEFltfudhVKA59eIoXMhDV1pD
=Alfv
-END PGP SIGNATURE-

___
support mailing list
support@lists.openmoko.org
https://lists.openmoko.org/mailman/listinfo/support


Re: [FSO] Wifi in M5 no longer functioning

2009-02-04 Thread Andy Green
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

Somebody in the thread at some point said:

| In my FSO M5-equiped FR eth0 visible by iwconfig and my WiFi points
| successfully scaned by iwlist.
|
| Since the previous Jan 5th basis I understood the kernel had, there are
| relatively few patches.  The main change is WLAN driver is now in modules.
|
| But, seems like I found another unpleasant bug :(

I never met a pleasant one :-)

| I tryed setting up direct WiFi network access (without hardware Access
| Points) for FR by hands (I successfully used such scenario for
| configuring wireless-networking for some OLPCs  traditional notebooks):
|
| # iwconfig eth0 mode ad-hoc
| # iwconfig eth0 essid my-wifi
| # iwconfig eth0
| eth0  AR6000 802.11g  ESSID:my-wifi
|   Mode:Ad-Hoc  Frequency:2.457 GHz  Cell: 4E:72:EC:8E:38:88
|   Bit Rate=1 Mb/s   Tx-Power=15 dBm   Sensitivity=0/3
|   Retry:on
|   Encryption key:off
|   Power Management:off
|   Link Quality:0/94  Signal level:-95 dBm  Noise level:-96 dBm
|   Rx invalid nwid:0  Rx invalid crypt:0  Rx invalid frag:0
|   Tx excessive retries:0  Invalid misc:0   Missed beacon:0
|
| # iwconfig eth0 channel 1
| # iwconfig eth0 key restricted 1234567890
| # iwconfig eth0
| eth0  AR6000 802.11g  ESSID:
|   Mode:Ad-Hoc  Frequency:2.457 GHz  Cell: 4E:72:EC:8E:38:88
|   Bit Rate=1 Mb/s   Tx-Power=15 dBm   Sensitivity=0/3
|   Retry:on
|   Encryption key:1234-5678-90   Security mode:restricted
|   Power Management:off
|   Link Quality:0/94  Signal level:-95 dBm  Noise level:-97 dBm
|   Rx invalid nwid:0  Rx invalid crypt:0  Rx invalid frag:0
|   Tx excessive retries:0  Invalid misc:0   Missed beacon:0
|
| I lost my ESSID at this step and try setting up this one once again:
|
| # iwconfig eth0 essid my-wifi
|
| After that system hangs up and blinking by red LED in AUX button.

Right, panicked.

| Seems like restricted key is key point for this situation -- with open
| key all works good.
|
| Can someone try to reproduce this situation? May be something wrong with
| my hardware...

No it shouldn't panic anyway.

Can you open a Trac bug about this, just cut and paste your description
here is fine.

If we can reproduce it with a debug board it's probably not too hard to
make the panic go away anyway.

Ad-hoc is relatively uncommon so it makes sense it's going to show more
bugs.

- -Andy
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org

iEYEARECAAYFAkmJutYACgkQOjLpvpq7dMpb/gCfZHfm8QsObsaVshBJM8UNosUb
kPwAn2TW7us9Iv5i7Yj3ug8apUFs7/PH
=m9Cl
-END PGP SIGNATURE-

___
support mailing list
support@lists.openmoko.org
https://lists.openmoko.org/mailman/listinfo/support


Re: [FSO] Wifi in M5 no longer functioning

2009-02-04 Thread Andy Green
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

Somebody in the thread at some point said:

| r...@om-gta02:~# ifup eth0
| WPA: Configuring Interface
| ioctl[SIOCSIWENCODEEXT]: Operation not supported
| ioctl[SIOCSIWENCODEEXT]: Operation not supported
| ioctl[SIOCSIWENCODEEXT]: Operation not supported
| ioctl[SIOCSIWENCODEEXT]: Operation not supported
| udhcpc (v1.13.2) started
| run-parts: /etc/udhcpc.d/00avahi-autoipd exited with code 1
| Sending discover...
| Sending discover...
| Sending discover...
| No lease, failing
|
| I don't _think_ it's a configuration issue, as my setup hasn't changed
| since M4 and I'm using the same config files.
|
| I'm afraid the new kernel broke something then. Can you try an older
one to
| make sure?

What's dmesg have to say about all this excitement?

# iwconfig
# iwlist eth0 scan

Since the previous Jan 5th basis I understood the kernel had, there are
relatively few patches.  The main change is WLAN driver is now in modules.

# lsmod

I guess check eth0 is still assigned to WLAN with

# ifconfig -a

as well.

- -Andy



-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org

iEYEARECAAYFAkmJc00ACgkQOjLpvpq7dMpX+gCfTgXFLwakVt0dVmIt1hf1n2yZ
LrkAn1Co5KbBV/xyBFANa8Ts9xT2jYaB
=B4zW
-END PGP SIGNATURE-

___
support mailing list
support@lists.openmoko.org
https://lists.openmoko.org/mailman/listinfo/support


Re: OM2008.12 audio problems during call

2009-02-03 Thread Andy Green
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

Somebody in the thread at some point said:
|
| just restore those .state files from whatever package they come from
| (don't recall) -- I git the same issue with that mixer -- better never
| click 'save' on the first tab ;)
|
| Thank you, Mr. Halchenko. I think this is what has happened.
|
| Just to be sure, the .state files I need to replace are:
| gsmhandset.state
| gsmheadset.state
| gsmspeakerout.state
| correct? Does anyone know where I can find the original 2008.12
versions of
| these?

For a few days sound was broken in the 2.6.29 kernels due to changes in
Alsa there, but it's fixed yesterday.

If it's the problem, alsactl restore -f
some-distributed-and-not-stored-to.state will blow an error on the
commandline about missing mixer element or somesuch.

- -Andy
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org

iEYEARECAAYFAkmIyXsACgkQOjLpvpq7dMojJgCfd6ro0I70uFpRzgYl5QIY9PSU
dogAn0IKmyM32ltpUv+fqawuSq4y7lk7
=zq4h
-END PGP SIGNATURE-

___
support mailing list
support@lists.openmoko.org
https://lists.openmoko.org/mailman/listinfo/support


Re: [Debian] Can't shutdown

2009-02-02 Thread Andy Green
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

Somebody in the thread at some point said:
| 2009/2/1 Jeffrey Ratcliffe jeffrey.ratcli...@gmail.com:
| shutdown now
|
| Figured it out. This should be:
|
| /etc/init.d/halt stop

Or shutdown -h now

- -Andy
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org

iEYEARECAAYFAkmG0B8ACgkQOjLpvpq7dMpZ1ACggjnavovc/fW6V1t3I1ENvbhj
azcAn2x2u4tJHPgc+bO+1cLh/mLBxVYH
=GJza
-END PGP SIGNATURE-

___
support mailing list
support@lists.openmoko.org
https://lists.openmoko.org/mailman/listinfo/support


Re: Qi

2009-01-31 Thread Andy Green
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

Somebody in the thread at some point said:
| 2009/1/31 Andy Green a...@openmoko.com:
| So need to remove what you have in partition 1, and add a kernel named
| /boot/uImage-GTA02.bin into partition 2 (since that's your rootfs),
| which should be ext2 or better ext3.
|
| Thanks. The Debian installer had put the kernel in the root of
| /dev/mmcblk0p1, not in a boot subdirectory. I added a boot
| subdirectory with symlinks to the kernel and now it's all working.
|
| Why is the journalling of ext3 better? Would the kernel also be better

Ext3 is better IMO because it can soak up a lot of casual damage that
ext2 can't.  With my usage pattern I am randomly powering things off all
the time I appreciate how ext3 never makes any difficulty for me about
that.  Whereas after a day of it ext2 is on its knees.

| off in an ext3 fs?

Splitting the kernel off in its own private partition is an aberration
we introduced because ext2 parser in U-Boot was broken for a while.  So
we had to make do with a VFAT partition first that U-Boot could parse,
then an ext2 one after that contained the rootfs.

But it makes trouble for updating the kernel in the case you have
multiple rootfs, and makes it possible to get your kernel and modules
out of sync.  It's an ugly kludge.

So the optimum situation is that the kernel is coming out of the same
filesystem that is the root filesystem, the package update for it is
simply to put in it /boot on current root filesystem which is easiest
for the package system.

It also readily allows different kernels in different rootfs if you are
making an SD with multiple rootfs.

- -Andy
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org

iEYEARECAAYFAkmErpcACgkQOjLpvpq7dMqjgQCdEstuzymmhmo8rMts3BB9cCa6
rAQAmwQI6bnr2ceW952zE8el5zL0dDC+
=mdqv
-END PGP SIGNATURE-

___
support mailing list
support@lists.openmoko.org
https://lists.openmoko.org/mailman/listinfo/support


Re: GTA02 is now an expensive brick / pcf50633 defaults

2009-01-06 Thread Andy Green
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

Somebody in the thread at some point said:
|
| Mike Montour wrote:
| If you program the PCF50633 as:
|   - charger enabled
|   - 100mA current limit
|   - do not wake on USB insertion (avoids the brownout issue)
|
| then it will have the behaviour that you want - even with a
| completely-dead battery, you can plug it in and it will slow-charge the
| battery in hardware. I think it still technically violates one of the
| USB standards by drawing current without talking to the host, but IMHO
| it's unlikely to cause any real-world problems.
|
|
| 1. This really ought to be default behaviour, the current way is really
| *really* bad.
|
| 2. Why stick to 100mA? I thought the USB standard was for ports to support
| 500mA?
|
| 3. I'm sure the FR doesn't need to negotiate with its charger.
|
| 4. How easy is it to make this change?

The issue is more complicated than this.

pcf50633 has more than one level of hardcoded default actions, some
registers get reset when you tell it to enter what it calls standby
but we would more naturally call off, and more get reset again when
the small backup battery that is also present on GTA02 becomes
exhausted, called NOPOWER state.

The real default settings for our variant that occur with NOPOWER
forces the charger disabled.  So if you stick your FR in a drawer, the
main battery is exhausted, the backup battery goes down a day or
whatever later, nothing that we did in previous session about charger
status sticks, the charger is off.  In addition, due to VB_SYS issue, we
can't get easily get to the point of running any code on A5 to enable it
without a loan from the battery to get us started.

Sometimes the battery is in cutoff and does have some power in the cell,
it's just disconnected from the terminals.  That's when any kind of
external kick will get us out of the hole.

There is no violation of USB standards if you only pull 100mA, but
another default set by going off even is that we pull 500mA from the
USB host until the CPU can get started, and that is in violation.  But
since we can do nothing about it, and we didn't find a host that cares
yet, it's OK.

There's two interesting ideas from Mike though, one is that disable USB
insert as ON will help by giving longer for VB_SYS to charge and the
other is leave charger enabled.  For both of these, they are defeated
(USB insert is ON action, charger disabled) by NOPOWER and we have no
control then.  If the backup battery had not decayed, it could help, but
it only has very partial impact and is not magically delivering the
behaviour you want, GTA02 with this issue still will not start if left
long enough for backup battery to fall below the pcf50633 threshold for
NOPOWER.

- -Andy
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org

iEYEARECAAYFAkljdhIACgkQOjLpvpq7dMpvDgCfSK+d1rosyd0cJIrP2dcIvkgz
yD0An1KMoTi57Z2oxRp0/8RrITsoPSF0
=LB9S
-END PGP SIGNATURE-

___
support mailing list
support@lists.openmoko.org
https://lists.openmoko.org/mailman/listinfo/support


Re: Phone wont boot Android

2008-12-20 Thread Andy Green
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

Somebody in the thread at some point said:
| I also tried installing Qi last night, for qi the red aux light blinks
once,
| and I get a black screen on the phone but the back light is on.
|
| with u-boot i just get the same black screen.
| This phone worked with the very first port of android but none of the
| updates.

You're booted into the kernel fine, since with Qi nothing brings up
backlight until then.

In Sean McNeil's kernel configs he has turned off the framebuffer
console completely because it caused flickering apparently.

So I guess you're all set but something is blowing chunks once Android
starts.  Maybe try a different kernel image like the ones in
http://people.openmoko.org/andy that does have a framebuffer console.

- -Andy

-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org

iEYEARECAAYFAklNVwEACgkQOjLpvpq7dMq2BACffhqT72EZ2lMyPwEMBgfG1CA5
SL0AnjYtz3AFkBsFCO4I2N1jn3W8vSlq
=nZHg
-END PGP SIGNATURE-

___
support mailing list
support@lists.openmoko.org
https://lists.openmoko.org/mailman/listinfo/support


Re: [FDOM, 2008.9] USB not working after suspend/resume-cycle [FIXED]

2008-12-18 Thread Andy Green
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

Somebody in the thread at some point said:
| The only thing that helped was removing and reloading the ohci_hcd
module,
| but that obviously only works with a kernel which has this one built
as a
| module.
| Somebody on some list said that doing an lsusb fixed it for him.
I've not
| even tried USB host mode myself.
|
| Brilliant - works like a charm! The moment I start lsusb, the keyboard
LEDs
| light up and I can use the device normally, just like that. Thanks a lot!
| I'll just add lsusb to my usbhostmode.sh script now. :)

Community debug assist is a marvellous thing.

- -Andy
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org

iEYEARECAAYFAklKPdwACgkQOjLpvpq7dMrTxwCeNFQOdyhW2kE4BDNjIiMHkmY9
vRkAn0/SGUTPtnF8Bnb00wIy66AP0qAk
=aTWu
-END PGP SIGNATURE-

___
support mailing list
support@lists.openmoko.org
https://lists.openmoko.org/mailman/listinfo/support


Re: [FDOM, 2008.9] USB not working after suspend/resume-cycle

2008-12-17 Thread Andy Green
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

Somebody in the thread at some point said:
| Hi there!
|
| Recently I got myself a usb keyboard for my freerunner, and it works quite
| nicely after putting the fr into host mode like described in the wiki.
Well,
| except for one little thing: It only works if the freerunner has not been
| suspended before. If it has, usb host mode seems to be broken.
| With the testing kernel, I was able to fix that by rmmoding and
modprobing the
| ohci_hcd module, but with earlier stable kernels, that module seems to be
| compiled into the kernel :(
|
| Any ideas how I can fix that without upgrading to a new kernel? Is
there some
| special file in /sys/ or /proc/ which I can use to restart the usb
subsystem?
|
| Or, if that isn't possible, how I can upgrade to a new kernel without
breaking
| my current distribution (which is FDOM from October 2008)?

Does it help if you reassert the two host mode things by hand in /sys?

http://wiki.openmoko.org/wiki/GTA02_sysfs#USB_Host_.2F_Device

- -Andy
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org

iEYEARECAAYFAklJf4kACgkQOjLpvpq7dMr2JQCaA9tShpvP7Z4/eYg7dH0Mbnj6
n6oAnR14WURU82hsEsEgpfZgWWEcDaoE
=RjhD
-END PGP SIGNATURE-

___
support mailing list
support@lists.openmoko.org
https://lists.openmoko.org/mailman/listinfo/support


Re: Sad Story

2008-12-12 Thread Andy Green
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

Somebody in the thread at some point said:

| Yeah, If I were going to file it a ticket, it would be this: 'Make
| Openmoko work all the time, in all these scenarios'. I'm sure my ticket

Well, every day I have to figure out what to spend time on, if you
aren't talking about specific issues you are having zero impact on my
choices.  Unless you just like to maintain having something to complain
about, that isn't helping you.

| See, if everything you have made until now were fixed immediately, the

If you stop and think about the opportunity represented by being able to
talk direct to the guy who has some control over hardware, bootloader
and kernel choices in Openmoko, are you sure that the general handwaving
complaining you have fallen back on is the best chance to solve whatever
it is that is bothering you?

- -Andy
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org

iEYEARECAAYFAklCLqgACgkQOjLpvpq7dMoBwwCbBg0YHnPTTQAR8KhV8smRJphC
Rt8AoIkVynqewojbNnhjIKWwWs2N5Y+5
=R2r7
-END PGP SIGNATURE-

___
support mailing list
support@lists.openmoko.org
https://lists.openmoko.org/mailman/listinfo/support


Re: Sad Story

2008-12-12 Thread Andy Green
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

Somebody in the thread at some point said:

| I clearly don't blame the developers here myself. I blame Openmoko's
| selfish management (probably it's board of directors or C*Os) for
| this. I am sure that a developer who talks back against this will just
| get fired/have to face bad consequences.
|
| What would be the best way to 'convince' or 'force' the management to
| change their decisions? Does anyone know a lawyer?

You should first form a list of what issues you have that can be fixed.
~ You'll need it for your lawyer, right?

By an amazing coincidence, that list of actual issues is what the
support mailing list is for.

- -Andy
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org

iEYEARECAAYFAklCuC8ACgkQOjLpvpq7dMpp/wCePEA7pXlQ1xIuMRm2aMVza6c5
4WkAmQHCXELZ+FcHULksEI8NY8PB0raM
=4gJf
-END PGP SIGNATURE-

___
support mailing list
support@lists.openmoko.org
https://lists.openmoko.org/mailman/listinfo/support


Re: Sad Story

2008-12-12 Thread Andy Green
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

Somebody in the thread at some point said:
| The website of Openmoko's launch will tell you when the Openmoko
| Freerunner was released.
| My bills will tell you when I have purchased the Freerunner.
| My Freerunner is available to prove the bugs in it.
| As for the list of Bugs, I'll just take the list from the Trac you
| have set up, with details of when 'each and every bug' was filed and
| resolved/never resolved.
|
| And, I have YOUR statement which proves that till date, many of those
| issues still exist.

Did someone from Openmoko tell you that GTA02 (unlike Windows, Mozilla,
$EVERYTHING_NONTRIVIAL) will have no bugs?

| | What would be the best way to 'convince' or 'force' the management to
| | change their decisions? Does anyone know a lawyer?

| Do you want more proof? Do you want more legal documents? Is this how
| you see this conversation heading to?

Yeah some guy mentioned a lawyer?  Oh: you.

- -Andy
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org

iEUEARECAAYFAklCu/0ACgkQOjLpvpq7dMpIFACeJfstMwX1cIaJx5vMxVvmfk0V
UmEAl17Z8/DccphXNamTkfoiXLpyJ/c=
=jW0L
-END PGP SIGNATURE-

___
support mailing list
support@lists.openmoko.org
https://lists.openmoko.org/mailman/listinfo/support


Re: Sad Story

2008-12-12 Thread Andy Green
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

Somebody in the thread at some point said:

| A quick thought...if Openmoko feels the hardware of GTA02 is inadequate
| to even bother making it a stable phone, should there not be some sort
| of recall and free replacement with a GTA03.

Yes.  That thought is still just as pointless a waste of time as the
first time you posted it 47 minutes before on the support mailing list.

- -Andy
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org

iEYEARECAAYFAklCxFwACgkQOjLpvpq7dMqHQgCfSwVz4ocy8biTpwOHOyJC6+tC
db4AnjmtIRX7/fE2NJWOUJMGh3q3g1kc
=NDKB
-END PGP SIGNATURE-

___
support mailing list
support@lists.openmoko.org
https://lists.openmoko.org/mailman/listinfo/support


Re: Sad Story

2008-12-11 Thread Andy Green
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

Somebody in the thread at some point said:
| Hi,
|
| For instance, the Openmoko GPS signal level is bummed because of a
| hardware fault, also there are merely quirks to solve them. The

... which was worked around in kernel once the aggressor signal was
identified.

https://docs.openmoko.org/trac/ticket/1542#comment:27

| Openmoko battery discharging, suspend and resume have not been ironed
| out fully yet. You would see these in _ANY_ basic phone out there. And
| people have been giving commands to get the GPRS modem to work. You
| see what I'm talking about? Nothing is connected, really.

The problem with charger restarting is solved

https://docs.openmoko.org/trac/ticket/1158#comment:33

and suspend and resume should be on a way more solid footing now on the
2.6.28-rc andy-tracking kernels.  Starting next week the root
filesystems will be adapted for these kernels and we start to provide
packages for them.

| Now, If you can't get to fix these today, how do u expect a person to
| program it well enough to rescue it back? You can't expect everyone to
| spend all their time getting their bare minimum phones to work, do
| you?

Well, three of the four things you mention are solved AFAIK.  It leaves
people have been giving commands to get the GPRS modem to work... it
doesn't quite sound like the terrible thing that supports the general
argument you're making.

- -Andy
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org

iEYEARECAAYFAklBQ4oACgkQOjLpvpq7dMpMqwCeJRix5jq8OH4JHR9OZiy+ylQS
KYIAnjwxIPRXmY/qiBsDF0yAXhg1AUIB
=JLh7
-END PGP SIGNATURE-

___
support mailing list
support@lists.openmoko.org
https://lists.openmoko.org/mailman/listinfo/support


Re: GSM no longer registers on GTA02 - how to troubleshoot?

2008-12-09 Thread Andy Green
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

Somebody in the thread at some point said:
| Hi everybody,
|
| I have been using Freerunner as a daily phone for some months now, and
| yesterday it suddenly stopped registering the GSM network. It was
| working fine in the morning, sending/receiving SMS, making phone
| calls, etc., then in the afternoon suddenly the phone portion was
| completely dead. I restarted qpe (running Om2008.9) a few times, then
| rebooted the phone, pulled out the battery and SIM and reseated
| everything, etc., but got nowhere. When I got home, I reflashed a
| couple of known good kernel/rootfs combinations, but got nowhere. I
| even tried an FSO image, since it uses a different phoneserver, but it
| didn't work either. I put my SIM into my old dumbphone (which is a
| little bit broken, in that it only works on speakerphone) and it was
| able to register with no problems.
|
| How can I troubleshoot GSM registration to see at what point the
| process is failing? I would hate for this to be a hardware problem,
| but I don't know what to do when reflashing doesn't work.
|
| I do have another issue, which I don't think is related: my NAND
| u-boot seems to be borked. Whenever I boot the GTA02 without booting
| it from the NOR u-boot, I get a scrambled image on the screen and the
| device is frozen. Often, the scrambled image that appears is whatever
| was in the framebuffer right before powering off. If I boot into NOR
| u-boot, I can boot into the OS fine, but if I just try to turn it on
| without holding down the AUX button or try booting into NAND u-boot, I
| get a freeze. I've tried reflashing u-boot several times, but it
| doesn't help. I download u-boot from downloads.openmoko.org and the
| md5 sum of my u-boot image is 40b8f908649f78e7d70358239da96f98. Is
| this a known good image? Why would u-boot fail after reflashing?

Your U-Boot environment can have some issue, that's why the problem
behaviour is so sticky.  You are into trying to get a USB serial
connection on ttyACM0 on your host and meddling from there.

About your main problem not much idea except confirm the SIM still works
on another phone I guess.

- -Andy

-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org

iEYEARECAAYFAkk+ne0ACgkQOjLpvpq7dMo0nwCeJFFUGrQUWaU9ENryQeq74ZEw
xC8AniEvrMspCUSAoOz+CF8D4XZRtewu
=kbxa
-END PGP SIGNATURE-

___
support mailing list
support@lists.openmoko.org
https://lists.openmoko.org/mailman/listinfo/support


Re: Neo due-2 - NG - last chance

2008-12-08 Thread Andy Green
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

Somebody in the thread at some point said:

| How about the NOR U-Boot, this has been quite reliable in bringing up
| backlight in my experience?  What happens with backlight if you hold
| down AUX during powerup to get into NOR U-Boot?

| When I'm booting to NOR U-Boot my FreeRunner is available with
| USB(dfu-util), but the screen is black — no backlight and not text (i
am use
| paltop with damage backlight - know ).


I blew up the backlight regulator on a GTA02 A5 here by hotplugging the
LCM connector... the flat cable has a 2 layer set of terminating pads on
it and it trashed the LED regulator on the PMU.  But if you didn't even
open the case I dunno what makes the problem, I never saw the regulator
have trouble without me causing it.  The LCD flex cable is held in place
by some sticky EMI tape so I doubt it moved.

NOR flash code should definitely start the LED regulator up without any
question.

- -Andy
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org

iEYEARECAAYFAkk9dPoACgkQOjLpvpq7dMoRxACfSARykgE2OhlPIbqYDcazdvjl
STUAn0lpVCzNKSizhmER6kD0/29yh7bi
=B1f6
-END PGP SIGNATURE-

___
support mailing list
support@lists.openmoko.org
https://lists.openmoko.org/mailman/listinfo/support


Re: dfu-util refusing to flash correct qi image

2008-12-07 Thread Andy Green
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

Somebody in the thread at some point said:

| That did it, allright.
|
| I never thought to try it, since the instructions said to check the
| dmesg (which told me to use qi-s3c2410-master_2ad3ce6ff57753e3.udfu) or
| to use the 6410 image...
|
| Mystery solved, thanks.

Yes you got good advice already but for the record:

qi-s3c2410 -- GTA01
qi-s3c2442 -- GTA02
qi-s3c6410 -- GTA03

- -Andy
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org

iEYEARECAAYFAkk7m2cACgkQOjLpvpq7dMqZigCeJij1iO39MAitiNqnzUoSuzaU
o6gAn3Tp0qJ0JAD+zVzz1OGOb0fPeFfw
=m1nu
-END PGP SIGNATURE-

___
support mailing list
support@lists.openmoko.org
https://lists.openmoko.org/mailman/listinfo/support


Re: WSOD solutions??

2008-12-04 Thread Andy Green
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

Somebody in the thread at some point said:
| Am Do  4. Dezember 2008 schrieb Andy Green:
| Somebody in the thread at some point said:
|
| | Further plans: modify driver to take down the complete LCM via LDO6
| instead of
| | entering deep_suspend. This seems to be the more reasonable
approach for
| | saving power while LCM disabled anyway, and it might cure WSOD as the
| LCM is
| | powered up from suspend-mode the same way as on boot. Side-effects
still
| | under investigation (sneak currents, any persistent data lost during
| power
| | down, timing issues for LCM coming up)
|
| Any reason none of this is on the Trac report?
| We decide on when to post any results to Trac. Emphasis on results.

Interested parties internal and external likely read either kernel list
or the Trac bug

https://docs.openmoko.org/trac/ticket/1841

or both, but not really support list.  For this kind of hard to
reproduce issue, you and the guys that are working on this (Jeremy?)
need to make a decision to work both with customers experiencing it
and other folks internal and external who already poke at the same bug.

| The hard reset line we operate into the LCM should make whether the
| power to it is up in suspend moot... neither does it explain why the
| WSOD comes with the blanking changes that are nothing to do with suspend.
| Thanks for concise summary of obvious conclusions resulting from my
previous
| posting.

But I didn't see you mention this obvious conclusion.

I'm also surprised if there's really nothing to this temperature thing.
~ Several users told it to reliably be the case on their bad LCM.

BTW I already removed LCM power at suspend already a couple of days ago
in andy-tracking

http://git.openmoko.org/?p=kernel.git;a=commitdiff;h=4a116494da4e43fdd8c40c08cc0b2e9cf3fda154

- -Andy
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org

iEYEARECAAYFAkk3lroACgkQOjLpvpq7dMqX9gCfX+VASZBAHvGWzrJSazUz7/cV
ROMAn3ZKmERKAD75/wBe2/nPghFF2ak1
=QnY4
-END PGP SIGNATURE-

___
support mailing list
support@lists.openmoko.org
https://lists.openmoko.org/mailman/listinfo/support


Re: Accelerometers jerky with new stable kernels?

2008-11-07 Thread Andy Green
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

Somebody in the thread at some point said:
| Timo Jyrinki wrote:
| 2008/11/6 Michael Zanetti [EMAIL PROTECTED]:
| What do you mean with working fine now? Are they working as they
were always
| or are they now really working (no crashes on suspend or after some
minutes of
| use)?
| Working fine as in accelerometer using applications (I have
| AccelGame and Doom) work, unlike with the patch (jerkiness). I don't
| use suspend with my own kernels since resume is always broken -
| probably because I only flash the self-compiled kernel and don't know
| how I could easily also copy the compiled modules into use on Neo.
|
| Using this reverting list [1] to get a woking stable kernel (image [2]).
| With itthe accelerometers work well also after suspend/resume (that
| works here, since it seems needed to the
| fix-glamo-mci-slow-clock-until-first-bulk.patch as reported in the
| kernel ML).
|
| I'm using my moko with OpenDoom in these days and all is completely
usable!

It's great you found some suspend stability, but on 2.6.24 / stable
there is zero doubt it is purely random chance build by build.  I would
guess that's why you can 'tune' stability by just removing patches that
affect the resume race by timing, even though the patches themselves
have no direct impact on suspend otherwise.  I found I could make or
break a 2.6.24 kernel for suspend merely by adding printk()s around.

Real solution is coming closer in stable-tracking for these problems
hopefully, although right now accel comms is completely broken in there.
~ But it should be straight in next couple of weeks kind of timeframe.

- -Andy
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org

iEYEARECAAYFAkkUAsIACgkQOjLpvpq7dMp1dwCdE5T8mmGHzDCg5qFD3bTBlZ4K
+0IAn2Y4lP/CSx26OduO+KY6RJ9aifa7
=7RO2
-END PGP SIGNATURE-

___
support mailing list
support@lists.openmoko.org
https://lists.openmoko.org/mailman/listinfo/support


Re: gps external antenna

2008-11-06 Thread Andy Green
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

Somebody in the thread at some point said:

| How do I tell the FR to use the external antenna instead of the
| internal? Nothing about that in the wiki.

As folks already said it's automatic: it detects the current eaten by
the LNA in the external antenna when it is connected, and switches over
to using that instead of internal antenna for the duration.

- -Andy
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org

iEYEARECAAYFAkkSsuUACgkQOjLpvpq7dMoGnQCgk26Z23UOtAn+UF2mNfScJ6Lz
JsAAn25OYGIuJBY7DPfwtUETiuOnltzz
=Va8M
-END PGP SIGNATURE-

___
support mailing list
support@lists.openmoko.org
https://lists.openmoko.org/mailman/listinfo/support


Re: Accelerometers jerky with new stable kernels?

2008-11-06 Thread Andy Green
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

Somebody in the thread at some point said:
| Hi,
|
| I get huge stalls with accelerometers if using the current kernel in
| testing or self-compiled stable branch kernel. Eg. in accelgame or
| opendoom, the fps is ca. 1-3, sometimes a few frame rapidly in
| succession and then 1 second of stalling. Anyone else experiencing the
| problem, possibly with a cure?
|
| During the weekend I tried one self-compiled kernel with all the
| lis302dl starting patches from
| http://git.openmoko.org/?p=kernel.git;a=shortlog;h=stable reverted,
| but it didn't help at all. Any idea which patch would be causing the
| problem?
|
| Okay, while writing this I noticed I didn't revert
| fix-lid302dl-bitbang-all-the-way-baby.patch, so it might be it, but
| I'd like to hear from others and also if it's a known problem to be
| fixed.
|
| -Timo, wants the accelerometers back (without losing other goodies in
| recent stable kernel)

How about this one:

http://git.openmoko.org/?p=kernel.git;a=commitdiff;h=35d910f40ce2c7c959edfbaa671f88e8da5fa6b1

- -Andy
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org

iEYEARECAAYFAkkSuygACgkQOjLpvpq7dMoQUgCggZ1GSNemdnhdV1BnegngF+3U
a5YAniuPHsER0WLhv2k6HSiKlMEhw8sn
=zn5m
-END PGP SIGNATURE-

___
support mailing list
support@lists.openmoko.org
https://lists.openmoko.org/mailman/listinfo/support


Re: Bad blocks in NAND

2008-10-22 Thread Andy Green
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

Somebody in the thread at some point said:
| As I said some days ago in a discussion about flashing via dfu-util or
| doing it after having booted from SD, I did it several times now during
| the last 2 days, and it works great. Another advantage is, after
| flashing the rootfs, you can adjust settings on the newly flashed fs
| (copy dropbear host keys, ssh authorized_keys, wpa_supplicant.conf and
| such) before booting the new system.
|
| However, I get this messages when erasing the rootfs:
|
| debian-gta02:/media/mmcblk0p6/Om2008.9# flash_eraseall /dev/mtd6
| Erasing 128 Kibyte @ f60 -- 99 % complete.
| Skipping bad block at 0x0f62
| Skipping bad block at 0x0f64
| Skipping bad block at 0x0f66
| Skipping bad block at 0x0f68
| debian-gta02:/media/mmcblk0p6/Om2008.9#
|
| Is it really always 128 kB that are marked as bad block? Anybody else
| got bad blocks already? It looks like dfu-util isn't reporting these
| errors from flash erase.

There's a bad block table floating around at the end of the flash, it is
falsely marked as being itself in bad blocks so it is left alone.  I
guess this is that.

- -Andy
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org

iEYEARECAAYFAkj/r8kACgkQOjLpvpq7dMoIrQCeOy8VtYO9La44nHahNW7srn4k
LJMAnjPIgyeiyXAzMlRKMruEHwILcHqU
=3LVb
-END PGP SIGNATURE-

___
support mailing list
support@lists.openmoko.org
https://lists.openmoko.org/mailman/listinfo/support


Re: [FDOM FSO] no sound in calls

2008-10-13 Thread Andy Green
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

Somebody in the thread at some point said:
| Hi all,
| I have my freerunner since around a month, and sound has always worked.
| As I've got FDOM in the flash, I tried to install the raster image,
| along with FSO. very nice peace of work !
| but I hadn't any sound in zhone, like some others in the mailing lists,
| so I switched back to my FDOM.
| Now I neither haver sound in FDOM  this used to work like a charm !
| I tried to replace the state files in /usr/share/openmoko/scenarios, but
| this does'nt work either ...
| Anyone has an idea on how to solve it ?

Confirm in dmesg that your audio is recognized in Alsa, something
interesting seems to be going on around that.

Bad:

~ ALSA device list:
~  No soundcards found.

Good:

~ ALSA device list:
~  #0: neo1973-gta02 (WM8753)

- -Andy
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org

iEYEARECAAYFAkjzOGYACgkQOjLpvpq7dMpCvQCghnBcKaGOTRHvkvSWAQc+ffv6
bPAAn3iooYBdkcMHHZiXv20bK2jOgu2p
=+Xxb
-END PGP SIGNATURE-

___
support mailing list
support@lists.openmoko.org
https://lists.openmoko.org/mailman/listinfo/support


Re: Om2008.9 suspend? - OM2007.2 battery life Om2008.9 battery life

2008-10-03 Thread Andy Green
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

Timo Jyrinki wrote:
 2008/10/1 Andy Green [EMAIL PROTECTED]:
 fixed this tonight on another tree and will update stable with it later
 or tomorrow.
 
 Just as a clarification, isn't the stable kernel tree only actually
 available in the unstable repository currently? Even testing has a
 kernel version from August (git a1e97c611), and the same version is in
 2008.8 (stable) repository.

Right -- I am talking about stable git branch for kernel itself.
Currently updating of kernel in packaging is rather conservative...
something I both approve of and am a bit dubious about at the same time.
 At least if folks can get at it from unstable it's OK, but I wonder
what the deal is with mixing stable basis for the distro with
cherrypicked kernel from unstable when it comes to module packages.

- -Andy
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org

iEYEARECAAYFAkjl2QIACgkQOjLpvpq7dMoJyACcDDJHyLxH6HxJpqfsJPK9Aro+
czMAn1TQ+YF9OrSRbmyKWqhG1odEAhE0
=aLSh
-END PGP SIGNATURE-

___
support mailing list
support@lists.openmoko.org
https://lists.openmoko.org/mailman/listinfo/support


Re: SD-Card Slot is Dead

2008-10-03 Thread Andy Green
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

Somebody in the thread at some point said:
| On Fri, Oct 3, 2008 at 2:00 PM, Andy Green [EMAIL PROTECTED] wrote:
| In the factory, SD card is used for production, so it is 100% tested
| (and fairly unlikely to die).
|
| As in, in the phone, or put in a card reader to verify it's working,
| or something else? No, this isn't important, I'm just curious :-)

As in the phone, we use SD Card boot to hold the production test software.

- -Andy
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org

iEYEARECAAYFAkjmCosACgkQOjLpvpq7dMp5oACfW32266X4Ny2iAu5+8VhUu+hg
xD4An3r/Fm8YboNpclmmkU19itgxCmP/
=R7nQ
-END PGP SIGNATURE-

___
support mailing list
support@lists.openmoko.org
https://lists.openmoko.org/mailman/listinfo/support


Re: battery life problem

2008-10-03 Thread Andy Green
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

Somebody in the thread at some point said:
| Just a quick addition to the power topic -
|
| I compiled and ran powertop on the phone yesterday. I know there's
been some
| previous discussion of using tickless kernels on the freerunner (and the
| problems encountered with those), and indeed the kernel was waking up to
| tick more than anything else.

Cool thing to do.

| However, about 25% of kernel waking came because of chatter from the
| accelerometers, which I don't use at the moment (they're a
bell/whistle for
| after I get the phone stable).
|
| Anyone know a way to turn them off completely?

If you don't have the event device nodes for them open, they do not run
the interrupts.

Use lsof -n to see who has them open.

- -Andy
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org

iEYEARECAAYFAkjmCwEACgkQOjLpvpq7dMrG9QCfefdDmSt3SzX9AI3rrWjFyFQB
KSIAniPdBRNp4bsbI1c3w2Kkqp0LJDDX
=8QXn
-END PGP SIGNATURE-

___
support mailing list
support@lists.openmoko.org
https://lists.openmoko.org/mailman/listinfo/support


Re: Om2008.9 suspend? - OM2007.2 battery life Om2008.9 battery life

2008-10-01 Thread Andy Green
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

Somebody in the thread at some point said:

| if it is - you are victim of the qpe halted suspend before it ever
got to the
| kernel bug. ompower is stuck waiting for apm -s to return
(indicating resume)
| at which time it will turn the light back on.

I guess I should have said if anything in userspace decides... then,
but it is evidently the right idea.

Just curious, what's qpe's POV about intercepting the suspend?  I guess
there are plans to have it not do it without a better reason?

- -Andy
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org

iEYEARECAAYFAkjjLFYACgkQOjLpvpq7dMpXowCeNvH5dqvXRaCPtWrh8nlD/g3o
ry0AniXta9j+dJ0ukIedgenCzULvlyXh
=hHhe
-END PGP SIGNATURE-

___
support mailing list
support@lists.openmoko.org
https://lists.openmoko.org/mailman/listinfo/support


Re: Om2008.9 suspend? - OM2007.2 battery life Om2008.9 battery life

2008-09-30 Thread Andy Green
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

Somebody in the thread at some point said:
| I was thinking of switching to the latest Om2008.9 from OM2007.2
| because Om2008.9 resumes a lot quicker from suspend.  However, the
| battery life with Om2008.9 is greatly reduced.  This leads me to
| believe that Om2008.9 either isn't truly suspending--more of a
| standby--or just isn't suspending everything that OM2007.2 does.
| Anyone else experiencing the same thing?
|
| The one difference is that Om2008.9 was running from the microsd card.
|  Would that make a difference as far as power usage during suspend?

There's this new userspace power daemon that makes decisions about
backlight and suspend now.  If it decides to just keep backlight off
for long periods, there's a bug in recent kernels that actually leaves
the backlight slightly on (trying to avoid another bug in the PMU).  I
fixed this tonight on another tree and will update stable with it later
or tomorrow.

It might not be the whole story but it should help.

- -Andy
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org

iEYEARECAAYFAkjioNAACgkQOjLpvpq7dMpzdQCdH14VCgoAi0mwZdt3zNo/cXnd
A3gAoIMKQmiYrZKE2FiGbwBG77gyntDC
=DmKW
-END PGP SIGNATURE-

___
support mailing list
support@lists.openmoko.org
https://lists.openmoko.org/mailman/listinfo/support


Re: neo1973 does not power on.

2008-09-24 Thread Andy Green
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

Somebody in the thread at some point said:
| Michael + Andy,
|   I haven't had it plugged in since I had the problem a couple days
ago. so I am
| pretty certain the back up battery is empty. I tried just powering it
up now
| that I've waited and nothing.
|
| I also tried the boot to NOR uboot procedure and nothing happened.
|
| It looks like the battery is a little low now so I'm going to see if
it will
| work after I let it charge up.
|
| Just to be clear this is a GTA01Bv4.

Michael pointed it out too -- I didn't notice we are talking about GTA01
instead of GTA02; it's no wonder you didn't mention NOR because there
isn't one on there.  Sorry for the noise.  (Kernel sources are the last
holdout where neo1973 tag refers to both GTA01 and GTA02, so seeing
neo1973 didn't trigger me to think about GTA01).

If something bad did happen to U-Boot bit of the NAND then you are going
to need a debug board to recover if I understood it, so let's hope you
get lucky after some charging.

- -Andy
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org

iEYEARECAAYFAkjZ2acACgkQOjLpvpq7dMr5ygCfUdrZCtMdnBJS9DPm+FFK6Hpc
7QIAnjTVtczC/g7WDmC9cVfmUxYyimZa
=z0tw
-END PGP SIGNATURE-

___
support mailing list
support@lists.openmoko.org
https://lists.openmoko.org/mailman/listinfo/support


Re: Battery Indicator problem

2008-09-24 Thread Andy Green
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

Somebody in the thread at some point said:

| @Andy, here some outputs:
|
| [EMAIL PROTECTED]:~#
| cat /sys/devices/platform/bq27000-battery.0/power_supply/bat/current_now
| cat: read error: Timer expired

Something's broken somewhere... there have been issues with HDQ
communication timing but that's only on 2.6.26+ branches IIRC.

The obvious thing to do would be swap out with another CC battery and
see if the problem belongs to that battery or the Freerunner, but I
guess you only have the one battery.

Timeout error as shown typically means either the battery HDQ pin is not
connected, but it can mean other things like our timing is wrong talking
to it, pin on CPU is dead, etc.

- -Andy
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org

iEYEARECAAYFAkjZ4zAACgkQOjLpvpq7dMpeUQCfcrZl+BNl2TF+qRHKtI28E7Mg
NuwAnA3ZCKw80pNJIa2FxWvAreFUnVqG
=bK6j
-END PGP SIGNATURE-

___
support mailing list
support@lists.openmoko.org
https://lists.openmoko.org/mailman/listinfo/support


Re: neo1973 does not power on.

2008-09-24 Thread Andy Green
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

Somebody in the thread at some point said:
| After letting the phone charge over night the battery voltage has dropped
| significantly. I'm starting to suspect the battery or else the
charging circuitry.
|
| which brings me to 2 questions:
| 1) If I want to take the battery out of the equation and run this off my
| variable power supply, what is a reasonable estimate for the max
current it
| should draw? If the third, middle, terminal ( I'm guessing a smart
battery line)
| isn't connected, should it still power up?

If the GSM side transmits, it pulls 2A at that time, otherwise
typically 300mA at battery.

| 2) Will the newer Debug board that comes with the GTA02 work with the
GTA01?

I think so, but I never used GTA01 so I can't give you totally straight
answer.

- -Andy
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org

iEYEARECAAYFAkjaTJcACgkQOjLpvpq7dMo64wCePU0wfFVIkhXtjQ2OTVIHM1GN
2EQAni2cM6NHX8aqCYaT6wL1kl77MyxM
=VwM8
-END PGP SIGNATURE-

___
support mailing list
support@lists.openmoko.org
https://lists.openmoko.org/mailman/listinfo/support


Re: neo1973 does not power on.

2008-09-23 Thread Andy Green
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

Somebody in the thread at some point said:
| After a recent reboot, the system never came back. Pushing the power
button
| causes nothing to happen. I don't even seem to get into uboot(no logo
screen). I
| don't have a debug board. I do not have a sim card. I've been running
Debian off
| of the SD card mostly with success. I reflashed Uboot a few weeks ago,
but have
| rebooted many times since then without incident.
|
| most recent changes (prior to last reboot):
| 1) I added the onboard fs to my fstab in my SD debian install so that
I could
| access it while booted into debian.
| 2)I changed from the debian X.org server over to the Xfbdev server
copied from
| the onboard fs. It worked fine when I did a startx. And I know it was
running
| the Xfbdev server because I was able to screen rotates with xrandr.
|
| Things I've tried:
| 1) I tried pulling the battery, to clear state. Then restarting.
| 2) Letting it sit plugged in to make sure the battery had good charge. I
| measured 4.1 volts across the battery terminals(out of the phone).
| 3) I removed the SD card. and repeated the above steps.
|
|
| Any suggestions as to what to try next?

See if this helps: pull the battery, wait a few seconds, then hold down
AUX key and keep it down, reapply battery, hold power button in too, let
go of AUX then let go of power button.

- -Andy
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org

iEYEARECAAYFAkjZMCgACgkQOjLpvpq7dMriwwCfTxga3hBT0qZWhG/9RIbVGPwm
weoAn3bDilTrPorc8gK+zp219s2E1sBT
=ZRcy
-END PGP SIGNATURE-

___
support mailing list
support@lists.openmoko.org
https://lists.openmoko.org/mailman/listinfo/support


Re: neo1973 does not power on.

2008-09-23 Thread Andy Green
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

Somebody in the thread at some point said:
| | After a recent reboot, the system never came back. Pushing the power
| button
| | causes nothing to happen. I don't even seem to get into uboot(no logo
| screen).
|
| See if this helps: pull the battery, wait a few seconds, then hold down
| AUX key and keep it down, reapply battery, hold power button in too, let
| go of AUX then let go of power button.
|
|
| Andy,
|
| For the curious, and for further enlightenment, can you explain what
| that does?

It just starts him up in NOR U-Boot instead of NAND.  The guy didn't
mention NOR, it can be he trashed the NAND U-Boot during meddling with
Debian and he will be able to recover via NOR if that's the case.

I found on some devices with various vintages of NOR U-Boot and / or
trouble you need to do that particular sequence to get it to start, so I
described it in detail just in case.

- -Andy
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org

iEYEARECAAYFAkjZUD0ACgkQOjLpvpq7dMom/ACeOaYV4iVNjoj1xMvYEGXsd2AK
AEIAoJHzZ02KACFV6r/e+AYrO3A05Kq1
=FSqm
-END PGP SIGNATURE-

___
support mailing list
support@lists.openmoko.org
https://lists.openmoko.org/mailman/listinfo/support


Re: Splash Issue

2008-09-23 Thread Andy Green
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

Somebody in the thread at some point said:
| On Saturday 20 September 2008 02:07:51 Armin ranjbar wrote:
| after Upgrading to om2008-update , it seems to i lost my 'splash' image ,
| today i have tried to write down om2008.9 splash by
| dfu-util -a splash -R -D Om2008.9.splash.gz
|
| but no luck , when i turn on device images from last state before
shutdown
| appear and no splash screen , any idea ?
|
|
| Is it supposed to still be compressed when you do that?

Yes, U-Boot unzips it.

The problem is probably with your U-Boot environment, it should have
some magic like this lying around

splashimage=nand read.e 0x3200 splash 0x5000; unzip 0x3200
0x0880 0x96000

If you do it by hand at U-Boot prompt it should show the splash in the
framebuffer.  It should be set in the environment to autoexecute
somewhere along the line but I have forgotten exactly how.

- -Andy


-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org

iEYEARECAAYFAkjZUXwACgkQOjLpvpq7dMopYQCfRpjxuEdpwYuWfe1nvNP5Dp0N
OwUAnRt5gUmlAfPpboSqR65yCUQQrw4l
=t0Dg
-END PGP SIGNATURE-

___
support mailing list
support@lists.openmoko.org
https://lists.openmoko.org/mailman/listinfo/support


Re: kernel panic on the second boot from the sd

2008-09-18 Thread Andy Green
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

Somebody in the thread at some point said:

| The problem is that sometimes when Linux tries to access the sd card,
| the card isn't ready yet and the first read doesn't get any data. After
| the first successful read in a session everything's gravy and you can
| read/write all you want, but in some circumstances software gives up at
| the first error so you never get a chance to try again.  For instance if
| that first access is attempting to get the partition table and fails,
| then the kernel is going to think no partitions exist.  In the case with
| the kernel panic on boot, the unknown-block (179,2) is exactly the
| same numbers I get and (I assume) it's trying to either access block 179
| of partition 2, or a partition starting at block 179 (the first block of
| partition 2?) and get sector 2 of it.  In any case what is going on here

No, these are the device numbers, where 179 is the mmc block devices.
It's basically saying that the second partition of that device doesn't
exist.

These would always be the numbers you see if you ask to boot from
mmcblk0p2 regardless of what the problem is.

I think you're right about the card not being ready somehow and timing
out though.

- -Andy
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org

iEYEARECAAYFAkjSsPQACgkQOjLpvpq7dMrA8wCghGAD0+c7RWR9uPGtn+7JEkWu
1EAAn0pO2FV9B03xbxBAPGm3IR1+ESpT
=cxox
-END PGP SIGNATURE-

___
support mailing list
support@lists.openmoko.org
https://lists.openmoko.org/mailman/listinfo/support


Re: some question about Debug Board

2008-09-14 Thread Andy Green
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

Somebody in the thread at some point said:
| Hi xiangfu:
|
| i receive a new debug board and the serial console is work.
| yesterday i try to use openocd but it is not work.
|
- --
|
| Bus 002 Device 015: ID 0403:6010 Future Technology Devices
| International, Ltd FT2232C Dual USB-UART/FIFO IC
|
| Your debug board is not programed or been cleaned eeprom, the USB ID is
| default FTDI ID 0403:6010
|
| Unable to find FTDI devices under given vendor/product id: 0x1457/0x5118
| Retrying with default FTDI id.
| Used eeprom space: 102 bytes
| FTDI write eeprom: 0
| Writing to file: neo1973_debug_board_v2.eeprom
| FTDI close: 0
|
| Same problem, it's not 0x1457/0x5118 anymore. so flashing fail.
|
| You could modify the openocd.cfg or openocd-debugv2.cfg file ,
| ftdi_vid_pid 0x0403 0x6010 make it work without refresh eeprom into
| correct USB ID.

Good job Tony answered, my experience is with exactly one rev2 debug
board that came from Willie with EEPROM set up so it's a much education
for me as you Xiangfu :-)

- -Andy
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org

iEYEARECAAYFAkjMzE4ACgkQOjLpvpq7dMoaogCfcIsCKLHzFPHQDSqUBPSboq3Q
rFIAoJMkHDM9DvoRFkdUyCqvX+oMFCao
=PQ5g
-END PGP SIGNATURE-

___
support mailing list
support@lists.openmoko.org
https://lists.openmoko.org/mailman/listinfo/support


Re: microSD slot fried?

2008-09-12 Thread Andy Green
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

Somebody in the thread at some point said:

| mmcblk0: mmc0:b360 USD   7070656KiB
|  mmcblk0:6glamo-mci glamo-mci.0: Error after cmd: 0x8310
...
| glamo_mci.sd_max_clk=1000
|
| This did help, I followed instructions at
| http://docs.openmoko.org/trac/ticket/1743
|
| How is it possible that a previously working card suddenly becomes
| non-working without the max clk, I wonder?

The issue here is that timeout on Glamo MCI is in units of MMC clocks,
but part of the latency for data access to the NAND in the card is in
absolute time.  So for example, it takes 100ms to get the data off the
memory in the card.

The absolute time that passes in the Glamo timeout depends entirely on
the clock speed, it's double the duration at 12.5MHz than it is at 25MHz
for example.

So reduction of clock here is just extending the absolute timeout period
to match what your (slow) card needs.

- -Andy
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org

iEYEARECAAYFAkjKLyYACgkQOjLpvpq7dMq43wCbBPyb/SyE1mROkhkKYK6csbKM
Qf8An3NzhubMZmuEMozI0vBme1LZks9J
=iz//
-END PGP SIGNATURE-

___
support mailing list
support@lists.openmoko.org
https://lists.openmoko.org/mailman/listinfo/support


Re: getting error while trying to flash my neo freerunner

2008-09-05 Thread Andy Green
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

Somebody in the thread at some point said:

| now where is the function read? and what does this function? I can't
find anything in the sam7dfu.c and in the included files, there is even
no function read.
| Only one file I can't check because I dunno where it is: config.h in
the svn trunk.

I don't think it makes the problem, but try

man 2 read

on a Linux box, read() is part of C standard library.

- -Andy
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org

iEYEARECAAYFAkjA984ACgkQOjLpvpq7dMpQLACgjrg8t0E8I5IG+qJnGl3bkNZ3
z44An3045iHl5ltoCahd8Opf3XUSOfBl
=BOxH
-END PGP SIGNATURE-

___
support mailing list
support@lists.openmoko.org
https://lists.openmoko.org/mailman/listinfo/support


Re: getting error while trying to flash my neo freerunner

2008-09-05 Thread Andy Green
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

Somebody in the thread at some point said:

| man 2 read
|
| on a Linux box, read() is part of C standard library.

| thanks for answer ;)
|
| but I get on my Ubuntu
| man 2 read
| No manual entry for read in section 2
| 

You can see it here --

http://www.linuxmanpages.com/man2/read.2.php

- -Andy
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org

iEYEARECAAYFAkjA+mIACgkQOjLpvpq7dMo6dQCeL3YxQ7ZMGl/DATfPNP2Up39O
1acAni7CVpR5BQeqLj6RtLvpjq98Rwf0
=G3GI
-END PGP SIGNATURE-

___
support mailing list
support@lists.openmoko.org
https://lists.openmoko.org/mailman/listinfo/support


Re: microSD slot fried?

2008-09-05 Thread Andy Green
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

Somebody in the thread at some point said:

| [3.165000] mmcblk0: mmc0:b368 SMI   498176KiB

That's the good part above, it talked to the card and got information
out of it (512MB card).

| [3.17]  mmcblk0:6glamo-mci glamo-mci.0:
| Error after cmd: 0x302

Then it all went wrong, 0x0302 is actually CRC error with data ready,
never seen it before.  After that it's dead meat.

Looks like bulk transfer on that card doesn't work under current default
conditions, eg max clock speed.

You can meddle with max clock speed on kernel commandline, try eg

glamo_mci.sd_max_clk=1000

- -Andy
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org

iEYEARECAAYFAkjB2fAACgkQOjLpvpq7dMrOZgCfe5ahhyOeWub+CQv4+Uhyo0bv
GbUAmQGKAu9r+VQ8klnNq74hqDPWPqdX
=YirJ
-END PGP SIGNATURE-

___
support mailing list
support@lists.openmoko.org
https://lists.openmoko.org/mailman/listinfo/support


Re: U-Boot DFU mode not working

2008-09-05 Thread Andy Green
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

Somebody in the thread at some point said:
| Hello,
|
| I've tried to upgrade software on my Freerunner, but when I'am trying
to enter
| DFU mode it doesn't work. I'am booting from NOR as described here:
| http://wiki.openmoko.org/wiki/Flashing_the_Neo_FreeRunner
| but when I'am runnig dfu-util it only shows USB Device Firmware Upgrade:

DFU never worked in a straightforward way for me until I wrote a little
workaround script, this one is the u-boot dfu upgrade one I use

#!/bin/bash
../../dfu-util/src/dfu-util -a 1 -d 0x1d50:0x5119 -D u-boot.udfu
if [ $? -eq 1 ] ; then
../../dfu-util/src/dfu-util -a 1 -d 0x1d50:0x5120 -D u-boot.udfu
../../dfu-util/src/dfu-util -a 1 -d 0x1d50:0x5119 -D u-boot.udfu
fi

Somehow it changes VID / PID after enumeration maybe, anyway if you
adapt this script it may help you out.

- -Andy
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org

iEYEARECAAYFAkjB+bwACgkQOjLpvpq7dMqbEACfdh+cosfbJFGYi7kMM9CfKcfp
xT8An3AO1/VlmCfPkmhUkOY6WE6F+HHR
=myFH
-END PGP SIGNATURE-

___
support mailing list
support@lists.openmoko.org
https://lists.openmoko.org/mailman/listinfo/support


Re: Can bad NAND blocks cause USB's device descriptor read/64, error -110 ?

2008-09-02 Thread Andy Green
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

Somebody in the thread at some point said:
| Hello dear all,
|
| My new FR (DateCode 20080725) is suspected to having bad NAND blocks:
| It had been suffering kernel panics etc. after as soon as I reflashed it
| with Om2008.8 on 20080831. Subsequent reflashings rendered my FR less
| unstable.

Copy of the panic backtrace would be interesting if you can get ahold of
it.  Usually every panic is a story although there's no guarantee we can
figure it out.

The kernel image itself is CRC-checked by U-Boot before it is run.  So
it is doubtful we can say that kernel is corrupted by bad NAND.

| I'm still suffering with USB cdc_ether, I'm getting `device descriptor
| read/64, error -110' and no solution found out there on the net helps.

This can be to do with the host PC USB device, cables, stack as well as
Freerunner.  Are there any other exotic variations of that in your host
PC dmesg or is it only that?

| Last time tried with:
| u-boot,rootfs (20080902) -- from Openmoko_update page
| kernel (20080902, too) -- from daily page.
| (!)With the 20080826 kernel from releases/_update page says wrong image
| format whatsoever.. (and still does if I try to revert to the 20080826
one)

That's a funny error to get, can it actually DFU anything reliably?
That is also using the dodgy USB connection of course...

- -Andy

-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org

iEYEARECAAYFAki98T4ACgkQOjLpvpq7dMpJHQCfW5H0o85lQdh8H1siLHev4Ml1
Jd0An0sDcAqCg3O6Xxam5xmp27ADEztI
=Wnwd
-END PGP SIGNATURE-

___
support mailing list
support@lists.openmoko.org
https://lists.openmoko.org/mailman/listinfo/support


Re: connecting a usb 3g modem

2008-09-01 Thread Andy Green
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

Somebody in the thread at some point said:
| On Sun, Aug 31, 2008 at 6:41 PM, Andy Green [EMAIL PROTECTED] wrote:
| Sometimes when I use it on normal laptop, I get bogus DNS in UK.
|
| You mean like if you use
|host somedomain.com
| on your Linux system, when using the modem as the DNS server, you get
| complaints like malformed response etc.?

No the dhcp client puts the DNS nameserver lines in, but they don't
always work.  There's another issue on my box somehow that later,
NetworkManager overwrites it with old DNS nameserver line somehow, but
now I realized why my connectivity breaks I stop NetworkManager for the
duration and that solves it.

| Surprisingly, that proprietary OS that most FOSS people hate to use...
| works just fine with the DNS implementation on the ADSL modems... a
| case of two brokens make one fixed? :-P

Dread to think of it.

- -Andy

-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org

iEYEARECAAYFAki7tLQACgkQOjLpvpq7dMr+UwCfeGhoqAP0Q5bAG8nnJs9WnVNs
fjgAnRNPXcFYBViVb2pm3KbqnOuaNqcz
=pj+t
-END PGP SIGNATURE-

___
support mailing list
support@lists.openmoko.org
https://lists.openmoko.org/mailman/listinfo/support


Re: connecting a usb 3g modem

2008-08-31 Thread Andy Green
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1


Somebody in the thread at some point said:
| The usb modem is recognised succesfully by the kernel,  just simply
| configure ppp to get it connecting to the net.
| I tried with the huawei 270 modem and I succesfully connected but I
| can't ping anything... it is because the modem need one usb 2.0 port or
| two usb 1.0 ports?

I have one of these, they are nice.

I would look at

route -n
ifconfig -a
cat /etc/resolv.conf

Sometimes when I use it on normal laptop, I get bogus DNS in UK.

| I have the wire to connect the modem to two usb 1.0, so just but a usb
| hub for the neo?

Dunno, see how you go, might be OK direct.  Or maybe the ping problem is
that... mine has an LED that goes blue solid when the modem is
authenticated and has comms to network provider.

- -Andy

-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org

iEYEARECAAYFAki6mIwACgkQOjLpvpq7dMoMXgCeK4N7Cg2yQ1z3ROWRi/TIMOor
ieQAnjIlJQy+VB/T95/UbIGQPQu7kHf4
=m9Fw
-END PGP SIGNATURE-

___
support mailing list
support@lists.openmoko.org
https://lists.openmoko.org/mailman/listinfo/support


Re: Official Debian port

2008-08-15 Thread Andy Green
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

Somebody in the thread at some point said:
| Am Freitag 15 August 2008 schrieb sparky mat:
| I noticed the 'Official Debian port' section on the debian page in the
| wiki. Wow! That's cool!
|
| Has someone tried this? If so, can someone update the wiki with a new
| section (or a new page) describing how one can get this working? I
believe
| they are calling for early adopters. As long as it can be installed
to the
| SD card, I'd like to give it a go.
|
| I tried it just yesternight. It didn't work.
|
| The SD(HC)-Card was prepard on my main machine like this:
| 1. partition: 8MB formatted vfat
| for uImage.bin (apart from that: empty). I wasn't sure which
kernel-Image to
| use so I took one from ASU.
|
| 2. partition: 2GB formatted ext2.
| I copied all the directories from the .tar.bz2 (Prebuilt Filesystem)
to this
| partition.
|
| wasn't sure which kernel-Image to use so I took one from ASU. It did
indeed
| boot, but when the kernel tried to start init, it panicked:
|
| Kernel panic - not syncing: No init found. Try passing init= option to
kernel.
|
| ... but I really don't know how to do that without screwing up Neo.

That's not so bad, it sounds like you mounted the rootfs OK.  Confirm
that you have /sbin/init on the ext2 partition.  I guess you ran it from
NOR flash... select USB console and connect to /dev/ttyACM0 on your
Linux host box via minicom or some such.  Do printenv and show the
results here.

- -Andy
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org

iEYEARECAAYFAkilcPMACgkQOjLpvpq7dMoAbgCfURRaHaI6xbhK2jv6gdITcnjU
orsAoI+KLu+fFe/XKD7rjTETgEozZtmg
=exbc
-END PGP SIGNATURE-

___
support mailing list
support@lists.openmoko.org
https://lists.openmoko.org/mailman/listinfo/support


Re: Flash a kernel with USB networking

2008-08-13 Thread Andy Green
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

Somebody in the thread at some point said:
| Hello,
|
|
| I did the same update as Thomas here :
http://lists.openmoko.org/pipermail/support/2008-August/000900.html
| And now I miss the modules g-ether and cdc-ether.
|
| His solution:
| My solution was to flash an older kernel, install the missing modules
with opkg, then re-flash the new kernel.
|
| So my question is how to do that :
| - what older kernel should I put on my sd card?
| - how to install the g-ether and cdc-ether modules from the sd card
with opkg?

There is a class of kernels now called moredrivers that have enough
drivers in the monolithic kernel that they do not rely on the presence
of modules themselves.  For example:

http://people.openmoko.org/andy/uImage-moredrivers-stable_88bf43840b9df0eb.bin

These guys have the Ethernet over USB (and other things) built-in
according to the old school.  You can DFU this in safely and then update
by packages.

- -Andy

-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org

iEYEARECAAYFAkijACEACgkQOjLpvpq7dMoXkgCfWB/qfaoZ1U2SP+UWHJmyAoCX
dpMAnRIgFJA+fuH029r3maDrEl5b1n/k
=L6Ml
-END PGP SIGNATURE-

___
support mailing list
support@lists.openmoko.org
https://lists.openmoko.org/mailman/listinfo/support


Re: gta02, 2007.2: frequent wakeups -- creating workaround

2008-08-13 Thread Andy Green
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1


| To what are you referring?  Did you see that in the linked qtopia code?
| no. i think you said that -- umount/mount on suspend/resume was a
| workaround except when resuming from gsm event. am i wrong?

| Oh, yes.  That is what it seemed like.  Perhaps I was just unlucky the
| first time.  I'll mess with it again after work to confirm that that
| truly is the difference.  But right now, I haven't corrupted my SD
| card for a few days.

I'm pretty sure this will make you immune for getting crapped on.  But
for SD Card rootfs case, it isn't really a solution.

- -Andy
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org

iEYEARECAAYFAkijGY4ACgkQOjLpvpq7dMrCSQCglCwDj9FwEP3k7gDZzC/NrOgj
GEYAn0lvdZw4ZdCqm4V4yYD1w4xcEyov
=Ijzp
-END PGP SIGNATURE-

___
support mailing list
support@lists.openmoko.org
https://lists.openmoko.org/mailman/listinfo/support


Re: Application to fade screen

2008-08-13 Thread Andy Green
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

Somebody in the thread at some point said:
| Hello all--
|
| I was wondering, is there some sort of application that will let me
| manually fade the openmoko screen to black? Similar to a screen saver?
| Thanks!

What are you, some kind of Goth?  :-)

echo 0
|/sys/devices/platform/s3c2440-i2c/i2c-adapter/i2c-0/0-0073/backlight/pcf50633-bl/brightness

- -Andy
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org

iEYEARECAAYFAkijHhMACgkQOjLpvpq7dMoB7wCfSm48iVMHo4x/UwzzXB/iLwZ9
BnYAn1sCvihXkGZe5x4+iZWFYHcA6816
=r6Y4
-END PGP SIGNATURE-

___
support mailing list
support@lists.openmoko.org
https://lists.openmoko.org/mailman/listinfo/support


Re: Application to fade screen

2008-08-13 Thread Andy Green
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

Somebody in the thread at some point said:

|
|/sys/devices/platform/s3c2440-i2c/i2c-adapter/i2c-0/0-0073/backlight/pcf50633-bl/brightness

Wow thunderbird thought it knew better... that's actually a greater-than
symbol not a pipe symbol.

- -Andy
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org

iEYEARECAAYFAkijHtMACgkQOjLpvpq7dMqsjgCfYUTrZ822dGrslNjDz4+idwq5
ZdEAn1YJb4Tnd57DGYs9hXcgEvLj9LT2
=HstU
-END PGP SIGNATURE-

___
support mailing list
support@lists.openmoko.org
https://lists.openmoko.org/mailman/listinfo/support


Re: Application to fade screen

2008-08-13 Thread Andy Green
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

Somebody in the thread at some point said:
| Andy Green wrote:
| Somebody in the thread at some point said:
| | Hello all--
| |
| | I was wondering, is there some sort of application that will let me
| | manually fade the openmoko screen to black? Similar to a screen saver?
| | Thanks!
|
| What are you, some kind of Goth?  :-)
|
| echo 0
|
|/sys/devices/platform/s3c2440-i2c/i2c-adapter/i2c-0/0-0073/backlight/pcf50633-bl/brightness
|
| -Andy
| Now there's a thought! A Goth theme with black buttons on a black
| background that light up black when pressed. And a black screensaver.

Just turn the thing off, we can say it is hardcore goth ;-)  and it
never fails to meet user expectations either...

- -Andy
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org

iEYEARECAAYFAkijXDEACgkQOjLpvpq7dMqP8QCeLFVz5FVzixmH6eaUiGktzX0n
0o4An3r3MTKpWX1zsa4zWhOJ2ltg6n9n
=xiu1
-END PGP SIGNATURE-

___
support mailing list
support@lists.openmoko.org
https://lists.openmoko.org/mailman/listinfo/support


Re: Differences between daily builds and Om2008.8

2008-08-12 Thread Andy Green
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

Somebody in the thread at some point said:
| I flashed the latest build (2008.8.11) from
| http://buildhost.openmoko.org/daily/freerunner/
|
| USB networking is gone. Glamo doesn't seem to be installed by default
| either. And, Om2008.8 doesn't update anything on 'opkg upgrade'.

What does that actually mean, both gone (what appears on host dmesg
when you plug it in) and Glamo doesn't seem installed by default...
you mean XGlamo?

Did you update the kernel at the same time as the rootfs image?

- -Andy
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org

iEYEARECAAYFAkihVQ0ACgkQOjLpvpq7dMrwPwCfRrEPOMtTTl1S+t64FbxXudz3
yo4An2akmm2FCJys8JwZl4kRq5VNQJhc
=gEnc
-END PGP SIGNATURE-

___
support mailing list
support@lists.openmoko.org
https://lists.openmoko.org/mailman/listinfo/support


Re: Differences between daily builds and Om2008.8

2008-08-12 Thread Andy Green
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

Somebody in the thread at some point said:
|
| What does that actually mean, both gone (what appears on host dmesg
| when you plug it in) and Glamo doesn't seem installed by default...
| you mean XGlamo?
|
|
| Yes, XGlamo doesn't seem to be used (the one provided by
| xserver-kdrive-glamo IIRC) since its slower than the release version and
| the 'loading' spinwheel is not there, the dropdown sliding animation is
| stuttered, etc.

I know nothing about that side of stuff, sorry.

| As for USB networking, its gone in the sense I get a usb0 on my ubuntu
| machine, but pinging the Freerunner gives me 'destination host
| unreachable'. Here's the dmesg output

That's what I call USB networking being there from my POV :-)  There is
some routing or IP allocation issue on Freerunner side that I read other
people mention seeing that would match this.

| [98557.649357] usb 4-1: new full speed USB device using uhci_hcd and
| address 18
| [98557.912376] usb 4-1: configuration #1 chosen from 2 choices
| [98557.922447] usb0: register 'cdc_ether' at usb-:00:1d.3-1, CDC
| Ethernet Device, 4a:8b:b5:17:5a:4b
|
| Please note that the same networking setup was working fine 2 hours back
| with the 2008.8 releae build.

Is there some way you can do

ifconfig usb0
route -n

on the Freerunner?

| Did you update the kernel at the same time as the rootfs image?
|
| Yup! Took the kernel and the rootfs jffs2 file from
| http://buildhost.openmoko.org/daily/freerunner/200808/20080811/

Fine -- I don't think the issue comes from that direction if you have
usb0 up on your host.  I guess it is somehow network interface config issue.

- -Andy
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org

iEYEARECAAYFAkihWfIACgkQOjLpvpq7dMqXiwCfYJSTKnYZPjtmHhr04zduGzMZ
X4kAn3jsL4fVNRlyFxvaBFViy+GWv+Dd
=Mz+y
-END PGP SIGNATURE-

___
support mailing list
support@lists.openmoko.org
https://lists.openmoko.org/mailman/listinfo/support


Re: Differences between daily builds and Om2008.8

2008-08-12 Thread Andy Green
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

Somebody in the thread at some point said:
|
| Fine -- I don't think the issue comes from that direction if you have
| usb0 up on your host.  I guess it is somehow network interface
| config issue.
|
|
| As in, network config issue on the Freerunner?

That's my guess from the behaviour.  It's what you would see if
Freerunner had no or wrong IP I think.

- -Andy
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org

iEYEARECAAYFAkihdcgACgkQOjLpvpq7dMp4MwCgiVXWR2Aa7P+AraHB/XlFfqZy
SEIAnjXaI8jVFyX6LnSLMl0e1dx+/eAN
=yKhJ
-END PGP SIGNATURE-

___
support mailing list
support@lists.openmoko.org
https://lists.openmoko.org/mailman/listinfo/support


Re: qtopia dialer completely broken?

2008-08-12 Thread Andy Green
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

Somebody in the thread at some point said:

| I have somewhere between 400 and 800 mb of openstreetmap data on the SD
| card. (As reported by du -hs; many of the files are  4KB...)

I guess it is a pathological filesystem for this thing, but still I
can't imagine why it wants to be looking inside anything unless it has a
recognized extension first, if that is what is happening not doing it
should sharply reduce the impact even if there are a lot of files.

- -Andy
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org

iEYEARECAAYFAkihvBoACgkQOjLpvpq7dMo4nACcD99SLXUF7p6sqHouWAONHYhF
1ysAmwWJPsRwHARnbW5QKkeRu8oaYR4F
=JZxJ
-END PGP SIGNATURE-

___
support mailing list
support@lists.openmoko.org
https://lists.openmoko.org/mailman/listinfo/support


Re: FreeRunner is completely dead

2008-08-10 Thread Andy Green
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

Somebody in the thread at some point said:
| Hi.
|
| I've been using my FreeRunner quite a lot since I got it and yesterday
| evening I played with USB host mode (the goal was to use a normal USB
| keyboard in the terminal). However, I reported my progress and then went
| to bed. [0]
|
| This morning when I got back to my computer, the FreeRunner's screen was
| black as is expected - the battery would be drained as it has been
| several nights/mornings before. But, today I just can't get it to boot
| anymore. The only difference from all other times (that I can think of)
| is that I left it in USB host mode.
|
| By now I've had it plugged into the wall charger that came with the
| phone for more than one hour, but to no avail - it still won't boot. :(
|
|
| Any tip, idea on what I should do or test?

This is a known issue... basically Freerunner cannot boot successfully
without the battery to help it along.  If the battery is dead, then it
won't succeed to boot.

What's actually happening is that there is a large instantaneous demand
for power, AKA inrush current, this causes the USB input to enter
current limit, the voltage drops on the Freerunner side, boot fails, it
restarts and so on.

It won't start to charge either until this early part of the startup
action completes successfully.

You can work around it with another battery that has charge; externally
charging you battery with external charger; or the not recommended but
I did it and there was no smoke in my case provide external 5V briefly
to battery + terminal (making sure they share 0V too) briefly until it
starts.  After that, if it stays awake it should charge OK.

- -Andy
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org

iEYEARECAAYFAkiey6gACgkQOjLpvpq7dMrlZQCdFS2QJSx1VL5wsBDN0GKSIv3c
MXsAn2sSqv73v0udrpsVtlFegpsAKF0A
=BCJ9
-END PGP SIGNATURE-

___
support mailing list
support@lists.openmoko.org
https://lists.openmoko.org/mailman/listinfo/support


Re: FreeRunner is completely dead

2008-08-10 Thread Andy Green
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

Somebody in the thread at some point said:
| You can work around it with another battery that has charge;
| externally
| charging you battery with external charger; or the not recommended but
| I did it and there was no smoke in my case provide external 5V
| briefly
| to battery + terminal (making sure they share 0V too) briefly until it
| starts.  After that, if it stays awake it should charge OK.
|
|
|
| Is it possible that this issue happened with the neo1973 as well, by
| any chance?  I have a neo1973 that also refuses to power on no matter
| what I do with it .. maybe I should use the charged batter of my
| freerunner to see if it can be brought to life?

Was neo1973 ever happy starting up without battery installed?  If not it
could have the same issue.

If you carefully listen to it trying to start with USB power and hear a
soft clicking a few times a second, that would suggest it can be the
same deal.

- -Andy
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org

iEYEARECAAYFAkifD8wACgkQOjLpvpq7dMo9HwCeKdru7/az8vdRtQo9BNTkazQX
398An0cwiJWamGaQac8PKjxXjLOG6Ct3
=StHS
-END PGP SIGNATURE-

___
support mailing list
support@lists.openmoko.org
https://lists.openmoko.org/mailman/listinfo/support


Re: FreeRunner is completely dead

2008-08-10 Thread Andy Green
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

Somebody in the thread at some point said:
| Hi andy-
| |
| | Is it possible that this issue happened with the neo1973 as well, by
| | any chance?  I have a neo1973 that also refuses to power on no matter
| | what I do with it .. maybe I should use the charged batter of my
| | freerunner to see if it can be brought to life?
|
| Was neo1973 ever happy starting up without battery installed?  If not it
| could have the same issue.
|
|
| neo1973 means GTA01, and don't have power from usb designed, so this
| issue looks like power related but may have different case than FR.

Thanks for the lesson Tony :-)  I saw it does have some charging
arrangement about USB power input, I just assumed it had the same plan
as GTA02.  But there is no VB_SYS.

Jay what I would do (but you have to factor in I do blow things up
sometimes) is remove the battery, cut up a USB cable, stick the A end in
my laptop and hold the black wire to 0V pin of battery connector, red
one to +ve pin of battery connector and see what happened when I powered
on.  pcf50633 is good for 6V abs max and USB is 5V.  If you think it is
dead already then this can only do nothing; make it deader; or send some
sign of life.

- -Andy
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org

iEYEARECAAYFAkifFPgACgkQOjLpvpq7dMqeIwCdEUjTu0Jt96eZ/fB2BeQInrru
97wAnAnGh8mrTCpLwzFKBUtRlO0hmCjt
=vt2G
-END PGP SIGNATURE-

___
support mailing list
support@lists.openmoko.org
https://lists.openmoko.org/mailman/listinfo/support


Re: OM2008.08 - Charging indicator?

2008-08-08 Thread Andy Green
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

Somebody in the thread at some point said:
| How can I tell/make sure my Freerunner is actually charging?
| There is what appears to be a lightning bolt in the top right corner,
| but that's there whether it's plugged in or not.
|
| And, since it goes into suspend after a short timeout (30 seconds?),
| does it continue to charge during suspend?

This is my fault, I broke it in the kernel yesterday.  It'll be fixed in
kernel update in the next couple of days.

For charging in suspend, yes, the charger in PMU is autonomous.  If you
change plugged state of USB cable, it will wake, otherwise it will stay
in charging readiness state that it was in before suspend during suspend.

During suspend same as during wake, separate charging state machine in
the PMU determines when charging starts and finishes, so it is pretty
unaffected by suspend action.

- -Andy
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org

iEYEARECAAYFAkicnqkACgkQOjLpvpq7dMob8QCfYElRuCZBCKSZRdNwYOroUOkf
AusAn0efO538iyxTlEMabhXoXDIhPzQt
=H+U8
-END PGP SIGNATURE-

___
support mailing list
support@lists.openmoko.org
https://lists.openmoko.org/mailman/listinfo/support


Re: OM2008.08 - Charging indicator?

2008-08-08 Thread Andy Green
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

Somebody in the thread at some point said:
| Sry I misread your post. Knowing if it's charging or not goes with the
command:
|
| cat /sys/devices/platform/bq2700-battery.0/power_supply/bat/status
|
| Also here the bug is responsilbe for no gui-view of this information.

Yep.

- -Andy

-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org

iEUEARECAAYFAkicnssACgkQOjLpvpq7dMo7+wCVHU3IQww5++y5mIMKC53Qq0iJ
8QCghWMg6OmOSpmReihbRsxLtGgEpZg=
=0ffa
-END PGP SIGNATURE-

___
support mailing list
support@lists.openmoko.org
https://lists.openmoko.org/mailman/listinfo/support


Re: Debian - How to install the kernel

2008-08-07 Thread Andy Green
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

Somebody in the thread at some point said:
| Here's a kernel that has the modules you need built into it already.
|
| http://warmcat.com/uImage-moredrivers-stable:2f775967a997e3b4.bin

| I get a 'Page load Error' when trying to access it. :-|


I guess your downloading app doesn't like the colon, because it
downloads fine here.  Try this.

http://warmcat.com/uImage-moredrivers-stable_507c3c1ba3921bb7.bin

- -Andy
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org

iEYEARECAAYFAkialecACgkQOjLpvpq7dMpI4ACgjP8qTYiWrKFFJMqTyt+GZ3Wl
Gx4An3SGactYEQA4SiSAjSb9V8UfK6zh
=4Hxu
-END PGP SIGNATURE-

___
support mailing list
support@lists.openmoko.org
https://lists.openmoko.org/mailman/listinfo/support


Re: Debian - How to install the kernel

2008-08-07 Thread Andy Green
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

Somebody in the thread at some point said:

| DFU this into the kernel partition and you should be good for
Debian, I
| use these kernels with it myself all day.
|
|
| How do I DFU this into the microSD's kernel paritition?

Sorry I was explaining the NAND case.  You can use it fine on microSD
just mount the microSD on your laptop / PC and copy that file over to
the correct partition as uImage.bin.  Then unmount, return it to
Freerunner and you should boot OK.

- -Andy
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org

iEYEARECAAYFAkialkYACgkQOjLpvpq7dMpSGgCfcRfuOIyiYc5B9SsHqJF00BXH
TrEAoIdQ31HLHkP/ZsJfuzOqlfBLePYk
=l7vu
-END PGP SIGNATURE-

___
support mailing list
support@lists.openmoko.org
https://lists.openmoko.org/mailman/listinfo/support


Re: Kernel with built-in usb drivers / building the kernel cheatsheet

2008-08-07 Thread Andy Green
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

Somebody in the thread at some point said:
| I understand that the usb tether drivers were moved out as kernel
| modules. However, for the Debian port to work, I need them as part of
| the kernel. Can someone point me to how I can re-compile the latest
| kernel with the drivers included?
|
| I am a OpenEmbedded newbie and have minimal experience with the Moko
| Makefile (i.e. I haven't done any modifications, just done a normal
build).

I don't think you need to do this, but of course it is a nice thing to
be able to do anyway.  You will know more about the OE side than me by
the sound of it, but I can give you the steps to build the kernel
standalone with the moredrivers config.

~ - git clone git://git.openmoko.org/git/kernel.git linux-2.6
~ - git checkout origin/stable
~ - edit ./build to point to your compiler path
~ - cp defconfig-2.6.24 .config
~ - ./build

That gets you a uImage.bin you can DFU or copy into your SD Card filesystem.

- -Andy
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org

iEYEARECAAYFAkial1oACgkQOjLpvpq7dMr4fgCcDW3gW/SUfaSoeDF13HNo9nNh
QJcAniA7Jcmy1mZvuoTwJ0Ts0GKPV+lb
=BdM7
-END PGP SIGNATURE-

___
support mailing list
support@lists.openmoko.org
https://lists.openmoko.org/mailman/listinfo/support


Re: No network after flashing 200808

2008-08-07 Thread Andy Green
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

Somebody in the thread at some point said:
| Hi,
|
| After flashing 200808 with kernel and rootfs from 20080806, I have no
| usb and no working wifi.
| Wifi connects, the wireless AP says so.  but a quick nmap of my
| network show no new IP picked up from DHCP.
|
| Should I reflash the kernel with an older one (which?) and add the
| usb-net modules and then reflash with 200808 kernal?

That may not be the cause if you have a U-Boot from a week or more ago.
~ It can knock out your USB functionality 50% of the time.  If that's the
case, just update U-Boot with DFU.

The issue with USB module should be resolved before yesterday in the
packaging, so the rootfs should have the modules, therefore I wonder if
it is the above.

Otherwise, this kernel (eventually intended to be a recovery / backup
kernel) has more modules built-in, including USB over Ethernet: you can
DFU it in and it will get you going again.

http://warmcat.com/uImage-moredrivers-stable_507c3c1ba3921bb7.bin

- -Andy
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org

iEYEARECAAYFAkiap1oACgkQOjLpvpq7dMrj2ACglOrsg72dR06Nc0icpYT03epZ
2A4AoIWrWc95xysTcX6sOn2KWpKQsZWR
=udAp
-END PGP SIGNATURE-

___
support mailing list
support@lists.openmoko.org
https://lists.openmoko.org/mailman/listinfo/support


Re: Debian : Touchscreen calibration

2008-08-07 Thread Andy Green
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

Somebody in the thread at some point said:
| Following the instructions at :
| http://wiki.openmoko.org/wiki/Debian#Get_touchscreen_working , I was
| able to get touchscreen to work as a mouse. But, the cursor is way off
| from where I am touching. I am guessing the calibration mentioned there
| is not accurate anymore? Has anyone managed to get it working on
Freerunner?

The instructions there are not right.  He gives canned numbers when
these vary by individual Freerunner.

You need to install tslib and its utilities (no idea what package, but
it probably is packaged), and run

ts_calibrate

when it exits after you touch the crosshairs in sequence, the magic
numbers it spits to console are the numbers you need to use for your
actual touchscreen.

- -Andy
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org

iEYEARECAAYFAkiasF8ACgkQOjLpvpq7dMpkxwCgkqMCAt3pkW1sGFWFcyc4KV1O
bbEAnjk9HC3dUi9i04JTcZ9UAcf8+qLp
=D7as
-END PGP SIGNATURE-

___
support mailing list
support@lists.openmoko.org
https://lists.openmoko.org/mailman/listinfo/support


Re: Debian - How to install the kernel

2008-08-07 Thread Andy Green
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

Somebody in the thread at some point said:
| Doesn't make any sense to me either... can you md5sum the broken
copy
| before overwriting it?
|
|
| Ok. Will do that the next time I get it. Also, I was using the ASU
| kernel so far. Switched to the one from the link you provided.

We have it configured in our defconfigs, dunno why the ASU one doesn't
have it.

CONFIG_INOTIFY=y

- -Andy
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org

iEYEARECAAYFAkiatbUACgkQOjLpvpq7dMpadQCeLg5yfP2JXdwOjvyH3Ky8DPdo
3icAoIz9JlJJXY+1v9/BCT5NzZqb2y9c
=1PJd
-END PGP SIGNATURE-

___
support mailing list
support@lists.openmoko.org
https://lists.openmoko.org/mailman/listinfo/support


Re: /qtopia/ weird problems after yesterdays' opkg upgrade

2008-08-07 Thread Andy Green
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

Somebody in the thread at some point said:

| 1. klogd went very busy and dmesg got full of
| evbug.c: Event. Dev: NULL, Type: 0, Code: 0, Value: 0
| evbug.c: Event. Dev: NULL, Type: 2, Code: 0, Value: 54
| evbug.c: Event. Dev: NULL, Type: 2, Code: 1, Value: 18

There's a module called evbug that is getting inserted for some reason,
I killed building it today in the configs we use so it should be gone
tomorrow.  It just seems to exist to spew in syslog.

| Aug  7 22:12:19 om-gta02 user.warn kernel: Alignment trap: mediaplayer
(1436) PC=0x42ef7d10 Instr=0xe592c008 Address=0x0009 FSR 0x013

| isn't alignment trap signals for memory management issues?

Alignment in ARM is quite interesting...

http://warmcat.com/_wp/2007/05/25/the-alignment-monster/

short story is this is just a warning and was handled.

| after a bit of waiting mediaplayer stopped crashing and started to
play the
| ringtone fine and then I started to hear ringtone again while
receiving a phone
| call just for once... on a next phone call -- same 'vibration only' (and
| mediaserver being very busy) and media player doesn't play -- just
sits waiting
| -- cpu is not busy.
|
| rebooted again (no busy klogd this time), didn't play tone, started media
| player -- started to play it and then started to play it while
accepting the
| phone calls.

Can't guess about that.

- -Andy

-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org

iEYEARECAAYFAkibh2cACgkQOjLpvpq7dMrw+ACfaM1cIwIOxra83ufEU5DzGLmi
h0QAnjn2/sMrmzXBHT6YxHRtzbCnnSDu
=yqaH
-END PGP SIGNATURE-

___
support mailing list
support@lists.openmoko.org
https://lists.openmoko.org/mailman/listinfo/support


Re: Uboot

2008-08-06 Thread Andy Green
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

Somebody in the thread at some point said:
| I had this problem when I put obexftp into the startup for my bluez.
  Turns out it doesn't auto-background, so it was hanging there just
after the startup.  Happily, I was able to ssh in, fix it, and kill the
obexftp to let it start up normal.  Can you ssh in via USB still?  If so
look at the last started process before your shell and see if it's hung.
|
|
|
| The process '/bin/ash /etc/rc5.d/S98qpe start' is constantly in the
| process list, so it looks like that's where it's hanging. If I kill this
| and restart it, it hangs again at the command-line as well until I ^C
| it. What is it about the most recent updates that could have caused this?

IIRC qpe is the thing that likes to sit there going through your SD card
contents after boot, making things slugglish.  Maybe that's why it is
hung?

- -Andy
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org

iEYEARECAAYFAkiZRSoACgkQOjLpvpq7dMpxXwCfbIGJXr8JhqOwR56NgtkOZLif
XzwAoJEzkoKd/LTGY3E/8W91IjuI/k5u
=FZFk
-END PGP SIGNATURE-

___
support mailing list
support@lists.openmoko.org
https://lists.openmoko.org/mailman/listinfo/support


Re: Led light stay on after boot

2008-08-06 Thread Andy Green
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

Somebody in the thread at some point said:
| Feydreva [EMAIL PROTECTED] writes:
|
| yeah, it turns off at one point... but it s not consistent...
|
| Well, it may be consistent to the fact that the phone eventually drops
| charging at one time even though the charger may be plugged-in (which
| is another bug) ?
|
| I'd bet for conjunction of several bugs ;)

It's not pretty in there.  I guess this is today's topic for debug action.

- -Andy
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org

iEYEARECAAYFAkiZSTEACgkQOjLpvpq7dMrCMQCeNpp5qDv6BYVO6ywWEy57u5GY
qxEAniyuLYwFeezKMTC++msXm4y1tSm8
=9kDU
-END PGP SIGNATURE-

___
support mailing list
support@lists.openmoko.org
https://lists.openmoko.org/mailman/listinfo/support


Re: Can't Boot Qtopia from microSD

2008-08-06 Thread Andy Green
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

Somebody in the thread at some point said:

| I changed the kernel name based on something I read stating it had to be
| called uImage.bin (is that wrong?)

It needs to match whatever you called it on the kernel commandline in
U-Boot, our ones use uImage.bin.

| 3. I upgraded my NAND U-boot to a post-July 23 version. I think I have
| the August 5 version now.

Great.

| 4. I edited by NAND U-boot menu to include:
| setenv menu_9 Boot from microSD part2 (ext2+ext2): setenv bootargs
| \${bootargs_base} rootfstype=ext2 root=/dev/mmcblk0p2 rootdelay=5
| \${mtdparts} ro\; mmcinit\; ext2load mmc 1 0x3200
| \${sd_image_name}\; bootm 0x3200

Ignoring the menu stuff for a moment, here is a canned one liner to do
something like what you want

setenv bootcmd  mmcinit \; ext2load mmc 2 0x3200 uImage.bin \;
setenv bootargs \${mtdparts} rootfstype=ext2 root=/dev/mmcblk0p2
console=ttySAC2,115200 loglevel=4 init=/sbin/init \; bootm 0x3200 ; boot

Notice on ext2load the first number is the partition index it should
use, counting from 1.

- -Andy
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org

iEYEARECAAYFAkiZXZwACgkQOjLpvpq7dMrZEACeIF7knB0jDkU2FkkYR03iKMGo
UAIAoIjFUKKff0Q+RAzA7j+v8L+ajji3
=+TDp
-END PGP SIGNATURE-

___
support mailing list
support@lists.openmoko.org
https://lists.openmoko.org/mailman/listinfo/support


Re: Unnecessary rant - Was: Re: USB networking in Ubuntu

2008-08-06 Thread Andy Green
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

Somebody in the thread at some point said:
| The problem I now have is that I am unable to get usb0 recognized as an
| interface by ifconfig.
|
| The command:
| ifconfig usb0 192.168.0.200
|
| results in
| SIOCSIFADDR: No such device
| usb0: ERROR while getting interface flags: No such device
|
| fr is plugged in and powered on? no suspend?
| if so, unplug it from the host pc, open a konsole/shell and do
|
| sudo tail -f /var/log/syslog
|
| have a look at the shell and plug in the fr again.
| any messages?
| there was a bug in one of the last u-boots, that broke about 50% of usb
| connections -- rebooting (the fr or the pc, couldn't figure out which,
try
| fr first) is supposed to help.
| the lattest u-boots are supposed to fix that.

Yes, there is another way to get this issue unfortunately too.  We moved
~ USB over Ethernet support on the Freerunner into a module.  But, it's
commonly written around that folks should update their kernel by DFU.
This was never a good idea for a modular kernel (it does nothing to
update the modules) and now it's a really bad idea.

If you update your kernel by the packaging from inside the Freerunner it
won't make this problem as it updates the modules at the same time (now,
anyway, there were other problems with that method until recently).

If that's what's happened (try the Uboot update first though because
that fits the symptom as well) best plan is nuke these guys on your
device (kernel + matching rootfs) with dfu

http://buildhost.openmoko.org/releases/Freerunner/

and start over.

- -Andy
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org

iEYEARECAAYFAkiZeFoACgkQOjLpvpq7dMqUXACcC7xnLGScuoq5b0RTmmwdZ79y
3hsAnRFCyGb9feZAtNUzDLKPDqmfD28J
=mbCO
-END PGP SIGNATURE-

___
support mailing list
support@lists.openmoko.org
https://lists.openmoko.org/mailman/listinfo/support


Re: Unnecessary rant - Was: Re: USB networking in Ubuntu

2008-08-06 Thread Andy Green
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

Somebody in the thread at some point said:

| Aug  6 11:25:36 ubuntu kernel: [  926.860809] cdc_acm 3-1:1.0:
ttyACM0: USB
| ACM device

| It seems the device is detected just as it is powering down.

Powering down or rebooting into U-Boot?

- -Andy
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org

iEYEARECAAYFAkiZgPMACgkQOjLpvpq7dMqcSwCeNrb3rGlDy2NgbZT3EzK2oMu5
OwEAn2vrHFjvDoJalRITickR6Mrn6Ki/
=y991
-END PGP SIGNATURE-

___
support mailing list
support@lists.openmoko.org
https://lists.openmoko.org/mailman/listinfo/support


Re: Can't Boot Qtopia from microSD

2008-08-06 Thread Andy Green
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

Somebody in the thread at some point said:

| You are right that I did not extract the rootfs, but that is because 1)
| I do not know how to extract the rootfs and 2) none of the various
| instructions on the wiki (which are inconsistent) worked for me.  For
| instance, when I do the various modprobe commands that are listed, all
| of them come back FATAL no such module, like so:
|
| [EMAIL PROTECTED]:/mnt/qtopia# modprobe mtdcore
| FATAL: Module mtdcore not found.
| [EMAIL PROTECTED]:/mnt/qtopia# modprobe jffs2
| FATAL: Module jffs2 not found.
| [EMAIL PROTECTED]:/mnt/qtopia# modprobe mtdram total_size=59024
| FATAL: Module mtdram not found.
| [EMAIL PROTECTED]:/mnt/qtopia# modprobe mtdchar
| FATAL: Module mtdchar not found.
| [EMAIL PROTECTED]:/mnt/qtopia# modprobe mtdblock
| FATAL: Module mtdblock not found.

These are all built-in to the monolithic kernels from us, so there are
no modules for them.  They're all needed to boot from mtd / jffs2 which
is normal for us.

| I should also emphasize to an earlier responder that I want to place
| both the kernel and the rootfs on the same ext2 partition, namely
| partition 2.  I believe this is possible, and I'd prefer it to having
| the kernel in the first FAT partition.

| Andy's email will likely be helpful here, but clearly I have to figure
| out how to extract the jffs2 first.

There should be a .tar.gz floating around for rootfs normally as well,
these are obviously more amenable to unpacking into the rootfs.

- -Andy
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org

iEYEARECAAYFAkiZ1/4ACgkQOjLpvpq7dMqdAgCgk37ZNuBM92BpaIPqs9OpWjKq
ScIAn05sPELF+ulrpwpD6wdw0ABX020I
=CsKG
-END PGP SIGNATURE-

___
support mailing list
support@lists.openmoko.org
https://lists.openmoko.org/mailman/listinfo/support


Re: Can't Boot Qtopia from microSD

2008-08-06 Thread Andy Green
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

Somebody in the thread at some point said:
| Andy Green wrote:
|
| setenv bootcmd  mmcinit \; ext2load mmc 2 0x3200 uImage.bin \;
| setenv bootargs \${mtdparts} rootfstype=ext2 root=/dev/mmcblk0p2
| console=ttySAC2,115200 loglevel=4 init=/sbin/init \; bootm 0x3200
; boot
|
| Notice on ext2load the first number is the partition index it should
| use, counting from 1.
|
| Minor correction - the parameter for ext2load and fatload is
| device:partition, so partition 2 would be ext2load mmc 1:2 
| IIRC this is documented in the online help for fatload but not for
| ext2load.
|
| For troubleshooting you can also run mmcinit; ext2ls mmc 1:2 / from
| the u-boot console to make sure that u-boot is able to access the
| partition, and to see what files are on the card. There is also a
| fatls command to look at a VFAT partition.

My mistake, I just edited my canned one I use for first partition off
the top of my head.  One way or another personally I only ever pulled
kernel from the first partition on SD Card myself, VFAT or EXT3.

- -Andy
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org

iEYEARECAAYFAkiZ2GcACgkQOjLpvpq7dMrApwCcCsMPGTV4FDanCxmZxUhEJLgs
OE0AninO8pLv3/lFF2UCWdPmeskpS7DK
=+KGy
-END PGP SIGNATURE-

___
support mailing list
support@lists.openmoko.org
https://lists.openmoko.org/mailman/listinfo/support


Re: Led light stay on after boot

2008-08-05 Thread Andy Green
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

Somebody in the thread at some point said:
| Ever since I flashed the latest firmware from
| http://buildhost.openmoko.org/daily/freerunner/200808/20080805/ , the
| led light stays light after boot.
|
| The files I used are :
|
http://buildhost.openmoko.org/daily/freerunner/200808/20080805/openmoko-openmoko-devel-image-glibc-ipk--20080805-om-gta02.rootfs.jffs2
|
http://buildhost.openmoko.org/daily/freerunner/200808/20080805/uImage-2.6.24+git27+1a95efdc23966343a3b432ceef2f6c24dc0e2052-r2-om-gta02.bin
|
| I have never updated my u-boot since I got the phone. Does this have
| something to do with that?

Yeah it indicates charging activity now.

- -Andy
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org

iEYEARECAAYFAkiX/gcACgkQOjLpvpq7dMocyQCfU3lQ936OXKUgJQn9HCmHstYX
cGkAn0ROh1rBAn3ht6DI+zE7F2+jAe/o
=dWnK
-END PGP SIGNATURE-

___
support mailing list
support@lists.openmoko.org
https://lists.openmoko.org/mailman/listinfo/support


Re: Led light stay on after boot

2008-08-05 Thread Andy Green
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

Somebody in the thread at some point said:

| Gta02v5 is the mass-production version.

Yes that one is good for A6 too when they come.  A6 is same as A5 except
for very minor changes in components.

| I took the plunge and flashed the 'fore mentioned u-boot. Boot seems
| faster. And USB networking works now!! :-)

Great.

- -Andy
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org

iEYEARECAAYFAkiYAxwACgkQOjLpvpq7dMplrwCffdIpjQJI4VpmeANWA0oJhE7k
PC4An1GQa7MvdMHY5AWyt5BBCNpG75j5
=wQHT
-END PGP SIGNATURE-

___
support mailing list
support@lists.openmoko.org
https://lists.openmoko.org/mailman/listinfo/support


Re: Buzzing sound

2008-08-05 Thread Andy Green
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

Somebody in the thread at some point said:
| 2008/8/5 Joerg Reisenweber [EMAIL PROTECTED]:
| Only thing you may try for now is changing GSM-provider, as it seems
to be
| very clear the issue is to be found on some GSM-networks, and not (or
less)
| on some others. Generally the 1800/1900 networks seem to be more
likely to
| cause this problem than the 850/900 networks. For Germany I got reports f
| noise only for E+ network (and resellers)
|
| My experience on Blau (E+) is that I hear the caller excellently - but
| the caller complains that buzzing is very bad.

Do they say the buzzing is consistent, or it comes and goes?

- -Andy
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org

iEYEARECAAYFAkiYKrQACgkQOjLpvpq7dMogwACfWP2yWOynrTBXpFmzK0q4cYES
zCYAn0TCfUURl2/GpHwGTFMkANSzo2FI
=1KrK
-END PGP SIGNATURE-

___
support mailing list
support@lists.openmoko.org
https://lists.openmoko.org/mailman/listinfo/support


Re: Buzzing sound

2008-08-05 Thread Andy Green
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

Somebody in the thread at some point said:
| 2008/8/5 Andy Green [EMAIL PROTECTED]:
| Do they say the buzzing is consistent, or it comes and goes?
|
| I've not asked in any detail. What frequency of coming and going has
| been reported? Are we talking seconds or minutes?

Like Ole says when I started looking at it, by calling a landline in the
same room and listening to its receiver lying on the desk, it varied
tremendously and not in a repeatable way.  Eg, it appeared to vary by
orientation of the phone, but when I traced the path backwards, the buzz
did not return.

Also, I had to keep whistling during this to provoke transmission
actions that seem to be the start of the issue.

Next I am gonna try rubbing my tummy at the same time :-O  Maybe a bag
of frozen peas on my head :-O

- -Andy
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org

iEYEARECAAYFAkiYMV4ACgkQOjLpvpq7dMpyxwCdFiK1KB3faqXMokVh0rCH6MK4
A1EAn0cTwP2ZEifuLIdMpRPrBZ55lZGP
=Nar7
-END PGP SIGNATURE-

___
support mailing list
support@lists.openmoko.org
https://lists.openmoko.org/mailman/listinfo/support


Re: Uboot

2008-08-05 Thread Andy Green
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

Somebody in the thread at some point said:
| I've found the daily u-boot images at
| http://buildhost.openmoko.org/daily/freerunner/200808/20080805/
|
| Which one do I use? There are 3 versions for the gta02v5 - what is the
| difference?

I guess this is an outcome of the buildhost issues that are happening at
the moment.  They're all the same git hash and same size, I guess they
are all identical.  Normally there'd be just the one a day.

- -Andy
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org

iEYEARECAAYFAkiYdgcACgkQOjLpvpq7dMrohwCfaqoW9TXs0eHRB2NSQjP+SdjN
t9AAn0sqeCS3BvCRl7n72Kp3Dim149or
=s3pA
-END PGP SIGNATURE-

___
support mailing list
support@lists.openmoko.org
https://lists.openmoko.org/mailman/listinfo/support


Re: Buzzing sound

2008-08-05 Thread Andy Green
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

Somebody in the thread at some point said:

| Just a note, on the skelephone that I use I replaced the GSM antenna
| with a short length of patch wire, it still works (and some guy was
| complaining normal GTA02 is for hardcore hackers).  I guess if I
| shortened it further, it can force base station to tell me to use max
| power all the time.  RF performance seems fine with $WIRE like 5cm long.

...

| the device can emit. So maybe this could be one of the possible fixes for
| this issue (a very nasty one, that probably even won't yield the desired
| results) to reduce gain by crippling the built-in antenna.

Gah no not a suggested fix, just to remove one of the variables about
provoking the problem in as consistently bad a form as possible :-))

- -Andy
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org

iEYEARECAAYFAkiYfmIACgkQOjLpvpq7dMohLACggbn5tKAPbdzBeelvEtQQ8tbf
YUsAoIm3rNels44Xn6vLDJB2GQHioXMZ
=dvKl
-END PGP SIGNATURE-

___
support mailing list
support@lists.openmoko.org
https://lists.openmoko.org/mailman/listinfo/support


Re: Buzzing sound

2008-08-05 Thread Andy Green
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

Somebody in the thread at some point said:
| Am Mi  6. August 2008 schrieb Andy Green:
| Somebody in the thread at some point said:
|
| | Just a note, on the skelephone that I use I replaced the GSM antenna
| | with a short length of patch wire, it still works (and some guy was
| | complaining normal GTA02 is for hardcore hackers).  I guess if I
| | shortened it further, it can force base station to tell me to use max
| | power all the time.  RF performance seems fine with $WIRE like 5cm
long.
|
| ...
|
| | the device can emit. So maybe this could be one of the possible
fixes for
| | this issue (a very nasty one, that probably even won't yield the
desired
| | results) to reduce gain by crippling the built-in antenna.
|
| Gah no not a suggested fix, just to remove one of the variables about
| provoking the problem in as consistently bad a form as possible :-))
|
| Yup, but as I elaborated in my last posting, the resulting RF field
will stay
| at same level (to the point where MaxTXPower of amp limits further
| adjustment), so probably will the noise. :-/

I didn't really get to believe it is specifically radiated RF coupling
yet, it can still be RF conducted via PSU.  Just something to bear in mind.

- -Andy
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org

iEYEARECAAYFAkiYl3sACgkQOjLpvpq7dMoSrQCfVPLV7pC07/+A7JHGxttLNq3p
//MAnjWcATY6UIV5cnIylInbdx6X5UP/
=U+ND
-END PGP SIGNATURE-

___
support mailing list
support@lists.openmoko.org
https://lists.openmoko.org/mailman/listinfo/support


Re: Upgrading u-boot needed ? - Was: Re: opkg upgrade breaks USB networking: g_ether module missing

2008-08-04 Thread Andy Green
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

Somebody in the thread at some point said:

| Could someone point to these valid u-boot*.bin files please ?

|
http://buildhost.openmoko.org/daily/freerunner/200808/20080804/uboot-gta02v5-latest.bin
| and got a message displayed advertising some april version (which I
| unfortunately didn't write down)... so I'm puzzled :(

Are you sure that it is not 4th Aug -- looks like 8th April in some
countries?  I doubt the latest symlink is broken for so many months.

| How could I check which version I'm using on my FR and if it's
| up-to-date or not ?

At the moment we don't stamp the image with the git head of the sources
it came from, we should move to that.  There's a build date on the menu
and coming out of the serial console as you saw.

At the moment this one is up to date for today and has the USB Insane
Spam Mode fixes.

http://buildhost.openmoko.org/daily/freerunner/200808/20080804/u-boot-gta02v5-1.3.1+gitr18+64eb10cab8055084ae25ea4e73b66dd03cc1a0cb-r0.bin

About the patchlevel, the git hash in the filename lets you check here

http://git.openmoko.org/?p=u-boot.git;a=shortlog;h=stable

and find out what patches are in and out.  In the stable trees (although
this isn't true for stable-2.6.26 yet) we only add patches, so if you
hover on each patch near the top (and look at the URL in your browser
status bar) you can find the matching git hash for the patch that was at
the top when the package was made.  Right now the topmost patch in git
is 64eb10cab80... mentioned in the package, so you know you have all the
patches in there.

- -Andy
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org

iEUEARECAAYFAkiWtDAACgkQOjLpvpq7dMoTdwCfXJ/SCs9i8NtyKknmE0tmUZj7
V5wAl1FBvODJVndD3Uo13ALcLcNu9lY=
=up7h
-END PGP SIGNATURE-

___
support mailing list
support@lists.openmoko.org
https://lists.openmoko.org/mailman/listinfo/support


Re: infinite pcf50633_irq messages , no X

2008-08-04 Thread Andy Green
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

Somebody in the thread at some point said:

| Thank you very much, I will play with those when i can connect again (i
| reflashed the root filesystem, which has the freerunner working as new
| again, but now i have lost usb0 on the host laptop).

:-/  Sorry about that, because the modules are packaged separately now
we pushed the Ethernet gadget into a module you need to install a couple
of new packages.

Thomas B. found you need to

# opkg install kernel-module-g-ether kernel-module-cdc-ether
# depmod

before you reboot into new kernel.

- -Andy
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org

iEYEARECAAYFAkiWtNkACgkQOjLpvpq7dMr+lwCfRtvTRmqnNp9TsDVlAICHuC4v
nWgAn17NT6xHKkhfRwficT0eeKDR0opE
=4e30
-END PGP SIGNATURE-

___
support mailing list
support@lists.openmoko.org
https://lists.openmoko.org/mailman/listinfo/support


Re: FSO Wifi - can ping but cannot download

2008-08-04 Thread Andy Green
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

Somebody in the thread at some point said:
| I've configured WiFi on a fresh FSO image the same way I had on a 2008.2
| image--by creating a wpa_supplicant.conf file and editing
| /etc/network/interfaces.  This configuration worked on the 2008.2 image
| but when I try implementing it on an FSO image, I cannot download
| anything.  I can ping sites and when using wget, I can connect to them,
| but downloading never commences.  It's not a DNS issue because ping and
| wget seem to be able to resolve correctly.  Any thoughts?

Sounds specific to tcp transport, ping is ICMP and DNS usually UDP.
Maybe connect to a local Linux box and run tcpdump on there to study
what it sees from the connection action.

- -Andy
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org

iEYEARECAAYFAkiWtaIACgkQOjLpvpq7dMrP8gCeNKkUgdWpiNNnrLXDD1shVpwJ
/FUAn0WZPY4IHJjjrERd8Ywr9yX+DGDL
=J4Ke
-END PGP SIGNATURE-

___
support mailing list
support@lists.openmoko.org
https://lists.openmoko.org/mailman/listinfo/support


  1   2   >