Issues writing on flash dsks

[dkintana]

Hi!

First of all, my name is David Kintana. I'm from Spain and I'm a fanatic of MS-DOS. I played with it since late 80's and, after some years of fun with it, I drove myself to Linux (fantastic, too) and Windows (With this one I had no fun at all, but it's mandatory at work today).
But some months ago I remembered those sunny days with MS-DOS while browsing FreeDOS website and I took the decisition of build a DOS station with some hardware that was at home. One of the musts was to have USB support, nd after some searching on the web, I found this great (very great) page. Thanks for your effort, Bret; your job is fantastic.

Ok. Now the problem.
I have several flash disks and, althought some of them works great with USBDRIVE, I have three that make extrange things.

The first one is a Kingstom DataTraveler 102. When I try to obtain a directory listing, the DIR comand gets stucked triying to obtain the remaining disk space at the end of the list (The directory listing gets fast and correct), and I have to reboot the computer to take control again, or to unplug the device and then plug another flash to obtain an error accesing the device and the message "Abort, Retry, fail". If I try to copy any file (I've tried with autoexec.bat), the operation takes several hours and end with an error accesing te device, with the same message as above.

The second one, is an adaptor from USB to micro SD. The simptoms are the same as above. Perhaps the Kingston One internally is some kind of adaptor from USB to SD?

The last one, is from "USBest Technologies" (Who knows?). This time, the directory listing works great. No probs. But when I try to copy autoexec.bat, I get the same problem: several hours waiting.

In all cases, the led gets blinking all the time while the operation takes place.

I've proved them with FreeDOS 1.0 and MS-DOS 7.1, with Intel and with VIA chipset. I've ran USBASPI before USBUHCIL but no luck.

An ideas of what's happening?

Thanks a lot!

[Bret]

The first one is a Kingstom DataTraveler 102. When I try to obtain a directory listing, the DIR comand gets stucked triying to obtain the remaining disk space at the end of the list (The directory listing gets fast and correct), and I have to reboot the computer to take control again, or to unplug the device and then plug another flash to obtain an error accesing the device and the message "Abort, Retry, fail".

I know there are problems with certain USB disks, and some of the ones made by Kingston (but not all of them) definitely have some problems. I will try and get this fixed in the next version of USBDRIVE. You'd actually be amazed at how many USB manufacturers are non-compliant with one or more of the standards they're supposed to follow.

Also, just so you know, if a drive is formatted in FAT32, it will always take a long time after the directory listing is displayed for it to display the disk space at the end, especially the first time you access it. DOS needs to read the entire FAT to figure out how much space there is, and this takes a long time (usually a few minutes) with FAT32. You can try increasing the /X value and see if that helps.

If I try to copy any file (I've tried with autoexec.bat), the operation takes several hours and end with an error accesing te device, with the same message as above.

Again, I know there are problems with certain disks, including Kingston's. A couple of people have sent me disks that don't work properly, and I will use them for testing before the next release of USBDRIVE. When I released the program before, it worked OK on all of the test disks I had at the time.

The second one, is an adaptor from USB to micro SD. The simptoms are the same as above. Perhaps the Kingston One internally is some kind of adaptor from USB to SD?

The Kingston one probably doesn't have anything to do with SD. Your SD converter probably just has some quirky timing issue or something like other problem disks have. I've tested it with some multi-port multi-media readers that use SD, and it has worked with those. So it's not just an SD thing.

The last one, is from "USBest Technologies" (Who knows?). This time, the directory listing works great. No probs. But when I try to copy autoexec.bat, I get the same problem: several hours waiting.

If you want to, you can send me the disks that don't work and I'll use them for testing. Just keep in mind that I'll probably need to keep them for many months, since I'm not sure when I'll start working on USBDRIVE again or when I'll be able to return them to you. I don't have a lot of time to dedicate to this project, though I'm working on it as I can.

[dkintana]

Bret:

Thanks a lot for your fast answer. I will play with the /X param and try to see the differences, if any, and update the post. I hope this post will help some others with the same issues.
I have valuable data in those flash disks, so I will try to make a copy. If It's possible to me to make that backup, I will send you the drives. Don't worry about sending back. It will be a present

Thanks again!

[Bret]

Cool - I like presents!

[dkintana]

Finally the drives worked!!!!

I bought a new usb card for connecting the front usb ports and to increase the number of them. The card was one that carries a VIA chip, wit 5 ports (One of them internal) and bought in eBay. eBay is plenty of exactlly this card. 15€ including shipping costs.

Well. The surprse has been when I have pluged one of the problematic drives in this new card and it has worked without any issue. Directory listings and copy of files to and from the device, without problems.

My card is an old PIII card with a BIOS that only supports usb keyboard and that makes extrange things with it, so, probably, it has a buggy implementation of the USB subsystem and thats what is making that some pen drives work fine while others not.

So, Bret, I think you will not receive any presents... At least pen drives

Thanks a lot!

[Bret]

That's great news!

It probably also means I need to make some improvements to the host controller drivers (USBUHCI & USBUHCIL), though I'm not exactly sure what those might entail.

[dkintana]

Bret, I can experiment with any piece of code that you may send me, If you dare.
I f you think Me and my hardware can be helpfull in any way, don`t hesitate to contact me. I will be more than pleased to help you in any way.

Thanks a lot.

[dkintana]

Bret, here is a detailed excerpt from one dos utility for obtaining the PCI info from a system. This excerpt is for the USB partt of the VIA chipset that is not running properly with your software. If you like, I can put all of the text or send it to you in any way:


Vendor 1106h VIA Technologies Inc
Device 0691h VT82C693A Apollo Pro 133 System Controller
Command 0006h (Memory Access, BusMaster)
Status A210h (Has Capabilities List, Received Master Abort, Detected Parity Error, Medium Timing)
Revision 44h, Header Type 00h, Bus Latency Timer 00h
Self test 00h (Self test not supported)
PCI Class Bridge, type PCI to HOST
Address 0 is a Memory Address (anywhere in 0-4Gb, Prefetchable) : D0000000h
New Capabilities List Present:
AGP Capability, Version 1.0 (AGP 1x and/or 2x support)
AGP Speed(s) Supported : 1x 2x
FW Transfers Supported : No
>4Gb Address Space Supported : No
Sideband Addressing Supported : Yes
Maximum Command Queue Length : 8 bytes
AGP Speed Selected : None Selected
FW Transfers Enabled : No
>4Gb Address Space Enabled : No
AGP Enabled : No
Sideband Addressing Enabled : No
Current Command Queue Length : 1 byte


Vendor 1106h VIA Technologies Inc
Device 3038h VT82C596B USB Controller
Command 0005h (I/O Access, BusMaster)
Status 2210h (Has Capabilities List, Received Master Abort, Medium Timing)
Revision 11h, Header Type 00h, Bus Latency Timer 20h
Self test 00h (Self test not supported)
Cache line size 32 Bytes (8 DWords)
PCI Class Serial, type USB (UHCI)
Subsystem ID 12340925h VIA USB Controller
Subsystem Vendor 0925h Known Bad Subsystem ID - no Vendor ID Available
Address 4 is an I/O Port : 0000D400h
System IRQ 9, INT# D
New Capabilities List Present:
Power Management Capability, Version 1.1
Does not support low power State D1 or D2
Does not support PME# signalling
Current Power State : D0 (Device operational, no power saving)


Thanks a lot!

[Bret]

Interesting, especially the "Known Bad Subsystem ID". Unfortunately, I don't think I can do much without having the hardware directly in front of me, and even then it could take a long time to figure it out.