[OLPC-devel] Re: LinuxBIOS and integrated system debug status (Friday night).

[Mitch Bradley]

Jim Gettys wrote:
> OK, here's where we stand after today's intensive debugging (Build 79).
>   1) the Marvell wireless patch to LB does not appear to work, so the
> Marvell wireless won't run from LinuxBIOS: the existing patch was known
> to possibly be problematical.  This still would block many users from
> conversion.
>   2) If a flash key/disk running ext3 is not shutdown cleanly, LinuxBIOS
> won't boot the file system.  This means you have to take the key or disk
> and clean it on a different machine to reboot the machine. This is not a
> blocker, but is sure irritating.
>   3) after following the directions to install an image to flash
> (uneventfully), the board I used will no longer boot anything; Samat saw
> this problem a couple weeks ago and again tonight, but no one else had
> reproduced it until I did.
Actually, I had that problem last week, but I misinterpreted it as being 
related to USB devices, because
last week's occurrence went away the next day after I fiddled with the 
USB devices.  But it happened to me again
this Thursday night.  I turned off the board and switched to a different 
board.  The next day, I tried the
"bad" board and it worked.
> The machine says "Jumping to LinuxBIOS" on
> the serial port and then does nothing.  Samat has seen this problem on a
> different board tonight as well; it started working again after being
> unplugged for a while (20 minutes). This smells like an uninitialized
> register in the NAND flash,
I have never knowingly used the NAND flash, so I question whether it's 
related to that.

I have added a trac ticket describing my experiences with this problem.  
When other people see it,
please update the ticket with detailed information concerning what you 
were doing.  Perhaps enough
such reports will give us a clue where to look.
> or some such problem. Not a blocker to a
> release, as we can tell people not to install onto NAND until we find
> and fix it.
>   4) the current sugar in build 79 looks to not scale itself properly to
> the 1024x768 resolution screen I chose for the default since that is the
> most likely to "just work on everyone's screen" size.
>
> On the good side:
>   o We have installation directions, that are now pretty good.  Whether
> they are good enough for non-native speakers of English to follow is not
> yet known.
>   
Now that the USB key image is hundreds of megabytes instead of 5, the 
save/overwrite/restore aspect
of the installation procedure is no longer a good idea.
>   o We determined that FC6 and Ubuntu Edgy cannot successfully build a
> ROM image; the resulting image hangs long into the boot sequence.  FC5
> and Ubuntu Dapper can build an image reliably.  So we'll stop beating
> our heads on this wall for the moment.  Half of today was lost on this
> one.
>   o The system now comes up 1024x768x16 at 60hz all the way into Sugar,
> using the Jordan's EXA driver.  The EXA driver runs x11perf without
> problems: rendercheck reports some problems.
>
> I don't think we're quite ready to ship this one yet. Close, very, very,
> very close, but not quite.  1) and 2) are the (potential) blockers.  I'd
> probably settle for the Marvell working, and live with the mount problem
> temporarily.  We can tell people not to use the NAND until we track it
> down. The sugar problem is pretty minor.
>
> Presuming we get 1) (and maybe 2) fixed, I'd like at least 3 more people
> who have *never* installed LinuxBIOS to do the installation and vet the
> instructions, before inviting more wholesale conversion.
>
> Thanks for everyone for their hard work.  This is making good progress,
> even if not quite as fast as we'd like.
>                           Regards,
>                                  - Jim
>