CTPCI and USB card

Discuss CT60/CT63, CTPCI, SuperVidel and EtherNAT hardware here.

Moderators: Mug UK, moondog/.tSCc., [ProToS], lp, Moderator Team

Latz
Captain Atari
Captain Atari
Posts: 169
Joined: Tue Mar 08, 2011 6:02 pm
Location: Germany
Contact:

Re: CTPCI and USB card

Postby Latz » Sat Sep 30, 2017 10:18 am

Kroll wrote:I replace mint060.prg in Auto folder but not working at all I can not start system, look at photo


When you test different FreeMinT versions you always have to replace mint060.prg in auto folder AND xaloader.prg + xaaes060.km
in /mint/1-*-*/xaaes folder. XaAES checks MinT kernel version at start!

Latz
Falcon/CT60, 14+128 MB RAM, CTPCI/Radeon 9250, EtherNeC, Carte Eiffel+LCD, 80GB IDE HD, DVD-RW.
Falcon/CT63, 4+128 MB RAM, CTPCI/Radeon 9250, NetUSBee, Carte Eiffel+LCD, 80GB IDE HD, CD-RW.

mikro
Atari God
Atari God
Posts: 1288
Joined: Sat Sep 10, 2005 11:11 am
Location: Brisbane, Queensland, Australia
Contact:

Re: CTPCI and USB card

Postby mikro » Sat Sep 30, 2017 10:22 am

Kroll wrote:I thing it is working, look at picture :)
This mint060prg from auto folder loading xaaes from 1-18-cur

OK, so usb_d80bd4cc is working (I hope you replaced also the usb binaries).

I replace mint060.prg in Auto folder but not working at all I can not start system, look at photo

Why only mint060? Why didn't you replace the usb binaries as well?

EDIT: I see Latz gets the idea, yes. Testing must be done in a very precise and systematic manner, my time is also limited for this kind of fun so if you say something is not working, it must be 100% clear in what conditions else whole testing can be trashed.

Kroll
Captain Atari
Captain Atari
Posts: 368
Joined: Fri Mar 09, 2012 10:07 am

Re: CTPCI and USB card

Postby Kroll » Sat Sep 30, 2017 10:39 am

mikro wrote:OK, so usb_d80bd4cc is working (I hope you replaced also the usb binaries).


Yes of course I replace not only Mint060.prg, but usb binaries too

mikro wrote:Why only mint060? Why didn't you replace the usb binaries as well?

I always replace all, you see the package usb_c17*.zip not start system at all

mikro wrote:EDIT: I see Latz gets the idea, yes. Testing must be done in a very precise and systematic manner, my time is also limited for this kind of fun so if you say something is not working, it must be 100% clear in what conditions else whole testing can be trashed.


Yes of course, I uderstand you and you can be sure in 100%

Latz
Captain Atari
Captain Atari
Posts: 169
Joined: Tue Mar 08, 2011 6:02 pm
Location: Germany
Contact:

Re: CTPCI and USB card

Postby Latz » Sat Sep 30, 2017 10:44 am

mikro wrote:EDIT: I see Latz gets the idea, yes. Testing must be done in a very precise and systematic manner, my time is also limited for this kind of fun so if you say something is not working, it must be 100% clear in what conditions else whole testing can be trashed.


Thank you! :D
I´m not a programmer but tested many Atari software; with MinT and all its files and dependencies it is indeed hard to keep an overview. :coffe:
But i think we´re on a good way -> thanks to Mikro!!!! :cheers:

Latz
Falcon/CT60, 14+128 MB RAM, CTPCI/Radeon 9250, EtherNeC, Carte Eiffel+LCD, 80GB IDE HD, DVD-RW.
Falcon/CT63, 4+128 MB RAM, CTPCI/Radeon 9250, NetUSBee, Carte Eiffel+LCD, 80GB IDE HD, CD-RW.

mikro
Atari God
Atari God
Posts: 1288
Joined: Sat Sep 10, 2005 11:11 am
Location: Brisbane, Queensland, Australia
Contact:

Re: CTPCI and USB card

Postby mikro » Mon Oct 02, 2017 9:34 am

Shame on me, I have totally overlooked Kroll's answer! And I'm just waiting for his test and waiting and waiting... stupid me.

Then, the "not starting at all". Man, try to be a bit more imaginative. ;) Of course there's no guarantee XaAES will work, it's a different kernel. Exactly as Latz has pointed out, you always need the combo of all parts. I was lazy to copy all the components by hand (xaaes*.km needs its own version of RSC files and all that crap) so I hoped you'll improvise and boot into a shell and run the usb loader from there. Is this an option for you?

Kroll
Captain Atari
Captain Atari
Posts: 368
Joined: Fri Mar 09, 2012 10:07 am

Re: CTPCI and USB card

Postby Kroll » Mon Oct 02, 2017 10:12 am

mikro wrote:Shame on me, I have totally overlooked Kroll's answer! And I'm just waiting for his test and waiting and waiting... stupid me.

Then, the "not starting at all". Man, try to be a bit more imaginative. ;) Of course there's no guarantee XaAES will work, it's a different kernel. Exactly as Latz has pointed out, you always need the combo of all parts. I was lazy to copy all the components by hand (xaaes*.km needs its own version of RSC files and all that crap) so I hoped you'll improvise and boot into a shell and run the usb loader from there. Is this an option for you?


Maybe something I do not understand you,
Sorry I have forgotten something, do some test, please remind me again I will do it
and maybe I missed something too.

User avatar
Ektus
Captain Atari
Captain Atari
Posts: 192
Joined: Mon May 24, 2010 2:58 am
Location: Germany
Contact:

Re: CTPCI and USB card

Postby Ektus » Mon Oct 02, 2017 10:32 am

Testing different versions is a PITA. From within the running system, you can't rename the 1-19-cur folder, which otherwise is the easiest way to switch versions. Rename mint.prg in AUTO-folder, rename corresponding system folder, reboot. No need to copy around kernel modules and stuff. But no, folder is write protected. So to switch versions, shut down and boot MagiC (to retain LFNs), rename the files, and do another reboot to start MiNT. Or run a boot manager, but I've never done that and they do add another layer of complexity.

Sorry for the rant, but it's one of the reasons I much prefer MagiC over MiNT.

With the current 1-19 from the other thread, I miss some boot messages on the screen (namely file system check). Witch 1-17 it automatically runs the check if the system was dirty, and shows a progress bar. With 1-19 and otherwise same settings, it just sits there for a minute or two with no output :-(

Regards
Ektus.
Schneider CPC464 (long retired), Atari Mega ST4 (retired), Falcon+CT2A, Falcon+CT63+CTPCI+Radeon 9250, Milan040+SCSI+Rage

mikro
Atari God
Atari God
Posts: 1288
Joined: Sat Sep 10, 2005 11:11 am
Location: Brisbane, Queensland, Australia
Contact:

Re: CTPCI and USB card

Postby mikro » Mon Oct 02, 2017 10:47 am

With newer kernels it's much easier -- every one of them has a unique ID instead of "CUR". That fsck issue is a known thing but for a world I can't remember whether it has been already fixed in recent kernels. If not, please report on https://github.com/freemint/freemint/issues, that's the only way to get your problems solved! :)

mikro
Atari God
Atari God
Posts: 1288
Joined: Sat Sep 10, 2005 11:11 am
Location: Brisbane, Queensland, Australia
Contact:

Re: CTPCI and USB card

Postby mikro » Mon Oct 02, 2017 10:49 am

Kroll wrote:Maybe something I do not understand you,
Sorry I have forgotten something, do some test, please remind me again I will do it
and maybe I missed something too.

What I mean is that instead of forcing XaAES to run with older kernels, go to mint.cnf, comment out the line with GEM= and uncomment the line with #INIT= and set it to something like INIT=/n/bin/bash (assuming 'n' is your / partition and you do have /bin/bash installed).

Then easily can do something like:

Code: Select all

cd /c/mint/1-19-cur/usb
./loader.prg

User avatar
Ektus
Captain Atari
Captain Atari
Posts: 192
Joined: Mon May 24, 2010 2:58 am
Location: Germany
Contact:

Re: CTPCI and USB card

Postby Ektus » Mon Oct 02, 2017 11:04 am

Testing USB, with the 1-19 kernel "post-pcibios" (from the network thread) the system froze upon connecting a thumb drive. Reset switch to the rescue, rebooting with the thumb drive still connected. System reaches desktop. Access LED on thumb drive blinks every 3s or so. There is no USB kernel module present. Connect USB keyboard -> system freezes again. Reset, both devices still connected. System freezes during startup (pressed "ENTER" key on USB keyboard after files system check). Reset. Press ENTER on normal keyboard to continue boot sequence after NFS message, system reaches desktop. Press any key on USB keyboard -> freeze up again.
Schneider CPC464 (long retired), Atari Mega ST4 (retired), Falcon+CT2A, Falcon+CT63+CTPCI+Radeon 9250, Milan040+SCSI+Rage

mikro
Atari God
Atari God
Posts: 1288
Joined: Sat Sep 10, 2005 11:11 am
Location: Brisbane, Queensland, Australia
Contact:

Re: CTPCI and USB card

Postby mikro » Mon Oct 02, 2017 11:23 am

Ektus: yep, that's more or less what Kroll is seeing, so far so good. If you find time and energy to test the stuff from usb.zip (see viewtopic.php?f=97&t=30680&start=25#p328544), that will help (I'm especially interested in the c17... version).

But same as Kroll, beware, don't try to mix it with your XaAES, let it boot into a shell.

User avatar
Ektus
Captain Atari
Captain Atari
Posts: 192
Joined: Mon May 24, 2010 2:58 am
Location: Germany
Contact:

Re: CTPCI and USB card

Postby Ektus » Mon Oct 02, 2017 11:48 am

Kroll wrote:
Ektus:
I noticed another problem, practically did not work left and right SHIFT key on my keyboard, no response after pressing under TOS and MagiC,
Under MiNT, when I press only left SHIFT on the screen shows "6", pressing right SHIFT letter "a".
What do you think about it ? Is this a problem in the TOS Drivers 1.02 ? or only with my keyboard.

The keyboard I checked on PC is OK
Unfortunately at this moment I have not other keyboard to check
In any question, please contact me, I hope now everything is clear.


Sorry, I missed this question. Strangely, my keyboard behaves the same. It used to be better. Currently running MagiC, lower case letters work, left-shift produces an "e", right-shift modifies something. CapsLock works.

My keyboard is a Cherry G270.

I've got a second keyboard, motion KB2609 with integrated USB 1.1 hub (2 sockets). The system won't boot with that one connected. It switches to the Radeon, but stays there with a blank white screen. That's got to be a bug in TOS.

Edit: Nope, no bug. Thermal problem, reboot sometimes won't work when running at 98MHz. I've reduced to 95MHz, the motion keyboard doesn't work, but doesn't interfere either.

Regards
Ektus.
Last edited by Ektus on Mon Oct 02, 2017 12:43 pm, edited 1 time in total.
Schneider CPC464 (long retired), Atari Mega ST4 (retired), Falcon+CT2A, Falcon+CT63+CTPCI+Radeon 9250, Milan040+SCSI+Rage

Kroll
Captain Atari
Captain Atari
Posts: 368
Joined: Fri Mar 09, 2012 10:07 am

Re: CTPCI and USB card

Postby Kroll » Mon Oct 02, 2017 12:06 pm

mikro wrote:Ektus: yep, that's more or less what Kroll is seeing, so far so good. If you find time and energy to test the stuff from usb.zip (see viewtopic.php?f=97&t=30680&start=25#p328544), that will help (I'm especially interested in the c17... version).

But same as Kroll, beware, don't try to mix it with your XaAES, let it boot into a shell.


MIkro:,
I repeat test for usb.zip, without loading XaEAS

First photo was done for usb_d80bd4cc and is working
Photo-0085.jpg


below photo for usb_c1790536, and what do you think ?
Photo-0086.jpg
You do not have the required permissions to view the files attached to this post.

mikro
Atari God
Atari God
Posts: 1288
Joined: Sat Sep 10, 2005 11:11 am
Location: Brisbane, Queensland, Australia
Contact:

Re: CTPCI and USB card

Postby mikro » Mon Oct 02, 2017 12:29 pm

Good, good. That certainly shows the bug is totally unrelated to PCI BIOS.

Try this one, it's somewhere in the middle between d80bd4c and c1790536. This will decide in which half of the commits the bug had been introduced. As the error reports some timeout, I wouldn't be surprised if that change was only about some value increased/decreased.

Tomorrow I'll prepare you the remaining set of builds to test (don't worry, not many left ;)), depending on the result from this one.
You do not have the required permissions to view the files attached to this post.

Kroll
Captain Atari
Captain Atari
Posts: 368
Joined: Fri Mar 09, 2012 10:07 am

Re: CTPCI and USB card

Postby Kroll » Mon Oct 02, 2017 12:55 pm

See picture of this test, I noticed that checking actually takes a little longer after running loader.prg, but Q:\ partitions are visible.

Photo-0089.jpg
You do not have the required permissions to view the files attached to this post.

User avatar
Ektus
Captain Atari
Captain Atari
Posts: 192
Joined: Mon May 24, 2010 2:58 am
Location: Germany
Contact:

Re: CTPCI and USB card

Postby Ektus » Mon Oct 02, 2017 3:15 pm

Testing usb_ba198af9 with a 16GB (INTENSO) brand USB thumb drive I can get a file listing of said drive in /q if it is plugged in during boot and without loading the mass storage driver. That's the built-in driver of TOS, I guess.
If I insert the drive after boot and then start loader.prg, it creates a new drive letter "y" but

Code: Select all

ls /y
ls: /y: Not a directory

I've got plenty partitions in that machine, the last regular drive letter being "W" or even "X".

Edit: Why W or X? 'Cause HDDRIVER may or may not assign a drive letter for the DVD drive. There are partitions C: through N: on the main IDE drive connected to the CTPCI IDE port, followed by partitions O: through R: on a 4GB SCSI drive, followed by S: on a ZIP 100, followed by T: through W: on a SCSI card reader, and then sometimes X: for the DVD (by HDDRIVER for use as DVD-RAM) and Z: for the DVD (assigned by Extendos).

When booting MagiC with a compatible thumb drive attached to USB, TOS first assigns letter Q: as that's the first free drive letter when HDDRIVER has used up all letters including P: (the limit for plain TOS without extensions). Then comes the second boot after the start of MagiC, where HDDRIVER re-scans all drives, assigns drive letters C: through W: and then the USB driver in TOS adds X: for the thumb drive.

Sadly, with MagiC, the partition on the thumb drive (type $0C, FAT32 Windows) cannot be read. HDDRUTIL sees the partition, but no drive and thus doesn't offer to format or partition it.

In case of MiNT, there's no re-scan and the thumb drive retains the letter Q:, but the following partitions are invisible.

What a mess :-(
Schneider CPC464 (long retired), Atari Mega ST4 (retired), Falcon+CT2A, Falcon+CT63+CTPCI+Radeon 9250, Milan040+SCSI+Rage

User avatar
Ektus
Captain Atari
Captain Atari
Posts: 192
Joined: Mon May 24, 2010 2:58 am
Location: Germany
Contact:

Re: CTPCI and USB card

Postby Ektus » Mon Oct 02, 2017 7:49 pm

I've found an ancient USB 2.0 18-in-1 card reader by Hama that is recognized by TOS and only accepts small cards (up to 2GB or some such). It is slow (xferrate reports 140kb/s), but otherwise working to access TOS-compatible media under both MagiC and MiNT. More modern card readers as well as some thumb drives are not supported.
Schneider CPC464 (long retired), Atari Mega ST4 (retired), Falcon+CT2A, Falcon+CT63+CTPCI+Radeon 9250, Milan040+SCSI+Rage

mikro
Atari God
Atari God
Posts: 1288
Joined: Sat Sep 10, 2005 11:11 am
Location: Brisbane, Queensland, Australia
Contact:

Re: CTPCI and USB card

Postby mikro » Tue Oct 03, 2017 12:04 am

Ektus: I'm sorry to say that but that kind of issues is probably never going to be fixed unless someone takes tosdrivers 1.01 package, bugfix it at least to the state where 1.02 is (to which are no sources available) and go from there.

Considering the number of CTPCI users out here, I wouldn't hold my breath for it. CTPCI is a dead product without latest source code and unavailable author, unfortunately.

mikro
Atari God
Atari God
Posts: 1288
Joined: Sat Sep 10, 2005 11:11 am
Location: Brisbane, Queensland, Australia
Contact:

Re: CTPCI and USB card

Postby mikro » Tue Oct 03, 2017 12:43 am

Hmm... totally accidentally, I have found a serious bug in the freemint usb code.

Please try the attached archive. If we are lucky, it will not work. Then try to rename storage_fixed.udx to storage.udd (delete/rename the original before). If that works -> voila!

If this wont work, no biggie, it's still a valid bugfix and we continue with testing other builds (perhaps via PM to avoid spaming this thread)
You do not have the required permissions to view the files attached to this post.

Kroll
Captain Atari
Captain Atari
Posts: 368
Joined: Fri Mar 09, 2012 10:07 am

Re: CTPCI and USB card

Postby Kroll » Tue Oct 03, 2017 1:34 am

Exactly,
I did as you said, and both versions storage.udd and storage_fixed.udd (after the first test storage.udd, delete one and change name from storage_fixed on storage.udd) work just as well,
I can read mass storage (see photo)
Photo-0071.jpg


What does it mean for us ?
Both versions behave exactly the same
You do not have the required permissions to view the files attached to this post.

mikro
Atari God
Atari God
Posts: 1288
Joined: Sat Sep 10, 2005 11:11 am
Location: Brisbane, Queensland, Australia
Contact:

Re: CTPCI and USB card

Postby mikro » Tue Oct 03, 2017 1:53 am

That means I have to build all the remaining kernels for you. :) Anyway, this bugfix isn't as important as I thought because that file was later replaced with something completely different (its C counterpart) so it wouldn't help us anyway.

I'll send you PM later today.

Kroll
Captain Atari
Captain Atari
Posts: 368
Joined: Fri Mar 09, 2012 10:07 am

Re: CTPCI and USB card

Postby Kroll » Tue Oct 03, 2017 2:02 am

Ektus wrote:I've found an ancient USB 2.0 18-in-1 card reader by Hama that is recognized by TOS and only accepts small cards (up to 2GB or some such). It is slow (xferrate reports 140kb/s), but otherwise working to access TOS-compatible media under both MagiC and MiNT. More modern card readers as well as some thumb drives are not supported.

You are right, I tested a few mass storage and only up to 2 GB working well.
On this photo you can see the first two are of 2GB capacity, the last one is 4GB and it is no visible when inserted into Netusbee.
Photo-0072.jpg
You do not have the required permissions to view the files attached to this post.

Kroll
Captain Atari
Captain Atari
Posts: 368
Joined: Fri Mar 09, 2012 10:07 am

Re: CTPCI and USB card

Postby Kroll » Tue Oct 03, 2017 2:06 am

mikro wrote:That means I have to build all the remaining kernels for you. :) Anyway, this bugfix isn't as important as I thought because that file was later replaced with something completely different (its C counterpart) so it wouldn't help us anyway.

I'll send you PM later today.

OK, as soon as I read the e-mail go to the test, but there is a big difference between us, we have a middle of the night :D

User avatar
Ektus
Captain Atari
Captain Atari
Posts: 192
Joined: Mon May 24, 2010 2:58 am
Location: Germany
Contact:

Re: CTPCI and USB card

Postby Ektus » Tue Oct 03, 2017 5:29 am

Just to clarify: I do have a netusbee, but currently there are no usb plugs connected (they were, but the cables are very stiff and the solder joints didn't last), so I'm only testing with a PCI USB card.

Meanwhile it's 07:30 here :-)
Schneider CPC464 (long retired), Atari Mega ST4 (retired), Falcon+CT2A, Falcon+CT63+CTPCI+Radeon 9250, Milan040+SCSI+Rage

Kroll
Captain Atari
Captain Atari
Posts: 368
Joined: Fri Mar 09, 2012 10:07 am

Re: CTPCI and USB card

Postby Kroll » Tue Oct 03, 2017 9:30 am

Ektus:,
If I presume, we have the same European time :)
So I'm currently testing on NETUSbee, because I found out earlier with eg Freemint 1-19, did not read mass storage at all.


Social Media

     

Return to “CT60 / CT63 Area”

Who is online

Users browsing this forum: No registered users and 4 guests