Rockbox Technical Forums

Installation / Removal => Manual Installation => Iriver - Installation/Removal/Flashing => Topic started by: nieroster on April 01, 2011, 01:02:31 PM

Title: bootloader 7pre4 behaviour
Post by: nieroster on April 01, 2011, 01:02:31 PM

whenever my iRiver H120 shuts down due to low battery the next time turned on the bootloader menu shows up, even when charging.

The bootloader menu even shows up when the Hold button is active. I would have assumed that an active Hold button prevents anything from happening, but obviously it is not a hardware solution but a button like any other.

The third thing I've noticed on my CF modded iRiver H120: when I plugin the USB cable prior to turning on the iRiver H120 (i.e. using th bootloader USB mode) the player does not boot correctly anymore after unplugging the USB cable. => ATA error:-80, no partition found. I have to do a reset in order to do anything with the player including swith off, which is a bit annoying when not having something with you that fits into the hole at the bottom.

Is all this intended behaviour? Are there currently any ongoing developments regarding the bootloader (making it a final 7)?

Title: Re: bootloader 7pre4 behaviour
Post by: AlexP on April 02, 2011, 05:54:47 AM
The bootloader menu shows up when you turn hold on, this is by design.

There are no current plans to update the H100 and H300 bootloaders, as if there is a problem in the bootloader on these players then you need specialist equipment to recover your player.  As a result, all official bootloaders are very heavily tested before release (note that v7pre4 is not official).  The person with the required equipment is not so active these days.  It is known that at least the H300 bootloader is broken in SVN, so there isn't any way that most people can test them.
Title: Re: bootloader 7pre4 behaviour
Post by: TriTL on June 26, 2011, 08:27:59 AM
Hi there,

I'm a bit sad about hearing that there are no plans to update the bootloaders.

After installing a 160 gb toshiba drive to my h120 i had a typical ata-32 error so that i only could boot to the orginal firmware to use the player/copy music etc. .
After installing the v7pre4 bootloader with the help of the old 20 gb drive the 160 gb drive worked fine.
I think integrating the new bootloader to rockbox will help many users.
Title: Re: bootloader 7pre4 behaviour
Post by: pondlife on June 28, 2011, 07:09:48 AM
As a warning, I can confirm that my H300 was killed merely by installing an updated bootloader.  This was using plain SVN code from about 18 months ago.

LinusN has managed to recover things with a bit of soldering, back to the standard bootloader, but we have no idea what is wrong with the later code; it may be that it merely got too large in some way, but I'm not risking any more attempts when it needs a round trip to Sweden and quite a lot of somebody else's time to fix it each failed attempt!


p.s. Of course, if you DO have the appropriate soldering skills, BDM equipment, coding skills AND lots of spare time, then I would indeed encourage you to tinker in this area and see if progress could ever be made.  ;D  Just don't go killing these lovely old boxes for my sake.
Title: Re: bootloader 7pre4 behaviour
Post by: gevaerts on June 28, 2011, 07:14:05 AM
LinusN now has my irivers with the express goal of adding the necessary connectors so they can be used for these tests.
Title: Re: bootloader 7pre4 behaviour
Post by: pondlife on June 28, 2011, 08:05:05 AM
That's great news.

I was attempting the first in this series of tests when I did what I did. ;)

FWIW, I built plenty of my own bootloaders which worked fine circa autumn 2007 (mainly for a working USB Bootloader mode and some remote control fixes).  

The last good one I built was constructed on 2/10/2007. It was probably based on SVN r14949 ("Thanks to LinusN, fix Coldfire bootloaders not booting - FS#7533") was likely the basis, but also contained some patches. The .hex file size is 4184540 bytes.

On 20/11/2009, aiming for an updated official bootloader, I built one clean/unpatched from SVN (probably r23674).  This one converted my H300 into a brick.  The .hex file size is also 4184540 bytes.

The file sizes don't mean much regarding the size of the bootloader content , but I have both of these .hex files if anyone wants a copy.

Title: Re: bootloader 7pre4 behaviour
Post by: TriTL on June 28, 2011, 06:47:42 PM
I have no idea about programing or compiling my own Rockbox but someday I'll learn how to :D

I use Rockbox since V 3.4 and really don't want to brick ma lovely player with it's high end sound quality.

Have someone got rockbox working with a harddrive bigger than 80 GB with the normal bootloader ?
(80 GB Toshiba drive worked fine in my player)
Or does anyone of the Developers can explain why there is a problem with bigger drives ?

sorry for my bad english.
Title: Re: bootloader 7pre4 behaviour
Post by: torne on June 29, 2011, 06:39:16 AM
Drives larger than 128GB need special handling which the stable bootloader doesn't include. There is no way to get it to work without installing a new bootloader. We can't easily update this bootloader at present because it's so difficult to test it. People are working on it.
Title: Re: bootloader 7pre4 behaviour
Post by: monoid on November 21, 2012, 10:31:17 AM
I would like to ask, what is the limit for disk size (for bootloader 7pre4)?

Is it 128 GB or more? Thanks!

And if only 128 GB, is there any hope there will be a new release of bootloader with higher limit? I am aware of the fact, there is a problem with bootloader source codes. :(
Title: Re: bootloader 7pre4 behaviour
Post by: torne on November 21, 2012, 02:22:15 PM
The patched 7pre4 version that people use supports any size of drive.
Title: Re: bootloader 7pre4 behaviour
Post by: monoid on November 22, 2012, 11:57:26 AM
Thanks a lot.

Does it (I mean - any size) include also SSD disk? I use 120 GB SSD without problem, so 240 GB SSD would be also without a problem?