Page 16 of 28
Re: DFB1 Support thread
Posted: 25 Nov 2022 16:18
by Badwolf
I did try patching TOS4.04 the way Hatari does to declare TT-RAM nice and early in the process, but can't find that image for the life of me.
For anyone with flash on their DFB1, that might be worth trying.
It could be hard coded to your specific board or, if someone's feeling particularly fly, the TT-RAM detect-and-test code from TOS3 could be patched in.
https://github.com/hatari/hatari/blob/m ... tos.c#L692
This overwrites some early code to jump to the patched area
https://github.com/hatari/hatari/blob/m ... tos.c#L693
This is the new code location.
https://github.com/hatari/hatari/blob/m ... tos.c#L185
And this is the actual code.
Anyone want to have a go with a hex editor and TOS404.IMG ?
BW
Re: DFB1 Support thread
Posted: 25 Nov 2022 16:44
by frank.lukas
I order some HN27C4096 ...
Re: DFB1 Support thread
Posted: 25 Nov 2022 17:17
by Badwolf
OK, I had another go.
This is untested other than in Hatari (with --patch-tos 0 set).
This applies most of the Hatari patch but hard codes the TT-RAM declaration to be 64MB rather than looking at a system variable.
I might have a go on this tomorrow with my card that has flash on it, but I can't test SCSI read/write, I'm afraid.
This does not set _FRB
BW
Re: DFB1 Support thread
Posted: 25 Nov 2022 19:49
by markus0321
Badwolf wrote: 25 Nov 2022 17:17
OK, I had another go.
This is untested other than in Hatari (with --patch-tos 0 set).
This applies most of the Hatari patch but hard codes the TT-RAM declaration to be 64MB rather than looking at a system variable.
I might have a go on this tomorrow with my card that has flash on it, but I can't test SCSI read/write, I'm afraid.
This does not set _FRB
BW
With HDDRIVER 11.13 it loads and freezes on screen as attached.
IMG_20221125_203709.jpg
Version 9.0 similarly only 4 bombs appear...
IMG_20221125_204430.jpg
Re: DFB1 Support thread
Posted: 26 Nov 2022 00:04
by Rustynutt
markus0321 wrote: 25 Nov 2022 08:51
Badwolf wrote: 24 Nov 2022 21:46
It's probably not worth repartitioning just to test an ancient bit of software, it was just on the off-chance the partition tables were compatible already.
I think it's time to raise the issue with Uwe.
BW
I did some more testing using the AHDI driver.
It works stably, the system does not hang when I have ALT-RAM memory active in TOS4.04. Too bad AHDI doesn't use more than 1GB of disk space.
EDIT:
I'll try to check if ICD PRO will work properly, I'm not sure but I think it supports drives up to 16GB and certainly up to 10GB, so it should be suitable for old SCSI drives.
Make sure if you partition with ICD Pro, and decide to return to HD Driver later, run the ICD fix patch included with HD Driver.
.There's a readme in the file archive that covers it.
Re: DFB1 Support thread
Posted: 26 Nov 2022 00:46
by Badwolf
markus0321 wrote: 25 Nov 2022 19:49
With HDDRIVER 11.13 it loads and freezes on screen as attached.
IMG_20221125_203709.jpg
Version 9.0 similarly only 4 bombs appear...
IMG_20221125_204430.jpg
You can’t also run MAPROM!
This will only work if HDDriver recognises TTRAM and adds the FRB. It’s an experiment, but it does MAPROM/FASTRAM’s job.
BW
Re: DFB1 Support thread
Posted: 26 Nov 2022 02:59
by Rustynutt
markus0321 wrote: 24 Nov 2022 21:33
Badwolf wrote: 24 Nov 2022 20:36
Great research, Markus. Thank you.
Does anyone who's actually experiencing this problem have an open ticket with Uwe at the moment? The above is extremely useful information if it could be passed on, please.
In the meantime, does ADHI recognise your partitions?
BW
I only found version 6.061 and it doesn't see the partition. I will have to prepare everything again on the second SCSI disk and transfer the files for testing.
EDIT:
It succeeded and only sees the first two partitions from this SCSI disk. I was able to install AHDI on drive C and boot from SCSI. From what I've managed to check, programs that didn't run before now work properly and I don't have to boot from a floppy disk. Unfortunately, I do not see the partition above the letter D, so there is no way to check if the archives are unpacking correctly. For sure I have to set up smaller partitions for AHDI or use a smaller disk size.
Here's an archive of AHDI, believe the latest as well as the latest HDX Tools. It's sort of a jumbled mess but served me well back then.
Just to add them out there, think you've found they don't solve the issue.
Re: DFB1 Support thread
Posted: 26 Nov 2022 12:48
by markus0321
Rustynutt wrote: 26 Nov 2022 02:59
markus0321 wrote: 24 Nov 2022 21:33
I only found version 6.061 and it doesn't see the partition. I will have to prepare everything again on the second SCSI disk and transfer the files for testing.
EDIT:
It succeeded and only sees the first two partitions from this SCSI disk. I was able to install AHDI on drive C and boot from SCSI. From what I've managed to check, programs that didn't run before now work properly and I don't have to boot from a floppy disk. Unfortunately, I do not see the partition above the letter D, so there is no way to check if the archives are unpacking correctly. For sure I have to set up smaller partitions for AHDI or use a smaller disk size.
Here's an archive of AHDI, believe the latest as well as the latest HDX Tools. It's sort of a jumbled mess but served me well back then.
Just to add them out there, think you've found they don't solve the issue.
Thanks for the files. I know I didn't solve the problem.
On AHDI, the problem is practically not revealed. With HDDRIVER ver <10.0 the crashes are sporadic but still noticeable from time to time. ICD drivers also don't work properly, yesterday I tried to run them, but I didn't even install them on the SCSI drive because when I started them from the floppy disk, I couldn't run the programs from the disk.
Re: DFB1 Support thread
Posted: 26 Nov 2022 20:46
by Badwolf
Unless you plan to switch to ICD or AHDI, there is no further point in testing them. We’ve proved HDDriver isn’t behaving as expected, so now a bug report needs to be raised with Uwe.
Specifically: HDDriver does not seem to honour the _FRB cookie when it is set after HDDriver initially loads.
Please someone who is experiencing this problem raise this on his forum.
BW
Re: DFB1 Support thread
Posted: 26 Nov 2022 22:29
by Badwolf
Badwolf wrote: 25 Nov 2022 17:17
This is untested other than in Hatari (with --patch-tos 0 set).
This applies most of the Hatari patch but hard codes the TT-RAM declaration to be 64MB rather than looking at a system variable.
This does not set _FRB
Just tried this on my DFB1 that has on-board flash and it does successfully declare 64MB TT-RAM very early in proceedings. Obviously FASTRAM/MAPROM should not be used with this image, therefore.
I don't know if this would help with the HDDriver problem or not as the examples Markus has tried were both with MAPROM still in the AUTO folder.
Incidentally, I've released a new version of DFB1's flashing application. No real difference to the binary that's been floating around, but the source is now also available:-
https://github.com/dh219/DFB/tree/main/DFB-Tools
Cheers,
BW