Rockbox mail archiveSubject: Re: Current state of "Flashbox"
Re: Current state of "Flashbox"
From: Doug <motiv01_at_earthlink.net>
Date: Fri, 29 Aug 2003 19:39:25 -0400
Actually, I've had this problem and mentioned that it exists several times,
but it never seems to get much response. I don't mean to keep pestering
about this problem, I'm just trying to help out people like you that have
the problem. It seems like there are several people that have this problem,
or at least that have mentioned it. It's a pain, but I am able to modify
the code and make my own builds that work, so it doesn't really bother me.
I'll paste the emails I posted about this problem one more time - your
problem sounds pretty much like the problem that I have, so as long as you
can make your own builds you can fix it. I have a build with bookmarking
that I built and have been using that I can email you if you'd like. It
looks like it's detecting your address as 0x300 just as it was mine. Here's
what I found out:
This message is from July 13th:
I have updated my Recorder 20 to boot from flash. I am able to use F1 and
boot the Archos firmware, however, I am getting ATA errors when trying to
boot rockbox from the second image. Most of the time I get ATA error: -21
after about 1 or 2 seconds of the splash screen, but occasionally I get ATA
error: -41 after about 10 seconds of the splash screen. I took a quick look
at the code and it seems like io_address_detect() and sometimes
freeze_lock() are failing. In both cases, the debug screen has master,
0x300 for ATA. Any idea of what might cause this? I have a 40GB IBM
Travelstar drive in the unit, but I'm not sure of the exact model. I have a
bleeding edge build (030714-0101) in flash also. Thanks a lot.
And this was from July 14th
Ok, here's what I did to get something that works. I don't quite understand
the low level ATA stuff enough to know why it works, but maybe someone will
know what needs to be changed in this detection code. In
io_address_detect() in ata.c, I basically forced it to the 'else' statement,
causing the address to be set to 0x200 instead of 0x300. I also commented
out the 'to be sure' check so the function would succeed. With these
changes, everything boots, very quickly :), and seems to work fine. I've
played around with it a little and I haven't noticed any side effects yet,
but I'm sure there could be some. Let me know if there's anything I can do
to help figure this out. Thanks a lot!
I hope this helps...
----- Original Message -----
From: "Owen Kalau" <owenk_at_canuck.com>
Sent: Friday, August 29, 2003 3:33 PM
Subject: Re: Current state of "Flashbox"
> On Fri, 29 Aug 2003 idc-dragon_at_gmx.de wrote:
> > - AFAIF, there are no ATA erors any more. This was only with the very
> > version.
> Actually, I had some problems with flashing when it first came out, so I
> reverted to the original firmware. I just tried it again today, and this
> is what I get:
> hit on - rockbox boot screen, cvs-030829...it waits for a bit, then it
> ATA error: -41
> Press ON to debug
> ON gives me the following:
> PADR: A27F
> PBDR: BDBF
> AN0: 3FF AN4: 00(0-1)
> AN1: 3E(2-5) AN5: 00(0-1)
> AN2: 01(C-D) AN6: 31(B-C)
> AN3: 17(4-7) AN7: 01(7-8)
> Batt: 5.14V 65%
> ATA: master, 0x300
Received on 2003-08-30